JPH05167857A - Image data restoring device - Google Patents

Abstract

PURPOSE:To shorten a processing time required for a reduced image (still image) display. CONSTITUTION:This image data restoring device is constituted so that an original image is divided by blocks consisting of plural picture elements, and at the time or receiving original image data consisting of code data derived with regard to the respective blocks, first of all, restoration image data (gradation value) related to each picture element is derived by a block unit, and subsequently, image data for a reduced display can be extracted from this restoration image data. This device is provided with a reduction rate setting part 7 for designating a reduction rate, a picture element thinning-out part 8 for outputting that which is formed by extracting the data of picture element parts of the number corresponding to the reduction rate concerned from the restoration image data by a block unit as the image data for a reduced display, a display image memory 9 for writing successively this output data. This device is constituted so as to display a reduced image of size corresponding to the reduction rate by using the contents of this display image memory 9.

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 device, and more particularly, it divides an original image into blocks each consisting of a plurality of pixels, and at least quantizes the tone values of each pixel in each block. When receiving the original image data composed of the coded data of the block unit obtained by encoding, the decoded image data is converted by the decoding unit and the inverse quantization unit to obtain the restored image data for each pixel of the block. The present invention relates to an image data restoration device capable of extracting reduced display image data in block units from restored image data.

Generally, in order to transmit (transmit, record, etc.) image data having a remarkably large amount of information, such as a halftone image or a color image, at high speed and with high quality, prior to transmission, the floor for each pixel is It is necessary to compress the key value with high efficiency to generate the encoded original image data, and as the encoding method,
For example, Adaptive Discrete Cosine Transform Coding
The crete cosine transform (hereinafter referred to as ADCT) is used.

In recent years, an image data which receives and reproduces such original image data and restores the original image from the original image data.
In the data restoration device, there is a demand for speeding up the display process for a reduced image (still image) consisting of some pixels in the original image, and the present invention responds to such a demand.

[0004]

2. Description of the Related Art A conventional coding apparatus using ADCT will be described with reference to FIG. In the figure, 41 is a block buffer that stores the gradation value of each pixel in an arbitrary block when the original image is divided into blocks (for example, 8 × 8), and 42 is each pixel The DCT transform unit 43, which transforms the tone value of the pixel into a DCT coefficient indicating the spatial frequency distribution of each pixel by performing a two-dimensional discrete cosine transform (hereinafter referred to as DCT), 43 Quantization unit 45 to be a quantized coefficient for each coefficient, 44 is a variable length coding unit for variable length coding the quantized coefficient according to the statistically obtained Huffman code table 46, and a variable length code At the output side of the conversion section 44, the code data for the gradation value of each 8 × 8 pixel is extracted, and the code data for all the blocks of the original image is obtained to obtain the code data for one screen. -The data is obtained.

FIG. 5 shows a restoration device for reproducing the gradation value of each pixel in each block of the original image from the encoded original image data, 51 is a variable length decoding unit, and 52 is a variable length decoding unit. Is an inverse quantization unit, 53 is an inverse DCT transform unit, 54 is a working image memory, 55 is a Huffman decoding table, and 56 is an inverse quantization matrix.

The decompressor executes a process such as variable length decoding-inverse quantization-inverse DCT conversion, which is the reverse of the previous encoding, and the original data is output to the output side of the inverse DCT conversion unit 53. The tone value of each pixel of the image is restored in block units, and the restored image data is generated by writing this tone value in the working image memory 54 for all blocks forming the original image. It will be.

Next, with reference to FIGS. 6 to 7, the manner in which the gradation value of each pixel in an arbitrary block is quantized will be described. That is, FIG. 6A shows an initial state before DCT conversion of each pixel of a block composed of, for example, 8 × 8 pixels, and each numerical value indicates a gradation value at that pixel. -FIG.6 (b) has shown the output of the DCT conversion part 42, ie, the DCT coefficient (coefficient of spatial frequency distribution) of each pixel.
The DCT coefficient is transposed by first performing a one-dimensional discrete cosine transform on the gradation value of each pixel of the block and then transposing the row and the column, and subsequently, for each pixel component after the transposition. , The one-dimensional discrete cosine transform is performed again, and then the row and the column are exchanged. FIG. 7 (a) shows the quantization matrix 45 consisting of the threshold value of each pixel for this DCT coefficient, and this threshold value is determined by visual experiments.・ FIG. 7 (b) shows the DCT coefficient shown in FIG. 6 (b).
The quantized coefficient after linearly quantized for each pixel by the threshold of (a) is shown, and the quantized coefficient of a pixel having a DCT coefficient of a value smaller than the threshold is “0”,
When observing the pixels of the entire block, the quantization coefficient has a DC component (pixel in 1 row and 1 column) and a slight AC component.

Then, the quantized coefficients of a two-dimensional array as shown in FIG. 7 (b) are converted into one-dimensional quantized coefficients by scanning in the direction of the arrow and then input to the variable length coding unit 44. In the variable-length coding unit 44, for the DC component, the difference between the DC component of the block (quantization coefficient of the first pixel) and the DC component of the previous block, and for the AC component, the effective coefficient (the value is 0 ”Quantized coefficient) and the run length of the invalid coefficient (quantized coefficient whose value is“ 0 ”) up to that point are respectively variable length coded by the Huffman code table 46 and coded in block units. Data is obtained, and the original image data is created by obtaining this code data for all the blocks constituting the original image.

When the original image data is restored by the restoring device shown in FIG. 5, first, the variable length decoding unit
In 51, a Huffman decoding table 55 having contents opposite to those of the Huffman code table 46 is used to calculate the difference or AC for the DC component.
The quantized coefficient of FIG. 7 (b) is generated by obtaining the value of the effective coefficient, the length of the run, etc. regarding the component. Then, the quantized coefficient is generated by the inverse quantizer 52 in FIG.
Inverse quantization matrix same as the quantization matrix of (a)
By inverse quantization using 56, the DCT of FIG.
The coefficient is obtained, and then, the DCT coefficient is orthogonally transformed by the inverse DCT transformation unit 53 to restore the gradation value of each pixel in FIG. 6A, that is, when the code data is created. Performs the reverse process for each block.

[0010]

As described above, in the conventional image data restoration device, the restored image data obtained in block units are sequentially written in the work image memory 54, and subsequently read out from this image memory. The restored image data is used to display the original image normally.

When the original image is reduced and displayed, the same restoration process as in the normal display, that is, the restored image data for one screen is stored in the large-capacity work image memory 54 which is generally slow in access speed. First, a process such as reading is performed, then, restored image data of a predetermined number of pixel portions is extracted from this restored image data as image data for reduced display, and subsequently, a reduced image (still image) is used using this image data. Since it is displayed, there is a problem that the processing time for reduced display becomes long.

Therefore, according to the present invention, only the restored image data of the pixel portion corresponding to the designated reduction ratio (from the restored image data consisting of the gradation value of each pixel of the block before being written in the image memory (for reduced display) is used. An image thinning unit for extracting and outputting (image data) is provided, and the reduced image data is written in the display image memory with a fast access speed, so that a reduced image (still image) used in, for example, retrieval. The purpose is to reduce the processing time required for display.

[0013]

According to the present invention, an original image is divided into blocks of, for example, 8 × 8 pixels, and the gradation value of each pixel of each block is quantized and then encoded. In the image data restoration device which receives the original image data generated by sequentially obtaining the code data in block units and collecting the relevant code data of all the blocks forming the original image, the original image data is stored in the original image data. When the reduced image of the original image is executed using the restored image data (gradation value) of each pixel in the block unit obtained by performing the processing such as decoding and inverse quantization, the restored image data of the block unit is Inside,
Only the restored image data of the pixel at the position corresponding to the reduction ratio designated in advance is taken out and written in the display image memory.

FIG. 1 is a block diagram showing the principle of the present invention. In the figure, 1 is a variable length decoding unit, 2 is an inverse quantization unit, and 3 is an inverse DC.
T transform unit, 4 is an image data restoration unit, 5 is a Huffman decoding table, 6 is an inverse quantization matrix,
These are the same as the conventional one shown in FIG. 7
Is a reduction rate setting unit for designating a reduction rate when reducing and displaying the original image.
Values such as 8 are often used. Reference numeral 8 denotes a pixel thinning unit, which is an output of the inverse DCT conversion unit 3, that is, a restored image data composed of gradation values restored for each pixel of the block.
From the data, only the restored image data (reduced display image data) of the pixel portion corresponding to the reduction ratio is taken out. 9
Is a display image memory in which reduced display image data output from the pixel thinning unit is written in block units.

Here, the inverse DCT conversion section 3 is required as follows.
This is because the DCT conversion for compressing the gradation value of each pixel with high efficiency is performed prior to the quantization at the time of encoding the image data, and the image data is encoded without this DCT conversion. In that case, the inverse DCT conversion unit 3 is omitted.

[0016]

According to the present invention, the original image data composed of the code data obtained by dividing the original image into blocks each consisting of a predetermined number of pixels such as 8 × 8 is obtained. When the still image of the original image is reduced and displayed on the screen by using the restored image data that is received and output to the output side of the image data restoration unit, the restored image data consisting of the gradation value of each pixel of the block , The pixel thinning unit 8 extracts only the gradation value of the pixel selected according to the reduction ratio specified in advance.
If, for example, the reduction ratio is 1/2, only the gradation values of the 1/4 pixel portion of the whole indicated by the shaded portion in the drawing are extracted and written into the display image memory 9. It is a thing.

That is, a reduced image (still image) of the original image.
When the image is restored and displayed on the screen, only the restored image data of the pixel portion that is actually used, that is, only the reduced display image data is written in the display image memory 9. Therefore, the display image memory 9 In this case, the writing and reading time can be shortened, the display processing of the reduced image can be speeded up, and the reduced image can be efficiently restored.

[0018]

Embodiments of the present invention will be described with reference to FIGS. FIG. 2 is an explanatory diagram showing a configuration example of a pixel thinning unit, 20 is provided in the image data restoring unit 4 and stores restored pixel data (gradation value) of each pixel in block units. A buffer memory from which the restored pixel data of the pixel specified by the output address of the address register 22 is taken out, and 21 temporarily stores the restored image data (gradation value) sent from the buffer memory 20. However, AND described later
When the output instruction signal from the circuit 26 is received, the latch 22 sends out the gradation value to the display image memory 9 as the reduced display image data, and 22 is the intra-block X address and the intra-block Y address of the next pixel to be processed. Is an effective address holding unit for storing the in-block address (effective X address and effective Y address) of each pixel selected according to the designated reduction ratio, and 24 and 25 are address registers. A comparison unit for checking whether the output address of 22 and the storage address of the effective address holding unit 23 are the same, 26 is an AND circuit for taking the logical product of the outputs of the comparison units 24 and 25, and 27 is the AND circuit 26. Based on the output signal (output instruction signal), a data transmission for transmitting a response signal ACK to the data request signal REQ from the memory control unit 31 to the memory control unit. Timing control unit, 28
Is a counter that integrates the output signal (output instruction signal) of the AND circuit 26, and 29 is the number of effective pixels set according to the reduction ratio, that is, the restored image data (gradation value) of each pixel of the block. ) Is a sending number holding unit that stores the number of pixels used as reduced display image data, and 30 checks whether the output signals of the counter 28 and the sending number holding unit 29 match, and if they match, restore processing is performed. Restored image data (gradation value) of the last effective pixel of the middle block
Signal LAST indicating that the image is transmitted to the display image memory 9
And a memory control unit 31 that outputs a data request signal REQ for the next effective pixel to the data transmission timing control unit 27, and receives the previous response signal ACK, the final signal LAST, and the like. Shown respectively.

Here, for example, when the block is composed of 8 × 8 pixels and the reduction ratio is specified to be 1/2, the effective address holding unit 23 stores "0" as both the effective X address and the effective Y address. , “2”, “4”, “6” are set, and “16 (= 8 × 8 ÷
4) ”is set, and the X address in the block and the Y address in the block are both“ 0 ”,“ 2 ”,“ 4 ”,“ 6 ”.
The output instruction signal from the AND circuit 26 is sent to the latch 21 only for the pixels that are
The restored image data (gradation value) of each pixel in the shaded area as shown by is written as reduced display image data.

Then, the output of the counter 28 becomes "16" and the final signal LAST from the comparator 30 is sent to the memory controller, and the restored display image data of the 16th pixel is displayed from the latch 21. When it is sent to the image memory 9, the reduced display processing for that block is completed, and then the same processing is repeated for the next block, so that the reduced image data of all the blocks forming the original image is displayed in the display image memory. 9 will be stored.

The number of pixels in one block can be arbitrarily set, and the reduction rate can be specified to any value. For example, the number of pixels of 16 × 16 and the reduction rate of 1/4 can be set. When specified, the valid X address and valid Y address set in the valid address holding unit 23 are both “0”, “4”, “8”, and “12”, and are set in the sending number holding unit 29. The value is 64.

FIG. 3 is an explanatory diagram showing a processing procedure when the reduced display image data is generated in block units and stored in the display image memory. That is, the restoration process of the block unit of the original image data in the image data restoration unit 4 is started, and the process proceeds to the next step. The in-block X address and the in-block Y address of the address register 22 are initialized and the process proceeds to the next step. The restored image data (gradation value) of the pixel at the X address in the block and the Y address in the block specified by the address register 22 is received from the buffer memory 20 and latched 21.
Write to and proceed to the next step. Whether or not this pixel is a pixel used for reduced display, that is, the address register
It is determined whether each of the in-block X address and the in-block Y address of 22 matches the valid X address and valid Y address of the valid address holding unit 23, and "YE
If "S", proceed to the next step, and if "NO", proceed to step. Upon receiving the output instruction signal from the AND circuit 26, the latch 21 sends the restored image data (gradation value) written therein to the display image memory 9 as reduced display image data, and the next step Proceed to. Whether all the reduced display image data of the block are stored in the display image memory 9, that is, the comparison unit 30
Of the last signal (LAST) is generated, the processing for the block is terminated if "YES", and the process proceeds to the next step if "NO". The in-block X address and the in-block Y address of the address register 22 are updated, that is, the values are set as the in-block address of the pixel to be subjected to the next thinning process, and the process returns to the step. The reduced display image data in block units is stored in the display image memory 9 by such a procedure.

By performing this processing on the restored image data (gradation values) of all the blocks forming the original image, all reduced display used when the original image is reduced and displayed as a still image. Image data for use is generated.

[0024]

According to the present invention, an original image is divided into blocks each composed of a plurality of pixels, and original image data composed of code data generated for each block is received and reproduced to obtain a still image of the original image. In the image data restoration device having the function of reducing the display, the image data restoration unit corresponds to the reduction ratio specified in advance from the restored image data (gradation value) for each pixel in block units. The process of taking out only the pixel portion of the position (image data for reduced display) and storing it in the display image memory,
Since it is configured to repeat for all blocks of the original image, the image memory for display is
Only the gradation value of the pixel actually used is stored as the reduced display image data, the processing time required for the reduced display is shortened, and the performance of the image data restoration device can be improved. it can.

[Brief description of drawings]

FIG. 1 is a principle configuration diagram of the present invention.

FIG. 2 is an explanatory diagram showing a configuration example of a pixel thinning unit of the present invention.

FIG. 3 is an explanatory diagram showing a processing procedure when generating reduced display image data in block units according to the present invention.

FIG. 4 is an explanatory diagram showing a conventional encoding device using ADCT.

5 is an explanatory diagram showing a conventional decoding device corresponding to the encoding device of FIG.

FIG. 6 is an explanatory diagram (No. 1) showing a general state in which the gradation value of each pixel of an arbitrary block forming an original image is quantized.

FIG. 7 is an explanatory diagram (No. 2) showing a general state in which the gradation value of each pixel of an arbitrary block forming an original image is quantized.

[Explanation of symbols]

 In FIG. 1, 1 ... Variable length decoding unit 2 ... Inverse quantization unit 3 ... Inverse DCT conversion unit 4 ... Image data restoration unit 5 ... Huffman decoding table 6 ... Inverse quantization Matrix 7 ... Reduction ratio setting unit 8 ... Pixel thinning unit 9 ... Display image memory

Claims (3)

[Claims] 1. An original image is divided into blocks made up of a plurality of pixels, and an original data made up of block-unit code data obtained by at least quantizing and encoding the gradation value of each pixel in each block. When the image data is received, the coded data is converted by the decoding unit and the dequantization unit to obtain the restored image data for each pixel of the block, and the reduced display image data is converted from the restored image data into the block unit. An image data restoration device capable of being taken out by means of a reduction rate setting unit for designating a reduction rate at the time of reduced display, and the reduction rate from the restored image data as the reduced display image data. Pixels for outputting the image data generated by extracting the restored image data of the number of pixels corresponding to the reduction ratio specified in the setting section A reduction image and a display image memory for sequentially writing the output data of the pixel reduction unit in the block unit, and the reduced display image for each of the blocks written in the display image memory. An image data restoration device characterized by outputting data. 2. The code data of the block unit is converted into a coefficient of a spatial frequency distribution for each pixel by DCT conversion prior to the quantization, and an inverse conversion corresponding to the DCT conversion is performed. The inverse DCT transform unit is provided on the subsequent stage side of the quantization unit.
The described image data restoration device. 3. The image data restoration device according to claim 1, wherein the reduction rate is designated to a power of 2.