DE69129066T2 - Method and device for a dodecahedral imaging system - Google Patents

Description

The invention relates generally to an image display system comprising a composite camera and projector construction in coordination with one another, and more particularly to the geometric arrangement of the elements of both the camera system and the projection system according to the twelve-sided dodecahedron which enables the projection of apparently continuous images over the interior of a dome or spherical movie theater. It also relates to panoramic television camera and projection systems.

Planetariums typically present the awe-inducing illusion of transporting viewers to an entirely new environment, a starry night outdoors, with the sky occupying the entire curved field of view. Laser shows such as Laser Jmages Inc.'s Lasenum also use these dome-shaped theaters to display colorful structures surrounding the viewer. But these dome-shaped theater shows do not include films that extend across the curved interior surface, because of the lack of an effective device that captures and projects a continuous image over such an extremely wide field of view with high resolution and minimal distortion, so that a complete environment is captured and projected.

The optimum image produced by a normal taking lens is circular in shape because increased distortion and loss of resolution typically depend on the distance from the optical center. This lens image is usually cropped to a rectangular shape, which may traditionally be derived from the shape of paintings, drawings and theater stages. However, this wastes a large portion of the usable lens image area. A rectangular image with an aspect ratio of 1:1.85, which is commonly used for motion pictures, can only cover 53% of this circular area. Widescreen Film formats such as Cinemascope, Vistavision and Panavision, which have aspect ratios of 1:2.2, use even less of this optimal image area. Therefore, the further an aspect ratio is from a circle, the worse the average image quality becomes.

In order to present a larger image that fills more of the audience's field of view while maintaining an acceptable level of resolution, large film formats such as 70mm and IMAX film have been introduced. But the motion picture footage that uses this film is produced for projection on flat screens and is unsuitable for a dome theater because a shot produced for a flat screen and shown on a curved surface such as a dome will distort the image unpleasantly. To create the illusion of a continuous image filling the interior of a hemisphere, a special shooting technique is required to make an image suitable for such a screen.

An image spanning the entire hemisphere can be captured by means of a fisheye lens and then projected back through the same type of lens onto a dome-shaped screen, using virtually the entire circular lens image. One such system for an inflatable dome was developed for Hitachi in the early 1970s. Another partial example is the OMNIMAX system of Imax Systems Corp., Toronto, Canada, which, like IMAX, uses a horizontal 70mm film image 15 perforations wide to project an elliptical image 180º wide, 100º above the horizon and 20 below. But the edges of the image are still visible, creating the effect of a film projection rather than a real environment. The extremely distorted image caused by the fisheye shot is difficult to decipher unless it is projected, and because this amount of barrel distortion is incompatible with normal film and television shooting, any additions or changes to the image, such as titling or special effects, are difficult to make. Also, because a single image can be distorted during the Because projection has to be stretched over such a large area, brightness and resolution typically suffer, especially at the periphery. Because of the film format and mechanical transport required, the camera also tends to be very bulky and heavy.

Two annular imaging and projection systems that use a single camera to produce annular images covering the horizon are described in US Patent US-A-3 209 073, which illustrates a panoramic television image using a spiral scanning system, and US Patent US-A-4 100 571 describes a laser imaging system used in a flight simulator. However, such systems are only for viewing around the horizon and neglect the space above and below the viewer. Also, both systems use low resolution scanning and projection techniques that would produce images of poor overall quality.

Panoramic film cameras have also been designed to scan a wide field of view at high resolution, usually by rotating the camera around a curved film plane to sequentially expose a panorama through a narrow, moving slit on the film. This produces high quality images. However, these views generally lie around the horizon, neglect the area above and below the viewer, and require a drum-shaped projection surface that corresponds to the cylindrical plane of the film image. The amount of mechanical movement involved in each exposure makes film shooting very difficult, and the amount of film involved makes the system cumbersome, and the extremely wide aspect ratio of the film image makes it unsuitable for projection using standard projection lenses.

In order to provide better resolution and brightness while working with recognizable individual images, multiple image systems have been developed that can cover a wide field of view. The use of multiple images also allows the choice between either different images on the different screens or a continuous panoramic scene. These multiple image systems usually comprise several rectangular screens arranged in a horizontal row. Examples are Cinerama with three screens and Circlevision with nine screens that form a complete circle around the viewer. However, this horizontal row of screens typically leaves the area above the viewer blank, and lens distortion and mechanical wobble at the edges of each rectangular image too often make the boundaries between one screen and another clear.

To cover an entire hemispherical field of view, a compound camera must have several cameras, all facing outward. The weight, size and jittery image of typical motion picture cameras make them unsuitable for use in such a system. The compound motion picture camera described in US Patent US-A-3 698 803 clearly shows how cumbersome such a camera would be.

On the other hand, television cameras produce a steady image, and they have become small enough to be bundled due to the development of charge-coupled-device (CCD) technology and fiber optic power. Wide-angle recording and projection using horizontal rows of rectangular television images has been developed for military and surveillance purposes, such as U.S. Patent US-A-4,772,942, which covers a four-camera binocular display system for tanks. This system, however, depends on well-designed pre-distortion of the image for the display system, because the cameras record parts of a curved field of view from the inside of the outward-facing curve, but the images are shown from the outside of the curve looking inward.

The projection of the camera image from the center of a sphere outwards is required, which corresponds to the orientation of the camera to the surrounding spherical field of view. The projection of television images onto wide screens with a brightness that is close to that of a cinema film projection was supported by the development of laser video projection systems, as used in US-A-4,796,961 and active phosphor screens responsive to polarized light beams as described in US-A-4,798,448. However, because a spherical surface cannot be completely divided into rectangles, the image must be reshaped or cropped for projection to produce a mosaic extending over one hemisphere. One such method of cropping is described in US-A-3,687,530 in which five projectors project triangular sections all the way around the periphery of a dome. However, the use of a triangle as the standard shape means that less than 41% of the available lens image could be used.

In Patent Abstracts of Japan, Vol. 8, No. 29 (E-226) (1466), February 7, 1984, Kurokawa discloses a method of constructing a composite projected image in a dome of pentagons and hexagons in a manner similar to that found in the casing of a soccer ball. This requires a very large number of projectors, lacks the ease of using a standard shape, and the disclosed method of producing these shapes by cutting off the four corners of a square screen cannot, in reality, produce either a true hexagonal or pentagonal shape.

The optimal cropping would produce a small number of images of the same shape, which could then be used to cover an entire spherical surface. Normal or Platonic polyhedra are the only geometric shapes that have this feature of equal facets. There are only five: the tetrahedron, the hexahedron or cube, the octahedron, the icosahedron, and the dodecahedron, a twelve-sided shape made of pentagons. An environmental photograph based on normal polyhedra is described in US Patent US-A-4 214 821, in which individual still photographs are exposed using a polyhedral camera mount and the copies are joined together to form a polyhedral solid. However, this only concerns the assembling of copies produced by still photography and seen from the outside as flat facets, and does not refer to the simultaneous recording of a moving moving image and its projection and the view of the interior of a spherical screen.

Existing pickup lenses produce an image with only one optical center. This does not allow more than one image for a composite camera or projection system to be extracted from the image produced by a single lens.

Existing standalone projectors for film or video rely on a static image format or "hard mask" to crop the camera or projector image. This does not allow for flexibility in the shape of the projected image or a composite projection system in which full or partial images add up to a variety of overall image shapes and formats.

It is therefore clear that the state of the art has several disadvantages, including:

(a) Single camera systems are limited to the resolution of that single camera.

(b) A rectangular image format for the recording is unsuitable for the uniform subdivision of a spherical surface and wastes the available lens image.

(c) Current cameras are not designed to fit together in a modular manner so that the edges of their respective images are automatically aligned.

(d) Compound motion picture cameras are bulky and heavy, have a limited recording time, and must be self-contained because light from the lens must strike the film recording medium directly. The mechanical movement also causes the image to shake, and the film recording medium itself cannot be expanded or changed in shape.

(e) Current composite image projectors do not use a single standard shape resembling a circle for constructing a composite dome screen image.

(f) Hard physical masks used as light barriers in existing film cameras mean that the camera or projector image shape cannot be controlled Fluidity can be changed.

(g) Due to limitations of spherical lens manufacturing, existing lenses have only one optical axis.

A combination of these disadvantages prevents the development of a practical method and device for recording and projecting a complete environment onto the interior of a spherical movie theater.

It is therefore the object of the present invention to describe a method and device for an environment imaging system comprising a modular camera and a matching projection system based on the twelve-sided dodecahedron for the recording and projection of high-resolution pentagonal images which appear to continue continuously over the interior of a surface covering a hemisphere or more, thereby obtaining and creating the impression of a total environment in a dome or a spherical movie theater.

Another object of the present invention is to describe the essential geometric alignment of these modular elements of both the camera and the projector according to the dodecahedron which make this possible.

Another object of the invention is to describe a standard pentagonal image shape which represents a more efficient use of the image formed by a lens than ordinary rectangular images and can be duplicated to extend completely over a hemisphere or even an entire sphere, with the resolution being equal both above and below the horizon.

A further object of the invention is to describe the elements of a video system in relation to existing hardware.

Another object of the invention is to describe a masking processor for further controlling the shape of individual facets sufficient to produce an apparent overall image of infinite size and shape.

A further object of the invention is to provide a modular To describe a camera system of interchangeable parts in which equipment sufficient to record an entire hemispherical field of view or more can be carried by a single cameraman.

Another object of the invention is to describe an electro-optical lens comprising a method of curving the video image produced by a camera module lens in such a way as to create the effect of a lens with two or three optical centers, thereby producing two or three pentagonal images from a camera module.

Another object of the invention is to describe an alternative virtual viewing system that takes into account a movable window of interest within a composite pentagonal image of a hemisphere.

Devices and methods according to the invention are specified in claims 1 to 39.

The present invention has many advantages over the prior art, including:

(a) A composite camera enables higher resolution and overall performance than a single-camera system.

(b) The use of a pentagon as a standard shape enables a spherical surface to be evenly and accurately divided and results in more efficient use of the available lens image than the existing rectangular formats, producing an average image quality per lens that is 42% higher than the usual cinema image format of 1:1.85.

(c) Using a standard alignment topology according to the dodecahedron creates a camera system of modular, interchangeable parts and a projection system that is modular and also interchangeable, both producing pentagonal images that always match along every edge.

(d) A composite video camera system is lightweight and portable compared to a composite motion picture camera because the composite video camera is its recording medium. The use of video allows for longer recording times, a more stable image than film recording and projection, and the ability to manipulate the image because the electrical signals representing that image are accessible before it reaches the recording medium and when it is played back from the recording medium and enters the video projector.

(e) The electronic control possible with an electronic image such as a video means that the aspect ratio of the projected image can be controlled fluidly by the application of control voltages or algorithms, resulting in the production of a total projected image of infinite shape.

(f) It is also possible to manipulate the signal representing the camera image to produce an image showing the effect of a lens with more than one optical center.

(g) Digitally stored composite multiple camera images can be assembled and viewed through a movable relevant window in a virtual image store.

The combination of these and other factors makes the present invention a practical method for wide-angle video capture and projection onto a spherical surface, up to and including the recording and display of apparently continuous environments that encompass a full spherical view.

Further objects, advantages and novel features of the present invention will become apparent from the following detailed description taken in conjunction with the accompanying drawings.

Fig. 1 is a perspective view of a dodecahedral solid with one of its sections removed, the boundaries between adjacent sections shown, and lines going from a central point to the corners of the outer pentagonal face of the missing section.

Fig. 2 is a perspective view of the dodecahedral arrangement of the optical axes in the lens holder of both the projection and camera systems and the common pentagonal image of both.

Fig. 3 is a schematic diagram showing a cross-section of the field of view covered by such a composite camera and projector system.

Fig. 4 is a perspective view showing a spherical surface divided into pentagonal sections corresponding to the modules of such a camera and projection system.

Fig. 5 is a schematic diagram showing a cross-section through a module of the camera system.

Fig. 6 is a schematic diagram showing a front view of a module of the camera system.

Fig. 7 is a block diagram of a module of the camera system.

Fig. 8 is an illustration of the typical pentagonal image shape and orientation for above-the-horizon imaging.

Fig. 9 is a block diagram of a module in the projection system.

Fig. 10 is an illustration of the types of image cropping performed by the masking processor.

Fig. 11 is a diagram comparing a pentagonal image with other film formats in terms of the optimal round lens image.

Fig. 12 is an illustration of the extraction of a pentagonal image from the original recording in various film and video formats, and the relationship of their dimensions to each other.

Fig. 13 is a schematic representation from inside a half dodecahedron showing the relationship between projected video scan lines and their pentagonal sections.

Fig. 14 is an illustration of an alternative method of dividing the film or video image into two pentagons made possible by an electro-optical lens.

Fig. 15 is a schematic representation of a barrel distortion pattern in an optical lens.

Fig. 16 is a schematic representation of dual counterdistortion of video raster scanning in an electro-optical lens.

Fig. 17 is a schematic representation of the final distortion pattern produced by the electro-optical lens, with the effect of a lens with more than one optical center, here equal to the centers of the two pentagons of Fig. 14.

Figure 18 shows an outline of the present invention, an illustration of capturing a portion of a spherical view in HDTV format by a modular camera with a dodecahedral lens mount, video distortion with an electro-optical lens producing more than one optical axis, and corresponding symmetrical pentagonal surfaces of optical congruence, including cropping to the edges of these surfaces by a masking processor, including the function of further cropping to the image as needed, then projecting the resulting video image onto a corresponding sphere from a modular laser HDTV video projector with lenses in an equivalent dodecahedral mount, thereby recreating the portion of a spherical field of view.

Fig. 19 is a perspective view of a portable composite video camera with contour grip that can cover almost a spherical field of view.

Fig. 20 is a perspective view of a hand holding this portable composite video camera.

Fig. 21 is a perspective view of a wearable composite video camera helmet capable of covering a hemispherical field of view.

Fig. 22 is an illustration of a camera or projector with three lenses aligned with the edges of the dodecahedron, where each lens can cover two screens, thus covering an entire hemispherical field of view with single resolution.

Fig. 23 is a top view of this edge-oriented camera or projector showing the arrangement of dual screens to cover a hemispherical field of view.

Fig. 24 is an illustration of the removal of three screens from the optimal lens image by an electro-optical lens.

Fig. 25 is a top view illustration of a dodecahedral vertex-based compound camera or projector for a hemispherical view.

Fig. 26 is a perspective view of a vertex-based camera or projector system showing the three screens covered by a camera or projector facet.

Fig. 27 is a circuit diagram of an arrangement of pentagons for planning footage and images for hemispherical recording and projection.

Fig. 28 is a schematic of an alternative arrangement of pentagons for planning imagery showing a geodesic grid .

Fig. 29 is a circuit diagram showing all screens adjacent by slight polar distortion and also an overlapping arrangement of dual screens.

Fig. 30 is a circuit diagram of a virtual image of six screens formed into a disk and used in a rotating electronic viewfinder system capable of covering a hemispherical view.

Fig. 31 is a schematic diagram showing a cross-section of a motion picture theater projection system using laser video projection.

Fig. 32 is a diagram showing a cross section of a composite projector using liquid crystal display panels, a motion picture theater light source, and two types of cooling systems.

Fig. 33 is a quarter section showing the characteristic angles and proportions of the dodecahedron.

Parts listed in the drawing:

41. displayed module

42. adjacent module

43. Line from center to corner #1

44. Line from center to corner #2

45. Line from center to corner #3

46. Line from center to corner #4

47. Line from center to corner #5

48. Centre of dodecahedron and system

49. Pentagonal surface of the dodecahedron (flat)

50. Dodecahedral lens mount

51. Disassembled module shown

52. Pentagonal image of the module projected onto the sphere

53. Lens axis for this pentagonal image

54. Vertical axis for upper module

55. Dodecahedron

56. Sphere

57. Cross-sectional angle of view of the screen (69.0948490)

57a. Angle between the optical axes directed towards the center (63.4349460)

58. hemispherical field of view (1800)

59. Standard optical axis for over-the-horizon imaging

60. Horizon line

61. Standard elevation above horizon line (26.5650540)

62. Standard axis for below-horizon shooting

63. Negative elevation for below horizon line (26.5650540)

64. Angle of lens field of view (74.7547520)

65. Circular lens image around the pentagon

66. Hard-cut pentagon image border

67. Pentagon image form

68. Part of the pentagon

69. Area outside the pentagon

70. Part to be shown, added to another screen

71. Added Screen (Full Screen)

72. Adjacent Part #2

73. Adjacent Part #3

74. Outer bezel ring

75. Inner surround ring

76. outer cone

77. Inner cone

78. Lens holder for single lenses

79. Automatic focus sensor

80. Automatic exposure sensor

81. Cable that carries focus and exposure indicators to the outside

82. Cable that routes focus and exposure control inside

83. Channel in the ring point area for conductive cables

84. Focus control motor

85. Exposure control motor

86. Charge Coupled Device (CCD) in camera

87. CCD control in camera

88. Control cable to synchronization source

89. Stronkabel

90. Amplifier and electro-optical converter

91. Fiber optic external camera cable

92. Closure end of the module

93. outer surface connection #1

94. outer surface connection #2

95. External surface of the module

96. Claw plug

97. Claw recording

98. Screw hole for removal

99. Lens front

100. Upper edge on module (ring)

101. Lower point on module (ring)

102. Adjusting screw #1 for ring

103. Adjusting screw #2 for ring

104. Camera module boundary

105. Lens symbol for camera

106. Exposure and focus control sensors

107. Exposure and focus evaluation processor

108. Exposure and focus control line

109. Motor-driven focus and exposure control

110. External synchronization source

111. Synchronization connection to other modules

112. Recorder

113. Image storage

114. Upper pentagonal edge (for shooting above the horizon)

115. Lower pentagon point (for shooting above the horizon)

116. Horizon line on the screen

117. Limit of the projection module

118. Video source, either camera or recorder

119. Masking processor

120. Mask Source

121. composite video signal

122. red light valve

123. green light valve

124. blue light valve

125. red light beam generator

126. green light beam generator

127. blue light beam generator

128. red fiber optic cable

129. green fiber optic cable

130. blue fiber optic cable

131. flat mirror

132. rotating mirror wheel

133. Projection lens

134. Projection screen area

135. Limit of the original raster scan

136. Innermost line of cutting for soft edge

137. Outermost line of cutting for soft edge

138. Optimal round lens image

139. Image format 1:1.33

140. Image format 1:1.85

141. Optical center of the lens and the original image

142. Image format 1:1.37

143. Image format 1:1.77 (9:16)

144. Video scan line of the displayed module

145. Video scanning line of the adjacent module

146. Zenith Umbrella Z

147. Upper adjacent screen at 12:00 A

148. Next screen clockwise at 2:24 B

149. Next screen clockwise at 4:48 C

150. Next screen clockwise at 7:12 D

151. Next screen clockwise at 9:36 E

152. Alignment arrow indicator

153. Left dual pentagon

154. Right dual pentagon

155. HDTV video limit

156. Lateral pentagonal tip

157. Grid line image

158. Tonnage

159. Raster scanning counterdistortion in an electro-optical lens

160. Final sum grating line image from the electro-optical lens

161. Left dual optical center

162. Right dual optical center

163. Distortion pattern of the left dual pentagon

164. Distortion pattern of the right dual pentagon

165. Video scan line example

166. Left dual optical axis

167. Right dual optical axis

168. HDTV image on the sphere

169. Spherical field of view

170. Dodecahedral camera or projector

171. Assembled camera from eleven modules

172. Contour grip

173. Exposure and focus point display

174. Trigger

175. Tripod thread

176. Cable bundle from the camera

177. Hand holding a camera on a contour grip

178. assembled camera from six modules

179. Holder in the form of a helmet

180. Ads in the cap peak

181. Cable from the outside surface

182. Cable from point channel

183. Cables coming out of the camera

184. Edge-based camera for single coverage

185. Electro-optical lens

186. Page edge-aligned axis

187. Left side of the dual screen field of view

188. Right side of the dual screen field of view

189. Top view of the camera

190. Upper edge target for alignment

191. Left margin for alignment

192. Right margin for alignment

193. Zenith screen of triple screen extraction

194. Left screen

195. Right umbrella

196. Concentric area indicated by other modules

197. Example of the edge of the displayed area

198. Area of reduced brightness

199. Vertex-based composite camera

200. Vertex-based camera module or facet

201. Electro-optical lens for vertex camera module

202. Perspective view of a vertex camera

203. Alignment axis to dodecahedron vertex

204. Axis to the Zenith

205. Flat umbrella outline

206. Horizon line on screen

207. Geodesic grid on screen

208. Circuit diagram of the interconnected shields

209. Individual outline of the outer screen with polar distortion

210. Virtual image of twelve screens as a disk

211. Compressed outline of the outer screen

212. Viewfinder window

213. Axes of movement

214. Group of three lasers

215. Group of three light valves

216. Fiber optic laser cable

217. Connection to the module on the projection head

218. White laser beam

219. Synchronizing distribution amplifier

220. Deflection plate between projection modules

221. Cinema projection bulb

222. Liquid crystal display transparency

223. Glass-coated water cooling barrier

224. Gas-subcooled cooling tubes

225. Radius of the pentagonal surface (1,000)

226. Height of the pentagonal surface (0.809017)

227. Half of a dodecahedral edge (0.587785)

228. Radius of the outer sphere (1.6472779)

229. Radius of the central sphere (1.5388416)

230. Radius of the inner sphere (1.3090167)

231. Half of the dihedral angle between the surfaces

232. Interior angle between edge center and pentagonal point

233. Interior angle between pentagonal point and center

234. Interior angle between pentagonal center and base

Optimum brightness and resolution for an image that spans the interior of a dome theater arises naturally when more than one camera and projector are used. The challenge is to find a simple standard format for each part of this multi-part image, one that ensures that the mosaic will fit together to produce a continuous image on the spherical screen with maximum resolution and minimum distortion.

The area of best image reproduction through a lens is typically a circle with the optical axis at the center, because power loss is typically proportional to the distance from the optical axis. The radius of this circle defines the radius of best image quality The optimal format would therefore be close to this round shape and could divide a hemisphere or a sphere into a reasonable number of equal sections.

The typical rectangular format wastes a lot of round area of the optimal lens image. A rectangular image with an aspect ratio of 1:1.33 can only encompass 61% of the circle. In addition, a circle cannot be evenly divided into rectangles.

The rectangular polyhedra or Platonic solids have the advantage of dividing a sphere into equal sections. They use triangular, square and pentagonal facets. Of them, a triangle can only cover 41% of the optimal circle. A square, the base of a cube, can cover almost 64%, but a hemisphere does not lend itself to equal division into squares. The pentagon, the base of the twelve-sided dodecahedron, covers the most of all, namely 75.71% of the optimal circle, and only six pentagons are required to cover half of a sphere. Therefore, a pentagonal image arranged according to the dodecahedron is clearly the best choice.

The present invention could be used for still image projection such as for slides, or for recording and projection with small format motion picture film such as 8 or 16 mm. However, video has many advantages. Video cameras are becoming smaller and video resolution is becoming increasingly better with the advent of high definition television (HDTV). The present invention will be described in terms of a portable composite video camera system and a compatible laser video motion picture theater projection system comprising means for recording and displaying an apparently continuous image over the area of a hemisphere or more. Continuing to utilize the advantages offered by video, a masking processor for producing complete screens of variable shape and size and an electro-optical lens that can be used to double either the coverage or resolution.

Fig. 1 shows a perspective view of a dodecahedral solid with one of its sections removed, with an indicated module at 41 and an adjacent module at 42, both representing modules of the composite camera system. At 43, 44, 45, 46 and 47 are lines going from a central point 48 to corners of the outer pentagonal face 49 of the missing section and representing the projection of an image from a central projection system onto a pentagonal screen. This defines the basic dodecahedral topology of the modules of both the camera system and the projection system of the present invention.

Fig. 2 shows the dodecahedral lens mount characteristics of both the compound camera and the compound projector at 50, where 51 is a single module disassembled in an exploded view. The pentagonal field of view produced by each module when projected onto a sphere is shown at 52, where 53 is the lens axis of the lens in that module.

Fig. 3 is a schematic showing the entire spherical field of view of this system and is shown as a cross section taken through the plane ABCD of Fig. 1. Note the asymmetry of cross sections of the dodecahedron. The optical axis 54 of the top module points straight up in the embodiment described here. Its advantage is that with only five hexagons the camera system can capture a complete hemispherical field of view down to the horizon in all directions and the projection system can reproduce this. For this reason this six module arrangement is described as above-the-horizon capture.

However, it is understood that other orientations of the portable camera are possible. The projector orientation could also be different, such as in a dome that is tilted to allow conventional cinema seating for spectators, as in the case of the OMNIMAX cinema.

The boundaries of the pentagonal screens of the projection system are defined by a dodecahedron 55 projected onto a screen 56. The angle of view of a single screen 52, from top to bottom, at 57 is 69 5' 41.4564" (69.094849). At 58, the entire field of view, which is 180, is covered by a total of six modules, three of which are shown in this cross section. At 59, a standard optical axis for taking and projecting images is located above the horizon line 60, except for the module which points directly upward. This optical axis has a height 61 above the horizon line of 26 33' 54.1944" (26.565054). For images taken below the horizon, the optical axes 62 have a negative elevation 63 of -26º 33' 54.1944" (-26.565054).

Fig. 4 is a perspective view showing a sphere divided into pentagons projected inward according to a dodecahedron. This shows the boundaries between adjacent parts of the field of view in the camera and projection system. The optical axis of the top module 54 points straight up in this embodiment.

In the camera system, the width of the field of view for the lens for a pentagonal image is shown at 64, which defines a circle 65 that touches the points of a projected pentagon 52. The optical axis AO for a single camera module is magnified by an angle 61 equal to 26 33' 54.1944" (26.565054).

In the video projection system, a video masking processor for each projection module crops the projected camera image by providing a video switcher-type cancellation signal to a pentagonal boundary 66 defining a central pentagonal shape 67 as shown in Fig. 11 and Fig. 12 when all projection modules are active. In addition to this pentagonal cropping, the masking processor may also be used to reshape the screen in accordance with additional cancellation signals from a video mask source. as also shown in Fig. 10 and Fig. 18. This may show one or more parts of a pentagon 68, or parts of the camera image 69 outside a pentagon if the original image includes it and the adjacent module 43 is inactive. This may also show a part of a pentagon 70 added to another pentagon 71, or several adjacent parts 70, 72 and 73. As long as adjacent screens respect their pentagonal, dodecahedral boundaries, the result will be an apparently continuous overall image on a screen of infinite shape or size.

Fig. 5 shows a cross-section of a module of the camera system. For clarity, the mounts for the lens elements and the internal electronic connectors have been omitted. This cross-section shows an idealized view with all components aligned along the cross-sectional plane. In practice, these parts would be distributed around the perimeter of each module.

The basic construction of the module is a cone with two pentagonal bezel rings attached. The outer bezel ring is at 74 and the inner bezel ring is at 75. The cone has an outer section 76 and an inner section 77 which also includes the lens mount 78, possibly a standard Canon B-4 or Fujinon B mount. The optical axis of the lens and module is shown at 69.

Sensors for automatic focus and exposure are shown at 79 and 80. A cable for relaying readings from the sensors for exposure and focus is at 81, and a cable for relaying control signals back to the motors to actually make the adjustments is at 82. These cables are shown connected to the side of the inner cone using a channel 83 in the inner bezel ring, made possible by the greater distance from the cone walls at the pentagonal points. The motor for controlling the focus on the lens by means of dual gears is shown at 84 and that for controlling the exposure at 85.

The focal length of the lens is determined according to the formula

F= DxA/0

where O is the size of the object in front of the camera, D is the distance from the object being photographed to the lens, A is the size of the image formed by the lens, and F is the focal length of the lens.

A charge coupled device (CCD) on which the image is formed is shown at 86. In the preferred embodiment of the invention, in which a 16 mm (0.63 inch) diameter circular image is formed on the CCD, having the same height as standard 35 mm film and approximately the same diagonal length as standard 16 mm film, this lens would have a focal length of 10.479 mm and an angle of view of at least 74.754752, shown at 64, which is usually described as a moderate wide angle lens. The lens shown here is based on the Leitz Elmarit -M f.28 lens with a field of view of 760.

A shorter focal length has the advantage of a much greater depth of field, so that the image is held in focus over a much greater range of distances. The near-far point of a 10.5mm lens is about 1 foot (30.48 cm) at f16, making everything beyond that distance to infinity within an acceptable 0.001" circle of focus, which is common for 35mm film shooting. The extreme depth of field means that the automatic focusing system described above could be set up as desired.

The electronic control switch for the CCD is located at 87; it is connected to an external synchronization source via control cable 88. This external synchronization source is used to connect all modules of the camera system together. Power for the module motors and electronics is provided via power cable 89.

In the preferred embodiment of the present invention, a high definition television signal is generated by the CCD. An array of three CCDS with a retrofocus lens and a dichroic prism can be used if required. used to extract parts of a color image or to increase resolution. An example of a small, high quality triple CCD camera is the Sony DXC-750. According to NHK Labs, an HCTV CCD would be a 25.4mm (1 inch) imaging system with 80 dB dynamic range and 25 MHz horizontal resolution. Since there are currently many competing proposals for HDTV standards, for the purposes of this discussion we will use equipment for the Sony HDTV standard known as HDVS or MUSE. It is to be understood, however, that the invention could easily be adapted to other television standards, including PAL and NTSC video.

An amplifier and electro-optical converter for converting the video information from the CCD into light pulses is located at 90. This converter performs the same function as the Sony HDFT-100 HD Optical Fiber Transmitter. The light pulses thus generated go into the fiber optic cable 91. This cable could be a camera cable for a Sony HDFC Optical Fiber Camera of suitable length. For a thinner cable capable of transmitting over long distances, this could be a single fiber optic cable, based on the techniques described in the paper "Transmission of HDTV and Audio Signals Over One Single-Mode Fiber" in the SMPTE Journal, Sep. 1989, pages 651-657.

In connection with the cables to and from the module, they are shown here coming out of the inner closure end 92. This has the advantage that the cables are automatically bundled into a central strand which exits downwards from the socket of the assembled camera. However, this depends on the cables being small enough and having enough flexibility to fit into this inner socket space. A practical alternative is to have these cables exit from one or more of the connectors 93 and 94 onto the outer surface 95 of the module at locations where they would still be invisible to the lens. The inner lens claw 75 could then be solid, extending into the center 48, allowing faster and more precise alignment of adjacent modules from more contact surfaces between them.

Modules are connected together at their points. This is accomplished by means of a claw connector 96 in conjunction with a claw receiver 97. Each claw connector has three tapered claws, each of which has a dovetail mandrel in a homologous stop, like leveling planks in a lashing. The release or coupling is accomplished by means of a screw hole 98 for a removal tool.

The preferred embodiment for the original image in each camera module is a rectangular image because it provides extra area outside the pentagon available for display selection and because rectangular CCDs and raster scanning are already common. However, a pentagonal CCD in the camera would require less video bandwidth in its recorder and transmitter.

Fig. 6 shows a front view of the camera module. The front element of the lens is located at 99 and the outer bezel ring is located at 74. This claw has a claw receiving slot 97 at each corner. They are used for a claw connector 96 composed of three dovetail terminals aligned in an equilateral triangle with a threaded hole in the center 98 for a removal tool. This terminal is slid between the points of adjacent modules to lock them together in the appropriate alignment.

Sensors for autofocus and exposure are shown at 79 and 80, and optional front panel connectors for cables are shown at 93 and 94.

For above-horizon shooting, except for the zenith-facing module, the horizontal edge 100 on both the inner and outer bezel rings is always at the top. For below-horizon shooting, the inner and outer jaws must be rotated 180º to the inner cone so that point 101 is at the top. Two set screws in a horizontal line, shown at 102 and 103 for the outer jaw, are used to lock the jaws in place for any orientation. The space that would be occupied by a camera module pointing straight down is reserved for cables to and from the composite camera socket. Up to 11 modules can be connected together in this way to cover almost a complete spherical view. With the camera cables attached to the outside of the modules as described above, up to 12 modules can be connected together to cover a complete spherical subjective.

Fig. 7 shows a block diagram of the camera module, showing the parts both inside and outside of the boundary 104 of the module itself. The optical axis of the module is at 53, and the angle of view of 74.7548 is shown at 64. For an electro-optical lens producing two or three screens as described below, this angle would be as high as 150. The lens is located at 105. A sensor for exposure and focus control is shown at 106, with its readings going to a central processing unit 107. The processor evaluates the exposure and focus readings for the various modules, using weighting to a module or zone if desired, and then adjusts the overall exposure and focus for all modules along line 108, which is accomplished by a motorized exposure and focus control 109. Using the same setting for all modules prevents obvious discontinuities between different screens due to either exposure or focus differences. However, with digital control of the pixels on the CCD, one can also have an exposure setting that varies across the area of a single screen. Toshiba and Hitachi are among the companies currently researching digital irises.

An external synchronization source 110, which is also connected to the other modules along line 111, controls a CCD controller 87 which scans the individual pixel light displays from the image formed on the surface of the CCD 86. This scanning function can be distorted by the electro-optical lens described below, and portions of the image information can be erased by the masking processor to be described. The electrical pulses, generated by this scanning of the pixels constituting a video image go to an amplifier and electro-optical converter 89 which converts the electrical pulses into light which is conducted through a fiber optic cable 91 directly to the projection system into a quiescent current device or into a recorder 112. When the distortion of the electro-optical lens and the cropping of the video image by the masking processor are not carried out during the scanning of the pixels on the CCD, they are effected in an image store 113 midway between the camera and the projection system either before or after a recorder.

The light pulses from each camera module could be converted back to electricity by a device such as an HDFR-100 Optical Fiber Receiver and passed through a processor such as an HDT-1000 HD TBC/Signal Processor to achieve maximum clarity before reaching the recorder. For optimal bandwidth with minimal average recording size, this recorder would be a HDTV (High Definition Television) digital video disk recorder. However, a Sony HDV-1000 Video Tape Recorder could also be used.

Fig. 8 shows a representation of the typical pentagonal image 67 as projected by the projection system. The orientation of the pentagonal image shown with edge 114 at the top and point 115 at the bottom is characteristic of each module, except the zenith in a six module camera system, which can capture a complete hemispherical view in any direction down to the horizon. For this reason, the orientation of the camera module and its pentagonal image is described as an above-the-horizon orientation. The level of the horizon is shown on the screen at 116.

Fig. 9 shows a block diagram of a module of the projection system, showing the parts both inside and outside the boundary 117 of the module itself. For the best optical results, the size of the projection module should match the size of the camera module as closely as possible. The source of the video image, either the camera module in a closed circuit system or a video recorder, is located at 118.

The masking processor 119 crops this video signal in the manner of a video switcher according to the templates provided by a video mask source 120. These templates include the main pentagonal mask and additional masks to show parts of the image together with other images so that overall images of variable shape and size can be created.

The masking processor then outputs the video image to be displayed in the form of a red, green and blue composite video signal 121 to its respective display generator in the projection system. In this display generator, the signals representing the red, green and blue portions of the video image are used to control three light valves shown at 122, 123 and 124, respectively, such as variable density filters, variable electrical resistors or variable reflection mirrors. These light valves vary the light intensity of three light beam generators for red, green and blue, shown at 125, 126 and 127, respectively.

In the preferred embodiment for projection and the system described here, the light generators for the projection system are large image ion lasers, one krypton laser for red and another argon laser for blue and green, producing a white beam. One example is the large-scale laser video projection done by Laser Creations of London, England, who recently demonstrated a large-scale Sony HCTV projection in Paris. Their Laser Video Projector (LVP) can even combine a raster generated video image with a vector generated laser effect. They can produce 9 watts of peak white light, equivalent to 8800 perceived lumens, with large image lasers for a maximum diagonal image size, with zero room light and with a 120' unit gain screen with infinite depth of field. They use Spectra Physics 2040 model large image lasers in an improved version of the Cavendish system, a 5 watt krypton laser for red and a 15-20 watt argon laser for blue and green, with an additional control for true colors, the by small-format krypton and argon lasers pumping tunable color lasers. The advantage of a laser projection system is that the laser light is the purest and most controllable color light available. It would theoretically produce the most intense image. The disadvantage is that large image lasers are currently expensive (about $60,000 each), require an extensive water cooling system (54 liters/min. at 480 kPa (70 psi) for each large-format laser), and require a large amount of power (three-phase, 208 volts, 80 KW per laser). Another disadvantage is that present laser projectors are not as bright as video projectors using bright movie theater light sources. There is also a tendency for a "sparkle" effect in the image caused by the laser creating the smaller irregularities in the projection screen. As reported, Visulux, San Jose, California, has an improved projector that does not have this problem.

For the sake of clarity, the beams produced by these lasers are separated here, being first guided to the projection head containing the lenses by the red, green and blue fiber optic cables shown at 128, 129 and 130 respectively. In this way, the size of the projection head can be kept quite small and approach the size of the camera head.

In the projection head, the beams are reflected by a mirror 131 which controls the vertical scan and a rotating octagonal mirror wheel 132 which controls the horizontal scan, producing a raster scan. These mirrors are controlled by an external synchronization generator 110 which is connected to all modules of the system along line 111. The reflected laser beams then pass through a projector lens 133 and strike a spherical image surface 134 which produces a video image. In combination with the images formed by the other projector modules, an apparently continuous image is produced over the interior of the hemisphere or more.

In the simplest and most straightforward embodiment of the present invention, there is a one-to-one correspondence between camera and projection modules, with both having an optical axis 53 identically aligned perpendicular to the center of a dodecahedral face and a matching field of view 57. However, when electro-optical lenses are involved, each camera module can cover more than one pentagonal section. These different sections can be assigned to their respective projection modules for individual coverage, or, if the projection modules can also cover more than one section, pentagonal images can be made to overlap for increased resolution.

Figure 10 shows the types of image cropping that is performed by the masking processor before a video image is projected. This cropping may be performed in the camera module system prior to image capture, but in the preferred embodiment described here it is described as part of the projection system.

At 135 is the boundary of the raster scan produced by the camera module. If the clipping described here is not done as a function of the CCD capture control, then the information representing this raster scan goes into an image memory 113 where it is available both to the masking processor described here and to the distortion of an electro-optical lens described below.

The primary task of the masking processor is to enforce a pentagonal dodecahedral boundary 66 between adjacent screens. It cuts out the image at a boundary defined by its central pentagonal part 67 if the projection module adjacent to that boundary 42 is active and the projection module is showing an image extending to that boundary. Thus, if all projection modules are active and showing their largest screens, all of those screens would be pentagons.

The masking processor performs all its cutting functions by applying erasure signals to the video raster in the manner of a video switcher. Cutting with a soft edge is preferred, with a linear progression from an innermost line 136 to an outermost line 137 where the IRE level is maintained at 0 for all erased areas, unlike the ordinary projected "video black" of IRE 7. This serves to prevent double exposures and to erase the boundaries of the original image, leaving an apparently continuous composite projected image.

In addition to defining the pentagonal boundary, this masking processor can also be used to create a screen of infinite shape by passing through different regions of the original image, according to additional cancellation signals generated by templates provided by a video mask source 120. Portions smaller than the pentagon 68, as well as portions 70 to be shown as distinct from images from other modules, can be displayed, thereby producing non-pentagonal overall projected images. Portions from outside the pentagon of a single module 69 can be shown extending into an adjacent area, so long as the original camera image included that portion of the image and the projection module for that adjacent area is inactive. As long as adjacent images remain divided by their pentagonal boundaries, the net effect will be to produce apparently continuous screens of any size or shape.

Fig. 11 shows an illustration of a division of the optimal lens image 138 into the pentagon 67, compared to the rectangular division into the usual television picture format 139 of 1:1.33 and the usual film picture formation 140 of 1:1.85. A pentagon comprises 75.71% of the round lens image, 23.9% more than can be achieved with a picture format of 1:1.33 and 42% more than a picture format of 1:1.85. Therefore, using the same lens, a pentagonal image produces an improved average resolution of up to 42%.

Fig. 12 shows the extraction of a pentagonal image from the rectangular raster scanned image of a camera module in one of the common film or television formats showing the relationship of dimensions between this projected pentagon and the various formats. The center of the pentagon matches the optical center 141 of the original rectangular image. The edge of the pentagon is at 66 and the point 115 rests on the floor edge. Any image information outside the boundary of the pentagonal image at 66 is not shown by the projection module unless adjacent projection modules are inactive. However, if this external image is recorded by the camera, it can be available as a reference when shown by another type of projector or playback screen.

The widths of several popular formats are shown in relation to the pentagonal image. At 139 is the limit of an aspect ratio of 1:1.33 used for NTSC video, while 142 shows the edge of the aspect ratio of 1:1.37 for nonanamorphic 35mm feature film. At 143 is shown the limit of an aspect ratio of 1:1.77 or 9:16 used in most common standards of HDTV video, including the Sony MUSE standard. At 140 is illustrated the limit of an aspect ratio of 1:1.85, common for most 35mm feature film. Other film formats, such as Super Panavision and 70mm, may be 1:2.2 or even wider. With an aspect ratio of 1:1.33, a pentagonal image covers 44.7% of the image area with a border at 66.

When more than six camera modules are grouped together, these additional modules can be described as taking pictures below the horizon. For the below-the-horizon modules, the pentagon is inverted, its point 115 is at the top instead of the bottom. This can be done in the projector's masking processor.

Fig. 13 shows the alignment of the video scan lines 144 and 145 for the adjacent screens of a hemispherical projection that is above the horizon. The view here is straight up from the projector, with the screens shown as flat facets. If projected onto a curved surface, the lines from the peripheral screens would also be curved, and be apparently continuous. The screens may be named starting from the zenith screen at 146 as Z. The screen adjacent its upper edge at 147 is A' then the others clockwise are designated at 148 as B, at 149 as C, at 150 as D and at 151 as E. The orientation of the image formed by each camera module is shown by an arrow 152 within each screen and points toward the top of the image. Note that the "tops" of the Z and A images are opposite each other. The video scan lines within each screen go from top to bottom. For a 1000-line HDTV video projector this would mean image information from line 1 to line 904. If the optical axes of both the camera and projector are properly aligned with the faces of a dodecahedron then these scan lines should match from side to side and give an apparently continuous image. If you have a dome with a diameter of 30 m (97.5') and a radius of 15 m (45.75'), each screen would have a height of 15.96 m (48.678'), giving each scan line on the screen a vertical height of 1.765 cm (0.695").

The overall resolution of such a system can be estimated by comparing it to existing film formats. An HDTV 1000 line image is usually said to be roughly comparable in apparent resolution to 35mm motion picture film. Therefore, a six-camera, six-projector system covering one hemisphere would have a photographic resolution equal to 70mm film with 7 perforations with a projection aperture of 48.565 mm (1.912") x 30.963 mm (1.219"), which is better than 70mm motion picture theater systems that use only 5 perforations.

Fig. 14 is an illustration of an alternative method of dividing a film or video image into pentagons, made possible by an electro-optical lens, in which for each camera module except the zenith, two pentagons 153 and 154 can be taken out of the original image with an aspect ratio of 1:1.77 or wider. The format 155 shown is the same as the aspect ratio of HDTV; it should be noted that it is exactly wide enough for two pentagons to fit into it, with the pentagonal side tip 156 dividing the rectangular side edge.

The challenge of this method of extracting pentagonal images, however, is the task of optically matching the two sides. The associated field of view for the camera lens would be a very wide 140º. In this case, the recess for the lens in each module face should be made wide and shallow to match this angle. With such a wide field of view, distortion usually results that increases concentrically from the optical center. And if concentric distortion occurs, then a view to the right of the optical center from one camera module would not match a view to the left of the optical center from an adjacent camera module. The limitations of lens manufacturing mean that optics alone do not produce the radical transformation of distortion to sharp boundaries that would be needed. The solution is a new form of electronic support of the optical lens image using the distortion of the video raster scan, here called an electro-optical lens.

Fig. 15 is a schematic representation of a grating 157 showing a barrel distortion pattern 158 induced by an optical lens.

Fig. 16 is a schematic representation of an example of a special electronic counterdistortion of an electro-optical lens of the video raster scan 159.

The special distortion could be carried out by applying a sequence of analog drive voltages to a frame store which holds the video image information representing the optical lens image, using a read-versus-write system such as that found in the Quantel MIRAGE video effects processor, wherein a frame store is used to store the information pattern for a video image; this information pattern is distorted and the distorted pattern is then re-sampled and re-evaluated using of a normal center-pass raster scan. This particular distortion could also be accomplished directly in the camera module by applying this pattern of drive voltages to the CCD pixel scan controller 87, which creates distortion in the order of display of these image units. With this electro-optical lens, the right and left halves would thereby be made optically congruent.

Fig. 17 is a schematic representation of the final distortion pattern 160 produced by the electro-optical lens and created by the application of the above special electronic distortion to the optical image, producing the effect of a lens with two optical centers at 161 and 162 which are equal to the centers of the left and right dual pentagons of Fig. 14, and distortion patterns 163 and 164 which are congruent.

An advantage of using an electro-optical lens to produce dual pentagonal images from the original recording is that one camera module could produce two images, so the total number of camera modules needed for full coverage could be halved. The number of video recorders would also be halved, since each would record two images side by side.

Another advantage is that with the same number of camera modules as before, and with projection modules each capable of displaying a field of view as wide as that of their respective camera modules, pentagonal images could be made to overlap so that each pentagonal screen would have twice the resolution because of double coverage. The overlapping images would match because of the congruent optics created by the electro-optical lens. An arrangement of these overlapping images is shown in Fig. 29. This increased resolution would give this video system the equivalent of 70 mm film with 12.6 perforations, which is close to the resolution of the largest film format, 70 mm IMAX and OMNIMAX with 15 perforations.

The two overlapping images for each screen could also used for the 3D effects, such as different filtration or polarization between the camera modules, or for a 3D effect without glasses, where one version of the image is shown slightly smaller and blurred in relation to the other. The spatial separation is achieved by having each overlapping screen be recorded from a different camera module and thus a slightly different viewpoint. The number of projection modules would remain the same as the number of camera modules, because each projection module would project the two adjacent images produced by the electro-optical lens.

It should be noted that the video scan lines, the direction of which is indicated by the sample line at 165, are parallel to the top and bottom edges of the original image and not parallel to the top edge of the pentagon as shown in Fig. 12. This means that the video scan lines of two superimposed images would be inclined at an angle of 36º to each other. These offset scan lines could also be used to separate the superimposed images for 3D effects, as in the use of two screen levels, each of which responds to only one scan line angle.

Fig. 18 is a summary of the present invention, an illustration of HDTV dual pentagonal images from an electro-optic lens in a dodecahedral mount projected after additional cropping by a masking processor. The optical axis of the left dual pentagon is shown at 166 and for the right dual pentagon at 167. At 168 an image having an aspect ratio equal to that of HDTV is shown as projected onto a spherical field of view 169 and captured or projected by a dodecahedral camera or projector 170. Note that with an electro-optic lens the optical center of the original capture 141 is perpendicular to a dodecahedral edge, midway between the dual pentagons, rather than perpendicular to the center of the dodecahedral pentagonal surface 49.

Pentagonal boundaries set by the masking processor are shown at 66. Further masking is also shown for producing non-pentagonal composite screens, including displaying one or more portions of a pentagon at 68, displaying an area outside the pentagon at 69 for use when the adjacent screen area is inactive, and displaying a portion adjacent to a boundary 70 either alone or in conjunction with another pentagonal screen 71 to produce a non-pentagonal composite image.

Fig. 19 is a perspective view of a portable composite video camera with eleven modules and a contoured grip that covers a nearly spherical field of view.

If each module produces dual views due to the use of electro-optical lenses, then each module except the zenith is actually aligned with a dodecahedral edge rather than the center of a face. This is the equivalent of a simple polar rotation of this inner camera dodecahedron by 360 with respect to the outer screen/field of view dodecahedron. The elevation angle of the optical axes should also be increased from 26.565054 to 31.717473.

The cables leading out of the assembled camera modules 171 are contained in a contoured handle 172 for the camera operator and includes an exposure level indicator 173 and the shutter release 174. A camera tripod mount 175 may be attached to the bottom of this handle and the cables exit from the side at 176. The camera modules are connected together by means of claw plug connectors 96. A maximum of eleven modules can be grouped together in this way, covering 91.6% of the total spherical field of view.

With a focal length of 10.5 mm and a camera module structure as described above in Fig. 5 and Fig. 6, each camera module would have a depth of approximately 6 cm (2.36") and twelve camera modules together would have a sphere of 15.5 cm (6") diameter, not including the camera mount, which with such a dimension could easily be carried by a single cameraman. can be.

The distance between the lens axes in such a camera would also be approximately 6 cm, which is the average distance between the pupils of the human eye. This intraocular distance between the lenses means that stereo images produced by this camera would have a natural three-dimensional quality. If each lens is electro-optic and used to produce overlapping images, such a camera could therefore produce realistic stereoscopic images in almost all directions.

Fig. 20 is a perspective view of a hand 177 holding this portable composite video camera by its contour grip. The space that the module pointing straight down would reach is reserved for the camera mount and for the cables leading into and out of the composite camera. However, if the cables to and from the modules are connected to the fronts of the modules, up to twelve modules can be connected together, covering an entire spherical field of view. However, because of the need for a projector mount, an entire sphere cannot be shown. It may also be desirable for the cameraman to remain out of the picture so that a smaller number of modules could be used for a smaller overall picture.

Fig. 21 shows a perspective view of a portable composite video camera mounted on a helmet, enabling coverage of a hemispherical field of view. The usual number of six camera modules 178 are shown assembled to cover a complete hemisphere. A helmet 179 supports this arrangement and contains a space for the supporting electronics. This also has the advantage that the cameraman is underneath and thus out of the panoramic image, and the camera arrangement can be worn comfortably for long periods, with exposure, focus and elapsed time indicators located on the inside of the cap visor 180. The cables from the camera modules are on their outer surfaces as they exit from the connectors. 181 and from the channels to the inside at the pentagonal points 182 (assuming means for connecting the modules together at locations other than the corners). These cables, which carry the video signals from the camera 183, lead out to a pickup device, either separate or carried by the operator, or directly to the projection system.

Fig. 22 is an illustration of a camera 184 with three electro-optical lenses, each lens 185 having an axis 186 aligned with the center of a dodecahedron edge and each capable of covering two screens 187 and 188, covering a total hemispherical field of view of six screens at a single resolution. Only three recorders are necessary because each can record two screens. Note that the top-facing module is different in shape from the side-facing camera modules.

Fig. 23 is an illustration of a top view 189 of this edge-facing camera, showing its arrangement of dual screens to cover a hemispherical field of view and the top edge target for alignment 190 as well as the left edge target 191 and the right edge target 192. The elevation angle for the side edge targets is 31 43' 2.9028" (31.717473) and for the top edge target is 58 16' 57.0972" (58.282527).

Fig. 24 is an illustration of the removal of three screens from the optimal lens image by an electro-optical lens. The three screens are shown here with slight polar distortion to make their edges contiguous. Instead of having the optical center of the original image on a dodecahedral edge midway between the dual pentagonal screens as in Fig. 14 and the edge-aligned dual screen camera system described above, this subdivision method aligns the optical center on a dodecahedral vertex so that three screens are removed, namely the zenith screen 193, the left side screen 194 and the right side screen 195. Each individual Screen is made optically congruent with the others, as shown in Fig. 17 for dual screens.

By using a vertex-based hemispherical camera system, with each of five cameras producing three screens, all the side pentagons would receive double coverage, and the zenith pentagon would be covered by five overlapping screens. However, because the vertical height of the video raster scan, which is equal to the height of the optimal lens image circle, is divided into two pentagon levels instead of one, as shown in Fig. 14, each pentagon in a triple system would have only half the resolution of a dual system, so that the double coverage of the side screens, while producing separate image layers that could be used for three-dimensional effects, would only have the same resolution as regular recordings. On the other hand, the zenith pentagon would have two and a half times the normal resolution due to its five overlapping screens. The projection of all three pentagons into an overlapping arrangement also depends on a projection module that also has this wide field of view and is also aligned with the vertex, meaning that it would have to have the same pyramidal structure as the camera.

The overlapping zenith screens could be shown as concentric disks of progressively smaller size 196 with soft edges through the use of the masking processor to gradually build up the resolution towards the center. Within the boundary of a displayed image 197 there would be one or more areas 198 of reduced brightness where the light level of the images would be adjusted to prevent luminance jamming due to subsequent overlapping images. This method of increased resolution at the center is consistent with the increased resolution found at the center of human vision and would serve to improve the quality of the overall presentation.

Fig. 25 is an illustration of a top view of a dodecahedral vertex-aligned compound camera 199 for a hemispherical view. It has the shape of a flat pyramid of five modules or facets, each Facet 200 carries an electro-optical lens 201 having a field of view of at least 1520. The surface of each facet has an inclination of 37 22'38.5536" (37.377376) and an elevation angle for the central axis of the lens of 52 371 21.4464F1 (52.622624).

Fig. 26 is a perspective view of a vertex-aligned camera 202 showing the three screens covered by a camera facet. At 203 is the central axis of the facet aligned with a dodecahedral vertex, and at 204, 166 and 167 are the three optical axes of the three separate screens with congruent optics created by the electro-optic lens.

Figure 27 shows an illustration of a guide for planning a shot of footage for a dodecahedral mosaic. The outline of a single flat screen is shown at 205, with the horizon line illustrated at 206. This guide is useful for freely dragging across sections along the horizon, ensuring that the pentagonal screens will match. If this guide is used for animation footage, it could be shot with a motion control animation camera that can rotate and move to capture each section in sequence. The camera should preferably move to maintain the same alignment of light on the footage throughout. This would produce a filmstrip in which the six screens would be shown in sequence for each frame of motion. The transfer of this film into six separate exposures, allowing simultaneous showing of the six screens, would be accomplished by step-by-step printing on either an optical printer or a film-to-tape transfer machine.

Fig. 28 shows an illustration of another type of guide for planning images for a dodecahedral mosaic showing a geodesic grid on each screen at 207. This type of guide is more accurate because it gives correct proportions and the lines that define large circular segments in each direction. However, because of its unusual geometry, it may take some time to get used to. and care must be taken to separate the sections to ensure that the artwork fits together along the edges.

Fig. 29 is a circuit diagram showing all the screens together and adjacent by slight polar distortion, as well as an overlapping array of screens from a dual electro-optical lens. This array of screens 208 can be accomplished by a linear progression of polar distortion up to 16.666% from the bottom to the tip of each outer pentagon, creating an outer pentagon shape shown at 209. Despite its slight disproportion, this guide can be very useful because it maintains the height of the outer screens and shows that all of their edges match up. Therefore, this is proportionally the most correct way to show the image produced by the six pentagonal screens as a flat image, and it is also the best way to divide an existing flat image so that it can be shown on six screens. In this way, a large overall shape can be subdivided for transmission to the individual masking processors for each screen.

Also shown in this figure is the arrangement of overlapping screens from a dual electro-optical lens, with the left and right screens produced by each module. Note that the central axis of each module in a dual screen system is opposite a dodecahedral edge. Note also that because of the structure of the dodecahedron, the zenith pentagon cannot be double-covered in a dual screen system, except by a separate single camera.

Fig. 30 is a circuit diagram of a virtual image 210 of six screens formed into a disk and used in a rotating electronic viewfinder system capable of covering a herniated sphere. It uses the screen ratio of Fig. 28 with further compression by 79.2% of the outer pentagons from their centers to their outer tips to reveal the outline at 211. Although this causes the horizon line on each screen to curve upwards, in the center, it makes the arrangement of the screens more suitable for an entire rotating electronic viewfinder. The image information in the screens is stored in a virtual memory, preferably at low resolution in a computer to enable real-time operation, with individual segments from attached playback machines or from the camera modules themselves. The window of the viewfinder 212 is an image memory which retrieves image information from the virtual memory as required. The directions of movement of this window 213 are controlled by a joystick, the vertical movement of which brings the window to the zenith and the horizontal movement of which causes rotation about the central point so that the scene pans along the horizon line. If possible, other control means can expand and contract the size of this window as required.

Fig. 31 is a schematic showing a cross section of a motion picture theater projection system using laser video projection. As can be seen from Fig. 3, cross sections of a dodecahedron are asymmetrical. For clarity, the parts in this view have been arranged along the cross-sectional plane as if there were six projectors around the periphery instead of five. Six modules are shown, but a total of eleven can be used in the projection system, with the added modules shown in dashed outline to achieve maximum area coverage.

The video source, either a VCR or a camera module, is located at 118. To mask out all but the central pentagonal portion of the image, the video signal passes through a masking processor 119 or through other portions of the image as described above. By using this processor in conjunction with the processors of other projector modules, non-pentagonal full screens in a variety of shapes and sizes can be shown.

Three light beams are emitted in red, green and blue. In a preferred embodiment, these light beams are laser beams. Other light sources can be used, such as found in other types of video projectors, but for this description we will assume laser beams. A group of three lasers emit red, green and blue laser beams at 214. The signals representing the red, green and blue color information of the video signal are used to control a group of three light valves for these laser beams at 215, regulating the amount of red, green and blue light which in the preferred embodiment is combined into a white light beam which passes through a fiber optic cable 216 which is connected at 217 to the outer surface of a projector module in the projection head. There the white laser beam 218 is reflected by a mirror 127 for vertical scanning and a rotating mirror wheel 132 for horizontal scanning, thereby producing a video raster scan. These and the other mirrors for the other modules for the projector are kept in sync by a house video synchronization generator 110 connected to the various modules through a distribution amplifier 219. The reflected laser beam passes through a projection lens 133 onto a projection surface 134 having a spherical curvature. Because of the infinite depth of focus of a laser, this projection surface could be egg-shaped or have some other curvature, but such a projected image would not be an undistorted reproduction of the camera view.

The optical axis 53 of each projection module is perpendicular to one of the faces of a dodecahedron when projecting a single screen, or is aligned with a dodecahedral edge for an overlapping double screen projection, or with a vertex for an overlapping triple screen projection. The angle of view of each projection lens matches its respective lens in the camera system.

Each module of the projector is enclosed in a baffle 200 to limit light leakage from one module to another. Additional modules, up to a total of eleven, can be added for below-horizon imaging for a maximum field of view of 91.6% of a sphere, which is shown here shown in dotted lines as previously described for the camera system. The only limitations are the need to provide space for the projector mount and for the audience area within the sphere where the laser projected images will not land for safety reasons. This is to avoid eye damage to spectators looking into the laser beam coming from the projector.

A projection surface should minimize angular reflection, which would tend to erase adjacent images. A gray surface or a porous white one with holes could be used to allow sound from the speakers to pass through behind the screen. The screen material should be used in geodesic sections. A lenticular projection screen material would also be useful to limit the scattering of projected light. However, the more directionally reflective the screen, the more the image would disappear when viewed at increasing angles away from the optical axis.

When the centers of each face of a dodecahedron are connected, an icosahedron is formed, which is the basis for R. Buckminster Fuller's geodesic dome construction system. This dodecahedral projection system is therefore consistent with the geodesic dome construction, and the screen boundaries, components for hanging a projection screen, and supports for placing speakers all match the structural elements of a typical geodesic dome.

If a porous projection screen is used, sound can pass through it, and the struts and parts of the geodesic dome structure would act as acoustic traps to minimize internal sound reflection within the dome.

If the viewers are outside the spherical projection surface 134 and look at an image projected from behind onto a translucent screen, seeing the projector is not a problem because the screen is between them and the projector. The image could then be very bright, depending on the light transmission of the screen, and it may a complete sphere could be shown. However, a viewer could not see every view at once, as would be the case inside by turning the head, and the viewing perspective would appear distorted. A projection with the viewers outside would be best for special purposes, such as showing a complete picture of the Earth.

Fig. 32 is an illustration showing a cross section of a composite projector using liquid crystal display (LCD) panels, a motion picture light source, and two types of cooling systems.

LCD projection involves the use of a motion picture theater projection lamp 221, such as a high power xenon, krypton or tungsten halogen lamp, to produce a strong white light. An LCD panel 222 in each projection module displays the color video signal and acts as an active transparency, blocking or passing the light as needed to produce an image, thereby controlling the amount of light passing through the strategically placed blue, green and red stripes or filters on the surface of the LCD panel with the colors passing through the projection lens 133 and combining on the screen. JVC and Panasonic are developing such projectors, and Sharp has introduced the XV-100 LCD video projector, which can project a 100-inch (2.54 m) image.

An LCD projector for the present invention may require only a small number of projection lamps, perhaps only one to be located behind the LCD panels, and does not require LCDs of exceptional size as long as they have good resolution. Because of its simple construction, low amount of power required, and because it would be brighter than a laser projector, LCD projection is a second choice as the preferred embodiment. The main disadvantages of this system are the limitations on resolution in such an LCD panel, the scattering of light between modules due to internal reflections from the LCD panel facets, and the heat generated by a strong projection lamp, particularly as it affects the top projection module and the alignment of the others. Denser LCDs with more pixels or combinations of multiple LCDs would improve resolution, opaque walls between the projection modules would prevent light scattering, and cooling could be provided by a transparent glass-encased barrier filled with circulating water 223 between the light and the LCD, as is currently used to protect the film in certain optical film printers, and could be augmented if necessary outside the projected image area by cooling tubes for a supercooled gas 224 such as liquid nitrogen.

Fig. 33 shows a quarter section of a dodecahedron, drawn according to a plane passing between D, E and F of Fig. 1. The figure shows the characteristic angles and proportions of the dodecahedron. The proportions of the external and internal measurements are shown, according to the fact that 1.00 is the radius of the pentagonal face. This radius is shown at 225. The height of the pentagonal section from the middle of an edge to the center is 0.809017, shown at 226, and one-half of a dodecahedron edge is 0.587785, shown at 227. The length of the radius of the outer sphere of the dodecahedron, touching all the outermost points, is 1.6472779, shown at 228. The middle sphere intersecting its edges is 1.5388416 and shown at 229; and the inner sphere touching the centers of its faces is 1.3090167 and shown at 230. One-half of the dihedral angle 116.565054 between dodecahedral faces is illustrated at 231. The interior angles are also shown; that between the center of an edge and a pentagonal point, 20.905151, is shown at 232, that between a point and a pentagonal center, 37.377376, is shown at 233, and that between a pentagonal center and a pentagonal base, 31.717473, is illustrated at 234.

This completes the basic description of the invention as a video system.

Using the principle of pentagonal images in a dodecahedral arrangement, other variations are also possible. For example, the geometry of the dodecahedron can be varied, and different types of image recording and projection devices can be used within the scope of the invention. Accordingly, the scope of the invention should not be determined by the illustrated embodiments, but by the appended claims and their legal equivalents.

If two basic pentagonal image shapes are assumed rather than just one, then as the zenith and base of the pentagons are enlarged, the standard shape for the screens both above and below the horizon becomes a flattened pentagon, more similar in shape to the usual rectangular screens.

Further subdivisions of the dodecahedron and the other polyhedra open up hundreds of variations of polyhedral geometry. By using masking processors, many interesting new shapes could be shown, such as the rhombicosadodecahedron, which has faces that are pentagons, squares, and equilateral triangles. Other spatial video processors such as the Quantel ADO Video Effects Processor and the Quantel MIRAGE Video Effects Processor could also be used to manipulate these individual facets prior to projection.

Other regular polyhedra have been described, including the deltahedra, a type of dodecahedron made up of equilateral triangles, the stem-shaped forms, and even the Kepler-Poinsot polyhedra. However, they all require more facets than the dodecahedron to record a complete spherical image. However, if signal bandwidth and module size allow, other types of polyhedral interchangeable modular movie camera and projection systems may be considered.

As new image recording and projection technologies emerge, any can be applied to the recording and projection principle described here, as long as it follows the essential dodecahedral geometry.

The present invention is also suitable for use in special effects because it divides the entire image into manageable pentagonal sections. Because each section is captured with a standard lens with a field of view of approximately 75º, the capture from different sources can easily be made consistent, particularly when using the image guides described. When the present invention is used as a video system, the composition of different images becomes even easier, particularly when using digital video. Therefore, the combination of regular capture for a dome theater and special effects such as time lapse capture can be more easily accomplished to create vivid impressions such as a dinosaur emerging from behind a tree in an image of the Brazilian rainforest.

An alternative method does not depend on a geodesic grid for generating imagery for pentagonal screens. For motion picture animation above the horizon, imagery can be created for every part of the dome screen except the zenith using a conventional rectangular field recording grid to then generate pentagonal image sections that match up, each section being recorded in sequence through a pentagonal mask in the camera, with the imagery on a animation table with the top of the table tilted away from the film plane by 260.34 minutes. The right and left edges of the imagery are constructed to match up. The animation for the zenith should be planned according to a concentric grid with lines emanating from a point based on several groups of five and recorded on a flat table so that it matches up with the top edges of the other sections. These sections, when transferred to video and shown by means of the projection system according to the present invention, should produce something like a continuous panoramic environment of animation. I created such trick shots on 35 mm 5247 color negative film using a 5369 soft-edge two-layer mask on a Mechanical Concepts animation table at Lumeni Productions in Hollywood, California.

The small size of the video camera described here is recommended for handheld use as well as for action scenes that would be impossible with a larger camera. If necessary, however, larger cameras would provide improved image quality.

Many improved video recording techniques are currently being developed, and video recorders are becoming smaller. Sony recently announced an HDTV optical video disk recorder with two-channel optical recording. Digital recording and transmission of video recordings is becoming the new standard. Magneto-optical disk (MOD) recordings using a polarizing laser, such as those introduced by Thomson of France and Canon of Japan, may become the preferred means of video storage because they allow random access and can become remarkably small and portable, since they do not require tape reels and recording drums. It is also the recording medium used by the NEXT computer and may become the standard for multimedia software of the future. Solid-state memory cards are also currently being developed for storing digital video information.

The limits of recording are dictated by proposed research into three-dimensional crystal matrix recorders. The phenomenon of photon echo, in which photons can be stored at specific locations in a crystal by two lasers, then released and detected using a third, may make three-dimensional recording possible. A proposal by Mark A. Friedlander to the U.S. Navy describes a square array of 10,000 tiny semiconductor lasers that can be focused into 1 um data cells in a 4" (100,000 um) cube of cooled rare earth doped lanthanum trifluoride. If ultraviolet lasers could be used to reduce the data cell size to 0.1 um, the capacity of such a recorder would be a proposed NEC standard Bandwidth for digital video (with CD sound quality) of 42 million bits per second at 754 years!

US Patent US-A-3 832 565 entitled "Holographic Memory with Dodecahedron Deteor Matrix" by P. Graf and M. Lang, assigned to Siemens AG, describes an apparatus for retrieving information stored in the form of sub-holograms by using a detection matrix having the shape of a dodecahedron. In other words, the detectors surround a recording medium in the shape of a dodecahedron that faces inward. In the present invention, cameras record an external image which projectors reproduce. Mainly for reasons of symmetry, this would appear to be the most appropriate means of storing and reproducing the image data of the present invention, particularly if each area could serve as a storage and retrieval unit corresponding to one of the modules of the present invention, and if simultaneous multiple recording and playback were possible, then it would be the preferred recording embodiment of the present invention.

In addition to the described embodiment of laser video projection, there are several possible alternative methods for the inventive projection system. The field of large-scale video projection is currently evolving, and further developments may emerge to provide the function of displaying images from the masking processor of each projection module onto a spherical surface. There are two basic approaches, video projection or video display on a direct-view screen. Projection reproduces the image from the camera system on a pentagonal screen from the inside of a dome outward, while direct-view screens reproduce the image on the pentagonal screen from the outside inward. In addition to the previously described laser and LCD projection, video projectors also include cathode ray tube (CRT) projectors. Examples of direct-view screens include televisions, light-emitting diode (LED) screens, and massive fiber optic domes.

In CRT projection, the second most preferred Projection means, the various red, green and blue video signals control the intensity of the electron currents in the cathode rays. In the more common variety, they are used to reproduce bright monochrome phosphor images on the surface of three cathode ray tube (CRT) screens. The light from these screens passes through the appropriate filter and a focusing lens and falls on the screen, producing a color video image. This is the most common type of video projection today. However, their projected image tends to be relatively faint and indistinct when thrown onto large screens. A successful variation of this type is the Eidophor of Gretag, Switzerland, which uses the cathode rays to deform thin layers of hot oil on a concave mirror, thus creating variable reflectivity to deflect the light from a powerful projection lamp through mirror rods and color filters, producing a large, bright image with good resolution. Recently, this company announced the Model 5177, which provides HDTV projection up to 12 m (40') wide. However, these types of projectors tend to be large and heavy, take time to warm up, and are difficult to maintain alignment because gravity deforms the various oil layer angles in different ways. This makes them currently unsuitable for the composite projection system described here.

Direct view methods of showing the images from the projection modules would mean that no central projector would be needed for the movie theater. Instead, the images would be projected from outside the projection screen to inside. Direct view projection systems include large screen TVs, LED screens, and massive fiber optic domes. Large screen TVs, particularly from Hitachi and Mitsubishi, now have larger screens than ever before, and provide good quality images. But these sets are still limited to a diagonal of about 1.83 m (72"). In addition, they are not manufactured with pentagonal-shaped screens, and the need for a black border to limit the picture tube section, it excludes use in a composite system of adjacent screens.

A large, steady video display screen made of LED elements has recently become more widely used in practice as the addition of blue LED elements to the existing red and green LEDs has been perfected. The L330RGB from Ledtronics, Torrace, California, is an example of an integrated red, green and blue LED package that can produce any color and therefore serves as a single pixel of a large screen. It is planned to make these pixels as small as the T1 size, which measures only 3 mm in diameter. The overall brightness of such a screen is limited by the brightness of the blue LED elements, which are typically much dimmer than the red and green LEDs. Other disadvantages of large LED screens result from heat, poor color matching due to manufacturing variations, and the limited life of the individual pixels.

Screens made from fiber optics are another alternative for a direct view system. Advance Display Technologies, Golden, Colorado, currently offers direct view screens for airports and stadiums made from solid rayon fiber bundles. The bundles get thicker the closer they get to the viewing surface and end at an angle to make a 6" x 6" (15.24 cm x 15.24 cm) "sign." These bundles or signs are constructed to form a large viewing screen for a video image projected into their other end. Excellent brightness and clarity can be achieved along with a very wide viewing angle, and resolution is limited only by the size of the fibers. However, these screens are currently limited to heavy flat rectangular screens measuring 9' x 12' (2.74 m x 3.65 m), and it would be a difficult task to make concave screens large enough to cover a dome theater and assemble them with proper alignment. Moreover, if this could be accomplished, the entire interior surface of the dome would essentially be a solid mass of fiber optics, and thus repair would be almost impossible if one of the bundles had to be replaced. If the main fibers carrying the image could be made to multiply and branch out to effectively expand the image into a separate fiber substructure close to the screen, then individual areas could be more easily created and repaired.

However the projection system of the present invention is designed, its large, steady image will give viewers the entirely new impression of not just "watching a movie" but actually "being there," while the increased resolution and three-dimensional effects made possible by the use of an electro-optical lens enhance the credibility of the projected image. In addition, the flexible shape of the composite image made possible by the use of masking processors provides creative variation in the expression of new ideas.

It is understood that due to the small, lightweight, portable nature, the position of the camera is extremely flexible. The translation from a module level to shooting above or below the horizon is for convenience of description and is not a rule for practice.

Likewise, a spherical theater is also capable of infinite manipulation of orientation. Unlike conventional structures, a sphere has no corners that prevent the free rotation of the entire structure, especially when built as a rigid sphere with rail guides.

Because of the small, compact size of the projection head made possible by the use of the invention in the preferred embodiment, the projection head on its fiber optic shaft is light enough to be controlled in movement and also independent of the seating plane; it can freely roam the theater. Because of its infinite focal point, a laser beam would always produce a sharp image regardless of its distance from the screen, and the masking processor allows image spots and fragments to be provided that the audience is always prevented from being directly exposed to the laser light. The rotation of the mirror wheels in the projection heads could act as a gyroscope to automatically provide stability in movement and a highly accurate scale for the orientation of the head.

The sound should of course be adapted to the presentation. Sound recording as part of the camera in the form of existing camcorders, is expected to be digital sound or sound for the future. Recent advances in the field of motion picture sound now include the practical, highly directional stereo Q sound developed by Archer Communications, Toronto, Canada.

Because of its symmetry, the dodecahedron has also been used for loudspeakers, namely in US Patent US-A-4 673 057 entitled "Geometric Transducer Arrangements" which describes the construction of a dodecahedral loudspeaker. If it is as suitable for use as a sound-producing structure as the present invention is as an image-producing structure, and the Graf and Lang dodecahedral recorder is used, then the dodecahedron may provide the key to the construction of the theater of the future.

The invention can be used for entertainment to provide a spectacular show. However, when such a large screen is used, different rules apply to those for the design of common television shows and films, and great care should be taken to avoid actual unpleasant physical side effects for the audience. The standard should be set by finesse rather than shock, and extreme movement or fast cuts should be avoided.

The present invention could also be used for commercial simulations, surveillance and flight simulators. As with motion pictures in general, the applications of such a powerful method for creating and presenting a living illusion are limited only by the talent of the users and the capacity of the audience.

A method and apparatus for capturing and projecting pentagonal images aligned with a dodecahedron using an electro-optical lens and a masking processor has therefore been described.

List of reference symbols Fig.7:

86 CCD

87 CCD control

89 Amplifiers

105 Lens

106 Sensor

107 Compare

109 Customize

110 Synchronizer

111 Recording or Projecting

113 Image storage

Fig.8:

110 Synchronizer

118 Video source

119 Masking processor

120 Mask

Fig. 10:

120 Mask Source

Fig. 31:

110 Synchronizer

118 Video source

119 Masking processor

120 Mask

214 Laser

215 light valves

Claims (39)

1. A method for photographing and projecting a field of view to a full comprising the steps: - centering a dodecahedron (55) in a concentric sphere representing a spherical field of view (169, 56) and extending the faces (149) of the dodecahedron onto the sphere, thereby forming twelve spherical sections divided by pentagonal boundaries (66), - photographing a composite image of a wide field of view up to a complete sphere, comprising a number of camera images and covering an area of more than one section, by means of at least one camera (104), each camera image being formed by a camera lens (105) whose optical axis (53) is aligned with the dodecahedron, the camera lens also having a field of view (64) sufficient to produce a camera image including at least one pentagonal section boundary (66), - Cropping the camera image to the pentagonal border, - aligning an array of projectors associated with the respective camera images, wherein each projector (117) has an optical axis (53) aligned with the dodecahedron, - Projecting the composite image by arranging projectors onto a concentric spherical screen (134), whereby a seemingly continuous composite image of the wide field of view up to a complete sphere is produced by the simultaneous display of adjacent camera images (52). 2. The method of claim 1, comprising the additional step of forming the camera and projector images by a video raster scan (144). 3. The method of claim 2, comprising the additional step of further randomly cropping the video raster scan (68-70) according to the application of cancellation signals from a raw video source (120). 4. The method of claim 2 or 3, with the additional step of aligning an optical lens to the center of an edge of the dodecahedron with a viewing angle sufficient to encompass two sections (153,154) and further comprising warping the video raster scan (159) to produce the effect of a lens with two optical centers equal to the centers of the two sections (161, 162). 5. A method according to claim 4, with the additional step of projecting overlapping images from adjacent projection modules (Figure 29). 6. The method of claim 2 or 3, with the additional step of aligning an optical lens (203) to a vertex of the dodecahedron with a viewing angle sufficient to encompass three sections (193-195) and further comprising warping the video raster scan to produce the effect of a lens with three optical centers equal to the centers of the three sections. 7. The method of claim 6, comprising the additional step of projecting overlapping images from adjacent projection modules (196). 8. Apparatus for photographing and projecting a wide field of view up to a complete sphere, comprising: - a composite camera comprising an array of interchangeable camera modules (104), each camera module having a Camera lens (105) having an optical axis (53, 186, 203) aligned with the dodecahedron and a conical viewing angle (65) of at least 74º sufficient to encompass at least one pentagonal subdivision (66) corresponding to a surface of the dodecahedron as projected onto a concentric spherical field of view (169, 56), - a cropping device (119) for limiting the camera image of the pentagonal division of each camera module to a pentagonal shape (67) corresponding to its respective area, so that the field of view portion of each camera module includes at least one pentagonal division, whereby the individual parts of a wide field of view up to a complete sphere are photographed, - a composite projector having an array of interchangeable projection modules (117) each associated with the camera modules, each of said projection modules having a projector lens (123) with an optical axis (53, 186, 203) aligned with the dodecahedron, and each of said projector modules operable to project its respective camera module image (52) onto a surface in a size and shape equal to said pentagonal divisions, thereby creating an apparently continuous image across the wide field of view up to a complete sphere by simultaneously displaying adjacent images. 9. The apparatus of claim 8, wherein both the camera module and its associated projection module are adapted to generate an image using a video raster scan (135, 144). 10. The apparatus of claim 8 or 9, wherein one or more charge coupled device (CCD) image sensors (86) are used for each camera module. 11. Device according to one of claims 8 to 10, wherein the signal representing the image from one or more camera modules is suitable for being recorded in a recording device (112). 12. Device according to one of claims 8 to 11, which is suitable for using a High Definition Television (HDTV) video standard with an aspect ratio of at least 9:16 (143). 13. Apparatus according to any one of claims 9 to 12, wherein each camera module comprises a lens aligned with the center of an edge of a dodecahedron, having a viewing angle of at least 140º sufficient to encompass two pentagonal divisions (153, 154), and further comprising distorting (159) the video image produced by the camera by applying drive voltages to produce the effect of a lens with two optical centers equal to the centers of the two pentagonal divisions (161, 162). 14. Apparatus according to any one of claims 9 to 12, wherein each camera module comprises a lens (203) aligned with a vertex of a dodecahedron having a viewing angle of at least 152º sufficient to encompass three pentagonal divisions (193-195), and further comprising distorting the video image produced by the camera by applying driver voltages to produce the effect of a lens with three optical centers equal to the centers of the three pentagonal divisions. 15. Apparatus according to any one of claims 9 to 14, wherein the cutting device (119) is adapted to reinforce a boundary corresponding to the pentagonal division only between the images produced by adjacent production modules. 16. Apparatus according to any one of claims 9 to 15, further comprising a masking processor (119) further adapted to limit the portion of the camera image displayed by the projection module, with boundaries (68-70) that can be dynamically controlled by the application of cancellation signals from a raw video source (120). 17. Apparatus according to claim 15 or 16, wherein the boundary corresponding to the pentagonal division (66) consists of a soft edge (136, 137) sufficient to conceal the boundary between adjacent images. 18. Apparatus according to any one of claims 9 to 17 which allows viewers to view the simultaneous display from inside the sphere (134) onto a front projection screen. 19. Apparatus according to any one of claims 9 to 18, wherein each projection module (117) comprises three light beam generators (125-127) comprising the red, green and blue video color components, and associated light valves (122-124, 215) for controlling the intensity of each light beam in accordance with a signal representing the strength of its respective color component in a video image (118) reproduces. 20. The apparatus of claim 19, wherein the light beam generators output laser beams (214). 21. Apparatus according to any one of claims 9 to 18, wherein each projection module comprises a liquid crystal display (LCD) screen (222) sufficient to produce a color video image from the respective camera module of the projection module, a projection lamp (221) for producing white light, a projection lens (133) and means for displaying the color video image on the liquid crystal display, whereby the white light of the projection lamp is colored and its intensity controlled in a manner sufficient to project the color video image through the projection lens onto a projection screen (134). 22. Apparatus according to any one of claims 9 to 21, wherein the lens of each projector module is aligned (186) with the center of an edge of a dodecahedron and is operable to display two adjacent pentagonal divisions (187, 188) adjacent the edge of a dodecahedron produced by a dual electro-optic lens. 23. Apparatus according to claim 22, adapted to overlap the dual pentagonal images from adjacent camera modules during projection (Figure 29), thereby providing two independently controllable layers of the projected image which can be manipulated for a stereoscopic effect. 24. The apparatus of claim 14, wherein the lens of each projection module is aligned (203) with a vertex of a dodecahedron and each projection module is operable to display three adjacent pentagonal divisions (193-195) adjacent to the vertex of a dodecahedron created by a triple electro-optic lens. 25. Apparatus according to claim 24, adapted to overlap the triple pentagonal images from adjacent camera modules during projection (196), thereby providing multiple independently controllable layers of the projected image that can be manipulated for stereoscopic effects. 26. Apparatus according to any one of claims 8 to 25, wherein the projection system comprises a screen having a vertical optical axis (54). 27. Device according to one of claims 8 to 17 or 19 to 26, which is suitable for making the projected image visible to the audience from the outside of the sphere through a translucent rear projection screen (134). 28. The method of claim 1, 2 or 3, wherein the optical axis (53) of the camera lens is centered on and perpendicular to a face of the dodecahedron and the camera lens has a field of view (65) sufficient to produce an image comprising the pentagonal boundary (66) of only the portion (52) corresponding to that face. 29. The method of claim 1, 2 or 3, wherein the optical axis (53) of the projector (117) is centered on and perpendicular to a face of the dodecahedron and the projector lens has a field of view (65) sufficient to produce an image comprising the pentagonal boundary (66) of only the portion (52) corresponding to that face. 30. Apparatus according to claims 8 to 12 or 15 to 21, wherein the orientation of the camera lens corresponding to a dodecahedron includes an optical axis centered on and perpendicular to a face of the dodecahedron and the camera lens has a field of view (65) sufficient to produce an image comprising the pentagonal boundary (66) of only the pentagonal division (52) corresponding to that face. 31. Apparatus according to any one of claims 8 to 12 or 15 to 21, wherein the orientation of the projector lens corresponding to the dodecahedron consists in having an optical axis centered on and perpendicular to a face of the dodecahedron, and the projector lens has a field of view (65) sufficient to produce an image comprising the pentagonal boundary (66) of only the pentagonal subdivision (52) corresponding to that face. 32. An optical device comprising: - a number of optical devices selected from the group comprising cameras and projectors, each having a lens mount and a lens with a viewing angle of at least 74º attached to the lens mount for producing a directed beam of light along an optical axis, - a number of interchangeable modular support structures (51), each modular support structure having at least five sides equidistant from a central axis (53), each side having at least one planar alignment surface having an average angle of approximately 58.28º with respect to a plane perpendicular to the central axis, and a holder means for allowing the modular support structure to contain one of the optical devices in such a way that the optical axis of the lens is nearly congruent with the central axis, - a fastening device (96) for securing the adjacent modular support structures (50) together along contacting planar alignment surfaces such that the optical axes of any optical devices contained in the adjacent modular support structures are aligned to meet near a central point (48), each optical axis also being aligned perpendicular to the center of a face of a pentagonal dodecahedron approximately concentric with the central point. 33. an optical device comprising: - a plurality of optical devices selected from the group comprising cameras and projectors, each comprising a lens holder and a lens (185) having a viewing angle of at least 140º attached to the lens holder for generating a directed light beam along an optical axis, - a number of rigid modular support structures with planar sides and mounting means for the optical devices (184), - a fastening device for securing adjacent modular support structures together, side by side, such that the optical axes of any optical devices contained in the adjacent modular support structures are aligned to meet near a central point (48), each optical axis (186) also being aligned perpendicular to the center of an edge of a pentagonal dodecahedron that is nearly concentric with the central point. 34. An optical device comprising: - a number of optical devices selected from the group comprising cameras and projectors, each comprising a lens holder and a lens (201) with a viewing angle of at least 150º, attached to the lens holder, for generating a directed light beam along an optical axis, - a plurality of rigid modular support structures (200) with planar sides and mounting devices for the optical devices, - a mounting device for mounting adjacent modular support structures side by side such that the optical axes (205) of any optical devices contained in the adjacent modular support structures are aligned to meet near a central point (48), each optical axis also intersecting a vertex of a pentagonal dodecahedron approximately concentric with the central point. 35. Apparatus according to any one of claims 8 to 17 or 30 or any one of claims 32 to 34, further comprising an electronic image viewing device for viewing a wide field of view, comprising: - virtual storage means for storing digital information forming a virtual image of six pentagonal screens (210) presenting a hemispherical view, the virtual image being formed in a disk by progressive polar expansion over the outer pentagons to bring their edges together (209), - Virtual memory viewing device comprising an image memory forming a movable window (212) capable of retrieving image information from the virtual memory, - Movement control means, such as a joystick, for the movable window for producing movement (213) along elevation and azimuth so that different directions can be viewed, with further controls to increase and decrease the size of the window as required with respect to the virtual image. 36. The apparatus of claim 35, further comprising compressing the spacing of the outer pentagons from their centers to their outer tips (211). 37. Apparatus according to claim 36, wherein the further compression is 79.2%. 38. A method for projecting a field of view up to a complete sphere, comprising the steps of: - Centering a dodecahedron (55) in a concentric sphere representing a spherical field of view (169, 56) and extending the surfaces (149) of the dodecahedron onto the sphere, thereby forming twelve spherical sections divided by pentagonal boundaries (66), - Projecting a composite image of a wide field of view onto a concentric spherical screen (134) to a complete sphere comprising a number of images of the spherical sections with adjacent images cropped onto the pentagonal boundaries, by means of an array of projectors associated with the respective images, each projector (117) having an optical axis (43) aligned with the dodecahedron, whereby an apparently continuous composite image is produced by simultaneously displaying the number of images (52). 39. A device for projecting a wide field of view up to a complete sphere, comprising: - a composite projector comprising an array of interchangeable projector modules (117), each projector module comprising at least one projector lens (133) having an optical axis (53, 186, 203) aligned with a dodecahedron and a field of view (57) sufficient to encompass at least one pentagonal subdivision (66) corresponding to a face of the dodecahedron as projected onto a concentric spherical field of view (169, 56), - image input device (118) comprising means for providing a pentagonal subdivision image for each projection module representing the appearance of at least one of the pentagonal subdivisions as applied to the concentric spherical field of view, - a clipping device (119) for limiting the projected image of the pentagonal subdivision image (67) according to its respective dodecahedron edges, whereby a seemingly continuous image of a wide field of view up to a complete sphere is produced by the simultaneous display of adjacent images.