CN100576933C - Method, device and autostereoscopic display for tracking optimal position - Google Patents

Abstract

The present invention relates to a method and apparatus for tracking the optimum position of a optimum position unit in a transmissive electronic display. The object of the present invention is to improve the reconstruction quality and illumination uniformity of such displays. The display contains, in addition to the image matrix (4), an optimal position unit consisting of an illumination matrix (1) and reconstruction elements. Once at least one observer's eye (6) position is determined by the control unit using inverse ray tracing, address data for switching on the lighting elements (LE) of the lighting matrix (1) are provided from the position data to the said observer's eye ( 6) Prepare to determine the best position (5). To improve the reconstruction quality, additional optical components are used in the ray paths of inverse ray tracing. In addition to processing the observer's viewing angle (α), the control unit also detects and takes into account certain angles (θ) of scattering or diffractive elements of a predetermined angular range. This enables additional address data to be turned on for the lighting elements (LE), and the determined optimum positions (5) can be illuminated in a uniform manner.

Description

Method, device and autostereoscopic display for tracking optimal position

The present invention relates to a method and apparatus for tracking the optimum position of an optimum position unit in a transmissive electronic display for displaying information, said method and apparatus projecting light through the display in a directional manner to the optimum position after the information has been modulated. position on at least one of the observer's eyes.

The invention may be used with monoscopic and/or autostereoscopic displays for one or more observers. The display device according to the present invention enables images to be selectively displayed in two-dimensional, three-dimensional or mixed modes. In this document, the term "autostereoscopic display" designates a display device with which at least one observer can view a three-dimensional image from a plurality of freely selectable positions without any additional assistance.

Viewed in the direction of light propagation, the display's sweet spot unit comprises an illumination matrix with a plurality of illumination elements that emit or transmit light, and an imaging device with imaging elements. The imaging device projects the light of the illumination elements with the illumination array turned on in one or more optimal positions, in an almost ideally parallel beam, into the eye of at least one observer. To this end, a plurality of illumination elements are assigned to each imaging element of the imaging device. The sweet spot is the area where the information provided by the info panel can be observed in high quality.

The uniformity of information displayed on the information panel must always be ensured at the optimum position, and information crosstalk between the eyes of the viewer must be prevented when viewing three-dimensional content. These conditions must be continuously fulfilled in order to continuously provide high quality monoscopic or stereoscopic image content to the observer if the observer changes his position in the space in front of the display device. For example, in a monoscopic display for a vehicle, passengers can watch a movie while a route map is provided to the driver. Both should be able to move within a certain range without losing their specific information.

Furthermore, it has been found that if the number of light sources switched on for a determined optimum position is only slightly too small or too large, interference and imaging defects are likely to occur. For example, observers may perceive crosstalk between the sweet spots, as well as deterioration of image quality such as uniformity and contrast. The human eye can easily detect this change.

Autostereoscopic displays are expected to render high-quality 3D scenes and exhibit features independent of the number of observers, such as observer freedom, discrete movement, and the option to obtain multiple maps in 2D and/or 3D path.

In order to be able to meet all these requirements as far as possible, a suitable tracking system is required, which provides information for the subsequent processing of the tracking device and the stereoscopic information display. Such a tracking system must be able to constantly detect the movement of observers in as large a viewing space as possible in front of the display device in order to always provide each observer with specific image information independent of their actual position. This places extremely high demands on the accuracy of the position detector, the quality of the individual components of the display and the overall image quality of the display.

Tracking systems exist for tracking using mechanical, optical and other means, or a combination thereof. However, these systems suffer from the disadvantage that the accuracy or suitability for real-time applications is adversely affected. Their designs are often bulky and have very limited viewing space that can provide information to the observer. Furthermore, the required computation time considerably increases the considerations in the process from position detection to information provision.

WO 03/053072A1 patent application document discloses a multi-user display with a tracking system for collecting three-dimensional position information and continuous stereoscopic image display. The display consists of a sequential arrangement of a three-dimensionally addressable backlight, a large-area imaging lens to focus light on the viewer, and a light modulator as an image matrix. The backlight is composed of a plurality of two-dimensional light source arrays, and the two-dimensional light source arrays are sequentially arranged in multiple planes one by one. A lighting element in one of the light source arrays of the backlight is turned on depending on the actual position of the viewer. This approach also allows the light source to track the distance between one or more observers and the display, respectively. Dynamically detect the three-dimensional position information of all observers' eyes, turn on the assignable lighting elements of the backlight, and focus the light beam and the modulated image to the observer's left/right eyes at the same time.

The disadvantage of such displays is low brightness because only one locally selectable point light source is available to illuminate the entire image for each observer's eye, and also because a portion of this light is absorbed by the array of unactivated light sources in the light path. In addition to its bulky size, such three-dimensional backlights are also difficult to manufacture.

US 6014164 patent discloses an autostereoscopic display for multiple observers with a light source tracking system. The light sources arranged in pairs for each observer can be moved in x, y and z directions to track changes in the observer's position with the help of a control system. The method can continuously provide the three-dimensional scene to the observer on the information panel. Because this system also uses the light source tracking method, the above-mentioned disadvantages also appear in this system. In particular, since the observer-specific pair of light sources must be tracked individually for each observer, expensive tracking devices are required. These devices do not allow the display device to achieve a flat design.

Known tracking methods are only able to deliver specific information to viewers at different positions in a rather constrained stereoscopic viewing space. The tracking range and brightness of the image matrix are limited. Furthermore, the displays are bulky and require expensive equipment, including computing equipment, to enable tracking. The more observers are located at different positions in front of the display, the greater the amount of data to be calculated and thus the greater the delay between position detection and the actual provision of the best position. As a result, it has become a common practice not to compute parts of the data in real time, but to store precomputed data in lookup tables that are looked up and processed when needed. Another disadvantage is that the storage capacity of the system will quickly be used up as the number of observers increases.

It is an object of the invention to prevent or reduce the above-mentioned disadvantages of the known technique by means of inverse ray tracing. The inverse ray tracing method used here determines the propagation of light based on the geometric arrangement of optical components. It exploits the property of light that the path of light is reversible, so that all light rays can be traced back from the eye to their point of origin.

In particular, it is an object of the present invention to provide a display with an optimal position unit that has the ability to rapidly and non-mechanically track an extended optimal position as a function of the position of the observer over a large viewing space. position and deliver observer-specific image information to a single sweet spot area in the image matrix while maintaining continuous high image quality and uniform illumination of the information panel.

At the same time, the tracking range is increased, and crosstalk is removed for each observer-specific image. Furthermore, the amount of data to be computed by the tracking method in real time is reduced while maintaining an acceptable computation time, and the amount of precomputed data stored in the lookup table is also kept as low as possible.

It is a well-known fact that scattering and diffraction by optical components in the light path greatly affects the image quality and thus the homogeneous illumination of the sweet spot. The selection of certain optical components may cause scattering and diffraction in the horizontal or vertical plane or in all planes. The purposeful use of scattering and diffractive devices further improves the uniformity of illumination and the quality of the sweet spot, especially if the structure of the illumination elements is very thin.

This object is solved in a manner according to the invention based on a tracking method for a display with an optimal position unit and an image matrix, in its multiple processing steps to detect in real time by position detectors, lighting elements The spatial position of the observer's eyes begins with the illumination elements arranged in a regular pattern in the illumination matrix and switched on in the optical pathway to provide the observer with a defined optimal position.

According to the invention, this object is solved by a tracing method comprising inverse ray tracing in which rays are traced from the viewer's eye through the image matrix and the imaging device to the illumination matrix. This enables the exact switching on of the lighting elements of the lighting matrix that are visible to a viewer looking through the image matrix. In conjunction with the imaging device, but independently of the type of imaging device used, the observer's eyes receive the directional illumination assigned to them at an optimal position.

After successive detections of the spatial position of at least one observer's eye by means of the position detector, the detected position information is transmitted to a control unit which performs inverse ray tracing. Depending on the accuracy of the position detector and/or other parameters, in particular the distance between the viewer's eyes and the display, the desired optimal position geometry is formed within the control unit by determining reference points in a separate step. The number and position of reference points for each observer's eyes can be freely selected. In the present invention, a square arrangement of reference points is preferred, that is, inverse ray tracing is performed on at least four points preferably forming a rectangle.

The number of reference points must be large enough so that the sweet spot is always evenly illuminated, but small enough that the required computing power and storage capacity are as low as possible. If the observer is located relatively far away from the display, for example, the range of optimal positions has to be larger, so that the accuracy of position detection is significantly reduced.

The control unit performs backward ray tracing from each reference point up to pixels arranged in the image matrix grid and determined by at least one parameter. It is better not to perform calculations on all pixels, but only on those pixels that lie in a column, so that it can be determined by only one parameter from a given image matrix grid.

Backward ray tracing from the pixels through the imaging device to the lighting elements. The calculation generates a data record with address information of the lighting element, which is projected by the corresponding imaging device to the respective reference point. In a subsequent processing step, a pattern of lighting elements to be switched on is generated on the basis of the address information, which pattern is imaged by the imaging device in the above-mentioned optimum position for each observer's eye.

The method according to the invention is applicable to both monoscopic and autostereoscopic displays for at least one observer.

While back ray tracing for each pixel is preferably performed in real-time for the central column of the image matrix, the control unit looks for other ray paths to precomputed data records stored in lookup tables for lookup lighting element address information. The address information of the lighting element pattern corresponding to the current observer position is looked up by the control unit by means of comparison data records and is used to generate the corresponding optimal position, where the viewing angle α determines the light rays from the reference point to the viewing pixels of the image matrix path. The precomputed ray paths of rays from each pixel in the grid of the image matrix to each lighting element of the lighting matrix are stored in data records in the look-up table.

Aberrations caused by the imaging device, such as field curvature, are best accounted for in the data recording. The well-known material and manufacturing tolerances as well as the temperature-dependent behavior of the optics used must also be taken into account.

Using viewing angles as a parameter for inverse ray tracing has the advantage that it reduces the amount of calculations to be performed, since certain viewing angles apply not only to one observer located at a certain distance in front of the display, but also to viewers located at different distances. multiple observers. If the same stereoscopic image is provided to all observers, the address information preferably only contains two separate sequences of lighting elements, one for each detected position of all left and right eyes of the observers. The lookup table contains a number of precomputed rays and view angles that are used to generate optimal positions for associated pixels of the image matrix and all lighting elements of the lighting matrix. These data records collectively represent the conversion data for the lookup table.

One embodiment of the invention includes an illumination matrix consisting of shutters with discretely controllable sub-pixels combined with a directional backlight. Alternatively, the lighting matrix can contain discrete controllable light-emitting lighting elements such as LEDs or OLEDs, thereby removing redundant backlighting and maintaining the flat design of the display.

Furthermore, the invention relates to a device for providing control signals in a display for the tracking of the best position of the best position unit, comprising suitable control and storage means to implement the method according to the invention.

According to another embodiment of the invention, at least one additional optical component in the form of a scattering or diffractive element is used in the light path, which is why there is an additional parameter for finding address information and thus improving the picture quality and illumination uniformity. The control unit also detects and takes into account the angle Θ (in the angular range), which determines the optical properties of the scattering or diffractive means.

The angle Θ is the scattering angle of a scattering device comprising at least one scattering element, or the diffraction angle of a diffractive device comprising at least one diffractive element.

Since multiple rays are required to generate the optimal position, there are multiple ranges of angles related to angle α for the calculation. The control unit preferably detects these angular ranges depending on the size of the lighting matrix grid actually used. To keep calculations simple, the angular extent information is not detected and taken into account depending on the position of scattering or diffractive devices in the light path, but is always at a position accessible during inverse ray tracing. To generate the determined optimal position, all lighting elements of the lighting matrix found during inverse ray tracing are turned on.

Due to the consideration of scattering or diffractive means during the inverse ray tracing according to the invention, additional lighting elements will be switched on in a simple manner. This further reduces or even eliminates image defects such as dark bands, crosstalk and lack of contrast.

The method for tracking the best position of the best position unit according to the present invention makes it possible to reduce the amount of conversion data stored in the look-up table and the required calculation time. In particular, if individual observers are simultaneously provided with different information in 2D and/or 3D mode, observer-specific image information can preferably be tracked in real time with high quality even when the observer changes his position in the viewing space.

Now, the method and corresponding device according to the present invention will be described in detail. in the picture

Figure 1 is a top view schematically representing a multi-user display with a sweet spot unit, an image matrix 4 and observer eyes 6 in a sweet spot 5,

Fig. 2 is a top view showing the light rays starting from the reference points _P1 to _P4 of the optimum position at the corresponding viewing angle α to the pixels _Dp and _Dr of the image matrix 4,

Figure 3a represents a sequence 7 of lighting elements forming a pattern M of lighting elements LE,

Fig. 3b shows a detail of a lighting matrix 1 with activated and non-activated lighting elements LE

Fig. 4 is a top view showing rays RT _i along the reverse ray tracing process, said rays are scattered by the scattering device SF in an angle range of ±θ, and the scattered rays RT _i0 to RT _i3 are directed to the lighting elements LE, and

FIG. 5 is a flow chart showing a backward ray tracing program in the display device.

The description is best referred to in autostereoscopic displays.

Now, in connection with an autostereoscopic display, and with reference to FIGS. The device is described in more detail. The method is based on the idea that for each observer's position only those lights are turned on Element LE. The object of the present invention is achieved by means of an inverse ray tracing method and a corresponding device.

Referring to FIG. 1 , an autostereoscopic display comprises the following main components: a sweet spot unit comprising an illumination matrix 1 with illumination elements LE _On to LE _qk and an imaging device comprising a lens 2 and a Fresnel lens 3 . A transparent image matrix 4 with pixels D _o to D _q , through which the unmodulated light of the sweet spot unit penetrates, is used to present a stereoscopic image. For the sake of simplicity, only one observer's eye 6 and an optimum position 5 determined by reference points _P0 to _Pn are shown in the figure.

Figure 2 illustrates the process of inverse ray tracing, which respectively represent the hypothetical ray paths projected from the reference points _P1 to _P4 at the best position to two pixels _Dr and _Dp randomly selected in the image matrix 4, and the corresponding four viewing angles α _r1 to α _r4 and α _p1 to α _p4 .

FIG. 3 a represents a sequence 7 of lighting elements, which forms a pattern M, containing all address information of the lighting elements LE of the lighting matrix 1 created by the control unit. The lighting elements LE (see Fig. 3b) generating the determined optimal position 5 will be switched on according to the pattern M described above.

Fig. 4 shows the calculated light RT _i emanating from a randomly selected reference point and its path scattered at a scattering angle θ by an exemplary scattering sheet SF. It illustrates how to turn on the additional lighting element LE based on the scattering angle Θ on the scattering or diffractive device when the additional optical element has an angular range of ±Θ and ±Θ/2. Here the angular range ±θ is detected only in the horizontal direction. Often, the process can also be done vertically.

Fig. 5 is a flowchart showing the main processing steps of the inverse ray tracing program.

The Lighting Matrix 1 is the main element of the multi-user display sweet spot unit. It provides an ideal, uniformly lit stereoscopic image to the observer continuously through the sweet spot 5, also as the observer moves.

As described in the main patent application, the position, number and extent of the optimal positions 5 to be generated can be controlled by a control unit and realized by means of a lighting matrix 1 consisting of a plurality of lighting elements LE _0n to LE _qk , which can Turn on discretely as shown in Figure 1. In the embodiment described, the lighting element LE is a monochromatic lighting element of an LCD illuminated by a backlight (not shown).

However, it can also be LEDs, OLEDs or similar point or line lighting elements arranged in a regular pattern and which can be turned on discretely.

The imaging device is a multi-part assembly consisting of a lens 2 representing the optical imaging system and a Fresnel lens 3 as an objective lens, wherein the Fresnel lens projects the sweet spot 5 onto the observer's eye 6 . Alternatively, the imaging device may be only the lens 2 . If desired, additional optics can be integrated into the imaging device to improve projection conditions. It is possible to supplement the lens 2 composed of vertically arranged microlenses with at least one lens composed of horizontally arranged microlenses. Combinations of more lenses are also possible. Instead of the lens 2, an optical system composed of a lens array arranged in a matrix or composed of prism elements may be used.

In another embodiment, the imaging device may further include a correction array for correction of the curvature of the image field. Depending on the type of image matrix 4 used, an additional retarder foil can be placed in the optical projection path to change the polarization of the light. A certain number of lighting elements LE is always assigned to each imaging element.

Referring to the top view in Fig. 1, the eyes 6 of an observer are located in the viewing space in front of the display, more precisely in the optimal position 5 extended in a given plane. First of all, the best position 5 does not exist in reality, but is predetermined in consideration of system parameters and viewing conditions. The range of optimal positions is illustrated by reference points _P1 to _Pn . The reference points P ₁ to P _n can also be arranged in any pattern, including three-dimensional patterns. The parameter n should be at least 4 to form a rectangle for inverse ray tracing to be able to achieve a clear sweet spot of 5. In this embodiment, n is 12. Depending on the accuracy of the position detector used and/or the position of the viewer's eye 6 relative to the display, the depth of the sweet spot 5 may be lower or greater than shown here. The more precisely the position detector works, the smaller the optimum position range can be. Point _P0 represents the eye position detected by the position detector. If there is more than one observer, the positions of all eyes 6 in the viewing space are dynamically detected and the corresponding position information is provided to the control unit for backward ray tracing.

After continuous real-time detection of the spatial position of the observer's eye 6 on point P ₀ , an imaginary optimal position 5 is formed around the eye by means of scattered reference points P ₁ to P _n . The ray paths RT 1 to RT n from each reference point P ₁ to P _n to the pixels D ₀ to D _q in the selected column of the image matrix 4 are calculated in _the control unit (see FIG. ₁ ). The image matrix 4 is subdivided into a grid of constant spacing, which forms the basis for pixel calculations. The grid spacing may or may not be the same as that of the image matrix 4 . However, it is also possible to use a grid consisting of several areas with different pitches. However, in the case of inverse ray tracing, it is better to use a pitch larger than that of the image matrix, as this significantly reduces the computational power required.

Pixels D ₀ to D _q are identified by their x-coordinates in the column in which they are located. In practice, the center column of the image matrix is used because the viewer prefers to look towards the center of the image display. Another parameter required for the calculation is the viewing angle α at which rays from the reference points P ₁ to P _n strike the pixels D ₀ to D _q of the grid. Experience has found that about 4000 viewing angles α are used to obtain sensible calculation results. If the number of views is considerably less than 4000, then the tracking accuracy will be adversely affected.

In the case of a two-dimensional imaging device, a pixel is not only defined by its x-coordinate, but also by its x and y-coordinates.

In the embodiment shown in FIG. 1, the lighting elements LE _On to LE _qk are monochromatic lighting elements of a light valve (not shown in the figure) illuminated by a backlight. The observer's eye 6 is located on the reference point P ₀ . The rays cast from the external reference points P _k and P _n to the external pixels D ₀ and D _q calculated by the backward ray tracing method are shown in the figure. The ray from the reference point P ₁ to the pixel D _p is represented together with the viewing angle α _p1 at which the ray hits a pixel of the image matrix 4 .

The light path from the pixel D _p terminates at the lighting element LE _p1 of the lighting matrix 1 through the imaging element, which is preferably located in the central column of the image matrix 4 . This calculation is done for all pixels D ₀ to D _q and a number of viewing angles α. This ensures that all lighting elements LE _On to LE _qk that must be switched on to obtain uniform illumination of the sweet spot 5 formed by the reference points P ₁ to P _n are switched on. The lighting elements LE _0n to LE _qk hit by the light will be switched on together with the corresponding bar.

If too few lighting elements LE _0n to LE _qk are switched on, the sweet spot 5 and the switched-on image will be under-illuminated. Conversely, if the lighting elements LE _On to LE _qk are turned on too much, the sweet spot 5 will additionally illuminate the other eye, resulting in crosstalk and a decrease in stereoscopic image contrast.

Another variation of the identified optimal position is shown in Figure 2. It can be seen how light rays project from four reference points P ₁ to P ₄ to two pixels D _r and D _p in the process of inverse ray tracing, thereby having different viewing angles α _r1 to α _r4 and _p1 to α _p4 respectively. If the observer is located very close to the position detector, it is best to use this optimal position structure in order to determine the position of the observer with high accuracy with very few reference points.

Real-time inverse ray tracing from the reference point P of the determined optimal position 5 to the pixel D of the grid of the corresponding image matrix 4 produces as input data the result for finding precomputed data stored in a look-up table (LUT) .

The lookup table contains precomputed data records, which represent the calculation results of a large number of ray paths that have been completed according to the same algorithm, and its real-time calculation may take a large amount of time. This is why the ray path from each pixel D in the grid of the image matrix 4 through the imaging elements to the two-dimensional coordinates of the lighting elements LE of the lighting matrix 1 is pre-calculated and stored in a data record in the look-up table .

However, it is also possible to calculate the ray path to the lens 2 in real time. This reduces the number of data records and thus saves storage capacity.

A comparison of the parameters of the data record calculated in real time with those pre-calculated and stored in the control unit yields address information of the lighting elements LE, which are projected to the corresponding reference points P through the lens 2 and the Fresnel lens 3 . Imaging elements may be hit multiple times by spurious light rays, as seen in sequence 7 of illumination elements shown in Figure 3a. The numbers in the sequence indicate how many times the lighting element LE was hit during the backward ray tracing starting at the reference point P. However, the number of hits when turning on the lighting element LE is not relevant. Generally, the lighting elements LE hit by the light RT at least once will be turned on. Based on the address information, the control unit generates a pattern M in which all lighting elements LE switch on the corresponding columns of the lighting matrix 1 (see FIG. 3b ). This pattern is now used to achieve an optimal position 5 for the observer's eye 6 at a defined position.

If there are several observers in front of the display, the order of the pattern M of lighting elements LE to be switched on corresponding to the actual number of observer's eyes is determined. For example, if all viewers want to watch the same content, first all left eyes can be provided with the required stereo information using inverse ray tracing, and then all right eyes.

If, according to the described inverse ray tracing method, the lighting elements LE are not switched on, the stereoscopic image perceived from the position of the sweet spot 5 has the above-mentioned disadvantages. For example, edge voids of individual microlenses will be perceived as dark bands in the image, and the illumination of the image matrix 4 will be uneven. It has been shown that it is best to additionally account for scattering or diffraction of light during inverse ray tracing.

According to another embodiment of the invention, an angle θ is introduced which may be a scattering angle or a diffraction angle. It is detected and taken into account during inverse raytracing within a defined range of angles. For the sake of simplicity, the invention will be described below with the aid of a scattering device. However, scattering means may similarly be used in other devices. The imaging device is preferably assigned a scattering device with at least one scattering element in the beam path. The scattering element can be a scattering sheet SF with a defined scattering angle. It can be located behind or in front of the Fresnel lens 3, or at other positions in the optical path. If multiple scatterers SF are used, each scatterer SF can have a different scatter angle, so that additional parameters within the range of angles can be detected and considered to find address information, as shown in FIG. 4 . This also applies similarly to the diffractive device and diffractive angle.

The invention also includes the possibility to detect and take into account angular ranges in both horizontal and vertical directions in order to find additional address information. Referring to the embodiment shown in FIG. 4 , it schematically illustrates how to determine the angle range of the traced light RT _i based on the known scattering angle θ of the scattering sheet SF. A selected ray RT _i originates from a random reference point P and passes through the image matrix 4 at pixel D. It is diffused by the diffuser sheet SF so that a large number of light rays RT _i0 to RT _i3 (indicated by arrows) strike the lighting elements LE _i−2 to LE _i+2 of the lighting matrix 1 . If the angular range ±θ is used for backward ray tracing, then every other lighting element will be turned on. If the angular range is divided by 2 (θ/2, as shown in FIG. 3 ), the lighting elements LE _i-1 and LE _i+1 will also be hit by light during the process of backward ray tracing. The number of lighting elements LE to be switched on can thus be found more precisely, which contributes to the lighting uniformity of the determined optimum position 5 , but also further reduces the risk of the lighting elements LE not being switched on. In fact, the number of additional rays to turn on is much greater.

The range of angles to be used in a particular case also depends on the grid size of the illumination matrix 1 used.

The smaller the size of the illumination matrix 1 is, the smaller the angle range for back ray tracing is determined based on the actual scattering or diffraction angle. However, it must be noted that the smaller the angle range, the more address information is looked up and the more computation time is required. This is why, when applying the inverse ray tracing method, the emphasis is on the setup and computing power, while still achieving good image quality and uniformity of illumination at the identified sweet spot 5 .

The control unit will detect and take into account the address information with all found values for a large range of angles. The address information includes the viewing angle α and the angle θ or θ/2 corresponding to the scattering angle or the diffraction angle in addition to the x-coordinate. The address information of the additional lookup increases the accuracy of generating the minimum number of lighting elements LE to switch on for the optimal position 5 .

Referring to FIG. 5 , it shows the flow of the inverse ray tracing process, which starts from the position detector detecting the position of the viewer's eye 6 until the pattern M of the lighting element LE is formed to generate the determined optimal position 5 .

The present invention also relates to a device, more specifically to a processor comprising various functional units for implementing the above-mentioned inverse ray tracing method described in the independent device claims.

The method of the present invention for tracking the optimal position of the optimal position unit preferably provides a display which has the ability to determine the optimal pattern of lighting elements with a minimum amount of data to generate the optimal position on the viewing plane, observing where the viewer views specific information provided by a matrix of images that is uniformly illuminated at all times. Since the aberrations are already taken into account in the backward ray tracing method, the display optimally shows only small optical errors.

The use of a look-up table has the advantage that it is not necessary to repeatedly recalculate the individual lighting elements needed to generate the optimal position.

Therefore, according to the movement of the observer's eyes, the optimal position and the corresponding stereoscopic image can be tracked quickly and accurately in real time, and at the same time, the range of tracking the observer can be increased.

Component Mark List

1- Lighting Matrix

2-lens

3-Fresnel lens

4- Image matrix

5- Best location

6- Observer eyes

7- Sequence of Lighting Elements

LE - lighting element

D-pixel

M-Pattern

P-reference point

RT-Ray

SF-diffuser

α-angle

θ-angle (scattering angle or diffraction angle)

Claims (15)

1. A method for tracking the optimum position of a display optimum position unit, It projects light from an illumination matrix oriented in a sweet spot (5) onto at least one observer's eye (6) using a sweet spot unit (1-3) comprising a plurality of an illumination matrix and an imaging device (2, 3) composed of a controllable illumination element (LE), the light is modulated with image information by the pixels of the image matrix (4); Use a position detector to detect the position (P ₀ ) of the observer's eyes and the corresponding position information; Using inverse ray tracing to detect controllable lighting elements that turn on based on the detected position of the viewer's eyes, and switching on said controllable lighting elements for the optimum position by means of a control unit controlled by positional information, It is characterized in that the control unit - determination of reference points (P ₀ to P _n ) by means of position information, which determine an extended optimal position (5) around each observer's eye (6), - perform inverse ray tracing for selected pixels (D ₀ to D _q ) of a given image matrix (4) grid, starting at each reference point (P ₀ to P _n ), through these selected pixels ( D ₀ to D _q ) and imaging devices (2, 3), to the illumination matrix (1), and - by calling the data records pre-calculated according to the parameters and stored in the storage device, looking up the address information of the lighting elements to be turned on, so as to turn on only these lighting elements that generate the determined optimal position (5) together with the imaging device ( LE _0n to LE _qk ). 2. Method according to claim 1, characterized in that the control unit determines the geometry of the optimal position (5) depending on the position of the observer's eyes (6) on the display. 3. The method according to claim 1, characterized in that it is performed in real time from each reference point (P ₀ to P _n ) to a selected pixel (D ₀ to D _q ) located in a column of the image matrix (4) Backward ray tracing. 4. The method of claim 1, wherein inverse ray tracing is performed using a storage device by including a look-up table storing data records obtained from precomputed ray paths from the image matrix ( 4) each pixel (D ₀ to D _q ) casts to each lighting element (LE _0n to LE _qk ), where the look-up table is addressable with the viewing angle (α) that determines the distance from the reference point The light path through the pixel to the lighting element. 5. The method according to claim 4, characterized in that the control unit determines the address information of the lighting element for each reference point, wherein the lighting element is assigned to the reference by looking up a data record stored in a look-up table according to backward ray tracing. point, where the lookup table is addressed with the parameters described. 6. The method of claim 5, wherein the address information comprises separate sequences of controllable lighting elements for the left and right eyes of the observer. 7. The method according to claim 4, characterized in that backward ray tracing is performed from one pixel to each reference point using a viewing angle (α), wherein the viewing angle is multiple for each pixel according to the number of determined reference points value. 8. The method according to claim 4, characterized in that the control unit additionally assigns to the data record a "scattering or diffracting angle" parameter (θ) within the angular range of the scattering or diffracting means arranged in the optical path, wherein the used The range of angles depends on the actual grid size of the lighting matrix used. 9. The method as claimed in claim 8, characterized in that an angular range is assigned to the data record independently of where the scattering device or diffractive device is located in the beam path. 10. Means for controlling the optimum position unit of the display, where the optimal position unit (1-3) contains an illumination matrix consisting of a plurality of controllable lighting elements (LEs) and imaging devices (2, 3) for orienting the optimal position by means of an image matrix comprising pixels The light emitted from the lighting matrix is projected onto the observer's eyes; having a position detector for detecting the position of the observer's eyes and corresponding position information; The controllable lighting element that is turned on according to the detected position of the observer's eyes is detected by using inverse ray tracing, and has a control unit that is controlled by the position information to turn on the controllable lighting element for the optimal position, characterized in that, The device contains - control means for supplying control signals for determining the reference points (P ₀ to P _n ) which determine the extended optimum position (5) around the eyes of each observer, and from each reference point (P ₀ to P _n ) through selected pixels (D ₀ to D _q ) in a given grid on the image matrix (4) and through the imaging device to the illumination elements (LE _0n to LE _qk ) perform backward ray tracing and look up address information to switch on only such lighting elements (LE _On to LE _qk ) forming the determined optimal position (5), and - storage means, which are controlled in dependence on parameters, for looking up address information for switching on the lighting elements, said storage means comprising a look-up table containing the storage of data records, wherein the data records are selected from the reference point according to The parameters of the given pixels (D ₀ to D _q ) and the light paths through the imaging device to the illumination elements (LE _On to LE _qk ) are pre-calculated. 11. The device according to claim 10, characterized in that at least one scattering or diffractive device with a scattering or diffractive angle (θ) is located at any position in the light path for the optical properties of the scattering or diffractive device And the lighting elements to be turned on additionally search for additional address information. 12. Device according to claim 10, characterized in that the lighting matrix (1) is a light valve combined with a directional backlight. 13. The device according to claim 10, characterized in that the lighting matrix (1) consists of light-emitting lighting elements (LE _On to LE _qk ) that can be switched on separately. 14. An autostereoscopic display comprising a device according to any one of claims 10 to 13. 15. An autostereoscopic display as claimed in claim 14, characterized in that the display generates optimal positions for a plurality of observers.

Images