JPH01228289A - Compressing and recording device and extending and reproducing device for picture signal - Google Patents

Abstract

PURPOSE:To prevent the deterioration of a picture according to the compression of data by forming a difference signal between the picture elements of the same color of a video signal image picked up by a single plate type color image sensor, compressing the encoding. CONSTITUTION:A preliminary data forming circuit 200 uses picture data obtained from the second picture element of the same color preceding by the prescribed number of the picture elements from the first picture element and the third picture element of the same color preceding by the prescribed number of lines from the first picture element according to a timing signal transmitted from a controller 36, when the picture data obtained from the first picture element is inputted to a subtracter 50 to calculated the predictive data of the picture data obtained from the first picture element and transmit the predictive data to the subtracter 50. In such a way, the difference signal between the picture elements of the same color of the video signal image picked up by the single plate type color image sensor is formed, compressed and encoded, so that the correlation of the picture data between the picture elements is enhanced to prevent the picture from being deteriorated due to the compression of the data.

Description

Detailed Description of the Invention Technical Field The present invention relates to an image signal compression/recording device and decompression/playback device, and more particularly to a compression/recording device for compressing image signal data and recording it on a recording medium, and a compression/recording device for compressing data of an imaged image signal and recording it on a recording medium, and The present invention relates to a decompression/playback device that reads and decompresses captured data and plays back captured images.

BACKGROUND ART For example, when storing an image signal captured by a solid-state image sensor such as a CCD in a storage device such as a memory card or a magnetic disk, the data of one image number is reduced to a small capacity in consideration of the storage capacity of the storage device. It is necessary to compress. An example of such an image data compression method is DPCM.
(predictive coding) is also known. DPIL:M aims to remove redundancy based on the correlation between image sample values, and predicts the next sample value from known sample values by performing appropriate @ calculations, and then calculates the predicted value and This encodes the difference signal from the actual sample value. In this method, the prediction accuracy of a signal with a large correlation between adjacent sample values increases, the effective amplitude of the difference signal can be reduced, and highly efficient coding becomes possible. Further, this method can be realized with a relatively simple device configuration, and in the case of image signals, it is possible to save 2 to 3 bits per pixel compared to ordinary PCM encoding.

By the way, conventional DPIJ calculates the difference signal to be encoded by calculating the R (red), Gl), B (blue) signals, or component signals such as the luminance signal (Y) and color difference signals (I signal and Q signal). This was done using data between adjacent pixels.

Therefore, if this method is used, for example, in a mid-plate color image sensor that extracts primary color video signals from a single solid-state image sensor, for example, green filters are arranged every other row in the image sensor, and red and blue filters are placed between them. When pixels are arranged alternately, the filter colors of adjacent pixels are different, so if you simply calculate the difference between the data of adjacent pixels, the error will become large and the image deterioration will be severe. Ta.

SUMMARY OF THE INVENTION The present invention solves the drawbacks of the conventional technique 11g1, and eliminates deterioration due to data compression even when an image signal captured by a single-chip color image sensor is compressed and encoded using DPC:H. An object of the present invention is to provide a compression recording device and a decompression playback device for image signals.

DISCLOSURE OF THE INVENTION According to the present invention, image data of the color image signal and an image are obtained by receiving an image signal captured by a color imaging means having a plurality of color filters on the front surface and corresponding to these color filters. Calculate the difference value between the data and the measured data,
An image signal compression recording device that compresses and encodes a difference value and records it on a recording medium includes a pixel delay unit that delays the input image data by a predetermined number of pixels and outputs the input image data when the image data is input; a line delay means for delaying input data by a predetermined number of lines and outputting the input data when the input data is input, a prediction data creation means for creating prediction data, and a prediction data and image created by the prediction data creation means. It has a subtraction means for calculating a difference value from the data, a color imaging means, and a control means for controlling the input/output timing of each data of the pixel delay means and the double delay means of the pre-a11 data creation means, and the prediction The data creation means follows the timing signal sent from the control means, and when the image data obtained from the first pixel is input to the subtraction means,
Image data obtained from a second pixel of the same color that is a predetermined number of pixels before the first pixel, and a third pixel of the same color that is a predetermined number of lines before the first pixel. The predicted data of the image data obtained from one pixel is calculated, and the predicted Jllll data is sent to the subtraction means.

Further, according to the present invention, the image data of the original color image signal is reproduced by decoding the compression-encoded difference value recorded on the recording medium by the above-mentioned image signal compression recording device and adding prediction data. The image signal decompression/reproduction device includes a pixel delay means that delays the input pixel data by a predetermined number of pixels when pixel data is input, and outputs the input pixel data after inputting the input pixel data. a line delay means for delaying and outputting the data by a predetermined number of lines, and a prediction data creation means for creating prediction data;
and addition means for adding the decoded difference value to the prediction data created by the pre-4111 data creation means, and the prediction data creation means adds the decoded difference value to the prediction data created by the addition means. Difference (add to 11ff) from data corresponding to a second pixel of the same color that is a predetermined number of pixels before the first pixel corresponding to the value, and a third pixel of the same color that is a predetermined number of lines before the first pixel corresponding to the value. It creates prediction data and sends the prediction data to the addition means.

DESCRIPTION OF EMBODIMENTS Next, embodiments of an image signal compression recording apparatus and decompression reproduction apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

Referring to FIG. 1, there is shown an embodiment in which an image compression recording apparatus according to the present invention is applied to a digital electronic still camera having a single-plate color image sensor. In addition,
Other parts of the camera that are not directly related to the description of the present invention, such as shutter, diaphragm, film, and other mechanisms, are not shown.

This device has a master lens 12, and the master lens 1
An imaging device 14 that converts an optical image of a subject captured by the master lens 12 from an optical signal to a video signal is arranged behind the lens 2 . The imaging device 14 is provided with a color filter 16 on its surface, and as shown in FIG. This is a so-called single-chip color image sensor in which (R) and blue filters (B) are arranged alternately.

The Tenoheisu 14 is connected to the signal line 1 from the controller 36.
The optical image is converted from an optical signal to a video signal according to the timing of a reference clock sent through 40. A video signal obtained from the imaging device 14 is sent to the AD conversion circuit 18 through a signal line 100. In addition, the imaging differential chair 1
Although not shown, comma correction or white balance processing means may be inserted between 4 and the AD conversion circuit 18.

The AD conversion circuit 18 is connected to the signal line 1 from the timing circuit 38.
According to the timing signal sent through 42, the analog video signal sent from the imaging device 14 is converted into a digital video signal. The color image signal converted into a digital signal by the AD conversion circuit 18 is transmitted to the signal line 10.
A difference signal is created between a digital color image signal sent to the subtracter 50 through the signal line 116 and a prediction signal X, which will be described later, is input to the other input through the signal line 116.
The signal is sent to the quantization circuit 20 through the quantization circuit 20. The quantization circuit 20 quantizes the difference signal produced by the subtracter 50 at a predetermined level, sends the quantized difference signal to an encoding circuit (not shown) through the signal Vj, 108, and the encoding circuit converts the quantized difference signal to an encoding circuit (not shown). A predetermined number of bits are assigned to the signal, encoded, and recorded in the semiconductor memory 34.

On the other hand, the difference signal quantized in the quantization circuit 20 is sent to the inverse quantization circuit 22 through the signal line 108 and to the signal line 1
Through 20]! ! The signal is sent to the quantization circuit 28.

The inverse quantization circuit 22 restores the difference signal quantized by the ψ quantization circuit 20 to the original difference signal. Inverse quantization circuit 2
The difference signal restored in step 2 is sent to the adder 52 of the predicted signal generation circuit 200, and the data of the difference signal delayed by two pixels divided by 1/2, which will be described later, is input to the adder 52 from the signal line 118. (hereinafter referred to as 2-pixel delayed difference signal). The difference signal added to the two-pixel delayed difference signal in the adder 52 is sent to the pixel delay circuit 24 through the signal line 112.

Pixel i! ! The delay circuit 24 delays the difference signal inputted from the signal vj112 by two pixels according to the timing signal inputted from the timing circuit 40 through the signal line 144, and outputs the delayed signal to the signal line 114. That is, for example, in the color filter 16 in FIG. 3, the digital color image signal for the pixel corresponding to C (filter (G)) is
If it is input as 0, the difference signal for the pixel corresponding to b (filter (G)) is the difference signal! ! 114.

The two-pixel delay difference signal outputted to the signal line 114 is divided by 1/2 in the divider 26 and outputted to the adder 54 through the signal line 115. On the other hand, the two-pixel delay difference signal divided by 1/2 in the divider 26 is transmitted to the signal line 11B.
It returns to the adder 52 through.

Further, the quantized difference signal sent to the dequantization circuit 28 through the signal line 120 is restored to the original difference signal in the dequantization circuit 28 . The difference signal restored by the inverse quantization circuit 28 is sent to the predicted signal generation circuit 2 through the signal line 122.
00 adder 56, and is sent from the signal line 130 to the adder 5
Data of the difference signal that is divided by 1/2 and delayed by 2 lines (hereinafter referred to as 2-line delayed difference signal) is input to 6.
is added. The difference signal added in adder 56 is sent to line memory 30 through signal line +24.

The line memory 30 delays the difference signal inputted from the signal line 124 by two lines according to the timing signal from the timing circuit 42 inputted from another input 146, and outputs the delayed signal to the signal line 126. Therefore, according to the example described above, C (filter (G)) of the color filter 16 in FIG.
The digital color image signal of the pixel corresponding to the subtracter 50
, a difference signal a for a pixel corresponding to a (filter (G)) two lines before is output to the signal line 128. The two-line delay difference signal outputted to the signal line 126 is divided by 1/2 in the divider 32 and outputted to the adder 54 through the signal line 128. On the other hand, the two-line delayed difference signal divided by 1/2 in the divider 32 returns to the adder 56 through the signal line 130.

The adder 54 receives the 2-pixel delayed difference signal divided by 2 input from the signal line 115 and the signal line ! 28 is added to the 2-line delayed difference signal divided by 2, and this is sent to the subtracter 50 through the signal line 118 as a prediction signal.

The controller 36 includes the imaging device 14, the AD conversion circuit 18. It is connected to the pixel delay circuit 24 , timing circuits 38 , 40 , 42 that generate signals to measure the timing of the operation of the line memory 30 , and the imaging differential 14 , and sends out a reference clock through the signal line 140 .

The operation of this device will be explained.

The optical image of the subject captured by the master lens 12 is converted from an optical signal to a video signal in the order of pixels corresponding to lea, 16b, . . . of the color filter 16 shown in FIG. 3 in the imaging device 14. The signal is sent to the AD conversion circuit 18 through the signal line 100. The video signal is converted into a digital signal by the AD conversion circuit 18 and sent to the subtracter 50 through the signal line 102. For example, when the video signal 1c of the pixel corresponding to C of the color filter 18 is sent to the subtracter 50, the video signal Tc is compared with the prediction signal X input from the prediction signal generation circuit 200 through the signal line 116 in the subtracter 50. , is converted into a difference signal Dc. Here, as described above, the predicted signal The signals are added together. That is, subtractor 5
2 of the video signal IC for the filter C input to 0
The 2-pixel delayed difference signal Db divided by 172 for filter b, which has 1M pixels, and the 2-line delayed difference signal Da, divided by 172 for filter a, which is 2 lines before, are calculated. It is something. This can be expressed as xl/2(Da+Db). Therefore, the difference signal [1c produced in the subtracter 50 is expressed by a linear equation.

Dc=Ic-1/2 (Da+Db) The difference signal Dc generated by the subtracter 50 is sent to the quantization circuit 20 through the signal line 104, and is quantized at a predetermined level in the quantization circuit 20. Quantized difference signal D
c is sent to an encoding circuit (not shown) through a signal line 106, where it is assigned a predetermined number of bits, encoded, and recorded in the semiconductor memory 34. - Sword, the difference signal Dc is a signal! ! It is sent to the inverse quantization circuit 22 and the inverse quantization circuit 28 through 1t108 and the signal line 120, and the prediction signal generation circuit 200 generates a prediction signal X for the next video signal.

The methyl image photographed by the electronic still camera in this manner is recorded in the semiconductor memory 34.

Figure 2 shows the image taken by the electronic still camera shown in Figure 1.
An example of a reproducing device that reproduces images recorded in the semiconductor memory 34 is shown.

The difference signal recorded in the semiconductor memory 34 is input from an input terminal (not shown), and is input to the representative value setting circuit 60 through a signal line 148. The representative value setting circuit 60 sets a representative value for restoring the original video signal using the difference signal input through the signal line 148, and decodes the input difference signal in a decoding circuit (not shown).

The adder 80 adds the difference signal input through the signal line 150 and a prediction signal y, which will be described later, input to another input from the signal line 184 to create an addition signal, and adds the signal M 1
52 to the image processing circuit 62.

The image processing circuit 62 inversely quantizes the addition signal produced by the adder 80 at a predetermined level to create original image data.
It is sent to the DA converter 64 through the signal line 154.

The DA converter 64 converts the digital video signal input through the signal line 154 into an analog signal, and converts the digital video signal input through the signal line 154 into an analog signal.
8 to output 68. The output 66 is, for example, a CRT.
The color image recorded in the semiconductor memory 34 is reproduced on the display device.

On the other hand, the addition signal output from the adder 80 is
60 and the signal line 168, the predicted signal generation circuit 3
The pixel delay circuit 6, which is sent to the pixel delay circuit 68 of 00 and the line memory 72, delays the addition signal input from the signal line 180 by two pixels and outputs it to the signal line 162.

The addition signal delayed by two pixels output to the signal line 162 is divided by 172 in the divider 70, and the signal is output to the signal line 162.
6 and is output to the adder 82.

The line memory 72 delays the addition signal input from the signal line 168 by two lines and outputs it to the signal line 170. The addition signal delayed by two lines and output to signal vj170 is divided by 1/2 in divider 74 and output to adder 82 through signal line 172.

The adder 82 receives the addition signal inputted from the signal line 168 and delayed by two pixels divided by 1/2, and the signal line 172
was divided by the input 1/2. The added signal delayed by two lines is added and sent to the adder 80 through the signal line 164 as the predicted signal y.

Next, the operation of the playback device will be explained.

The compressed difference signal recorded in the semiconductor memory 34 is read out by a connector (not shown) and input to the representative value setting circuit 60 through the signal line 14B.

The difference signal input to the representative value setting circuit 60 is decoded, and a representative value for restoring the original video signal is set. That is, in the apparatus shown in FIG. 1, a signal encoded by an encoding circuit (not shown) is decoded. The decoded difference signal is added with the prediction signal y created in the prediction signal creation circuit 300 in the adder 80, and is sent to the signal line 152.
The signal is sent to the image processing circuit 82 through the image processing circuit 82. The addition signal created by the adder 80 is dequantized in the 1-image processing circuit 82 to create original image data, which is sent to the DA converter 64 through the signal line 154. The digital image data created by the image processing circuit 62 is converted into an analog video signal by the DA converter 64 and sent to the output 66 through the signal v;j+58. The video signal converted into analog by the OA converter 64 is sent to a display device such as a CRT via an output 66 and reproduced as an original still image on one display device.

As mentioned above, according to the electronic still camera shown in FIG.
A video signal captured by a single-plate color image sensor 14 having a color filter 16 shown in the figure is transmitted to the DPCH
When 85! of one pixel is encoded and recorded in the semiconductor memory 34, When the image signal is input, the difference between the prediction signal X calculated from the signal two pixels before and the signal two lines before is calculated, and the difference signal is compressed and encoded and recorded in the memory 34.

In other words, a prediction signal X is created from pixels of the same color component, and a difference signal from this prediction signal X is created and compressed and encoded. For example, 113B, 16E, and 1 in FIG.
Unlike conventional methods in which processing is performed using adjacent pixels of different colors, such as when calculating a difference signal between 8Es, this method increases the correlation of pixel data between pixels and prevents image deterioration due to data compression. Can be done.

Furthermore, the output signal of the single-chip color image sensor 14 can be directly processed without being separated into RGB components, and since the compression process is performed using DPCM, the compression encoding process can be realized with a simple device. can.

Furthermore, according to the playback device shown in FIG. 2, when reading out the difference signal recorded in the semiconductor memory 34 in the electronic still camera shown in FIG. 1 and playing back the original video signal, the video signal of the l pixel is input. When the difference signal of 2 pixels before and 2 pixels
Since image processing is performed by adding the predicted signal y calculated from the difference signal before the line, the still image captured by the electronic still camera shown in FIG. 1 can be faithfully reproduced.

In this embodiment, in the electronic still camera shown in FIG. 1, the predicted signal For example, if pixels of the same color component are adjacent due to the RGB arrangement, it is also possible to calculate using a difference signal between the adjacent pixels.

Also, the imaging device 14. The signals to measure the timing of the operations of the AD conversion circuit 18, pixel delay circuit 24, and line memory 30 were created using the reference clock sent from the controller 36, but this timing signal was generated from the clock for the single-chip color image sensor 14. A sword that allows you to create one is more preferable.

Effects According to the present invention, a differential signal between pixels of the same color among the video signals captured by a single-chip color image sensor is created and compressed and encoded, so that the correlation of pixel data between pixels is increased. , it is possible to prevent image deterioration due to data compression.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the image signal compression/recording device according to the present invention applied to a digital electronic still camera. FIG. 2 is a block diagram showing an embodiment of the image signal decompression/reproduction device δ according to the present invention. FIG. 3 is a diagram showing an example of the arrangement of filters in the camera of FIG. 1. Explanation of symbols of main parts +4. , , , , Imaging device +8. , , , , , AD conversion circuit 20 , , , , , quantization circuit 22.28 , , , , , inverse quantization circuit 24 .
88 , , , , pixel delay circuit 30.72
, , , , , line memory 34 , , , , , ,
Semiconductor memory 36, , , controller 38.40.42. , , timing circuit eo,
, , , , Representative value setting circuit 62 , , , Image processing circuit B4 . , , , , , DA conversion circuit 200.300 , , , , Prediction signal generation circuit Patent applicant Fuji Photo Film Co., Ltd. Agent
Address: Takao Ohiba Takao 3 Concave

Claims (1)

[Claims] 1. Having color filters of a plurality of colors on the front side, receiving an image signal imaged by a color imaging means corresponding to these color filters, and transmitting image data of the color image signal and the image data of the color image signal. An image signal compression recording device that calculates a difference value between image data and predicted data, compresses and encodes the difference value, and records it on a recording medium. pixel delay means for delaying and outputting image data by a predetermined number of pixels; and line delay means for delaying and outputting the input data by a predetermined number of lines when the image data is input, a prediction data creation means for creating prediction data; a subtraction means for calculating the difference value between the prediction data created by the prediction data creation means and the image data; the color imaging means; and the prediction data creation means. control means for controlling input/output timing of each data of the pixel delay means and the line delay means, and the prediction data creation means is configured to control the subtraction means according to a timing signal sent from the control means. When image data obtained from one pixel is input, a second pixel of the same color, which is a predetermined number of pixels before the first pixel, and the first pixel are input.
The predicted data of the image data obtained from the first pixel is calculated using the image data obtained from a third pixel of the same color that is a predetermined number of lines before the pixel of A compression recording device for an image signal, characterized in that the image signal is sent to a subtraction means. 2. The compression-encoded difference value recorded on the recording medium is decoded by the apparatus according to claim 1, and the image data of the original color image signal is obtained by adding prediction data. A decompression/reproduction device for an image signal to be reproduced includes: a pixel delay unit that delays the input pixel data by a predetermined number of pixels when the pixel data is input, and outputs the input pixel data; a line delay means for delaying the input data by a predetermined number of lines and outputting the input data when input, a prediction data creation means for creating the prediction data; and prediction data created by the prediction data creation means. and an addition means for adding the decoded difference value to the first difference value, and when the decoded difference value is input to the addition means, the prediction data creation means adds the first difference value corresponding to the difference value. The predicted data to be added to the difference value is created from data corresponding to a second pixel of the same color that is a predetermined number of pixels before the pixel, and a third pixel of the same color that is a predetermined number of lines before the pixel. , an image signal decompression/reproduction device characterized in that the predicted data is sent to the addition means.