JPH0690363A - Predictive coding system coding device and decoding device - Google Patents

Abstract

(57) [Abstract] [Purpose] To obtain a high data compression effect for various types of image information and to increase the processing speed. [Structure] After creating a template according to the property of input image information, a predetermined encoding process is executed. [Effect] The prediction accuracy of pixel symbols is improved, and the data compression effect is enhanced. Further, since it is possible to prevent unnecessary surrounding pixels from being referred to, the processing speed is increased especially when processing is performed by software.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention predicts a symbol of each pixel of binary image information according to the states of surrounding pixels, and encodes image information based on the prediction result. , And a decoding device for restoring the code generated by the coding device to the original image information.

[0002]

2. Description of the Related Art In this type of predictive coding apparatus, when predicting each pixel of image information, the prediction accuracy improves as the number of surrounding pixels to be referred increases. Further, the higher the correlation between the pixel position of the peripheral pixel to be referred to and the pixel of interest, the higher the prediction accuracy. When the prediction accuracy is improved, the data compression effect of image information is also improved.

On the other hand, when such image information encoding processing is executed by software, if the number of surrounding pixels to be referred to is increased, the prediction processing will take time and the processing speed will decrease. Therefore, the number of reference pixels is usually set to about 6 to 10 pixels.

FIGS. 16 (a) to 16 (c) show examples of templates indicating the positions of surrounding pixels to be referred to. In the figure, the hatched frame shows the pixel of interest and the blank frame shows the surrounding pixels, and these three types of templates are
In each case, the number of reference pixels is 10.

Since the pixel positions of the surrounding pixels having a high correlation with the target pixel are different depending on the type of image, the matching templates are also different.

Generally, when a binary image such as a character or a graphic is processed, a template in which all the reference pixels are arranged near the target pixel is suitable as shown in FIG. Further, in the case of a pseudo-halftone image such as a picture or a photograph, a template in which reference pixels are dispersed is suitable as shown in FIG.

On the other hand, even when decoding the encoded image information, such a template is used to extract the surrounding pixels of the image information to be sequentially restored. In this case, the template used must be the same as that used in the encoding.

[0008]

By the way, in the case of transmitting an image like a facsimile machine, it is difficult for the transmitting side and the receiving side to use different templates at the same time depending on the type of image to be transmitted. Therefore, conventionally, the same template as shown in, for example, FIG. 16C is used, which can obtain a somewhat high data compression effect for both binary images and pseudo halftone images.

As a result, neither the binary image nor the pseudo-halftone image can be coded under the optimum condition, and the data compression effect is reduced.

Further, since a pixel at a fixed position is referred to regardless of the type of image, a pixel having a low correlation with the pixel of interest is also referred to. Since such a reference process is wasteful and requires a processing time, the processing speed of the encoding process is reduced.

As described above, conventionally, there has been a problem that a high data compression effect cannot always be obtained for various image information and the processing speed is slow.

The present invention solves the above-mentioned problems, always obtains a high data compression effect for various image information, and can increase the processing speed, and an encoding device and a decoding device of a predictive encoding system. The purpose is to provide.

[0013]

To this end, the present invention provides
When the image information is input, a template is created according to the property of the image information, and then a predetermined encoding process is executed.

[0014]

Since the template is created each time in accordance with the inputted image information, a high data compression effect can be always obtained for various image information. Further, in this case, it is possible to prevent unnecessary reference processing from being set in the template at the time of predicting the pixel symbol and not to perform unnecessary reference processing, so that the processing speed is increased.

[0015]

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

FIG. 1 is a block diagram of the transmitting side of an image transmitting apparatus according to an embodiment of the present invention. In the figure, the image information temporary storage unit 101 temporarily stores the input binary image information. The fixed template pixel extraction unit 102 sets each pixel of the input image information as a pixel of interest and extracts each pixel of a fixed peripheral range set in advance. The coincidence count counting unit 103 counts the number of coincidences of symbols between the pixel of interest and the extracted surrounding pixels. The change point counting unit 104 counts the change points of the symbols of the input image information.
The template creation unit 105 creates a template that specifies the positions of surrounding pixels to be extracted during the encoding process. The pixel extraction unit 106 extracts each pixel of the image information according to the created template. The arithmetic code generation unit 107 generates an arithmetic code by known arithmetic processing. The transmitting unit 108 transmits template specification information together with the generated code information.

FIG. 2 is a block diagram of the receiving side of the image transmitting apparatus. In the figure, a receiving unit 201
Is to receive each information transmitted from the transmitting side. The template creating unit 202 creates a template according to the specification information of the template transmitted from the transmitting side. The arithmetic code decoding unit 203 decodes the received code information by known arithmetic processing to restore the original image information. The pixel extraction unit 204 extracts peripheral pixels of the restored image information according to the template.

Next, the operation of the transmission side of the image transmission apparatus of the present embodiment having the above-mentioned configuration will be described.

As shown in FIG. 3, the transmitting side first starts inputting image information. This image information is binary image information obtained by scanning the image to be processed by main scanning and sub-scanning line by line, and this apparatus sequentially inputs the image information of each line pixel by pixel (process 301).

The image information temporary storage unit 101 sequentially stores the input image information. Fixed template pixel extraction unit 102
Pays attention to each input pixel and extracts each pixel around the target pixel. Pixels to be extracted are, as shown in FIG. 4, pixels D1 to D10 that are fixedly determined with respect to the pixel of interest indicated by hatching. The coincidence count counting unit 103 counts the number of coincidences between the extracted pixels D1 to D10 and the target pixel. As a result, the coincidence number counting unit 103 outputs the count values N1 to N10 corresponding to the pixels D1 to D10, as shown in FIG. Also, the change point counting unit 104
Monitors each pixel of the input image information and counts the number of change points every time it changes from white to black or from black to white (above, process 302).

After starting each of the above operations, this image transmission device monitors the number of change points counted (process 303),
If the number of change points is less than a preset fixed number (process 3
03 N), the operation is continued as it is (process 302).
What).

When the number of black-and-white change points of the image reaches the above-mentioned fixed number (Y in step 303), the count values N1 to N10 of the number of coincidences at that time are read. Then, among the count values N1 to N10, the pixel positions showing a predetermined constant value or more are determined, and a template showing those pixel positions is created (process 304).

The count values N1 to N10 indicate the magnitude of the correlation of each pixel position with respect to the target pixel. Therefore, the pixel position that shows a large value and the number of pixels are
It depends on the type of image information.

Now, for example, if image information of a binary image such as a character or a graphic is input, pixel positions showing a large value are concentrated near the pixel of interest and the number of pixels thereof is relatively small. . Thereby, for example, FIG.
A template as shown in is created. On the other hand, if the image information of a pseudo-halftone image such as a picture or a photograph is input, the pixel positions exhibiting a large value are dispersed in a rather wide range with respect to the target pixel, and the number of pixels is relatively large. Thereby, for example, a template as shown in FIG. 6B is created.

The transmitting unit 108 transmits the specifications of the created template, that is, the information indicating the extracted pixel position to the other party (process 305). The image information temporary storage unit 101 sequentially stores the image information until the input is completed. Pixel extraction unit 10
When the template is created, 6 sequentially reads the image information from the image information temporary storage unit 101 and extracts the surrounding pixels according to the template. Arithmetic code generator 1
Reference numeral 07 sequentially generates the code information of the arithmetic code based on the extracted surrounding pixels by the known arithmetic processing for predicting the target pixel. The transmission unit 108 sequentially transmits the generated code information (above, processing 306).

As a result, as shown in FIG. 7, the transmission side of the image transmission apparatus transmits the specification information of the template and the code information which is the encoded image information.

On the other hand, on the receiving side, as shown in FIG. 8, the receiving section 201 first receives the above-mentioned information to be transmitted. The template creation unit 202 reads template specification information from the received information and creates a template (process 401). The arithmetic code decoding unit 203 decodes the received code information by known arithmetic processing and sequentially restores the original image information. The pixel extraction unit 204 pays attention to each pixel of the image information that is sequentially restored, and extracts surrounding pixels according to the template. The arithmetic code decoding unit 203 predicts the pixel of interest with reference to the extracted pixel, and restores the image information pixel by pixel based on the prediction result. The original image information is sequentially restored by such a decoding process (process 40).
2).

As described above, in the present embodiment, the period until the change point of the symbol of the input image information reaches a fixed number,
The number of times the symbol of interest matches the pixel of each pixel in the surrounding constant range is counted for each surrounding pixel position, and a template indicating each pixel position having a large count value is created.

As a result, in the template, it is possible to correctly set only the pixel position having a high correlation with the pixel of interest in the image information that is always input. Since the predetermined encoding process is executed using such a template, the prediction accuracy of each pixel is improved, and a high data compression effect can always be obtained regardless of the type of image information.

Further, since no pixel position having a low correlation with the target pixel is set in the created template,
The number of reference pixels can be minimized. As a result, unnecessary reference processing is not executed, and particularly when the encoding processing is executed by software, the processing speed can be increased.

Further, the counting operation of the number of coincidences between the target pixel and the surrounding pixels is executed until the number of change points of the symbols of the input image information reaches a certain number, so that even in the case of image information having a small number of image information change points. , It is possible to correctly create a template according to the characteristics of image information. Furthermore, when the template is created, the encoding process is executed while continuing to input the image information, so that the encoding process can be started earlier.

FIG. 9 shows the operation of the transmitting side of the image transmitting apparatus according to the second embodiment of the present invention. In the figure, the same symbols as in FIG. 3 indicate the same processing. In the present embodiment, when the operations of storing image information, counting the number of coincidences between a target pixel and surrounding pixels, and counting the change points of image information are started (process 302), the input of image information is monitored (process 307). ) If the image information is continuously input, the above-described operations are continued (from N in step 307 to step 302).

When the input of the image information is completed (Y in processing 307), the number of times the target pixel and the surrounding pixels have been matched by the number-of-matches counter 103 and the number of change points counted by the change-point counter 104 are determined. Create a template based on.

That is, now, the count value of the number of coincidences is N1.
Suppose that N10 and the number of change points are M, the ratio of the number of coincidences to the number of change points, that is, N1 / M, N2 / M ...
-Calculate each value of N10 / M. Then, among the respective values, the pixel positions showing a predetermined fixed value or more are determined, and a template showing the pixel positions is created (process 308). After this, the process 305 of FIG.
The subsequent operations are similarly performed.

As described above, in this embodiment, the number of coincidences between the target pixel and the surrounding pixels is counted for all the input image information, and the template is created based on the count result. As a result, it is possible to correctly create a template according to the characteristics of the entire image information, and a higher data data compression effect can be obtained.

FIG. 10 shows the transmitting side of the image transmitting apparatus according to the third embodiment of the present invention. In the figure, the same reference numerals as those in FIG. 1 indicate the same parts, and the difference from FIG. 1 is that a template selecting section 109 is provided instead of the template creating section 105 and a new template storage section 110 is provided. That is the point. The template storage unit 110 stores in advance templates of various patterns as shown in, for example, FIGS. 11A to 11H, and the template selection unit 109 selects one of them. Is.

FIG. 12 shows the receiving side of the image transmitting apparatus of this embodiment. In the figure, the same reference numerals as in FIG. 2 indicate the same parts, and the difference from FIG. 2 is that a template number selecting section 205 is provided instead of the template creating section 202, and a new template storage section 206 is provided. That is the point. The template storage unit 206 is the same as the template storage unit 110, and the template number selection unit 205 is one for selecting one of the templates stored in the template storage unit 206.

FIG. 13 shows the operation on the transmission side of the image transmission apparatus of this embodiment. In the figure, the same symbols as in FIG. 3 indicate the same processing. In the present embodiment, when the number of change points of the input image information reaches a certain number (Y of processing 303), one matching one is selected from the stored various templates.

This selection is executed as follows, for example. That is, the count values N1 to N1 of the match counter 103
The count value corresponding to the pixel position designated by each template is read from N10 and the average value thereof is calculated.

For example, now, the coincidence counter 103 is
It is assumed that the count values N1 to N10 of each pixel shown in FIG. 5 are designated. Here, for example, considering the template of FIG. 11A, the ten pixel positions indicated by the template correspond to the ten count values N1 to N10 of FIG. 5, respectively. In this case, the average value in the template of FIG.
Calculate as / 10. Further, for example, considering the template of FIG. 11 (g), the seven pixel positions indicated by this template correspond to the count values N4 to N10 of FIG. In this case, the average value is (N3 + N4 ... + N
10) / 7 is calculated.

The average value of the count values is calculated in this way for each template, and one template having the maximum average value is selected (step 309).

Thereafter, as shown in FIG. 14, the selected template number is transmitted (process 310). Next, the code information generated by encoding the image information is transmitted (Process 3
06).

On the other hand, the receiving side reads the transmitted template number and selects the corresponding template. Then, using the selected template, a predetermined image information coding process is executed.

As described above, even if various templates are stored in advance and one that matches the image information to be processed is selected and used, a high data compression effect is always obtained, as in the above embodiment. And the processing speed can be increased.

Further, in this case, the transmitting side need only transmit the template number as the template information to the receiving side, so that the data amount of the transmitting information can be reduced.

In the above embodiment, when one template is selected, the count values corresponding to each template among the count values N1 to N10 are simply averaged. A more appropriate template can be selected by multiplying the count value of the pixel position having a stronger correlation by a coefficient, weighting the average value, and then averaging.

The number of coincidences between the pixel of interest and the surrounding pixels is counted until the number of change points of the image information reaches a certain number. However, as in the case of the second embodiment, the end of the image information is changed. You may make it count up to.

Furthermore, in each of the above-described embodiments, the protocol communication when the transmitting side and the receiving side communicate is not considered, but when executing the protocol communication, the transmitting side performs the device function of the receiving side. The image information may be transmitted after confirming and notifying the receiving side of the transmitting mode.

For example, like a G3 facsimile machine,
CCITT Recommendation T. When performing protocol communication conforming to 30, when the receiving side sends out the CED signal and the NSF signal in response to the incoming call, as shown in FIG.
The NSF signal notifies the transmitting side that it has a transmission function by arithmetic coding and a function of changing the template.

The transmitting side confirms the function of the receiving side by the NSF signal. Then, when transmitting the NSS signal and the TCF signal, the specification and the number of the template to be used are notified by the NSS signal.

According to the procedure described above, the recommendation T. The image transmission of each of the above-described embodiments can be performed by protocol communication conforming to 30.

In each of the above embodiments, the case of transmitting an image has been described, but the present invention can be similarly applied to the case of storing an image in a storage device such as an image filing device.

[0053]

As described above, according to the present invention, a template is created according to the property of input image information and a predetermined encoding process is executed. , A high data compression effect is always obtained, and unnecessary template processing is not performed in the template when predicting pixel symbols, and unnecessary reference processing is not executed, so that the processing speed is increased. It

[Brief description of drawings]

FIG. 1 is a block configuration diagram of a transmission side of an image transmission device according to an embodiment of the present invention.

FIG. 2 is a block configuration diagram of a receiving side of the image transmission device.

FIG. 3 is an operation flowchart of the transmission side of the image transmission device.

FIG. 4 is an explanatory diagram showing surrounding pixels fixedly extracted from input image information.

FIG. 5 is an explanatory diagram of a count value of the number of times of coincidence between the peripheral pixel and the target pixel.

FIG. 6 is an explanatory diagram showing an example of a created template.

FIG. 7 is an explanatory diagram of transmission information.

FIG. 8 is an operation flowchart of a receiving side of the image transmission device.

FIG. 9 is an operation flowchart of the transmission side of the image transmission device according to the second embodiment of the present invention.

FIG. 10 is a block configuration diagram of a transmitting side of an image transmitting apparatus according to a third embodiment of the present invention.

FIG. 11 is an explanatory diagram showing an example of various templates stored in advance in the embodiment.

FIG. 12 is a block configuration diagram of a receiving side of the image transmission device.

FIG. 13 is an operation flowchart of the transmission side of the image transmission device.

FIG. 14 is an explanatory diagram of transmission information in the above embodiment.

FIG. 15 is an explanatory diagram showing a procedure of protocol communication between a transmission side and a reception side in still another embodiment of the present invention.

FIG. 16 is an explanatory diagram showing various known templates.

[Explanation of symbols]

 101 Image Information Temporary Storage Unit 102 Fixed Template Pixel Extraction Unit 103 Match Count Counting Unit 104 Change Point Counting Unit 105, 202 Template Creation Unit 106, 204 Pixel Extraction Unit 107 Arithmetic Code Generation Unit 108 Transmission Unit 109 Template Selection Unit 110, 206 Template Storage unit 201 Reception unit 203 Arithmetic code decoding unit 205 Template number selection unit

Claims (7)

[Claims] 1. A template which shows each pixel position around a pixel of interest is provided, binary image information is sequentially input pixel by pixel, and each pixel of the input image information is focused, depending on the state of the surrounding pixels indicated by the template. In a coding device of a predictive coding system that predicts a symbol of a pixel of interest and generates a code based on the prediction result, template creating means for creating a template according to the property of input image information,
An encoding device of a predictive encoding system, comprising: an encoding execution unit that generates the above-mentioned code using the created template. 2. The template creating means comprises means for counting the change points of the pixel symbols of the image information that are sequentially input, and attention for each input pixel sequentially, and the pixel of interest and each pixel within a certain peripheral range. Means for counting the number of times of symbol coincidence for each pixel position within the surrounding constant range, and reading the number of coincidences for each pixel position when the count value of the changing point of the pixel symbol reaches a constant value. The coding apparatus according to claim 1, wherein the coding apparatus comprises a means and a means for creating a template indicating each pixel position where the read count value is relatively large. 3. The template creating means comprises means for counting the change points of the pixel symbols of the image information that are sequentially input, and sequentially focusing attention on each input pixel and the target pixel and each pixel in a peripheral fixed range. Means for counting the number of times of coincidence of symbols for each pixel position in the surrounding constant range, and for each of the count values of the changing points of the pixel symbols when the counting operation is completed for all input image information. 2. The method according to claim 1, further comprising: a unit for determining a ratio of the count value of the number of coincidences for each pixel position, and a unit for creating a template indicating each pixel position having a relatively large determined ratio. Predictive coding system coding device. 4. A plurality of templates having different pixel positions shown are provided, and the template creating means is means for selecting one template from the plurality of templates according to the property of image information input. The coding apparatus according to claim 1, wherein the predictive coding method is used. 5. The predictive coding system coding apparatus according to claim 1, further comprising means for adding information indicating specifications of a template used to the generated code and outputting the code. 6. A template showing each pixel position around a pixel of interest is provided, and when restoring image information from an input code, the symbol of the pixel of interest is predicted by the state of surrounding pixels shown by the template, and the prediction is performed. 2 based on the result
In a predictive coding method decoding device that restores value image information pixel by pixel, input means for inputting information indicating template specifications together with a code, template creating means for creating a template according to the input information, and created template And a decoding execution unit that restores the binary image information by using the decoding device. 7. A plurality of templates having different pixel positions shown are provided, and the template creating means is means for selecting one of the plurality of templates according to input information. A predictive coding method decoding device according to claim 6.