JPH10257308A - Image processing apparatus and method - Google Patents

Abstract

(57) [Problem] To improve the quality of an image obtained by processing based on a detected region by detecting a boundary region between a character / line drawing region and another image region in the image. A character / line image / halftone area is detected by a character / line image / halftone area detection unit.
24 detects a dot area. Detection signal B of both detectors
Signal PICT_I which is a logical sum of INGRA and AMI
In the image related to SO, non-detection areas are scattered in the boundary area between the halftone area and the character area. Therefore, the re-detection unit 202
7, it is determined whether or not the pixel of interest belongs to the non-detection area based on the degree of isolation of the white pixel, and the non-detection area is detected based on the result, whereby the signal PICT_ISO is corrected and the appropriate detection signal is detected. Generate a PICT.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an image processing apparatus and method, and more particularly to an image processing apparatus and method for extracting features of an image and processing the image based on the result.

[0002]

2. Description of the Related Art In recent years, color image data obtained by digitally processing color image data and outputting it to a color printer to obtain a color image, and color reading of a color original by color separation and electrical reading. A color printing system such as a so-called digital color image copying machine for copying a color image by printing out the image on recording paper has been remarkably developed.
With the spread of these, there is an increasing demand for printing quality of color images, especially for printing characters and line drawings more sharply.

[0003] Therefore, an apparatus for sharply printing a character or a line drawing by determining the thickness of a character or a line drawing and performing image processing according to the thickness has been proposed by the applicant of the present invention (Japanese Patent Application No. 2002-214,197). Hei 6-178501).

In this proposal, a character / halftone area detecting section and a halftone area detecting section are provided, and both the detecting sections detect a character / halftone area and a halftone area. To execute image processing corresponding to.

[0005]

However, in the above proposal, when a character or a line drawing is present in a halftone dot area, an image of a boundary portion where the halftone dot and the character or the like are in contact is detected by one of the detection units. May not be detected. When such a non-detection area exists, an erroneous determination is made when determining the size of the image area to which the pixel of interest belongs based on the detection results by the two detection units. In this case, improper image processing is performed on the image of the boundary portion, and the quality of the output image is reduced.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-described problems, and has been made by performing processing based on a detected area by detecting a boundary area between a character / line drawing area and another image area in an image. The objective is to improve the quality of the image obtained.

[0007]

SUMMARY OF THE INVENTION An image processing apparatus according to the present invention is an image processing apparatus for extracting features of an image and processing the image based on a result of the extraction. First detecting means for detecting a tone area, second detecting means for detecting a halftone area in the image, and second detecting means for detecting a boundary area between the character / line drawing / halftone area and the halftone area in the image. 3
And processing the image based on the detection results of the first to third detection means.

In the above-mentioned image processing apparatus, the third detecting means may detect a region of a character, a line drawing, a halftone region and a halftone dot region in an image from an area not detected by the first and second detecting units. Preferably, a boundary region is detected.

In the above-mentioned image processing apparatus, the third detecting means may be arranged such that each pixel has a character / line image / halftone area in the image based on a distribution of pixels not detected by the first and second detecting means. It is preferable to detect the boundary region by determining whether or not the pixel belongs to the boundary region between the pixel region and the halftone dot region.

In the image processing apparatus, the third detecting means may be configured such that each pixel is a character / line drawing / halftone area in the image based on a series of pixels not detected by the first and second detecting means. It is preferable to detect the boundary region by determining whether or not the pixel belongs to the boundary region between the pixel region and the halftone dot region.

In the above-mentioned image processing apparatus, the third detecting means may determine whether each pixel is a character, a line drawing, or a character in an image based on the number of consecutive pixels not detected by the first and second detecting means. It is preferable to detect the boundary area by determining whether or not the pixel belongs to the boundary area between the halftone area and the halftone area.

In the above-mentioned image processing apparatus, the third detecting means converts a character, a line drawing, and a character in the image from an image constituted by a set of detection results of the first or second detecting means.
It is preferable to detect a boundary region between the halftone region and the halftone dot region.

In the above-mentioned image processing apparatus, the third detecting means may be configured such that each pixel is formed based on a distribution of undetected pixels in an image formed by a set of detection results of the first or second detecting means. It is preferable to detect the boundary area by determining whether or not the pixel belongs to the boundary area between the character / line image / halftone area and the halftone area.

In the above-mentioned image processing apparatus, the third detecting means may be configured such that each pixel is formed based on a series of undetected pixels in an image constituted by a set of detection results of the first or second detecting means. It is preferable to detect the boundary area by determining whether or not the pixel belongs to the boundary area between the character / line image / halftone area and the halftone area.

In the above-mentioned image processing apparatus, the third detecting means may determine each of the undetected pixels based on the number of undetected pixels in an image constituted by a set of detection results of the first or second detecting means. Pixels are characters, line drawings,
It is preferable to detect the boundary area by determining whether or not the pixel belongs to the boundary area between the halftone area and the halftone area.

It is preferable that the image processing apparatus further includes a judging means for judging the size of the image area based on the detection result by the first to third detecting means, and processes the image according to the judgment result. .

An image processing apparatus according to the present invention is an image processing apparatus for extracting a feature of an image and processing the image based on the result, wherein a character / line drawing area in the image and a character / line drawing area other than the character / line drawing area are included. Main detecting means for detecting another image area, and sub-detecting means for detecting a boundary area between the character / line drawing area and the other image area. The method is characterized in that an image is processed based on the image.

In the image processing apparatus, it is preferable that the sub-detecting means detects the boundary area from an area not detected by the main detecting means.

In the image processing apparatus, the sub-detecting unit determines whether each pixel is a pixel belonging to the boundary area based on a distribution of pixels not detected by the main detecting unit. Preferably, a boundary region is detected.

In the image processing apparatus, the sub-detecting unit determines whether each pixel is a pixel belonging to the boundary area based on a series of pixels not detected by the main detecting unit. Preferably, a boundary region is detected.

In the image processing apparatus, the sub-detecting unit determines whether each pixel is a pixel belonging to the boundary area based on the number of consecutive pixels not detected by the main detecting unit. It is preferable to detect the boundary region by the above.

Preferably, each of the image processing apparatuses further includes an output unit that outputs a processed image.

It is preferable that each of the image processing apparatuses further includes an input unit for inputting an image to be processed.

In the image processing apparatus, the input means preferably includes means for reading and inputting a document image.

Each of the above image processing apparatuses preferably has a function of processing a full color image.

An image processing method according to the present invention is an image processing method for extracting a feature of an image and processing the image based on the result, and detecting a character, a line drawing, and a halftone area in the image. 1, a second detection step of detecting a halftone dot area in an image, and a third detection step of detecting a boundary area between a character / line drawing area and a halftone dot area in the image, The method is characterized in that the image is processed based on the detection results of the first to third detection steps.

An image processing method according to the present invention is an image processing method for extracting features of an image and processing the image based on a result of the extraction, wherein a character / line drawing region in the image and a region other than the character / line drawing region are included. A main detection step of detecting another image area, and a sub-detection step of detecting a boundary area between the character / line drawing area and the other image area. The method is characterized in that an image is processed based on the image.

The computer readable memory according to the present invention comprises:
A computer-readable memory containing a program code for extracting a feature of an image and processing the image based on the result, the computer-readable memory having a first detection step of detecting a character, a line drawing, and a halftone area in the image. Code, a code of a second detection step of detecting a halftone dot area in the image, and a code of a third detection step of detecting a boundary area between the character / line drawing area and the halftone dot area in the image. It is characterized by.

The computer readable memory according to the present invention comprises:
A computer readable memory containing a program code for extracting features of an image and processing the image based on the result, wherein a character / line drawing area in the image and another image area other than the character / line drawing area And a code for a sub-detection step for detecting a boundary area between the character / line drawing area and the other image area.

[0030]

Preferred embodiments of the present invention will be described below with reference to the accompanying drawings.

FIG. 1 is a diagram showing a sectional configuration of an image processing apparatus according to a preferred embodiment of the present invention. In FIG. 1, reference numeral 201 denotes an image scanner unit which reads a document and performs predetermined signal processing on a read image signal.
Reference numeral 228 denotes a host computer which generates an image signal and outputs the image signal via the controller 227 to the image scanner unit 2.
01. A printer unit 200 outputs a full-color image on a recording medium based on a document image signal read by the image scanner unit 201 or an image signal sent from the host computer 228.
That is, this image processing apparatus functions as both a copying machine and a printer.

In the image scanner unit 201, the original 2 placed on the original platen glass 203 by the original pressing plate 202
Light is irradiated by the halogen lamp 205 with the 04 pressed, the reflected light from the document 204 is reflected by the mirrors 206 and 207, and an image is formed on the three-line sensor (CCD) 210 by the lens 208.

The CCD 210 separates the reflected light from the document 204 into colors, reads red (R), green (G), and blue (B) components of full-color information, and sends the read signal to the signal processing unit 209. .

When reading an original, the halogen lamp 20
5 and the mirror 206 at a speed v, and the mirror 207 at a speed (１ ／) v, in a direction perpendicular to an electrical scanning direction (hereinafter, main scanning direction) in the three-line sensor 210 (hereinafter, sub-scanning direction). ), And scans the entire surface of the document 204.

The standard white plate 211 is a three-line sensor 21
R, G, B sensors 210-1 to 210-3 constituting 0
Is provided to generate correction data for correcting the read data by the. This standard white plate 211
Has substantially uniform reflection characteristics with visible light, and its color is white.

The image signal processing unit 209 outputs an original image signal obtained by reading the original 204 or the host computer 22.
8 is selected, and the selected image signal is decomposed into magenta (M), cyan (C), yellow (Y), and black (Bk) color components and subjected to image processing. The processing result is sent to the printer unit 200.

In the printer section 200, the image signals of M, C, Y, and Bk that have been subjected to the image processing are transmitted to the laser driver 212.
Sent to The laser driver 212 drives the semiconductor laser 213 while performing PWM modulation according to the image signal. The laser light emitted from the semiconductor laser 213 is applied to a polygon mirror 214, an f-θ lens 215, and a mirror 21.
6, the photosensitive drum 217 is scanned.
The printer unit 200 has a resolution of 200 lines /
400 lines can be switched.

The developing device is composed of a magenta developing device 219, a cyan developing device 220, a yellow developing device 221 and a black developing device 222. These four developing devices alternately contact the photosensitive drum, and Formed in M. The electrostatic latent images for C, Y, and Bk are developed with corresponding toner. The transfer drum 223 wraps the sheet supplied from the sheet cassette 224 or 225 so as to attract the sheet, and transfers the toner image formed on the photosensitive drum 217 to the sheet.

As described above, four of M, C, Y, and Bk
After the color toner images are sequentially transferred to the sheet, the sheet passes through the fixing unit 226 and is discharged.

FIG. 2 is a diagram showing a detailed configuration example of the image processing unit 209 of the image scanner unit 201. The image signal output from the three-line sensor 210 is input to the analog signal processing circuit 101, where gain adjustment and offset adjustment are performed, and then, in the A / D conversion circuit 102, an 8-bit digital image signal R 1 for each color signal. G1, B
Converted to 1. Digital image signals R1, G1, B1
Are input to the shading correction unit 103 and subjected to shading correction using a read signal of the standard color plate 211 for each color signal.

The clock generator 121 generates a clock signal for each pixel. The main scanning address counter 122 counts a clock signal supplied from the clock generation unit 121 and generates a main scanning address. The decoder 123 decodes the main scanning address from the main scanning address counter 122, and detects a line-by-line CCD drive signal such as a shift pulse or a reset pulse, or a one-line read signal supplied from the three-line sensor 210. VE signal indicating effective area, line synchronization signal HSYN
Generate C. The main scanning address counter 122
Is cleared by the HSYNC signal, and the counting of the main scanning address of the next line is started.

The sensors 210-1 to 210-3 constituting the three-line sensor 210 are arranged at a predetermined distance from each other. to correct. The input masking unit 106 converts the image into an NTSC standard color space.

In the image signal switching unit 1064, the image signals Rif,
The image signals R4, G4, and B4 related to reading by the Gif, Bif, or three-line sensor 210 are selected and output as image signals R40, G40, and B40.

In the background processing section 1065, the image signal R4
Detection and removal of background components of 0, G40, and B40 are performed.
Whether or not to perform background removal can be instructed by the operation unit 101.

Light / density converter (LOG converter) 107
Includes a look-up table ROM, and the luminance signals of R5, G5, and B5 are referred to as C0, M
It is converted into a density signal of 0, Y0. Line delay memory 10
8 is an image signal R in a black character determination unit 113 described later.
40, G40, UCR generated based on B40, F
The image signals C0, M0, and Y0 are delayed so as to match the generation timing of the determination signals of ILTER and SEN.

The masking UCR unit 109 includes C1, M
A black signal (K) is extracted from the three primary color signals of Y1, M1, C1, and K, and a black signal (K) is extracted from the three primary color signals. (Eg, 8 bits) image signals Y2, M2, C
2 and K2 are output in a frame-sequential manner.

The main scanning direction scaling circuit 110 scales the image signals Y2, M2, C2, K2 and the black character determination signals FILTER, SEN in the main scanning direction by a known interpolation operation, and outputs the image signals Y3, M2, C2. , K2 and black character determination signals FILTER ', SEN'.

Output filter (spatial filter) unit 111
Selects edge enhancement or smoothing processing based on a black character determination signal FILTER 'instructing selection of edge enhancement processing / smoothing processing, performs the selected processing on the image signals Y3, M3, C3, and K3, The signals are output as signals Y4, M4, C4, and K4.

The M4, C4, Y thus generated
4 and K4, and a black character determination signal SEN 'for instructing selection of 200 lines / 400 lines are sent to the laser driver 212 of the printer unit 200,
At 00, an output image expressing the density by PWM is formed.

Next, the operation of the black character determination section 113 will be described. In the black character determination section 113, the image signals R40, G40, B4 output from the image signal switching section 1064 are output.
0, the black character determination signals UCR, FILTER,
Generate SEN.

[Explanation of Edge Detector] Image signals R40, G40, B output from image signal switcher 1064
40 is input to the R, G, and B terminals of the edge detection unit 115 of the black character determination unit 113, where edges in the image are detected. FIG. 3 is a block diagram illustrating a configuration example of the edge detection unit 113.

The luminance calculation circuit 250 outputs the image signal R (R4
0), G (G40), and B (B40), for example, the luminance is calculated by Expression (1).

Y = 0.25R + 0.5G + 0.25B (1) FIG. 4 is a block diagram showing a detailed configuration example of the luminance calculation circuit 250. The luminance calculation circuit 250 receives the input signals R,
The results obtained by multiplying G and B by predetermined coefficients by multipliers 401 to 403 are added by adders 404 and 405 to obtain a luminance signal Y.

The edge min direction detecting circuit 251 determines a direction in which the absolute value of the edge amount takes the minimum value (hereinafter referred to as “edge min direction”) from the input luminance signal Y.

FIG. 5 is a diagram for explaining the function of the edge min direction detection circuit 251. The edge min direction detection circuit 251 delays the input luminance signal Y one line at a time by FIFOs 501 and 502 and performs a well-known Laplacian filter process of 3 × 3 pixels. The Laplacian filters 503 to 506 are filters in the vertical direction, diagonal direction, horizontal direction, and diagonal direction, respectively, as illustrated. The edge min direction detection circuit 251 finds the direction in which the absolute value of the edge amount, which is the output value of the filter in the four directions, takes the minimum value, and sets the direction as the edge min direction.

Edge min direction smoothing circuit 252
Performs smoothing processing on the obtained edge min direction. By this processing, only the direction having the largest edge component can be stored, and the other directions can be smoothed. That is, the feature of the halftone dot component having a large edge component in a plurality of directions is reduced because the edge component is smoothed. On the other hand, features of characters and thin lines having edge components in only one direction are preserved. By repeating this processing as necessary, the separation of the character / line component and the halftone dot component is performed more effectively, and the character component existing in the halftone dot that cannot be detected by the general edge detection method is also detected. It becomes possible.

In the edge detection circuit 253, for an image relating to the luminance signal Y directly input from the luminance calculation circuit 250, the pixel value of a pixel equal to or less than a predetermined threshold th_edge1 is set to “0”, and the pixel value of the pixel exceeding the threshold th_edge1 is set. An edge detection image is generated by setting the value to “1”. Also,
The image related to the signal that has passed through the edge min direction smoothing circuit 252 is subjected to the above-described Laplacian filter processing to set the pixel value of a pixel equal to or smaller than a predetermined threshold th_edge2 to “0”, and the pixel of the pixel exceeding the threshold th_edge2 An edge detection image is generated by setting the value to “1”. Thus, by preparing an image having two kinds of properties and using an image that does not pass through the edge min direction smoothing unit 252 for characters on a white background, it is possible to detect edges even for finer character details. On the contrary, for a character in a halftone dot, the edge min
By performing edge detection using an image that has passed through the direction smoothing unit 252, only characters and lines can be detected without detecting halftone components.

FIG. 6 is a diagram showing an example of processing for detecting an edge in an image. (A) is an example of an image relating to luminance data Y, and (b) is an example of an edge detection image.

The edge detector 253 further includes a threshold value t
The detection result (edge detection image) by h_edge1 is 7
As a result of expansion using a block size of × 7, 5 × 5, 3 × 3, a result indicating that no edge is detected, a result indicating that no edge is detected, and a three-bit signal EDGE representing the detection result using the threshold th_edge2 in seven codes, Output. Here, the expansion of the signal refers to performing an OR operation on all pixel values in the block.

[Description of Saturation Determining Unit] FIG. 7 is a block diagram showing a detailed configuration example of the saturation determining unit 116. In the saturation determination unit 116, the image signals R40, G40, and B40 output from the image signal switching unit 1064 are input to the R, G, and B terminals of the maximum value detection unit 601 and the minimum value detection unit 602, and the maximum value Max is output. (R, G, B) and minimum value Min
(R, G, B) is extracted.

FIG. 8 shows an LUT (lookup table).
FIG. 603 is a diagram illustrating characteristics of a saturation determination look-up table in 603. The LUT 603 has a maximum value Max
(R, G, B) and the minimum value Min (R, G, B) are the area BK (black), the area GRY (intermediate between color and black), the area CO
It is determined which of L (color) and area W (white) it belongs to, and the result is output as a 2-bit determination signal COL.

[Description of Character Thickness Determination Unit] FIG. 9 is a block diagram showing a detailed configuration example of the character thickness determination unit 114. The image signals R40, G40, and B40 output from the image signal switching unit 1064 are input to the R, G, and B terminals of the minimum value detection unit 2011.

The minimum value detection unit 2011 outputs the image signal R (R
40), the minimum value Min of G (G40) and B (G40)
(R, G, B) is detected and output as a MINRGB signal. Then, the average value detection unit 2022 calculates the average value AV of the MINRGB signals of the target pixel and 5 × 5 pixels in the vicinity thereof.
E5 and average value AVE of MINRGB signals of 3 × 3 pixels
3 is calculated.

The character / line image / halftone detecting unit 2023 determines that the pixel of interest is a character / line image or a part of the halftone area based on the density of the pixel of interest and the density change from the pixel of interest toward its periphery. It is determined whether or not.

FIG. 10 shows a character / line drawing / halftone detection circuit 2
FIG. 23 is a block diagram showing a detailed configuration example of No. 023. letter·
In the line drawing / halftone detection circuit 2023, an appropriate offset value OFST1 is added to AVE3 by the adder 2030, and the value is compared with AVE5 by the comparator 20310.
The comparator 2032 compares the output from the adder 2030 with an appropriate limit value LIM1. Then, output values from the respective comparators are input to the OR circuit 2033.

Output signal BINGRA of OR circuit 2033
Becomes "1" when AVE3 + OFST1> AVE5 or AVE3 + OFST1> LIM1. That is, the character / halftone detection circuit 2023 uses the local area (3 × 3
When there is a density change from the pixel area) to the periphery (5 × 5 pixel area) (the edge portion of the character / line image) or a local area (3 × 3 pixel area) including the pixel of interest , If the density is higher than the specified value
The signal BINGRA indicating that it is a character / line drawing / halftone area becomes "1" in the line drawing and in the halftone portion.

FIG. 11 is a diagram showing a detailed configuration example of the dot area detection unit 2024. In the halftone dot area detection unit 2024, the minimum value detection circuit 2021
Is added to the value of the signal MINRGB generated by
41, an appropriate offset value OFST2 is added, and the result is compared with AVE5 in the comparator 2041. In the comparator 2042, the adder 2040
Is compared with an appropriate limit value LIM2.
The output value from each comparator is input to the OR circuit 2043.

Output signal BINAMI of OR circuit 2043
Becomes “1” when MINRGB + OFST2> AVE5 or MINRGB + OFST2> LIM2. Using the BINAMI signal, the edge direction detection circuit 2044 obtains the edge direction of each pixel.

FIG. 12 is a diagram showing rules for detecting the edge direction in the edge direction detection circuit 2044. It should be noted that the pixel of interest is the pixel at the center of the 3 × 3 pixel group in FIG. When the pixel values of the eight pixels near the target pixel satisfy any of the conditions (0) to (3) shown in FIG. 12, the edge direction detection circuit 2044 sets bits 0 to 0 of the 4-bit edge direction signal DIRAMI. 3
1 ".

The opposing edge detection circuit 2045 at the next stage detects opposing edges in a 5 × 5 pixel area centered on the target pixel. FIG. 13 shows the D of the target pixel.
FIG. 4 is a diagram illustrating a 5 × 5 pixel area in which an IRAMI signal is A33.

Here, as a rule for detecting the opposing edge,
When any of the following (Condition 1) to (Condition 4) is satisfied, a rule is adopted that detects an opposite edge. When detecting the opposite edge, the opposite edge detection circuit 2045 sets the EAAMI signal to “1”. (Condition 1): A11, A21, A31, A41, A5
Bit 0 of any one of A1, A22, A32, A42 and A33 is "1" and A33, A24, A34 and A4
Bit 1 of any one of 4, A15, A25, A35, A45, and A55 is “1”. (Condition 2): A11, A21, A31, A41, A5
1, A22, A32, A42, A33 has bit 1 set to “1” and A33, A24, A34, A4
Bit 0 of any one of A4, A15, A25, A35, A45, and A55 is “1”. (Condition 3): A11, A12, A13, A14, A1
5, A22, A23, A24, or A33 has bit 1 set to “1” and A33, A42, A43, A4
Bit 3 of any one of A4, A51, A52, A53, A54, and A55 is “1”. (Condition 4): A11, A12, A13, A14, A1
Bit 3 of any one of A5, A22, A23, A24, and A33 is “1” and A33, A42, A43, and A4
4, any one of bit 2 of A51, A52, A53, A54, and A55 is “1”.

The expansion circuit 2046 expands the EAAMI signal by 3 pixels × 4 pixels, and if there is a pixel whose EAAMI is “1” in the 3 × 4 pixel area near the target pixel, EAAMI signal is replaced with “1”.

Further, the contraction circuit 2047 and the expansion circuit 20
Using 48, an isolated detection result in a 5 × 5 pixel area is removed to obtain an output signal EBAMI. Here, the contraction circuit 2047 is a circuit that outputs “1” only when all the input signals are “1”.

Next, the counter 2049 counts the number of pixels for which the output signal EBAMI of the expansion circuit 2048 is "1" in a window having an appropriate size, and outputs a count value. Here, 5 including the pixel of interest
A x64 pixel window is employed.

FIG. 14 is a diagram showing an example of this window. In FIG. 14, there are nine sample points in the window in every four pixels in the main scanning direction and a total of 45 points for five lines in the sub-scanning direction. By moving the windows in the main scanning direction for one target pixel, nine windows (1) to (9) in FIG. 14 are prepared. In other words, this means that a region of 5 × 64 pixels centered on the target pixel.

The comparator 2050 includes a counter 204
9 indicates that the EBAMI signal is "" in each window.
The number of "1" is counted, and when the number at which EBAMI becomes "1" exceeds an appropriate threshold value LIM3, a halftone dot area detection signal AMI indicating a halftone dot is set to "1".

The halftone dot image detected as a set of isolated points (isolated white pixels) by the character / line drawing / halftone detection unit 2013 by the halftone dot area detection circuit 2014 is appropriately detected as a halftone dot area. Can be. Characters, line drawings,
A signal BINGRA indicating that a halftone area has been detected and a halftone area signal AMI are subjected to an OR operation in an OR circuit 2025, and as a result, a binary signal PI obtained by binarizing the input image
CT_ISO is generated. More specifically, the binarized signal PICT_ISO is a binary image signal in which a character, a line drawing, a halftone area, and a halftone area are detected and the pixel value is set to “1”, and the other pixels are set to “0”. is there.

However, this binarized signal PICT_
When a character or a line drawing area exists in a halftone dot or a halftone area, in other words, when a character or a line drawing area and a halftone dot or a halftone area are in contact with each other, the boundary portion is defined as an image area (character, As a line drawing, a halftone or a halftone dot area), a case where it can be detected by the character / line drawing / halftone detection unit 2013 or the halftone dot region detection unit 2014 and a case where it cannot be detected are mixed.

As a result, the binarized signal PICT_ISO is partially dotted with an area that has failed to be detected as a character, line drawing, halftone area or halftone area.

Therefore, it is necessary to detect an image area (non-detection area) in which the detection has failed. Redetection unit 20
27 detects a non-detection area, corrects the binarized signal PICT_ISO by this, and corrects the binarized signal PICT_ISO.
(Signals indicating a character, a line drawing, a halftone area, and a halftone dot area).

FIG. 15 is a diagram showing a detailed configuration example of the redetection section 2027. When the target pixel of the binarized signal PICT_ISO is “1” (black), the target pixel has already been detected.
It is set to “1” which is the same as the _ISO signal. On the other hand, the binary signal P
When the pixel of interest in ICT_ISO is “0” (white), the pixel of interest may belong to a non-detection area. Therefore, this pixel of interest is located in an area to be detected as a character / line drawing / halftone area or halftone area. It is determined whether it belongs.

This determination is made based on the distribution of pixels (white pixels) having a pixel value of “0”. In this embodiment, the degree of isolation (the degree to which white pixels are isolated) is checked, and if the degree of isolation is large, it is determined that the area is a non-detection area (area to be detected). Specifically, the number of white pixels connected to the pixel of interest having a pixel value of “0” (the number of white links) is checked. If the number of white links is small, the degree of isolation is high. It is determined that the pixel of interest belongs to the non-detection area (the binarized signal PICT is set to “1”). It is determined that the target pixel does not belong to the non-detection area (the binary signal PICT is set to “0”).

Various rules can be applied as the rules for checking the number of white links. The re-detection unit 2027 determines whether or not the pixels to be inspected around the target pixel (for example, eight surrounding pixels and four pixels at the top, bottom, left, and right) are white pixels. If there is, the same processing is further performed on the inspection pixel around the white pixel, and the number of white pixels obtained by repeating this processing in a chain is the white connection number.

Hereinafter, a description will be given of a method of inspecting the number of white links in the re-detecting section 2027.

The white connection number detecting section 2091 outputs the binarized signal P
The number of white links is detected based on ICT_ISO. The white connection number detection unit 2091 selects the pattern of the inspection target pixel based on the flag signal con_flg supplied from the CPU 102. FIG. 16 shows the flag signal con_flg.
FIG. 6 is a diagram showing an example of a pattern of inspection target pixels selected based on the pattern.

FIG. 16A shows a pattern in which four pixels at the top, bottom, left and right of the center pixel 3002 are to be inspected, and FIG. 16B shows eight pixels around the center pixel 3003 to be inspected pixels. It is a pattern.

First, an example of the operation of the white connection number detection unit 2091 when the pattern shown in FIG. 16A is used will be described with reference to FIG. FIG. 17 shows the binarized signal PICT.
FIG. 3 is a diagram illustrating an example of a binary image related to _ISO. Here, it is assumed that is a white pixel and the other pixels are black pixels, and the target pixel is.

First, the central pixel 3002 of the pattern shown in FIG.
It is determined whether the pixel corresponding to 001 is a white pixel. In this case, pixels are white pixels.

Next, the center pixel 3002 of the pattern shown in FIG. 16A is sequentially matched with the pixel determined to be a white pixel in the previous process, and the pixel corresponding to the inspection target pixel 3001 is a white pixel. It is determined whether or not. in this case,
Pixels and are newly discovered white pixels.

Next, the center pixel 3 of the pattern shown in FIG.
002 are sequentially matched, and it is determined whether or not the pixel corresponding to the inspection target pixel 3001 is a white pixel. In this case, the pixel is a newly discovered white pixel.

Next, the center pixel 3002 of the pattern shown in FIG.
Are sequentially matched, and it is determined whether or not the pixel corresponding to the inspection target pixel 3001 is a white pixel. In this case, there is no newly discovered white pixel. Therefore, the process of detecting the number of connected whites ends.

By this processing, it is detected that the white pixels connected to the target pixel are six pixels of pixel,. Therefore, the number of connected white pixels is 7, and this connected number is input to the next-stage comparator 2092 as a connected number signal CONT.

Next, an example of the operation of the white connection number detection unit 2091 will be described with reference to FIG. 17, using the pattern shown in FIG. 16B as an example.

First, the central pixel 3002 of the pattern shown in FIG.
It is determined whether the pixel corresponding to 001 is a white pixel. In this case, pixels are white pixels.

Next, the central pixel 3002 of the pattern shown in FIG. 16B is sequentially matched with the pixel determined to be a white pixel in the previous process.
It is determined whether the pixel corresponding to 1 is a white pixel. In this case, pixel and are newly discovered white pixels.

Next, the center pixel 3 of the pattern shown in FIG.
002 are sequentially matched, and it is determined whether or not the pixel corresponding to the inspection target pixel 3001 is a white pixel. In this case, there is no newly discovered white pixel. Therefore, the processing for detecting the number of white links is terminated.

By this processing, it is detected that the white pixels connected to the target pixel are the eight pixels from pixel to.
Therefore, the number of connected white pixels is nine, and this connected number is used as the connected number signal CONT in the comparator 2092 in the next stage.
Is input to

FIG. 18 is a flowchart showing the procedure of the inspection process of the number of white links. FIG. 19 is a diagram showing a circuit configuration for executing the procedure shown in FIG. CPU19
10 operates based on the program 1921 which codes the procedure according to the flowchart of FIG. 18, and performs the white connection number inspection process.

When an image related to the binarized signal PICT_ISO is prepared in the image memory 1901, step S18 is performed.
01, the CPU 1910 outputs the flag signal con_
The pattern of the inspection target pixel designated by flg is applied, and it is determined whether each inspection target pixel around the target pixel is a white pixel. In step S1802, the pixel determined to be a white pixel is mapped. Mapping to memory 1902.

In step S1803, step S18
02, the white pixel mapped to the mapping memory 1902 is set as the central pixel 3002, and the image memory 19
In step S 01, it is determined whether the inspection target pixel 3001 located around the center pixel 3002 is a white pixel. In step S 1804, the pixel determined to be a white pixel is mapped to the mapping memory 1902.

In step S1805, step S18
In step 03, it is determined whether a white pixel has been found. If a white pixel has been found, the process returns to step S1803. If a white pixel has not been found, the process proceeds to step S1806.

In step S1806, the number of white pixels mapped in the mapping memory 1902 is counted, and the number of connected white pixels is obtained, and the number of connected white pixels is written to the output port 1930. The connection number signal CONT is output to the comparator 2092.

The comparator 2092 outputs the white connection number signal C
The ONT compares the indicated white connection number with a threshold signal con_th supplied from the CPU 102, and outputs a comparison result signal COMP as a result. Threshold signal con_th
Can be appropriately changed by the CPU 102.

The white connection number signal CONT is equal to the threshold signal con.
_Flg, in other words, when the degree of isolation of the white pixel is higher than a predetermined value, the comparator 2092
Indicates that a non-detection area (area to be detected) was detected. "
On the other hand, if the white connection number signal CONT has a value equal to or greater than the threshold signal con_flg, in other words, if the degree of isolation of the white pixel is equal to or less than a predetermined value, the comparator 2092 determines that the target pixel is not detected. "0" indicating that the area is not an area (area to be detected) is output.

For example, the flag signal con_flg specifies the pattern shown in FIG. 16A as the pattern of the pixel to be inspected, and the binary signal PICT_ISO constituting the binary image shown in FIG. 17 is input. In this case, the white connection number signal CONT indicates “7”. In this case, for example, if the threshold signal con_th is “8”, the comparison result signal C
OMP is “1”, that is, indicates that the target pixel belongs to a non-detection area (area to be detected).

Also, for example, the flag signal con_flg
However, when the pattern shown in FIG. 16B is designated as the pattern of the pixel to be inspected, and the binary signal PICT_ISO shown in FIG.
The white connection number signal CONT indicates “9”. In this case, for example, if the threshold signal con_th is “8”, the comparison result signal COMP indicates “0”, that is, indicates that the target pixel does not belong to the non-detection area (the area to be detected).

The OR circuit 2093 outputs the binarized signal PICT
OR operation of the value of _ISO and the comparison result signal COMP,
The binarized signal PICT after re-detection of the non-detection area is output.

This binarized signal PICT is input to area size determination section 2026 (see FIG. 9).

FIG. 20 is a block diagram showing a detailed configuration example of area size determination section 2026. The area size determination unit 2026 includes a plurality of contraction circuits 2081 and expansion circuits 2
082, and the size of the region referred to by each pair is different from each other. The binarized signal PICT is first input to the contraction circuit group 2081 after being line-delayed according to the size of the contraction circuit. In the illustrated configuration example, 23 ×
It has seven types of contraction circuits from 23 pixel size to 35 × 35 pixel size.

The output of the contraction circuit group 2081 is input to the expansion circuit group 2082 after being line-delayed. In the configuration example shown in FIG.
It has seven types of expansion circuits from 7 × 27 pixel size to 39 × 39 pixel. The expansion circuit group 2082 outputs a 7-bit signal PICT_FH including seven expansion circuits.

If the pixel of interest is a part of a character / line drawing, a signal PICT_FH corresponding to the thickness of the character is generated. FIG. 21 is a diagram for describing an example of a process of generating the signal PICT_FH. The illustrated example shows a case where the image related to the input binarized signal PICT is a band-shaped image having a width of 26 pixels. 27 ×
When a contraction of a size larger than 27 is performed and the expansion is performed by a corresponding expansion circuit, the output of the expansion circuit becomes “0”, and after performing a contraction of a size smaller than 25 × 25,
When this is expanded by the corresponding expansion circuit, the output of the expansion circuit becomes "1". Therefore, bits (0) and (1) of signal PICT_FH are “1”, and bits (2) to (6)
Becomes "0". In this case, the signal PICT_FH is 3
This indicates that the width is 0 pixels.

The encoder 2083 outputs the signal PICT_F
By encoding H, an image area signal ZONE_P indicating the size of the image area to which the pixel of interest belongs is determined.

FIG. 22 is a diagram showing an example of an encoding rule in the encoder 2083. In the illustrated example, a 3-bit signal is used as the image area signal ZONE_P, and the thickness of an image such as a character or a line drawing is indicated in eight stages based on the signal PICT_FH. As shown in the drawing, the signal ZONE_P is "
0 ", and for images such as thickest characters and line drawings,
The signal ZONE_P becomes "7". With this process,
For a photographic image, a halftone image, or the like in which the binary signal PICT is “1” in a wide area, the signal ZONE_P is “
7 "(maximum value), and for an area where the binarized signal is" 1 "is smaller than the maximum value (for example, an image such as a character or a line drawing), the signal is determined according to the thickness of the area. ZO
The value of NE_P is defined.

By the way, when the re-detection unit 2027 is not provided, that is, when the binarized signal PICT_ISO is directly input to the area size determination unit 2026, for example, if a character or a line drawing exists in a halftone dot, The above-mentioned non-detection area is scattered at the boundary between the area and the area such as a character. The binarized signal PICT_ISO constituting the image in which the non-detection areas are scattered as described above is transmitted to the area size determination unit 2.
When the contraction is performed in the contraction circuit 2081 at 026, the plurality of non-detection areas scattered are combined to form a large area, and even after the expansion processing by the next-stage expansion circuit 2082 is performed, the large area remains Remains. Therefore, for example, it is erroneously determined that a thin character area is scattered around a character or the like in the halftone dot area, although the area should be determined to be a large halftone dot area.

On the other hand, as described above, the re-detection section 2027 is provided to re-detect the non-detection area, and the binarized signal PICT_I
The erroneous determination can be avoided by determining the area based on the binarized signal with the corrected SO.

The signal ZONE_P is output to the ZONE correction unit 20.
84 is input. FIG. 23 shows the ZONE correction unit 2084.
3 is a block diagram showing a detailed configuration example of FIG. Signal ZONE
_P is a line delay unit 2111 having a plurality of FIFOs
, And is input to an average value calculation unit 2111 that calculates the average value of the signal ZONE_P in a region of 10 × 10 pixels, for example. The value of the signal ZONE_P is
Thicker characters and line drawings are larger, and thinner characters are smaller. Therefore, the value of the signal ZONE, which is the average of the values of the signal ZONE_P, also increases as the characters and line drawings are thicker.
The thinner, the smaller. It is desirable that the block size used for this correction be determined according to the block size for determining the thickness of characters and the like.

By performing the subsequent processing using the corrected ZONE signal, even in a portion where the thickness of a character or a line suddenly changes, the determination result of the thickness changes smoothly. Deterioration of image quality due to a change in processing can be further improved.

Here, as described above, the signal ZONE is output.
Can be regarded as a halftone area or a halftone area. Therefore, the signal ZONE and the signal EDGE
Characters and lines existing inside halftones and halftone dots
It can be distinguished from characters and lines in other areas (white background).

Next, the process of detecting the edge of an image will be described with reference to FIG.

First, for the binarized signal PICT, 5 ×
The expansion processing is performed by the expansion circuit 2111 of No. 5. With this process, the detection result can be corrected for a halftone dot region in which incomplete detection may be performed.

Next, the output signal of the expansion circuit 2111 is subjected to a contraction process by an 11 × 11 contraction circuit 2112, and the result is output as a signal FCH. The signal FCH generated in this manner is a signal obtained by contracting the signal PICT by three pixels.

FIG. 24 is a diagram showing an example of an algorithm for detecting a character or the like in a halftone or halftone dot area. As described above, since the signal FCH is a signal obtained by contracting the signal PICT, by combining the FCH, the signal ZONE, and the signal EDGE, the edge in the white background is
Edges in the halftone / halftone area can be distinguished,
In the case where a character or the like exists in a halftone dot image, black character processing can be performed without emphasizing a halftone dot component and without processing a portion that does not require black character processing such as an edge of a photograph.

[Explanation of LUT] The LUT 117 includes a character thickness determination unit 114, an edge detection unit 115, and a saturation determination unit 1.
The black character determination signals UCR and FIL are output according to the values of the signals ZONE, FCH, EDGE, and COL, which are the 16 output signals, and the reference table partially shown in FIG.
Determine the values of TER and SEN. These are signals for selecting masking UCR coefficients, spatial filter coefficients, and printer resolution. The value of each signal has the following meaning.

EDGE = 0: not determined as an edge at threshold th_edge1 1: determined as an edge at threshold th_edge1 (no expansion) 2: determined as an edge at threshold th_edge1 (3 × 3 expansion) 3: determined as an edge at threshold th_edge1 (5 × 5 expansion) 4: Judgment as an edge with threshold th_edge1 (7 × 7 expansion) 5: Not judged as an edge with threshold th_edge2 6: Judgment as an edge with threshold th_edge2 (no expansion) FCH = 0: Image edge 1: Image No edge UCR = 0: more black 1: ・ 2: ・ ・ 6: ・ 7: less black FILTER = 0: smoothing 1: strong edge enhancement 2: middle edge enhancement 3: weak edge enhancement SEN = 0: 200 lines 1: As described above, according to this embodiment, the re-detection unit 20
27, the non-detection area is re-detected, and the binarized signal PICT_ISO is corrected, so that the area to which the pixel of interest belongs (for example, a halftone area or a halftone area)
Can be correctly determined. Therefore, a high-quality output image can be formed.

In the above embodiment, the re-detection unit 2
027, the binary signal PICT_ISO is used.
The signal BINGRA, which is the output of 23, may be used. In this case, the output of the re-detection unit 2027 and the halftone dot region detection unit 20
The logical sum of the input data and the data of the input data 24 may be input to the area size determination unit 2026. Further, as an input of the re-detection unit 2027, a signal AMI output from the halftone dot region detection unit 2024 may be used. In this case, the logical sum of the output of the re-detection unit 2027 and the output of the character / line drawing / extraction completion detection unit 2023 may be input to the area size determination unit 2026.

The present invention can be applied to a system composed of a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), but it can be applied to a single device (for example, a copier, Facsimile machine, etc.).

Further, an object of the present invention is to provide a storage medium storing a program code of software for realizing the functions of the above-described embodiments to a system or an apparatus, and to provide a computer (or CPU) of the system or apparatus.
And MPU) read and execute the program code stored in the storage medium.

In this case, the program code itself read from the storage medium implements the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention.

As a storage medium for supplying the program code, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD
-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

When the computer executes the readout program code, not only the functions of the above-described embodiment are realized, but also the OS (Operating System) running on the computer based on the instruction of the program code. ) May perform some or all of the actual processing, and the processing may realize the functions of the above-described embodiments.

Further, after the program code read from the storage medium is written into a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, based on the instruction of the program code, It goes without saying that the CPU included in the function expansion board or the function expansion unit performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

[0132]

According to the present invention, by detecting a boundary region between a character / line drawing region and another image region in an image,
An image obtained by the processing based on the detected area can be improved in quality.

[0133]

[Brief description of the drawings]

FIG. 1 is a diagram showing a cross-sectional configuration of an image processing apparatus according to a preferred embodiment of the present invention.

FIG. 2 is a diagram illustrating a detailed configuration example of an image processing unit of the image scanner unit.

FIG. 3 is a block diagram illustrating a configuration example of an edge detection unit.

FIG. 4 is a block diagram illustrating a detailed configuration example of a luminance calculation circuit.

FIG. 5 is a diagram for explaining a function of an edge min direction detection circuit.

FIG. 6 is a diagram illustrating an example of an edge detection process in an image;

FIG. 7 is a block diagram illustrating a detailed configuration example of a saturation determination unit.

FIG. 8 is a diagram showing characteristics of a look-up table for determining saturation in an LUT (look-up table).

FIG. 9 is a block diagram illustrating a detailed configuration example of a character thickness determination unit.

FIG. 10 is a block diagram illustrating a detailed configuration example of a character / halftone detection circuit.

FIG. 11 is a diagram illustrating a detailed configuration example of a halftone dot area detection unit.

FIG. 12 is a diagram showing a rule of edge direction detection in an edge direction detection circuit.

FIG. 13 is a diagram showing a 5 × 5 pixel area in which a DIRAMI signal of a target pixel is A33.

FIG. 14 is a diagram illustrating an example of a window to be counted.

FIG. 15 is a diagram illustrating a detailed configuration example of a re-detection unit.

FIG. 16 is a diagram showing a pattern of an inspection target pixel selected based on a flag signal con_flg.

FIG. 17 is a diagram illustrating an example of a binary image of a binary signal PICT_ISO.

FIG. 18 is a flowchart illustrating a procedure of a white connectivity degree inspection process.

FIG. 19 is a diagram showing a circuit configuration for executing the procedure shown in FIG. 18;

FIG. 20 is a block diagram illustrating a detailed configuration example of an area size determination unit.

FIG. 21 is a diagram illustrating an example of a process of generating a signal PICT_FH.

FIG. 22 is a diagram illustrating an example of an encoding rule in an encoder.

FIG. 23 is a block diagram illustrating a detailed configuration example of a ZONE correction unit.

FIG. 24 is a diagram illustrating an example of an algorithm for detecting a character or the like in a halftone or halftone dot area.

FIG. 25 is a diagram illustrating an example of a reference table in an LUT of a black character determination unit.

Continued on the front page (72) Inventor Yuki Uchida 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon Inc.

Claims (23)

[Claims] 1. An image processing apparatus for extracting features of an image and processing the image based on the result, comprising: first detection means for detecting a character, a line drawing, and a halftone area in the image; A second detecting means for detecting a halftone dot area in the image, and a third detecting means for detecting a boundary area between the character / line drawing / halftone area and the halftone dot area in the image. An image processing apparatus for processing an image based on a detection result by a third detection unit. 2. The third detecting means detects a boundary area between a character / line drawing / halftone area and a halftone area in an image from an area not detected by the first and second detecting means. The image processing apparatus according to claim 1, wherein: 3. The apparatus according to claim 3, wherein each of the pixels includes a character / line drawing / halftone area and a halftone area in the image based on a distribution of pixels not detected by the first and second detection means. 2. The image processing apparatus according to claim 1, wherein the boundary area is detected by determining whether or not the pixel belongs to the boundary area with the pixel. 4. The image processing apparatus according to claim 1, wherein each of the pixels includes a character / line image / halftone area and a halftone area in the image based on a series of pixels not detected by the first and second detection means. 2. The image processing apparatus according to claim 1, wherein the boundary area is detected by determining whether or not the pixel belongs to the boundary area with the pixel. 5. The image processing apparatus according to claim 3, wherein each of the pixels includes a character, a line drawing, and a halftone area in the image based on the number of consecutive pixels not detected by the first and second detection means. The image processing apparatus according to claim 1, wherein the boundary area is detected by determining whether or not the pixel belongs to a boundary area with the halftone dot area. 6. The image processing apparatus according to claim 1, wherein the third detecting means determines a character / line drawing / halftone area and a halftone area in the image from an image formed by a set of detection results of the first or second detecting means. The image processing apparatus according to claim 1, wherein a boundary area is detected. 7. The image processing apparatus according to claim 1, wherein each of the pixels detects a character or a character in the image based on a distribution of undetected pixels in an image formed by a set of detection results of the first or second detection means. 7. The image processing apparatus according to claim 6, wherein the boundary area is detected by determining whether or not the pixel belongs to a boundary area between the line drawing / halftone area and the halftone dot area. 8. The image processing apparatus according to claim 1, wherein each of the pixels detects a character or a character in the image based on a series of undetected pixels in an image formed by a set of detection results of the first or second detection means. 7. The image processing apparatus according to claim 6, wherein the boundary area is detected by determining whether or not the pixel belongs to a boundary area between the line drawing / halftone area and the halftone dot area. 9. The image processing apparatus according to claim 5, wherein each of the pixels is included in the image based on the number of undetected pixels in the image formed by the set of detection results of the first or second detection means. The image processing apparatus according to claim 6, wherein the boundary area is detected by determining whether or not the pixel belongs to a boundary area between the character / line drawing / halftone area and the halftone area. 10. The image processing apparatus according to claim 1, further comprising: a determination unit configured to determine a size of an image area based on a detection result of the first to third detection units, and processing the image according to the determination result. The image processing apparatus according to claim 1. 11. An image processing apparatus for extracting a feature of an image and processing the image based on the result, wherein a character / line drawing area in the image and an image area other than the character / line drawing area are detected. Main detecting means, and a sub-detecting means for detecting a boundary area between the character / line drawing area and the other image area, and processes an image based on a detection result by the main detecting means and the sub-detecting means. An image processing apparatus characterized by the above-mentioned. 12. The image processing apparatus according to claim 11, wherein the sub-detection unit detects the boundary region from a region not detected by the main detection unit. 13. The sub-detection unit detects the boundary region by determining whether each pixel belongs to the boundary region based on a distribution of pixels not detected by the main detection unit. The image processing apparatus according to claim 11, wherein: 14. The sub-detection unit detects the boundary region by determining whether each pixel belongs to the boundary region based on a series of pixels not detected by the main detection unit. The image processing apparatus according to claim 11, wherein: 15. The sub-detecting unit determines whether each pixel is a pixel belonging to the boundary area based on the number of consecutive pixels not detected by the main detecting unit, thereby determining the boundary. The image processing apparatus according to claim 11, wherein an area is detected. 16. The image processing apparatus according to claim 1, further comprising an output unit configured to output the processed image. 17. The image processing apparatus according to claim 16, further comprising input means for inputting an image to be processed. 18. The image processing apparatus according to claim 17, wherein said input means includes means for reading and inputting a document image. 19. The method according to claim 1, wherein processing is performed on a full-color image.
An image processing apparatus according to the item. 20. An image processing method for extracting features of an image and processing the image based on the result, comprising: a first detection step of detecting a character, a line drawing, and a halftone area in the image; A second detection step of detecting a halftone dot area in the image, and a third detection step of detecting a boundary area between the character / line drawing area and the halftone dot area in the image. An image processing method, wherein an image is processed based on a detection result in a detection step. 21. An image processing method for extracting features of an image and processing the image based on the result, wherein a character / line drawing area in the image and an image area other than the character / line drawing area are detected. A main detection step, and a sub-detection step of detecting a boundary area between the character / line drawing area and the other image area, and processes an image based on a detection result by the main detection unit and the sub-detection step. An image processing method comprising: 22. A computer readable memory containing a program code for extracting a feature of an image and processing the image based on the result, wherein the computer readable memory detects a character, a line drawing, and a halftone area in the image. A code of a first detection step, a code of a second detection step of detecting a halftone dot area in an image, and a code of a third detection step of detecting a boundary area between a character / line drawing area and a halftone dot area in the image. And a computer readable memory comprising: 23. A computer readable memory storing a program code for extracting a feature of an image and processing the image based on the result, wherein a character / line drawing area in the image and a part other than the character / line drawing area are included. A computer-readable memory comprising: a code for a main detection step of detecting another image area; and a code of a sub-detection step for detecting a boundary area between the character / line drawing area and the other image area. .