JP3295507B2 - Binary image encoding / decoding method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a binary image encoding / decoding method for encoding binary image data in blocks of a predetermined shape.

[0002]

2. Description of the Related Art In general, black-and-white binary image data has a very large amount of information. Therefore, if it is handled as it is, for example, the capacity of a recording medium required for storage becomes enormous or transmitted. Inconveniences such as a very long time for the operation occur.

Therefore, in an information processing apparatus that handles such black-and-white binary image data, the image data is stored in a state of being encoded and compressed, or data is transmitted.
As a method (encoding method) for encoding and compressing such image data, for example, in a group 3 facsimile apparatus, a one-dimensional encoding method called an MH method or a two-dimensional encoding method called an MR method is adopted. Further, a group 4 facsimile apparatus employs a two-dimensional coding system called an MMR system.

In recent years, JBIG (Joint Bi)
-Level Image Coding Explorer
ts Group), a coding method having a large coding efficiency is being standardized, and this coding method is called a JBIG method. In this JBIG system, for example, 400
With a resolution of (dots / 24.5 (mm)), the encoding compression rate can be improved by about 30% compared to the MMR method.

[0005]

However, even if such a latest encoding method is applied, the amount of information of the original image data is enormous, so that the amount of code data after encoding and compression is still quite large. Become. In particular, in recent years, there has been a demand to improve the resolution of the images handled,
It is required to further improve the encoding compression rate.

The present invention has been made in view of such circumstances, and has as its object to provide a binary image encoding / decoding method capable of further improving the encoding compression rate.

[0007]

According to the present invention, there is provided a binary image encoding / decoding method for encoding binary image data in units of blocks having a predetermined shape. For the target pixel, a prediction process of predicting the color of the target pixel based on the black-and-white distribution of a plurality of surrounding pixels adjacent to the target pixel is applied, and a pixel in which the prediction result of the prediction process does not match is determined in the same block. When the number is equal to or more than the predetermined number, a predetermined binary image encoding process is applied to the block to form code data of the block, and when the prediction result of the above-described prediction process does not match the number of pixels in the same block that is less than the predetermined number. Does not apply the binary image encoding process to the block,
At the time of decoding, for a block in which code data is formed, a predetermined binary image decoding process is applied to the code data to form image data of the block, and for a block in which code data is not formed. In the above, the above-described prediction processing is executed, and image data in the block is formed based on the prediction result.

In a binary image encoding / decoding method for encoding binary image data in block units of a predetermined shape, at the time of encoding, for each block, a target pixel to be encoded is assigned to the target pixel. Applying a prediction process of predicting the color of the target pixel based on the black-and-white distribution of a plurality of adjacent peripheral pixels, and when there is a predetermined number or more of pixels in which the prediction results of the prediction process do not match in the same block,
A predetermined binary image encoding process is applied to the block to form code data of the block, a first identification mark is added to the head of the code data, and pixels for which the prediction result of the prediction process does not match Is less than the predetermined number in the same block, the binary image encoding process is not applied to the block, the second identification mark is used as code data of the block, and the first identification When the code data to which the mark is added is detected, a predetermined binary image decoding process is applied to the code data to form the image data of the block, and when the second identification mark is detected, 1 The above-described prediction processing is executed for pixels of a block, and image data in the block is formed based on the prediction result.

In a binary image encoding / decoding method for encoding binary image data in block units of a predetermined shape, at the time of encoding, for each block, a target pixel to be coded is assigned to the target pixel. Applying a prediction process of predicting the color of the target pixel based on the black-and-white distribution of a plurality of adjacent peripheral pixels, and when there is a predetermined number or more of pixels in which the prediction results of the prediction process do not match in the same block,
A predetermined binary image encoding process is applied to the prediction error data obtained for the block to form code data of the block, and the number of pixels whose prediction results do not match in the prediction block is less than a predetermined number in the same block. If there is, the above-mentioned binary image encoding process is not applied to the block, and at the time of decoding, the above-mentioned prediction process is executed for one block of pixels for the block in which code data is formed, and And a predetermined binary image decoding process is applied to the code data to calculate a prediction error value for one block, and the pixel prediction value and the prediction error value are calculated in pixel units. To form image data for one block,
For a block in which code data is not formed, the above-described prediction processing is executed, and image data in the block is formed based on the prediction result.

In a binary image encoding / decoding method for encoding binary image data in block units of a predetermined shape, at the time of encoding, for each block, a target pixel to be coded is assigned to the target pixel. Applying a prediction process of predicting the color of the target pixel based on the black-and-white distribution of a plurality of adjacent peripheral pixels, and when there is a predetermined number or more of pixels in which the prediction results of the prediction process do not match in the same block,
A predetermined binary image encoding process is applied to the prediction error data obtained for the block to form code data for the block, a first identification mark is added to the beginning of the code data, and the prediction process is performed. When there is less than a predetermined number of unmatched pixels in the same block in the same block, the above-described binary image coding process is not applied to that block, and the second identification mark is used as the code data of the block to perform decoding. At this time, when detecting the code data to which the first identification mark is added, the prediction process is performed on the pixels of one block to calculate a pixel prediction value of one block, and the detected code data is detected. A predetermined binary image decoding process is applied to calculate a prediction error value for one block, and the pixel prediction value and the prediction error When the second identification mark is detected, the prediction processing is performed on the pixels of one block, and the image data in the block is obtained based on the prediction result. Is formed.

[0011]

Therefore, when the pixel values in the block can be almost completely generated by the prediction processing, no code data is created, so that the coding compression efficiency becomes very large. Further, by encoding the prediction error value in block units, it is possible to realize the encryption of the code data.

[0012]

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

FIG. 1 shows a group 3 facsimile apparatus according to an embodiment of the present invention.

In FIG. 1, a control unit 1 performs a control process of each unit of the facsimile apparatus and a facsimile transmission control procedure process.
The control unit 1 stores a control processing program to be executed, various data necessary for executing the processing program, and the like, and constitutes a work area of the control unit 1. This is for storing various information unique to the facsimile machine.

The scanner 4 reads an original image into a black and white binary image at a predetermined resolution.
Is for recording and outputting an image at a predetermined resolution. The operation display unit 6 is for operating this facsimile machine, and is composed of various operation keys and various displays.

The binary image encoding / decoding section 7 encodes and compresses the image signal and decodes the encoded and compressed image information into the original image signal. , MH, MR, MMR, and J
The BIG system is provided. The image storage device 8 is for storing a large number of image information in a coded and compressed state.

The group 3 facsimile modem 9 is for realizing a group 3 facsimile modem function, and exchanges transmission procedure signals with a low-speed modem function (V.21 modem) and mainly exchanges image information. High-speed modem function (V.33 modem, V.33
29 modem, V.29. 27ter modem etc.).

The network controller 10 is for connecting the facsimile apparatus to a public telephone network and has an automatic transmission / reception function.

The control unit 1, the system memory 2,
Parameter memory 3, scanner 4, plotter 5, operation display unit 6, binary image encoding / decoding unit 7, image storage device 8,
The group 3 facsimile modem 9 and the network control device 10 are connected to a system bus 11, and data exchange between these components is mainly performed through the system bus 11.

Data exchange between the network controller 10 and the group 3 facsimile modem 9 is directly performed.

In this embodiment, the binary image data read by the scanner 4 is encoded and compressed by the following method.

First, as shown in FIG. 2A, image data is divided into 8 × 8 size blocks BLK, and encoding is performed in units of these blocks. In this encoding, the template TMP having the shape shown in FIG. 2B is converted into a target pixel to be encoded (indicated by X in the template).
To extract 10 surrounding pixels A to J of the target pixel X, apply a prediction process of a predetermined prediction algorithm to the surrounding pixels A to J, and obtain a prediction value (white or black) of the target pixel X. ) Is calculated, the predicted value is compared with the value of the target pixel X, and the number of cases where the values do not match is formed for each block.

When the number of mismatches is smaller than a predetermined value, it is determined that the pixel values included in the block can be automatically generated by the above-described prediction processing, and Is not performed. At the same time, when the number of mismatches is equal to or greater than a predetermined value, the binary image encoding process is applied to the block. In this case, JBIG encoding processing is applied.

After the encoding process, the block code data as shown in FIG. 3A is formed for each block BLK. This block code data is obtained by adding a flag bit F indicating the type of code data of the block to the head of the code data of one block.

As shown in FIG. 2B, only the flag bit F of data "0" is arranged as the block code data for the block that has not been subjected to the coding process by the above-described coding process. You. As for the block on which the encoding process has been performed, the flag bit F of the data “1”, the code data of the block, and the block end code EOB indicating the end of the code data, as shown in FIG. Are formed in this order to form block code data.

Here, the template TMP described above is
This corresponds to the one applied in the predictive encoding process adopted in the JBIG system, and the above-described prediction process uses the predictive encoding process adopted in the JBIG system. Further, the block code data of the block on which the encoding process has been performed is equivalent to code data itself obtained by encoding and compressing according to the JBIG method, to which flag bits F have been added.

As described above, for a block in which a pixel value can be almost generated by the prediction process, the block code data is only one bit of the flag bit F, so that the coding compression rate is greatly improved.

The encoding method in this case is not a method of directly encoding the image data of all blocks, and is a so-called lossy encoding method in which image data is lost. By appropriately setting a predetermined value used for determining whether or not to perform the encoding process, it is possible to realize an encoding method with little deterioration in image quality and high encoding efficiency. Further, when it is desired to suppress the deterioration of the image quality, a smaller value is used as the predetermined value, and when it is desired to further improve the coding efficiency, a larger value is used as the predetermined value.

FIG. 4 shows an example of encoding processing of one page of binary image data.

First, a predetermined value is set to a constant R used to determine whether or not to encode a block (Process 1).
01). Next, the value of a counter i for counting the number of pixels that do not match the predicted value in each block is initialized to 0 (process 102), and one block of binary image data to be processed is extracted (process 102). Process 103).

Then, a target pixel to be processed is selected from the block to be processed (step 10).
4) The above-described prediction processing is executed by applying the above-described template TMP to the target pixel, and a predicted value corresponding to the target pixel is calculated (process 105).

Next, it is checked whether or not the value of the target pixel and the predicted value do not match (judgment 106). If the result of the judgment 106 is YES, the value of the counter i is increased by one (process 107). .

As described above, when the processing for one pixel is completed, it is checked whether the processing for one block is completed at that time (decision 108), and when the result of the decision 108 is NO, the processing returns to the processing 104. , The processing for the next pixel is executed.

When the processing for one block is completed,
When the result of Step 8 is YES, the value of the counter i is
It is checked whether it is equal to or less than the constant R (judgment 10
9). When the result of the determination 109 is YES, the flag bit F is set as the block code data of the block.
Is set to data "0" to generate 1-bit data, and the block code data is stored in a predetermined code data buffer for temporarily storing one page of encoded data (process 110).

If the result of the determination 109 is NO, the binary image encoding / decoding section 7 applies the JBIG encoding process to the binary image data of the block, and obtains the encoded data And data "1"
Is added to form block code data of this block, and the block code data is stored in the above-described code data buffer (steps 111 and 11).
2).

When the encoding process for one block is completed in this way, it is checked whether the encoding process for all the blocks for one page is completed (decision 11).
3) If the result of determination 113 is NO, process 1
Returning to 02, the encoding process of the next block is executed. When the result of the determination 113 is YES, the code data for one page stored in the code data buffer at that time is output to, for example, the next processing process (processing 114).

In this way, one page of binary image data obtained by the scanner 4 is encoded, and one page of encoded data is formed. In this case, the image quality of the decoded binary image data can be adjusted to some extent by changing the value of the constant R. For example, when the image quality is specified by the user, The value of the constant R may be set to the calculated value. It is desirable that the relationship between the type of image quality and the value set for the constant R be determined by an experiment.

FIG. 5 shows an example of processing for decoding original image data from one page of code data formed by the coding method of FIG.

First, the flag bit F is detected (step 20).
1) It is checked whether the value is 0 (decision 202). When the result of the decision 202 is YES,
Based on the binary image data formed by the previous decoding process and stored in the predetermined image buffer, the prediction process using the above-described template TMP is executed,
One block of binary image data is generated (process 20).
3) The binary image data thus obtained is stored in the data area of the above-described image buffer corresponding to the arrangement area of the block to be processed at that time (process 204).

If the result of the determination 202 is NO, one block of code data following the flag bit F is extracted (step 205), and the extracted code data is converted to a binary image coding / decoding unit. 7 to form original binary image data by decoding in the JBIG system (process 20).
6) The obtained binary image data for one block is stored in the data area corresponding to the arrangement area of the block to be processed at that time in the above-described image buffer data area (processing 207). ).

In this way, when one block of binary image data is formed and stored in the corresponding data area of the image buffer, it is determined whether the processing of one page of code data has been completed at that time. (Decision 208), decision 20
When the result of step 8 is NO, the process returns to step 201,
The decoding process of the code data of the next block is executed. When the result of the determination 208 is YES, the one-page binary image data stored in the image buffer is output to, for example, the next processing process (processing 2).
09).

In the above-described embodiment, the binary image data is directly encoded to form the encoded data. However, a prediction error value can be used as the encoding target. FIG. 6 shows an example of the encoding process of the binary image data for one page in this case.

First, a predetermined value is set to a constant R used to determine whether or not to encode a block (process 3).
01). Next, the value of a counter i for counting the number of pixels that do not match the predicted value in each block is initialized to 0 (step 302), and one block of binary image data to be processed is extracted (step 302). Process 303).

Then, a target pixel to be processed is selected from the block to be processed (step 30).
4) Apply the above-described template TMP to the pixel of interest, execute the above-described prediction processing, and calculate a prediction value corresponding to the pixel of interest (processing 305).

Next, it is checked whether or not the value of the target pixel and the predicted value do not match (judgment 306). If the result of the judgment 306 is YES, the value of the counter i is increased by one (step 307). ), The prediction error value for the pixel is set to data “1” (process 308). If the result of determination 306 is NO, the prediction error value for that pixel is set to data “0” (step 30).
9).

As described above, when the processing for one pixel is completed, it is checked whether the processing for one block is completed at that time (decision 310), and when the result of the decision 310 is NO, the processing returns to the processing 304. , The processing for the next pixel is executed.

When the processing for one block is completed, a decision 31 is made.
When the result of 0 is YES, the value of the counter i is
It is checked whether it is equal to or less than the constant R (decision 31
1). When the result of decision 311 is YES, flag bit F
Is set to data "0", and 1-bit data is generated, and the block code data is stored in a predetermined code data buffer for temporarily storing encoded data for one page (process 312).

If the result of the determination 109 is NO, the binary image encoding / decoding section 7 applies an encoding process according to the JBIG method to the binary image data of the block, and obtains the encoded data And data "1"
To form block code data of this block, and the block code data is stored in the above-described code data buffer (processing 313, 31).
4).

When the encoding process for one block is completed as described above, it is checked whether the encoding process for all the blocks for one page is completed (decision 31).
5) If the result of determination 315 is NO, process 3
Returning to 02, the encoding process of the next block is executed. When the result of the determination 315 is YES, the code data for one page stored in the code data buffer at that time is output to, for example, the next processing process (processing 316).

FIG. 7 shows an example of processing for decoding original image data from one page of code data formed by the coding method of FIG.

First, the flag bit F is detected (Step 40).
1) It is checked whether the value is 0 (decision 402). When the result of the decision 402 is YES,
Based on the binary image data formed by the previous decoding process and stored in the predetermined image buffer, the prediction process using the above-described template TMP is executed,
One block of binary image data is generated (Step 40).
3) The binary image data obtained thereby is stored in the data area of the above-described image buffer corresponding to the arrangement area of the block to be processed at that time (process 404).

When the result of the determination 402 is NO, one block of code data following the flag bit F is extracted (step 405), and the extracted code data is converted to a binary image coding / decoding section. 7 to form the prediction error value data for one block of the original image by decoding in the JBIG system, and store the value of the prediction error value data for the one block in a predetermined buffer (process 406). Next, based on the binary image data formed by the previous decoding process and stored in the predetermined image buffer, the above-described prediction process using the template TMP is executed, and the binary image for one block is executed. Data is generated (process 407). Then, the stored prediction error value data formed in the processing 406 and the prediction image data for one block formed in the processing 407 are combined to form the original binary image data. The binary image data for the block is stored in the data area corresponding to the arrangement area of the block to be processed at that time among the above-described image buffer data areas (processing 408).

In this way, when one block of binary image data is formed and stored in the corresponding data area of the image buffer, it is determined whether or not the processing of one page of code data has been completed at that time. (Determination 409), determination 40
When the result of No. 9 is NO, the process returns to the process 401,
The decoding process of the code data of the next block is executed. If the result of the determination 409 is YES, the one-page binary image data stored in the image buffer is output to, for example, the next processing process (processing 4).
10).

By the way, the code data thus formed is data obtained by coding the prediction error value. Therefore, the original image data cannot be formed without knowing that the code data is a prediction error value, so that the data can be encrypted.

Also, since the correlation between the prediction error values between adjacent pixels is almost nil, in this case, the encoding method for encoding one block of image data is the one of the JBIG system. In some cases, a run-length encoding scheme such as the MH scheme is more suitable. However, in this case, since the MH coding method itself does not have the concept of block coding, when decoding, when the number of pixels of the decoding result reaches 64 pixels for one block, one block is not used. It is necessary to determine the delimiter of the code data for one block by performing the determination of extracting the code data.

In the above-described embodiment, the size of one block is set to the same 8 × 8 size as in the JPEG system, but other sizes can be applied. Further, the template and the prediction processing method applied to the prediction processing are not limited to those described above, and another template or prediction processing method suitable for the present invention can be configured. Further, in the above-described embodiment, the present invention is applied to the group 3 facsimile apparatus, but the present invention can be similarly applied to other data communication terminal apparatuses and data processing apparatuses.

[0057]

As described above, according to the present invention,
If the pixel values in the block can be almost generated by the prediction processing, no code data is created, so that the coding compression efficiency becomes very large. In addition, by encoding the prediction error value in block units, there is an effect that encryption of encoded data can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a group 3 facsimile apparatus according to one embodiment of the present invention.

FIG. 2 is a schematic diagram illustrating an encoding method.

FIG. 3 is a schematic diagram showing an example of block code data.

FIG. 4 is a flowchart illustrating an example of an encoding process.

FIG. 5 is a flowchart showing a processing example when decoding the code data formed in FIG. 4;

FIG. 6 is a flowchart showing another example of the encoding process.

FIG. 7 is a flowchart showing a processing example when decoding the code data formed in FIG. 6;

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H04N 1/41-1/419

Claims (4)

(57) [Claims] 1. A binary image encoding / decoding method for encoding binary image data in block units of a predetermined shape, wherein, at the time of encoding, for each block, a target pixel to be coded is assigned to the target pixel. When a prediction process of predicting the color of the target pixel based on the black and white distribution of a plurality of adjacent peripheral pixels is applied, and when the prediction result of the prediction process does not match a predetermined number of pixels in the same block, the block , A predetermined binary image coding process is applied to form code data of the block, and when there are less than a predetermined number of pixels in the same block where the prediction results of the prediction process do not match, At the time of decoding, without applying the value image encoding process, for a block in which code data is formed, a predetermined binary image Applying the image decoding process to form image data of the block, and executing the above-described prediction process for a block where no code data is formed, and forming image data in the block based on the prediction result. A binary image encoding / decoding method. 2. A binary image encoding / decoding method for encoding binary image data in block units of a predetermined shape, wherein, at the time of encoding, for each block, a target pixel to be encoded is assigned to the target pixel. When a prediction process of predicting the color of the target pixel based on the black and white distribution of a plurality of adjacent peripheral pixels is applied, and when the prediction result of the prediction process does not match a predetermined number of pixels in the same block, the block , A predetermined binary image coding process is applied to form code data of the block, a first identification mark is added to the head of the code data, and pixels for which the prediction results of the above-described prediction process do not match are identical. If there is less than the predetermined number in the block, the binary image coding process is not applied to the block, and the second identification code is used as the code data of the block. When decoding the code data to which the first identification mark is added at the time of decoding, the predetermined binary image decoding process is applied to the code data to form image data of the block. Wherein when the second identification mark is detected, the prediction processing is executed for one block of pixels, and image data in the block is formed based on the prediction result. Method. 3. A binary image encoding / decoding method for encoding binary image data in block units of a predetermined shape, wherein, at the time of encoding, for each block, a target pixel to be encoded is assigned to the target pixel. When a prediction process of predicting the color of the target pixel based on the black and white distribution of a plurality of adjacent peripheral pixels is applied, and when the prediction result of the prediction process does not match a predetermined number of pixels in the same block, the block A predetermined binary image encoding process is applied to the prediction error data obtained for to form code data of the block, and the number of pixels in which the prediction result of the prediction process does not match is less than a predetermined number in the same block. Sometimes, the binary image coding process is not applied to the block, and at the time of decoding, one block is used for a block in which code data is formed. The above-described prediction process is executed for the pixels for the lock, a pixel prediction value for one block is calculated, and a predetermined binary image decoding process is applied to the code data to calculate a prediction error value for the one block. Calculate and combine the pixel prediction value and the prediction error value on a pixel-by-pixel basis to form one block of image data. For blocks where no code data is formed, execute the prediction process, A binary image encoding / decoding method characterized by forming image data in the block according to a result. 4. A binary image encoding / decoding method for encoding binary image data in units of a block having a predetermined shape, wherein at the time of encoding, for each block, a target pixel to be encoded is set to a target pixel. When a prediction process of predicting the color of the target pixel based on the black and white distribution of a plurality of adjacent peripheral pixels is applied, and when the prediction result of the prediction process does not match a predetermined number of pixels in the same block, the block A predetermined binary image encoding process is applied to the prediction error data obtained for the block to form code data for the block, a first identification mark is added to the beginning of the code data, and When the number of unmatched pixels is less than the predetermined number in the same block, the binary image coding process is not applied to the block, When the second identification mark is used as the code data and the code data to which the first identification mark is added is detected at the time of decoding, the above-described prediction processing is executed for the pixels of one block, and the pixels of the one block are decoded. A prediction value is calculated, a predetermined binary image decoding process is applied to the detected code data, a prediction error value for one block is calculated, and the pixel prediction value and the prediction error value are calculated in pixel units. To form image data for one block, and when the second identification mark is detected, the prediction processing is executed for the pixels for one block, and the image data in the block is calculated based on the prediction result. A binary image encoding / decoding method characterized by forming.