JP5293482B2 - Image processing apparatus, image processing method, and information recording medium - Google Patents

Description

  The present invention relates to an image processing device, an image processing method, and an information recording medium.

  Conventionally, there is an image processing technique for reducing a binary image. For example, when the resolution of an image handled by a printer, scanner, copier, or the like is high, but a high-resolution image is not required, a reduced image is generated. Further, for example, when a high-resolution image is browsed by an image viewer or the like, a reduced image such as a thumbnail or an image that fits on the screen of the display device is generated.

  As a technique for reducing a binary image, for example, in Japanese Patent Laid-Open No. 7-123246 (Patent Document 1), when a binary image is reduced, it is converted into one multivalued pixel for each block according to the reduction ratio. A binary image reduction device that binarizes the multi-valued pixel is disclosed. In the binary image reduction device disclosed in Patent Document 1, when binarizing a multi-value pixel, an exception is made depending on the arrangement of the pixels of the binary image data in the block that is the basis of the multi-value pixel. Processing is provided.

  Japanese Patent Application Laid-Open No. 2002-354233 (Patent Document 2) generates a multi-valued image once reduced when a binary image is reduced, and uses dither or error diffusion to express a halftone therefrom. An invention such as an image reduction device that performs binarization is disclosed.

  Incidentally, some binary images include many tables, lines, and the like, such as forms. Forms and the like often do not contain many figures and photos. When such a binary image of a form is reduced, fine lines may be cut or blurred, and if OR compression is performed to avoid this, black may increase excessively.

  However, in the invention such as the image reduction device described in Patent Document 2, since binarization is performed by dithering, error diffusion, or the like, if processing such as thin lines is performed, the structure of the table or the like in the image is collapsed. May be. Further, in the invention of the binary image reduction device described in Patent Document 1, an exception process is provided according to the arrangement of the pixels for each block of the binary image data, so the arrangement of the pixels must be verified, Processing becomes complicated.

  The present invention has been invented to solve these problems in view of the above points, and a reduced image that maintains the shape of an image element in an original image of a binary image as much as possible with a simple configuration. An object is to provide an image processing apparatus, an image processing method, and an information recording medium.

In order to achieve the above object, the image forming apparatus of the present invention employs the following configuration. Includes a block dividing unit for dividing the binary image into blocks, each said block, and it converts the multi-level pixel, a reduced image generation unit for generating a reduced image, for each of the multi-level pixel, the multivalued pixels An average density of multi-valued pixels in a predetermined range is calculated, a black pixel ratio storage threshold value indicating a value obtained by converting the minimum value into a density is set as the minimum value of the number of black pixels with the number of pixels equal to or higher than the average of the density A black pixel ratio storage threshold determination unit to determine, a density threshold determination unit to determine a threshold of density at which one multi-value pixel becomes a white pixel,
On the basis of the black pixel ratio storing threshold and the threshold value of the density, have a, a binarization unit for binarizing the pixels in which the reduced image has, the binarization unit, the concentration of the multi-level pixel Is greater than or equal to the black pixel ratio storage threshold corresponding to the multi-value pixel, and the density of the multi-value pixel is smaller than the black pixel ratio storage threshold corresponding to the multi-value pixel, When the density is larger than the average density of the pixels around the multi-value pixel and the density of the multi-value pixel is equal to or higher than the density threshold, the multi-value pixel can be configured as a black pixel. .

  Accordingly, it is possible to provide an image processing apparatus that generates a reduced image that maintains the shape of an image element in an original image of a binary image as much as possible with a simple configuration.

  In order to solve the above problems, the present invention may be an information recording medium storing an image processing method in the image processing apparatus and a program for causing a computer to execute the image processing method.

  According to the image processing device, the image processing method, and the information recording medium of the present invention, an image processing device that generates a reduced image that maintains the shape of an image element in an original image of a binary image as much as possible with a simple configuration, An image processing method and an information recording medium can be provided.

FIG. 1 is a diagram illustrating an example of a functional configuration of the image processing apparatus according to the present embodiment. FIG. 2 is a diagram illustrating an example of divided blocks. FIG. 3 is a diagram illustrating an example of a black pixel included in a block serving as one black pixel. FIG. 4 is a flowchart showing the image processing method (part 1) according to the present embodiment. FIG. 5 is a flowchart showing the image processing method (part 2) according to the present embodiment. FIG. 6 is a block diagram showing the configuration of the image processing apparatus according to the present embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following embodiment, an example will be described in which the density of pixels of a multi-valued image is in a range from 0 to 255, 0 is white, and 255 is black. In the binary image, 0 represents a white pixel and 1 represents a black pixel.

Embodiment of the present invention
(Example of functional configuration of image processing apparatus according to this embodiment)
FIG. 1 is a diagram illustrating an example of a functional configuration of the image processing apparatus according to the present embodiment. 1 includes a block dividing unit 10, a reduced image generating unit 20, a black pixel ratio storage threshold determining unit 30, a density threshold determining unit 40, a binarizing unit 50, an image memory 81, and a work memory 82. Are connected by a bus.

  The block dividing unit 10 divides the binary image into blocks corresponding to the reduction ratio. The reduced image generation unit 20 generates a multi-value reduced image by generating one multi-value pixel for each block. The black pixel ratio storage threshold determination unit 30 determines a black pixel ratio storage threshold that is a threshold used when binarizing a multi-valued reduced image. The black pixel ratio storage threshold is obtained by converting the number of black pixels included in the predetermined range to the pixel density of the multi-valued image when expressing pixels in a predetermined range of the multi-valued reduced image in binary. Value.

  The density threshold value determination unit 40 determines a density threshold value of a multi-value pixel that becomes a white pixel when converting a pixel included in the multi-value reduced image into a binary value. The density threshold is determined based on the reduction ratio.

  Based on the black pixel ratio storage threshold determined by the black pixel ratio storage threshold determination unit 30 and the density threshold determined by the density threshold determination unit 40, the binarization unit 50 For each pixel, the density is converted to binary.

  The image memory 81 stores input binary image image data. The work memory 82 stores temporary data when the block division unit 10, the reduced image generation unit 20, the black pixel ratio storage threshold value determination unit 30, the density threshold value determination unit 40, and the binarization unit 50 perform processing. Store. The image memory 81 and the work memory 82 may be configured as one memory.

  Details of the processes of the block dividing unit 10, the reduced image generating unit 20, the black pixel ratio storage threshold determining unit 30, the density threshold determining unit 40, and the binarizing unit 50 will be described below.

(Processing of block dividing unit 10)
The block dividing unit 10 divides the original binary image into blocks. FIG. 2 is a diagram illustrating an example in which the block dividing unit 10 divides an original binary image into blocks corresponding to a reduction ratio. FIG. 2 shows an example in which the reduction ratios in the vertical direction and the horizontal direction are 40%, respectively. When the reduction ratio is 40%, one pixel of the reduced binary image corresponds to a block of 2.5 pixels vertically and 2.5 pixels horizontally. The block b1 includes four pixels A, B, F, and G. The block b1 further includes four pixels C, H, K, and L, each of which is ½ pixel. Further, the pixel M is included in 1/4 pixel.

(Processing of the reduced image generation unit 20)
The reduced image generation unit 20 calculates the average density of the included pixels for each block divided by the block division unit 10. For example, when the reduction ratio is 50% in both vertical and horizontal directions, 2 × 2 pixels form one block. Therefore, the number of black pixels in 2 × 2 pixels is counted to obtain one 4-level multi-value pixel. In the case of the block shown in FIG. 2, the area average of the pixel values in the block is obtained by setting black to 255 and white to 0. Thereby, one value is obtained for each block.

(Processing of the black pixel ratio storage threshold value determination unit 30)
The black pixel ratio storage threshold value determination unit 30 calculates a threshold value for storing the black pixel ratio storage threshold value. The black pixel ratio storage threshold value determination unit 30 calculates an average density α of pixels in a predetermined range for each pixel of the reduced image generated by the reduced image generation unit 20 according to the following equation (1). The predetermined range is, for example, a range including 8 pixels around the target pixel, pixels of several lines above and below the target pixel, or the entire reduced image. Note that the predetermined range may be determined according to the reduction ratio.

Next, the black pixel ratio storage threshold value determination unit 30 determines the number of pixels Y that is the average density α or the number of pixels exceeding the average density α when the Y pixel is black in the X pixel, It calculates by following Formula (2).

  Since the partial area (X) actually contains pixels having pixel values of 0 to 255, this value Y is a virtual value. If the value Y is only white and black pixels in the partial area (X) and there are Y black pixels in total, the average density of the partial area (X) is the average density of the multivalued pixels. Means the same or slightly darker. That is, the value Y is the minimum value of the number of black pixels.

Next, the black pixel ratio storage threshold value determination unit 30 sets the minimum value Y of the number of black pixels included in the partial region X based on the density histogram of the density C (k) of the surrounding partial region X pixels including the target pixel. Determine the corresponding concentration. The following expression (3) is an expression representing the average density α by the density histogram hist (n), and the following expression (4) is an expression representing a value N obtained by converting the minimum value of the number of black pixels into a density. .

  The range of the value N is equal to the concentration range, and in this embodiment, 0 ≦ N ≦ 255. This value N is referred to as “black pixel ratio storage threshold”. When binarizing the partial area (X), by binarizing the density N as a threshold value with black as the density N and the density value less than N as white, (X) of the partial area after binarization Is substantially equal to the average density of the partial area (X) of the multi-valued image.

  Note that the black pixel ratio storage threshold value determination unit 30 may determine the range in which the average density α is acquired based on the accuracy and the capacity of the used memory. For example, when the average density α is acquired, the entire image is used, so that it becomes a stable parameter for the entire image, and the accuracy is higher when determining the black pixel ratio storage threshold.

  When saving the capacity of the memory used for processing or when realizing the image processing apparatus according to the present embodiment with a small memory, it is preferable to determine the range for obtaining the average density α in consideration of the processing efficiency. . For example, pixel values for a predetermined number of lines of the reduced multi-valued image may be held, and the average density may be acquired from the data. For example, the number of lines may be 3 using the front and back of the target pixel. As a result, in the binarization process described later, the process in the vertical direction is the same as the determination process of using the surrounding 8 pixels (rule 2), and the range in the vertical direction is the same. it can.

(Processing of density threshold value determination unit 40)
The density threshold value determination unit 40 determines a density threshold value, which is another threshold value used for binarization. The density threshold is a value at which one multi-value pixel becomes a white pixel. The density threshold M is obtained by the following equation (5).
M = ((number of gradations) / 2) × (magnification) (5)
However, (magnification ratio) employs a magnification ratio having a larger value among the vertical magnification ratio and the horizontal magnification ratio. Note that the value of the scaling factor at the time of reduction is smaller when the image after reduction is smaller.

For example, when the reduction ratio W is 40%, the density threshold M is
M = 128 × 0.4 = 51
It becomes. In the above formula, the calculation result is rounded off.

Equation (5) is an empirically determined value. More specifically, for example, there are the following backgrounds (a) to (c).
(A) If the magnification ratio is close to 100%, the density threshold M = 128 is determined, and monochrome is determined in a majority manner.
(B) When the scaling factor is 50%, the density threshold M = 64. That is, in the case of 50% reduction, 2 × 2 pixels of the original image correspond to one pixel of the result image, and the density threshold M = 64≈255 / 4 is “a black pixel in the 2 × 2 pixel region of the original image” If there is one, the result image is black.
(C) If the reduced image is even smaller, for example 1 / N, the line is stored as “the result image is black if there is one line of length N / 2 in the N × N region of the original image”. It becomes a regular decision rule. In FIG. 3, 2 pixels out of 4 × 4 pixels are black pixels. Therefore, the example of FIG. 3 shows the threshold value of the lowest density at which the pixels of the image after reduction based on this region become black pixels.

(Processing of binarization unit 50)
The binarization unit 50 performs binarization using the black pixel ratio storage threshold N and the density threshold M. The rules when the binarization unit 50 binarizes multi-valued pixels are the following (Rule 1) and (Rule 2).

(Rule 1) The density of the pixel of interest in the multi-valued image is C, and if C ≧ N, the pixel is a black pixel. Otherwise go to (2).
(Rule 2) The density C of the target pixel is larger than the average density of the eight adjacent points, that is,
C × 8> concentration sum of adjacent 8 points holds, and
C ≧ M
If so, the pixel is a black pixel. Otherwise, the pixel is a white pixel.

  According to (Rule 2), the case where the density threshold value M is applied to form a black pixel is limited to the case where the point of interest is darker than the surrounding average density. As a result, it is possible to suppress deterioration that crushes a bright void area compared to the surroundings and deterioration that excessively increases the line width.

  FIG. 4 is a flowchart showing an image processing method according to the present embodiment. In step S101 in FIG. 4, the block dividing unit 10 divides the input binary image into blocks corresponding to the reduction ratio. Progressing to step S102 following step S101, the reduced image generation unit 20 acquires the value of one multi-value pixel for each block divided in step S101. Thereby, a reduced multi-valued image is generated.

  Progressing to step S103 following step S102, the black pixel ratio storage threshold value determination unit 30 determines the black pixel ratio storage threshold value N. Proceeding to step S104 following step S103, the density threshold value determination unit 40 determines the minimum density threshold value M. Progressing to step S105 following step S104, the binarization unit 50 binarizes each pixel of the reduced multi-valued image.

  In the processing of FIG. 4, the minimum density threshold value M is determined by the reduction ratio, and therefore may not be in the order shown in FIG. 4. For example, it is possible to use the value in the binarization of other pixels by obtaining the value at the beginning of the process and holding it.

  FIG. 5 is a flowchart showing processing for generating a multi-valued image reduced for each line. In step S201 in FIG. 5, the block dividing unit 10 divides the input binary image into blocks corresponding to the reduction ratio. In step S202, the density threshold value determination unit 40 determines the lowest density threshold value M based on the reduction ratio. This value is held in the work memory 82, for example.

  Progressing to step S203 following step S202, the block dividing unit 10 generates a block corresponding to three lines of the input binary image, and the reduced image generating unit 20 reduces the multi-value image reduced by the divided blocks. Is generated. For the upper and lower lines of the image, a known technique such as a process of supplementing virtual pixels by a filter process or the like may be used. More specifically, the processing may be performed assuming that a line having the same pixel as the first line of the image is immediately above the first line.

  Progressing to step S204 following step S203, the black pixel ratio storage threshold value determination unit 30 calculates a black pixel ratio storage threshold value. This threshold is the black pixel ratio storage threshold N. Progressing to step S205 following step S204, the multi-valued image corresponding to the three-line block obtained in step S203 is binarized.

  Progressing to step S206 following step S205, it is determined whether or not the target line is the last multi-value line. In the case of the last line, the process proceeds to step S209 and the process ends. If it is not the last line, the process proceeds to step S207. In step S207 following step S206, the target line to be processed is shifted by one.

  Progressing to step S208 following step S207, in the new target line, the block dividing unit 10 divides the block, and the reduced image generating unit 20 generates a reduced multi-value image. After the process of step S208, the process returns to step S204 to repeat the process.

  FIG. 6 is a block diagram showing the configuration of the image processing apparatus according to the present embodiment. The image processing apparatus in FIG. 6 can execute the image processing method according to the present embodiment by digital processing such as a microprocessor.

  The image processing apparatus in FIG. 6 includes a CPU 1, a memory 2, a work area RAM 3, a program storage ROM / RAM 4, a scanner 5, a printing device 6, a display 7, and a CD-ROM / FD drive 8.

  The CPU 1 reads and executes a program stored in the program storage ROM / RAM 4 or a program read by the CD-ROM / FD drive 8 from a recording medium such as a CD-ROM / FD, and uses the work area RAM 3. Image processing including block division processing for document image data is executed.

  The CPU 1 stores the processed image data in the memory 2. The CPU 1 also displays the image data on the display 7 or causes the printing device 6 to output it.

  The memory 2 stores image data. The work area RAM 3 is provided when the CPU 1 executes a program. The program storage ROM / RAM 4 stores a program executed by the CPU 1.

  The scanner 5 optically reads a document image original and converts it into image data that is an electrical signal. In the printing device 6, image data is printed by the CPU 1. The display 7 displays information including the image data 7. In the CD-ROM / FD drive 8, a recording medium such as a CD-ROM or a floppy (registered trademark) disk is inserted, and information stored in the recording medium is read out.

  According to this embodiment, thin ruled lines are difficult to fade even when reduced at high magnification. More specifically, when it is desired to obtain a binary image by reducing the binary image, even if the reduction is performed at a high magnification at a high speed, the frame and the table are configured with thin ruled lines. A binary image that can be viewed and processed without breaking the structure of the table than before can be obtained.

  Although the best mode for carrying out the invention has been described above, the present invention is not limited to the embodiment described in the best mode. Modifications can be made without departing from the spirit of the present invention.

1 CPU
2 Memory 3 Work area RAM
4 Program storage ROM / RAM
5 Scanner 6 Printing Device 7 Display 8 CD-ROM / FD Drive 10 Block Dividing Unit 20 Reduced Image Generating Unit 30 Black Pixel Ratio Storage Threshold Determination Unit 40 Density Threshold Determination Unit 50 Binarization Unit 81 Image Memory 82 Work Memory 100 Image Processing apparatus

JP 7-123246 A JP 2002-354233 A

Claims (7)

A block dividing unit for dividing the binary image into blocks ;
For each of the blocks, and converts the multi-level pixel, a reduced image generation unit for generating a reduced image,
For each multi-value pixel, the average of the density of multi-value pixels in a predetermined range including the multi-value pixel is calculated, the number of pixels equal to or higher than the average of the density is set as the minimum value of the number of black pixels, and the minimum value is A black pixel ratio storage threshold value determination unit for determining a black pixel ratio storage threshold value indicating a value converted into density;
A density threshold value determining unit that determines a density threshold value at which one multi-value pixel is a white pixel;
On the basis of the black pixel ratio storing threshold and the threshold value of the density, have a, a binarization unit for binarizing the pixels in which the reduced image has,
The binarization unit includes:
When the density of the multi-value pixel is equal to or higher than the black pixel ratio storage threshold corresponding to the multi-value pixel, and the density of the multi-value pixel is smaller than the black pixel ratio storage threshold corresponding to the multi-value pixel. When the density of the multi-value pixel is larger than the average density of the pixels around the multi-value pixel and the density of the multi-value pixel is equal to or higher than the density threshold, the multi-value pixel is set as a black pixel . An image processing apparatus.   The block dividing unit determines the number of pixels in the vertical direction of the block based on a vertical reduction ratio, and determines the number of pixels in the horizontal direction of the block based on a horizontal reduction ratio. The image processing apparatus according to claim 1. The image processing according to claim 1, wherein the density threshold determination unit determines the density threshold based on a reduction ratio having a larger value out of a reduction ratio in the vertical direction and a reduction ratio in the horizontal direction. apparatus. The black pixel ratio storing the threshold value determining unit, the predetermined range, the image processing apparatus according to any one of claims 1 to 3, characterized in that the entirety of the binary image. The black pixel ratio storing the threshold value determining unit, the predetermined range, any one of claims 1 to 3, characterized in that a predetermined number of lines of pixels of preceding and succeeding pixels in which the reduced image has An image processing apparatus according to 1. A block dividing step for dividing the binary image into blocks ;
For each of the blocks, and converts the multi-level pixel, a reduced image generation step of generating a reduced image,
For each multi-value pixel, calculate the average density of multi-value pixels in a predetermined range including the multi-value pixel, set the number of pixels equal to or greater than the average of the pixels as the minimum value of the number of black pixels, and set the minimum value to A black pixel ratio storage threshold value determining step for determining a black pixel ratio storage threshold value indicating a value converted into density;
A density threshold value determining step for determining a density threshold value at which one multi-value pixel is a white pixel;
On the basis of the black pixel ratio storing threshold and the threshold value of the density, have a, a binarization step for binarizing the pixels in which the reduced image has,
In the binarization step, the density of the multi-value pixel is equal to or higher than the black pixel ratio storage threshold corresponding to the multi-value pixel, and the density of the multi-value pixel corresponds to the multi-value pixel. When the density of the multi-valued pixel is smaller than the black pixel ratio storage threshold, the density of the pixels around the multi-valued pixel is greater than the average, and the density of the multi-valued pixel is equal to or higher than the density threshold, An image processing method, wherein a value pixel is a black pixel . A block dividing step for dividing the binary image into blocks ;
For each of the blocks, and converts the multi-level pixel, a reduced image generation step of generating a reduced image,
For each multi-value pixel, calculate the average density of multi-value pixels in a predetermined range including the multi-value pixel, set the number of pixels equal to or greater than the average of the pixels as the minimum value of the number of black pixels, and set the minimum value to A black pixel ratio storage threshold value determining step for determining a black pixel ratio storage threshold value indicating a value converted into density;
A density threshold value determining step for determining a density threshold value at which one multi-value pixel is a white pixel;
On the basis of the black pixel ratio storing threshold and the threshold value of the density, have a, a binarization step for binarizing the pixels in which the reduced image has,
In the binarization step, the density of the multi-value pixel is equal to or higher than the black pixel ratio storage threshold corresponding to the multi-value pixel, and the density of the multi-value pixel corresponds to the multi-value pixel. When the density of the multi-valued pixel is smaller than the black pixel ratio storage threshold, the density of the pixels around the multi-valued pixel is greater than the average, and the density of the multi-valued pixel is equal to or higher than the density threshold, An information recording medium recording a computer program for causing a computer to execute an image processing method in which a value pixel is a black pixel .

Images