JPH1125272A - Picture identification device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To determine characteristics for the change of gradation information for a picture in which a multilevel PDL (business graphics) picture and a natural picture are present. SOLUTION: A connected lines counting means 1a receives a multilevel picture data 11, scans each picture element, and compares the value of a picture element with the value of a scanned adjacent picture element, operates binarization into '1' when they are identical, and '0' when they are different, and outputs the number of connected lines which is 12 for each picture element under consideration. A picture identifying means 1b receives the number of the connected lines which is 12 outputted from the connected lines counting means 1a, calculates the total number of the connected lines within a prescribed reference range, determines a reversible encoding area by comparing it with a prescribed threshold value, and outputs identification information 13 being the determined result. A reference range controlling means 1c inputs resolution information 14, and sets a reference range 15 for determining the number of the connected lines according to resolution.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image identification technique for identifying characteristics of each area of an image, and more particularly to an image identification technique which can be used as a preprocessing for adaptive encoding for switching a plurality of encoding systems according to the characteristics of an image. About technology. The present invention
Particularly, in an apparatus that inputs and encodes an image in which an image generated from a page description language (hereinafter referred to as PDL) and an image scanned in from a scanner are mixed, a PD is used.
The present invention relates to an image identification technique that can be applied when identifying an application area of an encoding method suitable for an L-generated image.

[0002]

2. Description of the Related Art First, related techniques will be described. 1. Lossless and irreversible encoding Digital images include computer generated CG
(Computer Graphics) Business graphics (hereinafter collectively referred to as PDL images) described in images and PDL, images scanned in from a scanner (hereinafter, referred to as PDL images).
There is an image in which a PDL image and a natural image are mixed.

A PDL image is an image generated by overwriting a large number of lines or planes with a certain pixel value. Therefore, there is a high probability that the same pixel continues in a specific direction, the pixel value changes sharply, and the edge becomes sharp. It has the property of being clear. With this characteristic, the PDL image can be efficiently compressed by the lossless encoding method. When a PDL image is compressed by the irreversible encoding method, image quality is easily deteriorated such as blurred edges.

[0004] On the other hand, a natural image has a characteristic that noise is included, a pixel value often changes locally, and an edge is not clear. The lossy coding method removes redundant components from image data that do not affect the image quality, whereas the lossless coding method does not remove redundant components due to noise or local pixel changes. For this reason, when a natural image in which the same pixels are not continuous is compressed by the lossless encoding method, the compression ratio becomes low. Natural images cannot be efficiently compressed unless the lossy coding method is used.

[0005] As can be understood from the above, when encoding an image in which a PDL image and a natural image are mixed, efficient compression is not possible unless the encoding method is switched according to the characteristics of the image in the page. Since the lossless encoding method can perform encoding without deteriorating the image quality, it is particularly desirable to identify a region that can be efficiently compressed by the lossless encoding method.

When identification information indicating a region between a PDL image and a natural image is given in advance, the encoding process may be switched based on the identification information. When there is no identification information, for example, when an image in which a PDL image and a natural image are mixed is input as raster data, it is necessary to identify the lossless encoded area before performing the encoding process.

When the same pixel continues even in a natural image,
Compression can be efficiently performed by the lossless encoding method. That is, PD
Even when the L image and the natural image do not coexist, each encoding method corresponds to a local characteristic variation. Therefore, P
It is desirable to identify a region that can be efficiently encoded by the lossless encoding method regardless of the type of the DL image or the natural image.

The above is the switching of the encoding system for the purpose of improving the coding efficiency. In addition, there is the switching of the encoding system for the purpose of improving the image quality.
For example, a character region is cut out from an image in which characters and photos are mixed, and the character region is losslessly encoded so that edges do not deteriorate, and the photograph region is encoded by a lossy encoding method. The character region has a characteristic of edge information, and the photograph region has a characteristic of gradation information identifying each region.

The processing other than the coding is switched according to the characteristics of the image. For example, if you want to output an image containing both halftone and non-halftone dots on a low-resolution printer,
The dot area is smoothed so that moire does not occur. A halftone dot is represented by a binary area gradation, and the number of halftone dots has a characteristic that depends on a halftone dot period and a region area.

As described above, there are various switching processes according to the characteristics of an image, but all of them are based on the technology of identifying an area by focusing on the characteristics of an image. 2. As a conventional technique for identifying areas having different characteristics, there is a technique for identifying a halftone area in an image in which halftone and non-halftone dots are mixed. For example, Japanese Patent Laid-Open Publication No. Hei 2-103885 discloses that, based on the number of regions where black pixels are connected (hereinafter referred to as black regions) and regions where white pixels are connected (hereinafter referred to as white regions), in a predetermined block. A method for identifying a halftone dot region is described.

The halftone dots represent the density by area gradation according to the size of the black area. Most of the low-density areas are white areas, and the respective black areas are isolated. Therefore, the number of areas is larger in the black area than in the white area. On the other hand, in an area having a high density, the number of areas in the white area is larger than that in the black area.

In order to measure the number of black areas and white areas, a threshold value is set for black and white according to the average density of the block,
If the pixel value is larger than the black threshold, it is converted to black; if it is smaller than the white threshold, it is converted to white; if it is between the black and white thresholds, it is converted to gray, and the black and white regions in the block are separated. Vertical or horizontal 2
Assuming that the connection of two adjacent pixels is line connection, and the connection forming a plane with four adjacent pixels, two in each of the vertical and horizontal directions, is plane connection, the number of black or white areas in the halftone area is the number of pixels, the number of lines The number of connections and the number of surface connections are counted, and can be obtained by the following equation.

[0013]

## EQU1 ## (Number of regions) = (Number of pixels)-(Number of line connections) +
(Number of Surface Connections) An advantage of the above-described method is that the number of regions can be calculated only by counting local lines or surfaces without calculating pixel connections globally as in labeling.

In the halftone dot area, the number of areas is the number of halftone dots and is determined by the interval between the halftone dots. On the other hand, in the non-dot area, the number of areas varies depending on the content of the image. As described above, the halftone area is identified by utilizing the difference in the number of areas between the halftone area and the non-halftone area.

The outline of the process for identifying a halftone dot region will be described with reference to FIGS. FIG. 6 is a configuration diagram of the halftone dot region identification processing, and FIG. 7 is an explanatory diagram of the region number measurement. In FIG. 6, first, image information 51 is stored in a buffer memory 5a, and density data 52 is ternarized into black, gray, and white by dynamic ternarization 5b. Next, the number-of-regions measurement 5c calculates the calculation elements 6c, 6d, 6 in units of predetermined blocks.
e and 6f are counted, and the number of areas is measured. At this time, the periphery of the block is masked with the same pixel value as the pixel value of the measurement target region as indicated by the thin line portion in FIG. 7B, and the region in contact with the edge of the block is not counted. For the white area and the black area, the number of areas is measured, and the larger number of areas is adopted, and a temporary halftone dot candidate 54 is determined by comparing with a predetermined threshold value. Subsequently, based on the results corrected by the respective correction processes of 5d, 5e, and 5f, a halftone dot region and a non-halftone dot region are identified. 3. Discrimination Between Multi-Valued PDL Image and Natural Image Since a halftone dot region is represented by a binary area gradation, a change in gradation information within a halftone dot has no meaning. Similarly, only the edge information of the character area is significant. As described above, in the halftone dot area and the character area, the edge at which the pixel value changes is important image information, and the conventional image identification uses this characteristic.
However, in both the PDL image and the natural image, image information is represented by a change in gradation information. In the conventional image identification, the characteristic with respect to the change of the gradation information is not determined.
The DL image cannot be distinguished from the natural image.

For example, in the prior art, a pixel value is ternarized into black, gray, and white by a threshold value of black and white, and separated into a black region and a white region. If this is applied to an image in which a PDL image and a natural image coexist, the pixel values in each threshold segment are masked with the same value, so that the characteristics and pixel values of the PDL image in which the same pixels continue are locally varied. The characteristics of the natural image cannot be determined.

FIG. 8 shows an example in which a region where the same pixel continues cannot be identified. FIG. 8 is a diagram showing a change in a pixel value in a certain horizontal direction in an image in which a PDL image and a natural image are mixed. 7a is an example of an image in which a PDL image and a natural image are mixed, 71 is a PDL image area where the same pixels are continuous, and 72 is a natural image area where the pixel value changes. 7b is a horizontal pixel value at a vertical position B in the PDL image area, and 7c is a horizontal pixel value at a vertical position (B + 1) in the natural image area. 7b,
7c shows a change in the pixel value from the horizontal position x1 to the horizontal position x2. In this example, the ternarization of the pixel value
Both the 71 PDL image and the 72 natural image are gray areas, and the 71 PDL image area and the 72 natural image area are identified as the same area. Thus, a pixel value of 3
When the value is converted to a value, it is impossible to identify a region where the same pixel continues in an image in which a PDL image and a natural image are mixed.

[0018]

As described above, in the conventional image identification, since the area identification is performed using characteristics other than the gradation information, PDs having different characteristics of the gradation information are used.
A lossless coding area cannot be identified for an image in which an L image and a natural image are mixed.

An object of the present invention is to identify a lossless coding area of an image in which a multi-level PDL image and a natural image are mixed.

Another object of the present invention is to identify a characteristic based on gradation information of a multi-valued image and optimize an encoding method, a printing method, and other image processing methods according to the identification result. I have.

[0021]

According to the present invention, a region where the same pixel continues is identified for each pixel value in order to determine the characteristic with respect to a change in gradation information.

That is, according to the present invention, in order to achieve the above object, each pixel of the input multi-valued image data is scanned by the image identifying device, and the value of the pixel of interest and the adjacent pixel that has been scanned are scanned. And a line connection counting unit that outputs the number of line connections that connect pixels having the same value, and a line connection number that is output from the line connection counting unit. Calculating a total number of line connections related to each other, comparing the total number of line connections with a predetermined threshold value, and providing image identification means for identifying a characteristic of each area of the multi-valued image data based on a result of the comparison. ing.

In this configuration, by identifying a region where the same pixel continues for each pixel value, it is possible to determine the characteristic of the multi-valued image data with respect to the change in the gradation information. Then, for example, the coding method can be switched based on the change in the gradation information. Also, the printing method is switched.

Further, in this configuration, the image identification means may determine whether each of the areas of the multi-valued image is a lossless coding area based on the result of the comparison. Further, a reference range for determining the number of line connections may be set based on resolution information of the multi-valued image data.

According to the present invention, in order to achieve the above-mentioned object, for each pixel of the multi-valued image data input to the image discriminating apparatus, the value of the pixel and the adjacent pixel in a predetermined direction And a line connection counting unit that outputs the number of line connections that connect pixels having the same value, and a line connection number that is output from the line connection counting unit. Image identification means for calculating the total number of line connections for pixels, comparing the total number of line connections with a predetermined threshold, and identifying characteristics of each region of the multi-valued image data based on the result of the comparison. I have to.

Also in this configuration, by identifying the area where the same pixel continues for each pixel value, it is possible to determine the characteristic of the multivalued image data with respect to the change in the gradation information.

Further, according to the present invention, in order to achieve the above-mentioned object, in the multi-valued image data input to the image identification device, the area formed by pixels having the same value as the value of the pixel concerned And an image identification means for calculating the total number of the areas within a predetermined reference range and identifying the characteristics of each area of the multi-valued image data based on the total number.

Also in this configuration, by identifying the area where the same pixel continues for each pixel value, it is possible to determine the characteristic of the multivalued image data with respect to the change in the gradation information.

Further, according to the present invention, in order to achieve the above-mentioned object, for each pixel of the input multi-valued image data, the multi-valued image data is stored in a predetermined direction. And a line connection counting unit that outputs the number of line connections that connect pixels having the same value, and a line connection number that is output from the line connection counting unit. Calculate the total number of line connections for pixels in the range,
Means for comparing the total number of line connections with a predetermined threshold value and determining whether each of the areas of the multi-valued image is a lossless coding area based on the result of the comparison, and executing based on the determination result And a lossless encoding means.

In this configuration, the characteristic of the multivalued image data with respect to the change in the gradation information can be determined by identifying the area where the same pixel continues for each pixel value. Then, lossless encoding is selectively applied when the same pixel value continues, whereby encoding can be performed at a high compression ratio even when a PDL image and a natural image coexist.

[0031]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail.

In the present invention, in order to determine the characteristic of the change of the gradation information for each pixel value, the target pixel and the adjacent pixel are compared and set to “1” if they are the same value and “0” if they are different. 2
The value is converted to the number of line connections for each pixel of interest. Then, the number of line connections is calculated in a predetermined range, and the size of the surface of the region where the same pixels are continuous is represented by the number of line connections in the region. The characteristics of the lossless coding region are determined by the number of line connections without discriminating the direction in which the same pixels continue, and the lossless coding region is identified.

FIG. 1 shows the structure of the present invention.
In this figure, the image identification device includes a line connection counting unit 1.
a, image identification means 1b and reference range control means 1c. The line connection counting means 1a receives the multi-valued image data 11, scans each pixel, compares the value of the pixel of interest with the value of the scanned adjacent pixel, and determines "1" if the values are the same and "1" if the values are different. It is binarized to 0 ", and the number 12 of line connections is output for each pixel of interest. Image identification means 1b
Is the number 12 of line connections output from the line connection counting means 1a.
, The total number of line connections within a predetermined reference range is calculated, the lossless coding area is determined by comparison with a predetermined threshold value, and the identification information 13 as the determination result is output. The reference range control unit 1c receives the resolution information 14 and sets a reference range 15 for determining the number of line connections according to the resolution.

In this configuration, a lossless coding area can be identified for an image in which a multi-valued PDL image and a natural image are mixed, and adaptive coding can be performed. Further, by setting the reference range according to the resolution, it is possible to identify the characteristics of local line connection asymptotic to the global characteristics.

Of course, the present invention can be applied to other than the switching of the encoding method. For example, the present invention can be applied in various ways such as switching of a printing method and switching of image processing.

Next, the present invention will be described in more detail with reference to examples. Embodiment 1 Embodiment 1 of the present invention will be described with reference to FIGS.

FIG. 2 is a configuration diagram of the first embodiment of the present invention. First, an outline of the identification processing will be described with reference to FIG. In FIG. 2, a thick line indicates data, and a thin line indicates control information.

In FIG. 2, in order to perform the identification processing, first, image information 29 is input to an image input interface 2b from a scanner or a transmission line, and output to a buffer memory 2c as multi-bit image data 210 of 8 bits per pixel. Thereafter, the control unit 2a is notified of the input notification 21.

When receiving the input notification 21, the control unit 2a instructs the buffer memory 2c with a block address 22 for cutting out a block of m pixels × n pixels. The buffer memory 2c outputs the block image 211 corresponding to the block address 22 to the block memory 2d. Further, the control unit 2a sends the scan instruction 23 to the block memory 2d.
Is transmitted to the Euler number measuring device 2e.
Is sent. The control unit 2a sets the number of pixels 26 of the block in the Euler number measuring device 2e in advance before sending the calculation instruction 25.

The block memory 2d sequentially scans the pixels in the block according to the scan instruction 23 from the control unit 2a, and stores the pixel of interest and the pixel data 212 of the reference pixel referenced by the template shown in FIG. Output to the number measuring device 2e. When the scanning of all the pixels in the block is completed, the block memory 2d notifies the control unit 2a of the scanning end notification 24.

The Euler number measuring device 2e measures the Euler number 213 in accordance with the procedure shown in FIG. 4 according to the calculation instruction 25 from the control section 2a, and outputs it to the threshold value judging device 2f.

Upon receiving the scan end notification 24, the control unit 2a sets a predetermined threshold value 28 in the threshold value judgment unit 2f,
A judgment instruction 27 is sent.

The threshold value judging device 2f receives the Euler number 213 of the block image 211 measured by the Euler number measuring device 2e and the threshold value 28 according to the judgment instruction 27 from the control section 2a.
Are compared with each other, and if they are smaller than the threshold, they are identified as lossless coding areas, and if they are larger than the thresholds, they are identified as lossy coding areas, and predetermined identification information 214 assigned to each is output.

Next, the processing of the Euler number measuring device 2e will be described with reference to FIGS.

FIG. 3 is an explanatory diagram of line connection. 3a is
Templates 3b and 3c for comparing the value of the pixel of interest and the value of the scanned adjacent pixel are line connections when 2 pixels × 2 pixels are the same pixel, and 3d to 3n are examples of counting the number of line connections. . To calculate the Euler number of the line connection, 3b
When 2 pixels × 2 pixels are the same pixel as shown in
Make sure that the lines representing the connections do not cross. For example, in the line connection of 3b, the line connection in the diagonal direction is limited to only the pixel 36 and the pixel 37 as in 3c, and the pixel 35 and the pixel 38 are not regarded as the line connection even if they are the same pixel.

When there are three different pixel values in the reference range of 3 pixels × 3 pixels as shown in 3d, the total number of connections in the reference range is obtained as follows. First, scanning is performed sequentially from the upper left, and the target pixel is moved as shown by 3e to 3m. In the case of 3e, there is no pixel included in the template of 3a, so the number of connections is 0. In the case of 3f,
Since the value of the target pixel is A and the value of the left adjacent pixel included in the template 3a is also A, the value is “1”, and the number of connections is 1. Similarly, in the case of 3 g, the number of links is one. In the case of 3h, the upper and upper right neighboring pixels are included in the template 3a. However, since the value A is different from the value B of the target pixel, the number of connections is 0. Hereinafter, similarly, the number of connections in the case of 3i, 3j, 3k, 31 and 3m is as follows:
They are 1, 2, 2, 2, and 0, respectively. Then, the above number of line connections is accumulated over the reference range and the number of connections along the line is 9
(3n).

FIG. 4 is a flowchart for counting the number of line-connected Eulers for a block image of m pixels × n pixels. The number of branches in the flowchart indicates the number of line connections.

The Euler number measuring device 2e calculates the Euler number of the line connection. In FIG. 4, each pixel of the block image of m pixels × n pixels is scanned from the 0th column to the (n−1) th column for each row from the 0th row to the (m−1) th row. First, initialization is performed in step 4a,
At s and 4t, the line number is incremented from 0 to (m-1). Also, in steps 4c and 4d, the column number of the 0th row is incremented from 0 to (n-1), and in step 4e
Reset to 0 with. Then, in steps 4q and 4r, the first
Similarly, the column numbers in the (m-1) th row from the row are set to 0 to (n
-1), and the column number is reset to 0 in step 4u. Thus, scanning of the pixel is performed.

For the 0th row, the value of the target pixel and the value of the pixel to the left of the target pixel are compared, and if the values are the same, 1 is added to the number of branches (step 4b). From the first row to the (n
In the 0th column up to the -1) row, the value of the pixel of interest and the value of the pixel above the pixel of interest and the value of the pixel on the upper right of the pixel of interest are compared. (Steps 4f, 4g, 4h, 4i). From the first column to the (n−2) th column, the value of the target pixel and the pixel to the left of the target pixel, the value of the pixel at the upper left of the target pixel, the value of the pixel above the target pixel, and the target The values of the pixels at the upper right of the pixels are compared, and if they are the same value, 1 is added to the number of branches (steps 4f, 4g, 4j, 4j).
k, 41, 4m). For the (n-1) th column, the value of the pixel of interest and the pixel to the left of the pixel of interest, the value of the pixel at the upper left of the pixel of interest, and the value of the pixel above the pixel of interest are each compared. If they are the same value, 1 is added to the number of branches (steps 4f, 4g, 4n, 4o, 4p). However, as shown in 3b of FIG. 3, when the value of the pixel on the left of the target pixel is equal to the value of the pixel on the target pixel,
Even if the value of the target pixel and the value of the pixel at the upper left of the target pixel are the same, they are not added to the number of branches. When the scanning of all the pixels in the block is completed, the number of lines in the block is subtracted from the number of pixels set by the control unit 2a to obtain the Euler number of line connection.

Second Embodiment Next, a second embodiment of the present invention will be described with reference to FIG. In the second embodiment, the image identification device of the present invention is applied to an image encoding device. In FIG. 5, portions corresponding to those in FIG. 1 are denoted by corresponding reference numerals, and detailed description thereof is omitted.

In FIG. 5, first, the multi-valued image data 11 in which the PDL image and the natural image are mixed is determined by the line connection counting means 1a.
The resolution information 14 is input to the reference range control unit 1c. The reference range control unit 1c outputs the resolution information 14
, A reference range for determining the number of line connections is determined, and is output to the line connection counting means 1a as a reference range 15. The line connection counting unit 1a scans each pixel of the multi-valued image data 11 corresponding to the reference range 15, compares the value of the pixel of interest with the value of the scanned adjacent pixel, and if the value is the same, "1"; It is binarized to “0” and the number of line connections is 12 for each pixel of interest.
Is output to the image identification means 1b. Image identification means 1b
The number 12 of line connections output from the line connection counting means 1a is input, and the total number of line connections in the reference range 15 is calculated.
The lossless coding area is determined by comparison with a predetermined threshold value, and the identification information 13 as the determination result is output to the coding unit 1d. The encoding unit 1d performs PDL according to the identification information 13.
An encoding method for efficiently compressing an image and an encoding method for efficiently compressing a natural image are selected, the multi-valued image data 11 is encoded using the selected encoding method, and encoded data 16 is output. . In the embodiment of FIG. 5, a multi-valued image in which a PDL image and a natural image are mixed is referred to as a local reference range 15.
Is identified as the PDL image or the natural image according to the line connection number 12 of FIG. 1, and the encoding method is switched according to the identification information 13 to perform encoding. Therefore, the multi-valued image in which the PDL image and the natural image are mixed Can be efficiently encoded.

It should be noted that the present invention is not limited to the above-described embodiment, but can be variously modified. For example, the line connection counting means 1a may determine the number of regions having the same pixel value and determine the characteristics of the image.

[0053]

The effects of the present invention are as follows. 1. A multi-valued PD is used to identify a region where the same pixel continues for each pixel value and identify the characteristics of the change in gradation information.
It is possible to identify a lossless encoded area of an image in which an L image and a natural image are mixed. 2. Since the sum of the line connections is calculated within a predetermined range, the lossless coding area can be identified regardless of the direction in which the same pixels are continuous. 3. By determining the reference range for determining the number of line connections according to the resolution, it is possible to identify the characteristics of the local line connections asymptotic to the global characteristics.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of the present invention.

FIG. 2 is a block diagram illustrating a configuration of a first exemplary embodiment of the present invention.

FIG. 3 is a diagram illustrating line connection in the first embodiment of the present invention.

FIG. 4 is a flowchart illustrating the calculation of the Euler number of line connection in Embodiment 1 of the present invention.

FIG. 5 is a block diagram illustrating a configuration of a second exemplary embodiment of the present invention.

FIG. 6 is a block diagram showing a configuration of a conventional example for separating a halftone dot region.

FIG. 7 is a diagram illustrating the number of regions in a conventional example for separating a halftone dot region.

FIG. 8 is a diagram illustrating an example in which a PDL image cannot be distinguished from a natural image.

[Explanation of symbols]

1a Connection number counting means 1b Image identification means 1c Reference range control means 1d Encoding means 11 Multi-valued image data 12 Number of line connections in multi-value area 13 Identification information 14 Resolution information 15 Reference range 16 Code data 2a Control unit 2b Image input Interface 2c Buffer memory 2d Block memory 2e Euler number measuring device 2f Threshold judgment unit 21 Input notification 22 Block address 23 Scan instruction 24 Scan end notification 25 Calculation instruction 26 Block pixel count 27 Judgment instruction 28 Image information 29 Multi-valued image data 210 Block image 211 Pixel data 212 Euler number 213 Identification information 3a Adjacent pixel value comparison template 3b Same pixel of 2 × 2 pixels 3c Line connection when 2 × 2 pixels are the same pixel Example of counting the number of line connections 3d to 3n 31-34 adjacent Comparison pixel position of prime value comparison template 35-38 Pixels of the same value 4a-4v Processing for calculating Euler number of line connection 5a Buffer memory 5b Dynamic ternarization 5c Number of areas measurement 5d Lump size correction 5e By peripheral pixels Correction 5f Edge correction 51 Image information 52 Density data 53 Ternary data 54 Temporary halftone dot candidate 55 Temporary halftone dot 56 Halftone dot candidate 57 Pixel in halftone area 58 Pixel in non-halftone area 59 Pixel in photo area 6a Number of areas Example of measurement 6b Example of block peripheral mask in area number measurement 6c to 6f Connected element for calculating the number of areas 7a Example of image in which PDL image and natural image are mixed 7b Pixel value of PDL image in which identical pixels continue 7c Pixel value Of the natural image in which the image changes 71 Area of the PDL image 72 Area of the natural image

Claims (6)

[Claims] 1. Scanning each pixel of input multi-valued image data, comparing the value of a pixel of interest with the value of a scanned adjacent pixel, and outputting the number of line connections connecting pixels of the same value. Line connection counting means, and the number of line connections output from the line connection counting means are input, the total number of line connections for pixels within a predetermined reference range is calculated, and the total number of line connections is compared with a predetermined threshold. And an image identifying means for identifying a characteristic of each area of the multi-valued image data based on a result of the comparison. 2. The image encoding apparatus according to claim 1, wherein the image identification unit determines whether each of the areas of the multi-valued image is a lossless encoding area based on a result of the comparison. 3. The image identification device according to claim 1, wherein a reference range for determining the number of line connections is set based on resolution information of the multi-valued image data. 4. For each pixel of the input multi-valued image data, the value of the pixel is compared with the value of an adjacent pixel in a predetermined direction, and the number of line connections connecting pixels of the same value is output. Line connection counting means, and the number of line connections output from the line connection counting means is input, the total number of line connections for pixels within a predetermined reference range is calculated, and the total number of line connections is a predetermined threshold An image identification unit that performs comparison and identifies characteristics of each area of the multi-valued image data based on a result of the comparison. 5. A means for detecting an area formed by pixels having the same value as the value of the input multi-valued image data, and calculating a total number of the areas within a predetermined reference range; An image identification unit that identifies characteristics of each area of the multi-valued image data based on the total number. 6. For each pixel of the input multi-valued image data, the value of the pixel is compared with the value of an adjacent pixel in a predetermined direction, and the number of line connections connecting pixels of the same value is output. Line connection counting means, and the number of line connections output from the line connection counting means is input, the total number of line connections for pixels within a predetermined reference range is calculated, and the total number of line connections is a predetermined threshold Comparing means for determining whether each of the areas of the multi-valued image is a lossless coding area based on the result of the comparison; and lossless coding means executed based on the determination result. Multi-valued image coding apparatus.