JPH07131787A - Method and device for coding picture data - Google Patents

Abstract

PURPOSE:To encode only a valid block and to obtain code data of uniform quantity as to each frame by deciding respectively a quantization step used for coding processing in response to a valid block number for each frame so as to quantize each frame information. CONSTITUTION:Each of a series of still picture is divided into block and a discrimination means 102 discriminates the presence of correlation of a clock corresponding to a picture of a preceding frame latched in a reference frame latch means 101. Then a count means 122 counts number of valid block for each frame. A quantization step decision means 143 obtains the quantization step corresponding the counting result. A coding means 103 encodes the picture based on the decided quantization step and the coded picture is sent from a transmission means 104.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image data encoding method and apparatus for encoding a moving image by encoding a series of still images for each block consisting of a plurality of pixels.

The data amount of a multi-valued image (hereinafter, simply referred to as an image) such as a halftone image or a color image is so huge that it cannot be treated as it is like other numerical data. For this reason, the image is encoded and the amount of data is compressed with high efficiency, and then stored in the image database or transmitted through the communication line.

As a method of compressing the amount of image data,
There are various methods such as a two-dimensional discrete cosine transform method (DCT method) and a predictive coding method for coding a difference between image data representing the color of each pixel and a predicted value obtained from surrounding image data.

The DCT method is one in which an image is subjected to cosine transform for each block of 8 × 8 pixels and the obtained transform coefficient is encoded. In particular, the adaptive two-dimensional cosine transform method (ADCT method) quantizes transform coefficients using a threshold value adapted to vision, and then encodes
It is known that a high compression rate can be obtained.

This ADCT method is an international standard encoding method (JPEG) for still images by the Joint Photographic Experts Group (JPEG), which is a working group jointly established by the International Telegraph and Telephone Consultative Committee and the International Organization for Standardization.
Method) has been adopted. The JPEG method uses a Huffman code table obtained by statistically examining the appearance frequency of codes,
This is a method of encoding the quantized coefficient obtained by the ADCT method described above.

[0006] The JPEG system is a coding system corresponding to still images, but since it is simple and can obtain high image quality, it is expected to be applied to applications such as video conferences and video telephones for transmitting moving images. .

[0007]

2. Description of the Related Art When the JPEG method is applied to a moving image, the moving image is regarded as composed of a series of still images, and each successive frame is encoded as an independent still image. However, since a moving image is composed of a huge number of frames, simply coding each frame as a still image as described above would result in a huge amount of coding.

As a technique for suppressing the code amount when compressing a moving image by applying the JPEG system, the present applicant has filed Japanese Patent Application No. 4-77957 "Image data encoding / decompression method and apparatus thereof". Has already applied.

According to this technique, when each frame is coded, the correlation with the previous frame is checked for each block, only the block judged to have little correlation is coded as a valid block, and block information indicating the position of the valid block is recorded. It is sent with.

FIG. 16 shows an example of the configuration of an image data encoding device to which this technique is applied. In FIG. 16, the frame memory 201 holds image data corresponding to the image of the previous frame as a reference frame, and the block determination unit 202 compares the image data of each frame with the corresponding block of the reference frame for each block. Then, the presence or absence of correlation is determined. The block discriminating unit 202 discriminates a block determined to have a small correlation as a valid block, outputs discriminating information indicating whether each block is a valid block, and the encoding processing unit according to the discriminating information. 204 is
Only the effective block is encoded, and the frame memory 201
Updates the reference frame with new image data only for valid blocks.

Further, the block information creating unit 203 creates block information indicating the position of the effective block based on the determination for each block of one frame, and the code obtained by the encoding processing unit 204 is the code. The data is sequentially stored in the buffer 205.

In this way, when the processing for all blocks of one frame is completed, the multiplexer 206 first sends the block information obtained by the block information creating unit 203, and then the code buffer 205 stores the block information. The sign of is output.

According to this technique, only the part that has changed from the previous frame is coded in block units, and the code corresponding to the part that does not change much, such as the background part, is not generated, so the code amount of each frame is Can be significantly compressed.

Here, in the above-mentioned encoding processing unit 204, the DCT conversion unit 241 performs a two-dimensional DCT conversion on the image of one block to calculate the DCT coefficient indicating the spatial frequency distribution of the image data, and this DCT coefficient. Quantizer 242
Quantize to obtain the quantized coefficient. Next, the code generation unit 243 reads the obtained quantized coefficients in an order called zigzag scanning, and determines the continuous length of the effective coefficient having a value other than “0” and the invalid coefficient having the subsequent value of “0”. It is configured to be converted into a combination and Huffman-encode these combinations using a code table.

When quantizing DCT coefficients, JPEG recommends applying a quantization step proportional to each component of the quantization matrix shown in FIG. 17 to each spatial frequency component. This quantization matrix is obtained from a value obtained from an experiment in which the human visual sensitivity to each spatial frequency component is examined, and is stored in the quantization matrix holding unit 244 in advance as shown in FIG. , Used to calculate the quantization step.

Generally, before starting the encoding process, a value called a scaling factor SF is input to the register 245, and the space corresponding to the space by the arithmetic processing unit 246 is calculated from the scaling factor SF and each component of the quantization matrix. The quantization step for the frequency component is obtained and used for the quantization in the above-mentioned quantization unit 242.

Here, the above-mentioned scaling factor SF
Is a control coefficient for controlling the quantization step width.In the field of still image coding, the user sets the value of this scaling factor SF in consideration of the characteristics of the image to be coded and the purpose of use. It is often decided and entered. The arithmetic processing unit 246 described above calculates the quantization step by performing, for example, an operation of dividing the product of each component of the quantization matrix and the scaling factor SF by a constant (for example, 50).

On the other hand, the coded data obtained by the above-mentioned coding device is restored to the original image by the image data restoration device shown in FIG. 18, the input code data is divided into block information and a code of a valid block by the code division unit 301, and the code of the valid block is input to the decoding processing unit 302, and the decoding processing unit 30
2, the restoration processing of the image of the corresponding block is performed in the same manner as in the related art. Further, the address generator 303
Based on the block information divided by the code division unit 301, the addresses of effective blocks are sequentially calculated and sent to the restored image holding unit 304. Therefore, the restored image holding unit 304 can restore the image of one frame by sequentially storing the images of valid blocks obtained by the decoding processing unit 302 based on the address from the address generation unit 303. .

[0019]

By the way, in a videophone system, a videoconference system, etc., in order to restore a natural moving image on the receiving side, the image data encoding device on the transmitting side uses the coded data of each frame. Must be sent within a certain time.

However, the above-mentioned Japanese Patent Application No. 4-779.
When the technique of No. 57 is applied as it is, the number of effective blocks changes for each frame, and thus the code data amount for each frame largely changes. Therefore, if the code data obtained by this technique is sent as it is, many frames are sent in a sequence in which frames with few changes are continuous, whereas in a sequence in which large changes are made, Only a few frames were sent out, and it was difficult for the receiving side to restore a natural moving image.

On the other hand, in the field of moving image compression, various techniques have been proposed as techniques for keeping the coded data amount of each frame at a constant value (for example, CCITT SGXV Recommendation H.261).
See).

However, these techniques have a very complicated structure such that the quantization step width can be changed even within one frame in order to keep the flow rate of code data strictly constant. Also, since it is assumed that all blocks of all frames are encoded,
It cannot be directly applied to the technique of Japanese Patent Application No. 4-77957 mentioned above.

It is an object of the present invention to provide an image data encoding method and apparatus which encodes only effective blocks included in each frame of a moving image and obtains an equal amount of encoded data for each frame. To do.

[0024]

FIG. 1 shows the principle of the image data coding method of the first aspect. The invention of claim 1 divides each of a series of still images into blocks composed of a plurality of pixels, and among blocks included in the image of each frame, an effective block having no correlation with the corresponding block of the previous image. In the image data encoding method of determining the effective number of effective blocks and selectively transmitting only effective blocks, the number of effective blocks determined for each frame is counted and used in the encoding process according to the number of effective blocks. It is characterized in that each quantization step is determined and the information of each frame is quantized using the quantization step.

FIG. 2 shows the principle of the image data coding method of claims 2 and 5. The invention according to claim 2 divides each of a series of still images into blocks composed of a plurality of pixels, and among blocks included in the image of each frame, an effective block having no correlation with the corresponding block of the preceding image. In the image data coding method of discriminating the effective block, selectively encoding and transmitting only the effective block, the number of the effective blocks determined for each frame is counted, and the effective block of the effective block is set using the set quantization step. The encoding process is performed, and the setting value of the quantization step used in the encoding process is changed according to the number of effective blocks.

FIG. 3 shows the principle of the image data encoding method of claim 3. According to a third aspect of the invention, in the image data coding method according to the first aspect, when the number of valid blocks exceeds a predetermined threshold value, all blocks of the corresponding frame are set as valid blocks and correspond to the frames. It is characterized in that a predetermined value larger than the standard quantization step is set as the quantization step.

FIG. 4 shows the principle of the image data encoding method of claim 4. According to a fourth aspect of the present invention, in the image data encoding method according to the first aspect, the number of consecutive frames in which the number of valid blocks is equal to or less than a predetermined threshold value is counted based on the number of valid blocks of each frame, and When the number of consecutive frames to be exceeded exceeds a predetermined threshold, a predetermined value smaller than the standard quantization step is set as the quantization step of the current frame, and all blocks of the current frame are set as valid blocks. It is characterized by

According to a fifth aspect of the present invention, in the image data encoding method according to the second aspect, a new quantization step is obtained based on the number of effective blocks of each frame and the code amount thereof, and the quantization step is set. It is characterized by changing the value.

FIG. 5 shows the principle of the image data encoding method of claim 6. According to a sixth aspect of the present invention, in the image data encoding method according to the second aspect, the number of consecutive frames in which the number of valid blocks is equal to or less than a predetermined threshold is counted based on the number of valid blocks of each frame, When the number of consecutive times of exceeds a predetermined threshold value, it is decided to perform refresh in the next frame, and the setting value of the quantization step is changed by a predetermined value smaller than the standard quantization step, and All the blocks of the frame are set as effective blocks.

FIG. 6 shows a block diagram of the principle of the image data encoding device according to the seventh aspect. According to a seventh aspect of the present invention, each of the series of still images is divided into blocks composed of a plurality of pixels, and for each block of the image of each frame, the corresponding block of the image of the previous frame held in the reference frame holding means 101. Whether or not there is a correlation with is discriminated by the discriminating means 102, and only effective blocks having no correlation are encoded by the encoding means 1.
In the image data encoding device for encoding, the image data encoding device for transmitting the code corresponding to the effective block and the block information indicating the position of the effective block by the transmitting means 104 temporarily encodes the image data of the current frame to be encoded. The current frame holding unit 111 that holds and sequentially outputs the image data of each held block according to the input of new image data, and the discrimination information obtained by the discrimination unit 102 for each block of the current frame. Discrimination information holding unit 112 that holds the same and outputs the corresponding discrimination information to the encoding unit 103 and the sending unit 104 in synchronization with the output of the image data from the current frame holding unit 111.
A quantization step setting means 113 for obtaining a quantization step corresponding to the number of effective blocks in the current frame based on the discrimination information of each block of the current frame is provided, and the encoding means 103 is obtained by the quantization step setting means 113. Discrimination information holding means 112 using the determined quantization step.
It is characterized in that it is a configuration for quantizing the information of the effective block indicated by the output.

FIG. 7 shows a block diagram of the principle of the image data encoding device according to the eighth and eleventh aspects. According to the invention of claim 8, each of the series of still images is divided into blocks composed of a plurality of pixels, and for each block of the image of each frame, the corresponding block of the image of the previous frame held in the reference frame holding means 101. Whether or not there is a correlation with is discriminated by the discriminating means 102, only the effective block having no correlation is encoded by the encoding means 103, and the transmitting means 104 outputs the code corresponding to the effective block and the block information indicating the position of the effective block. In the image data encoding device to be transmitted, the encoding means 103 is configured to include the quantization step holding means 121 for holding the setting value of the quantization step used when quantizing the information of the effective block,
A counting means 122 for counting the number of effective blocks for each frame based on the discrimination information from the discrimination means 102;
A quantization step updating means 123 for obtaining a quantization step corresponding to the counting result by the counting means 122, and updating the content of the quantization step holding means 121 with a new quantization step after completion of the encoding processing of each frame. It is characterized by that.

FIG. 8 shows the essential parts of the image data coding apparatus according to claims 9 and 10. According to a ninth aspect of the present invention, in the image data encoding apparatus according to the seventh aspect, it is determined whether the number of valid blocks in the current frame exceeds a predetermined threshold value based on the content of the determination information holding unit 111. Discrimination information holding means 1 according to the means 131 and the input of the determination result indicating that the number of valid blocks exceeds a predetermined threshold value.
In place of the output of 12, the switching means 132 for sending the determination information indicating that all blocks are valid is provided, and the quantization step setting means 113 determines that the number of valid blocks exceeds a predetermined threshold value. It is characterized in that a predetermined quantization step larger than the standard quantization step is output according to the input of the result.

According to a tenth aspect of the invention, in the image data coding apparatus according to the seventh aspect, the discrimination information holding means 112 is provided.
A frame detecting means 133 for detecting a frame in which the number of effective blocks is equal to or less than a predetermined threshold value, and a frame counting means 134 for counting the number of consecutive corresponding frames in accordance with the output of the frame detecting means 133. , The determination information holding unit 112 according to the determination unit 135 that determines whether or not the number of consecutive corresponding frames exceeds a predetermined threshold, and the determination result that the number of consecutive corresponding frames exceeds the predetermined threshold. In place of the output of, the switching unit 136 that sends the determination information that all the blocks are valid is provided, and the quantization step setting unit 113 indicates that the number of consecutive corresponding frames exceeds a predetermined threshold. It is characterized in that a predetermined quantization step smaller than the standard quantization step is output according to the input of the determination result.

The invention of claim 11 is the image data encoding device according to claim 8, further comprising a measuring means 141 for measuring the code amount of the current frame obtained by the encoding means 103, and the quantization step updating means. 123 is 1 based on the code amount of the current frame and the counting result by the counting means 122.
Code amount calculating means 14 for calculating the code amount per block
2, and a quantization step determining unit 143 that determines a new quantization step based on the counting result of the counting unit 122 and the code amount per block.

[0035]

According to the first aspect of the present invention, the information of each frame is quantized by using the quantization step according to the number of effective blocks of each frame. For example, AD the effective block
When encoded using the CT method, the code data amount of each effective block is substantially the same in a sequence in which similar images are continuous, so it is considered that the number of effective blocks and the code data amount are almost proportional. Therefore, it is possible to equalize the code data amount by using the quantization step according to the number of effective blocks.

According to the second aspect of the invention, the quantization step of the next frame is changed according to the number of effective blocks in each frame. In this case, it is not possible to strictly equalize the code data amount of each frame, but as a whole,
It is possible to make the flow rate of code data close to uniform.

According to the third aspect of the present invention, a frame including a predetermined number or more of effective blocks is detected, all the blocks of this frame are set as effective blocks, the restored image on the receiving side is refreshed, and standard quantization is performed. The information of this frame is encoded using a quantization step larger than the step. As a result, compared with the case of randomly refreshing, refreshing can be performed more efficiently, and at this time, by using the quantization step that can obtain a high compression rate, the coded data amount of the corresponding frame can be reduced. The flow rate of the code data can be made uniform by suppressing the flow rate.

According to a fourth aspect of the present invention, when a predetermined number of frames with a small number of effective blocks continue, all blocks of the current frame are set as effective blocks, and a quantization step smaller than the standard is applied to the current frame. To do. In this way, by performing refreshing when the flow rate of code data is small, the transmission rate of the transmission line can be effectively used. In addition, at this time, since the quantization step that can obtain high image quality is used, the image quality of the restored image can be improved by refreshing, and particularly, in the part where the change is scarce such as the background part, that part becomes an effective block. High image quality is maintained until it is determined.

The fifth aspect of the present invention changes the quantization step of the next frame based on the number of effective blocks and the code amount of each frame. In this way, by determining the quantization step in consideration of the code amount, it is possible to deal with the change in the code amount per block due to the change in the property of the image.

According to a sixth aspect of the present invention, when a frame having a small number of effective blocks continues for a predetermined number, all blocks of the next frame are set as effective blocks and a quantization step smaller than the standard is used for the frame. Is applied. Thereby, like the invention of claim 4,
It is possible to effectively use the transmission rate of the transmission path and improve the quality of the restored image.

According to a seventh aspect of the present invention, the current frame holding means 1 is provided.
The image data and the discrimination information are respectively held by 11 and the discrimination information holding unit 112, and after the discrimination processing of the block for one frame is completed, the quantization step setting unit 113 responds to the number of effective blocks. The quantization step is set, and the encoding means 103 performs the encoding process using this quantization step. As a result, the invention of claim 1 can be applied to an image data encoding device that encodes only valid blocks and sends them together with block information, and the coded data amount of each frame can be made uniform.

According to the eighth aspect of the invention, the quantization step updating means 123 calculates the quantization step according to the number of effective blocks based on the counting result of the counting means 122, and the quantization provided in the encoding means 103. By updating the step holding means 121, it is possible to apply the invention of claim 2 and bring the code data closer to uniform as a whole.

According to the ninth aspect of the invention, the switching means 132 and the quantization step setting means 113 operate according to the determination result of the determination means 131, so that the invention of the third aspect is applied to efficiently restore the image. With refreshing
It is possible to suppress the protrusion of the code data amount.

According to a tenth aspect of the invention, the frame detecting means 1
33, the frame counting means 134, and the determining means 135 detect a place where consecutive frames having a small number of effective blocks are detected, and the quantization step setting means 113 and the switching means 136 are operated in accordance with the determination result of the determining means 135. By operating, it is possible to apply the invention of claim 4 to effectively utilize the transmission rate of the transmission path for refreshing and improve the image quality of the restored image.

In the eleventh aspect of the invention, the code amount calculation unit 141 calculates the code amount per block based on the code amount and the number of effective blocks obtained by the measurement unit 141, and the code amount per block. The quantization step determining means 142 determines the quantization step to be applied to the next frame based on the number of effective blocks and the number of effective blocks. As a result, by applying the invention of claim 4, it is possible to make the code data uniform while flexibly responding to changes in the characteristics of the image.

[0046]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 9 shows a block diagram of an embodiment of the image data encoding apparatus of claim 7.

In FIG. 9, the image data coding apparatus according to claim 7 has a frame memory 211, a discrimination information holding section 212 and a control coefficient setting section 220 added to the conventional image data coding apparatus shown in FIG. Is configured.

This frame memory 211 corresponds to the current frame holding means 111, has a capacity equivalent to one frame of image data, and responds to the input of each block of a new frame to the previous frame. The image data of the corresponding block is sequentially sent to the encoding processing unit 204. Further, the discrimination information holding unit 212 corresponds to the discrimination information holding means 112, and in response to the input of the discrimination information about each block of the new frame,
The discrimination information of the corresponding block of the previous frame is output and sent to the block information creation unit 203.
For example, the discrimination information holding unit 212 may be configured using a FIFO having a bit length corresponding to all blocks in one frame.

Further, the frame memory 201 corresponds to the reference frame holding means 101, and in the same manner as in the conventional case, the image data of the block inputted according to the output of the block discriminating section 202 corresponding to the discriminating means 102. Is used to update the image of the reference frame. On the other hand, an encoding processing unit 204 corresponding to the encoding unit 103
Is configured to encode only valid blocks according to the discrimination information output from the discrimination information holding unit 212.

That is, the frame memory 211 and the discrimination information holding section 212 delay the image data and the discrimination information by one frame, and the control coefficient setting section 220 obtains them in parallel with the discrimination processing of the effective block of the next frame. The encoding processing unit 204 is configured to perform encoding processing of each frame using the obtained scaling factor SF.

The control coefficient setting section 220 corresponds to the quantization step setting means 113. In the control coefficient setting unit 220, the counter 221 is configured to count the number of blocks validated by the block determination unit 202 for each frame and send the counting result Nc to the divider 223 and the register 222. . For example, the counter 221 may be configured to output the count result Nc in response to a frame end signal indicating that the input of one frame of image data has ended, and reset the count value to the initial value “0”.

The register 222 described above uses the newly input count result Nc as the number of valid blocks in the previous frame.
It is configured to hold as No and output the held contents. Therefore, the divider 223 divides the count result Nc of the counter 221 by the output of the register 222 to obtain the ratio D of the number of effective blocks of the current frame and the previous frame. Next, the multiplier 225
Multiplies the ratio D of the number of effective blocks by the scaling factor SF of the previous frame held in the register 224, inputs the multiplication result as a new scaling factor SF into the register 245 of the encoding processing unit 204 and the register 224. To store.

Now, using the same quantization step, D
When the CT coefficient is quantized, the code amount per block often becomes almost the same value, and therefore the code amount per frame is proportional to the number of effective blocks in that frame. Therefore, as described above, if the scaling factor SF is calculated according to the ratio D of the number of effective blocks to the previous frame and is input to the register 245 of the encoding processing unit 204, the quantization is performed according to the variation of the number of effective blocks. The size of the steps can be controlled. This makes it possible to apply a large quantization step to a frame with many effective blocks and conversely to a small quantization step to a frame with few effective blocks. You can

Before encoding the first frame,
The scaling factor SF adapted to a standard image may be set as the initial value in the registers 224 and 245 described above, and the standard number of valid blocks may be set in the register 222 as the initial value. Further, the image of the first frame may be encoded using the initial value of the scaling factor SF described above, and the scaling factor SF may be set as described above for the second and subsequent frames.

The code data of each block thus obtained is sequentially held in the code buffer 205, and when the coding process for one frame is completed, the blocks are processed by the multiplexer 206 in the conventional manner. It is transmitted subsequently to the block information obtained by the information creating unit 203, and is transmitted to the transmission path via the transmission buffer 207. That is, the function of the sending unit 104 is realized by these respective units.

Further, by transmitting the information about the scaling factor SF set together with the block information, the receiving side can perform the inverse quantization process using the same scaling factor SF. For example, the set scaling factor SF may be input to the block information creation unit 203, and information regarding this scaling factor SF may be added before or after the block information.

The maximum value SF of the scaling factor SF
_MAX set in advance to, the control coefficient setting unit 220, when the scaling factor SF obtained by the multiplier 225 exceeds this maximum value SF _MAX is the maximum value SF _MAX
It may be configured to be replaced with and transmitted to the encoding processing unit 204.

In this case, if the value of the scaling factor SF corresponding to the minimum image quality required in the applied system is set as the above-mentioned maximum value SF _MAX , the scaling factor SF will not increase without limit. The code amount can be made uniform while maintaining the image quality to some extent.

Further, by applying the invention of claim 9, when the number of valid blocks exceeds a predetermined number, all the blocks are set as valid blocks, and the control coefficient setting unit 220 sets the predetermined scaling factor SF _a . It may be configured to be set.

In this case, as shown in FIG. 10, a multiplexer (symbol MP in the figure is added to the control coefficient setting unit 220.
226 is added and the counter 221 is added.
The comparator 213 compares the counting result Nc of the above with a predetermined threshold value Nth, and the multiplexer 226 performs a predetermined scaling instead of the output of the multiplier 225 according to the comparison result indicating that the counting result Nc exceeds the threshold value Nth. It may be configured to output the factor SF _a .

In this case, the function of the determination means 131 is realized by the comparator 213, and the function of the quantization step setting means 113 of claim 9 is realized by adding the multiplexer 226 to the control coefficient setting section 220. ing.

Here, the above-mentioned scaling factor SF
_{As a} , _a value with which a high compression rate is obtained may be set, and for example, the maximum value SF _MAX of the scaling factor SF corresponding to the necessary minimum image quality may be set. As the threshold value Nth described above, for example, a number corresponding to 95% of the total number of blocks is set, and the comparator 2
13 may be configured to output a logic "1" when the counting result Nc exceeds this threshold Nth.

The output of the comparator 213 and the output of the discrimination information holding unit 212 are input to the OR gate 214, and the output of the OR gate 214 is used as the discrimination information of each block of the current frame in the block information creating unit 203 and the code. It suffices to input to the conversion processing unit 204.

As a result, when the output of the comparator 213 indicates that the counting result Nc exceeds the threshold value Nth, a logical "1" is output as the discrimination information of all blocks,
The function of the switching unit 132 is realized by the OR gate 214, and the discrimination information indicating that all the blocks of the corresponding frame are valid blocks can be transmitted.

In response to the output of the OR gate 214, the coding processing section 204 performs coding processing of all blocks of the corresponding frame, and the block information creating section 20.
3 creates block information based on this discrimination information.

As described above, when most of one frame is regarded as a valid block, all the blocks are coded as valid blocks so that the restored image on the receiving side is refreshed by the information of this frame. , The error accumulated in the background part can be eliminated.
Further, at this time, by increasing the quantization step width, it is possible to minimize the increase in the code amount.

Further, as shown in FIG. 11A, the control coefficient setting section 220 comprises a counter 221 and a look-up table (indicated by LUT in the figure) 227, and the count result Nc by the counter 221 is looked up. The address may be input to the table 227 to obtain the scaling factor SF corresponding to the number of effective blocks.

In this case, as the relationship between the number of effective blocks Nc and the scaling factor SF, not only a simple proportional relationship but also various functions can be applied. Also,
Since an arithmetic circuit is not required, the circuit configuration can be simplified, and since the information on the number of effective blocks of the previous frame is not required, it can be applied to the encoding process of the first frame as it is. . Further, as shown in FIG. 11B, the control coefficient setting unit 220 is composed of a counter 221, a comparison circuit 228, and a multiplexer (indicated by the symbol MPX in the drawing) 229, and the comparison circuit 228 determines the number of effective blocks Nc. A configuration may be adopted in which a plurality of thresholds Th1 to Thn are compared to determine which of the sections divided by these thresholds the effective block number Nc belongs to, and the multiplexer 229 is controlled according to the determination result.

In this case, the above-mentioned n threshold values Th _{1 to} Th
_If the scaling factors SF _{1 to} SF _{n + 1} corresponding to the n + 1 sections divided by n are set in the multiplexer 229 in advance, the scaling factors corresponding to the corresponding section according to the value of the number of effective blocks Nc. SF is obtained.
Also in this case, since the arithmetic circuit is unnecessary, the circuit configuration can be simplified and the speed can be increased.

Further, the invention of claim 10 may be applied so that all the blocks are coded by using a quantization step for high image quality when frames with few effective blocks are continuous.

In FIG. 12, the image data encoding device of claim 10 is the same as the image data encoding device shown in FIG.
Frame counter 230, selector 215, and OR gate 2
And 14 are added.

The frame counting section 230 corresponds to the frame detecting section 131 and the frame counting section 134, and the comparator 231 compares the counting result Nc of the counter 221 of the control coefficient setting section 220 with a predetermined threshold value Nr. Then, the corresponding frame is detected, and the counter 232 counts the number of consecutive frames having a small number of valid blocks according to the comparison result. As the above-mentioned threshold Nr, for example, the number of blocks corresponding to 10% of all blocks is set, and the counter 232 sets the count value to 1 according to the comparison result indicating that the number of effective blocks Nc is equal to or less than the threshold Nr. The count value may be reset and the count value may be reset to the initial value “0” according to the comparison result indicating that the number of valid blocks Nc exceeds the threshold value Nr. Further, the comparator 233 corresponds to the judging means 135, compares the count value of the counter 232 with another threshold value Fth, and when the number of consecutive frames with a small number of blocks exceeds the threshold value Fth, the logic “1” is set. Is output, and when it is less than or equal to the threshold value Fth, a logical “0” is output and sent as a switching instruction to the selector 215. Further, the counter 232 may reset the count value according to the comparison result indicating that the number of consecutive frames having a small number of blocks exceeds the threshold Fth.

When logic "0" is input as the switching instruction, the selector 215 selects the scaling factor SF obtained by the control coefficient setting unit 220, and the predetermined scaling is performed according to the input of logic "1". The configuration is such that the factor SFr is selected and sent to the encoding processing unit 204.
That is, the quantization step setting means 1 according to claim 10 is controlled by the control coefficient setting section 220 and the selector 215.
It is configured to realize 13 functions. Here, as the above-mentioned predetermined scaling factor SFr, a value with which high image quality can be obtained may be set.

By inputting the output of the comparator 254 and the output of the discrimination information holding unit 212 into the OR gate 214, when the number of consecutive frames having a small number of blocks exceeds the threshold Fth, the discrimination information holding unit A logical "1" is obtained as the output of this OR gate 214 regardless of the output of 212. That is, the function of the switching unit 136 is realized by the OR gate 214, and the determination information indicating that all the blocks are valid for the corresponding frame is the block information creation unit 203 and the encoding processing unit 204.
Sent to.

In response to this, all blocks of the current frame held in the frame memory 211 are coded as valid blocks, and the block information creation unit 203 blocks all the blocks to be valid blocks. Information is created.

Here, when frames with a small number of effective blocks are consecutive, it is expected that there will be a margin in the storage capacity and transmission rate of the transmission buffer 207. For example, if a frame having a code amount which is a fraction of the standard one frame continues, a space corresponding to the code amount of the refresh frame is considered to be created in the transmission buffer 207 and the transmission path. Therefore, considering the ratio of the transmission rate and the code amount, the threshold value Fth of the number of consecutive frames described above is
If the above is determined, the restored image can be refreshed in consideration of the transmission rate and the margin of the transmission buffer 207, and compared with the case where the refresh is simply performed periodically.
The capacity of the transmission buffer 207 and the transmission path can be efficiently used.

Further, at this time, the image quality of the refreshed restored image can be improved by using the scaling factor SFr capable of obtaining the high image quality. In particular, with respect to a portion having a small change such as a background portion, high image quality can be maintained until the portion is updated as an effective block.

Generally, some deterioration in image quality is hard to be sensed in a portion where the change is drastic, but deterioration in image quality is sensitively sensed in a portion where the change is small, such as the background, which may cause a discomfort to the user. Therefore, as described above, by encoding the refresh frame using the scaling factor SFr for high image quality, it is possible to give the user the impression that the image quality is generally improved.

Next, an embodiment of the image data coding apparatus according to claim 8 will be described. FIG. 13 shows a block diagram of an embodiment of the image data encoding apparatus of claim 8. In FIG. 13, the image data encoding device according to claim 8 is the same as the conventional image data encoding device shown in FIG.
After the control coefficient setting unit 220 shown in FIG. 2 is added and the encoding processing for the one frame by the encoding processing unit 204 is completed, the control coefficient setting unit 220 stores the new scaling factor SF in the register 245 of the encoding processing unit 204. It is configured to input.

Here, the control coefficient setting section 220 shown in FIG. 9 or 11 is configured to calculate the scaling factor SF according to the frame end signal indicating that the input of the image of one frame is completed. Therefore, it can be applied as it is to the image data encoding device according to claim 8. In FIG. 13, the control coefficient setting unit 2 shown in FIG.
The case where 20 is applied is shown.

In this case, it is necessary to set the initial value of the scaling factor SF in the register 245 of the encoding processing unit 204, and the quantization step using this initial value of the scaling factor SF is applied in the first frame. To be done. In each frame after the second frame, the quantization step using the scaling factor SF according to the number of effective blocks in the previous frame is applied.

In this way, when the quantization step is changed according to the number of effective blocks of the previous frame, strictly speaking, the code amount of each frame cannot be made uniform,
The flow rate of code data as a whole can be made to approach uniformly. Further, in this case, since it is not necessary to delay the image data and the discrimination information by one frame, it can be realized with a simple circuit configuration.

By the way, in a sequence in which similar images are continuous, the code amount per effective block is almost constant, but before and after a scene change, the features of the image are often completely different. The code amount per effective block is often significantly different.

According to the eleventh aspect of the image data encoding device, the quantization step can be set in consideration of the variation of the code amount per block as described above. In Figure 14,
The block diagram of the Example of the image data coding apparatus of Claim 11 is shown.

In FIG. 14, the image data coding apparatus according to claim 11 is the same as the conventional image data coding apparatus shown in FIG. 16, except that the control coefficient setting unit 220 shown in FIG. Scaling factor SF from the fluctuation of
And a control coefficient setting unit 2 for calculating
20 or a selector 216 that selectively outputs the output of the control coefficient calculation unit 250.

In the control coefficient calculation unit 250, the divider 251 corresponds to the code amount calculation means 142, and the information about the code amount for one frame from the code buffer 205 and the counter 221 of the control coefficient setting unit 220. The number of effective blocks Nc is input, and the code amount Bc per block of the frame is calculated. In this case, the function of the weighing means 141 is realized by the code buffer 205 notifying the code amount for one frame.

The code amount Bc described above is input to the divider 252 as a dividend and is also sent to the register 253. By this register (denoted by symbol R in the figure) 253, the code amount Bo per block of the previous frame is obtained. Retained. Further, at this time, the register 253 inputs the held code amount Bo as a divisor of the divider 252. Therefore, the divider 252 causes the current frame 1
The ratio K of the code amount per block to the previous frame is obtained. Further, the multiplier 254 multiplies the current scaling factor SF held in the register (indicated by symbol R in the figure) 255 by the above-mentioned code amount ratio K, and the multiplication result is used as a new scaling factor SF in the selector 21.
It is configured to be sent to 6.

Further, the comparator 256 compares the above-mentioned code amount ratio K with predetermined thresholds Th _k1 and Th _k2 , and compares the code amount ratio K.
Is included in the section indicated by the thresholds Th _k1 and Th _k2 , and the selector 21 uses the determination result as a switching instruction.
It is configured to be sent to 6.

This selector 216 selects the scaling factor SF from the control coefficient setting unit 220 according to the result of the determination that the code amount ratio K is included in the above-mentioned section,
The scaling factor SF from the control coefficient calculation unit 250 is selected and output according to the determination result that it is not included in the above-described section.

The output of the selector 216 is input to the register 224 of the control coefficient setting unit 220 and the register 255 of the control coefficient calculation unit 250 and held as the current scaling factor SF.

In this way, when the code amount per block greatly changes, the selector 216 is switched to select the scaling factor SF calculated according to the code amount per block. , The scaling factor SF can be set according to the variation of the number of effective blocks with reference to this scaling factor SF.

That is, the function of the quantization step determining means 143 is realized by the control coefficient setting unit 220, the selector 216, and the control coefficient calculating unit 250, and not only the number of valid blocks but also the block per block due to the change of the characteristics of the image. It is possible to set an appropriate scaling factor SF in consideration of the change in the code amount of.

As a result, due to a scene change, etc.
Flexible even when the characteristics of the image change significantly,
The flow rate of code data can be made uniform. When the invention of claim 6 is applied to the image data encoding device of claim 8, the reconstructed image is refreshed using a quantization step that can obtain high image quality when a frame with a small number of effective blocks continues. It is also possible.

For example, as shown in FIG. 15, a frame counting section 230, a selector 215, and an OR gate 214 may be added to the image data coding apparatus shown in FIG.

In this case, the selector 215 outputs the output of the control coefficient setting unit 220 and the predetermined scaling factor SFr according to the output of the comparator 233 of the frame counting unit 230, as in the embodiment shown in FIG. You can select either of the above. Further, the OR gate 214 may be configured to output the logical sum of the output of the block discrimination unit 202 and the output of the comparator 233.

As a result, when the number of consecutive frames with few valid blocks exceeds the threshold Fth, all blocks of the next frame are coded as valid blocks using the quantization step obtained from the scaling factor SFr. Therefore, it is possible to perform refreshing while considering the margin of the transmission rate and improve the quality of the refreshed restored image.

[0097]

As described above, according to the present invention, when only effective blocks are discriminated and encoded from a series of images, by using a quantization step according to the number of effective blocks of each frame, Since it is possible to equalize the code amounts of, the natural moving image can be reproduced on the receiving side. Further, when most of one frame is an effective block, all blocks of the frame are encoded by using a quantization step that can obtain a high compression rate, so that the code amount is made uniform and the restored image is reproduced. You can refresh efficiently.

By changing the quantization step of the next frame according to the number of effective blocks in each frame, the flow rate of code data can be made uniform as a whole as a whole. Further, in this case, by considering both the number of effective blocks and the code amount of each frame, the quantization step of the next frame is determined, so that the change of the code amount due to the change of the property of the image is flexible. Accordingly, the flow rate of code data can be made uniform.

Further, when a frame having a small number of effective blocks continues, the transmission rate is efficiently used by encoding all the blocks of the frame or the next frame by using the quantization step for high image quality. Thus, the restored image can be refreshed and the image quality of the restored image can be improved.

[Brief description of drawings]

FIG. 1 is a diagram showing a principle of an image data encoding method according to claim 1;

FIG. 2 is a diagram showing the principle of the image data encoding method of claims 2 and 5;

FIG. 3 is a diagram showing the principle of the image data encoding method of claim 3;

FIG. 4 is a diagram showing the principle of the image data encoding method of claim 4;

FIG. 5 is a diagram showing the principle of the image data encoding method of claim 6;

FIG. 6 is a block diagram showing the principle of the image data encoding device according to claim 7;

FIG. 7 is a block diagram showing the principle of the image data encoding device according to claims 8 and 11;

FIG. 8 is a diagram showing a main part of the image data encoding device according to claims 9 and 10;

FIG. 9 is a configuration diagram of an embodiment of the image data encoding device of claim 7;

FIG. 10 is a configuration diagram of an embodiment of the image data encoding device of claim 9;

FIG. 11 is a block diagram of another embodiment of a control coefficient setting unit.

[Fig. 12] Fig. 12 is a configuration diagram of an embodiment of the image data encoding device of claim 10.

FIG. 13 is a configuration diagram of an embodiment of the image data encoding device of claim 8;

[Fig. 14] Fig. 14 is a configuration diagram of an embodiment of the image data encoding device of claim 11.

FIG. 15 is a configuration diagram of another embodiment of the image data encoding device of claim 8;

FIG. 16 is a configuration diagram of a conventional image data encoding device.

FIG. 17 is a diagram showing a quantization matrix.

FIG. 18 is a diagram showing a configuration example of a conventional image data restoration device.

[Explanation of symbols]

 101 Reference Frame Holding Means 102 Discriminating Means 103 Encoding Means 104 Sending Means 111 Current Frame Holding Means 112 Discrimination Information Holding Means 113 Quantization Step Setting Means 121 Quantization Step Holding Means 122 Counting Means 123 Quantization Step Updating Means 131, 135 Determination Means 132, 136 Switching means 133 Frame detecting means 134 Frame counting means 141 Metering means 142 Code amount calculating means 143 Quantization step determining means 201, 211 Frame memory 202 Block discriminating section 203 Block information creating section 204 Encoding processing section 205 Code buffer 206, 226, 229 Multiplexer (MPX) 207 Transmission buffer 212 Discrimination information holding unit 213, 231, 233, 256 Comparator 214 OR gate 215, 216 Selector 220 Control coefficient setting unit 221,232 Counter 222,224,245,253,255 Register 223,251,252 Divider 225,254 Multiplier 227 Look-up table (LUT) 228 Comparison circuit 230 Frame counter 241 DCT converter 242 Quantization unit 243 Code generation unit 244 Quantization matrix holding unit 246 Arithmetic processing unit 250 Control coefficient calculation unit

Claims (11)

[Claims] 1. A series of still images is divided into blocks each composed of a plurality of pixels, and an effective block having no correlation with a corresponding block of a preceding image is determined from blocks included in the image of each frame. Then, in the image data encoding method for selectively encoding and transmitting only the valid blocks, the number of valid blocks determined for each frame is counted and used for the coding process according to the number of valid blocks. An image data encoding method, characterized in that each quantization step is determined, and the information of each frame is quantized using the quantization step. 2. A series of still images is divided into blocks each composed of a plurality of pixels, and an effective block having no correlation with the corresponding block of the preceding image is determined from the blocks included in the image of each frame. Then, in the image data encoding method for selectively encoding and transmitting only the effective block, the number of effective blocks determined for each frame is counted, and the effective block of the effective block is set by using the set quantization step. An image data encoding method comprising performing an encoding process and changing a setting value of a quantization step used in the encoding process according to the number of effective blocks. 3. The image data coding method according to claim 1, wherein when the number of valid blocks exceeds a predetermined threshold, all blocks of the corresponding frame are set as valid blocks and the quantum corresponding to the frame is set. An image data encoding method, wherein a predetermined value larger than a standard quantization step is set as the encoding step. 4. The image data encoding method according to claim 1, wherein the number of consecutive frames in which the number of valid blocks is equal to or less than a predetermined threshold is counted based on the number of valid blocks of each frame, and the corresponding frame When the number of consecutive times exceeds a predetermined threshold, a predetermined value smaller than the standard quantization step is set as the quantization step of the current frame, and all blocks of the current frame are set as valid blocks. Characteristic image data encoding method. 5. The image data coding method according to claim 2, wherein a new quantization step is obtained based on the number of effective blocks of each frame and the code amount thereof, and the setting value of the quantization step is changed. An image data encoding method characterized by: 6. The image data coding method according to claim 2, wherein the number of consecutive frames in which the number of valid blocks is equal to or less than a predetermined threshold is counted based on the number of valid blocks of each frame, When the number of consecutives exceeds a predetermined threshold, it is decided to refresh in the next frame, and the set value of the quantization step is changed by a predetermined value smaller than the standard quantization step. An image data encoding method, characterized in that all blocks of a frame are effective blocks. 7. A series of still images are each divided into blocks composed of a plurality of pixels, and for each block of the image of each frame, a corresponding block of the image of the previous frame held in the reference frame holding means (101). Whether or not there is a correlation with is discriminated by the discriminating means (102), and only the effective block having no correlation is encoded by the encoding means (103),
In an image data encoding device for transmitting a code corresponding to the effective block and block information indicating the position of the effective block by a transmission means (104), temporarily holds the image data of the current frame to be encoded. And input new image data,
A current frame holding means (111) for sequentially outputting the held image data of each block, and for each block of the current frame, holding the discrimination information obtained by the discrimination means (102), In synchronization with the output of the image data from the holding means (111), the corresponding discrimination information is sent to the encoding means (1
03) and the discrimination information holding means (112) to be transmitted to the transmission means (104) and the quantization step for obtaining the quantization step corresponding to the number of effective blocks in the current frame based on the discrimination information of each block of the current frame. And a step setting means (113), wherein the encoding means (103) uses the quantization step obtained by the quantization step setting means (113) to output the discrimination information holding means (112). An image data encoding device having a configuration for quantizing the information of the indicated effective block. 8. A series of still images are each divided into blocks composed of a plurality of pixels, and for each block of the image of each frame, the corresponding block of the image of the previous frame held in the reference frame holding means (101). Whether or not there is a correlation with is discriminated by the discriminating means (102), and only the effective block having no correlation is encoded by the encoding means (103),
In an image data encoding device for transmitting a code corresponding to the effective block and block information indicating a position of the effective block by a transmission means (104), the encoding means (103) quantizes the information of the effective block. This is a configuration including a quantization step holding unit (121) that holds a setting value of a quantization step used when performing quantization, and based on the discrimination information from the discrimination unit (102), the effective block of each frame is determined. Counting means (122) for counting the number and a quantization step corresponding to the counting result by the counting means (122), and the content of the quantization step holding means (121) after the coding process of each frame is completed. And a quantization step update means (123) for updating the image data with a new quantization step. 9. The image data encoding device according to claim 7, wherein it is determined whether the number of valid blocks in the current frame exceeds a predetermined threshold value based on the content of the determination information holding unit (111). In response to the input of the means (131) and the determination result indicating that the number of valid blocks exceeds a predetermined threshold, instead of the output of the determination information holding means (112), it is determined that all blocks are valid. And a switching means (132) for sending information, wherein the quantization step setting means (113) receives a standard quantization step in response to an input of a determination result indicating that the number of effective blocks exceeds a predetermined threshold value. An image data encoding device having a configuration for outputting a predetermined quantization step larger than the above. 10. The image data coding apparatus according to claim 7, wherein a frame detection unit (133) detects a frame whose number of valid blocks is equal to or less than a predetermined threshold value based on the content of the discrimination information holding unit (112). ) And a frame counting means (1) for counting the number of consecutive corresponding frames according to the output of the frame detecting means (133).
34), a determination means (135) for determining whether or not the number of consecutive corresponding frames exceeds a predetermined threshold, and a determination result indicating that the number of consecutive corresponding frames exceeds a predetermined threshold. In addition to the output of the discrimination information holding means (112), there is provided a switching means (136) for transmitting discrimination information indicating that all the blocks are valid, and the quantization step setting means (113) corresponds to the above. An image data code having a configuration in which a predetermined quantization step smaller than the standard quantization step is output in response to an input of a determination result that the number of consecutive frames exceeds a predetermined threshold value. Device. 11. The image data coding apparatus according to claim 8, further comprising: a measuring means (141) for measuring the code amount of the current frame obtained by the encoding means (103), and a quantization step updating means ( 123) is a code amount calculating means (142) for calculating a code amount per block based on the code amount of the current frame and the counting result by the counting means (122), and counting by the counting means (122). An image data coding apparatus, comprising: a quantization step determining means (143) for determining a new quantization step based on a result and the code amount per block.