JP2005301457A - Image processor, program, and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To achieve an increased processing efficiency when executing image processing to a binarized image by selectively using a bit-map form and a run-length form. <P>SOLUTION: An image processor 1 compares a processing cost in the case of a bit-map form and a processing cost in the case of a run-length form concerning a binarized image to which image processing is to be executed, selects the data form of which processing cost is the lower and executes the image processing to the binarized image with it, and outputs the binarized image that has been processed. An image processing device 2 compares a processing cost in the case of the image processing in which entropy coded binarized image data are decoded to binarized image data of the bit-map form and a processing cost in the case of the image processing in which the entropy coded binarized image data are decoded to binarized image data of the run-length form, decodes the entropy coded binarized image data to the binarized image data of the data form of which processing cost is the lower, executes the image processing, and outputs the binarized image data that have been processed. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a technique for performing image processing on a binary image.

  A binary image typified by a monochrome image is often handled in a bitmap format. For example, if there is a certain difference in the appearance frequency of white pixels and black pixels in a monochrome image, the monochrome image is converted to a run-length format. In addition to being able to reduce the amount of data, image processing such as enlargement, reduction, and rotation can be processed in a shorter time than in the case of the bitmap format. Various proposals have been made for techniques for performing image processing such as enlargement, reduction, and rotation on binary images in the run length format. For example, Patent Documents 1 to 3 disclose techniques relating to enlargement and reduction. Documents 4 and 5 propose techniques relating to rotation.

Japanese Patent Laid-Open No. 05-250468 Japanese Patent Application Laid-Open No. 06-044369 JP 08-33288 A JP-A-01-149185 Japanese Patent Laid-Open No. 02-297679

  For example, if a black-and-white image contains a natural image expressed in hatches or pseudo-halftones, and the area ratio of such a hatch or natural image is large, the number of black runs contained in the black-and-white image is remarkably large. It will increase. When such a monochrome image is handled in the run length format, the data amount of the monochrome image increases as compared with the bitmap format, and a long time is required for image processing such as enlargement, reduction, and rotation.

  For example, assuming that 1% of all pixels in a black-and-white image are black pixels, the average run length of black runs is 4, and the amount of data per black run is 16 bits. When the image data amount is 100, the data amount in the run-length format is only 8. However, assuming that 16% of all pixels in a monochrome image are black pixels, the average run length of black runs is 2, and the amount of data per black run is 16 bits. When the amount is 100, the data amount in the run-length format is 256, and the data amount is significantly increased as compared with the bitmap format. As described above, depending on the image characteristics of the binary image, when it is handled in the run length format, the amount of data increases, and it may take a long time for image processing such as enlargement, reduction, and rotation.

  The present invention has been made in view of the circumstances described above, and its purpose is to increase the processing efficiency when performing image processing on a binary image by properly using a bitmap format and a run-length format. is there.

  In order to solve the above-described problems, the present invention provides a specifying means for specifying a data format of a binary image to be subjected to image processing, and a processing amount when image processing is performed on the binary image. First calculation means for calculating using the information obtained from the binary image for each of the case where the data format of the image is the bitmap format and the case of the run length format, and the first calculation means The selection means for comparing the processing amount in the case of the bitmap format and the run length format and selecting a data format with a small processing amount, and the data format specified by the specifying means are selected by the selection means Conversion means for converting the data format of the binary image to the data format selected by the selection means when the data format differs from the selected data format, and the data format selected by the selection means To provide an image processing apparatus comprising image processing means for performing image processing on the binary image, and output means for outputting the binary image processed by said image processing means.

  According to another aspect of the present invention, there is provided a specifying means for specifying a data format of a binary image to be subjected to image processing, and a processing amount when the image processing is performed on the binary image. For each of the case where the format is a bitmap format and the case where the run length format is used, a calculation means for calculating using information obtained from the binary image, and a case of the bitmap format calculated by the calculation means; The selection means for comparing the processing amount in the case of the run length format and selecting a data format with a small processing amount, and when the data format specified by the specifying means is different from the data format selected by the selection means, Conversion means for converting the data format of the binary image into the data format selected by the selection means; and the binary image of the data format selected by the selection means. It provides image processing means for performing image processing, a program for functioning as an output means for outputting the binary image processed by said image processing means and. The present invention also provides a computer-readable recording medium on which the program is recorded.

  According to the present invention, the image processing apparatus (computer) compares the processing amount in the bitmap format with the processing amount in the run-length format for the binary image subjected to image processing, and the processing amount is small. A data format is selected, image processing is performed on the binary image, and the processed binary image is output.

  Further, the present invention relates to the amount of processing when the entropy-encoded binary image data is decoded and subjected to image processing, when the binary image data is decoded into bitmap-format binary image data. The first calculation means for calculating each of the decoded binary image data in the length format and the amount of processing in the bitmap format and the run length format calculated by the first calculation means are compared. Selection means for selecting a data format with a small amount of processing, decoding means for decoding the entropy-encoded binary image data into binary image data of the data format selected by the selection means, and the decoding Image processing means for performing image processing on the binary image data decoded by the means, and output means for outputting the binary image data processed by the image processing means. To provide an image processing apparatus for Bei.

  According to the present invention, the amount of processing when the computer decodes entropy-encoded binary image data and performs image processing, the binary image data is decoded into bitmap-format binary image data. A calculation means for calculating each of the case and the case of decoding into binary image data in the run length format, and comparing the processing amount calculated in the bitmap format and the run length format calculated by the calculation means, Selecting means for selecting a data format with a small amount; decoding means for decoding the entropy-encoded binary image data into binary image data in the data format selected by the selecting means; and decoding by the decoding means Image processing means for performing image processing on the processed binary image data, and output means for outputting the binary image data processed by the image processing means To provide a program for which to function. The present invention also provides a computer-readable recording medium in which the program is recorded.

  According to the present invention, an image processing apparatus (computer) decodes entropy-encoded binary image data into bitmap-format binary image data and performs image processing, and a run length. Compare the amount of processing when image processing is performed after decoding binary image data in a format, and after decoding into binary image data in a data format with a small amount of processing, the binary after processing is subjected to image processing Output image data.

  According to the present invention, it is possible to increase processing efficiency when image processing is performed on a binary image.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[A. First Embodiment]
FIG. 1 is a block diagram showing main parts of an image processing apparatus 1 according to the first embodiment of the present invention. As shown in FIG. 1, the image processing apparatus 1 includes an image storage unit 11, a run length conversion unit 12, a run length processing unit 13, a bitmap conversion unit 14, a bitmap processing unit 15, and a data format selection unit. 16, an image input / output unit 17, and a control unit 18. The image processing apparatus 1 is a CAD system, a copier, a facsimile, a personal computer, or the like having a function of performing image processing such as enlargement, reduction, rotation, and reversal on a binary image.

  The image storage unit 11 is a memory for storing binary image data to be subjected to pixel image processing and binary image data during or after image processing. The run-length conversion unit 12 converts binary image data in bitmap format into binary image data in run-length format. The run length processing unit 13 performs image processing such as enlargement, reduction, rotation, and reversal on the binary image data in the run length format. The bitmap converter 14 converts the run-length format binary image data into the bitmap format binary image data. The bitmap processing unit 15 performs image processing such as enlargement, reduction, rotation, and reversal on the binary image data in the bitmap format.

  The data format selection unit 16 performs processing cost (processing amount) when image processing is performed in a bitmap format on binary image data subjected to image processing, and processing cost when image processing is performed in a run length format. And select a data format with a low processing cost. The image input / output unit 17 inputs binary image data from the outside of the image processing apparatus 1 and stores the binary image data in the image storage unit 11, or outputs the binary image data after image processing to the outside of the image processing apparatus 1. . The image input / output unit 17 includes a scanner unit for capturing a binary image, a display unit for displaying the binary image after image processing, and the like. The control unit 18 controls the units 11 to 17 of the image processing apparatus 1.

  Next, binary image data in the bitmap format and the run length format will be described. FIG. 2 shows a binary image (13 vertical pixels × 16 horizontal pixels) and 2 in the bitmap format when the pixel value of the white portion of this binary image is 0 and the pixel value of the black portion is 1. It is a figure which shows value image data and binary image data of a run length format. Binary image data in bitmap format is the pixel value of each pixel in units of 8 bits from the 0th pixel to the 7th pixel and from the 8th pixel to the 15th pixel for each scanning line (horizontal direction in the figure). (Binary number) is converted to a hexadecimal number. For example, looking at the scanning line in the second row from the top of the binary image shown in the figure, the pixel values from the 0th pixel to the 7th pixel are “011111011” and the pixels from the 8th pixel to the 15th pixel. The value is “11001100”. When these are converted into hexadecimal numbers, “7b” and “cc” are obtained. In the bitmap format binary image data shown in the figure, “0x” described at the head is an identifier indicating that the subsequent data is a hexadecimal number.

  The run-length binary image data is composed of information indicating the number of black runs included in the scan line, the start position of each black run, and the run length for each scan line. For example, when looking at the second scanning line from the top of the binary image shown in the figure, the number of black runs included in this scanning line is “3”, and the start position of the first black run is “1 (pixels). ”), The run length is“ 4 (for pixels) ”, the start position of the next black run is“ 6 ”, the run length is“ 4 ”, and the start position of the last black run is“ 12 ”. Is “2”. Instead of the number of black runs described for each scan line, for example, (start position, run length) = (0, 0), run data that does not actually exist is changed to the end of the data for each scan line. May be added to indicate the end of the scan line.

  Hereinafter, in this specification, the binary image is a monochrome image, the pixel value of a white pixel is 0, and the pixel value of a black pixel is 1. For the run-length format binary image data, the data structure shown in FIG. 2 is used. Of course, these are merely examples, and for example, the pixel value of a white pixel may be 1 and the pixel value of a black pixel may be 0. Further, the binary image data in the run length format may be composed of information indicating the number of white runs, the start position of each white run, and the run length for each scanning line. Further, instead of the start position and the run length, information indicating the start position and the end position or the start position and the change position of the pixel value may be used.

  FIG. 3 is a block diagram showing the configuration of the data format selection unit 16. As shown in the figure, the data format selection unit 16 includes a run number / run length counting unit 61, a run length conversion cost calculation unit 62, a run length processing cost calculation unit 63, and a bitmap conversion cost calculation unit 64. A bitmap processing cost calculation unit 65 and a processing procedure control unit 66.

  The run number / run length counting unit 61 counts the total number of black runs included in the binary image and the total run length. The run-length conversion cost calculation unit 62 calculates a processing cost when the run-length conversion unit 12 converts binary image data in the bitmap format into binary image data in the run-length format. The run-length processing cost calculation unit 63 calculates a processing cost when the run-length processing unit 13 performs image processing such as enlargement, reduction, rotation, and reversal on binary image data in the run-length format. The bitmap conversion cost calculation unit 64 calculates the processing cost when the bitmap conversion unit 14 converts binary image data in run-length format into binary image data in bitmap format. The bitmap processing cost calculation unit 65 calculates a processing cost when the bitmap processing unit 15 performs image processing such as enlargement, reduction, rotation, and inversion on binary image data in the bitmap format. The processing procedure control unit 66 controls the units 61 to 65 of the data format selection unit 16.

  The processing cost (processing amount) includes the amount of calculation required for data format conversion processing and image processing, the calculation time, the amount of load applied to the calculation processor, and the like.

  FIG. 4 is a flowchart illustrating the operation of the image processing apparatus 1 according to the first embodiment. For example, when the user designates binary image data and the contents of image processing (enlargement, reduction, rotation, inversion, etc.) performed on the binary image data, first, the data format selection unit 16 (Processing procedure control unit 66) determines whether or not the designated binary image data is in a bitmap format (step S1). As a result, when the run length format is used instead of the bitmap format (step S1: NO), the process proceeds to step S8. On the other hand, in the case of the bitmap format (step S1: YES), the data format selection unit 16 performs the designated image processing on the binary image data (bitmap format) 15. Is calculated using the bitmap processing cost calculator 65 (step S2).

Here, the bitmap processing cost calculation unit 65 includes a cost calculation coefficient Pb determined in advance according to the performance of the bitmap processing unit 15 and the image size (horizontal width W, vertical width H) of the binary image to be subjected to image processing. Are used to calculate the bitmap processing cost according to (Equation 1) shown below. Note that the data of the horizontal width W and the vertical width H is added to the binary image data as header information, for example.
Bitmap processing cost = Pb × W × H (Formula 1)

  Note that the bitmap processing cost may be calculated using a cost calculation formula that incorporates factors other than the image size (for example, the total number of pixels of the binary image and the proportion of black pixels). In addition, if it is attempted to calculate an accurate value for the bitmap processing cost, it takes time to calculate the cost itself, thereby reducing the overall performance. For this reason, in the above-described step S2, it is only necessary that the bitmap processing cost value having a certain degree of accuracy can be easily calculated.

  Next, the data format selection unit 16 counts the total number of black runs included in the binary image (bitmap format) subjected to image processing using the run number / run length counting unit 61 (step S3). Here, when counting the total number of black runs included in a binary image (bitmap format), the number of black runs is counted while checking each pixel value in the binary image data bit by bit in accordance with the scanning order. However, this requires a long time to count black runs. Therefore, in the image processing apparatus 1 according to the present embodiment, the total number of black runs can be counted in a short time by using the method described below.

  That is, first, in the memory not shown in the image processing apparatus 1, a total of 256 pixel values from “00000000” to “11111111” are read out from the binary image data in the bitmap format in units of 8 bits according to the scanning order. Tables 80a and 80b (see FIG. 5) indicating the number of black runs included in the 8 bits are stored for each pattern. The tables 80a and 80b are a table 80a in which the last pixel value (hereinafter referred to as “the value of the immediately preceding pixel” in the 8-bit pixel value read out last time) is 0 and 1 And the table 80b. In addition, the “id” item includes values obtained by converting 8-bit pixel values into decimal numbers for a total of 256 patterns from “00000000” to “11111111”. Stores the number of black runs.

  The image processing apparatus 1 reads out pixel values by 8 bits from the binary image data in the bitmap format according to the scanning order. Further, the image processing apparatus 1 determines whether the value of the immediately preceding pixel is 0 or 1 and determines whether to refer to the table 80a or the table 80b. Note that the value of the immediately preceding pixel is determined by an AND operation of “0x01” with the previously read 8-bit pixel value, and if the value of the immediately preceding pixel is 0, the table 80a is referenced. Then, the image processing apparatus 1 converts the read 8-bit pixel value into a decimal number, acquires the number of black runs from the table 80a or the table 80b according to the value of the immediately preceding pixel using the converted value as an argument, By repeating this sequentially, the total number of black runs included in the binary image (bitmap format) subjected to image processing is obtained.

  More specifically, for example, as shown in FIG. 6A, the value of the previous pixel is 0 (white), and the 8-bit data (8-bit pixel value) read this time is “11100100”. If there is, the image processing apparatus 1 uses the value “229” obtained by converting the current 8-bit data “11100100” into a decimal number from the table 80a where the value of the previous pixel is 0 and uses the black run number “2” as an argument. obtain. As is clear from FIG. 6A, the number of black runs in the current 8-bit data is two. Next, as illustrated in FIG. 6B, when the value of the previous pixel is 0 and the 8-bit data read this time is “00111111”, the image processing apparatus 1 sets the value of the previous pixel to 0. From the table 80a, the black run number “0” is obtained with the value “63” obtained by converting the current 8-bit data into a decimal number as an argument. In FIG. 6B, there is one black run in the current 8-bit data, but since this black run continues to the end of the current 8-bit data, it is read next. May be continuous with black runs in 8-bit data. Therefore, in the case shown in FIG. 6B, the image processing apparatus 1 counts the number of black runs in the current 8-bit data as zero.

  Further, as shown in FIG. 6C, when the value of the immediately preceding pixel is 1 (black) and the 8-bit data read this time is “11100100”, the image processing apparatus 1 determines the value of the immediately preceding pixel. The number of black runs “2” is obtained from the table 80b for 1 with the value “229” obtained by converting the current 8-bit data into a decimal number as an argument. In FIG. 6C, since there is one black run following the previous 8-bit data and one other black run in the current 8-bit data, the image processing apparatus 1 The number of black runs in the current 8-bit data is counted as 2.

  On the other hand, as shown in FIG. 6D, when the value of the previous pixel is 1 (black) and the 8-bit data read this time is “00111111”, the image processing apparatus 1 determines the value of the previous pixel. The number of black runs “1” is obtained from the table 80b for 1 with the value “63” obtained by converting the current 8-bit data into a decimal number as an argument. In FIG. 6D, since one black run existing in the current 8-bit data continues to the end of the 8-bit data, there is no black run in the current 8-bit data. However, since the immediately preceding pixel is black (1) and the first pixel in the current 8-bit data is white (0), the black run at the end of the previous 8-bit data is completed. I understand that. For this reason, in the case shown in FIG. 6D, the image processing apparatus 1 counts the number of black runs in the current 8-bit data as 1.

  Note that the immediately preceding pixel is regarded as white (0) because the immediately preceding pixel does not exist for the head portion of each scanning line. For the last 8-bit data of each scanning line, when the last pixel is black (1), the black run count is incremented by one.

  As described above, in step S3 shown in FIG. 4, the run number / run length counting unit 61 makes the number of black runs 8 bits at a time for bitmap image binary image data with reference to the tables 80a and 80b. Are obtained and added together to obtain the total number of black runs included in the binary image (bitmap format). Therefore, compared with the case where the number of black runs is counted while checking each pixel value in the binary image data bit by bit according to the scanning order, the total number of black runs included in the binary image is reduced in a short time. You can count it out.

  Here, the case where the number of black runs is counted in units of 8 bits has been described, but the number of bits per time for counting the number of black runs is not limited to 8 and is arbitrary. The larger the number of bits per time for counting the number of black runs, the greater the number of pixels that can be checked at a time, so the total number of black runs can be counted in a shorter time. In step S3, the run number / run length counting unit 61 may count the number of black runs while checking each pixel value in the binary image data (bitmap format) one bit at a time in accordance with the scanning order. Good.

  The image processing apparatus 1 also has a case where the binary image data input to the image processing apparatus 1 is in the bitmap format, or the binary image data created in the image processing apparatus 1 is in the bitmap format. Alternatively, the total number of black runs included in the binary image may be counted, and the total number of black runs counted may be stored in the image storage unit 11 in association with the binary image data. In this case, in step S3 described above, the run number / run length counting unit 61 reads out the total number of black runs stored in the image storage unit 11 in association with the binary image to be subjected to image processing. The total number of black runs included in the value image can be obtained.

  Now, after counting the total number of black runs in step S3, the data format selection unit 16 then converts the binary image data (bitmap format) subjected to image processing into binary image data in the run length format. The processing cost of the run length conversion unit 12 is calculated using the run length conversion cost calculation unit 62. Further, the data format selection unit 16 calculates the processing cost of the run length processing unit 13 when the specified image processing is performed on the binary image data after being converted into the run length format, as a run length processing cost calculation unit. It calculates using 63 (step S4).

Here, the run length conversion cost calculation unit 62 uses the cost calculation coefficient Tr determined in advance according to the performance of the run length conversion unit 12 and the total number R of black runs obtained in step S3 as follows. The run length conversion cost is calculated according to the following (Equation 2).
Run-length conversion cost = Tr × R (Formula 2)

Note that the run-length conversion cost may be calculated using (Expression 3) shown below instead of (Expression 2). This (Equation 3) is the processing cost required for reading binary image data and run detection (proportional to the total number of pixels of the binary image), and the initial processing for each scan line required for run detection. (Which is proportional to the vertical width H of the binary image), Trs and Trh are cost calculation coefficients determined in advance according to the performance of the run-length conversion unit 12.
Run-length conversion cost = Tr × R + Trs × W × H + Trh × H (Equation 3)

The run length processing cost calculation unit 63 uses the cost calculation coefficient Pr determined in advance according to the performance of the run length processing unit 13 and the total number R of black runs obtained in step S3 as follows. The run length processing cost is calculated from (Equation 4).
Run-length processing cost = Pr × R (Formula 4)

Note that the run-length processing cost may be calculated using (Expression 5) shown below instead of (Expression 4). This (Equation 5) considers the cost of initial processing for each scanning line (proportional to the vertical width H of the binary image), and Prh is determined in advance according to the performance of the run length processing unit 13. This is a cost calculation coefficient.
Run-length processing cost = Pr × R + Prh × H (Formula 5)

  Also, as with bitmap processing costs, if you try to calculate accurate values for run-length conversion costs and run-length processing costs, the cost calculation itself will take time, reducing overall performance. Will end up. For this reason, in the above-described step S4, it is only necessary that the run-length conversion cost and the run-length processing cost with a certain degree of accuracy can be easily calculated.

  Next, the data format selection unit 16 adds the run-length conversion cost calculated in step S4 and the run-length processing cost, and compares the added cost with the bitmap processing cost calculated in step S2 (step S2). S5). As a result, when the bitmap processing cost is smaller (step S5: YES), the data format selection unit 16 reads binary image data (bitmap format) to be subjected to image processing from the image storage unit 11 and performs bitmap processing. To the unit 15. The bitmap processing unit 15 performs designated image processing such as enlargement, reduction, rotation, and reversal on the received binary image data (step S6). Then, the binary image data processed by the bitmap processing unit 15 is displayed or output to the outside of the image processing apparatus 1 for printing, for example, by the image input / output unit 17.

  On the other hand, when the bitmap processing cost is higher (step S5: NO), the data format selection unit 16 reads out binary image data (bitmap format) to be subjected to image processing from the image storage unit 11 and executes a run length conversion unit. 12 to send. The run-length conversion unit 12 converts the received binary image data into binary image data in the run-length format and transmits it to the run-length processing unit 13. The run-length processing unit 13 performs designated image processing on the received binary image data (run-length format) (step S7). Then, the binary image data processed by the run length processing unit 13 is output by the image input / output unit 17.

  On the other hand, when it is determined in step S1 that the binary image data is not in the bitmap format but in the run length format (step S1: NO), first, the data format selection unit 16 performs binary processing for image processing. The total number of black runs included in the image (run length format) and the total run length are counted using the run number / run length counting unit 61 (step S8).

  Here, in the case of the run length format, the total number of black runs included in the binary image is the number of runs described at the beginning of each scan line in the binary image data in the run length format shown in FIG. It can be obtained by adding all of them. As described above, instead of describing the number of runs at the head of each scan line, data such as (0, 0) is added as the end code to the end of the data for each scan line. The total number of black runs can be calculated by subtracting the total number of end codes from the total number of (start position, run length). The total run length can be obtained by adding all the run length values in the case of binary image data in the run length format. Further, when the run data is not (start position, run length) but, for example, (start position, end position), each run length can be obtained from end position-start position + 1. All the run length values may be added.

  Next, the data format selection unit 16 calculates the run-length processing cost as the processing cost of the run-length processing unit 13 when the specified image processing is performed on the binary image data (run-length format) on which image processing is performed. It calculates using the part 63 (step S9). The process in step S9 is the same as the run length process cost calculation process in step S4 described above, and the run length process cost calculation unit 63 uses the above-described (expression 4) or (expression 5) to execute the run length process. Calculate the cost.

  Thereafter, the data format selection unit 16 converts the processing cost of the bitmap conversion unit 14 when the binary image data (run-length format) subjected to image processing into bitmap format binary image data is converted into a bitmap. Calculation is performed using the cost calculation unit 64. Further, the data format selection unit 16 determines the processing cost of the bitmap processing unit 15 when the designated image processing is performed on the binary image data after being converted into the bitmap format, as a bitmap processing cost calculation unit. It calculates using 65 (step S10).

Here, since the processing cost required for conversion from the run-length format to the bitmap format depends on the total number of black runs and the total run length included in the binary image, the bitmap conversion cost calculation unit 64 performs step S8. Using the total number R of black runs R and the sum N of run lengths and cost calculation coefficients Tbr, Tbn determined in advance according to the performance of the bitmap conversion unit 14, the bitmap is expressed by the following (Equation 6): Calculate the conversion cost. Further, the bitmap processing cost is calculated by the above-described (Equation 1).
Bitmap conversion cost = Tbr × R + Tbn × N (Expression 6)

  Of course, it is only necessary that the bitmap conversion cost and the bitmap processing cost with a certain degree of accuracy can be easily calculated in step S10. Further, since the total length N of run lengths matches the total number of black pixels included in the binary image, the total number of black pixels included in the binary image may be used instead of the total length N of run lengths.

  Next, the data format selection unit 16 adds the bitmap conversion cost calculated in Step S10 and the bitmap processing cost, and compares the added cost with the run length processing cost calculated in Step S9 (Step S9). S11). As a result, when the run-length processing cost is smaller (step S11: YES), the data format selection unit 16 reads binary image data (run-length format) to be subjected to image processing from the image storage unit 11 and executes the run-length processing. To the unit 13. The run length processing unit 13 performs designated image processing such as enlargement, reduction, rotation, and reversal on the received binary image data (step S12). Then, the binary image data processed by the run length processing unit 13 is output by the image input / output unit 17.

  On the other hand, when the run-length processing cost is higher (step S11: NO), the data format selection unit 16 reads binary image data (run-length format) to be subjected to image processing from the image storage unit 11 and performs a bitmap conversion unit. 14 to send. The bitmap conversion unit 14 converts the received binary image data into binary image data in the bitmap format, and transmits it to the bitmap processing unit 15. The bitmap processing unit 15 performs designated image processing on the received binary image data (bitmap format) (step S13). Then, the binary image data processed by the bitmap processing unit 15 is output by the image input / output unit 17.

  As described above, according to the present embodiment, the image processing apparatus 1 compares the processing cost in the case of the bitmap format with the processing cost in the case of the run length format for the binary image subjected to image processing, A data format with a low processing cost is selected, image processing is performed on the binary image, and the processed binary image is output. Therefore, by using the bitmap format and the run length format properly, it is possible to increase the processing efficiency when image processing is performed on the binary image.

[B. Second Embodiment]
Next, a second embodiment will be described. In addition, in this embodiment, the same code | symbol is used for the part which is common in 1st Embodiment, and the description is abbreviate | omitted.

  FIG. 7 is a block diagram showing a main part of the image processing apparatus 2 according to the second embodiment of the present invention. The image processing apparatus 2 shown in the figure is different from the image processing apparatus 1 described in the first embodiment in that a run length decoding unit 19 is used instead of the run length converting unit 12 and a bit map is used instead of the bitmap converting unit 14. The decoding unit 20 includes a data format selection unit 21 instead of the data format selection unit 16. In this embodiment, the binary image data is encoded using an entropy encoding method such as an MH (Modified Huffman) method or an LZ (Lampel-Ziv) method. The run-length decoding unit 19 decodes entropy-encoded binary image data into run-length format binary image data. In addition, the bitmap decoding unit 20 decodes the entropy-encoded binary image data into bitmap image binary image data.

  As shown in FIG. 8, the data format selection unit 21 according to the present embodiment is different from the data format selection unit 16 described in the first embodiment in that the run number estimation is performed instead of the run number / run length counting unit 61. The unit 67 includes a run length decoding cost calculation unit 68 instead of the run length conversion cost calculation unit 62 and a bitmap decoding cost calculation unit 69 instead of the bitmap conversion cost calculation unit 64.

  The run number estimation unit 67 uses the data size (data amount) after decoding the entropy-encoded binary image data, for example, information indicating the image size given as header information (horizontal width W, vertical width H). The binary image data is calculated without decoding, and the compression rate of the binary image data is calculated from the value and the data size of the entropy-encoded binary image data. In addition, the run number estimation unit 67 estimates the total number of black runs included in the binary image from the calculated compression rate. Here, in entropy coding, generally, the more drastic changes in pixel values in a binary image are, that is, the greater the number of black runs included in a binary image and the shorter the average run length for black runs. , The compression rate decreases. Therefore, for example, as shown in FIG. 9, a table 81 showing the relationship between the compression rate and the ratio of the total number of black runs to the total number of pixels of the binary image is stored in the memory (not shown) of the image processing apparatus 2. The run number estimation unit 67 refers to the table 81 and estimates the total number of black runs included in the binary image from the compression rate.

  For example, when the compression rate is 0.45 and the total number of pixels of the binary image is 1000, the run number estimation unit 67 refers to the table 81 shown in FIG. 9 and is included in this binary image. Estimate the total number of black runs to 20. The table 81 is created based on a statistical value regarding the compression ratio and the ratio of the total number of black runs to the total number of pixels, which is obtained by investigating a large number of binary images (entropy coding and bitmap format) in advance. Has been.

The run-length decoding cost calculation unit 68 calculates the processing cost of the run-length decoding unit 19 when entropy-encoded binary image data is decoded into run-length format binary image data. The run-length decoding cost calculation unit 68 calculates a decoding cost coefficient Dr per run determined in advance according to the performance of the run-length decoding unit 19 and the total number R of black runs obtained by the run number estimation unit 67. Then, the run-length decoding cost is calculated by (Equation 7) shown below.
Run-length decoding cost = Dr × R (Expression 7)

In consideration of the code data reading process, the run-length decoding cost is calculated by (Expression 8) shown below instead of (Expression 7) using the encoded data amount D and the read cost coefficient Dd per unit data amount. It may be calculated.
Run-length decoding cost = Dr × R + Dd × D (Equation 8)

  Also, as with the run-length conversion cost, if an accurate value is calculated for the run-length decoding cost, it will take time to calculate the cost itself, and the overall performance will be reduced. It is only necessary that the run-length decoding cost with a certain degree of accuracy can be easily calculated.

On the other hand, the bitmap decoding cost calculation unit 69 calculates the processing cost of the bitmap decoding unit 20 when the entropy-encoded binary image data is decoded into the binary image data in the bitmap format. The bitmap decoding cost calculation unit 69 uses a decoding cost coefficient Db per run determined in advance according to the performance of the bitmap decoding unit 20 and the total number R of black runs obtained by the run number estimation unit 67. Then, the bitmap decoding cost is calculated by (Equation 9) shown below. Of course, as shown in (Equation 8) described above, the bitmap decoding cost may be calculated in consideration of the code data reading process and the like.
Bitmap decoding cost = Db × R (Equation 9)

  The bitmap decoding cost is considered to be related to the total run length (= total number of black pixels included in the binary image). However, in order to obtain the sum of the run lengths from the entropy-encoded binary image data, the processing cost almost equal to the processing for decoding into the binary image data in the bitmap format is often required. Assuming that the average run length is substantially constant in equation (9), the processing cost corresponding to the coefficient Db is incorporated, or a function depending on the compression rate is set and added to (equation 9).

  In the image processing apparatus 2 according to the present embodiment, for example, the user performs binary image data (entropy coding) and contents of image processing (enlargement, reduction, rotation) performed on the binary image data. First, the data format selection unit 21 calculates a run length decoding cost, a run length processing cost, a bitmap decoding cost, and a bitmap processing cost. Next, the data format selection unit 21 compares the run length decoding cost + run length processing cost with the bitmap decoding cost + bitmap processing cost, and selects a data format with a low processing cost. As a result, it is determined whether the image processing is performed by decoding the binary image data in the run length format or the image processing is performed by decoding the binary image data in the bitmap format. Thereafter, binary image data (entropy coding) is decoded into binary image data in the data format selected by the data format selection unit 21 and subjected to designated image processing. Then, the binary image data after the image processing is output by the image input / output unit 17.

  As described above, according to the present embodiment, the image processing apparatus 2 processes the entropy-encoded binary image data by decoding the binary image data in the bitmap format and performing image processing. Is compared with the processing cost when the image processing is performed by decoding the binary image data in the run length format, and the image processing is performed after the decoding to the binary image data in the data format having a low processing cost. The later binary image data is output. Therefore, by using the bitmap format and the run length format properly, it is possible to increase the processing efficiency when image processing is performed on the binary image.

[C. Modified example]
(1) When the data format of the binary image after the image processing is designated as the bitmap format or the run length format, the image processing is performed in consideration of the conversion cost to the designated data format. The data format may be selected. For example, when the binary image data before image processing is in the bitmap format and the data format after image processing is also designated as the bitmap format, the image processing apparatus 1 according to the first embodiment The bitmap processing cost is compared with the run length conversion cost + run length processing cost + bitmap conversion cost, and a data format with a low processing cost is selected as a data format for performing image processing. In addition, when the binary image data before image processing is in the bitmap format and the run length format is designated as the data format after the image processing, the image processing apparatus 1 performs the bitmap processing cost + run The length conversion cost is compared with the run length conversion cost + run length processing cost, and a data format for performing image processing is selected.

(2) When the value of the bitmap processing cost or the run length processing cost changes according to the type of image processing such as enlargement, reduction, rotation, inversion, etc., for example, a coefficient prepared for each type of image processing is used as the cost. A configuration may be adopted in which the bitmap processing cost and the run length processing cost are calculated by being incorporated in the calculation formula. For example, when enlargement is specified as the type of image processing, the total number R of black runs in (Equation 2) described above is multiplied by the enlargement magnification Z, and each of the horizontal width W and vertical width H indicating the image size is multiplied. What is necessary is just to multiply the enlargement magnification Z.

(3) When decoding entropy-encoded binary image data into binary image data in bitmap format, first, it is decoded into binary image data in run-length format, and then this is converted into a binary image in bitmap format. A method of converting to data is often adopted. Therefore, as shown in FIG. 10, a bitmap conversion unit 14 may be provided instead of the bitmap decoding unit 20, and the calculation of the bitmap decoding cost may also be calculated as the run length decoding cost + the bitmap conversion cost.

(4) The image processing apparatuses 1 to 3 (computers) execute the processing according to the present invention (for example, the processing shown in FIG. 4) according to a program installed in a memory (not shown). May be provided to the image processing apparatuses 1 to 3 by communication. Alternatively, such a program may be provided by being recorded on a recording medium.

1 is a block diagram illustrating a main part of an image processing apparatus 1 according to a first embodiment of the present invention. 4 is a diagram for explaining binary image data in a bitmap format and a run-length format according to the embodiment. FIG. 4 is a block diagram showing a configuration of a data format selection unit 16 according to the same embodiment. FIG. 4 is a flowchart illustrating an operation of the image processing apparatus 1 according to the embodiment. It is a figure which concerns on the same embodiment and shows the data structure of tables 80a and 80b. It is a figure for demonstrating the operation | movement which counts a black run concerning the embodiment. It is a block diagram which shows the principal part of the image processing apparatus 2 which concerns on 2nd Embodiment. 4 is a block diagram illustrating a configuration of a data format selection unit 21 according to the embodiment. FIG. It is a figure which shows the data structure of the table 81 which showed the relationship between the compression rate and the ratio of the total number of black runs with respect to the total number of pixels of a binary image concerning the embodiment. It is a block diagram which shows the principal part of the image processing apparatus 3 which concerns on a modification (3).

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1, 2, 3 ... Image processing apparatus, 11 ... Image storage part, 12 ... Run length conversion part, 13 ... Run length processing part, 14 ... Bit map conversion part, 15 ... Bit map processing part, 16, 21 ... Data format Selection unit, 17 ... Image input / output unit, 18 ... Control unit, 19 ... Run length decoding unit, 20 ... Bitmap decoding unit, 61 ... Run number / run length counting unit, 62 ... Run length conversion cost calculation unit, 63 ... Run length processing cost calculation unit, 64 ... Bitmap conversion cost calculation unit, 65 ... Bitmap processing cost calculation unit, 66 ... Processing procedure control unit, 67 ... Run number estimation unit, 68 ... Run length decoding cost calculation unit, 69 ... Bitmap decoding cost calculation unit, 80a, 80b, 81... Table.

Claims (13)

Specifying means for specifying the data format of a binary image to be subjected to image processing;
The amount of processing when image processing is performed on the binary image is obtained from the binary image when the data format of the binary image is a bitmap format and when it is a run-length format. First calculating means for calculating using information;
Selection means for comparing the processing amount calculated by the first calculation means in the case of the bitmap format and in the case of the run length format, and selecting a data format having a small processing amount;
Conversion means for converting the data format of the binary image into the data format selected by the selection means when the data format specified by the specification means is different from the data format selected by the selection means;
Image processing means for performing image processing on the binary image in the data format selected by the selection means;
An image processing apparatus comprising: output means for outputting a binary image processed by the image processing means. Of the bitmap format and the run length format, the processing amount when the data format of the binary image is converted to a data format different from the data format specified by the specifying means is obtained using information obtained from the binary image. A second calculating means for calculating
The first calculation means uses the second calculation means for the processing amount of the data format different from the data format specified by the specifying means among the processing amounts in the case of the bitmap format and the run length format. The image processing apparatus according to claim 1, wherein the processing amount calculated by the step is added. The first calculation means sets the processing amount when image processing is performed on the binary image, and the image size of the binary image when the data format of the binary image is a bitmap format. The image processing apparatus according to claim 1, wherein the calculation is performed using The first calculation means includes a processing amount when image processing is performed on the binary image, and a case where the data format of the binary image is a run-length format. The image processing apparatus according to claim 1, wherein the calculation is performed using the number of runs for one of the pixel values. The second calculation means uses the number of runs for any one of the pixel values included in the binary image to calculate the processing amount when converting the data format of the binary image to a run-length format. The image processing apparatus according to claim 2, wherein the image processing apparatus is calculated. The second calculation means determines the amount of processing when the data format of the binary image is converted to the bitmap format, the number of runs for any one of the pixel values included in the binary image, and the pixel The image processing apparatus according to claim 2, wherein calculation is performed using the number of pixels having a value. One of the pixel values included in the binary image when the data format of the binary image input to the image processing device or the binary image created in the image processing device is a bitmap format Counting means for counting the number of runs for
Storage means for storing the number of runs counted by the counting means in association with the binary image;
The first calculating means stores the processing amount when image processing is performed on the binary image, and is stored in the storage means when the data format of the binary image is a run-length format. The image processing apparatus according to claim 4, wherein the calculation is performed using the number of runs. Reading means for reading out pixel values n bits (n is an integer of 2 or more) from the binary image data in the bitmap format according to the scanning order;
Discrimination means for discriminating a pixel value immediately before the n-bit pixel value read by the reading means;
Storage means for storing, for each n-bit pixel value, data having two types of run numbers for any one of the pixel values included in the n-bit pixel value according to the immediately preceding pixel value;
Included in the binary image data based on the n-bit pixel value read by the reading means, the previous pixel value determined by the determining means, and the data stored in the storage means A counting means for counting the number of runs;
The first calculating means determines the processing amount when the binary image is subjected to image processing when the specifying means specifies the bitmap format, and the data format of the binary image. The image processing apparatus according to claim 4, wherein the run length format is calculated using the number of runs counted by the counting unit. The amount of processing when the entropy-encoded binary image data is decoded and subjected to image processing is the same as when the binary image data is decoded into the binary image data in the bitmap format and the binary image in the run length format. First calculating means for calculating each of the data when decrypted;
Selection means for comparing the processing amount calculated by the first calculation means in the case of the bitmap format and in the case of the run length format, and selecting a data format having a small processing amount;
Decoding means for decoding the entropy-encoded binary image data into binary image data in the data format selected by the selection means;
Image processing means for performing image processing on the binary image data decoded by the decoding means;
An image processing apparatus comprising: output means for outputting binary image data processed by the image processing means. A specifying means for specifying a data amount after decoding entropy-encoded binary image data;
Second calculating means for calculating a compression rate of the binary image data using the data amount of the entropy-encoded binary image data and the data amount specified by the specifying means;
Storage means for storing information indicating the number of runs for one of the pixel values included in the binary image for each compression rate;
The first calculation means calculates the amount of processing when decoding into binary image data in bitmap format and when decoding into binary image data in run length format by the second calculation means. The image processing apparatus according to claim 9, wherein calculation is performed using a number of runs specified based on a compression rate and information stored in the storage unit. Computer
Specifying means for specifying the data format of a binary image to be subjected to image processing;
The amount of processing when image processing is performed on the binary image is obtained from the binary image when the data format of the binary image is a bitmap format and when it is a run-length format. A calculation means for calculating using information;
Selection means for comparing the processing amount calculated in the case of the bitmap format and the run length format calculated by the calculation means, and selecting a data format with a small processing amount;
Conversion means for converting the data format of the binary image into the data format selected by the selection means when the data format specified by the specification means is different from the data format selected by the selection means;
Image processing means for performing image processing on the binary image in the data format selected by the selection means;
A program for functioning as output means for outputting a binary image processed by the image processing means. Computer
The amount of processing when the entropy-encoded binary image data is decoded and subjected to image processing is the same as when the binary image data is decoded into the binary image data in the bitmap format and the binary image in the run length format. A calculation means for calculating each of the decrypted data,
Selection means for comparing the processing amount calculated in the case of the bitmap format and the run length format calculated by the calculation means, and selecting a data format with a small processing amount;
Decoding means for decoding the entropy-encoded binary image data into binary image data in the data format selected by the selection means;
Image processing means for performing image processing on the binary image data decoded by the decoding means;
A program for functioning as output means for outputting binary image data processed by the image processing means.   The computer-readable recording medium which recorded the program of Claim 11 or 12.

Images