Cameras

Details Cameras Cameras

1997-10-01

2003-10-14

Moe, Aung S. (Department: 2612)

Television

Camera, system and detail

With single image scanning device supplying plural color...

C348S273000, C348S217100

Reexamination Certificate

active

06633333

ABSTRACT:

FIELD OF THE INVENTION
This invention relates to cameras.
The invention is particularly applicable to cameras which produce colour video outputs (including false colour) such as may be fed to a colour display.
The invention is also particularly applicable to such cameras in which a single radiation-sensitive sensor is used to produce the colour video outputs. Typically the sensor is scanned in a raster or sampled as in a solid state array.
BACKGROUND OF THE INVENTION
Various proposals have been made to solve the problem of how to reproduce colours using a single sensor. One of the earliest proposals was Baird's system of aiming a monochrome television camera through a rotating filter, successive segments of which passed red, blue and green light. The receiver used a similar rotating filter synchronised with the one at the camera.
U.S. Pat. Nos. 2,733,291 and 3,378,633 disclose the concept of exposing monochrome light-sensitive devices (a vidicon in the case of the former and cine film in the case of the latter) to produce colour television outputs, each exposing the light-sensitive device through a filter consisting of stripes of two colours which repeat in the scanning direction at different frequencies. A tuned circuit is used to select the respective colour components in the output. In the latter patent, the filter consists of alternate clear and cyan vertical stripes, overlying alternate clear and yellow stripes inclined relative to the vertical stripes. This provides a grouping of four elemental colour areas which is repeated over the entire area of the filter and is imaged over the entire area of each frame of the cine film. Where the transparent stripes overlap, the light-sensitive device produces a signal corresponding to the full luminance signal. Where cyan (which passes green and blue light) overlaps yellow (which passes red and green light), only the green component of the scene is image. Where the cyan and yellow stripes overlap the transparent stripes, the light-sensitive device images the green and blue, and green and red components of the scene, respectively. It is sufficient for the repetition frequency of the yellow and cyan stripes in the direction in which the cine film is ultimately scanned in order to produce signals for a colour television receiver to be different (for example, by making one set of stripes vertical and the other set of dimensionally identical stripes inclined), for it to be possible to extract the components in the output of the scanned cine film by means of tuned circuits. Sufficient information is provided for a low resolution luminance component to be produced, in addition to three components corresponding to the primary colours (red, green and blue).
A disadvantage of the use of tuned circuits to extract the individual components in the output of the video waveform is that only modest pictures are produced due to cross-talk (cross-luma and cross-chroma).
When each cine frame is raster scanned in U.S. Pat. No. 3,378,633, the grey scale value of successive pixels on each line corresponds to that for the respective colour components i.e. white, yellow, green, cyan, in a fixed sequence. It has been proposed (Albert Macovski—Spatial-Frequency Encoding Techniques Applied to a One-Tube Colour Television Camera, IEEE Transactions on Broadcasting, Vol BC-16, No. 4, December 1970) to sample the grey scale of each pixel on each video line corresponding to the respective colour components, rather than to discriminate between the components using tuned circuits. However, geometrical errors resulting in indefinite registration between the filter and the cine film, and between the cine film and its scanner, would make it difficult to predict which colour components any group of pixels represents.
It may be noted that this disadvantage does not apply in a consumer video camera, in which successive pixels along each line of the solid state imager are covered by gelatine filters which are yellow, cyan, green or transparent, because the filters are physically secured to the imager in this case. This means that, say, the first pixel on line
1
represents the scene imaged through a yellow filter, and accordingly outputs of the imager corresponding to the respective colour components can be derived with certainty.
However, such an arrangement would not be possible if the sensor included an image intensifier. In this case the filters would have to be positioned in front of the image intensifier rather than in front of the solid state sensor, since the output of the image intensifier is itself monochrome.
The geometrical distortion produced by an image intensifier is such that it would be impossible to predict the colour sequence for each line of the solid state imager based upon the sequence of elemental colour areas in the filter.
In the case of a single-tube colour television image pick-up apparatus exposed through a colour stripe filter, it has been proposed to compensate for non-linearity in the tube's deflection system by the use of a frame memory (GB-A-2 135 853). The frame memory stores video signals corresponding to illumination of the tube with primary colours. These stored signals are used to compensate for non-linearities in the image pick-up mode of the tube. However, the video signals, while corrected for non-linearities caused e.g. by drift, are nevertheless low resolution.
It has been proposed (“Interplex—A New Versatile Full Resolution Single-Tube Colour TV Camera System”, M Koubek, IEEE Transactions on Broadcasting Vol BC-22 No 3 September 1976, pp 30-35) to produce a high resolution single-tube camera. This is done by deriving a luminance signal which occupies a substantial proportion of the video bandwidth. A single-tube camera is exposed through a colour stripe filter, and separate colour outputs are produced, as in GB-A-2 135 853. The output of the single-tube is such a luminance signal, but the problem is that the stripe pattern is superimposed, and drastic filtering has been used in the past to remove the effect of the stripes. Because GB-A-2 135 853 uses vertical stripes, the respective colour information appears in the video bandwidth in the form of harmonics of the line frequency (which contains luminance information) and hence there is no way of separating the chrominance information and the luminance information, necessitating severe filtering of the bandwidth to produce a luminance signal. Koubek, however, uses obliquely orientated stripes, so the chrominance information is interleaved with the harmonics of the line frequency in the video bandwidth, and a comb filter is used to separate the chrominance and luminance information. This results in a luminance signal which is free of the shading pattern of the filter and thus has a bandwidth commensurate with the video bandwidth. However, the system relies on using a linear scan camera tube (i.e. good geometry). If the stripe pattern on the target became distorted, the bandwidth of the chrominance signals would increase and would no longer interleave the line harmonics.
SUMMARY OF THE INVENTION
It is an aim of the invention to permit high resolution colour video signals to be obtained using a single sensor, even if accompanied by an image intensifier, which would suffer from significant geometric distortion.
The invention provides a camera comprising a sensor for receiving radiation forming an image of a scene, filter means positioned in the path of radiation incident on the sensor, the filter means being arranged to pass different spectral regions in different spatial regions, so that different spatial regions of the sensor are exposed to radiation of different spectral regions, decoding means for producing separate outputs from the sensor corresponding to the different spectral regions, the decoding means being arranged to use stored signals derived from the sensor output corresponding to exposure of the sensor through the filter means by radiation of reference spectral regions, and a circuit for using the outputs corresponding to the different spectral regions to rem

Affiliated with

Also associated with

Rating

Cameras does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Cameras, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Cameras will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3169813