CA2096095A1

CA2096095A1 - Method for producing a composite image from two different images

Info

Publication number: CA2096095A1
Application number: CA002096095A
Authority: CA
Inventors: Naamen Keskes
Original assignee: Individual
Current assignee: Societe Nationale Elf Aquitaine Production SA
Priority date: 1992-05-13
Filing date: 1993-05-12
Publication date: 1993-11-14
Also published as: FR2691316B1; FR2691316A1; NO931728D0; EP0570258A1; NO931728L; JPH0638023A

Abstract

ABSTRACT OF THE DISCLOSURE
Method for producing a composite image from two different images.
This method is of the type consisting in using each original image broken down into pixels, each pixel being assigned a code with which is associated a color whose primary components are determined by a palette of colors capable of generating said primary components for reconstituting said pixel on a medium, wherein the composite image is generated by juxtaposing, on the medium, pixels or groups of pixels of the different original images distributed in such a way that a pixel or group of pixels occupies the same position in space in the composite image as it did in the original images.
Applications notably for the production of a composite image from two images of geophysical parameter charts.

Description

209~
METHOD FOR PRODUCING A COMPOSITE IMAGE FROM
TWO DIFFERENT IMAGES.
BACKGROUND OF THE INVENTION
1 - Field of the Invention This invention relates to a method for producing a composite image from two different images, and more particularly from two images of geophysical parameter maps.

2 - Description of the Prior Art Until recent years, the philosophy behind computer-assisted applications comprising graphical outputs was to reserve a graphical screen to visualize the outputs of the application and to use an alphanumeric terminal for man-machine dialog. With the advent of multiwindowing (X-windows), everything can be generated on a same screen and several applications can be run simultaneously on one single screen. The images plotted in a memory by the application are comprised of elementary points (pixels) to which are attributed color codes defined with a number B of bits enabling 2B
colors to be defined. Conventionally, each code Ci is associated with a color characterized by its three components, Ri red, Vi green, Bi blue, by means of a correspondence established in a table still referred to as a color palette (or LUT or Look-Up Table, false color table, colormap, ...) linked to the screen. The contents of the color palette define a transfer function between the codes and the colors, i.e. its function is to define the colors associated with the different codes. For each code, there is, at output, a predetermined group of three e.g. 8-bit numbers defining precisely the primary components R, V and B of the color for the corresponding pixel. The 8-bit numbers are converted by digital-to-analog converters 209~39~
whose outputs constitute the electric signals R, V and B. The utilisation of the color palette at the memory output defines a memory architecture enabling the image to be represented with a reduced storage capacity. It further enables the correspondence between the codes and colors to be modified quickly, interactively and independently of the image, without modifying the contents of the table and completely rewriting the image memory. This interactive modification of the distribution of colors in the table is performed by means of a special applicatlon enabling the transfer function to be modified.
French patent application No. 2,668,276 proposes a color operating method for simultaneous representation on a screen of plural color images by means of a single color palette in which the palette is divided into code ranges, the number of ranges being at most equal to the number of images to be represented. To each image is assigned a range of palette codes and for each image an application is applied to the palette enabling the color correspondences to be modified so as to interactively vary the colors displayed on the screen for each pixel.
Such a method constitutes a very notable progress and enabled images to be simultaneously and separately visualized on a screen while having the possibility of applying the color palette to one or all of them, so as to more particularly enhance any feature of the images.
However, the need arose to blend or mix two different types of images on a screen medium.
There is a known method enabling two images to be mixed according to which, by computation, the intensity of the primary components R, V, B of the color is modulated for each pixel of an image by the intensity 2~9~95 of the color of the corresponding pixel in the other image.
The method consists in breaklng down a first image into a certain number of points, the color of each point, which can be a pixel or group of pixels, being defined by its primary monochrome components R, V and B. These components are in fact electrical signals which are respectively assigned to the channels of a color receiver. During a second stage, the maximum color intensity divergence of the second image is determined. For instance, in the case of a second image in black and white and if it is decided that white has an intensity equal to 0 and black an intensity equal to 1, an intensity value of 0 to 1 is defined for each point or pixel of said second image. During a third stage, a pixel of the first image is made correspond with a pixel of the second image and the amplitude of the signals R, V and B of each pixel in the first image is modified by assigning them the intensity value or coefficient of the corresponding pixel in the second image. During a fourth stage, the modulated monochrome signals R, V and B are respectively introduced into the channels R, V and B of the color receiver.
Subsequently, the R, V and B triplet of each screen pixel, excited by the modulated signals R, V and B, is integrated by the human eye and brain to provide a predetermined color.
The method briefly described above is a computation method which notably requires a breakdown of each colored pixel of the first image into its primary components R, V and B and a computation of the intensity coefficient of the color of the corresponding pixel in the second image and of the modulated signals R, V , B.

2 ~ 5 Another method of mixing two images is the one used conventionally for television. However, this method is only applicable to monochrome images R, V and B and only enables a minimal adjustment of the contrast and intensity of the image.
This invention aims to obviate the above-mentioned drawbacks and to provide a simple and fast method of mixing at least two colored images.
osJEcT OF THE INVENTION
The main object of this invention is to provide a method for producing a composite image from at least two different original images, of the type consisting in using each original image broken down into pixels, each pixel being assigned a code with which is associated a color whose primary components are determined by a color palette capable of generating said primary components for recomposing said pixel on a medium, wherein the composite image is generated by juxtaposing, on the medium, pixels or groups of pixels of different original images distributed in such a way that a pixel or group of pixels occupies the same position in space in the composite image as it did in the original images.
A further object of the invention is to associate at least one color palette with the images to be mixed, so as to vary the colors representing the intensities of the primary components of the colors in at least said images prior to injecting them into the channels R, V and B of the medium, and so as to thus enable the colors of the composite image to be varied according to whether such and such a feature of said composite image is to be enhanced or attenuated.
SUMMARY OF THE INVENTION

209~095 Accordingly, with such a structure, the operator is master of the choice of signals representative of the colors to be mixed in each of the two images and therefore of the colors of the composite image. He can S e.g. modify the signals of the colors in one image or in both, as he chooses, until the required result is obtained on the composite image.
BRIEF DESCRIPTION OF THE DRAWINGS
Further features and advantages of the invention will be apparent from the following particular description of a preferred embodiment of this invention as illustrated in Lhe corresponding accompanying drawings in which:
- Figure 1 is a schematic representation of a screen on which are projected two images to be mixed and the composite image resulting from the mixture.
- Figure 2 is a synoptic and schematic representation of the means for mixing the images.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
The method embodying the invention consists in mixing two images M1 and M2 to obtain a composite image M3, the three images M1, M2 and M3 being simultaneously projected onto a screen or medium 14.
During a first stage, each image M1 and M2 is broken down into pixels or group of pixels, each pixel being, as is well known by experts, composed of one or more triplets each comprised by the primary components R, V and B, said pixels being distributed regularly along horizontal and vertical lines. The first image M1 comprises pixels or groups of pixels P1, materialized by crosses in figure 1, and the second image M2 comprises pixels or groups of pixels P2 materialized by circles in the same figure 1. For the purposes of simplification, only four horizontal lines H1 to H4 and , 209~a3~
: five vertical columns Dl to D5 have been represented in each image Ml and M2, but it goes without saying that, in reality, each image Ml and M2 comprises a very large number of pixels distributed over a larger number of : 5 lines and columns.
~: In the schematic representation in figure 1, the consecutive pixels in the images have been separated by a :relatively large gap, again for the~ purposes of clarlty.
- :: 10 ~ Each:pixel Pl and :P2 is assigned a~code with which ;~ ~ is~associated:a color whose primary components R, V and -~: : : B are determined~ by at least one color~::palette 7 ~ : represented in figure 2. ~It :is ~possible~to~ use two :~ distinct color palettes with different:~ codes, one :15~ palette being assigned to~image Ml~and~the other palette to image M2, or to use a:single:~palette of :~ ~: which the code is divided into two parts, one part ~ ~:
: : assigned to image Ml and the~:other part assigned to : image M2 as indicated :in French patent:~appllcation:No.
:2,668,276.~:That ~application is~ incorporated~into this description as regards : the ~means and method for transforming:: and recomposing the~ pixels with their primary components at output of said pa}ette(s).
According to the invention, the composite image M3 :25 :is generated by juxtaposing, on the screen 14, pixels ~ ~ or groups of pixels of the two images Ml and M2 which ;~ : - are also distributed along horizontal and vertical lines:but in such: a way that a pixel Pl from image Ml alternates with a pixel P2 from image M2 both horizontally and vertically while preserving the position in space they had in the original images Ml : . and M2. In fact, and in reference to the simplified figure 1, the only pixels processed and projected are : : ~ the odd-number ranked pixels Pl beginning with the ::

, :
:`~' '' ~ , , ' ,.,, . , , " .
.: ,: - . ,, , . ~
,'' ., ' ' ' : ~:
~:
::

2d9~9~
flrst one in the first horizontal line H1 counting from the top, in image M1, which alternate with the even-number ranked pixels P2 in the first horizontal line H'1, in the second image M2 and beginning with the second one. For the second horizontal line H2 of image M1, the only pixels processed and projected are even-number ranked pixels P1 whereas for the second horizontal line H'2 of image M2, it is the odd-number ranked pixels P2 which are processed and projected.
The other horizontal lines of images M1 and M2 are processed similarly to obtain the alternation required.
The composite image M3 obtained by the mixing of images M1 and M2 comprises pixels P2 which occupy the same position in spàce as they have in the original image M2, as regards both the horizontal and vertical lines. According to another feature of the invention, the pixels P1 of image M1 are distributed in the composite image M3 in such a way that, on two consecutive horlzontal or vertical lines, they are arranged in quincunx, as represented in figure 1. In this figure, pixels P1 and P2 also found in the composite image M3 have been connected by broken lines.
The composite image M3 has a given usable area, only limited by the dimensions of the screen 14, with the pixels or groups of pixels juxtaposed to form the composite image M3 taking up the entire usable area.
In figure 2, an arrow F1 represents the connection between the image M1 and a device 1 which is notably capable of modifying the intensity of the signal representative of said image Ml. Likewise, arrow F2 represents the connection between image M2 and a device 2 which can be identical or different to device 1, but of which one of the functions is to modify the intensity of the signal representative of image M2.

20~95 In thi.s way, the devices 1 and 2 supply at output signals representative of the derived images M'1 and M'2 comprising the same pixel distribution as the original images M1 and M2.
According to this invention, a system 3, called a sieve or selector, receives at its inputs 4 and 5 the signals representative of the derived images M'1 and M'2 e.g. according to the mode described in relation to figure 1.
A signal representative of the pixel selected by the sieve 3 is available at the output 6 of said sieve - 3. The signal supplied by the sieve 3 is addressed to the input of a color palette 7, which comprises a means for generating, from the input signal, the primary components R, V and B of the pixels of the input signal. At output of the palette 7, there are electrical signals 8, 9 and 10 representative of the components R, V and B of the pixels selected by the sieve 3.
The electrical signals R, V and B of each pixel to be projected on the screen are injected into the corresponding red 11, green 12 and blue 13 channels of the screen 14. As the intensity of the primary components R, V and B varies from one pixel to another, "colored compositions" are obtained on the screen which are different from one point to another and which depend on the resolution of said screen, said "colored compositions" being integrated by the eye which only discerns a point of a certain color.
When images M1, M2 and M3 are displayed on the screen 14, it is still possible to vary the colors, or more precisely the "colored compositions" of the pixels, and therefore of the images, by acting on the ` 20~6a3~
color palette 7. The method embodying the invention i8 therefore interactive.
When a modification of the colors of the composite image M3 is required, the color palette 7 is acted upon until the required result is obtained on the composite image M3, since acting upon the ~color palette 7 modifies the components R, V, B and~therefore the color of the~ pixels projected on the screen ~14. There is therefore interaction between the three~ images ~lO~ ~appearing on the screen 14. ~
Furthermore, the~ interaction takes place~in real - ~ - time, ~as the operator has the~ result of the manipulations~ performed direetly and~immediately~on the composite image M3. ~ ~ ~
15 Such a feature is particularly~ interesting for geophysical applications, and notably when~an~isochrone map lS mixed with an azimuth map.
An azimuth map is ~an image~ on which the notion of ridge or valley is~ difficult to identify,~even for an expert. ~However, the~corresponding ~isochrone map or image contains the aItitude~data. The~mixlng, ~according to the inventioni~ of azimuth ~and~;isochrone~;~ images enables the azimuth image to~be enhanced by conferring a clear meaning to the variations and it ~becomes possible to clearly differentiate ridges, valleys or ,; ~ , faults.
The composite image obtained produces a relief effect and can also be further enhanced by reinforcing certain of its aspécts, by way of a modification to the colors from the palette for the isochrone image and/or the azimuth image.
of course, a zoom effect, when available on the équipment used, can be applied to the images displayed on the screen 14. However, better results are obtained ' ;~ , .. . . .
, :, ,, : , . . . ::
. .. : , , , : , . . , : :
::: , -.. , - :- ", , -, . :,-, . ,. , ,, . - :
~.: , -: - ., , -20~6~
when the zoom effect is applied to the images before they are mixed.
It goes without saying that devices 1 and 2 described in relation to figure 2 can be omltted when the contrasts of images Ml and M2 to be mixed are sufficient.

Claims

1 - A method for producing a composite image from at least two different original images, of the type consisting in using each original image broken down into pixels, each pixel being assigned a code with which is associated a color whose primary components are determined by a color palette capable of generating said primary components for recomposing said pixel on a medium, wherein the composite image is generated by juxtaposing, on said medium, pixels or groups of pixels of different original images distributed in such a way that a pixel or group of pixels occupies the same position in space in said composite image as it did in the original images.

2 - The method as claimed in claim 1, wherein said composite image has a given usable area, with the juxtaposed pixels or groups of pixels taking up the entire usable area.

3 - The method as claimed in either claim 1 or 2, wherein, in said composite image, a pixel or group of pixels belonging to one of said original images is separated from a pixel or group of pixels from said same original image by a pixel or group of pixels from another original image.

4 - The method a claimed in claim 3, wherein, in said composite image, the pixels or group of pixels are distributed along horizontal and vertical lines.

5 - The method as claimed in claim 4, wherein the pixels or group of pixels from a same original image on two consecutive lines are in quincunx.

6 - The method as claimed in any one of claims 1 to 5, wherein the images are two images of geophysical parameter maps, such as an isochrone map and an azimuth map, said images having the same resolution.