JP3100383B2 - Character region separation method and apparatus - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character area separation method and apparatus for accurately separating a character area in an image from other picture areas.

[Conventional technology]

Generally, in an image called a document image, characters (including line drawings), photographs, halftone dots, and the like are usually mixed.
When such an image is reproduced by, for example, a digital copying machine or a facsimile, the image quality of the reproduced image is improved. (A) Characters and lines are reproduced clearly so as not to be cut off.

(B) For photographs and halftone dots, emphasis is placed on gradation, and halftone dots are further removed from moire.

It is desirable to take such measures. For example, in the case of reproducing an image in monochrome binary, for example, in order to realize the above (a), the edges of characters and lines are emphasized, and then binarization processing is performed.

In order to realize the above (b), dither processing is performed after smoothing a photograph or a halftone dot.

And the like. However, in order to perform this processing, first, before the processing, the character area (the fifth
It is necessary to accurately separate the part surrounded by the dotted line in the drawing example) from the picture area (= photograph part + halftone part).

Conventionally, a method of separating a character area from a photographic area (for example, Japanese Patent Publication No. 58-115975) or a method of separating a character area from a halftone dot area (for example, Japanese Patent Publication No. 60-80365) have been found. There has been no method for directly separating the character area from the picture area. Therefore, in order to separate a character area and a picture area in an image, the above two methods are combined, and a logical sum (OR) of a photograph area separated and extracted by the former method and a halftone dot area separated and extracted by the latter method is used. ), (Photo area +
(Halftone dot area) as a picture area.

Therefore, the present applicant has previously applied for a character area separation method for directly separating the character area from the picture area. In this character area separation method, input image information is converted into digital multi-valued data and then smoothed, edge pixels are extracted from the smoothed digital multi-valued data, and the number of edge pixels is counted for each predetermined local area. When the density of the edge pixels exceeds a predetermined threshold value, a predetermined pixel in the local region is detected as a character pixel, and when the density is equal to or less than the threshold value, a region is detected as a picture pixel other than a character.

[Problems to be solved by the invention]

The character area separation method proposed by the present applicant can directly separate a character area and a picture area in an image. However, since the character area and the other picture areas are separated based on the density of the edge pixels for each predetermined local area, the edge pixels of the halftone dot are erroneously determined as the character pixels depending on the type of the halftone dot and its distribution state. It could be detected and was not perfect.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a character area separating method and apparatus which improves the above character area separating method and further enhances the accuracy of separating a character area from a picture area.

[Means for solving the problem]

In order to achieve the above object, the present invention provides a character area separating method for separating a character area from an input image, performing a smoothing process on the input image data, extracting edge pixels from the smoothed image data, The character area in the image is provisionally determined based on the density information of the edge pixel in the local area, and the target pixel belonging to the provisionally determined character area is separated from the target pixel by a predetermined distance in a predetermined direction. It is determined whether or not the target pixel is a regular character pixel based on attribute information of a plurality of pixels present at the position.

In a character area separating apparatus for separating a character area from an input image, a smoothing means for performing a smoothing process on the input image data; an edge pixel extracting means for extracting edge pixels from the smoothed image data; Character region determining means for temporarily determining a character region in an image based on the density information of the edge pixels in both local regions of the local region; and a target pixel belonging to the provisionally determined character region and a predetermined direction from the target pixel in a predetermined direction. And a region correction unit that determines whether or not the pixel of interest is a regular character pixel based on attribute information of a plurality of pixels existing at positions separated by a distance of.

(Operation)

According to the present invention, an area correction is performed using the property of a character. That is, in general, a character is formed by connecting the lines forming the character vertically, horizontally, or diagonally. Therefore, when the pixel of interest is a regular character pixel, the line segment of the character is surrounded by the pixel of interest. Are usually present.

The method of the present invention utilizes the continuity of the line segments forming the character described above, determines whether or not a character pixel detected based on the edge density is a regular character pixel based on surrounding pixel information, and If the character pixel is not the character pixel, the region correction is performed by deleting the character pixel from the character region. Thereby, the separation accuracy between the character area and the picture area is improved.

〔Example〕

FIG. 1 shows an embodiment of a character area separating apparatus constructed by applying the method of the present invention.

The input image signal unit 1 is a circuit for raster-scanning an original image, converting the original image into digital multi-value data, and taking in the data. If the reproduced image is monochrome, a luminance signal is input. If the reproduced image is color, an RGB signal after color separation or a YMC signal after color correction is input. In the case of color, the processing described below may be performed in parallel for each color signal.

The smoothing processing unit 2 is a circuit that smoothes the digital multi-valued data sent from the input image signal unit 1 using a smoothing filter having a predetermined weight coefficient. As the smoothing filter, for example, a filter expressed by the following equation (1) is used.

The coefficient k is about 1 to 10, and an optimum value is selected according to the size of the character to be separated, the number of halftone dots, and the scanner or printer used.

The differentiation processing unit 3 is a circuit that differentiates the smoothed data sent from the smoothing processing unit 2 in order to extract edge pixels in an image. As the differentiation circuit, for example, the following (2) A Laplacian filter or the like represented by the equation is used.

The binarization processing unit 4 is a circuit that binarizes the differentiation result of the differentiation processing unit 3 with a predetermined threshold value TH1, and extracts pixels equal to or greater than TH1 as edge pixels.

The count processing unit 5 includes a predetermined local region, for example, N ×, which is predetermined for the edge pixels extracted by the binarization processing unit 4.
The number of edge pixels P is counted in units of a block of N pixel size, and the count value P is compared with a predetermined threshold value TH2 to determine whether the pixel of interest in the block belongs to a character area or a picture area. In this area detection, one of the following detection conditions is adopted depending on whether the block is processed while being moved in units of one pixel or while being processed in units of one block.

When the count value P is equal to or more than a predetermined threshold value TH2, the central pixel in the block is a character area, and when the count value P is equal to or less than the threshold value TH2, the central pixel in the block is a picture area.

(When moving by one block) When the count value P is equal to or more than a predetermined threshold value TH2, all pixels in the block are set as a character area, and when the count value P is equal to or less than the threshold value TH2, all pixels in the block are set as a picture area. .

The correction processing unit 6 determines whether or not the character pixel detected by the counting processing unit 5 is a regular character pixel by using the above-described character property, and according to the determination result, determines the character area and the pattern. This is a circuit for correcting the detection result of the area. The surrounding pixels used for this correction include a plurality of pixels located at a predetermined distance from the pixel of interest in a predetermined direction, for example, as shown in FIG.
Eight pixels that are positioned vertically, horizontally, and diagonally above the target pixel in a block of n pixel size (n = 3 in the figure) are adopted. When the central pixel is a character pixel, if at least one character pixel is present among the eight pixels surrounding the central pixel, the central pixel is determined to be a regular character pixel, and the eight peripheral pixels are determined. If no character pixel is present, the center pixel is determined to be a picture pixel.

The operation of the above embodiment will now be described with reference to the flowchart of FIG. The target image is an image in which a photograph part, a halftone part, and a character part are mixed, and a character area and a picture area (photo part + halftone part) are separated from the mixed image. The halftone dots to be separated are about 65 to 200 lines,
Characters are about 7th grade or larger, for example, 400dp
i, It shall be read by a scanner of about 64 gradations.

The image signal is converted into digital multi-value data in the input image signal unit 1 and stored. The digital multi-valued data is sent to the smoothing processing unit 2 and smoothed by a smoothing filter as shown in equation (1) (step [1]). (Step [2]). The binarization processing unit 4 binarizes the differential data with a predetermined threshold value TH1 to extract edge pixels in the image (step [3]).

The counting section 5 sets the edge pixel information detected as described above for each of N lines (step [4]), and sets N ×
The number P of edge pixels in the block is counted while applying a block of N pixel size to each pixel (step [5]). Then, the count value P for each block is compared with a predetermined threshold value TH2, and when the count value P is equal to or greater than the threshold value TH2, the central pixel of the block is set to a character pixel. The center pixel is detected as a picture pixel (step [6]), and the character area and the picture area are separated based on the density information of the edge pixels.

Each process up to the point described above is the same as the character region separation method previously proposed by the present applicant. Thus N ×
By performing region separation based on the edge pixel density in an N-pixel size block, even if a part of a character in an input image has blurring, the blurring portion can be effectively detected as a character region. Even if there is a halftone dot edge pixel, a considerable portion can be removed because the halftone dot edge pixel has a low density.

However, even if the regions are separated as described above, edge pixels of the halftone dots may be erroneously detected as character regions depending on the types of the halftone dots and their distribution states. Therefore, in the present invention, the area correction is performed on a character pixel that is erroneously detected as described below.

That is, the correction processing unit 6 sets the detection result of the counting processing unit 5 for every n lines (step [7]). For example, when region correction is performed using a 3 × 3 pixel size block shown in FIG. 2, n = 3 lines are set, and while the 3 × 3 pixel size block shown in FIG. 2 is moved in units of one pixel, The number Q of character pixels in eight surrounding pixels located vertically, horizontally, and diagonally above the center pixel of each block is counted (step [8]).

If the center pixel is a character pixel and there is at least one character pixel in the eight surrounding pixels, it is determined that the center pixel is a regular character pixel (steps [9] and [1]).
0]), a character area signal is output to the character pixel without performing area correction.

On the other hand, when there is no character pixel in the surrounding eight pixels, it is determined that the center pixel is a pixel which is erroneously detected as a character pixel while being a pixel which originally constitutes a picture. [11]) For the center pixel, a picture area signal is output instead of the character area signal. Therefore, as a result of the area correction of the picture pixel erroneously detected as the text pixel to the original picture pixel, the separation accuracy of the text area and the picture area is improved.

In the above embodiment, the correction process is performed in units of one pixel, but may be performed in units of blocks. In this case,
All the pixels in the block may be corrected to either character pixels or picture pixels.

FIG. 4 shows an example of a copying machine constructed using the character area separating apparatus of the present invention.

In the figure, reference numeral 7 denotes the character area separating device shown in FIG. 8 is a sharpening processing circuit for edge enhancement for character area processing, 9 is a binarization processing circuit, 10 is a smoothing processing circuit for picture area processing, and 11 is a vortex emphasizing the gradation for producing halftones. Type dither processing circuit, 13 is a character area separating device 7
Is an image signal selection circuit for selecting an output signal of either the binarization processing circuit 9 or the dither processing circuit 11 in accordance with the character / pattern area signal output from. In FIG. 4, the above-described processing by the sharpening processing circuit 8 and the binarization processing circuit 9 is performed.
The above-described processing is realized by the smoothing processing circuit 10 and the dither processing circuit 11, respectively.

The image signal is input in parallel to each of the character area separating device 7, the sharpening circuit 8, and the smoothing circuit 10. The character area separation device 7 determines whether all pixels of the input image signal are a character area or a picture area as described above, and sends the determination result to the image signal selection circuit 12.

The image signal selection circuit 12 selects the output of the binarization processing circuit 9 when a character area signal is given from the character area separation device 7, and outputs the dither processing circuit 11 when a picture area signal is given. Select As a result, an image sharpened by the sharpening processing circuit 8 and the binarization processing circuit 9 is output in the character area of the input image, and the smoothing processing circuit is output in the picture area.
The image subjected to pseudo halftone processing by the dither processing circuit 10 is selectively switched and output.

Therefore, if an image is copied and reproduced using the image signal output from the image signal selection circuit 12, the character area in the image is sharpened, and the picture area other than the text, such as a photograph or a halftone dot, has a natural feeling. To obtain a high-quality image subjected to pseudo halftone processing. Note that the smoothing processing circuit 10 in FIG. 4 and the smoothing processing unit 2 in FIG. 1 can be shared.

〔The invention's effect〕

As is apparent from the above description, when the character separation method and apparatus of the present invention are used, it is possible to further improve the separation accuracy between a character area and a picture area in an image.

[Brief description of the drawings]

FIG. 1 is a diagram showing an embodiment of the present invention, FIG. 2 is a diagram showing an example of surrounding pixels used in correction processing, FIG. 3 is a flowchart of the above embodiment, and FIG. FIG. 5 is a diagram illustrating an example of a copying machine configured by using the apparatus, and FIG. 5 is a diagram illustrating an example of a character area in an image.

──────────────────────────────────────────────────続 き Continued on the front page (56) References JP-A-63-263974 (JP, A) JP-A-63-240175 (JP, A) JP-A-64-81080 (JP, A) JP-A-1- 227573 (JP, A) (58) Field surveyed (Int. Cl. ⁷ , DB name) H04N 1/40-1/409 H04N 1/46 H04N 1/60

Claims (6)

(57) [Claims] In a character area separating method for separating a character area from an input image, a smoothing process is performed on the input image data, and edge pixels are extracted from the smoothed image data.
A character region in the image is provisionally determined based on the density information of the edge pixel in the predetermined local region, and a target pixel belonging to the provisionally determined character region is separated from the target pixel by a predetermined distance in a predetermined direction. A region correction for determining whether or not the pixel of interest is a legitimate character pixel based on attribute information of a plurality of pixels present at the specified position. 2. The method according to claim 1, wherein the plurality of pixels located at a predetermined distance from the target pixel in a predetermined direction are eight pixels located in the vertical, horizontal, and oblique directions of the target pixel. Item 1. The character area separating method according to Item 1. 3. The character area separating method according to claim 2, wherein when there is no character pixel in the eight pixels, it is determined that the pixel of interest is not a regular character pixel. 4. A character region separating apparatus for separating a character region from an input image, wherein: a smoothing means for performing a smoothing process on the input image data; and an edge pixel extracting means for extracting edge pixels from the smoothed image data. Character region determining means for temporarily determining a character region in an image based on density information of the edge pixel in a predetermined local region; and a target pixel belonging to the provisionally determined character region and a predetermined Region correction means for determining whether or not the pixel of interest is a regular character pixel based on attribute information of a plurality of pixels located at positions separated by a predetermined distance in the direction. Character area separation device. 5. A plurality of pixels located at a predetermined distance from the pixel of interest in the area correction means in a predetermined direction in a predetermined direction are eight pixels located in the vertical, horizontal, and oblique directions of the pixel of interest. 5. The character area separating apparatus according to claim 4, wherein: 6. The character region separating apparatus according to claim 5, wherein when there is no character pixel in the eight pixels, the pixel of interest is determined to be not a regular character pixel.