JP2853141B2 - Image area identification device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to a binary area in a document image in which a binary image represented by a character line image or the like and a halftone image represented by a photographic image or the like are mixed. The present invention relates to a document image area identification device that identifies a document image and a halftone area.

(Prior Art) When a binary image and a halftone image included in a document are reproduced well as a digital image, it is necessary to perform processing according to the characteristics of the binary region and the halftone region. Therefore, it is necessary to identify a character line image and a photo image. Further, in consideration of improvement of image compression in communication, image identification is indispensable.

As a conventional system, for example, there is one described in Japanese Patent Application Laid-Open No. 58-205376. As shown in the upper part of FIG. 2, the density characteristic of the character line image shows a characteristic in which the density difference between adjacent pixels is large. On the other hand, as shown in the upper part of FIG.
The density difference between the adjacent pixels becomes gentle. In the above-mentioned conventional method, an image is identified by utilizing such a difference in density characteristics between a binary image and a halftone image. That is, M
In this method, the maximum density difference in a block is determined in units of a block of × N pixels, and when the density difference is larger than a threshold value, the area is identified as a binary area, and when the density difference is smaller than a threshold value, it is identified as a halftone area.

(Problems to be Solved by the Invention) However, in the above-described conventional method, there is a problem that an erroneous recognition occurs because an area having a large density difference also exists in a halftone area, thereby deteriorating image quality. The misidentified areas in the halftone area are an edge area from white to black (black to white) as seen in a character area and an edge area from light black to dark black (dark black to light black). In particular, the latter has a drawback that the image quality of the halftone area is deteriorated because the image quality is conspicuous.

SUMMARY OF THE INVENTION An object of the present invention is to solve the drawbacks of the prior art, and an object of the present invention is to make it possible to identify a character line image and a photographic image in a document image with a simple hardware configuration. Furthermore, compared to the conventional method, it is possible to prevent an edge in a black region of a halftone region, that is, an edge that changes from light black to dark black and from dark black to light black, and obtain a good reproduced image. It is intended for.

(Means for Solving the Problems) As shown in FIG. 1, the present invention provides a binarizing unit 1 for binarizing an original image in which a character line image and a halftone image are mixed, Edge detecting means 2 for performing an edge detecting process on the image binarized by the converting means 1, edge detecting means 3 for performing an edge detecting process on the original image, and an edge detected by the edge detecting means 3. When a contour component detected by the contour detecting means 2 is present in the area near the component, the area component is determined as a character area, and the other area is determined as a halftone area.

(Operation) The input n-bit image data in which the character line image and the halftone image are mixed is binarized by the binarizing unit 1 using a fixed threshold, and thereafter, the outline detecting unit 2 converts the image from white to black (black). The image is subjected to contour detection processing (binary image edge detection processing) for finding a point at which the color changes from white to white. On the other hand, the edge component is obtained from the input n-bit image data by the edge detecting means 3, and the result is input to the area determining means 4. The area determining means 4 determines whether the pixel of interest is a character area or a halftone area based on the input edge detection data.

In the present invention, light black to dark black in the halftone area,
Alternatively, even if the region where dark black changes to light black is misidentified as an edge by the edge detecting means, the edge of the same image is binarized and the image is detected and compared with this. Misidentification is eliminated, and high-quality reproduction is possible.

(Description of Examples) Hereinafter, the contents of the present invention will be described in detail based on examples.

As shown in the upper part of FIGS. 2 and 3, the input image signal has density characteristics corresponding to the respective attributes, and the edge detection processing or the edge detection is performed on the image having such characteristics. The results are shown in the lower part of the figure. In the character area, almost all black areas are detected as edge areas. By performing character processing in the vicinity of the edge area, a substantially clear character image is reproduced. On the other hand, most of the halftone area is subjected to halftone processing because the halftone area does not contain much edge components. However, since the edge area is included to some extent, the area is subjected to character processing and the image quality is degraded. The edge of the halftone area is characterized by including many edges from light black to dark black (dark black to light black) in addition to edges changing from white to black (black to white).

This is considered in terms of image quality. The phenomenon in which the edge from white to black (black to white) is subjected to character processing corresponds to so-called contour emphasis, and does not cause a deterioration in image quality. On the other hand, if an edge from light black to dark black (dark black to light black) is subjected to character processing, the area is not the part corresponding to the contour, and the black and white will become unnaturally conspicuous, deteriorating the image quality It is a factor to make it.

Accordingly, the present invention uses the information of the outline (edge component of the binarized image) together with the area identification data to erroneously recognize the edge from light black to dark black or from dark black to light black, which is a factor of image quality deterioration. Is to be able to prevent.

As a method for obtaining an outline component of an image, an outline can be extracted by obtaining an edge component of an image binarized by a fixed threshold value. The difference between the result of contour extraction and the result of edge detection will be described with reference to FIGS. 4 and 5.

In the character area, as shown in FIG. 4, a contour component obtained by detecting an edge after binarizing the original image is:
Due to the density characteristics of the character, it is almost equal to the edge component directly obtained from the density difference for the original image.

On the other hand, in the halftone region, as shown in FIG.
The outline component obtained by detecting the edge after binarizing the original image is only an edge component from white to black (black to white). It does not include edge components from to dark black (dark to light black). Therefore, it is possible to improve the accuracy of image identification by checking whether or not there is a region corresponding to the contour of the image in a nearby region based on the result of the edge component.

In the contour extraction, only one pixel of the contour of the black area is extracted.
Simply taking AND (logical product) significantly deteriorates the image quality of the character area. Therefore, high-quality area separation can be performed by enlarging the area of the image whose contour has been detected, and then performing AND with the edge detection result.

The hardware configuration of the image area identification device according to the present embodiment based on the principle of area identification described in detail above will be described.

As shown in FIG. 6, the image area identification device includes a binarization circuit 11 for binarizing a multi-valued input image read from a document in which a character image and a halftone image are mixed, and a binarization circuit 11
, A line memory 12 for supplying the output of each of the three lines in parallel to the contour detection circuit 13, a contour detection circuit 13 for detecting the contour of the binary image output from the binarization circuit 11, and the contour detection circuit 13
, A line memory 14 for supplying the output of each of the five lines in parallel to a region enlarging circuit 15, an area enlarging circuit 15 for enlarging the outline region detected by the outline detecting circuit 13, and an edge detecting circuit 17 for input lines of three lines in parallel. A line memory 16 for supplying
An edge detection circuit 17 that detects edges of the input image using an edge detection filter such as a Laplacian filter,
A line memory 18 for adjusting the timing of input to the area determination circuit 19, and an area determination circuit 19 for performing area determination based on edge-detected data and area-expanded data
Assuming that the input image is a character image, all processes appropriate for the attributes of the character image are performed, the character processing circuit 20 that supplies the processing result to the image switching circuit 22, and the input image is regarded as a halftone image. A halftone processing circuit 21 that performs processing suitable for the attribute of the halftone image and supplies the processing result to the image switching circuit 22;
An image switching circuit 22 that switches and outputs an image suitable for the attribute of each image based on the output of the area determination circuit 19.

The image binarized by the binarization circuit 11 is supplied to the contour detection circuit 13, and the contour detection circuit 13 performs a contour detection process. In the contour detection processing, a contour is detected by referring to surrounding pixels based on data for three lines. FIG. 7 shows an example of the contour detection circuit.

The In Figure 7, an image is input binarized data is sequentially output by scanning the, by storing two lines line memory 12 _1, 12 ₂ to the scan lines, the line memory 12 ₁ of the output when the target pixel output b, outputs c (adjacent pixels upward) 1 line before the corresponding pixel of the target pixel from the line memory 12 _2, the corresponding pixel (downward after one line of the target pixel input data Is output as an adjacent pixel a). Also outputs a pixel of interest e and preceding pixel adjacent pixels d and neighboring pixels f after one pixel by using the latch 30, 31 from the output of the line memory 12 _1.

The exclusive OR operation (EXOR) circuit 25 performs an exclusive OR operation on the target pixel b and the adjacent pixel a in the downward direction, and performs an AND operation on the output g and the target pixel b with an AND circuit 6
, A signal h representing the lower contour portion of the image is obtained. Further, the exclusive OR operation of the pixel of interest b and the adjacent pixel c in the upward direction is performed by the EXOR circuit 27, and the AND operation of the output i and the pixel of interest b is performed by the AND circuit 28. Obtain a signal j representing the part.
The signals i and j are mixed by an OR circuit 9 to obtain a signal k representing the upper and lower contours of the image.

The EXOR circuit 32 determines the target pixel e and the immediately preceding pixel d adjacent
And an exclusive OR operation with the output 1 and the pixel of interest e
By performing a logical AND operation with AND circuit 33,
The symbol m representing the outline on the right side of the image is obtained. Also,
The exclusive OR operation of the pixel of interest e and the adjacent pixel f on the left is performed by the EXOR circuit 34, and the logical product of the output n and the pixel of interest e is performed by the AND circuit 35. Is obtained. The signals m and o are mixed by the OR circuit 36 to obtain a signal p representing the left and right outline portions of the image.

An OR circuit 38 outputs a signal representing an outline by performing a logical OR operation on a signal k representing the upper and lower contours of the image and a signal p representing the left and right contours. Note that the latch 37 is used to adjust the timing of the pixel of interest b and e since there is a time lag of 1 pixel.

By sequentially performing the above processing while scanning the image, the contour portion can be accurately obtained. Since this contour detection circuit performs sequential processing by hardware using only simple logical operations, the processing is fast and real-time processing is possible.

As described above, since only one pixel of the outline of the black area is extracted in the outline extraction, the image quality of the character area is remarkably deteriorated by simply ANDing the result of the edge detection image and the outline extraction image. Therefore, the area detected by the contour detection circuit 13 is enlarged by the area enlargement circuit 15. FIG. 8 shows an example of the area expanding circuit. In this example, processing for enlarging a black pixel of one dot to 25 dots of 5 × 5 is performed. That is, by using the four line memories 41, the pixel values of the adjacent five dots in the vertical direction are obtained, and
The OR operation is performed by the circuit 42 to expand the contour area in the vertical direction. The output of the OR circuit 42 is taken out in parallel by the 4-stage shift register 43 to obtain pixel values of 5 dots in the horizontal direction. By performing the sum operation, the outline area is expanded in the horizontal direction.

The edge detection circuit 17 detects a portion having a large density gradient as an edge with respect to a multi-valued image input, and can be configured by using a known edge detection filter such as a Laplacian filter.

The area determination circuit 19 obtains an AND of the data whose edge is detected by the edge detection circuit 17 and the data obtained by enlarging the outline area by the area expansion circuit 15, and determines that the area is an edge area and a character area if the area is an outline area. The region other than is determined to be an intermediate region.

The character processing circuit 20 performs processing on the input image data according to the characteristics of the characters, and the halftone processing circuit 21 performs processing on the input image data according to the characteristics of the halftone image.

The image switching circuit 22 switches the output of the character processing circuit 20 and the output of the halftone processing circuit 21 based on the output of the area determination circuit 19, and outputs the processed image suitable for each area.

(Effects of the Invention) According to the present invention, a character area and a halftone area can be identified for a document image in which a character line image and a photographic image are mixed. In particular, it is possible to prevent erroneous recognition of an edge that changes from light black to dark black (dark black to light black), which has caused deterioration in image quality in the conventional method, and thus it is possible to reproduce a high-quality image. Further, according to the present invention, a character line image and a photographic image in a document image can be identified with a simple hardware configuration.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a basic configuration of the present invention. FIG. 2 shows an example of a density characteristic of a character line image and an edge detection result. FIG. 3 shows an example of a density characteristic of a halftone image and an edge detection result. FIG. 4 is a diagram for explaining the operation of determining the area of the character line image. FIG. 5 is a diagram for explaining the operation of determining the area of a halftone image. FIG. 6 is a block diagram showing the overall configuration of one embodiment of the present invention. FIG. 7 is a diagram showing an example of the contour detection circuit 13. FIG. 8 is a diagram showing an example of the area enlarging circuit 15. 11 ... binarization circuit, 12, 14, 16, 18 ... line memory,
13 ... contour detection circuit, 15 ... range expansion circuit, 17 ... edge detection circuit, 19 ... area determination circuit, 20 ... character processing circuit, 21 ... halftone processing circuit, 22 ... image switching circuit .

Continuation of the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G06T 7/40 H04N 1/40

Claims (1)

(57) [Claims] 1. A binarizing means for binarizing an original image in which a character line image and a halftone image are mixed, and performing an outline detection process on the image binarized by the binarizing means. Contour detection means; edge detection means for performing edge detection processing on the original image; and, if a contour component detected by the contour detection means is present in a region near the edge area detected by the edge detection means, the area is determined. And a region determining means for determining the other region as a halftone region.