JP2918360B2 - Inverse quantization method and image data restoration device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image data restoration apparatus for restoring an original image from code data, and to an inverse quantization method for inverse quantization of a quantized coefficient obtained by decoding code data, and an inverse quantization method for the same. The present invention relates to an image data restoration apparatus to which the method is applied.

[0002] Adaptive Discrete Cosine Transform Coding Method Using Two-Dimensional Orthogonal Transform As an encoding method for compressing a data amount of a multi-valued image such as a halftone image or a color image without deteriorating its characteristics. Transfo
rm, hereinafter referred to as ADCT method) is widely used.

In the ADCT method, a multi-valued image is divided into blocks each having a predetermined number of pixels (for example, 8 × 8 pixels), and image data is orthogonally transformed for each block to obtain a transform coefficient (hereinafter referred to as DCT coefficient). ) Is obtained, and each component of the matrix is quantized using a corresponding visual adaptation threshold value (described later) and then subjected to variable-length coding, thereby compressing the data amount.

[0004]

2. Description of the Related Art FIG. 11 shows a configuration of an image data compression apparatus to which a conventional ADCT method is applied. Also, in FIG.
An example of a block obtained by dividing a multi-valued image will be described.

[0005] A multi-valued image read by an image reading device or the like is sequentially input to the DCT converter 611 for each block described above, and the DCT converter 611 performs a two-dimensional discrete cosine transform (hereinafter referred to as DCT transform). Through the processing, the matrix is converted into an 8 × 8 matrix (hereinafter, referred to as DCT coefficient D) including DCT coefficients corresponding to the spatial frequency components. FIG. 13 shows an example of the DCT coefficient D.

Each component of the DCT coefficient D is quantized by the linear quantization unit 620 using a component corresponding to the quantization threshold value Q _TH . The above-described quantization threshold Q _TH is
It is obtained from the visual adaptation threshold value corresponding to each spatial frequency and the quantization control parameter SF. This visual adaptation threshold is predetermined based on an experimental result regarding the sensitivity of visual perception to each spatial frequency component, and is given as a quantization matrix V _TH (see FIG. 14). Further, the quantization control parameter SF is a coefficient that determines the quantization accuracy of the image, and according to the image quality required for the restored image,
The setting is performed by an operator prior to the encoding process of the image data for one screen.

Here, the value of each component of the above-described quantization matrix V _TH is determined by the absolute value of the component corresponding to the low spatial frequency as shown in FIG. On the contrary, the absolute value of the component corresponding to the high spatial frequency is set to be large. For this reason, the quantization coefficient D _QU obtained by quantizing the DCT coefficient D by the linear quantization unit 620 is, as shown in FIG. 15, a component at the upper left corner of a matrix indicating a DC component (hereinafter, referred to as a DC component).
And only a very small number of AC components indicating low spatial frequency components around this DC component are effective coefficients having a value other than zero, and most AC components are often invalid coefficients having a value of zero. .

When the quantized coefficients D _QU obtained in this manner are scanned by using a scanning sequence called zigza scan shown in FIG. 16, a one-dimensional array in which a series of effective coefficients are followed by invalid coefficients. Is obtained. The one-dimensional array is converted by the encoding unit 631 into a combination of an effective coefficient (index) and a continuous length (run) of an invalid coefficient continuing before the index, and each combination is converted based on the code table 632. Image data is compressed by substituting codes corresponding to the frequencies of occurrence and performing variable-length coding.

The coded data obtained in this way is restored to image data by the image data restoration device shown in FIG. Decoding unit 711 of image data restoration device
Is provided with a decoding table 712 indicating a combination of a run and an index corresponding to a code, which is opposite to the above-described code table 632, and decodes sequentially input codes to obtain a combination of an index and a run. And input to the inverse quantization unit 720.

In the inverse quantization unit 720, the above-described one-dimensional array is restored from the input index and run by the array restoring unit 721, and each component of the one-dimensional array is sequentially input to the multiplier 722. You. The quantization threshold holding unit 723 holds the quantization threshold Q _TH , and the corresponding quantization threshold is sent to the multiplier 722 in synchronization with the input of the quantization coefficient to the multiplier 722. I have. Thus, each component of the quantization coefficient D _QU is inversely quantized by the multiplier 723 using the corresponding quantization threshold, and D
The CT coefficient D is restored.

Next, the DCT coefficient D is input to an inverse DCT conversion section 731, and the inverse DT
By performing the CT conversion process, the image data of the corresponding block is restored.

[0012]

By the way, in the above-mentioned conventional system, inverse quantization processing is performed by multiplying all quantization coefficients by a quantization threshold. However, the multiplication process of the invalid coefficient whose value is zero and the quantization threshold is a useless process in which the multiplication result is known in advance. That is, conventionally, regardless of whether the processing is useless or not, 64 times of multiplication processing is performed for the inverse quantization of the quantization coefficient D _QU of one block. Processing time was long.

As a technique for eliminating such useless processing and speeding up the inverse quantization processing, the present applicant has already filed an application for Japanese Patent Application No. 1-316708 "Image Data Restoration Method". . FIG. 18 shows a schematic configuration of an inverse quantization processing unit according to this technique.

In this technique, an address calculation unit 724 obtains an address of a corresponding effective coefficient from an input run value, and a quantization threshold value holding unit 723 stores a quantization threshold value corresponding to this address in a multiplier 722. The effective coefficient is only inversely quantized by multiplying the quantized threshold value and the index by the multiplier 722.
In this case, the multiplier 72 corresponds to the address described above.
By storing the result of multiplication by 2 in the DCT coefficient storage unit 725, the DCT coefficient D of 8 rows and 8 columns is restored.

According to this technique, in the inverse quantization process of each block, the inverse quantization result is written only at the address corresponding to the effective coefficient of the DCT coefficient holding unit 725.
Before performing the inverse quantization process on the next block, it is necessary to perform an initialization process for clearing the DCT coefficients obtained by the inverse quantization process on the previous block. For example, the DCT coefficient holding unit 7
It is sufficient to write the initial value "0" in all the 64 addresses of 25.

However, since the content of the address corresponding to the invalid coefficient of the previous block is zero, it is useless to write zero again by the initialization process. Also, since the address corresponding to the effective coefficient of the quantization coefficient D _QU of the next block is updated by writing the corresponding effective coefficient at the time of the inverse quantization processing of the next block,
It is not necessary to initialize these addresses in advance.

That is, regardless of whether the distribution of the effective coefficients in the quantized coefficients D _QU corresponding to two consecutive blocks coincides or does not coincide, the initialization processing for all 64 addresses is performed as described above. Because of the time required for unnecessary initialization processing, the effect of shortening the time required for the inverse quantization processing by reducing unnecessary multiplication processing is lost.

The present invention reduces the useless multiplication processing and also reduces the useless initialization processing in consideration of the distribution of effective coefficients in the quantization coefficient matrix of each of two consecutive blocks. It is an object to provide a method and an image data restoration device.

[0019]

FIG. 1 shows the principle of an inverse quantization method according to the present invention. According to the first aspect of the present invention, the original image is orthogonally transformed into blocks each including a plurality of pixels,
Decode the code data when restoring the original image from the code data by the orthogonal transform coding method that codes the quantization coefficient matrix obtained by quantizing each component of the coefficient matrix composed of the conversion coefficient with the corresponding quantization threshold. In the inverse quantization method of obtaining a coefficient matrix from the quantized coefficients obtained by the above, in the quantization coefficient matrices corresponding to each block and the block before it, a difference in distribution of effective coefficients having a value other than zero is determined. Then, the effective coefficients included in the quantization coefficient matrix corresponding to each block are multiplied by the corresponding quantization threshold to selectively inverse quantize, and according to the effective coefficient address indicating the position of the effective coefficient in the quantization coefficient matrix. In addition to holding the obtained inverse quantization results ,
Corresponds to the invalid coefficient and exists in the previous block.
The initialization process is performed on the current block invalid address corresponding to the effective coefficient, and the coefficient matrix corresponding to each block is restored.

FIG. 2 shows the configuration of the image data restoration apparatus according to the second and third aspects. According to a second aspect of the present invention, a quantized coefficient matrix obtained by orthogonally transforming an original image for each block composed of a plurality of pixels and quantizing each component of a coefficient matrix composed of transform coefficients with a corresponding quantization threshold is used. Decoding means 101 for continuously receiving code data according to an orthogonal transform coding scheme to be decoded, decoding the code data, and sequentially outputting decoded data representing a quantization coefficient matrix corresponding to each block; Address calculating means for obtaining effective coefficient addresses indicating positions occupied by the effective coefficient having a value other than zero included in the quantization coefficient matrix of each block in the quantization coefficient matrix based on the data, and outputting the effective coefficient address as a current block address. An inverse quantization unit 112 for performing inverse quantization by multiplying the effective coefficient by a quantization threshold indicated by the corresponding current block address; Means 1
12, a transform coefficient storage unit 113 for storing the result of inverse quantization as a transform coefficient in the current block address, and sequentially storing the current block address and storing the effective coefficient address held in the processing of the previous block in the previous block address. And an address holding unit 114 which outputs a current block invalid address corresponding to an invalid coefficient in each block and a current block invalid address corresponding to a valid coefficient in the previous block based on the previous block address and the current block address. Means 115, an initializing means 116 for writing and initializing an initial value in a storage location of the transform coefficient storage means 113 indicated by the current block invalid address, and a coefficient matrix composed of the transform coefficients stored in the transform coefficient storage means 113. Inverse orthogonal transformation means 102 for performing inverse orthogonal transformation and restoring image data; Characterized in that was.

According to a third aspect of the present invention, in the image data restoring apparatus according to the second aspect, the transform coefficient storage means 113 comprises:
A plurality of areas having a capacity corresponding to a coefficient matrix for one block are switched to store a coefficient matrix corresponding to a continuous block.

FIG. 3 shows the construction of the image data restoring device according to the fourth and fifth aspects. According to a fourth aspect of the present invention, in the image data restoration apparatus according to the second aspect, a DC block represented only by a DC component is detected based on the number of effective coefficients included in a quantization coefficient matrix corresponding to each block. And
It is characterized by including a block detection means 121 for subjecting this detection result to inverse orthogonal transformation processing by the inverse orthogonal transformation means 102.

According to a fifth aspect of the present invention, in the image data restoring apparatus according to the second aspect, based on the number of effective coefficients included in the quantization coefficient matrix corresponding to each block, an invalid block including no effective coefficient is determined. It is characterized by including a block detection means 131 for detecting the detection result and subjecting the detection result to inverse orthogonal transformation processing by the inverse orthogonal transformation means 102.

[0024]

According to the first aspect of the present invention, only the effective coefficients of the quantized coefficient matrix corresponding to each block are selectively inversely quantized and held at the corresponding effective coefficient addresses.
The initialization process can be performed while considering the distribution of effective coefficients in the quantization coefficient matrix of one block. This makes it possible to reduce unnecessary initialization processing for the case where the previous block and the effective coefficient address match or for the address corresponding to the invalid coefficient in the previous block.

According to a second aspect of the present invention, there is provided an address calculating means based on the decoded data obtained by the decoding means.
11 and the inverse quantization means 112 selectively inverse quantizes the effective coefficients included in the quantized coefficient matrix of each block, and stores them in the transform coefficient storage means 113 according to the effective coefficient addresses. Further, the address holding unit 114 and the determination unit 115 determine the current block invalid address, and the initialization unit 116 writes the initial value in the storage location of the transform coefficient storage unit 113 indicated by the current block invalid address. The coefficient matrix corresponding to each block can be restored and subjected to inverse orthogonal transform processing by the inverse orthogonal transform means 102.

This makes it possible to omit useless initialization processing in consideration of the distribution of effective coefficients in the quantization coefficient matrix of each of two consecutive blocks. According to the third aspect of the present invention, the process of storing the transform coefficient in any one of the plurality of regions of the transform coefficient storage unit 113 and the inverse orthogonal transform process on the coefficient matrix stored in another region can be performed in parallel. it can.

According to a fourth aspect of the present invention, the block detecting means 12
1, the DC block is detected, so that the inverse orthogonal transform unit 102 can perform processing suitable for the DC block according to the detection result of the block detection unit 121.

According to a fifth aspect of the present invention, the block detecting means 13
Since the invalid block is detected by 1, the inverse orthogonal transform unit 102 can perform a process adapted to the invalid block according to the detection result by the block detecting unit 131.

[0029]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 4 shows an embodiment of the image data restoring apparatus according to the second aspect.

In FIG. 4, the decoding unit 711 and the decoding table 71
2 corresponds to the decoding means 101, and the decoding unit 711 searches the decoding table 712 for the corresponding combination of the index and the run in accordance with the input of each code, as in the related art, and is obtained. The search result is transmitted to the inverse quantization unit 210 as decoded data.

In the inverse quantization unit 210, the decoded data described above is separated into a run and an index by the demultiplexer 201 and sent to the address calculation unit 111 and the inverse quantization unit 112, respectively.

The address calculating means 111 comprises an integrating circuit 211 and a block detecting section 212.
The integrating circuit 211 is configured to sequentially integrate the value obtained by adding the numerical value “1” to the run value. Also, the run is input to the block detection unit 212, and in response to the input of the run “R _EOB ” indicating the end of the block, a detection signal indicating that the end of the block has been detected is output.
The integrated value of 11 is reset to an initial value “0”.

In this way, a scanning order by zigzag scanning (see FIG. 16) is obtained as an integration result corresponding to each index. This scanning order indicates the position occupied by the effective coefficient corresponding to each index in the quantization coefficient D _QU of 8 rows and 8 columns.
The scanning order obtained by 11 may be transmitted as the current block address.

In FIG. 4, the inverse quantization means 112 is composed of a multiplier 221 and a quantization threshold value holding section 222. The quantization threshold value holding section 222 has a function corresponding to the scanning order by zigzag scanning. , A quantization threshold value corresponding to each DCT coefficient is held. Therefore, in response to the input of the current block address obtained by the above-described address calculation unit 111, the quantization threshold holding unit 222 sends the corresponding quantization threshold to the multiplier 221 and the multiplier 221
By multiplying the input index by the quantization threshold, the effective coefficient indicated by the index is inversely quantized to obtain a DCT coefficient.

In this case, the demultiplexer 201
Thus, the run and the index are separated in advance and input to the address calculation unit 111 and the inverse quantization unit 112, respectively. However, the address calculation unit 111 and the inverse quantization unit 112 It may be configured to include means for extracting an index.

In FIG. 4, the DCT coefficient obtained by the above-described inverse quantization means 112 is input to one of the input ports of the multiplexer 231, and the other input port of the multiplexer 231 has a numerical value “ 0 ”has been entered. That is, the multiplexer 231 responds to a switching instruction described later, and
In this configuration, any one of the output of S <b> 21 and the numerical value “0” is selected and sent to the buffer 232.

The buffer 232 corresponds to the transform coefficient storage means 113, and is configured to store data input via the multiplexer 231 at a specified address.

In FIG. 4, the memory 241 and the counter 242 form an address holding means 114. The counter 242 increments the count value by "1" each time a valid coefficient address is input to the memory 241. It is configured to add. The memory 241 stores the counter 2
The effective coefficient address is held at the address indicated by the count value of 42.

For example, as shown in FIG. 5B, the address corresponding to the effective coefficient indicated by hatching in FIG.
The addresses are stored in the addresses of the memory 241 corresponding to the input order.

The effective coefficient address held in the memory 241 in this way is transferred to the memory 251 shown in FIG. 4 in accordance with the detection signal from the above-described block detection section 212. It is configured to be retained. At this time, the count value of the counter 242 is transferred to the register 254 shown in FIG.

The above-mentioned memory 251, comparator 252, discrimination processing unit 253 and register 254 are connected to discrimination means 115.
The comparator 252 compares the previous block address stored in the memory 251 with the current block address from the address calculation unit 111, and based on the result of the comparison, the discrimination processing unit 253 described later. It is configured to perform distribution determination processing.

The result of the determination by the determination means 115 is input to the multiplexer 231 and is used for the input port selection processing by the multiplexer 231. Further, the previous block address output from the memory 251 is input to the selector 255 together with the current block address from the address calculation unit 111. The selector 255 responds to the determination result by the determination unit 115 described above. , One of the previous block address and the current block address, and
2 is designated as a write address.

That is, the multiplexer 231 and the selector 255 operate according to the result of the determination by the determination means 115, and write a numerical value "0" as an initial value in the address indicated by the result of the determination. The function of the initialization means 116 is realized by the 231 and the selector 255.

Next, regarding the inverse quantization operation by the inverse quantization unit 210 shown in FIG. 4, decoded data representing the quantized coefficient D _QU shown in FIGS. 5A and 5C is continuously input. The case will be described in detail as an example.

In this case, the above-mentioned memory 251 stores
The effective coefficient address in the quantized coefficient D _QU shown in FIG. 5A has already been held as the previous block address. Further, in accordance with the input of the combination of the index and the run of the decoded data representing the quantized coefficient D _QU shown in FIG. 5C, the effective coefficient address is calculated by the address calculating means 111 as described above, and the current block is calculated. The addresses are sequentially input to the memory 241 and the comparator 252.

FIG. 6 is a flowchart showing the inverse quantization operation. In response to the input of the current block address (step 301), the determination processing unit 253 specifies the address of the memory 251 and instructs the output of the previous block address. For example, the determination processing unit 253 may be configured to control a pointer that specifies an address of the memory 251. The pointer may be set at the start address of the memory 251 when the inverse quantization process of each block is started, and the pointer may be sequentially advanced.

Next, the current block address (current AD) and the previous block address (previous AD) are compared by the comparator 252, and based on the result of the comparison, first, the current block and the previous block are located at the same position. It is determined whether the effective coefficients are distributed (step 302).

For example, since the current block address and the previous block address which are input first are both addresses "0" corresponding to the DC component, the comparator 252 makes the current block address and the previous block address coincide. Is obtained. In this case, in both the current block and the previous block, it is determined that effective coefficients are distributed corresponding to the corresponding address, and step 30 is performed.
The determination may be affirmative in step 2.

In this case, the determination processing unit 253 instructs the multiplexer 231 to select the output of the multiplier 221 and instructs the selector 255 to select the current block address. As a result, the DCT coefficient obtained by the multiplier 221 is stored in the current block address (for example, the address corresponding to the DC component) of the buffer 232 (Step 303). At this time, the discrimination processing unit 25
3 updates the pointer designating the address of the memory 251 (step 304), and after instructing the memory 251 to output the next effective coefficient address, performs the processing of the next decoded data.

If the value of the pointer before update matches the content of the register 254 in step 304, it indicates that there is no effective coefficient in the previous block after the current block address input in step 301. ing. Therefore, after that, according to the input of the current block address, a process of storing the output of the multiplier 221 at the corresponding address may be performed.

On the other hand, if a negative determination is made in step 302, it is determined that there is a difference in the distribution of effective coefficients between the current block and the previous block, and it is determined whether the current block address is smaller than the previous block address ( Step 30
5).

Here, when the current block address is smaller than the previous block address, it indicates that the current block address corresponds to the invalid coefficient in the previous block. Accordingly, in the case of an affirmative determination in step 305, the determination processing unit 253 determines whether the multiplexer 231 and the selector 25
5, the DCT coefficient is stored in the current block address of the buffer 232 (step 306), and the processing of the next decoded data may be performed.

On the other hand, if the current block address is larger than the previous block address, it indicates that the previous block address corresponds to the invalid coefficient in the current block. That is, the previous block address is a current block invalid address.

For example, it is determined that the current block address “3” input second is larger than the second previous block address “2”, and a negative determination is made in step 305. In this case, the determination processing unit 253 determines that the current block invalid address has been detected, and
31 is instructed to select the numerical value “0”, and the selector 255 is instructed to select the previous block address. As a result, the numerical value “0” is stored in the corresponding address of the buffer 232 (for example, the previous block address “2”), and the effective coefficient written in the inverse quantization processing of the previous block is cleared (step 307). ).

In the case of a negative determination in step 305 described above, there is a possibility that a current blockless invalid address exists in addition to the invalid coefficient indicated by the previous block address.

Therefore, the discrimination processing unit 253 updates the pointer after the end of the above-described step 307 (step 3).
08), after instructing the memory 252 to output the next previous block address, the process may return to step 302 again to perform a process of detecting a difference in distribution of effective coefficients between the current block and the previous block. As a result, the current block invalid address can be detected without fail.

In step 308 described above,
If the pointer before update matches the final address set in the register 254, the process of storing the DCT coefficient in the current block address is performed in the same manner as in step 303, instead of the pointer update process. May be performed in accordance with the input of the current block address to store the output of the multiplier 221 at the corresponding address.

Steps 301 to 308 described above
Is repeated, the DCT coefficients D in the portion before the effective coefficient address corresponding to the effective coefficient included in the quantization coefficient D _QU of the current block are sequentially restored.

Further, according to the detection signal from the block detection section 212, the last previous block address is compared with the last current block address, and the determination processing section 253 performs the following processing according to the comparison result. Thus, the portion of the DCT coefficient D after the last current block address is restored.

If the last previous block address is larger than the last current block address, a valid coefficient address in the previous block exists after the last current block address. In this case, the determination processing unit 253
However, the contents of the memory 251 may be sequentially read out until the contents of the register 254 and the pointer match, and the initialization process for each of these addresses in the buffer 232 may be performed in the same manner as in step 307 described above. In other cases,
Since the processing is performed in step 304 or step 308 described above, the processing may be ended as it is.

As described above, the inverse quantization result is stored in the current block address of the buffer 232, and the current block invalid address of the buffer 232 is selectively initialized, so that the DCT coefficient D of 8 rows and 8 columns is obtained. Can be restored.

As a result, unnecessary multiplication processing of the invalid coefficient and the quantization threshold is reduced, and useless initialization processing for the address corresponding to the invalid coefficient in the previous block and the valid coefficient address in the current block is also reduced. It is possible to do.

Here, as shown in FIGS. 5 (a) and 5 (c), the effective coefficient distribution is often similar in the quantized coefficients D _QU corresponding to two consecutive blocks. ,
In step 305 described above, only a very small number are detected as the current block invalid address. For example, in the quantized coefficient D _QU shown in FIG. 5C, only three hatched addresses are detected as current block invalid addresses requiring initialization processing. Therefore, the time required for the inverse quantization process of the quantized coefficient D _QU is the time required for the multiplication process for the seven effective coefficients and the time required for the process of writing the initial value “0” to the above three addresses. And the quantization coefficient D _{QU for} one block
Can be greatly reduced.

As described above, in the process of writing the DCT coefficient obtained by the multiplier 221 to the buffer 232, the process of writing the initial value “0” to the address detected in step 306 is performed, so that one block There is no need to perform the initialization process again after the minute inverse quantization process ends.

Accordingly, in response to the detection signal from the block detection section 212, the discrimination processing section 253 performs the above-described processing, and then indicates that the restoration processing of the DCT coefficient D has been completed.
What is necessary is just to set it as the structure which notifies to the T conversion part 731 and instructs the start of an inverse DCT conversion process.

In response to this start instruction, the inverse DCT converter 7
31 is the DC stored in the buffer 232 as in the prior art.
By performing the inverse DCT transform processing on the T coefficient, 1
The image data for the block is restored. Further, by repeating the above-described inverse quantization process and inverse DCT transform process for each block, image data for one screen is restored.

After all the valid coefficients of the current block have been stored in the current block address of the buffer 232, the initialization processing for the current block invalid address of the buffer 232 may be performed. For example, memories 241 and 251 having 64 addresses are provided, and for each of the quantization coefficient D _QU of the current block and the previous block, information indicating whether or not each address is a valid coefficient is stored, and each is stored in a valid state. A map representing the distribution of coefficients may be created, and the current block invalid address may be obtained from the comparison result of this map to perform the initial processing.

Further, since the inverse quantization process of each block and the inverse DCT transform process of each block can be performed independently, a configuration in which the inverse quantization process and the inverse DCT transform process are performed in parallel is also possible. Good.

FIG. 7 shows an embodiment of the inverse quantization unit of the image data restoration apparatus according to the third aspect. In FIG. 7, the conversion coefficient storage unit 113 includes two buffers 233a and 233.
b, selector 234 and multiplexer (MPX) 235
And the pipeline control unit 236. The output of the multiplexer (MPX) 231 is input to the buffers 233a and 233b. In the conversion coefficient storage unit 113, the pipeline control unit 236
33a, 233b, selector 234, and multiplexer 2
35 to control the write operation and the read operation for the buffers 233a and 233b. Hereinafter, when the buffers 233a and 233b are collectively referred to, they are simply referred to as a buffer 233.

In this case, the above-described discrimination processing unit 253
Transmits a notification to the effect that the processing for restoring the DCT coefficient D has been completed to the pipeline control unit 236, and the pipeline control unit 236 responds to the notification by writing the buffer 233 in which the writing of the DCT coefficient D has been completed. Is set to the read valid state, the other buffer 233 is set to the write valid state, and the selector 234 may be instructed to switch the output port. In response to this, the selector 234 inputs the address from the selector 255 to the buffer 233 which has been newly written, and the DCT coefficient writing process is performed.

At this time, the pipeline control unit 23
6 instructs the multiplexer 235 to select the input port corresponding to the buffer 233 in the read enabled state, and instructs the inverse DCT transform unit 731 to start the inverse DCT transform process. May be instructed to the decoding unit 711 to start the decoding processing of the next block in response to the end notification.

As described above, since the inverse quantization processing of the next block can be performed in parallel with the inverse DCT transformation processing, the restoration processing of the entire image data can be further speeded up.

As a technique for shortening the time required for restoring the entire image, the present applicant has disclosed a technique disclosed in Japanese Patent Application No. Hei.
No. -290304, "Image Data Restoration Method" has already been filed, and by implementing this technique in combination with the inverse quantization method of the present invention, the image data restoration processing can be further speeded up.

In Japanese Patent Application No. 1-290304, "Image Data Restoration Method", when the effective coefficient included in the quantization coefficient D _QU or DCT coefficient D is only the DC component, the block contains the effective AC component. It is determined that the DC block does not have the DC block, and instead of performing the inverse DCT transform processing,
By obtaining the image data of the corresponding block by multiplying the component by a constant, the restoration processing of the entire image is speeded up.

FIG. 8 shows an embodiment of the image data restoring apparatus according to the fourth aspect. 8, the image data restoration apparatus has a configuration in which a comparator 261 corresponding to the block detection unit 121 is added to the inverse quantization unit 210 shown in FIG. 8, the inverse DCT transform unit 731, the shift circuit 262, and the multiplexer 263 form the inverse orthogonal transform unit 102.

The above-mentioned comparator 261 is configured to compare the count value of the counter 242 of the address holding means 114 with the threshold value “1” in response to the input of the detection signal from the block detection section 212. 263 is configured to operate according to the comparison result.

Here, when the comparator 261 determines that the number of effective coefficients included in the quantization coefficient D _QU is one, the effective coefficient corresponds to the DC component of the DCT coefficient D. That is, the block corresponding to the quantization coefficient D _QU is a DC block having no effective AC component, and the image data of this block is all the DC data described above.
Since the value is obtained by dividing the component by the constant “8”, the shift circuit 262 described above stores the D value stored in the buffer 232.
By shifting the C component right by 3 bits, the image data of this block can be obtained.

That is, the multiplexer 263 switches between the output of the inverse DCT converter 731 and the output of the shift circuit 262 according to the output of the comparator 261, so that the inverse orthogonal transform means 102 The inverse DCT transform processing can be replaced with shift processing by the shift circuit 262.

As described above, by providing the inverse quantization unit 210 with the block detecting means 121 and detecting the DC block, the inverse orthogonal transform means 102 can adapt to the DC block having no effective AC component. This makes it possible to reduce the time required for image data restoration processing as a whole.

In this case, since only the DC component of the DCT coefficient D is used for the image data restoration processing, the address corresponding to the AC component of the previous block is initialized to 8
There is no need to restore the DCT coefficients D in row 8 column. Therefore,
When a DC block is detected, the initialization process for the address corresponding to the AC component of the block before this DC block is skipped, and then, when performing the inverse quantization process of the block including the AC component, The inverse quantization process may be performed as described above, using the address of the corresponding AC component as the previous block address.

For example, in response to a detection result indicating that a DC block has been detected, the transfer operation from the memory 241 to the memory 251 and the transfer operation of the count value of the counter 242 to the register 254 are stopped, and the block before the DC block is stopped. May be stored in the memory 251 and the register 254, respectively. In this case, at the time of inverse quantization of the quantization coefficient D _QU of the block having an effective AC component, the effective coefficient address held up to the address of the memory 251 indicated by the contents of the register 254 is replaced with the previous block. The DCT coefficient D used as an address and corresponding to the corresponding block is restored as described above.

As a result, the number of times the initialization process is executed can be suppressed to the minimum limit, so that the time required for the inverse quantization process can be further reduced. Further, the inverse quantization method of the present invention may be applied to a hierarchical restoration type image data restoration apparatus.

FIG. 9 shows an embodiment of a hierarchical restoration type image data restoration apparatus to which the inverse quantization method of the present invention is applied.
9, the image data restoration apparatus is configured by adding a layer restoration processing unit 270 to the image data restoration apparatus shown in FIG. 4, and includes the decoding unit 711 and the inverse quantization unit 2 described above.
10 and the inverse DCT transform unit 731 restore the image in each layer from the input code data, and based on the obtained image and the image of the previous layer, the layer restoration processing unit 27
0 restores an image including all information up to the current hierarchy. The hierarchy restoration processing unit 270 holds, for example, image data of the previous hierarchy, adds the image data to the image data obtained by the inverse DCT transform unit 731, and obtains all information up to the current hierarchy. May be obtained so as to obtain image data including.

When performing hierarchical reconstruction, a code obtained by selectively encoding only a part of the components of the quantization coefficient D _QU of 8 rows and 8 columns is input as a code for one block of each hierarchy. For example, in the spectral selection method, first, only the DC component of each block is selectively encoded and transmitted, and then the second and third components are zigzag scanned.
Components corresponding to sequentially higher spatial frequencies are selectively encoded and transmitted, such as selecting a component to be scanned first. For this reason, in each hierarchy, the distribution of the effective coefficients in the quantization coefficient D _QU of each block almost overlaps, so that the effective block address corresponding to the invalid coefficient in the current block is the effective coefficient address in the previous block. The addresses are very few.

Therefore, when the inverse quantization method of the present invention is applied to the image data restoration apparatus of the hierarchical restoration method, the effect of reducing the time required for the inverse quantization processing of the quantization coefficient D _QU corresponding to each block is obtained. Especially large.

The present applicant has already filed Japanese Patent Application No. 2-180473 as a technique for shortening the time required for restoring the entire image data at the time of hierarchical restoration.
This technique detects an invalid block in which the quantization coefficient D _QU does not include a valid coefficient, omits the inverse DCT transform processing and the update processing of the image data for the invalid block, and achieves high-speed restoration processing of the entire image data. It is intended to make it.

FIG. 10 shows an embodiment of the image data restoring apparatus according to the fifth aspect. 10, the image data restoration apparatus has a configuration in which a comparator 271 corresponding to the block detection means 131 is added to the image data restoration apparatus shown in FIG. 9, and an inverse DCT transform unit 731 and a hierarchical restoration processing unit 270 constitutes the inverse orthogonal transform means 102.

The above-mentioned comparator 271 receives the count value of the counter 242 of the address holding means 114 and the threshold “0” in response to the input of the detection signal from the block detection section 212.
_Are compared, and when they match, an invalid block signal indicating that the effective coefficient is not included in the quantization coefficient D _QU of the corresponding block is output.

For example, in the spectral selection method, since components corresponding to high spatial frequencies are sequentially transmitted as described above, in a high-order hierarchy, all the selected components may be invalid coefficients.

In such a case, as the code for one block, "E" indicating that the code for one block is completed.
Only the “OB” code is input, and in response to this, the comparator 271 outputs an invalid block signal.

In this case, all the components of the quantization coefficient D _QU corresponding to the corresponding block are invalid coefficients.
Since all the components of the corresponding DCT coefficients D are also invalid coefficients, the inverse quantization process and the inverse DC
The T conversion process is a useless calculation process. In this case, since new information relating to the image of the current layer is not input, it is not necessary for the above-described layer restoration processing unit 270 to update the image including information up to the previous layer.

For example, in response to the above-mentioned invalid block signal, the discrimination processing unit 253 replaces the notification that the restoration processing of the DCT coefficient D has been completed, and informs the inverse DCT conversion unit that the corresponding block is an invalid block. 731. At this time, similarly to the processing of the DC block described above, the memory 2
The transfer processing to the register 51 and the register 254 may be stopped, and the contents of the memory 251 and the register 254 may be retained and used for the inverse quantization processing of the next block.

In addition, inverse DCT transform section 731 skips the inverse DCT transform processing in response to the notification that the block is an invalid block, and relays this report to hierarchical restoration processing section 270, and in response to this, The restoration processing unit 270 may skip the above-described hierarchy restoration processing and output the image data of the previous hierarchy as it is.

As described above, by providing the inverse quantization unit 210 with the block detecting means 131 to detect invalid blocks, the inverse orthogonal transform means 102 can adapt to invalid blocks that do not contain valid information. This makes it possible to reduce the time required for image data restoration processing as a whole.

[0095]

As described above, according to the present invention, when the effective coefficient is selectively dequantized and the coefficient matrix is restored, the distribution of the effective coefficient in the quantized coefficient matrix corresponding to two consecutive blocks is determined. By performing the initialization processing while taking into account, it is possible to reduce unnecessary initialization processing for the address corresponding to the invalid coefficient of the previous block and the address corresponding to the valid coefficient in the current block, so that the inverse quantization is performed. The time required for the processing can be reduced, and the speed of the restoration processing of the image data can be increased.

Further, by performing the inverse quantization and initialization processing and the inverse DCT transform processing in parallel, the processing time per block can be reduced and the restoration processing can be performed at higher speed.

Further, by detecting a DC block and an invalid block and performing processing adapted to these blocks by the inverse orthogonal transform means, the time required for the inverse quantization processing of each block is reduced, and the entire image is restored. The time required for processing can be significantly reduced.

[Brief description of the drawings]

FIG. 1 is a diagram showing the principle of the inverse quantization method of the present invention.

FIG. 2 is a diagram showing a configuration of an image data restoration device according to claims 2 and 3;

FIG. 3 is a diagram showing a configuration of an image data restoration apparatus according to claims 4 and 5;

FIG. 4 is a diagram showing a configuration of an embodiment of an image data restoration apparatus according to claim 2;

FIG. 5 is an explanatory diagram of an effective coefficient address.

FIG. 6 is a flowchart illustrating an inverse quantization process.

FIG. 7 is a diagram showing a configuration of a main part of an embodiment of an inverse quantization unit of the image data restoration device according to claim 3;

FIG. 8 is a diagram showing a main configuration of an embodiment of an inverse quantization unit of the image data restoration apparatus according to claim 4;

FIG. 9 is a diagram showing a main configuration of an embodiment of an image data restoration apparatus of a hierarchical restoration method to which the inverse quantization method of the present invention is applied.

FIG. 10 is a diagram showing a main configuration of an embodiment of an inverse quantization unit of the image data restoration apparatus according to claim 5;

FIG. 11 is a configuration diagram of a conventional image data compression device.

FIG. 12 is a diagram illustrating an example of a block.

FIG. 13 is a diagram illustrating an example of a DCT coefficient D.

FIG. 14 is a diagram showing a quantization matrix.

FIG. 15 is a diagram illustrating a quantization coefficient D _QU .

FIG. 16 is an explanatory diagram of a zigzag scan.

FIG. 17 is a configuration diagram of a conventional image data restoration device.

FIG. 18 is a schematic configuration diagram of a conventional inverse quantization processing unit.

[Explanation of symbols]

 Reference Signs List 101 decoding means 102 inverse orthogonal transformation means 111 address calculation means 112 inverse quantization means 113 transform coefficient storage means 114 address holding means 115 discrimination means 116 initialization means 121, 131 block detection means 201 demultiplexer (DMPX) 211 integration circuit 212 block Detector 221, 722 Multiplier 222, 723 Quantization threshold holding unit 231, 235, 263 Multiplexer (MPX) 232, 233 Buffer 234, 255 Selector 236 Pipeline controller 241, 251 Memory 242 Counter 252, 261, 271 Comparator 253 Discrimination processing unit 254 Register 262 Shift circuit 270 Hierarchical restoration processing unit 611 DCT transformation unit 620 Linear quantization unit 631 Encoding unit 632 Code table 711 Decoding unit 712 Decoding table 20 inverse quantization unit 721 sequences restorer 724 address calculating unit 725 DCT coefficient holding unit 731 inverse DCT unit

Continuation of front page (56) References JP-A-3-177163 (JP, A) JP-A-3-151176 (JP, A) JP-A-4-70060 (JP, A) JP-A-3-293865 (JP) , A) (58) Field surveyed (Int. Cl. ⁶ , DB name) H04N 1/41-1/419 H04N 7/24-7/68

Claims (5)

(57) [Claims] 1. A quantization coefficient matrix obtained by orthogonally transforming an original image for each block composed of a plurality of pixels and quantizing each component of a coefficient matrix composed of transform coefficients with a corresponding quantization threshold. When restoring the original image from code data according to the orthogonal transform coding method, in the inverse quantization method of obtaining a coefficient matrix from quantization coefficients obtained by decoding the code data, each block and the previous block In the quantization coefficient matrix corresponding to each block, a difference in distribution of effective coefficients having a value other than zero is determined, and the effective coefficients included in the quantization coefficient matrix corresponding to each block are multiplied by the corresponding quantization threshold. And selectively inversely quantize, according to an effective coefficient address indicating a position of the effective coefficient in the quantization coefficient matrix, and holding the obtained inverse quantization result. Both are invalid coefficients in each block
And the effective coefficient in the previous block
An inverse quantization method, wherein initialization processing is performed on a corresponding current block invalid address to restore a coefficient matrix corresponding to each block. 2. A quantization coefficient matrix obtained by orthogonally transforming an original image for each block composed of a plurality of pixels and quantizing each component of a coefficient matrix composed of transform coefficients with a corresponding quantization threshold value. Decoding means (101) for successively inputting code data according to the orthogonal transform coding scheme, decoding the code data, and sequentially outputting decoded data representing a quantization coefficient matrix corresponding to each block; Based on the decoded data, an effective coefficient address indicating a position occupied by the effective coefficient having a value other than zero included in the quantization coefficient matrix of each block in the quantization coefficient matrix is obtained, and an address to be output as a current block address. Calculating means for multiplying the effective coefficient by a quantization threshold indicated by a corresponding current block address to perform inverse quantization; 1
12), transform coefficient storage means (113) for storing the result of inverse quantization by the inverse quantization means (112) in the current block address as a transform coefficient, and sequentially storing the current block address, Address holding means (1) for outputting the effective coefficient address held during the processing of the block as the previous block address.
14) determining means for determining, based on the previous block address and the current block address, a current block invalid address corresponding to an invalid coefficient in each block and a valid coefficient in a preceding block; ), Initialization means (116) for writing and initializing an initial value in a storage location of the conversion coefficient storage means (113) indicated by the current block invalid address, and stored in the conversion coefficient storage means (113). An inverse orthogonal transform unit (102) for inversely orthogonally transforming a coefficient matrix including the transformed coefficients to restore image data. 3. The image data restoration device according to claim 2, wherein the conversion coefficient storage means (113) switches a plurality of areas having a capacity corresponding to a coefficient matrix for one block to correspond to a continuous block. An image data restoring apparatus characterized by storing a coefficient matrix to be processed. 4. The image data restoration device according to claim 2, wherein a DC block represented by only a DC component is detected based on the number of effective coefficients included in a quantization coefficient matrix corresponding to each block, This detection result is converted into an inverse orthogonal transform means (102)
Block detecting means (12
An image data restoring apparatus characterized by comprising (1). 5. The image data restoration device according to claim 2, wherein an invalid block containing no effective coefficient is detected based on the number of effective coefficients included in a quantization coefficient matrix corresponding to each block, and the detection is performed. Block detection means (131) for subjecting the result to inverse orthogonal transformation processing by inverse orthogonal transformation means (102)
An image data restoration device comprising: