JP2003092764A - Image signal processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate moire caused when interleave compression is ununiformly applied to an image picked up by a solid-state imaging device in a columnar direction. SOLUTION: A CCD image sensor divides pixel arrays where R, G are alternately arranged by 6 pixels each and sums information electric charges by R, G included in each division and provides an output. Columnar positions of initial sampling points to which the summed pixel values (R), (G) correspond are ununiform. The signal processing circuit receiving the ununiformly compressed image obtains a pixel value at target sampling points arranged at an equal interval by interpolation arithmetic operations to provide outputs of pixel values (R)', (G)' configuring a uniformly compressed image. For example, when the target sampling point is made coincided with the initial sampling point to which the pixel value (R) is located, the (R)' is coincident with the (R), and the (G)' is calculated by a weight average arithmetic operation on the basis of two (G) values placed on and under the target sampling point by using a reciprocal of a distance between the positions and the target sampling point for a weighting coefficient.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image signal processing apparatus for generating an image compressed in a column direction in which pixels are thinned out in the column direction with respect to an original image composed of pixels arranged in a matrix. .

[0002]

2. Description of the Related Art An electronic still camera or the like is equipped with a high-resolution solid-state image pickup device having several million pixels. These cameras have a display unit for previewing an image to be captured. Since the preview image has a small screen and is mainly used for the purpose of confirming a subject, it does not require the resolution of the still image recorded on the recording medium. Further, in moving image shooting, since it is necessary to repeatedly process and record an image at a short frame rate, the process cannot catch up if the number of pixels to be processed is large. Furthermore, due to the human visual characteristics of moving images, the resolution of still images is not required.

Therefore, in the pre-view shooting and the moving picture shooting, a CCD (Char that enables driving to output an image signal in which pixels are thinned out in the column direction, that is, the vertical direction) is used.
ge Coupled Device) There is an image sensor.

Next, the processing when the column-direction compression is not performed and the conventional column-direction compression processing will be described with reference to the drawings.

FIG. 5 is a schematic diagram for explaining the processing when the compression in the column direction is not performed. Here, in the image pickup area of the CCD image sensor, for example, R (Re
A mosaic type color filter in which three colors of d), G (Green) and B (Blue) are arranged in a predetermined order is attached. In this color filter, R and G are alternately arranged in odd columns, and G and B are alternately arranged in even columns. Vertically arranged R ₁ , G ₂ , R ₃ , G ₄ , ... In FIG. 5 represent the arrangement of pixels in an odd column of the CCD image sensor and the amount of information charges generated in the pixels, and R _{2i- 1} is the second (2i
-1) The information charge amount of the R pixels arranged in the row (i is a natural number), and G _2i represents the information charge amount of the G pixels arranged in the 2i-th row of the same column.

When the compression is not performed, the information charges of each row are sequentially read from the CCD image sensor. That is, R ₁ , G ₂ , R ₃ , ... On the illustrated columns are all output. At the stage of being output from the CCD image sensor, only one of R and G signals can be obtained for each row, but in the signal processing circuit for this output, both R and G signals are assigned for each row by interpolation processing. Therefore, the resolution in the vertical direction can be improved. Specifically, R _{2i + 1}
When the R signal relative to the position of the referred 'to _{2i + 1,} G signal G' R and _{2i + 1,} which are given by the following equation.

[0007]

## EQU1 ## _{R'2i + 1} = _{R2i + 1 G'2i + 1} = ( _G2i + _{G2i + 2} ) / 2 That is, G _2i and G in which the G signal is located above and below R _{2i + 1
It} is set to the average value of _{2i + 2} . R signal R for similarly position of G _2i When _'2i, a G signal G' referred to _2i, which are given by the following equation.

[0008]

[Number 2] _{R '2i = (R 2i-} 1 + R 2i + 1) / 2 G' 2i = G 2i ie, R signals are located above and below the G _2i R _2i-1 and R
_It is set to the average value of _{2i + 1} .

When not compressed in this way, each pixel arranged in the vertical direction becomes a sampling point that determines both the R signal value and the G signal value.

FIG. 7 is a schematic diagram for explaining column-direction compression in a CCD image sensor, and R ₁ , G _2, etc. are the same notations as in FIG. In the CCD image sensor, imaging is performed by the interlace method, and even rows and odd rows are read by separate imaging. Therefore, the compression is performed by adding and synthesizing pixels corresponding to the same color component by a predetermined number of rows. In the example shown in the figure, the column is divided into 6 pixels,
A set of R and G addition pixel values is generated by the addition and synthesis processing for every 6 pixels. In other words, two added pixel values are generated from the 6 pixel values and compressed in the column direction by 1/3.

Here, two pixels R _2i-1 , adjacent to each other in the column direction,
If G _2i is defined as a pixel set, 3 rows of pixel sets are included in 6 rows corresponding to a set of R and G added pixel values, and this is defined as a set group. In the example shown in the figure, the pixels R ₁ to
G ₆ constitutes a first set group, pixels R _{7 to} G ₁₂ constitute a second set group, and pixels R _{13 to} G ₁₈ constitute a third set group.

In addition composition, pixel values included in each set group are added for each color. The added pixel values of R and G generated from the kth set group are <R> _2k−1 and <G> _{2k, respectively.}
Is written. For example, the addition pixel value of the first set group is given by the following equation.

[0013]

[Equation 3] <R>₁= (R₁+ R₃+ R_Five) <G>₂= (G₂+ G_Four+ G₆) From a CCD image sensor that has an additive synthesis function,
Corresponding to the column <R> ₁, <G>₂, <R>₃, <G>_Four，
... are output in order. That is, <R>_2k-1, <G>
_2kDefines the CCD output signals of the (2k-1) th row and the 2kth row.
To achieve.

The signal processing circuit for this output performs the same interpolation processing as in the case without compression described above, and assigns both R and G signals to each row of the CCD output to generate a compressed image with improved vertical resolution. To do. Specifically, <R>
_{When the} R signal given by the signal processing circuit is written as <R> ′ _{2k + 1} and the G signal is written as <G> ′ _{2k + 1 at the position of 2k + 1} , these are given by the following equations.

[0015]

[Equation 4] <R>' _{2k + 1} = <R> _{2k + 1} <G>' _{2k + 1} = (<G> _2k + <G> _{2k + 2} ) / 2 That is, the G signal is <R> _2k Located above and below ₊₁ <G>
_It is set to the average value of _2k and <G> _{2k + 2i} . Likewise the R signal relative to the position of the <G> _2k <R> when referred to _{2k '2k,} <G> the G signal', which are given by the following equation.

[0016]

[Equation 5] <R>' _2k = (<R> _2k-1 + <R> _{2k + 1} ) / 2 <G>' _2k = <G> _2k That is, the R signal is positioned above and below <G> _2k. <R>
_It is set to the average value of _2k-1 and <R> _{2k + 1} .

In this way, the signal processing circuit receives the column compressed image signals from the CCD image sensor in which rows are thinned out, and the R signal value and G are respectively applied to the pixels arranged in the vertical direction, that is, the sampling points. Generate a new column-wise compressed image with both signal values defined.

[0018]

SUMMARY OF THE INVENTION
Sequential output from the CCD image sensor included <R>
₁, <G>₂, <R>₃, <G>_Four, ... that are associated with
The positions of the sampling points in the column direction are not evenly spaced. example
For example, in the above example, <R>₁Sampling point corresponding to
The position in the column direction is <R>₁Pixel R that is the source of₁, R₃，
R_FiveIs the third row, which is the central row position, and <G>₂Corresponding to
The position of the sampling point in the column direction is <G>₂With
Pixel G₂, G_Four, G₆In the fourth line, which is the line position in the center of
is there. Similarly, <R>₃Is the original pixel R₇, R₉，
R₁₁Is associated with the ninth row, which is the central row position of
<G>_FourIs the original pixel G₈, G_Ten, G₁₂Center line of
The position is the 10th row. Therefore, with CCD output
<R>_{2k + 1}<G> obtained from the lines before and after_2kAnd <G>
_{2k + 2}The simple average value of is <R>_{2k + 1}Of the G signal value at the position
The accuracy is low as an estimated value.
<G>_2k<R> obtained from the lines before and after _2k-1And <R>
_{2k + 1}The simple average value of is <G>_2kOf the R signal value at the position
The accuracy is low as a fixed value. Therefore, conventional signal processing
The column-compressed image generated by the path is liable to cause moire.
It had the problem of poor quality. This image quality deterioration
Might have been acceptable if the compression ratio was low,
The compression rate increases the number of pixels in the image pickup area of the CCD image sensor.
It has become higher with the increase in size, and this problem cannot be ignored.
It

The present invention has been made in order to solve the above problems, and an object of the present invention is to provide an image signal processing device for generating a column-direction compressed image in which deterioration of image quality is reduced.

[0020]

An image signal processing apparatus according to the present invention accumulates information charges in a plurality of light-receiving pixels in which a plurality of color components are associated in a predetermined order, and among the predetermined number of pixels in the column direction, The second compressed image is generated by subjecting the first compressed image obtained from the solid-state image pickup device that synthesizes the information charges in the column direction to the row position of the center pixel of the first pixel as an initial sampling point to interpolate in the column direction. In the image signal processing device, a line memory that holds at least one row of the first compressed image output from the solid-state imaging device, and a target sampling point set at equal intervals in the column direction of the first compressed image To define the second compressed image, and use the interpolated pixel value calculated based on the data held in the line memory to supplement the pixel data in the column direction of the second compressed image. And a interpolation processing unit that, the interpolation processing unit calculates the interpolation pixel value by the weighted average calculation using the weighting coefficient corresponding to the reciprocal of the distance between the target sampling point and the initial sampling point.

In a preferred aspect of the present invention, the first compressed image has first color components and second color components alternately arranged in a predetermined column, and the interpolation processing section causes the first color component and the second color component to be alternately arranged. The image signal processing device is characterized in that the initial sampling point of either the color component or the second color component is used as the target sampling point.

In another preferred aspect of the present invention, the first compressed image has first color components and second color components alternately arranged in a predetermined column, and the interpolation processing section is arranged to perform the first color component and the second color component. The initial sampling point of either one of the first color component or the second color component is set as a first target sampling point, and the first target sampling points adjacent in the column direction are equally spaced from each other. The image signal processing device is characterized in that the row position to be set is the second target sampling point.

[0023]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described with reference to the drawings.

[First Embodiment] FIG. 1 is a block diagram showing the schematic arrangement of an image pickup apparatus having a signal processing circuit for generating a compressed image according to the present invention. This image pickup device includes a CCD image sensor 2, a CCD driver circuit 3, a timing control circuit 4, an analog signal processing circuit 5, and an A / D conversion circuit 6.
And a digital signal processing circuit 7.

The CCD image sensor 2 is configured to have a plurality of light receiving pixels in the image pickup area, and accumulates information charges generated in response to incident light in each light receiving pixel.

In the image pickup area of the CCD image sensor 2, for example, a mosaic filter composed of three colors of RGB is arranged as shown in FIG. The color C (α, β) of the light receiving pixel designated by the row number α and the column number β is determined as follows, for example.

[0027]

## EQU6 ## C (2i-1,2i-1) = R C (2i, 2i) = B ... (1) C (2i-1,2i) = C (2i, 2i-1) = G ( However, i is a natural number) In the filter array defined by the equation (1), the odd-numbered columns of the light-receiving pixels arranged in a matrix have a pixel set consisting of R and G, such as R, G, R, G ,. Repeatedly placed. On the other hand, in the even-numbered columns, a pixel set of G and B is repeatedly arranged, such as G, B, G, B, ....

The CCD driver circuit 3 drives the CCD image sensor 2 and reads out an image signal associated with each color component for each pixel. For example, when the CCD image sensor 2 is an interline type, the frame transfer clock φf is applied to the imaging unit of the CCD image sensor 2 to transfer the frame to the storage unit, and the vertical transfer clock φv is applied to the storage unit. The frame-transferred information charges are line-transferred to the horizontal transfer register. Then, the horizontal transfer clock φh is applied to the horizontal transfer register to horizontally transfer the information charges, and the output unit converts the information charges into a voltage value according to the charge amount and outputs the image signal as an image signal. The frame transfer clock φf, vertical transfer clock φv, and horizontal transfer clock φh are generated based on each timing signal supplied from the timing control circuit 4.

The timing control circuit 4 is composed of a plurality of counters for counting the reference clock CK having a constant cycle, and divides the reference clock CK to divide the frame shift timing signal FT, the horizontal synchronizing signal HT and the vertical synchronizing signal VT.
To generate.

The analog signal processing circuit 5 takes in the image signal output from the CCD image sensor 2 and performs various analog signal processing such as sample hold and gain adjustment. The A / D conversion circuit 6 takes in the image signal subjected to the analog signal processing, converts it into a digital signal for each pixel, and outputs it as image data. The digital signal processing circuit 7 performs predetermined matrix processing on the image data to generate brightness data and color data, and performs processing such as contour correction and gamma correction on the generated data. That is,
Similar to the conventional technique, at the time of preview display or moving image capturing, the image data is compressed according to the convenience of the display device and the compressed image is output.

Here, the CCD image included in the image pickup device is used.
The di-sensor 2 itself is similar to that of the prior art. Obey
For example, here also, in the case of column-direction compression operation, a 3-pixel
Information for each of the two colors included in each set group consisting of
It is assumed that the charges are added and combined and output. That is, CCD
The image sensor 2 performs additive synthesis for each 6 lines. The k-th
Group (that is, the (6k-5) to 6k rows of the light receiving pixel)
As a result of the addition synthesis to the CCD image sensor
Add pixel value <R> to the horizontal transfer register of server 2._{2k -1}，
<G>_2kAre sequentially output and the horizontal transfer register is started from the even column.
Add pixel value <G>_2k-1, <B>_2kAre output sequentially
Be done. By the way, the horizontal transfer record of the CCD image sensor 2
From the register, pixel signals that are continuous in the row direction are read out.
It This is because the CCD image sensor 2 is compressed in the column direction.
It is the same even if it is.

The digital signal processing circuit 7 has an interpolation processing section 10 for performing interpolation processing in the column direction on the output of the CCD image sensor 2 during the compression operation in the column direction. This interpolation processing is performed using the addition pixel values that are adjacent to each other in the column direction as described later, but as described above, the addition pixel values that are adjacent to each other in the column direction are continuously obtained on the output signal of the CCD image sensor 2. Absent. Therefore, the digital signal processing circuit 7 has a CCD
A line memory 12 capable of holding at least one row of output signals of the image sensor 2 is prepared. Accordingly, it is possible to perform an interpolation calculation using a certain added pixel value and the added pixel value that is output in the same column as the added pixel value and preceded by one horizontal scanning period.

As described above, the added pixel value output by the CCD image sensor 2 in the column-direction compression operation corresponds to the non-uniform sampling points (initial sampling points) in the column direction. From the added pixel values <R>, <G>, and <B> at the initial sampling point, the digital signal processing circuit 7 interpolates pixel values <R> ′ at sampling points (target sampling points) at equal intervals in the column direction. , <G>',<B>' are obtained, and the image signal represented by the interpolated pixel value is output. This allows
The nonuniformly compressed image represented by the added pixel value at the initial sampling point is converted into the uniformly compressed image represented by the interpolated pixel value at the target sampling point.

FIG. 2 is a schematic diagram for explaining a compressed image generation process in the digital signal processing circuit 7, and illustrates a process corresponding to one odd column. In the drawing, the symbol strings R ₁ , G _{2 and the} like arranged vertically on the left side are the pixel values 20 of the odd-numbered columns in the image pickup area of the CCD image sensor 2. Further, the symbol strings <R> ₁ , <G> _{2 and the} like arranged vertically to the right are added pixel values 22 obtained by adding and synthesizing the pixel values 20 for each 6 rows (that is, for each set of 3 pixels), and this is the CCD image. It is output from the sensor 2. Pixel values 20 and 22 generated and output by the CCD image sensor 2 are shown in FIG.
It is the same as the conventional one shown in.

The column direction position of the sampling points corresponding to the added pixel values <R> _2k−1 , <G> _2k (k is a natural number) is the row of the central pixel of the three added original pixels. Can be position. That is, the sampling position of <R> _2k−1 is the (6k−3) th row on the imaging area,
<G> The sampling position of _2k is the (6k-2) th row. Therefore, the sampling positions <R> _2k-1 and <G> _2k
And the interval between <G> _2k and <R> _{2k + 1} are not equal, and the initial sampling points of both <R> and <G> are not located at equal intervals. However, the initial sampling points of <R> alone or <G> alone are equally spaced.

Therefore, for example, the interpolation processing unit 10 is <R>.
One of the initial sampling point corresponding to <1> and the initial sampling point corresponding to <G> is set as the target sampling. For example, the line (6k-3), which is the position of the initial sampling point corresponding to <R>, is the target sampling point.

The interpolation processing unit 10 adds the added pixel value 22 (<R>
_2k−1 , <G> _2k ), the interpolated pixel value 24 (<R> ′ _k , <G> ′ _k ) at the target sampling point is calculated and output. Specifically, <R>' _k and <G>' _k are calculated by the following equations.

[0038]

[Equation 7]   <R> '_k  = <R>_2k-1                              ……… (2)   <G> '_k  = (<G>_2k-2+5 <G>_2k) / 6 ………… (3) Where <R>_2k-1The sampling position of is the target sample
Since it coincides with the ringing position, the interpolation processing unit 10_kTo
On the other hand, <R>_2k-1Is used as is. On the other hand, <G> '_k
Is sampling at a position above and below the target sampling position
<G>_{2k -2}And <G>_2kIs obtained by interpolating and.
This interpolation calculation is performed with the target sampling point and <G>._2k-2，
<G>_2kCorresponding to the reciprocal of the distance from each sampling point of
It is a weighted average calculation using the same weighting factor.
Accurately estimates the G pixel signal at the target sampling point
To be done. Specifically, the first sampling point where the target sampling point is located
Line (6k-3) and <G>_2k-2The sampling point of
5 lines from the (6k-8) line, while the target sump
Line (6k-3) where the ring point is located and <G>_2kThe sun
Distance from line (6k-2) where pulling point is located is 1
And based on this, as shown in equation (3),
<G>_2k-2And <G>_2kAnd are combined at a ratio of 1: 5
<G> '_kIs determined.

In order to obtain <G>' _k , the added pixel value <G> _2k-2 output from the CCD image sensor 2 in advance is required. Therefore, the digital signal processing circuit 7 stores the added pixel value <G> _2k-2 output from the CCD image sensor 2 in the line memory 12, and the interpolation processing unit 10 outputs the required added pixel value <from the line memory 12 <. G> _2k-2 is read and the above interpolation processing is performed. here,
<G> from the line memory 12 to calculate the equation (3)
After reading _2k-2 , <G> _2k is stored in the line memory 12 <G> in preparation for the interpolation processing in the next target sampling.
It is stored at the address where _2k-2 was stored.

In the above description, the odd-numbered column is taken as an example, but
The digital signal processing circuit 7 also performs the interpolation processing on the even columns.

By the processing described above, one target sampling point is set at equal intervals for every 6 rows of the original image obtained in the image pickup area of the CCD image sensor 2. That is, a column-compressed image that is uniformly 1 / 6-compressed in the column direction is generated.

[Second Embodiment] FIG. 3 is a block diagram showing the schematic arrangement of an image pickup apparatus having a signal processing circuit according to a second embodiment of the present invention. The components having the same functions as those in the above-described embodiment are designated by the same reference numerals to simplify the description. In the digital signal processing circuit 7 of the above-described embodiment, the point corresponding to any one of the initial sampling points is set as the target sampling point.
In addition to the sampling point, the midpoint is also set as the target sampling point. Digital signal processing circuit 30
Has an interpolation processing unit 32 and a work memory 34.

FIG. 4 is a schematic diagram for explaining a compressed image generation process in the digital signal processing circuit 30, and illustrates a process corresponding to one odd column. Interpolation processing unit 3
In the same manner as in the above-described embodiment, 2 sets the target sampling to either one of the initial sampling point corresponding to <R> and the initial sampling point corresponding to <G>. For example, the line (6k-3), which is the position of the initial sampling point corresponding to <R>, is set as the first target sampling point 40. The interpolation processing unit 32 further defines points that divide the two adjacent first target sampling points 40 at equal intervals as second target sampling points. For example, in FIG. 4, the 6th row, which is the midpoint between the two first target sampling points, is the second target sampling point 42.

The interpolation processor 32 adds the added pixel value 22 (<R>
_2k−1 , <G> _2k ), the interpolated pixel value 44 at the first and second target sampling points is calculated. The sixth (6
The interpolated pixel values at the first target sampling point located in the (k-3) th row are <R>' _2k-1 and <G>' _2k-1 , which correspond to the equations (2) and (3) in the above embodiment. It is calculated by the following formula.

[0045]

[Equation 8] <R>' _2k-1 = <R> _2k-1 ... (4) <G>' _2k-1 = (<G> _2k-2 +5 <G> _2k ) / 6 ......... (5) On the other hand, the interpolated pixel values at the second target sampling point located on the 6th row are <R>' _2k and <G>' _2k .
<R>' _2k is obtained at initial sampling points located above and below the second target sampling point, and <R> _2k-1 and <R>
_{2k + 1} and is interpolated, and <G>' _2k is obtained from the initial sampling points located above and below the second target sampling point, and <G> _2k and <G> _{2k + 2} are obtained. It is obtained by interpolation. This interpolation calculation is a weighted average calculation similar to the equation (5). That is, <R>' _2k and <G>' _2k are calculated by the weighted average calculation represented by the following equation using the weighting coefficient according to the reciprocal of the distance between each initial sampling point and the second target sampling point. To be done.

[0046]

[Equation 9] <R>' _2k = (<R> _2k-1 + <R> _{2k + 1} ) / 2 (6) <G>' _2k = (2 <G> _2k + <G> _{2k +2} ) / 3 (7) In order to obtain <R>' _2k and <G>' _2k , the added pixel value <R> _2k-1 output from the CCD image sensor 2 in advance. , <G> _2k is required. Therefore, the digital signal processing circuit 30 stores the added pixel values <R> _2k−1 , <G> _2k output from the CCD image sensor 2 in the working memory 34, respectively, and the interpolation processing unit 32 causes the working memory 34 to operate. The necessary addition pixel value <R> _2k-1 is read out from the working memory 34 and the necessary addition pixel value <G> _2k is read from the working memory 34 to perform the interpolation processing of the formula (7). To do.

The flow of processing will be described. Interpolation processing unit 32
When <R> _{2k + 1} is input from the CCD image sensor 2, the <R> _2k-1 stored in the working memory 34 is read out and interpolation is performed based on the equation (6). 'Output _2k . On the other hand, <R> _{2k + 1} is stored in the work memory 34. Although this is processing for a certain odd number column, the same processing is performed for other columns.

Next, from the CCD image sensor 2 to <G>
_{When 2k + 2} is input, the interpolation processing unit 32 causes the work memory 3
The <G> _2k stored in 4 is read out, interpolation is performed based on the equation (7), and <G>' _2k is output. On the other hand, <G>
_{2k + 2} is stored in the work memory 34. The same processing is performed for the other columns.

Thereafter, the digital signal processing circuit 30 outputs <R> _{2k + 1} from the working memory 34 based on the equation (4).
Is read out and is output as <R>' _{2k + 1} . A process corresponding to this is similarly performed for other columns.

Further, thereafter, the digital signal processing circuit 3
0 from the working memory 34 <G> _2kAnd <G>_{2k + 2}To
Read out and interpolate based on equation (5), then <G> '
_{2k + 1} Output as. Corresponding processing to other columns
The same goes for the following.

By the above processing, two target sampling points are set at equal intervals for every 6 rows of the original image obtained in the image pickup area of the CCD image sensor 2. That is, a column-compressed image that is 1 / 3-compressed uniformly in the column direction is generated.

[0052]

According to the image signal processing apparatus of the present invention, it is possible to generate a compressed image having good image quality with less moire from image data which is compressed unevenly in the column direction.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an image pickup apparatus including a signal processing circuit according to a first embodiment of the present invention.

FIG. 2 is a schematic diagram illustrating a compressed image generation process in the signal processing circuit according to the first embodiment.

FIG. 3 is a block diagram showing a schematic configuration of an image pickup apparatus including a signal processing circuit according to a second embodiment of the present invention.

FIG. 4 is a schematic diagram illustrating compressed image generation processing in a signal processing circuit according to a second embodiment.

FIG. 5 is a schematic diagram illustrating processing when compression in the column direction is not performed.

FIG. 6 is a schematic view showing an arrangement of color filters.

FIG. 7 is a schematic diagram illustrating column-direction compression in a CCD image sensor.

[Explanation of symbols]

2 CCD image sensor, 3 CCD driver circuit,
4 timing control circuit, 5 analog signal processing circuit,
6 A / D conversion circuit, 7,30 Digital signal processing circuit, 10,32 Interpolation processing unit, 12 line memory, 3
4 Working memory.

Continued front page    F term (reference) 5B057 BA02 BA12 CA01 CA08 CA12                       CA16 CB01 CB08 CB12 CB16                       CC01 CD07 CD10 DC09                 5C065 AA03 BB13 CC01 DD07 DD17                       FF03 GG11 GG13 GG17 GG18                       GG21

Claims (3)

[Claims] 1. Information charges are accumulated in a plurality of light receiving pixels in which a plurality of color components are associated with each other in a predetermined order, and the information charges are stored with a row position of a central pixel among a predetermined number of pixels in a column direction as an initial sampling point. In the image signal processing device for performing the interpolation processing in the column direction on the first compressed image obtained from the solid-state image pickup element for synthesizing in the column direction to generate the second compressed image, the image signal processing apparatus outputs the second compressed image. A line memory that holds at least one row of the first compressed image, and the second compressed image is defined by sampling the first compressed image at target sampling points set at equal intervals in the column direction. An interpolation processing unit that interpolates pixel data in the column direction of the second compressed image using an interpolation pixel value calculated based on the data held in the line memory, the interpolation processing unit The image signal processing device, wherein the interpolation pixel value is calculated by a weighted average calculation using a weighting coefficient according to the reciprocal of the distance between the initial sampling point and the target sampling point. 2. The image signal processing device according to claim 1, wherein in the first compressed image, first color components and second color components are alternately arranged in a predetermined column, and the interpolation processing unit , The first color component, or the second color component
The image signal processing device, wherein the initial sampling point of any one of the color components is used as the target sampling point. 3. The image signal processing device according to claim 1, wherein in the first compressed image, first color components and second color components are alternately arranged in a predetermined column, and the interpolation processing unit , The first color component, or the second color component
The first sampling point of one of the color components of
The first sampling point that is adjacent in the column direction
The image signal processing device, wherein the row positions that divide the target sampling points from each other at equal intervals are used as the second target sampling points.