Method for the Opto-Electronic Transmission of an Image Original

It is known to televise polychrome image originals by forming the image optically, for example three times adjacently, on one and the same photo-sensitive coating of a camera tube and by scanning these images in raster configuration by means of an electron beam. The images will be situated in uniform alignment in a block like postage stamps.

Identical points in the image will be situated asymmetrically with respect to the longitudinal axis of the optical elements and with respect to the geometrical centre axis of the camera tube. Some elements generate symmetrical shading and geometrical distortions. The camera tube is subject to substantial geometrical distortion and changes of sensitivity, both of which act symmetrically with respect to the camera axis. To reproduce the image, the three images must be brought into flush alignment and each image point must be reproduced at the correct ratio of its intensity. This calls for comprehensive optical and/or electronic means for correcting the geometry and shading.

It is the object of the present invention to provide a method, capable of functioning without the correcting means which require exceptionally difficult adjustment and readjustment and permitting the use of camera tubes with substantial geometrical distortion, for example image amplifier tubes.

The method according to the invention is associated with the generation of two images situated adjacently and symmetrically with respect to each other on a surface which is at least approximately plane and by said images having a brightness distribution between them which is symmetrical for a monochrome image original and such images are interrogated so that the scanning rasters are also situated at least approximately symmetrically with respect to each other.

To achieve transmission compatible with broadcasting in accordance with the present invention it is advantageous to use two scanning rasters which repeat after two vertical scanning strokes and produce an interlace scan of at least one of the two images.

To achieve sharply focused transmission it is advantageous to employ two scanning rasters, which repeat after four vertical scanning strokes and each comprises four quarter rasters which nest within each other and are locally offset from each other by a quarter line.

To obtain the simplest possible optical and scanning systems it is advantageous for the two images to be generated on the photo-sensitive coating of a camera tube and to be scanned directly or indirectly by means of an electron beam.

For reasons of compatibility with conventional television systems two images are generated in flush aligned mirror image configuration on the photo-sensitive coating.

To obtain two images which are symmetrical with respect to the longitudinal axis of the camera tube the images are generated on the photo-sensitive coating are interrogated so that the scanning rasters are disposed symmetrically and more particularly in mirror image configuration with respect to each other and that in the case of frame and line scanning the latter takes place after two half field rasters, which are offset by a half line with respect to each other and are nested within each other. To this end it is advantageous to generate the two images by means of a frame splitter with a geometric-optically homogenous, partially transmissive, reflective plane which passes each incident pencil of rays partially without deflection and partially reflects it and by the use of two planes each of which performs reflective deflection and one of the planes is disposed upstream and the other downstream of the partially transmissive, reflective plane so that the line of intersection between the two deflecting planes is situated in the partially transmissive plane and their angles of inclination (.alpha.) are of equal magnitude with respect to the latter.

It is also convenient to employ an image splitter with two optical 30.degree./60.degree./90.degree. prisms, fixedly interconnected by means of a partially transmissive stratum.

The method according to the invention operates with only two images. However, three colour separations are necessary to provide correct colour reproduction. This can be achieved by means of a lens for generating an intermediate image which is projected by an intermediate optical system and by the adjoining image splitter on the photo-sensitive coating and image limitation is advantageously performed in the intermediate image plane.

To this end it is advantageous that at least one image is produced by striated filtration on the photo-sensitive coating by means of at least one striation filter which filters two different wave lengths ranges and/or distributions and the optically encoded electric video signals, produced by direct or indirect scanning, are distributed in accordance with the coding by electric decoding means.

To achieve the simplest possible signal transmission it is convenient that a striation filter be disposed in the intermediate image plane and a uniformly filtering filter be disposed in one optical path between the image splitter and the photo-sensitive coating or two uniformly filtering filters which are different from each other are disposed in both optical paths and the filter colour is selected so that electric video signals, whose coding components correspond to at least three different colour separations, are generated after scanning of the photo-sensitive coating. To avoid flickering images it is convenient for the images to be interrogated in time sequence by means of a switching mechanism and that at each instant only one of the electrical video signals is transmitted.

Since the image generated from the intermediate memory usually has unsharp edges it is convenient to generate, by means of a lens, an intermediate image which is projected on the photo-sensitive coating by an intermediate optical system and by means of the adjoining image splitter and image limitation advantageously takes place in the intermediate image plane. To obtain optimum geometrical area utilization of the photo-sensitive coating two images are generated with an axis of symmetry parallel with the line direction and the individual signals are stored, corresponding to a half field scan, in the intermediate memory, at least for the duration of a half field period.

When using an interlace scan it is advantageous to employ an identification signal to coordinate the electric video signals to each scanned colour separation and to form the identification signal from the horizontal and vertical synchronizing signals by means of a coincidence circuit.

The invention will be explained by reference to the accompanying drawing in which

FIG. 1 shows in diagrammatic form an exemplified embodiment of a system for performing the method according to the invention,

FIG. 2 the scanning rasters employed for interrogating the double image,

FIG. 3 the reproduction rasters employed for image reproduction,

FIGS. 4a through 4d the production of a rhombus by different decoding means,

FIG. 5 a perspective view of an exemplified embodiment of an image splitter,

FIG. 6 a section through a second exemplified embodiment of an image splitter arranged on a camera tube and a block circuit diagram of the connected signal processing system,

FIGS. 7 and 8 exemplified filter coefficients,

FIG. 9 a diagrammatic view of the use of the method according to the invention using a head amplifier tube and

FIG. 10 a modification of one detail of the schematic shown in FIG. 9.

FIG. 1 shows in diagrammatic form that the object 1 is projected by the lens 2 and by the image splitter 3 as an image 4 and 5 on the camera tube 6. The images 4 and 5 on the other hand are converted into the video signals 9 and 10 by scanning in accordance with the rasters 7 and 8. The images 4 and 5 are situated on the generally plane surface 11.

The lens 2 and the image divider 3 distort the images 4 and 5, for example in pin-cushion form. Since the images 4 and 5 are situated symmetrically with respect to the axis 12, the distortions are also symmetrical. The generated distorted images can readily be brought into flush alignment by two likewise symmetrical scans 7 and 8. This can be achieved most readily if the axis of symmetry of the scan 7,8 coincides with the axis of symmetry 12 of the images 4 and 5.

A photo-cell mosaic, for example a charge coupled device (CCD) can be employed in place of a camera tube 6 and scanning can be performed image spot by image spot and line by line. Scanning 7,8 can be performed by the interlace process (illustrated) so that the signal is broadcast-compatible. Other modes of scanning, for example PPI scanning for use with radar, are of course also possible.

FIG. 2 shows the horizontal scans H formed from the horizontal synchronizing pulses SH.

Two symmetrically disposed vertical scans VG and VX are formed from the vertical (frame) synchronizing pulses SV.

Identification synchronization SJ alternately renders one of the above-mentioned vertical scans operative to provide a vertical deflection VNP.

FIG. 3 shows the line trace produced by the scanning described with reference to FIG. 2 for reproduction of the G and X image. It can be seen that both images are reproduced with a quadrupled interlace scan. The images therefore contain all video data but provide the impression of drifting lines unless line repetition is employed.

FIG. 4a shows the reproduction of a rhombus without repetition of any data. Owing to the persistence of human vision this provides the impression of a raster which drifts at the velocity v.

FIG. 4b shows the reproduction of the same rhombus but each line is repeatedly reproduced on a line by line basis in order to avoid drifting lines. It can be seen that the edge sharpness of image reproduction suffers substantially.

FIG. 4c is a reproduction of the same rhombus with line repetition in half field configuration. It can be seen that the image reproduction is substantially improved compared with that of FIG. 4b.

FIG. 4d shows an image reproduction in which data is repeated frame by frame. It can be seen that image reproduction sharpness becomes perfect and surprisingly the image does not flicker.

FIG. 5 is an exemplified embodiment of a symmetrical image splitter and shows its method of operation. The lens 2 projects the object 1 (without image splitter) in inverted form on the image plane 18. The image splitter contains a partially transmissive and a partially reflective plane 14. In the absence of the mirrors 15, the said plane projects the object 1 on the image plane 17. The two images 4 and 5, which are symmetrical with respect to each other and are situated in the plane 11, are generated by the mirrors 15 and 16.

The two images 4 and 5 have a symmetry axis 12 which corresponds to the straight line intersection of the image plane 11 with the semi-transmissive plane 14.

To prevent the two images 4 and 5 overlapping, part of the semi-transmissive plane 14, which is not used as a partially reflecting mirror and is nearest to the image plane 11, is embodied as being opaque and light-absorbent.

The mirrors 15 and 16 can be replaced by totally reflecting prism surfaces.

FIG. 6 shows another embodiment of the image splitter, disposed on a camera tube, and a block circuit diagram of the connected signal processing system.

To generate the two images, this embodiment makes use of a geometrically-optically homogeneous, partially transmissive, reflective plane 14 which partially transmits each incident pencil of rays without deflection and partially reflects it and employs two planes 15 and 16, each effecting reflective deflection. The partially transmissive plane 14 is disposed between the longer perpendicular surfaces and the reflecting planes 15 and 16 are disposed on the hypotenuse surfaces of two 30.degree./60.degree./90.degree. prisms 19 and 20 which are fixedly joined to each other, for example by means of epoxy resin.

Two raster filters 27 and 28 are disposed on the one hand between the prisms 19 and 20 and on the other hand between the planes on which the images 4 and 5 are generated.

The coded video signals 9 and 10 are produced when the images 4 and 5 are scanned by the electron beam 32 in the camera tube 46. The two raster filters 27 and 28 generate striated images in which the striae correspond to two different colour separations because of the locally different transmissivity of the raster filters 27,28. Scanning by means of the electron beam 32 produces a video signal which contains the image information of polychrome colour separations is frequency multiplex-nested form. The electron beam 32 is moved by the deflection signals from the synchronizing centre 33 over the striated polychrome images. Said beam thus generates the video signals 9 and 10, each of which corresponds in frequency multiplex to two colour separations. The frequency selective system 29 splits each of the frequency multiplex signals 9 and 10 into its components. Each of these components is connected by the switches 30 and 31 to the four inputs U.sub.1 *, U.sub.2 *, U.sub.3 * and U.sub.4 * of the matrix 35 in accordance with the scanning of images 4 and 5. The matrix 35 therefore generates the video signals R,G and B which at each instant correspond to the colour separation that has just been scanned.

To ensure flicker-free reproduction of monochrome places of the original image each of two images, filtered in such striated manner, are projected on the photo-sensitive coating of the camera tube 46 so that each of the two decoded electric video signals corresponds to a pair of complementary colours.

To ensure that polychrome parts of the original image, i.e. of the object 1, are also reproduced with the least amount of flicker, the system employs striation filters which are such that all complementary colours are situated on the same curve with a constant, relative brightness reference value for body colours with optimum colours and their studio light illumination, as can be seen by reference to FIG. 7.

To provide the best subjective colour reproduction striation filters are used as illustrated in FIG. 8, with filter coefficients which correspond at least approximately to the primary colours U.sub.1, U.sub.2, U.sub.3 and U.sub.4 of human vision and correspond substantially to the points of intersection of the small u axis and v axis of television with the colour triangle (R.sub.e), (G.sub.e), (B.sub.e) used in television. The u and v axes are defined by

and E.sub.y =0.3.multidot.E.sub.R +0.59.multidot.E.sub.G +0.11.multidot.E.sub.B.

The striation filters 27 and 28 of FIG. 6 or 9 must be disposed close to the image plane. It is however also possible to place a two-coloured striation filter into the intermediate image plane 23 and to dispose two differently coloured, optically homogeneous light filters between the prisms 18 and 19 and the image planes in which the image 4 and 5 is produced.

Substantial geometric distortion, provided it occurs only symmetrically, can be tolerated, because of the symmetrical arrangement of the images 4 and 5 and by virtue of the image scanning by the electron beam 32.

In practice it was hitherto impossible to employ video amplifier tubes for colour reproduction because of their severe geometrical distortion. As indicated in FIG. 9, they can be used because of the symmetry properties of the present method. The video amplifier 48 amplifies the images 4 and 5 generated by the fibre glass optics 47 on the photosensitive coating 11. The images 50 and 51, which can then be scanned in time sequence by the electron beam 32, and are still disposed symmetrically with respect to the tube axis, are thus generated on the intermediate stratum 49.

FIGS. 9 and 10 indicate storage of the video signals 9 and 10 by the memory devices 41 and 42 and 43 respectively, to avoid flicker effects.

It is also possible to generate two images as shown in FIG. 1 and to store the video signals, corresponding to one raster line, for the duration of a line period in an intermediate memory.

It is also possible to generate two images as illustrated in FIG. 1 and to store the video signals, corresponding to one half field scan, in an intermediate memory, at least for the duration of one half field.

Memory systems have a limited resolution. Image reproduction therefore appears to be less sharp. This can be improved by the insertion of a crispening circuit 44 upstream of the green signal output.

In addition to the memories 41, 42 or 43, image repetition calls for a switching mechanism 39 or 40. Switching must be performed in synchronism and in phase with the scanning of images 50, 51. According to FIG. 9, a coincidence system 38 is provided to obtain synchronous switching in phase. In conjunction with interlace scanning this permits unequivocal coordination of the synchronizing signal J from the coincidence of the horizontal and vertical synchronizing signals H and V. The last-mentioned are obtained by the synchronizing separator 37 from the video signals 9 and 10 respectively.

Owing to unavoidable band jitter, a video signal obtained from a program memory 36 no longer has a precise time axis. The memory time is altered by the horizontal pulse H and the stored video signal is called up from the memory so that it coincides precisely in time with the direct video signal, so that the video signals stored in the intermediate memories 41, 42 and 43 respectively nevertheless generate an image which coincides with the directly generated image reproductions.

Since a relatively small frequency band is required during transmission of the optical encoded video data (image 4 in FIG. 9), if this corresponds only to a red and a blue colour separation, then during transmission of, for example, green video data (image 5 in FIG. 9) it is possible for additional data, more particularly several speech data, to be co-transmitted during the transmission time of the first image 4. This is due to the fact that only a third of the band width required for transmission of green data is required for transmitting red data and only an eighth of the band width required for the transmission of green data is sufficient for the transmission of blue data. An auxiliary carrier, higher than the highest red data to be transmitted, can therefore be amplitude modulated with blue data and can be transmitted together with red data. The band width of this combined signal is only slightly wider than half the band width of the green signal.

Similar conditions obtain for digital transmission.