JPH0654180A - Method and device for image area separation - Google Patents

Abstract

PURPOSE:To accurately separate the border of character and photographic areas by deciding both of the areas by whether a no-background picture element extracted by comparing a value obtained by calculating the difference between an average signal level and a maximum signal level within a window including an attention picture element and a threshold value satisfies a prescribed condition or not. CONSTITUTION:Concerning input document image data, the difference between the average signal level and the maximum signal level within the window of 4X4 picture element with the attention picture element in a center is calculated by their calculation circuits 1 and 2. A no-background picture element separation circuit 3 compares the calculation result and the prescribed threshold value to decide whether the attention picture element is no-background color or not. When the deicision is the no-background picture element, an edge picture element deciding circuit 4 decides whether the attention picture element is an edge picture element or not by the calculation result of the circuits 1 and 2. Next, an area deciding circuit 7 calculates through the use of the run length value and the edge counting value of run and edge counters 5 and 6 to decide whether to be the characer or photographic image area in accordance with the values satisfy the prescribed condition or not. Thus, the of both of the areas can accurately be separated.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image area separating apparatus for separating a character area and a photographic area of a multi-gradation document image in which a character portion and a photograph portion are mixed in an image processing apparatus such as a copying machine.

[0002]

2. Description of the Related Art When an image processing apparatus such as a copying machine processes a document image including a binary image typified by a character image and a gradation image typified by a photographic image, a change in signal level occurs in the character image. In contrast to a steep image, a photographic image has a gradual change in signal level, and therefore the character image and the photographic image have different characteristics. Therefore, when copying, it is necessary to separately process the character area and the photographic area, such as performing edge enhancement on the character image and desired image processing such as color conversion on the photographic image.

A conventional area dividing method for equipment mounting will be described below. As an image-related device equipped with an image area division algorithm, a device integrated with a scanner and a printer, such as a facsimile machine or a digital copying machine, can be considered. In a digital copying machine, a document image read by a scanner unit is recorded as it is or after a predetermined image processing is performed on a sheet by a printer unit. On the other hand, in the facsimile machine, the rough flow of transmitting the original image read by the scanner part through the communication line and recording it on the paper by the printer part in the receiving side facsimile device is followed. It is common with the digital copying machine in that data flows.

The reason why the area division is necessary for the image processing in the image-related equipment is that the original image includes a binary image such as characters and figures, an intermediate image such as a photograph, and a pseudo halftone image by halftone dot formation. Can be mentioned. Hereinafter, for simplicity, the binary image is referred to as a character image, the halftone image is referred to as a photographic image, and the pseudo halftone image is referred to as a halftone image. In many facsimile machines and digital copiers, the scanner section reads the entire original as a halftone image, but the printer section records it as a binary image, so it is necessary to perform binarization processing on the read original image. Occurs. Further, in order to prevent deterioration of image quality during binarization, it is necessary to save resolution for character images, reproduce gradation for photographic images, and reduce moire fringes for halftone images according to image characteristics. However, it is difficult to properly binarize these images by one method. Therefore, it is general that the read image is divided into areas to separate character images, photographic images, and halftone dot images, and binarization processing suitable for each image is performed.

Further, in order to mount an image processing apparatus such as area division on image-related equipment, the following restrictions are often taken into consideration in consideration of the characteristics of the equipment. (1) Processing is performed with the minimum memory used. (2) Perform high-speed processing in real time on input data from the scanner section. (1) is a constraint for reducing the overall apparatus cost, and is particularly required for processing by a facsimile apparatus. In addition, (2) requires high-speed processing, and at the same time (1)
There are some restrictions. In order to realize the processing with a small memory, it is necessary to sequentially process the data input from the scanner or the communication line every several lines. In addition to these conditions, the accuracy of area division needs to be guaranteed. An area dividing method suitable for mounting on a device, which has already been reported as a method for separating each area from a character image in which a character image and a photographic image are mixed, will be described below.

Conventional Example 1. Method of using run length of non-background pixels When an image in which a character image and a photographic image are mixed is scanned in the main scanning direction, the run length of non-background pixels is short and the pixels that make up a photographic image Is a method using the feature that the run length of the non-background pixel is long. There is a feature that real-time processing can be performed by line sequential processing, and the system is simple and the circuit can be easily constructed. FIG. 22 shows an example of a block diagram for implementing this method. The outline of the algorithm will be described with reference to FIG. In the hue detection unit, Y, M, C, K,
Any of the eight hues of W, B, G and R is detected. The blocking unit blocks each pixel into several pixels, for example, 4 × 4 to 8 × 8 pixels. The maximum frequency color is determined by the block determination unit for each block created here, and is determined as the block color. For non-background blocks, the main scan run count unit measures the run length of the block, and the judgment and extraction unit judges it as a character image candidate if it is below a predetermined threshold, and if it is above the threshold, it is judged as a photographic image candidate To do. The sub-scanning edge detection unit determines a character image portion candidate when there is a block including an edge in the range of several blocks near the target block, and determines a photographic image portion candidate when there is no block including an edge. . This process
This is used in the sense of correction in order to prevent the lateral line included in a character from being erroneously determined by the photographic image portion in the run length determination in the main scanning direction. By using this method, photographs and characters are well separated. However, since a plurality of pixels are divided into blocks for determination, the boundary between the photograph and the background is not strictly determined. This is appropriate for each area 2
When the binarization is performed, it causes block distortion in the boundary area due to the difference in the binarization method. In addition, it is difficult to set an appropriate run length threshold depending on the character size and photo size, and it is necessary to introduce a dynamic threshold determination method for generalization.

Conventional example 2. Method that uses the maximum density difference in a small area Character image has a sharp density change with peaks in the background and non-background pixels, and a photographic image has a relatively flat density distribution and a gentle density change Is. In this method, an input image is processed in a block made up of several pixels, or in window units in which several pixels around a pixel of interest are referred to in order to sequentially process pixels. Further, since the lines are sequentially processed, real-time processing can be performed, and at the same time, it can be realized by a memory of about 4 lines and a simple logic circuit. FIG. 23 shows a block diagram for implementing this method. The outline of the algorithm will be described with reference to FIG. The processing system has a line buffer of 3 lines and stores an input image signal. This is centered on the pixel of interest, for example 5 ×
Divide the window of 3 pixels. Next, in the image area separating unit (image area separating unit), if the maximum density difference in the window is equal to or larger than a predetermined threshold value T1, it is determined that there is a possibility of a halftone dot and a halftone dot determination is performed. The outline of halftone dot judgment is 0 → 1 after binarization.
Alternatively, the number of change points of 1 → 0 is measured, and if the number of change points is a predetermined threshold value or more, it is determined as a halftone dot candidate, and if it is less than the threshold value, it is determined as a non-dot dot candidate. At this time, the determination using pattern matching is also used in order to improve the accuracy of the halftone dot determination. Next, a window including pixels determined to be a character or a photographic image candidate by comparison with the threshold value T1 has a maximum density difference in the window of T4 or more, which is a value larger than T1, and is less than T4 in the character area. When subdivided into the photo area. The image area separation A variable is a newly introduced variable and represents the degree of belonging to the halftone dot area. For a certain halftone dot area candidate, the halftone dot detection with hysteresis is performed by referring to the image separation A variable of the neighboring pixels stored in the memory 2.
Similarly, for a pixel having a character area candidate, the image area separation B of the surrounding 8 pixels stored in the memory 3 is referred to a variable to detect a character area having hysteresis. One of the purposes of providing the area detection with hysteresis is to reduce the erroneous determination area. When the character area and the photograph area are separated by using the maximum density difference in the window using this method, the area in the photograph where the density change is sharp is erroneously determined as a character. This is reduced by using the image area separation B variable, but it cannot be eliminated. This is not considered to be a problem when the image areas are finally binarized and integrated again, but there is room for improvement in terms of separation accuracy between the character image and the photographic image.

Conventional example 3. Other conventional methods As a method of separating images including character images and photographic images,
Those already reported are shown in FIG. All of the methods in FIG. 24 perform processing for each block, and there are a method of using the maximum density difference in the block, a method of using pattern matching, and a method of calculating the amount of pixels forming an edge. For the separation of text area and photo area, see #
1 uses the maximum density difference in the window including the block or the pixel of interest. As described above, according to this method, the portion in the photograph where the density change is sharp is erroneously determined as a character. #
In No. 2, since a block that does not include pixels forming an edge is determined to be a photograph, pixels at the edge portion in the photograph are erroneously determined to be characters or halftone dots. In # 4, since the image type is determined using the edge pixel density in the block, pixels including edges in the photograph are erroneously determined to be characters or halftone dots, and further blurred characters cannot be detected. Furthermore, since any of these methods is limited to a block size of 3 × 3 or 4 × 4, high-precision separation is impossible.

Conventional example 4. Next, FIG. 25 is a block diagram showing a conventional image area separation system disclosed in, for example, Japanese Patent Laid-Open No. 58-3374.
Is an image memory, 111 is a block memory, 112 is a binarization circuit, 113 is an image quality judgment circuit, 114 is a threshold memory for systematic dither, 115 is a fixed threshold memory, and 1 is a memory.
Reference numeral 16 is a threshold value memory switching circuit, 117 and 118.
Is a switch.

Next, the operation will be described. First, when the switches 117 and 118 are at the positions shown in the figure, the image information read from the document is stored in the image memory 109 via the side a of the switch 117. 1 block is 4 × 4
When the image information corresponding to four scanning lines is stored in the case of the configuration of 6 pixels, the switch 117 is switched from the a side to the b side and the switch 118 is switched from the b side to the a side, and the image information is written in the image memory 110. Meanwhile, the contents of the image memory 109 are sequentially processed at the same time. Regarding the contents of the image memory 109, first, surface information corresponding to one block is transferred to the block memory 111. Next, from the contents of the block memory 111, the image quality judgment circuit 113 discriminates each pixel in the block having the maximum density level and the one having the minimum density level, and the difference between them is determined by a predetermined threshold value m.
It is determined whether it is a photograph area or a character area by comparing with.

The switching circuit 116 sets a threshold value for binarization on the basis of the judgment result of the image quality judging circuit, as a threshold value memory 114 for systematic dither in the photograph area, or a fixed threshold value in the character area. Any of the memories 115 of FIG. The binarization circuit 112 compares the threshold value data of either the threshold value memory 114 or 115 thus selected with the density level of each pixel in the block memory 111, and outputs the comparison result. Is output as a binary signal.

When all the contents of the image memory 109 have been processed and the image information corresponding to the next four scanning lines has been stored in the image memory 110, the switch 117 moves from side b to side a and the switch 118 moves from side a to side b. Switching, image memory 11
The image information is sequentially binarized such that the contents of 0 are processed.

[0013]

Since the conventional image area separation method is configured as described above, an area where the density level of the object such as the contour of a photographic image is sharply changed is erroneously determined as a character area. Thus, a region such as a faint character, which is a binary image and has a gentle change in density level, is erroneously determined as a photographic region. Therefore, there is a problem that the region separation accuracy is low in the entire image. Further, since the image is divided into blocks each having a plurality of pixels for processing, there is a problem that the accuracy of the area determination in the boundary area of the photographic image or the character image is low.

The present invention has been made in order to solve the above problems, and correctly determines a region in a photographic image region where the density level changes sharply as a photographic region and a blurred character as a character region. It is an object of the present invention to provide an image area separating apparatus and method capable of separating the boundary between a character area and a photo area with accuracy.

[0015]

An image area separating apparatus according to a first aspect of the present invention uses an image information consisting of a plurality of types of non-background areas having different pictures and characters and a background area where the non-background areas do not exist. This is intended to separate a plurality of types of non-background regions having different characters, characters, etc., and has the following elements. (A) Input means for inputting by scanning image information having a plurality of types of non-background areas, such as a background area composed of background color pixels and non-background color pixels and having different pictures, characters, etc. (B) non-background color pixel separation means for detecting non-background color pixels from the image information input for each scan by the input means, and (c) in the scanning direction separated by the non-background color pixel separation means. Run detecting means for detecting the length of one or more continuous non-background color pixels as one run, and (d) the non-background color pixels separated by the non-background color pixel separating means constitute an edge of the non-background area. Edge pixel determining means for determining whether the pixel is an edge pixel to be processed, or (e) based on a ratio of non-background color pixels determined as edge pixels by the edge pixel determining means in the run detected by the run detecting means. Region determining means for determining attributes of the run indicating which run belonging to the non-background area of any type of the different kinds of non-background regions are like photos or text run that is detected by the run detection means.

The image area separating apparatus according to the second invention further refers to the image area separating apparatus according to the first invention by referring to the attributes of a plurality of adjacent runs judged by the area judging means to each other. This is provided with a region attribute correcting means for correcting.

The image area separating method according to the third invention has the following steps. (A) An input step of scanning and inputting image information composed of a plurality of types of non-background areas such as photographs and characters and other background areas line by line, and (b) inputting the image information input by the input step. A non-background area detecting step of detecting a non-background area from each line, and (c) an edge detecting step of detecting a portion which becomes an edge of the non-background area from the non-background area detected by the non-background area detecting step. An area determination step of determining the type of the non-background area detected by the non-background area detection step based on the ratio of the detected non-background area and the edge portion thereof by the non-background area detection step and the edge detection step .

[0018]

In the image area separating device according to the first aspect of the present invention, the attribute of the run detected by the run detecting means in the area determining means is determined. Here, the run attribute is information indicating which of the plurality of types of non-background areas such as a photo area and a text area the run belongs to, and the area determination means determines this attribute. It is determined whether the edge pixel determined by the edge pixel determination means is a character area or a photograph area based on the ratio of how much the edge pixel is included in the run. If the proportion of edge pixels in the run detected by the run detection means is high, it is determined as a character area, and if the proportion of edge pixels in the run detected by the run detection means is small, a photograph is taken. Judge as an area. In general, characters are clearly divided into black and white, whereas in photographs, white and black often change gently. Therefore, when the ratio of the non-background color pixels determined as the edge pixels among the continuous non-background color pixels belonging to one run by the run detection means is large, it can be determined as a character. On the other hand, in the case of the photo area, unlike the text area, non-background color pixels are continuous, and since there are few parts that are determined to be edge pixels because they exhibit a smooth change, they are treated as edge pixels in the run. When the determined ratio of the non-background color pixels is small, it is possible to determine that it is a photograph area. As described above, the area determination unit determines a run in the main scanning direction composed of non-background color pixels as a character area when the edge pixel content rate in each run is high, and determines the run as a character area when the edge pixel content rate is low. To judge.

The image area separating apparatus according to the second invention further comprises area attribute correcting means in addition to the image area separating apparatus according to the first invention, and the area attribute correcting means comprises:
The area determination unit refers to the attributes of a plurality of adjacent runs that have been determined, and corrects the attribute of the run. Considering the case where the attributes of the run cannot be correctly determined only by the area determination unit described above. , The attribute of a run is corrected by referring to the attribute of the run adjacent to the run, and the attribute of the run is correctly determined. For example, as described above, a run in the main scanning direction composed of non-background color pixels is determined to be a character region if the edge pixel content rate in each run is high, and is determined to be a photographic area if the edge pixel content rate is low. However, the area attribute correction unit corrects the attribute of the run of interest by referring to whether the adjacent run in the sub-scanning direction is a character region or a photo region.

In the image area separating method according to the third aspect of the invention, after inputting image information composed of a background area and a plurality of types of non-background areas for each line in the input step, a non-background area detecting step is performed for each line. Detect non-background areas from inside.
Further, the edge detecting step detects a portion which is an edge of the non-background area from the non-background area detected by the non-background area detecting step. Then, the area determination step determines whether the type of the non-background area is a photo area or a text area based on the ratio of the edge portion of the non-background area. The area type determination method performed by the area determination step uses the same principle as the area type determination method in the area determination means described in the first aspect of the invention, and a description thereof will be omitted here.

[0021]

EXAMPLES Example 1. First, an outline of the image area separating apparatus according to this embodiment will be described. In this embodiment, image data is input, an average signal level value Dav within the window and a maximum signal level difference Ddif are calculated for an N × N window including a target pixel, and Dav is set to a predetermined threshold value T1.
When the non-background color pixel is extracted by comparing with the non-background color pixel, and the run length Crun of the non-background color pixel column in the main scanning direction up to that point is 0, Dav <T1 / 2
Or Ddif> T2 (T2 is a predetermined threshold value), Cr
When un is not 0 and Ddif> T2 is satisfied, the pixel of interest is determined to be a pixel forming an edge in the image. If the number Cedg of pixels forming an edge included in the same run by the main scanning is equal to or more than a predetermined ratio with respect to the run length Crun, this run is determined as a character region in the image, and if it is less than that, the photo region is determined. To do. (However, “<,>” here is valid when the image signal is represented by a luminance value.)

An embodiment of the present invention will be described below with reference to the drawings. In FIG. 1, 1 is an average signal level calculation circuit for calculating an average signal level within a window centered on a pixel of interest, 2 is a maximum signal level difference calculation circuit for calculating a maximum signal level difference within the window, and 3 is an average. A non-background color pixel separation circuit 4 for determining whether or not the pixel of interest is a non-background color pixel based on the average signal level in the window calculated by the signal level calculation circuit 1. Reference numeral 4 constitutes the edge part in the image of the pixel of interest. Edge pixel determination circuit for determining whether or not a pixel
Reference numeral 5 is a run counter that counts the number of consecutive pixels in the main scanning direction that are determined as non-background color pixels by the non-background color pixel separation circuit 3, and 6 is a pixel that is determined by the edge pixel determination circuit 4 as a pixel forming an edge. Is an edge pixel counter that counts the number of pixels in the same run, and 7 is a photograph showing whether the run is a character area based on the count values of the run counter 5 and the edge counter 6 for a run consisting of non-background color pixels. An area determination circuit for determining whether the area is an area, and 8 is a line buffer for holding an input image signal for four lines, and constitutes a ring buffer for each line.

Next, FIG. 2 is a diagram showing the structure of a pixel window handled by the image area separating apparatus according to this embodiment. Figure 2
, The pixel window consists of 4 × 4 pixels, and
Is the pixel of interest in the pixel window.

Next, the operation of the image area separation device according to the present invention will be described with reference to the case where the reference pixel window size is set to the size of 4 × 4 pixels as shown in FIG. A description will be given according to 1.

When document image data consisting of multiple gradations of characters and photographs are input to the image area separating device, data for four lines is stored in the line buffer 8. From the line buffer 8, 4 × 4 pixel luminance data centered on the pixel of interest is input to the average signal level calculation circuit 1 and the maximum signal level difference calculation circuit 2. The average signal level calculation circuit 1 calculates the average luminance level Dav by performing the calculation of the following (formula 1), and sends the calculation result to the non-background color pixel separation circuit 3 and the edge pixel determination circuit 4.

[0026]

[Equation 1]

Here, Dij is a value indicating the pixel brightness, and the pixel brightness is 0 to 255 here as shown in FIG.
Shall take the value of. That is, when the image information is composed of white and black, the case where the brightness is 0 is black, the case where the brightness is 255 is white, and the interval is gray. It should be noted that here, the brightness is an opposite concept to the density described in the conventional example. When the brightness is 0, the density is 255, and when the brightness is 255, the density is 0. .

In the maximum signal level calculation circuit 2, the maximum brightness value Dmax and the minimum brightness value Dm of the pixels in the window are set.
The following formula (2) is calculated using in, the maximum brightness difference Ddif in the window is calculated, and the calculation result is sent to the edge pixel determination circuit 4. Ddif = Dmax-Dmin (Equation 2) For example, when this (Equation 2) is described with reference to FIG. 2, when there are 16 types of pixel window luminances D11 to D44, D32 is the maximum luminance. Value Dm
ax and D44 is the minimum brightness value Dmin,
Ddif will be calculated in D32-D44.

Next, in the non-background color pixel separation circuit 3, the equation (3) is satisfied by comparing the average density value Dav of the window obtained from the average signal level calculation circuit 1 with a predetermined threshold value T1. At this time, the pixel of interest is determined to be a non-background color pixel. Dav <T1 (Equation 3) For example, this (Equation 3) will be described with reference to FIG.
When the brightness is set to 230, the range between the brightness 230 and 255, that is, the range of B is determined as the background color pixel,
Between the brightness 0 and the brightness 230, that is, in the range F, the pixel is determined as a non-background color pixel.

When it is determined that the target pixel is a non-background pixel, the edge pixel determination circuit 4 uses the average density value Dav and the maximum brightness difference value Ddif to determine whether the target pixel is a pixel forming an edge of the image. To judge.

Next, the operation of the run counter 5 will be described. In FIG. 4, X is a background area in the image information, and Y is a non-background area. Here, in order to simplify the description, the non-background area Y is a rectangular area, and the run length in the main scanning direction is L. Further, R indicates each run constituted by the non-background area Y.
P1 shows the case where the pixel of interest of the pixel window is at the edge of the non-background region Y, and P2 shows the case where the pixel of interest of the pixel window is inside the region of the non-background region Y.
The run counter 5 has a value of 0 until the pixel window reaches the position of P1, and when the pixel window reaches the position of P1, the non-background area Y is started, and thus the run counter is counted up. Begin to. After that, the pixel of interest in the pixel window moves from P1 to P2,
Further, the run counter is counted up until it passes through the non-background area Y, so that the number of non-background color pixels included in the run is counted by the run counter.

The fact that the run counter 5 is 0 means that the scan portion up to now is not the non-background color pixel but the run counter 5
When 1 becomes 1, it means that the pixel of interest is an isolated point or the start point of the run. The edge pixel determination circuit 4 refers to the run counter 5 and executes (equation 4) and (equation 5) when the value of the run counter 5 is 0, and executes when the value of the run counter 5 is not 0 (equation 5). . Dav <T1 / 2 (Equation 4) Ddif> T2 (Equation 5)

FIG. 5 is a diagram for explaining the meaning of (Equation 4) and (Equation 5). FIG. 5 (a) is a diagram showing an example of a photographic region composed of non-background color pixels, and FIG. 5 (b) is a diagram showing an example of a character region. First, the meaning of (Equation 5) will be described using the example of the character area in FIG. When the white and black pixels frequently change as in the case of the character area in FIG. 5B, the maximum brightness difference Ddif of the pixel window has a large value. Therefore, when the character area starts and when the character area ends, the edge can be detected using (Equation 5). However, when an edge is to be detected using only (Equation 5), as shown in FIG.
In the case of a photographic area in which white and black pixels change gently, there is a possibility that the maximum brightness difference Ddif does not exceed the threshold value T2 in any pixel window. For example, in FIG. 5A, although the actual run length starts from the position of S1,
When only (Equation 5) is used for edge determination, in the worst case, the maximum brightness difference Ddif of the pixel window does not exceed T2 and is not determined as an edge at all positions from S1 to S2. Exists. (Equation 4) is used to improve such inconvenience. That is, (Equation 4) is the average luminance level Da of the pixel window.
When v is smaller than the threshold value T1 / 2, that is, when the average luminance of the pixel window becomes smaller than a certain luminance and becomes closer to black, the pixel is determined as an edge. For example, in FIG. 5A, if the pixel window is the threshold value T1 in S3.
When the average brightness level Dav smaller than / 2 is calculated, this position can be determined as an edge. In this way, the average density value Da of the window is
Even when v is larger than T1 / 2, the pixel of interest corresponds to an edge. When it is determined that the target pixel is a pixel forming an edge, the count values of the run counter 5 and the edge counter 6 are incremented by 1. That is, the following calculation is performed. Run length value Crun = Crun + 1 Edge count value Cedg = Cedg + 1 When it is determined that the pixel of interest is not a pixel forming an edge, only the run counter 5 is incremented by 1.

With the above operation, the target pixel is set to 1 in the main scanning direction.
The non-background color pixel separation circuit 3 sequentially shifts the pixels one by one until the target pixel is determined to be a background pixel. When the run counter 5 is not 0, that is, when the runs of the non-background color pixels are continuous, when the next pixel of interest is determined to be the background pixel by the non-background color pixel separation circuit 3, the run counter 5 up to that point
The run length value Crun and the edge count value Cedg of the edge counter 6 are used to calculate the (Expression 6) in the area determination circuit 7 to obtain the edge pixel content rate in the run, and the predetermined threshold value T3 (Expression 6). ), The run is determined to belong to the character image area. When (Equation 6) does not hold, it is determined that the run belongs to the photographic image area. Cedg * 10 / Crun> T3 (Equation 6)

The meaning of the above (formula 6) will be described with reference to FIG. As shown in FIG. 5, in the case of the photograph area, it is assumed that only E1 is determined as the edge. When the run length from S1 to S2 is 10 and it is determined that there is one edge, (Equation 6) becomes 1 × 10/10, and T3
If is set to 5, (Equation 6) does not hold, so
This run is determined to be a photographic area. Also, FIG. 5 (b)
As shown in, when it is determined that the eight non-background color pixels E1 to E8 are edges, (Equation 6) is 8 × 10 /
If it is 10 and T3 is also 5,
(Equation 6) is established, and it is determined that this run belongs to the character image area. The original image is separated based on the attribute of whether the run is a text area or a photo area obtained in this way, and image processing suitable for each of the character image and the photographic image such as binarization and sharpening of the image is performed for each area. Give.

As described above, according to this embodiment, with respect to the window of N × N pixels centering on the pixel of interest in the image,
The maximum signal level difference detecting means for obtaining the difference between the maximum level value and the minimum level value of the image signal in the window, and the average signal level detecting means for obtaining the average level value of the image signal in the window are provided. Non-background color pixel determining means for separating non-background color pixels using a value, and an edge for determining whether the target pixel is a pixel forming an edge in an image using the average signal level value and the maximum signal level difference value Determining means, run length measuring means for measuring run lengths of consecutive non-background color pixels, edge pixel number means for counting pixels determined to form an edge by the edge determining means in the same run, and Based on the measurement results of the length measuring means and the edge pixel number counting means, whether the run consisting of the non-background color pixels of interest is a character area or a photograph area is determined. Characterized by comprising a run attribute determination means for constant.

According to this embodiment, in the image data input to the image area separation device, N × N including the target pixel is included.
Since the processing is performed using the pixel window, the character area and the photograph area can be separated with high accuracy by referring to the surrounding pixels, as compared with the case where the area attribute determination is performed only for one pixel of interest. In addition, image noise points in the document will not be erroneously extracted as characters or photographic area candidates. In addition, the separation accuracy in the image region is improved as compared with the case of processing in blocks of N × N pixels including the target pixel.

The average signal level value D in the window
av, when the run length Crun of the non-background color pixel row in the main scanning direction is 0 for the maximum signal level difference value Ddif, Dav <T1 / 2 or Ddif> T2 (T
2 is a predetermined threshold), Drun when Crun is not 0
When f> T2 is satisfied, it is possible to extract the pixel in the vicinity of the edge in the image with good accuracy by determining that the pixel of interest is a pixel forming the edge in the image.

Further, the number of pixels Cedg forming the edge included in the same run by the main scan is the run length Crun.
On the other hand, when the ratio is equal to or higher than the predetermined ratio T3, the run is determined as the character region in the image, and when it is less than the photo region, the region determination is performed with reference to the connection information between pixels in the main scanning direction. Since it can be performed, the accuracy of the area determination is improved as compared with the case where the area determination is performed only for the pixel of interest or in small area units such as a square block.

Example 2. In the above-described embodiment, the window including the target pixel is a 4 × 4 pixel square window.
Any size such as 3 × 3 or 8 × 8 may be used, and the shape may be other than square.

Example 3. In the above embodiment, when calculating the average signal level as a feature of the image in the window,
Although the image signal level in the window is simply taken as the arithmetic mean, it may be referred to by weighting according to the positions of the target pixel and the reference pixel.

Example 4. In the above embodiment, the threshold value T used for pixel separation in the non-background color pixel separation circuit 3 is used.
Although a predetermined value is used as 1, the threshold value for separating the background pixel and the non-background pixel is changed from the data obtained by scanning the upper few lines of the document image or the feature obtained by scanning the entire image once. You may make a decision.
For example, when a character, a photograph, or the like is input from a newspaper or a colored paper surface, the background color pixel may not show the luminance 255, that is, white, depending on the color of the newspaper paper or the color that is colored in advance. In this case, the first few lines of the document image are analyzed in advance. That is, since there are generally few non-background color pixels in the first few lines of a document image, it can be considered that the data of the first few lines are composed of background color pixels. . Alternatively, by scanning the entire image once, it is possible to know the brightness of the entire background color pixel. In this way, by knowing the average luminance of the background color pixels in advance, for example, when it is found that the luminance is about 235 instead of 255, when the threshold value T1 is 230 as before, The range considered to be the background color is luminance 230 and 2
It becomes a very narrow range of 35, and although it is actually a background color, it is often determined as a non-background color pixel due to some blur or dust. . In such a case, by changing the value of the threshold T1 from 230 to 200, for example,
As described above, it is possible to prevent the separation error between the background color pixel and the non-background color pixel.

Example 5. Further, in the above-described embodiment, the one in which the image signal is represented by the brightness value is used, but other pixel data expressing method such as the density value may be used.

Example 6. In the above-described embodiment, the maximum signal level difference value in the window is used for the edge pixel determination in the edge pixel determination circuit 4, but other edge detection means such as Laplacian may be used.

Example 7. Next, an improved example of the first embodiment will be described. First, an outline of the image area separation device according to this embodiment will be described. In this embodiment, image data is input, an average signal level value Dav in the window and a maximum signal level difference Ddif are calculated for an N × N window including a target pixel, and Dav is set to a predetermined threshold value T1.
When the non-background color pixel is extracted and the run length Crun of the non-background color pixel column in the main scanning direction up to that point is 0, Dav <T1
/ 2 or Ddif> T2 (T2 is a predetermined threshold value),
When Crun is not 0, when Ddif> T2 is satisfied, the pixel of interest is determined to be a pixel forming an edge in the image. If the number Cedg of pixels forming an edge included in the same run by the main scanning is equal to or more than a predetermined ratio with respect to the run length Crun, this run is determined as a character region in the image, and if it is less than that, the photo region is determined. To do. Further, the determination result regarding the run including the non-background color pixel of interest is corrected by referring to whether the run including the non-background color pixels in the preceding and following rows adjacent in the sub-scanning direction is a character area or a photograph area. (However, “<,>” here is valid when the image signal is represented by a luminance value.)

An embodiment of the present invention will be described below with reference to the drawings. In FIG. 6, 1 to 8 are the same as those described in the first embodiment, and the description thereof is omitted here. A region attribute 9 stores a total of three lines of data including a line including a pixel of interest and an adjacent line as a result of determining whether the run of interest is a character region or a photo region by the region determination circuit 7. Buffer, 10
Refers to the area attribute information of all the runs contained in the row in the main scanning direction of interest stored in the area attribute buffer 9, and the area attribute of the run adjacent to the run of interest contained in the adjacent row. This is a region attribute correction circuit for performing correction.

FIG. 7 shows the structure of the area attribute buffer. Here, the area attribute buffer is indicated by info_buf. The area attribute buffer always stores data for three lines including the line of interest, the line preceding the line of interest, and the next line. There is a run number buffer of size n which is about 1/3 of the number of pixels in the main scanning direction for each line. When the run is detected by scanning the line of interest in the main scanning direction in the area dividing unit, the run number 0,
The address and attribute of the start point of the run are sequentially written in the buffer areas 1 ,. When the run detection is completed, the address of the end point of the run is written. Pixel numbers in the main scanning direction are recorded as the start address and the end address. Further, as the attribute, a numerical value representing a number corresponding to the character area or the photograph area is stored. Further, when the scanning of the line of interest is completed, a numerical value indicating the end of data is written in the start address area of the next run number for the processing in the attribute correction unit.

Next, the operation of the image area separating apparatus according to this embodiment will be described. Since the respective parts 1 to 8 shown in FIG. 6 perform the same operations as those shown in the first embodiment, the operation will be omitted here. Description is omitted, and particularly different parts will be described below. As described in the first embodiment, the area determination circuit 7 determines whether the run belongs to the character image area or the photographic image area, and outputs the information regarding the attribute. The information on the attribute of whether the run belongs to the character area or the photo area and the line including the run for which the address value of the run in the main scanning direction is noted and the line adjacent to the line are stored in the area attribute buffer 9. save. The area attribute correction circuit 10 obtains information about area attributes of all runs included in the line of interest stored in the area attribute buffer 9 and information about area attributes of runs of preceding lines and runs of subsequent lines adjacent to the run. Reference is made and correction is performed according to the rule shown in FIG.

The correction using the attribute information of the adjacent area will be described below. As a result of the processing of the area dividing unit, when the attribute information for three lines including the target line in the main scanning direction and the lines before and after the target line is stored in the attribute buffer, the attribute correction for the target line is performed. The attribute correction is performed by referring to the attributes of the runs included in the run included in the target line and adjacent to the line immediately preceding and the line after the target line. FIG. 7 shows the rule of attribute correction. Here, the run of the target line for which attribute correction is to be performed is the run of interest, the run adjacent to the target run by the run of the preceding line is the preceding run, and the run adjacent to the target run is the run of the subsequent line. Is called the next run.

When neither the preceding run nor the following run exists for a certain target run, that is, when the target run is an isolated region, no attribute correction is performed. When the attribute of the preceding run is the photo area and the run of interest is the text area, the run of interest is corrected to the photo area. This is because it is unlikely that the photo area and the character area are adjacent to each other in the document image, and in particular, it has a meaning of correction for erroneously determining the boundary portion of the photo area as a character area due to area division. Therefore, the attribute of the preceding run, which has already been attribute-corrected, is prioritized and the line of interest is corrected to the photographic area. Also, when the next run is a photograph area, the attribute of the noticed run is corrected to a photograph. This is also for correcting the attribute of the boundary of the photo area that is erroneously determined to be a character area by the area dividing unit.

When the attributes of the preceding run and the run of interest are both in the photographic area, the next run is likely to be in the photographic area and requires attribute correction. However, since the correction is performed by the correction process when the next line is focused on in the flow of the entire process, no correction is performed here. Furthermore, whether the attribute is represented by * is the character area or the photograph area, the attribute correction is performed by correction using another combination or correction for the next scan line. Therefore, at the stage of attribute correction focusing on a certain line, these may be either the character region or the photo region.

By referring to the information regarding the area attribute of the run of interest corrected as described above, the character area and the photograph area are separated from the original image, and 2 are separated for each separated image data area. Image processing such as digitization is performed.

As described above, in this embodiment, the run adjacent to each other in the sub-scanning direction is a run forming a character area or a run forming a photograph area, and the run of interest is the character area. The present invention is characterized in that a run attribute correction means for correcting the attribute of a run indicating whether there is a photo area or a photo area is provided.

According to this embodiment, the area attribute in the sub-scanning direction is corrected by referring to the information regarding the area attribute of the adjacent run in the sub-scanning direction with respect to the target run for which the area determination has been performed. Since the area determination can be performed with reference to the connection information of, the accuracy of area separation is improved.

Example 8. In the above embodiment, when the area attribute correction circuit 10 corrects the area attribute of the run of interest,
Although the information about the area attributes for three lines before and after the line of interest stored in the region attribute buffer 9 is used, a total of four lines before the line of interest and two lines after the line of interest may be referred to. Not.

Example 9. Further, in the above-mentioned embodiment, the area attribute correction is carried out for all the runs consisting of the non-background color pixel row. However, the area judgment in the area judgment circuit 7 is performed next, instead of separating the character area and the photograph area by the threshold value T3. In addition, a plurality of thresholds are set to classify into a run that is clearly a text region and a run that cannot be determined to be a text region or a photo region based on the edge pixel content rate, and a run that is clearly a photo region. The area attribute correction circuit 10 may determine the area attribute only for a run that cannot be determined as a photograph area.

Example 10. Next, a specific example of the simulations of Examples 1 and 7 will be described. First, the simulation environment will be described. In FIG. 9, 50 is an image scanner, 51, 52 and 53 are computers,
Reference numeral 54 is a color printer. To perform this simulation, the workstation 52 is first used to create and execute the program. The color image scanner 50 is used to capture the image data. This image scanner has a maximum read line density of 320
The resolution of the dpi is 256 gradations for each color of RGB. In addition, a color printer 54 is provided for hard copy of the processing result.
To use. This printer has a recording linear density of 150 dp
i, the color separation accuracy is 256 gradations for each color of YMC. Also,
The program was created in C language, and a general-purpose window system library was used so that it could be displayed on a CRT.

10 to 12 show examples of recording the original image used in the simulation. Each image uses an image read by a scanner with each pixel having 8 bits / color. FIG. 10 shows 198 of the Electrophotographic Society test chart.
8th edition No. The image size is 592 × 700 pixels. 11 to 12 are sentence images created by combining sentences with SCID image samples, and FIG. 11 shows 661 × 904.
The pixel has a size of 682 × 925 pixels in FIG.

Next, the simulation result will be described. First, the optimization of the threshold will be described. Figure 1
13 and 14 show the results of obtaining the average detection rates of the image regions with respect to the threshold values T2 and T3 for the three types of images of 0 to FIG. FIG. 13 shows the detection rate of the photo area, and FIG. 14 shows the detection rate of the character area. From FIG.
Although the detection rate of a photograph changes depending on T2 and T3, when the optimum T2 for each T3 is selected so that the detection rate becomes maximum, the maximum detection rate is about the same value. The reason for this is that when the edge pixel determination threshold value T2 is decreased with respect to the increase of the edge content rate threshold value T3, the number of edge pixels included in a certain non-background run increases as a result. Therefore, the influence of T3 increase on the determination result can be absorbed by optimizing T2. In addition, FIG.
From this, it can be seen that the character detection rate decreases as T2 increases for each T3, and that the character detection rate decreases as T3 itself increases. This is because as T2 and T3 increase, the criterion for determining that the pixel is an edge pixel becomes stricter, and a region in which non-background pixels are relatively dense among characters is erroneously determined as a photographic region. From these results, the optimum threshold value group in which the detection rates of the character area and the photograph area are both high is when T2 is 40 and T3 is 5. The area detection rate at this time is 99.3% for the photograph area and 100% for the character area.

Next, the image division result will be described. FIGS. 15 to 17 show the results obtained by performing area division on the original images of FIGS. 10 to 12 using this method. Here, the respective threshold values were T1 = 230, T2 = 40, and T3 = 5. 15 to 20, in FIGS.
Is the result of extracting only the photographic area from the original image. Further, FIGS. 16, 18 and 20 show the results of extracting only the character area from the original image. From the area division result, it can be seen that the character area, the photograph area and the background can be separated well.
However, an area for several lines in the main scanning direction at the upper end of the photograph area is erroneously determined as a character area. This is a stage of feature extraction processing using a window, and since the pixel array at the end points of the photographic area is as shown in FIG. 21, the Ddif value becomes large, and the pixel of interest is determined to be a pixel including an edge.
In the non-background run in the main scanning direction including this pixel of interest, all the pixels that make up the run are determined as edge pixels, and Cedg
Since the value increases and (Equation 6) is satisfied, the character area is determined. On the other hand, in the attribute correction accompanying the processing in the sub-scanning direction, correction for only one line is performed, and it is not possible to completely correct erroneous determination of the character area at the upper end of the photo area. For example, when the entire image is binarized, the end points of the photograph are sharpened similarly to the character area, so this erroneous determination does not pose a problem. Further, in FIG. 16, the underline of the title character is determined to be the character area due to the same cause. Also in this case, there is no problem in the determination result considering the nature of the cusps. On the other hand, for example, when compressing and coding the result of area division and using it, the character area containing much visual information in the small area emphasizes the image quality after decompression rather than the compression rate, and the photo area It is common to use an encoding method that emphasizes the compression rate. In this case, when an image including ruled lines such as a diagram is divided into regions, the ruled lines are determined as character regions in this method. A ruled line having a small amount of visual information should be encoded by a method in which the compression rate is more important than the image quality after decompression. Therefore, it is not preferable to determine the ruled line as a character area in which the image quality after decompression is more important than the compression rate, and there is room for improvement when the present method is used as preprocessing for image compression and encoding.

Next, speeding up of the algorithm will be described. As a result of the optimization of the threshold value described above, it was confirmed that a good area division result was obtained when T3 was 5. T3
Is a threshold value for the ratio of pixels including edges in the referenced window. Therefore, when T3 is 5, (Equation 6) can be replaced by (Equation 7). If (Crun >> 1) -Cedg ≧ 0, run of photograph area (Equation 7) If (Crun >> 1) -Cedg <0, run of character area Here, “>> 1” is 1 to the right. Represents a bit shift. By replacing (Equation 6) with (Equation 7), the multiplication and division performed for each run in the image can be replaced with bit shift and subtraction, so that the algorithm can be speeded up. When an algorithm in which (Equation 6) is replaced by (Equation 7) is applied to the original images shown in FIGS. 10 to 12 and a simulation is performed, the processing speed is improved by 10% on average. Further, this makes it possible to simplify the circuit even when the circuit of this system is implemented.

Further, (Equation 4) can be replaced by (Equation 8). Dav <(T1 >> 1) (Equation 8) Here, “>> 1” represents 1-bit shift to the right.
By substituting (Equation 8) for (Equation 4), division (T1
/ 2) can be replaced with a bit shift to speed up the algorithm, and since the threshold T1 can be used in common with (Equation 3), a storage area for storing the threshold in advance Need less

Example 11. In the above embodiment, the case where the non-background area includes the character area and the photograph area is shown, but the character area is an example of the binary image area, and even in the case of other binary image areas such as symbols, figures, and tables. Good. The photographic area is an example of a halftone image area, and may be another halftone image area such as a picture, an illustration, or a television image. Also,
Not only the case where the binary image area and the halftone image area exist completely separately, but also the case where they exist in the same area.

Example 12 In the above embodiment, the case of black and white has been described as an example, but the case of color may be used. For example, in the case of color, the operation described in the above embodiment is performed for each of the three primary colors of cyan, magenta, and yellow (or blue, red, and yellow), and a comprehensive judgment is made on these three types of results. You can do this.

[0065]

As described above, according to the first to third inventions, a binary image such as a character and a halftone image such as a photograph can be extracted with better accuracy than in the conventional case. There is an effect. By accurately separating the areas of the binary image and the halftone image in this manner, the image area separating device and method thereof are used in a facsimile machine, a digital copying machine, etc., so that an output image faithful to the original image can be obtained. It is possible to obtain a device to obtain.

[Brief description of drawings]

FIG. 1 is a block circuit diagram showing a configuration of an image area separation device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of a pixel window handled by an image area separation device according to an embodiment of the present invention.

FIG. 3 is a diagram showing a pixel determination region according to an embodiment of the present invention.

FIG. 4 is a diagram for explaining an operation according to the embodiment of the present invention.

FIG. 5 is a diagram showing the brightness of a photograph area and a character area according to an embodiment of the present invention.

FIG. 6 is a block circuit diagram showing a configuration of an image area separation device according to an embodiment of the present invention.

FIG. 7 is a diagram showing an example of an area attribute buffer according to an embodiment of the present invention.

FIG. 8 is a diagram showing an example of rules for correcting a region attribute for a run of interest in a region attribute correction circuit of an image region separating apparatus according to an embodiment of the present invention.

FIG. 9 is a diagram for explaining a simulation environment showing a specific example of the present invention.

FIG. 10 is a diagram showing an original image used in a simulation according to an embodiment of the present invention.

FIG. 11 is a diagram showing an original image used for a simulation according to an embodiment of the present invention.

FIG. 12 is a diagram showing an original image used for a simulation according to an embodiment of the present invention.

FIG. 13 is a diagram showing a relationship between an edge determination threshold value T2 and a detection rate of a photo area according to an embodiment of the present invention.

FIG. 14 is a diagram showing a relationship between an edge determination threshold value T2 and a character area detection rate according to an embodiment of the present invention.

FIG. 15 is a diagram showing a result of extracting a photographic area from the original image shown in FIG. 10 according to an embodiment of the present invention.

16 is a diagram showing a result of extracting character regions from the original image shown in FIG. 10 according to an embodiment of the present invention.

FIG. 17 is a diagram showing a result of extracting a photographic area from the original image shown in FIG. 11 according to an embodiment of the present invention.

FIG. 18 is a diagram showing a result of extracting a character area from the original image shown in FIG. 11 according to an embodiment of the present invention.

FIG. 19 is a diagram showing a result of extracting a photographic area from the original image shown in FIG. 12 according to an embodiment of the present invention.

20 is a diagram showing a result of extracting a character area from the original image shown in FIG. 12 according to the embodiment of the present invention.

FIG. 21 is a diagram showing a mechanism of a portion erroneously determined according to an embodiment of the present invention.

FIG. 22 is a block diagram of a conventional area division method using run length.

FIG. 23 is a block diagram of conventional area division using density change.

FIG. 24 is a diagram showing another conventional area dividing method for a document image.

FIG. 25 is a block diagram showing a conventional image area separation method.

[Explanation of symbols]

 1 Average Signal Level Calculation Circuit 2 Maximum Signal Level Difference Calculation Circuit 3 Non-Background Color Pixel Separation Circuit 4 Edge Pixel Judgment Circuit 5 Run Counter 6 Edge Counter 7 Area Judgment Circuit 8 Line Buffer 9 Area Attribute Buffer 10 Area Attribute Correction Circuit 11 Target Pixel

Claims (3)

[Claims] 1. An image area separating device having the following elements: (a) Input for inputting image information having a background area composed of background color pixels and a plurality of different kinds of non-background areas composed of non-background color pixels. Means, (b) non-background color pixel separating means for detecting non-background color pixels from the image information input by the input means, (c) one or more in a predetermined direction separated by the non-background color pixel separating means Run detection means for detecting the length of continuous non-background color pixels as runs, and (d) the non-background color pixels separated by the non-background color pixel separation means are edge pixels forming edges of the non-background area. (E) based on the ratio of non-background color pixels determined as edge pixels by the edge pixel determination means in the run detected by the run detection means, Emissions are different types of area determination means for determining an attribute of the run indicating which run belonging to the non-background area of any type of non-background regions. 2. The image area separation device further comprises area attribute correction means for correcting the attribute of a run by referring to the attributes of a plurality of adjacent runs determined by the area determination means. Image region separation device. 3. An image area separating method having the following steps: (a) an input step of inputting image information composed of a background area and a plurality of types of non-background areas line by line; (b) inputting by the input step. A non-background area detecting step of detecting a non-background area from each line of the image information, and (c) an edge detecting a portion which becomes an edge of the non-background area from the non-background area detected by the non-background area detecting step. Detection step, (d) determining the type of the non-background area detected by the non-background area detection step based on the ratio of the non-background area detected by the non-background area detection step and the edge detection step to the edge portion thereof Region determination step to be performed.