JPH06334988A - Compressed picture evaluating method and picture data compressing and displaying device - Google Patents

Abstract

PURPOSE:To provide the compressed picture quality evaluating method, which easily and surely evaluates the picture quality of a reproduced picture, and the picture data compressing device which determines an optimum compression parameter without paying operator's attention to the compression parameter. CONSTITUTION:These compressed picture evaluating method and picture data compressing and display device consist of an original picture display memory 2 into which an original picture is taken, a compressing part 3 which compresses picture data, an expanding part 5 which expands compressed data, a reproduced picture display memory 6 where the reproduced picture is recorded, a picture quality degradation calculating part 7 which calculates the degradation in quality of the reproduced picture, a degradation deciding and coloring part 8 which converts the degree of degradation in picture quality to corresponding color information, and a picture quality degradation degree display memory 9 where converted color information is recorded.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to lossy compression in which the compression ratio and the image quality are changed by changing the compression parameter, and in particular, the degree of image quality deterioration of a compressed / decompressed image (hereinafter referred to as a reproduced image) with respect to an original image. The present invention relates to an image data compression apparatus capable of displaying an image on a screen and selecting an optimum compression parameter by objective judgment without considering the numerical value of the compression parameter. Further, the present invention relates to an image data compression / display device that performs simple and reliable compression and high-speed decompression in a system such as an image database system that compresses a large amount of image data into a database and searches and displays at high speed.

[0002]

2. Description of the Related Art Conventionally, various compression methods using orthogonal transform coding have been known as lossy image compression methods for compressing multivalued image information such as natural images at a high compression rate. Generally, image information is divided into blocks of N × N pixels, orthogonally transformed for each block, and the orthogonally transformed data is quantized by each frequency component using human visual characteristics. Quantize with step size. Further, the quantized data is subjected to variable length coding such as Huffman coding to realize high compression of image information.

However, in this orthogonal transform coding, since image information is truncated when quantizing by utilizing human visual characteristics, it is constituted by a quantizing step width set for each frequency component. When the table (hereinafter referred to as the quantization table) is changed, the image quality and the compression rate change significantly.
Therefore, some methods have been proposed for determining the optimum quantization table for the image information to be compressed.

As an example thereof, as described in Japanese Patent Laid-Open No. 3-16489, an operator is made to input a parameter for determining a quantization table, a quantization step width is calculated from the input parameter, and the quantization step width is calculated. A method of creating a quantization table, performing compression / expansion using the created quantization table, displaying the result on the screen, and allowing the operator to evaluate the image quality to determine the quantization table, JP-A-1-128674 As described in (1), there is a method of creating reduced image data by thinning pixels from the original image data, compressing and expanding the reduced image data, evaluating the image quality, and determining the quantization step width.

[0005]

In the above prior art, when the image quality of a reproduced image is evaluated with respect to the original image and the compression parameter is determined, the original image and the reproduced image are visually compared and evaluated, or the original image and the reproduced image are reproduced. The mean square error (hereinafter referred to as S / N) of the entire image is obtained, and the image quality is numerically evaluated and determined. However, in the case of visually evaluating the original image and the reproduced image, the operator subjectively makes a judgment, so that an oversight or a wrong judgment is inevitable. In addition, in the numerical judgment by S / N measurement of the entire screen, it is possible to make an objective judgment, but when a part of the reproduced image is extremely deteriorated in image quality, there is a drawback that it is not reflected in the S / N measurement result. . If the result of image quality evaluation does not achieve the desired image quality, the operator must input the compression parameter again, and the operator must know the relationship between the compression parameter and the image quality or compression rate. There is a problem that the optimum compression parameter cannot be set.

The present invention solves the above problems, displays the degree of image quality deterioration in block units used in compression processing, and makes a judgment of image quality deterioration of a reproduced image by objective judgment, and a compression parameter. It is an object of the present invention to provide an image data compression display device capable of performing image data compression at a desired image quality and compression rate without being aware of the above.

[0007]

To achieve the above object, in a compressed image evaluation method, an image data capturing means for capturing an image to be compressed, an original image recording means for temporarily recording the captured image data, and a temporary image recording means are provided. A compression unit that compresses the recorded image data with the set compression parameter and a compression ratio calculation that compares the image data size before compression and the image data size after compression to calculate the compression ratio and generate compression ratio information. Means, decompressing means for decompressing compressed image data, replay image recording means for temporarily recording decompressed replay image data, and reading out and comparing the original image and the replay image in block units, An image quality deterioration calculating unit for calculating deterioration, an image quality deterioration degree recording unit for temporarily recording the degree of image quality deterioration in block units as color information, and an image quality deterioration unit in block units. Depending on the situation, the deterioration determination display means for writing the deterioration degree display information to the image quality deterioration degree recording means, the information recorded in the original image recording means, the reproduced image recording means, the image quality deterioration degree recording means, and the compression rate information are required. Display control means for performing display processing according to the above.

Further, in the image data compression display device, in addition to the compressed image evaluation method, a compression parameter generation means for generating a plurality of compression parameters and a multi-screen for facilitating comparison of a part of the decompressed reproduced image are provided. It is provided with a reproduced image cutting-out means for displaying, a selecting means for selecting the optimum compression parameter from the multi-screen, and a file recording means for recording the image data compressed by the optimum compression parameter.

[0009]

In the compressed image evaluation method, the original image to be compressed is captured by the capturing means and recorded in the original image recording means. A compression parameter is set in the compression means, and the original image recorded in the original image recording means is compressed. A compression rate is calculated by the compression rate calculation means and displayed by the display control means. The original image compressed by the expanding unit is expanded and recorded in the reproduced image recording unit. The image quality deterioration calculating unit reads out the original image and the reproduced image recorded in the original image recording unit and the reproduced image recording unit in block units, and performs arithmetic processing such as S / N to quantify the image quality deterioration. . The image quality deterioration degree quantified by the image quality deterioration calculation means is judged by the deterioration judgment display means in block units, converted into deterioration degree display information according to the deterioration degree, and corresponds to the image quality deterioration degree recording means. Write to the block position. The display control means selectively displays the recording information recorded in the reproduction image recording means, the reproduction image recording means, or the image quality deterioration degree recording means according to an instruction from the operator.

As a result, the operator who evaluates the image quality gives the display control means the record information of the image quality deterioration degree recording means in which the degree of the image quality deterioration of the reproduced image is recorded in the color unit of each block. By displaying it, the degree of image quality deterioration and the position of image quality deterioration can be easily identified. Further, by switching and displaying the information of the original image recording means and the reproduced image recording means and the compression rate information as necessary, it is possible to judge the image quality deterioration while referring to the original image, the reproduced image and the compression rate.

Further, in the image data compression display device, the original image to be compressed is captured by the capturing means and recorded in the original image recording means. The image quality deterioration position of the reproduced image is determined by using the compressed image evaluation method, and a portion of the reproduced image where the image quality is significantly deteriorated is instructed to the reproduced image cutout unit. The compression parameter generation means prepares compression parameters for the number of screens displayed in multi-screen. The original image recorded in the original image recording means is repeatedly compressed and expanded by the compression means and the expansion means by changing the compression parameter by the number of screens of the multi-screen to be displayed as a list. At the same time, the compression ratio calculation unit obtains the compression ratio of each reproduced image, and the reproduction image cutout unit cuts out only the portion instructed by the operator, and the specified range of each reproduction image and the compression ratio information are combined. The screen is displayed. Further, the image quality deterioration degree of each reproduced image displayed on the multi-screen is displayed by using the evaluation method of the compressed image. The operator evaluates the reproduced image displayed on the multi-screen, the image in which the image quality deterioration degree is displayed, and the compression rate, and selects the one having the least image quality degradation and the good compression rate by the selection means. The original image data recorded in the original image recording unit is compressed using the compression parameter selected by the selecting unit. The compressed image data is recorded in the file recording means together with the compression parameter.

Thus, the operator objectively judges the screen on which the image quality deterioration degree is displayed in block units and the compression ratio and selects the one with the least image quality deterioration, and the optimum compression parameter is selected. The image data can be compressed and recorded with the selected compression parameter.

[0013]

Embodiments of the present invention will now be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing the structure of a compressed image evaluation method according to an embodiment of the present invention. In the compression method of the present embodiment, orthogonal transformation is performed for each block of N × N pixels, and compression is performed using N × N quantization tables. In addition, the compression parameter is information including information related to an image (data size, image size, etc.) and information related to compression (quantization table, Huffman code table, etc.).

In the figure, 1 is an image input unit for inputting an image and converting it into a digital image, 2 is an original image display memory for temporarily recording the original image data input from the image input unit 1, and 3 is an original image display memory. The compression unit 4 for compressing the original image data of the image display memory 2 with the set compression parameter compares the original image data size of the original image display memory 2 and the compressed data size of the compression unit 5 with a compression rate. Is calculated to generate compression rate information, 5 is a decompression section that decompresses the compressed data compressed by the compression section 3 with the set compression parameter, and 6 is the reproduction image data decompressed by the decompression section 5. A reproduced image display memory which is temporarily recorded, and 7 is a unit for reading out the image data recorded in the original image display memory 2 and the reproduced image display memory 6 for each block to calculate the image quality deterioration of each block. An image quality deterioration calculation unit for quantifying the degree of the image quality, a deterioration determination coloring unit for determining the numerically determined image quality deterioration degree calculated by the image quality deterioration calculation unit 7 for each block, and converting it to corresponding color information; The image quality deterioration degree display memory for temporarily recording the color information converted by the deterioration determination coloring unit 8 in block units is an image written in the original image display memory 2, the reproduced image display memory 6, and the image quality deterioration degree display memory 9. A display control unit that controls the display of the data and the compression information obtained by the compression calculation unit 4, 11 is a display that displays the video signal output from the display control unit 12, and 12 is an instruction input unit through which the operator inputs an instruction. Reference numerals 13 and 13 are CPUs for controlling the whole of the present invention, and 14 is a system bus for exchanging data and control commands of each portion of the present invention.

Next, the operation of each part of this embodiment will be described.

The image input unit 1 digitizes image information captured by a camera, a scanner or the like and digitizes the original image display memory 2
Write in. The compression unit 3 compresses the original image data of the original image display memory 2 using the set compression parameter.
The compression rate calculation unit 4 obtains the original image data size of the original image display memory 2 and the compressed data size compressed by the compression unit 3, and adds the compressed data size by the original image data size to calculate the compression rate. . The compressed data compressed by the compression unit 3 is expanded by the expansion unit 5 and converted into reproduced image data.

FIG. 2 shows a block diagram of the compression unit 3 and the expansion unit 5.

Reference numeral 21 is an orthogonal transformer for cutting out the block of the image data to be compressed and orthogonal transformation of the cut out block, and 22 is a quantizer for quantizing each block subjected to the orthogonal transformation with a set quantization table. , 23 is an encoder for encoding the quantized data, and 24 is an encoder 2
Decoder for decoding the compressed data encoded in 3.
5 is a quantizer 2 for compressing the compressed data decoded by the decoder 24;
An inverse quantizer that performs inverse quantization using the same quantization table as that in 2 and an inverse orthogonal transformer 26 that performs inverse orthogonal transformation of each block inversely quantized by the inverse quantizer 25. Compressor 3
When the compression parameter is set, extracts the quantization table information and sets it in the quantizer 22, and sets the information regarding coding in the encoder 23. Next, the original image data is read from the original image display memory 2 and divided by the orthogonal converter 21 into blocks of N × N pixels, and then orthogonal conversion is performed. The method of orthogonal transform is discrete cosine transform,
Hadamard transform, Fourier transform, or the like can be used. Next, each block of the orthogonally transformed image data is
Quantization is performed by the quantizer 22 using a quantization table set for each block. Further, each quantized block is variable-length coded by the encoder 23 to be compressed data. As a variable length coding method, Huffman coding, arithmetic coding, or the like can be used.

By the operations of the orthogonal transformer 21, the quantizer 22 and the encoder 23, the original image data read from the original image display memory 2 is compressed by the compression unit 3 into irreversible compressed data.

In the decompressing unit 5, when the compression parameters used for compression in the compressing unit 3 are set, the quantization table information is extracted and set in the dequantizer 25, and the information about the coding is sent to the decoder 24. Set. Next, when the compressed data compressed by the compression unit 3 is received, it is decoded by the decoder 24, and the inverse quantizer 25 performs inverse quantization for each block. The inversely quantized compressed data is inversely orthogonally transformed for each block by the inverse orthogonal transformer 26, decompressed into reproduced image data, and written in the reproduced image display memory. Decoder 2
By the operations of 4, the inverse quantizer 25, and the inverse orthogonal transformer 26, the compressed data compressed by the compression unit 3 is expanded to the image data before compression although there is a loss of information.

Next, returning to FIG. 1, the operation of the image quality deterioration calculation unit 7 and thereafter will be described.

The image quality deterioration calculation unit 7 uses the original image display memory 2
Of the original image data and the reproduced image data of the reproduced image display memory 6 are divided into blocks of N × N pixels used for compression and expansion, and then read in block units to calculate the degree of image quality deterioration in block units. I do. The calculation method for digitizing the degree of image quality deterioration between the original image and the reproduced image is S / N.
Measurement may be mentioned, but other calculation methods may be used.

The deterioration determination coloring section 8 is composed of an image quality deterioration calculating section 7
The degree of image quality deterioration calculated in step S4 is determined for each block, converted into color information according to a color conversion table prepared in advance, and written into the corresponding block position of the image quality deterioration degree display memory 6. For example, if the color conversion table is set to convert color information closer to red, which is a warning color, as the degree of image quality deterioration is greater, and to color information closer to green, which is a safe color, as the degree of image deterioration is smaller, Degradation determination coloring section 8
According to the calculation result of the image quality deterioration calculation unit 7, blocks having large image quality deterioration are converted to colors close to red, blocks having small image quality deterioration are converted to colors close to green, and blocks having intermediate image quality deterioration are close to yellow. After converting to a color, the block range at the corresponding block position of the image quality deterioration degree display memory 6 is filled with the converted color.

The display control unit 10 selectively reads out the original image of the original image display memory 2, the reproduced image of the reproduced image display memory 6 and the compression information obtained by the compression ratio calculation unit 4 according to the display command of the CPU 13, It is converted into an image signal and output to the display 4. The instruction input unit 12 includes a pointing device, a keyboard, and the like, and outputs an instruction from the operator to the CPU 13
To enter. For example, when the operator inputs an instruction to display the image quality deterioration degree information from the instruction input unit 12, the CPU 13 commands the display control unit 10 to display the contents of the image deterioration degree display memory 9. The display control unit 10 reads out the image information from the image deterioration degree display memory 9 and displays it on the display 1
Display in 1. As a result, the operator can confirm the image quality deterioration degree information in which a block having a larger image quality deterioration is displayed in a color closer to red and a block having a smaller image quality deterioration is displayed in a color closer to green on the display 11.

Next, the control operation of the CPU 13 from the acquisition of the original image to the evaluation of the image quality of the reproduced image in this embodiment will be described.

FIG. 3 is a flow chart showing the flow of the compressed image evaluation processing according to this embodiment.

First, the original image is digitized by the image input section 1 and is taken into the original image display memory 2 (S1).
A compression parameter for performing compression evaluation is set in the compression unit 3, the original image data recorded in the original image display memory 2 is compressed, and compressed data is generated (S2). Here, the compression parameter is set by the operator using the instruction input unit 12 to set information such as a compression range, a quantization table, a Huffman code table, and the like. The compression rate calculation unit 4 calculates the compression rate (S
3). The compression parameter used for compression is set in the decompression unit 5, the generated compressed data is decompressed, and the reproduced image data is written in the reproduced image display memory 6 (S4).

Next, the degree of image quality deterioration of each block is measured, the measurement result is subjected to color conversion, and color information is written in the image quality deterioration degree display memory 9 in block units to perform image quality deterioration degree coloring processing. Here, S / N measurement is used in the process of digitizing the degree of image quality deterioration of each block, and the degree of image quality deterioration is divided into four levels, and red, orange, yellow, and green color conversion is performed in the order of increasing image quality deterioration. A case will be described.

First, the read blocks of the original image display memory 2 and the reproduced image display memory 6 are first set to the block positions for S / N measurement, and the write block positions of the image quality deterioration display memory 9 for writing the color converted color information are set. The initial position is set (S5). The original image data and the reproduced image data in the read blocks of the original image display memory 2 and the reproduced image display memory 6 are read (S6). The S / N is calculated from the read original image data and reproduced image data (S7). When the calculation result is 40 or more (S8: Yes), the writing block of the image quality deterioration display memory 9 is filled with green (S).
9). When the calculation result is less than 40 and is 30 or more (S10:
Yes), the writing block of the image quality deterioration display memory 9 is filled with yellow (S11). 2 if the result is less than 30
If 0 or more (S12: Yes), the writing block of the image quality deterioration display memory 9 is filled with orange (S1).
3). If the calculation result is less than 20, the writing block of the image quality deterioration display memory 9 is filled with red (S14).
Filling processing for writing blocks (S9, S11, S1)
When S3, S14) is completed, it is determined whether there is a next read block (S15). If there is a next read block (S15: Yes), the read block position of the original image display memory 2, the reproduced image display memory 6, and the write block position of the image quality deterioration display memory 9 are moved to the next block position (S16). Next, the processing from the processing S6 is repeated to continue the image quality deterioration degree coloring processing. If there is no next read block (S15: No), the image quality deterioration degree coloring process ends. By performing the image quality deterioration degree coloring process, the degree of image quality deterioration of the reproduced image data with respect to the original image data is color-coded in block units and written in the image quality deterioration display memory 9. Further, when the operator inputs an instruction to display the degree of image quality deterioration from the instruction input unit 12, the display control unit 10 displays the contents of the image quality deterioration display memory 9 on the display 11 by an instruction from the CPU 13.

By the CPU 13 performing the above operation, the operator intuitively displays the deterioration of the image quality of the reproduced image when the original image is compressed with the set compression parameter by displaying the degree of the image deterioration in different colors. Can be evaluated.
Further, when the operator inputs a display instruction of the original image, the reproduced image, and the compression ratio information from the instruction input unit 12, the display control unit 10 compresses the original image display memory 2, the reproduced image display memory 6, and the compression ratio calculation unit 4. Since the display is switched to the rate information, it is possible to refer to the original image, the reproduced image, and the compression rate information.

In the above embodiment, the degree of image deterioration is divided into four levels and displayed in four colors. However, if the number of colors can be displayed in the image quality deterioration degree display memory 9, the image quality deterioration is divided into several levels. You can color-code the degree of. Furthermore, although the image quality deterioration degree is converted into color information, it may be converted into information such as symbol information and character information that can display the state of image quality deterioration.

The image quality deterioration calculation unit 7 in FIG. 1 performs a calculation for digitizing the degree of image quality deterioration for each block of N × N pixels used for compression, but has a unit of pixels that is completely unrelated to the compression process. You may use the block of.

In FIG. 1, the original image memory 2, the reproduced image display memory 6, and the image quality deterioration degree display memory 9 are dedicated to each image data.
Although one display memory is used, if a recording area for each image data is prepared, the present embodiment can be realized with one general-purpose display memory. Further, the image quality deterioration degree display memory 9 may display as many colors as the number of colors to be displayed.

The display control unit 10 of FIG. 1 selectively switches between the original image display memory 2, the reproduced image display memory 6, and the image quality deterioration degree display memory 9 to display, but the necessary portions of each display memory are simultaneously displayed. It is also possible to perform display control of reading and overlapping display so that the contents of each display memory are displayed simultaneously.

Next, a second embodiment will be described with reference to FIG.

In the present embodiment, a part of the reproduced image compressed and expanded with different compression parameters for the same original image is displayed in a multi-screen, and the reproduced image displayed in the multi-screen is selected to determine the compression parameter. The image data compression device is configured to perform the image quality deterioration degree coloring process described in the above-mentioned embodiment on each reproduced image displayed on a multi-screen and intuitively evaluate the image quality of the reproduced image for each compression parameter. Is.

FIG. 4 is a block diagram of an image data compression apparatus according to this embodiment. In FIG. 4, the components denoted by the same reference numerals as those in FIG. 1 are the same. The configuration newly added and changed in the figure will be described. Reference numeral 41 denotes a compression parameter generation unit that generates compression parameters for the number of multi-screens to be compared and sets the compression parameters in the compression unit 3 and decompression unit 5, and 42 cuts out a necessary portion of the reproduction data decompressed by the decompression unit 5. A reproduction image cutout unit that controls writing so that a multi-screen display is performed, and a file recording unit 43 that records compressed compressed data together with compression parameters.

Next, the operation of the added portion will be described.
The compression parameter generation unit 41 is given an instruction from the CPU 13 through the system bus 14 and generates a compression parameter according to the given instruction. The compression parameter is a parameter including information such as a compression range, a quantization table, and a Huffman code table. Here, the quantization table that affects the image quality of the reproduced image is the compression parameter generation unit 4
A description will be given of a case where the information such as the compression range and the Huffman code table generated in step 1 is given by the CPU 13 and set.

FIG. 5 is a block diagram when the quantization table of the compression parameter generator 41 is generated. Reference numeral 51 is a basic quantization table recording area in which a basic quantization table prepared in advance according to the present invention is recorded, and 52 is a selection quantization table recording in which a quantization table in the compression parameters selected by the operator is recorded. A region, 53 is a quantization table selection arithmetic circuit for selecting a required quantization table from the tables recorded in the basic quantization table recording region 51 or the selection quantization table recording region 52.
Numeral 4 is a quantization table recording arithmetic circuit for recording a newly selected quantization table in the selected quantization table recording area 52 or for calculating the respective recorded quantization tables to generate and record a new quantization table. Is.

From the CPU 13 to the compression parameter generator 41
When a command for preparing the specified number of compression parameters is given to, the quantization table selection arithmetic circuit 53 selects the required number of quantization tables from the basic quantization table recording area 51 prepared in advance. . The selected quantization table is sequentially set in the compression unit 3 and the decompression unit 5 by a setting command from the CPU 14. For example, when 64 quantization tables are prepared in the basic quantization table recording area 51 in advance and an instruction to prepare 8 basic quantization tables is given, the 64 quantization tables are selected. Select the required eight quantization tables. On the contrary, when eight quantization tables are prepared in the basic quantization table recording area 51 in advance and an instruction to prepare 64 basic quantization tables is given, the eight prepared quantization tables are prepared. 56 quantization tables are newly generated from the quantization tables, and 64 quantization tables are selected together with the prepared quantization tables. As the quantization table generation method, two pieces are selected from the eight prepared quantization tables, and each corresponding element (quantization step width) of the selected quantization table is calculated (for example, the average). Is performed to generate a new quantization table.

From the CPU 13 to the compression parameter generator 41
When a command for recording the selected compression parameter is given to the CPU 1, the quantization table recording arithmetic circuit 54
The quantization table in the compression parameters given by No. 3 is recorded in the selected quantization table recording area 52.

From the CPU 13 to the compression parameter generator 41
When a command for preparing a specified number of compression parameters is given to, the quantization table recording operation circuit 54 performs an operation between all the quantization tables recorded in the selected quantization table recording area 52, and the required number Is generated and recorded, and is selected by the quantization table selection arithmetic circuit 53. The selected quantization table is sequentially set in the compression unit 3 and the decompression unit 5 by a setting command from the CPU 13.
For example, when seven quantization tables are recorded in the selected quantization table recording area 52 and an instruction to prepare one selected quantization table is given, calculation is performed among the seven quantization tables. Perform and record one quantization table. In the calculation between all the quantization tables, the minimum value is extracted from the corresponding elements (quantization step widths) of all the quantization tables to generate a new quantization table. The calculation for each element may be not only the minimum value but also the maximum value or the average value. On the contrary, when three quantization tables are recorded in the selective quantization table recording area 52, if an instruction to prepare eight selective quantization tables is given, eight out of three quantization tables are given. Generate and record the quantization table of. As the quantization table generation method, two pieces are selected from the eight prepared quantization tables, and each corresponding element (quantization step width) of the selected quantization table is calculated (for example, the average). Is taken) to generate a new quantization table.

The compression parameter generation unit 41 has a CPU 13
By executing the above-mentioned operation in accordance with the above command, the quantization table required for compression comparison according to the present invention is generated.

Next, returning to FIG. 4, the reproduced image cutout unit 4
The write operation of No. 2 will be described.

Upon receiving the reproduced image data decompressed by the decompressing unit 5, the reproduced image cutout unit 42 cuts out only the range set by the CPU 13 from the received reproduced image data, and reproduces the cutout in the reproduced image display memory 6. Write control is performed so that image data is displayed on a multi-screen. At this time, the writing position of the reproduced image data to be written is controlled so that it does not overlap for each reproduced image data to be compressed and compared. As a result, the partial image data of the plurality of reproduced images compressed and expanded by the plurality of quantization tables generated by the compression parameter generating unit 41 is written so as to be displayed on a multi-screen.

When the compressed data recording command is given from the CPU 13, the file recording unit 43 records the compressed data compressed by the compression unit 3 and the compression parameters used for compression as a compressed image file.

Next, the control operation of the CPU 13 in the present embodiment until the original image is captured, the compression parameter for optimum compression is determined, and the compressed data is recorded will be described.

FIG. 4 is a flow chart showing the flow of processing according to this embodiment.

First, the operator sets basic compression parameters (S21). The original image is digitized by the image input unit 1 and fetched in the original image display memory 2, and the entire original image data is compressed and expanded by the set compression parameter and written in the reproduced image display memory 6 to perform the image quality deterioration degree coloring process ( S22). Here, the image quality deterioration degree coloring process is the process described with reference to FIG. 3 of the first embodiment.
Next, the operator evaluates the reproduced image in which the degree of the image quality deterioration is color-coded and displays, and designates the portion of which the image quality deterioration is large and the compression parameter is changed to improve the image quality from the instruction input unit 12 (S23). Further, the operator designates the number of multi-screens to be displayed and compared at one time from the instruction input unit 12 (S2
4). When the number of multi-screens is input, a command for preparing the basic quantization table for the number of multi-screens is issued to the compression parameter generation unit 41, the quantization tables for the number of multi-screens are prepared, and the prepared quantization tables are used. Then, compression and decompression for the number of multi-screens are performed (S25). Further, the compression ratio in each compression / expansion process is performed by the compression ratio calculation unit 4, the reproduction image data is cut out by the reproduction image cutout unit 42, and the partial image of each reproduction image data and the compression ratio information are multiplexed. A screen is displayed (S26).

The internal processing flow of the processing S25 and the processing S26 for realizing the multi-screen display will be described with reference to the block diagrams of the compression unit 3 and the decompression unit 5 of FIG. 2 and the flowchart of FIG.

First, the first quantization table prepared by the compression parameter generation unit 41 is set in the quantizer 22 of the compression unit 3 and the dequantizer 25 of the decompression unit 5 (S4).
1). Next, the entire screen of the original image data on the original image display memory 2 is orthogonally transformed by the orthogonal transformer 21 (S42). The quantizer 22 quantizes the orthogonally transformed data using the quantization table set in the process S41 (S43).
The quantized data is encoded by the encoder 23 to generate compressed data (S44). At this time, the compression ratio calculation unit 4
The compression rate is calculated from the compressed data size and the original image data size to obtain the compression rate information (S45). Next, the compressed data compressed by the decoder 24 is decoded (S46). The dequantizer 25 dequantizes the quantized data using the quantization table set in the process S41 (S4).
7). The inversely quantized data is inversely orthogonally transformed by the inverse orthogonal transformer 26 (S48). Next, from the reproduced image data compressed and expanded by the first quantization table, the reproduced image cutout unit 42 cuts out the reproduced image data in the range designated by the operator in step S23 of FIG. 6, and displays the reproduced image data. It is written in the first position of the multi-screen display of the memory 6 (S48). Further, the compression rate information calculated in the process S45 is written in the first position of the multi-screen display of the reproduction image display memory (S50). Here, in the compression / expansion processing using the second and subsequent quantization tables, similarly, the reproduction image cutout unit 42 and the compression ratio calculation unit 4 perform the writing in the second and subsequent display positions of the multi-screen. next,
When the partial image data of the reproduced image written by the reproduced image cutout unit 42 is the last reproduced image data of the multi-screen display (S51: YES), the compression rate of each reproduced image data displayed in the multi-screen is checked. The display order of the multi-screens is rearranged in the order of smaller compression rate (S53). Process 5
After step 3, the multi-screen display process with compression information ends. When the partial image data of the reproduced image written by the reproduced image cutout unit 42 is not the last data of the multi-screen display (S51: NO), the compression parameter generation unit 4
The next quantization table prepared in 1 is set in the quantizer 22 of the compression unit 3 and the dequantizer 25 of the decompression unit 5 (S5).
2). Next, the processing from the processing S42 is repeated to continue the multi-screen display of the reproduced image.

By performing the above-described processing in processing S25 and processing S26, the compressed / decompressed partial image data is displayed on a multi-screen.

Next, returning to FIG. 6, the processing from the processing S27 will be described.

Image quality deterioration degree coloring processing is performed on each reproduced image partial image data of the multi-screen created in step S26, and color information corresponding to the degree of image quality deterioration of each reproduced image partial image data is obtained. CPU 1 writing to display memory 9
When the display control unit 10 displays the contents of the image quality deterioration degree display memory 9 in accordance with the command No. 3, the reproduced image displayed with the image quality deterioration degree colored is displayed on the multi-screen (S27). In step S27, the operator evaluates the compression ratio information corresponding to the image quality deterioration degree coloring display image displayed on the multi-screen, and the operator selects and instructs the image quality deterioration degree coloring display image with the least image quality deterioration and the appropriate compression ratio. (S28). Here, when the image quality deterioration degree coloring display image displayed on the multi-screen is selected, the quantization table (compression parameter) used for compressing the corresponding reproduced image is selected. Further, the quantization table selected in the process S28 is recorded in the selected quantization table recording area 52 of FIG. When the operator inputs an instruction to display the reproduced image displayed on the multi-screen from the instruction input unit 12, the CPU 13 issues a command to the display control unit 10 to display the content of the reproduced image display memory 6. As a result, the playback image displayed on the multi-screen is displayed instead of the image deterioration degree coloring image displayed on the multi-screen.
The image quality can be evaluated by checking the actual reproduced image. In step S28, image quality deterioration is evaluated, and when the quantization table selection by the image selection displayed on the multi-screen is completed, it is checked how many quantization tables are recorded in the selected quantization table recording area 32, and the quantization table is If a plurality of CPUs are selected (S29: YES), the CPU
A command to prepare for the number of multi-screens from the selected quantization table is issued to the compression parameter generation unit 41 by 13, and a quantization table for displaying a new multi-screen is prepared (S30). Next, the processing from step S22 is repeated, and an instruction is given to select a display image with the image quality deterioration degree having the smallest image quality deterioration and an appropriate compression ratio. When only one quantization table is selected (S29: NO), the operator is allowed to select whether to evaluate the image quality of other positions of the original image, and when the image quality of other positions is evaluated ( S31: YES)
Sets the currently selected compression parameter as a basic compression parameter (S32). Next, the processing from the processing S22 is repeated to evaluate the image quality after compression / expansion of the other position of the original image data and select the compression parameter. When the image quality evaluation at other positions is not performed (S31: NO), an instruction to prepare only one quantization table is issued to the compression parameter generation unit 41, the selected quantization table 52 is recalculated, and the entire original image data is calculated. To prepare a quantization table for compressing (S11). The recalculated quantization table (compression parameter) is used to compress the entire area of the original image data (S33). The compressed data obtained by compressing the original image data in the process S33 and the compression parameter used for the compression are recorded as a file in the file recording unit 43 (S34).

By the CPU 13 performing the above operation, the operator can select the image having the smallest image quality deterioration and the highest compression ratio from the images displayed on the multi-screen, and is optimal for compressing the original image data. The compression parameters can be determined.

Although the number of screens for multi-screen display is input in the processing S24 of FIG. 6 of the above embodiment, the number of screens for multi-screen display is automatically calculated from the range in which compression / expansion of original image data is evaluated. It does not matter if you set it later.

Next, a third embodiment will be described below with reference to FIG.

The present embodiment is an image data compression display device for compressing a large amount of image data, recording it together with search information into a database, and searching and displaying the image data from the search information. It is configured such that simple and reliable image data compression registration using the degree coloring process and high-speed reproduction by image data reproduction management can be performed.

FIG. 8 is a block diagram of an image data compression display device according to this embodiment. In FIG. 8, the components denoted by the same reference numerals as those in FIG. 1 are the same.
The configuration newly added and changed in the figure will be described. 8
Reference numeral 1 is a reduced image generation unit for generating a reduced image referred to when searching for image data, 82 is an image data management unit that manages registration and search of image data, and 83 is a reproduction management unit that manages reproduction of image data. Is.

Next, the operation of the added portion will be described.
The reduced image generation unit 81 is configured to control the system bus 1 from the CPU 13.
A reduced image generation command is given through 4, and according to the given command, processing such as thinning is performed from the original image recorded in the original image display memory 2 to generate reduced image data of a specified size. The generated reduced image data is passed to the compression unit 3 and compressed, added to the compressed image file as the search information of the image data, and recorded in the file recording unit 43. When the image data management unit 82 receives an image data registration command from the CPU 13 through the system bus 14, the image data management unit 82 adds the search information required for the search to the compressed data generated by the compression unit 3 and records it in the file recording unit 43. . When a search command is given from the CPU 13, the file registered in the file recording unit 43 is searched according to the given search information. Here, the search information includes character information such as the creation date of the image data and the characteristics of the image data, and the reduced image generated by the reduced image generation unit 81. The reproduction management unit 83 manages reproduction of partial image information associated with the compressed image file searched by the image data management unit 82.
Follow the playback time and playback instructions. The partial image information that accompanies the compressed image file is information in which a portion with severe image quality deterioration is compressed and recorded with another compression parameter.

Next, the control operation of the CPU 13 when the image data registration and the registration image search are performed in this embodiment will be described.

FIG. 9 is a flow chart showing the flow of processing according to this embodiment.

First, the operator selects an execution mode (S61). When the image data registration is selected as the execution mode (S62: Yes), the image data to be registered is fetched from the image input unit 1 (S62). The compression processing of the captured image data is interactively performed with the operator by using the multi-screen display on which the image quality deterioration degree coloring processing described in the second embodiment is performed, and the optimum image data compression is performed (S6).
4). The optimum compressed image data is added with search information such as a reduced image and character information necessary for the search by the image data management unit 82 (S65). The compressed image data to which the search information is added is recorded in the file recording unit 43 (S66).
To end the image data registration (S67: Yes),
The process returns to the execution mode selection of step S61. When the image data registration is to be continued (S67: No), the process S63 to the process S
The process of 66 is repeated to register another image data.

Here, the explanation of the optimization processing of the image data compression of the processing 64 will be given with reference to FIG. In FIG. 10, the processes designated by the same reference numerals as those in FIG. 6 indicate the same processes. Further, FIG. 6 shows a process of determining one compression parameter for all the screens of the image data to be compressed, but in FIG. 10, the basic compression parameter is used for the compression of all the screens, which causes a large deterioration in image quality. The part shows a process of selecting an optimum compression parameter as a partial image and performing compression recording. In the part of FIG. 10 where processing different from that of FIG. 6 is performed, when one display with the least image quality deterioration is selected from the multi-screen image quality deterioration degree coloring display, the original image is designated by the selected compression parameter. Only the range is compressed (S81). The compressed data is recorded as a partial image file together with the compression parameter (S82). When the other range of the original image is also checked (S83: Yes), the process S61 is performed.
The process S82 is repeated to compress another part of the original image, which has a large deterioration in image quality, with another compression parameter and temporarily records it as partial image information. When the other range of the original image is not checked (S83: No), the entire original image is compressed with the basic compression parameter (S84). The compressed data is recorded in the file recording unit 43 in a form managed by the image data management unit 82 together with the compression parameter and the temporarily recorded partial image information (S85). As a result, the entire screen of the original image is compressed with the basic compression parameter, and the portion with large image quality deterioration is compressed with the optimal compression parameter as the partial image, and the search information is stored in the file recording unit 43 as the accompanying information of the entire compressed image data. Recorded in a controlled manner.

Next, returning to FIG. 9, a description will be given of a case where the image search execution is selected in the process 61.

When the image search execution is selected as the execution mode (S68: Yes), the operator's search condition input is accepted (S69). The image data management unit 82 searches the image data registered in the file recording unit 43 according to the input search conditions (S70). Optimal display of the retrieved image data is performed (S71). When the image search execution is ended (S72: Yes), the process returns to the execution mode selection of the process S61. When continuing image search execution (S7
2: No) repeats the processing from the processing S69 to the processing S71 to perform another image data search.

Here, the description of the optimum display process of the process 71 will be given with reference to FIG.

First, the file name of the image data to be reproduced is received (S91). The file information of the image data of the received file name is read (S92). The image data of the entire screen is expanded and displayed according to the read file information (S93). When the reproduction end instruction is input (S94: Yes), the process 71 is ended. When the reproduction end instruction is not input (S94: No),
It is checked whether or not the separately compressed partial image data exists. When the separately compressed partial image data exists (S9)
5: Yes) decompresses and displays the partial image data (S
96), and returns to step S94. If the separately compressed partial image data does not exist (S95: No), the process 71 ends. By the determination processing of processing S94 and processing S95, while the reproduction end instruction is not input, the partial image data attached to the image file designated for display is sequentially displayed, and the image data displayed for a long time is added with the attached partial image data. By doing so, the image quality improves. Further, when the display speed is emphasized rather than the image quality, an excessive partial image data is not reproduced by inputting a reproduction end instruction, so that high-speed display is possible.

Finally, returning to FIG. 9, when the end is selected in the selection process S61 in the execution mode (S73: Yes)
Will explain. The processing of image data registration and registration image search is ended.

By the CPU 13 performing the above operation, the operator selects image data registration or image search to select image data deterioration degree color-processed multi-screen display, and image data registration for which compression parameters can be determined. It is possible to execute an image search capable of high-speed reproduction by managing reproduction of image data.

In the above embodiment, the image quality deterioration degree coloring process and the compression parameter selection process by the multi-screen are used for the image database system which performs the registration search of the image data, but it may be used for the image editing system, the presentation system or the like. Absent.

Although the instruction input from the outside is used to determine the reproduction end in FIG. 11, the reproduction display time may be set to determine the reproduction end.

In FIG. 10, when the original image is compressed by the compression parameter for the full screen, only the range in which the image quality deterioration is particularly bad is compressed by another compression parameter and recorded in the file recording section 43. As in FIG. 6, one compression parameter is selected for all screens, and the file recording unit 43
You may record it in. At this time, FIG.
The optimum display processing of is not necessary. The original image data may be divided into several parts and one compression parameter may be determined for each divided screen.

FIG. 9 shows all the components necessary for compression recording / reproduction, but the compression recording device and the reproduction device may be separately configured.

[0076]

As described above, according to the compressed image evaluation method of the present invention, when the operator sets the original image to be compressed and the compression parameter, the compression image is expanded by the automatically set compression parameter. The degree of image quality deterioration of the reproduced image is displayed in color units of blocks. As a result, the operator who evaluates the image quality of the compressed image can distinguish the degree of image quality deterioration and the image quality deterioration position by intuitive and reliable image quality evaluation that only evaluates the displayed color information. Further, by displaying the original image, the reproduced image, and the compression rate information as necessary, it is possible to determine the image quality deterioration while referring to the original image, the reproduced image, and the compression rate.

Further, according to the image data compression / display apparatus of the present invention, when it is desired to evaluate a plurality of compression parameters for the original image and determine the optimum compression parameter, the degree of deterioration of the image quality displayed in different colors is evaluated. Then, when the operator specifies a part of the original image data, the compressed image is automatically expanded with a plurality of compression parameters, and the reproduced image with each compression parameter is displayed on a multi-screen. Further, the degree of image quality deterioration of each reproduced image displayed on the multi-screen is displayed in different colors. As a result, the operator can determine the optimum compression parameter for the original image to be compressed simply by evaluating the color information according to the image quality deterioration and selecting the color-coded display screen with the least image quality deterioration. Further, even when a plurality of compression parameters are selected, a new compression parameter is automatically calculated from the selected compression parameters, so that the operator can determine the compression parameters without being aware of the compression parameters.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration for realizing a compressed image evaluation method according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a compression unit and a decompression unit that realize compression and decompression according to the present invention.

FIG. 3 is a flowchart illustrating the operation of the compressed image evaluation method according to the present invention.

FIG. 4 is a block diagram showing a configuration of an image data compression display device showing a second embodiment of the present invention.

FIG. 5 is a block diagram showing a configuration of a compression parameter generation unit according to a second embodiment of the present invention.

FIG. 6 is a flowchart illustrating a compression parameter determination operation of the second embodiment of the present invention.

FIG. 7 is a flow chart illustrating an operation of displaying reproduced image data and compression information on a multi-screen according to the second embodiment of the present invention.

FIG. 8 is a block diagram showing a configuration of an image data compression display device showing a third embodiment of the present invention.

FIG. 9 is a flowchart illustrating an image data registration / retrieval operation according to the third embodiment of the present invention.

FIG. 10 is a flowchart illustrating a compression parameter determination operation according to the third embodiment of this invention.

FIG. 11 is a flow chart illustrating a display operation of compressed image data according to the third embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Image input part, 2 ... Original image display memory, 3 ... Compression part, 4 ... Compression ratio calculation part, 5 ... Decompression part, 6 ... Reproduction image display memory, 7 ... Image quality deterioration calculation part, 8 ... Deterioration determination coloring part , 9 ... Image quality deterioration degree display memory, 10 ... Display control unit, 11 ... Display, 12 ... Instruction input unit, 13 ... CPU, 14 ... System bus, 41 ... Compression parameter generation unit, 42 ... Reproduced image cutout unit, 43 ... File recording unit, 81 ... Reduced image generation unit, 82 ... Image data management unit, 83 ... Reproduction management unit.

Claims (9)

[Claims] 1. An irreversible image data compression method in which the compression ratio and the image quality are changed by changing the compression parameter, and a compression means for compressing the digital image data with the set compression parameter, and a compression means for compressing the compressed image data. Decompression means for decompressing, image quality deterioration computing means for computing the image quality degradation of the compressed and decompressed reproduced image data with respect to the original image data on a block-by-block basis, and a degree of image quality degradation of each block is determined to display according to the image quality degradation. Using the deterioration determination display means for converting into information, the digital image data is compressed and expanded by the set compression table, and the image quality deterioration calculation means calculates the degree of image quality deterioration of the reproduced image with respect to the original image in block units, A compressed image evaluation method characterized in that the deterioration determination display means displays image quality deterioration information according to the degree of image quality deterioration. 2. The compressed image evaluation method according to claim 1, further comprising: a deterioration determination coloring unit that determines the degree of image quality deterioration of each block and converts it into color information according to the image quality deterioration.
A compressed image evaluation method characterized by displaying an image quality deterioration evaluation screen color-coded for each block. 3. The compression image evaluation method according to claim 1, wherein a deterioration determination coding unit that determines the degree of image quality deterioration of each block and converts the image quality information into character / graphic information according to the image quality deterioration is used in block units. A compressed image evaluation method, characterized in that an image quality deterioration evaluation screen coded and displayed on is displayed. 4. The compressed image evaluation method according to claim 1, wherein image quality deterioration information is displayed in block units by using a compression ratio calculation means for calculating a compression ratio when the image is compressed with a set compression parameter. A compressed image evaluation method characterized by displaying an image quality deterioration evaluation screen and compression ratio information. 5. An image data compression apparatus for compressing digital image data by selecting an optimum compression parameter by an irreversible image data compression method in which the compression ratio and the image quality are changed by changing the compression parameter. , A decompression means for decompressing the compressed image data in an arbitrary range, a compression parameter generation means for generating a compression parameter, and a multi-screen for displaying a list of decompressed reproduced images. Display means,
Image quality deterioration calculating means for calculating the image quality deterioration of the reproduced image data displayed in a list on a block-by-block basis, deterioration judgment display means for judging the degree of image quality deterioration of each block and converting it into display information, and a list-displayed multi-screen Using a selecting means for selecting a screen determined to have little image quality deterioration among the plurality of compression parameters generated by the compression parameter generating means, the plurality of reproduced images are displayed by the multi-screen display means. Is displayed in a list, the image quality deterioration calculating unit calculates the degree of image quality deterioration of the reproduced image with respect to the original image in block units, and the deterioration determination display unit displays image quality deterioration information in block units according to the degree of image quality deterioration. The displayed image quality deterioration evaluation screens are displayed as a list, and the displayed image quality deterioration evaluation screens are evaluated and selected by the selecting means, and the pressure associated with the selected screen is displayed. Image data compression display apparatus characterized by compressing the digital image data parameter. 6. The image data compression apparatus according to claim 5, wherein the compression rate calculation means for calculating each compression rate when the image data is compressed with a plurality of set compression parameters is used, An image data compression display device, which displays a list of image quality deterioration evaluation screens and reproduced image data that are color-coded and displayed in block units, and compression ratio information. 7. An image data compression display device for compressing and recording a plurality of image data using an irreversible image data compression method in which a compression rate and an image quality are changed by changing a compression parameter. A compression unit for performing compression in an arbitrary range, a decompression unit for decompressing compressed image data in an arbitrary range, a compression parameter generation unit for generating compression parameters, and a multi-screen display for displaying a list of decompressed reproduced images. Means, an image quality deterioration calculation means for calculating the image quality deterioration of the reproduced image data displayed in a list in block units, and a deterioration determination display means for judging the degree of the image quality deterioration of each block and converting it into display information. The selection means for selecting the screen judged to have little image quality deterioration from the multi-screens and the image for adding the search information to the compressed image data. Data management means, recording means for recording compressed data to which search information is added, search means for searching recorded compressed image data, and reproduction management means for managing reproduction of the searched compressed image data , Compressing and decompressing with a plurality of compression parameters generated by the compression parameter generating means, displaying a list of the plurality of reproduced images on the multi-screen display means, and causing deterioration of image quality of the reproduced image with respect to the original image by the image quality deterioration calculating means. The degree is calculated in block units, and the deterioration determination display means displays a list of image quality deterioration evaluation screens in which image quality deterioration information is displayed in block units according to the degree of image quality deterioration, and the displayed image quality deterioration evaluation screens are displayed. The image data is evaluated and selected by the selection means, compressed by the compression means using the compression parameter attached to the selected screen, and the compressed compressed image data Search information is added by the image data management means, compressed image data with search information added by the recording means is recorded, and compressed image data recorded by the recording means is searched by the search means, An image data compression display device, which reproduces the compressed image data retrieved by the reproduction management means. 8. The image data compression display device according to claim 7, wherein the compression means compresses the whole image and the partial image with different compression parameters, and a plurality of compressed image data are compressed by the image data management means. The image data compression display device is characterized in that the compressed image data is sequentially reproduced until the reproduction management means gives a reproduction end instruction. 9. The image data compression display device according to claim 7, wherein the compression means divides one original image data into partial images, and compresses each divided image with different compression parameters. An image data compression display device, characterized in that a plurality of compressed image data that have been compressed are managed in association with each other by the image data management means, and the compressed image data are sequentially reproduced until a reproduction end instruction is given by the reproduction management means.