JP2010041251A - Video signal processor, imaging apparatus, and video signal processing method - Google Patents

Abstract

An object of the present invention is to solve the problem of false color generation when a hue signal is obtained from a single-plate image sensor output by interpolation.
A first primary color signal is generated at a predetermined pixel position from a video signal obtained by an image sensor, a second primary color signal interpolation value is generated at a predetermined pixel position, and a hue signal at the predetermined pixel position is generated. Applicable when generating. The processing configuration includes a second primary color signal interpolation value created from the second primary color signals at a plurality of positions used in the hue signal generation process, and the second primary color signal or second second existing at the position where the hue signal is to be generated. The similarity is determined from the absolute value of the difference from the primary color signal interpolation value. Then, the reliability of the hue signal is determined based on the determined similarity. Further, based on the determined reliability of each position, an addition ratio of hue signals at a plurality of positions is determined, and an interpolation process for obtaining a hue signal at an arbitrary position is performed.
[Selection] Figure 5

Description

The present invention relates to a video signal processing device and a video signal processing method applied to a case where a hue signal is generated from a primary color signal obtained by a single-plate image sensor provided in an imaging device such as a video camera, and the video signal processing. The present invention relates to an imaging apparatus to be performed.
When obtaining a video signal using a single-plate image sensor, a color signal obtained based on the arrangement of the color filters on the image sensor is used to generate a color signal at an insufficient pixel position by interpolation processing. The present invention relates to a technique to be applied during the interpolation process.

  There are various types of imaging devices such as a CCD (Charge Coupled Device) type image sensor, a CMOS (Complementary Metal Oxide Semiconductor) type image sensor, etc., provided in an imaging device such as a video camera. A so-called single-plate type imaging apparatus using one (one) of these image sensors has a configuration shown in FIG. 9, for example.

  In FIG. 9, an imaging apparatus 200 includes an imaging lens 201, an imaging element 202, an analog / digital converter (A / D converter) 203, a correction system circuit 204, a color separation circuit 205, and a video signal processing circuit. 206.

  The imaging lens 201 forms image light of a subject on the imaging surface of the imaging element 202. The image sensor 202 is configured by a CDD image sensor, a CMOS image sensor, or the like, and includes a plurality of two-dimensionally arranged photoelectric conversion elements. Each photoelectric conversion element constitutes a pixel, and the position occupied by each photoelectric conversion element on the imaging surface corresponds to the pixel position. A color filter 202a is arranged on the two-dimensionally arranged pixels of the image sensor 202, and a signal value of one color per pixel is obtained.

Here, the color filter 202a is, for example, an array of red (R), green (G), and blue (B) that functions as an image sensor with a special Bayer array.
That is, as shown in FIG. 10A, the color filter 202a includes, for example, a first line in which green and red are alternately arranged, and a second line in which green and blue are alternately arranged, with green being a checkered pattern. Are arranged alternately in a vertical direction so as to form a checkered pattern. In such an array, the green (G) signal is extracted as it is, and the red (R) signal is output by averaging the red data of two pixels adjacent in the horizontal direction. As a configuration. Similarly, for the blue (B) signal, the blue data of two pixels adjacent in the horizontal direction are added and averaged and output.
Therefore, as shown in FIG. 10B, the pixel arrangement in the special Bayer arrangement as seen from the output state of the image sensor 202 is one pixel for each horizontal line with respect to green data in each horizontal line. The placement position shifts.
The red data is obtained at the same position as the green data by the averaging of two pixels. The blue data is also obtained at the same position as the green data by the averaging of two pixels. The red data and blue data are output once every four pixels.
As shown in FIG. 10, the red data and the blue data are averaged over two pixels, so that the SN ratio of the red data and the blue data can be improved and the sensitivity can be improved.

  Returning to the description of FIG. 9, the A / D converter 203 converts the output signal of each pixel of the image sensor 202 from an analog signal to a digital signal, and converts the red, green, and blue color data R, G, and B into Output. The correction system circuit 204 performs optical system correction such as shading correction and imager correction such as defect correction on the color data R, G, and B output from the A / D converter 203.

  The color separation circuit 205 responds to each pixel position of the image sensor 202 by interpolation processing from the red, green, and blue color data R, G, and B that are output from the correction system circuit 204 and are spatially out of phase. Red, green, and blue color data R, G, and B are obtained. The video signal processing circuit 206 performs comma correction, Takashiro enhancement, etc. on the red, green, and blue color data R, G, and B obtained by the color separation circuit 205, and then the luminance signal Y and the red difference signal Cr. (R−Y) and the blue color difference signal Cb (B−Y) are generated and output.

  The operation of the imaging apparatus 200 shown in FIG. 9 will be described. The subject image light is incident on the imaging surface of the imaging element 202 through the imaging lens 201, and an optical image of the subject is formed on the imaging surface. In the imaging device 202, imaging processing is performed in a state where an optical image of the subject is formed on the imaging surface, and an output signal (red signal, green signal or red signal) having a value corresponding to the image light of the subject. A blue signal) is obtained. An output signal of each pixel output from the image sensor 202 is converted from an analog signal to a digital signal by the A / D converter 203 and then supplied to the correction system circuit 204.

  In the correction system circuit 204, optical system correction such as shadedder correction and imager correction such as defect correction are performed on the color data R, G, and B output from the A / D converter 203. The color data R, G, B after being corrected by the correction system circuit 204 is supplied to the color separation circuit 205. The color separation circuit 205 generates a green difference signal from the Bayer-type arrayed red, green, and blue color data R, G, and B, and interpolates the red data R and the blue data B using the green difference signal. Red, green, and blue color data R, G, and B of a full color image are obtained.

  The spatially in-phase red, green and blue color data R, G and B obtained by the color separation circuit 205 are supplied to the video signal processing circuit 206. In this video signal processing circuit 206, the red, green, and blue color data R, G, and B obtained by the color separation circuit 205 are subjected to comma correction, Takashiro enhancement, and the like. Based on the green and blue color data R, G, and B, a luminance signal Y and color difference signals Cr and Cb are generated and output. Up to this point, the overall processing operation as the imaging apparatus has been described.

Next, the processing state in the color separation circuit 205 required when processing the output of the image sensor of the color filter array of the special Bayer type array shown in FIG. 10 will be described.
As described above, the color separation circuit 205 creates a green difference signal from the red, green, and blue color data R, G, and B in the special Bayer arrangement, and uses the green difference signal to generate the red data R and blue data. Interpolation of B is performed to obtain color data R, G, B of a full color image in which each pixel has the same phase. In the following description, two types of green difference signals necessary for processing in the color separation circuit are referred to as green difference signals U and V. The green difference signal U is a signal indicated by the difference between the blue signal and the green signal (i.e., the BG signal), and the green difference signal V is a signal indicated by the difference between the red signal and the green signal (i.e., RG). Signal).

As a process for obtaining the green difference signals U and V, conventionally, for example, a color difference signal at each pixel position is obtained by a linear interpolation process according to the distance. FIG. 11 is a diagram showing the state of this linear interpolation.
FIG. 11A shows a part of a pixel array (a line in which G and R pixels are arrayed) of a horizontal line in which an image sensor is provided. As described above, the output red pixel data R is as described above. Is data (R1, R2, R3,...) Obtained by averaging the adjacent red data.
FIG. 11B shows the interpolated green data (G1, G2, G3, G4, G5, G6...) At the pixel positions without the green filter, which are estimated by interpolating the green data from the image sensor. .
FIG. 11C shows a hue (color difference) signal obtained from the red data and the interpolated green data obtained as described above.
That is, the green difference signal V at the red output position R1 is obtained by the calculation of R1- (G1 + G2) / 2 using the red signal R1 at that position and the interpolated green signals G1 and G2 on both sides of R1. Similarly, the green difference signal V at the red output position R2 is obtained by the calculation of R2- (G3 + G4) / 2 using the red signal R2 at that position and the interpolated green signals G3 and G4 on both sides of R2. .

Assume that the level of the hue signal obtained in this way is the signal level indicated by the solid line arrow in FIG. The green difference signal V is obtained at the positions of the red output pixels R1, R2, and R3. At other pixel positions, processing for obtaining is performed by interpolation.
As processing to be obtained by this interpolation, as shown in FIG. 11D, the color difference signal level obtained by actual calculation is shown by a broken line at each pixel position so as to be a level indicated by a broken line. This is a process of performing so-called linear interpolation with the level of the arrow.

Patent Document 1, Patent Document 2 and Patent Document 3 output from an image sensor in which a color filter in which red (R), green (G), and blue (B) are arranged in a Bayer type or the like is arranged on the imaging surface. Describes a technique for obtaining spatially in-phase red, green, and blue color data corresponding to each pixel position of an image sensor from red, green, and blue color data that are spatially out of phase with each other. Yes.
Tokuma 2005-217478 Tokuma 2006-174485 Tokuma 2002-027487

When the hue signal is generated by the linear interpolation shown in FIG. 11 described above, an inappropriate hue signal may be generated depending on the picture pattern. This will be described with reference to FIG.
First, as shown in FIG. 12 (a), if there is a pixel array on the image sensor, and the luminance level fluctuates greatly at one of the two lines indicated by the color jump and boundary line on the pixel. To do. In FIG. 12A, R, G, and B are pixel colors, and numbers such as R11 indicate pixel numbers.
In order to obtain the hue signal of the pixel R33 at the boundary position where the luminance level changes in a state where there is such a luminance change, as shown in FIG. 12B, the signals of the surrounding pixels are required. Become.

FIG. 12C shows a change state of the red signal R and the green signal G in the vicinity of the pixel R33 on a horizontal line where the pixel R33 is located.
As for the red signals R31 and R33, as shown in FIG. 12C, an average signal of both levels obtained by averaging is output. On the other hand, for the green signals G32 and G34, a green signal is obtained directly from these pixels, and the original signal level indicated by a circle in FIG. 12C is detected.
Here, if there is a large level fluctuation between the pixel G32 and the pixel R33, a large error occurs in the estimated value at the location where the level fluctuation occurs.

  Accordingly, as shown in FIG. 13, a hue signal (color difference signal) is generated using a red output value having a large error in addition of the red signal R31 and the red signal R33 due to level fluctuations and a green interpolation signal having a large estimation error. As a result, a signal having a level different from the original hue signal is obtained.

  In this way, when there is a large error in the hue signal estimated at each position, a so-called false color that becomes a color different from the actual color at the contour portion where the luminance greatly varies when an actual image is displayed. Will occur. As the false color, for example, assuming an image with a white object in a black background, the boundary between the white object and the background should change from white to black. Is a phenomenon that appears at the boundary.

  The present invention has been made in view of these points, and an object of the present invention is to solve the problem of false colors.

The present invention provides a first primary color at a predetermined pixel position from a video signal obtained by an imaging device in which color filters of primary colors of green, red and blue are alternately arranged in a predetermined arrangement on a plurality of pixels arranged two-dimensionally. This is applied when the second primary color signal interpolation value is generated at a predetermined pixel position and the hue signal at the predetermined pixel position is generated.
The processing configuration includes a second primary color signal interpolation value created from the second primary color signals at a plurality of positions used in the hue signal generation process, and the second primary color signal or second second existing at the position where the hue signal is to be generated. The similarity is determined from the absolute value of the difference from the primary color signal interpolation value. Then, the reliability of the hue signal is determined based on the determined similarity. Further, based on the determined reliability of each position, an addition ratio of hue signals at a plurality of positions is determined, and an interpolation process for obtaining a hue signal at an arbitrary position is performed.

  Thus, by determining the addition ratio for interpolation based on the reliability of each position, for example, when the signal level fluctuates relatively suddenly, the reliability decreases. For signals with low reliability, it is possible to take measures to lower the addition ratio. On the other hand, for signals with high reliability, the addition ratio may be increased.

  According to the present invention, it is possible to control the addition ratio of a plurality of signals for obtaining a hue signal based on the calculated reliability, and effectively suppress the occurrence of false colors when there is a sudden color variation. can do.

Hereinafter, a first embodiment of the present invention will be described with reference to FIGS.
FIG. 1 is a diagram illustrating an overall configuration example of an imaging apparatus 100 according to the present embodiment.
Referring to FIG. 1, the imaging apparatus 100 includes an imaging lens 101, an imaging element 102, an analog / digital converter (A / D converter) 103, a correction system circuit 104, a color separation circuit 105, a video signal. And a processing circuit 106.

  The imaging lens 101 forms image light of a subject on the imaging surface of the image sensor 102. The image sensor 102 is configured by a CDD image sensor, a CMOS image sensor, or the like, and includes a plurality of two-dimensionally arranged photoelectric conversion elements. Each photoelectric conversion element constitutes a pixel, and the position occupied by each photoelectric conversion element on the imaging surface corresponds to the pixel position. A color filter 102a is arranged on the two-dimensionally arranged pixels of the image sensor 102, and a signal value of one color is obtained per pixel.

For example, as shown in FIG. 3, the color filter 102a is a color filter in which red (R), green (G), and blue (B) are arranged in a special Bayer type.
In the special Bayer array here, as already described with reference to FIG. 10, in the horizontal line with the red pixel, the output position of the red pixel is the same position as the green pixel by addition averaging, and there is a blue pixel. In the horizontal line, the output position of the blue pixel is the same position as the green pixel by addition averaging.

  The A / D converter 103 converts the output signal of each pixel of the image sensor 102 from an analog signal to a digital signal, and outputs red, green, and blue color data R, G, and B. The correction system circuit 104 performs optical system correction such as shading correction and imager correction such as defect correction on the color data R, G, and B output from the A / D converter 203.

  The color separation circuit 105 obtains a hue signal from the red, green, and blue color data R, G, and B that are output from the correction system circuit 104 and are spatially shifted from each other, and performs interpolation using the hue signal. Thus, red, green, and blue color data R, G, and B that are spatially in phase are obtained. In the present embodiment, the obtained hue signals are a green difference signal U indicated by the difference between the blue signal and the green signal, and a green difference signal V indicated by the difference between the red signal and the green signal. Processing for obtaining the green difference signals U and V is performed by the conversion unit 111 in the color separation circuit 105. Then, the interpolation unit 112 uses the green difference signals U and V generated by the conversion unit 111 and the color data R, G, and B supplied from the circuit 104 in the previous stage, so that the red color is spatially in phase. , Green and blue color data R, G, and B are obtained. When the interpolation unit 112 performs processing for obtaining red, green, and blue color data R, G, and B that are spatially in phase, in this example, the reliability of the color difference signal generated by the reliability generation unit 113 is used. Interpolation processing is performed by changing the addition ratio using data. Details of the processing configuration using the reliability generation unit 113 will be described later.

  The spatially in-phase red, green and blue color data R, G and B obtained by the color separation circuit 105 are supplied to the video signal processing circuit 106. In the video signal processing circuit 106, comma correction, Takashiro emphasis, and the like are performed on the red, green, and blue color data R, G, and B obtained by the color separation circuit 105. Based on the green and blue color data R, G, and B, a luminance signal Y and color difference signals Cr and Cb are generated and output. The video signal processing circuit 106 performs video signal processing such as comma correction and Takashiro enhancement on the supplied data and then outputs the processed data. The output video signal is supplied to a display unit or a recording unit (not shown), and processing such as display and recording is performed.

FIG. 2 is a diagram showing a more detailed configuration example of the color separation circuit 105 of FIG.
The signal obtained at the input terminal 10 is supplied to the RGB separation unit 11 and separated into red, green, and blue color data R, G, and B. The color data R, G, and B separated by the RGB separation unit 11 are supplied to the hue signal generation unit 12.
The hue signal generation unit 12 calculates the green difference signal V indicated by the difference between the red signal and the green signal at the position where the red pixel of the image sensor exists, and also the blue signal and the green signal at the position where the blue pixel exists. The green difference signal U indicated by the difference is calculated and each color difference signal is output.
The output green difference signals U and V are supplied to the correction unit 13 so that color signals at positions other than the respective pixel positions are generated by interpolation, and red, green, and blue of the same phase are spatially generated. Color data R, G, and B are obtained. The obtained color data R, G, B are output from the output terminal 20 to the subsequent circuit.

  Then, the color difference signal generated by the hue signal generation unit 12 is supplied to the similarity generation unit 14, and adjacent color difference signals are compared to generate similarity data. The generated similarity data is supplied to the reliability determination unit 15 to determine the reliability coefficient of the color difference signal at each position. The reliability coefficient of each determined pixel position is supplied to the correction unit 13 and used for processing when generating a color difference signal by interpolation. Here, a color difference signal is generated by interpolation in a process called reliability coefficient α blend.

The flowchart of FIG. 4 is a diagram showing a specific example of the processing state in the configuration of FIG.
The example of FIG. 4 shows a processing example for generating the green difference signal V indicated by the difference between the red signal and the green signal.
First, using the red data R and the adjacent interpolated green data G on both sides, a green difference signal V is generated at the red pixel position (step S11). The interpolated green data G is obtained by interpolating the green data G at the pixel position where no green filter exists from the interpolation of the green data at the adjacent positions. The interpolated green data G1, G2,... Described with reference to FIG. It corresponds to.
Next, the similarity is determined from the absolute value of the difference between the average value of the two green data G used to generate the green difference signal V and the actual green data G of the green pixel (step S12). Here, it is assumed that the similarity is higher as the absolute value of the difference is smaller, and the similarity is lower as the absolute value of the difference is larger.

Then, the similarity degree at the determined pixel position is compared with the similarity degree at the adjacent pixel position to calculate a similarity coefficient. Here, the two similarities are compared to determine which pixel position has high reliability, and the reliability is calculated from the ratio of the two similarities (step S13). The reliability is calculated as a reliability coefficient α.
Using the reliability coefficient α at each pixel position calculated in this way, the addition ratio of the color difference signal at each pixel position is determined, and so-called α blending is performed from the output of the hue signal generation unit 12. The color difference signal at the missing pixel position is calculated (step S14).

FIG. 5 is a diagram showing a specific example of color difference signal generation processing.
FIG. 5A shows a part of a horizontal pixel array (a line in which G and R pixels are arranged) having an image sensor, and because of the special Bayer array, the output red pixel data R is , Data obtained by adding and averaging adjacent red data (R1, R2, R3...).
FIG. 5B shows interpolated green data (G1, G2, G3, G4, G5, G6...) At pixel positions without a green filter, estimated by interpolating green data from the image sensor. .
FIG. 5C shows a hue (color difference) signal obtained from the red data and the interpolated green data obtained as described above.
For example, the green difference signal V1 at the red output position R1 is obtained by the calculation of R1- (G1 + G2) / 2 using the red signal R1 at that position and the interpolated green signals G1 and G2 on both sides of R1. The green difference signal V2 at the red output position R2 is obtained by the calculation of R2- (G3 + G4) / 2 using the red signal R2 at that position and the interpolated green signals G3 and G4 on both sides of R2. The color difference signal V3 at the red output position R3 is obtained by the calculation of R3- (G5 + G6) / 2 using the red signal R3 at that position and the interpolated green signals G5 and G6 on both sides of R3.

Here, when attention is paid to the green difference signal V1 at the pixel position R1 and the green difference signal V2 at the pixel position R2, and the respective similarities are calculated, the state shown in FIG. 5D is obtained.
The similarity h1 of the green color difference signal V1 is h1 = | G1− (G1 + G2) / 2 |. In calculating the similarity h1 of the green difference signal V1, the signal levels of the pixel positions G1 and G2 adjacent to the pixel position R1 are used.
Similarly, the similarity h2 of the green difference signal V2 is h2 = | G3− (G3 + G4) / 2 |. For the calculation of the similarity h2 of the green difference signal V2, the signal levels of the pixel positions G3 and G4 adjacent to the pixel position R2 are used.

The reliability coefficient α is in the state shown in FIG.
That is, the reliability coefficient α1 of the green difference signal V1 is α1 = h2 / (h1 + h2). The reliability coefficient α2 of the color difference signal V2 is α2 = 1−α1 = h1 / (h1 + h2).

Thus, when the reliability coefficient is obtained, a green difference signal is calculated at a pixel position (position between R1 and R2) between the green difference signal V1 and the green difference signal V2.
For example, as shown in FIG. 5F, as the color difference signal C3 at the position of the green pixel G3,
Calculated as C3 = α1 · V1 + (1−α1) · V2.

In addition to calculating the reliability coefficient α sequentially by such calculation, a correspondence table of the two similarities h1 and h2 and the similarity α is prepared, and multiple levels of reliability can be obtained by referring to the table. A configuration may be selected from the coefficient α.
In addition, the interpolation processing by α blending using the reliability coefficient α shown here is an example, and other interpolation processing may be used as long as the interpolation processing is performed using the reliability.

By calculating the color difference signal in this way, the error in the color difference signal at the pixel position where the color signal greatly varies can be reduced, and the occurrence of false colors can be reduced.
That is, for example, as shown in FIG. 6A, it is assumed that the red signal R and the green signal G change in a certain horizontal line. In this example, it is assumed that the signal level suddenly increases at a specific position.
It is assumed that the green difference signals Vx, Vy, Vz detected at the position of the red pixel at this time are in the state shown in FIG. As can be seen from FIG. 6B, the color difference signal level also fluctuates greatly at the level change point.
In such a case, the similarity and reliability shown in FIG. 5 are calculated, so that the reliability of the signal after the level change is lowered, and when calculating the color difference signal interpolated by α blending, Signals with high reliability are added at a high ratio, and signals with low reliability are added at a low ratio. Accordingly, a color difference signal close to the original signal level can be obtained, and the generation of a false color near the change point can be effectively prevented. In particular, a great effect can be obtained when processing the output of an image sensor having a special Bayer arrangement in which red pixels and blue pixels are thinned as shown in FIG. Although the example applied to the special Bayer array has been described here, there is an effect of generating false colors also when applied to an image sensor having a Bayer array other than the special Bayer array.

Next, a second embodiment of the present invention will be described with reference to FIGS. 7 and 8, the same reference numerals are given to the portions corresponding to FIGS. 1 to 6 described in the first embodiment.
FIG. 7 showing the overall configuration of the imaging apparatus 100 will be described. In the present embodiment, the color separation circuit 105 ′ is changed from the example of FIG. 1 of the first embodiment.

  In the present embodiment, an interpolation unit 114 that performs linear interpolation is provided as the color separation circuit 105 ′ in addition to the interpolation unit 112 that uses the reliability coefficient α described in the first embodiment. Then, according to the reliability generation state in the reliability generation unit 113, the outputs of the two interpolation units 112 and 114 are switched by the changeover switch 115 and supplied to the video signal processing circuit 106. As the linear interpolation processing in the interpolation unit 114, for example, as already described in FIG. 9, a plurality of pixel signals are added at a ratio corresponding to the distance from the pixel position.

The flowchart of FIG. 8 is a diagram showing a processing state in the color separation circuit 105 ′.
In the flowchart of FIG. 8, the same step numbers are assigned to the same processes as those in the flowchart of FIG. 4.
First, using the red data R and the adjacent interpolated green data G on both sides, a green difference signal V is generated at the red pixel position (step S11). The interpolated green data G is obtained by interpolating the green data G at the pixel position where no green filter exists from the interpolation of the green data at the adjacent positions. The interpolated green data G1, G2,... Described with reference to FIG. It corresponds to.
Next, the similarity is determined from the absolute value of the difference between the average value of the two green data G used to generate the green difference signal V and the actual green data G of the green pixel (step S12).

  Next, the determined similarity is compared with the similarity between adjacent positions (step S15). In this comparison, it is determined whether or not the difference between the two similarities is equal to or less than a preset threshold value (step S16). Here, when the difference between the two similarities is equal to or less than the threshold value, the changeover switch 115 is set as the output side of the interpolation unit 114 that performs linear interpolation, and the color difference signal by linear interpolation is adopted and the color difference signal is output. (Step S17).

  On the other hand, when the difference between the two similarities exceeds the threshold value, the changeover switch 115 is set as the output side of the interpolation unit 112. In this way, the color similarity signal is obtained by calculating the similarity coefficient in step S13 and the color difference signal of the α blend in step S14.

In the case of the second embodiment, when the difference in similarity is large, interpolation processing using α blend is performed, and when the difference in similarity is small, conventional linear interpolation processing is performed. It will be.
Interpolation processing using alpha blending is highly effective in preventing false colors, but there is a possibility that the image quality may be deteriorated, for example, in the case of a picture with a continuous high frequency. For this reason, as in the configuration of FIG. 7, in combination with linear interpolation, by performing an interpolation process using α blend when there is a difference in similarity, except for a situation where there is a high possibility that false colors will occur. Linear interpolation is performed, and the image quality can be further improved.

  Note that the example of switching based on the difference in similarity shown in the processing of FIG. 8 is an example, and other interpolation processing may be switched based on other determinations.

In addition, the pixel arrangement and the like described in each embodiment described above are examples, and are not limited to those described in each embodiment. The hue signal (color difference signal) generation process of the present invention can be applied to various pixel arrangements.
In the processing described with reference to FIG. 5 and the like in the embodiment, the processing using the green difference signal V indicated by the difference between the red signal and the green signal has been described. The same processing may be performed using the green difference signal U indicated by the difference of.

1 is a block diagram illustrating a configuration example of an imaging apparatus according to a first embodiment of the present invention. It is a block diagram which shows the example of a color separation structure by the 1st Embodiment of this invention. It is explanatory drawing which shows the example (example of a special Bayer arrangement | sequence) of the pixel arrangement | sequence of an image sensor. It is a flowchart which shows the example of a color separation process by the 1st Embodiment of this invention. It is explanatory drawing which shows the example of the color separation process state by the 1st Embodiment of this invention. It is explanatory drawing which shows the example of the level of the hue signal corresponding to the example of a change of the red signal R and the green signal G by the 1st Embodiment of this invention, and the change. It is a block diagram which shows the structural example of the imaging device by the 2nd Embodiment of this invention. It is a flowchart which shows the example of a color separation process by the 2nd Embodiment of this invention. It is a block diagram which shows the structural example of the conventional imaging device. It is explanatory drawing which shows the example (example of a special Bayer arrangement | sequence) of the pixel arrangement | sequence of an image sensor. It is explanatory drawing which shows the example of formation of the hue signal by linear interpolation. It is explanatory drawing which shows the example of the state which the estimation error of a color signal generate | occur | produces. It is explanatory drawing which shows the example of generation | occurrence | production of a false color signal.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Input terminal, 11 ... RGB separation part, 12 ... Hue signal generation part, 13 ... Correction part, 14 ... Similarity generation part, 15 ... Reliability determination part, 20 ... Output terminal, 100 ... Imaging device, 101 ... Imaging Lens 102, imaging device 103 analog / digital converter 104 104 correction system circuit 105, 105 ′ color separation circuit 106 video signal processing circuit 111 conversion unit 112 112 interpolation unit 113 trust Degree generator, 114 ... interpolator, 115 ... switch

Claims (6)

A first primary color signal is generated at a predetermined pixel position from a video signal obtained by an image sensor in which green, red and blue primary color filters are alternately arranged in a predetermined arrangement on a plurality of pixels arranged two-dimensionally. A hue signal generation unit that generates a second primary color signal interpolation value at a predetermined pixel position and generates a hue signal at the predetermined pixel position;
A reliability determination unit that determines the reliability of the hue signal based on the similarity determined by the similarity determination unit;
A video signal processing apparatus comprising: an interpolation processing unit that determines a hue signal at an arbitrary position by determining an addition ratio of hue signals at a plurality of positions based on the reliability of each position determined by the reliability determination unit . The interpolation processing unit multiplies a hue signal at a plurality of positions by a coefficient of reliability at each position, using the reliability determined by the reliability determination unit as a coefficient, adds the multiplication signal, and adds an arbitrary position. The video signal processing apparatus according to claim 1, wherein a hue signal is obtained. The video according to claim 1, wherein the reliability determination unit determines that the reliability is higher as the absolute value of the difference determined by the similarity determination unit is smaller, and the reliability is lower as the absolute value of the difference is larger. Signal processing device. A linear interpolation processing unit different from the interpolation processing unit using reliability,
The linear interpolation processing unit obtains a hue signal at an arbitrary position by linear interpolation of a plurality of hue signals generated by the hue signal generation unit,
When the similarity difference for the nearby hue signal determined by the similarity determination unit is equal to or less than a threshold, the hue signal obtained by the linear interpolation processing unit is employed,
The video signal processing apparatus according to claim 1, wherein when the threshold value is exceeded, a hue signal obtained by the interpolation processing unit using reliability is employed. An image pickup unit having an image pickup element in which color filters of primary colors of green, red, and blue are alternately arranged in a predetermined arrangement on a plurality of pixels arranged two-dimensionally;
A hue for generating a first primary color signal at a predetermined pixel position from a video signal obtained by the imaging unit, generating a second primary color signal interpolation value at a predetermined pixel position, and generating a hue signal at the predetermined pixel position A signal generator;
A second primary color signal interpolation value created from the second primary color signals at a plurality of positions used for the hue signal generation, and a second primary color signal or an interpolation value of the second primary color signal existing at a position where the hue signal is to be generated; A similarity determination unit that determines the similarity from the absolute value of the difference between,
A reliability determination unit that determines the reliability of the hue signal based on the similarity determined by the similarity determination unit;
An imaging apparatus comprising: an interpolation processing unit that determines an addition ratio of hue signals at a plurality of positions based on the reliability of each position determined by the reliability determination unit and obtains a hue signal at an arbitrary position. A first primary color signal is generated at a predetermined pixel position from a video signal obtained by an image sensor in which green, red and blue primary color filters are alternately arranged in a predetermined arrangement on a plurality of pixels arranged two-dimensionally. A hue signal generation process for generating a second primary color signal interpolation value at a predetermined pixel position and generating a hue signal at the predetermined pixel position;
A second primary color signal interpolation value created from the second primary color signals at a plurality of positions used for the hue signal generation, and a second primary color signal or an interpolation value of the second primary color signal existing at a position where the hue signal is to be generated; Similarity determination processing for determining similarity from the absolute value of the difference between,
A reliability determination process for determining the reliability of the hue signal based on the similarity determined in the similarity determination process;
A video signal processing method for performing interpolation processing for determining a hue signal at an arbitrary position by determining an addition ratio of hue signals at a plurality of positions based on the reliability of each position determined in the reliability determination process.

Images