JPH02222394A - Picture signal compression encoder - Google Patents

Abstract

PURPOSE:To enlarge the compressibility of picture data and to obtain comparatively satisfactory picture quality by setting a normalization coefficient concerning a luminance signal smaller than a normalization coefficient concerning a chrominance signal and compressing chrominance signal data by the compressibility larger than that of luminance signal data. CONSTITUTION:A two-dimensional orthogonal transform part 14 executes the two-dimensional orthogonal transform of picture data for each block of a luminance signal Y and color difference signals R-Y and B-Y to be successively inputted. Then, a normalization part 16 normalizes the picture data for which the two-dimensional orthogonal transform is executed in the two-dimensional orthogonal transform part 14. Normalization coefficient setting means 44, 46, 48 and 50 are provided to set the normalization coefficient. Then, by setting the normalization coefficient to be used concerning the luminance signal data to a smaller value than the normalization coefficient to be used concerning the chrominance signal data and executing normalization, the chrominance signal data are compressed by the compressibility larger than that of the luminance signal data. Thus, the compression of the data can be efficiently executed and data quantity can be reduced. Then, the degradation of the picture quality can be prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image signal compression encoding device, and more particularly to an image signal compression encoding device for efficiently compressing and encoding color image signals. When storing digital image data such as transferred image data in memory, in order to reduce the amount of data and reduce the storage capacity of the memory,
Various compression encoding methods are used. In particular, two-dimensional orthogonal transform encoding is widely used because it allows encoding at a high compression rate and suppresses image distortion associated with encoding.

In such two-dimensional orthogonal transform encoding.

The image data is divided into a predetermined number of blocks, and the image data within each block is subjected to two-dimensional orthogonal transformation. The orthogonally transformed image data, that is, the transformation coefficients, are compared with a predetermined threshold, and portions below the threshold are truncated (coefficient truncation). As a result, the conversion coefficient below the threshold is
Thereafter, the transform coefficients that have been subjected to the primary coefficient truncation and are processed as 0 data are divided by a predetermined quantization step value, that is, a normalization coefficient, and quantization, that is, normalization, is performed according to the step width. Thereby, the value of the conversion coefficient, that is, the dynamic range of the amplitude can be suppressed. After that, encoding such as two-dimensional Huffman encoding is performed to obtain compressed data.

When performing such two-dimensional orthogonal transform encoding on a color image signal, the luminance signal Y and the color difference signals R-Y, B-Y, which constitute the color image signal, are each subjected to two-dimensional orthogonal transform and normalized. Normalize by coefficient. In that case,
Luminance signal Y and color difference signal R-Y, B-Y to be processed
The ratio of the amount of data is usually Y :RY :B-Y
= 4 + 2: 2. That is, the ratio of the luminance signal to the color difference signal consisting of both Y and B-Y is l:1. Therefore, the data amount ratio between the luminance signal and the color difference signal obtained by orthogonal transform encoding is also l:l.

Conventionally, since the luminance signal and the chromaticity signal were compressed and encoded at the same compression rate, it was not possible to perform efficient data compression taking into account the characteristics of these signals. That is, for example, when the amount of data after compression is limited, there is a problem in that reducing the amount of data causes a corresponding reduction in image quality.

-m The present invention solves the problems of the prior art and efficiently converts color image signals by using normalization coefficients according to luminance signals or chromaticity signals in normalization after two-dimensional orthogonal transformation. An object of the present invention is to provide an image signal compression encoding device that performs orthogonal transform encoding.

According to the present invention, an image signal compression encoding device performs two-dimensional orthogonal transformation encoding of color image data constituting one screen for ti degree signal data and chromaticity signal data, respectively. orthogonal transformation means for two-dimensional orthogonal transformation of image data; normalization means for normalizing the data orthogonally transformed by the orthogonal transformation means; encoding means for encoding the data normalized by the normalization means; normalization coefficient setting means for setting normalization coefficients for the signal data and chromaticity signal data, the normalization coefficient setting means for setting normalization coefficients used for the t11 signal data, for the chromaticity signal data; The normalization coefficient is set to a smaller value than the normalization coefficient, and the normalization means performs normalization using the normalization coefficient set by the normalization coefficient setting means, thereby compressing the chromaticity signal data to a greater degree than the luminance signal data. It is compressed by the ratio.

Embodiments of the image signal compression encoding device according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 shows an embodiment of an image signal compression encoding device according to the present invention.

The device has an input terminal IO, and from three terminals of the input terminal IO, a luminance signal Y and a color difference signal RY are output.

B-Y are input sequentially. Luminance signal Y and color difference signal R
-Y and B-Y are created by separating the video signals of a still image captured by an image sensor into five colors and converting them into a matrix. In addition, the luminance signal Y and the color difference signals R-Y, B-Y are processed by a blocking unit (not shown).
Still image data for a frame is divided into a plurality of blocks, and pixel data as shown in FIG. 3 is input for each block.

The luminance signal Yj15 and the color difference signals RY and B-Y input from the input terminal IO are sent to the switch 12. The switch 12 is switched according to a control signal from the control section 40 and outputs either the luminance signal Y1 and the color difference signal RY or B-Y inputted from the input terminal 10 to the two-dimensional orthogonal transformation section 14.

The two-dimensional orthogonal transform unit I4 performs two-dimensional orthogonal transform on the image data of each block of the luminance signal Y and the color difference signals RY and BY, which are sequentially manually input. As a two-dimensional orthogonal transformation,
Well-known orthogonal transformations such as discrete cosine transformation and Hadamard transformation are used.

2D orthogonal transformation part! The image data for each block subjected to the two-dimensional orthogonal transformation in step 4 is arranged vertically and horizontally, with low-order data arranged in the upper left part and higher-order data moving toward the lower right. The output of the two-dimensional orthogonal transform section 14 is sent to the normalization section 16.

The normalization unit 16 performs normalization after truncating coefficients on the image data that has been subjected to two-dimensional orthogonal transformation in the two-dimensional orthogonal transformation unit 14, that is, the transformation coefficients. is compared with a predetermined threshold, and the portion below the threshold is discarded. Normalization is performed by dividing the truncated transform coefficient by a predetermined quantization step value, that is, a normalization coefficient, and performing quantization using the normalization coefficient.

The normalization unit 16 is connected to the weight table 4 via a switch 44.
6.48!3 and 50 are connected. switch 44
is switched according to a control signal from the control unit 40,
Weight tables 46, 48 and 50 are used to normalize the luminance signal Y1 and the color difference signals R-Y and B-Y, respectively. This is a table of data of normalization coefficients used in . As shown in FIG. 4, these weight tables consist of the same number of data as the number of pixels forming a block. For the conversion coefficient, the low-frequency components are important as data, and the high-frequency components are less important.
In the weight table shown in FIG. 4, data is assigned so that low frequency component data is emphasized and high frequency component data is discarded.

The weight table 46 is table data for the luminance signal Y, and the weight table 48 is for the color difference signals R-Y and B-Y.
and data smaller than 50.

Therefore, the image data of the m-degree signal Y is the color difference signal RY
Since the normalized luminance signal Y data is divided by the table data having a smaller value than the image data of and B-Y, the data of the luminance signal Y after normalization becomes larger than the data of the color difference signals R-Y and B-Y. As a result, the luminance signal Y is changed to the color difference signals RY and B.
- Orthogonal transform encoding is performed with a compression rate smaller than Y. In other words, the color difference signals R-Y and B-Y are compressed and encoded to be larger than the luminance signal Y.The weight ratio between the weight table 46 and the weight tables 48 and 50 is, for example, the luminance signal that is encoded and output. Y and color difference signal R-Y and 11
-Y (71) It is preferable to set the ratio so that the total ratio is 3=1, that is, the color difference signal R-Y or B-Y is each one-sixth of the luminance signal Y.

The normalization unit 16 normalizes the image data for each block of the luminance signal Y1, color difference signals R-Y and B-Y using the data of the weight table 46, 48 or 50 selected and sent by the switch 44. weight table 4
Divide by the data from 6.48 or 50. 1
Normalization is performed using data from the weighting table 46 for the 11 signal Y, data from the weighting table 48 for the color difference signal RY, and data from the weighting table 50 for the color difference signal B-Y.

Note that when performing normalization by dividing the conversion coefficient by the table data T, if 1/T is calculated in advance and the conversion coefficient is multiplied by this value, the number of dividers can be reduced. The scale of the device can be reduced.

The normalized transform coefficients are arranged in blocks in the same way as the pixel data shown in FIG. 3, and are output after being scanned in a zigzag pattern starting from the low frequency component as shown in FIG. 5.

The output of the normalization unit 16 is output to a two-dimensional Huffman encoding unit 28. The two-dimensional Huffman encoding unit 28 often has consecutive zeros in the normalized transform coefficients that are scanned and input in a zigzag manner as described above. Detect run length,
A run length of zero and an amplitude of non-zero are determined and subjected to two-dimensional Huffman encoding. The output from the two-dimensional Huffman encoding unit 28 is output to an output terminal 32. The data sent to the output terminal 32 is transmitted to a transmission path (not shown) or recorded on a recording medium such as a magnetic disk.

The control unit 40 is a control unit that controls each functional unit of this device, and in particular inputs the luminance signal Y, color difference signals R-Y, and B-Y, which are sequentially input from the input terminal IO, to the two-dimensional orthogonal transformation unit 14. Therefore, switching of the switch 12 is controlled. In addition, the luminance signal Y and color difference signal R-Y input manually to the normalization section
and weight table 46.48 or 5 depending on B-Y
In order to send the data from 0 to the normalization unit 16, switch 4
According to this device, the input terminal l
The switch 12 is switched so that data of the luminance signal Y is first inputted from O, and the switch 44 is switched so that table data from the weight table 46 is inputted to the normalization unit 16. As a result, the data of the luminance signal Y is input from the input terminal lO through the switch 12 to the two-dimensional orthogonal transformer 14, where it is orthogonally transformed and the normalizer 16
sent to. The normalization coefficient data for the luminance signal γ is input from the weight table 46 to the normalization unit 16, and normalization is performed using this data.6 The data of the normalized luminance signal Y is encoded by the two-dimensional Huffman encoding unit 2B. output terminal 3
Output to 2.

Next, the switch 12 is switched so that the data of the color difference signal RY is inputted from the input terminal 10, and the switch 44 is switched so that the table data from the weighting table 48 is inputted to the normalization section 16. As a result, the data of the color difference signal RY is inputted from the input terminal 10 through the switch 12 to the two-dimensional orthogonal transformation unit I4, subjected to orthogonal transformation, and sent to the normalization unit 16. Normalization coefficient data for color difference No. 8 RY is input from the weight table 48 to the normalization unit 16, and normalization is performed using this data. The data of the normalized color difference signal RY is encoded by the two-dimensional Huffman encoder 28 and output to the output terminal 32.

Similarly, the data of the color difference signal B-Y is input from the input terminal IO, normalized by the data of the weight tilt pull 50 for the color difference signal B-Y, and output to the output terminal 32.

As mentioned above, the luminance signal Y, the color difference signals R-Y and B-Y
The weighting tables 48 and 50 used to normalize the six color difference signals R-Y and B-Y are each normalized using a different weighting table. is set to a large value compared to . Therefore, one color difference signal RY and B
In the normalization of −Y, the data compression rate is higher than in the normalization of the luminance signal Y. For example, the data of the encoded color difference signals The total is one-third of the data of the signal Y.

In the reproduced image, the color difference signals RY and B-Y have less influence on image quality than the luminance signal. Therefore, even if the compression ratio of the color difference signals Y and B-Y is increased as described above, the image does not deteriorate much. As described above, according to the present apparatus, it is possible to increase the compression ratio of image data while maintaining relatively good image quality.

FIG. 2 shows another embodiment of the image signal compression encoding device according to the present invention.

In this device, the image data for each block of the luminance signal Y and the color difference signal R-Y, [1-Y, which are manually input from the input terminal 10, are also sent to the activity calculation section 2o. The activity calculation unit 2o calculates the activity for each block of the luminance signal Y and the color difference signals RY, [1-Y, and calculates the total value for each block, thereby calculating the total activity of the entire image from the luminance signal Y and the color difference signal RY, [1-Y]. Calculate each color difference signal R-Y and B-Y. The output from the activity calculation section 20 is sent to the control section 40.

Tables 52, 54, and 56 store normalization coefficient data for the luminance signal Y, color difference signals RY, and BY, respectively, as in the apparatus shown in FIG. Table 52, 5
Weight setting units 58, 60, and 62 are respectively connected to 4.56. Each of the weight setting units 58, 60, and 62 outputs weight data to be multiplied by data T in tables 52, 54, and 56. The control unit 40 is the activity calculation unit 2
Luminance signal Y and color difference signals RY and B sent from 0
- According to the value of the activity of Y, the weight setting unit 58.
Six weight setting units 58, 60, and 62 output signals instructing weights to units 60, 62, and set weights in accordance with signals from the control unit 40. To set the weight according to the activity value, for example, the weight of the normalization coefficient is set according to the relationship shown in FIGS. 6A and 6B. Weight setting section 5
The weights set by 8.60.62 are output to tables 52.54.56, respectively. Table 52.5
At 4.56, the table data is multiplied by the weight sent from the 1 weight setting unit 58, 60, 62, and the switch 44
sent to. The switch 44 is switched by a control signal from the control section 40 in the same manner as the device shown in FIG. Accordingly, the data in tables 52, 54, or 56 is output to the normalization unit 16.

Further, in this device, the output from the two-dimensional Huffman encoder is sent to a multiplexer 64. On the other hand, the outputs of the weight setting units 58, 60, and 62 are also sequentially output to the multiplexer 64. The multiplexer 64 receives input from the two-dimensional Huffman encoding unit 28 and the weight setting unit 58.
．． 60. Select the input from 62 and output it to the output terminal 32. As a result, compression-encoded image data and weight data for normalization are sent to the output terminal 32.

The rest of the operation of this device is similar to that of the device shown in FIG. 1, so a description thereof will be omitted.

According to this device, as in the device shown in FIG.
When the color difference signal Y or B-Y is input to the normalization section 16, a table 52, 54 or 56 is selected accordingly, and normalization is performed. Therefore, in the normalization of the color difference signals R-Y and B-Y, the data compression rate is higher than in the normalization of the luminance signal Y, and the output encoded data is also Since the compression rate increases, it is possible to increase the compression rate while maintaining relatively good image quality.

Moreover, the weight setting unit 58.6
Since the weight of the normalization coefficient is set at 0.62,
Appropriate compression encoding can be performed according to the frequency components of each signal.

Further, since weight data for each signal for normalization is output together with the encoded image data, the reproduction device can perform decoding using the normalization coefficients.

肱-1 According to the present invention, the compression encoding device sets the normalization coefficient for the luminance signal smaller than the normalization coefficient for the chromaticity signal in the normalization after orthogonal transformation, and converts the chromaticity signal data into a luminance signal. Compress with a higher compression rate than the signal data. Therefore, since the luminance signal data is compressed and encoded with a small compression rate and the chromaticity signal data is compressed and encoded with a large compression rate, relatively good image quality can be achieved even if the compression rate of the image data is increased.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the image signal compression encoding device according to the present invention. FIG. 2 is a block diagram showing another embodiment of the image signal compression encoding device according to the present invention. 4 is a diagram showing an example of pixel data constituting a block; FIG. 4 is a diagram showing an example of 1-weight table data. FIG. 5 is a diagram showing the order in which run lengths and non-zero amplitudes are encoded; FIGS. 6A and 6B are examples of lookup tables for converting the total activity value into normalization coefficient weights. FIG. -14, Two-dimensional orthogonal transformation unit 16, Normalization unit 20, Activity calculation unit 46.48.
5G, weight table 52, 54.56. Table 58.60.628 Weight setting unit 64, , , , multiplexer

Claims (1)

[Claims] 1. An image signal compression encoding device that performs two-dimensional orthogonal transformation encoding of color image data constituting one screen for luminance signal data and chromaticity signal data, the device comprising: orthogonal transformation means for two-dimensional orthogonal transformation of image data; normalization means for normalizing the data orthogonally transformed by the orthogonal transformation means; and encoding means for encoding the data normalized by the normalization means. , normalization coefficient setting means for setting normalization coefficients for the luminance signal data and chromaticity signal data, and the normalization coefficient setting means for setting the normalization coefficients used for the luminance signal data for the chromaticity signal data. The normalization coefficient is set to a value smaller than a normalization coefficient used for the signal data, and the normalization means normalizes the chromaticity signal using the normalization coefficient set by the normalization coefficient setting means. An image signal compression encoding device characterized in that data is compressed at a higher compression rate than the luminance signal data. 2. The apparatus according to claim 1, further comprising an activity calculation means for calculating the activity of the luminance signal data and the chromaticity signal data, and the normalization coefficient setting means is configured to calculate the activity of the luminance signal data and the chromaticity signal data. An image signal compression encoding device characterized in that a normalization coefficient is set for the luminance signal data and chromaticity signal data according to the obtained activities of the luminance signal data and chromaticity signal data. 3. The apparatus according to claim 1 or 2, wherein the normalization coefficient setting means includes means for setting a coefficient according to the activity, and means for selecting a weighting table for the luminance signal data and the chromaticity signal. An image signal compression encoding device, wherein a value obtained by multiplying the set coefficient by the selected weight table is set as the normalization coefficient.