JP2002281302A - Image area separating device, image area separating method, recording medium, image processor, image processing method and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image area separating device to perform a processing by attaching importance to characters in the case of monochrome reproduction, however, not to degrade the quality of a reproduced image even in the case of a pattern, especially dot processing. SOLUTION: The image area separating device to detect a character area from inputted image data is provided with at least one of a discrimination performance control part 10 to control discrimination based on the kind of the reproduced image and a judgment processing part to judge image characteristics due to an area in the inputted image data. An edge area detecting part 20 to detect the edge area is provided as the judgment processing part and a judgment parameter in the edge area detecting part is controlled in the direction to detect an edge more when the inputted image is reproduced in a monochrome image in the discrimination performance control part 10.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image area separating apparatus, an image area separating method and a recording medium, an image processing apparatus, an image processing method and a recording medium for improving the quality of an output image with respect to digital image processing. It is suitable for application to image processing of digital image data obtained through a digital copying machine or a network.

[0002]

2. Description of the Related Art For example, there is a technique described in Japanese Patent Publication No. 7-8007 for improving image quality of an output image with respect to digital image processing. The method according to this technique detects a black character area from two means, a means for detecting a character edge using pixel continuity and a means for determining a black portion, and performs processing for character processing on the detected area. This is to improve the image quality. In such a character edge detecting means, the tendency is particularly strong in halftone dot printing, but there are many areas that are erroneously detected as character areas in a picture, and this erroneous detection causes deterioration in image quality. In order to reduce this problem, a paper “Image area separation method for mixed images of characters / pictures (dots, photographs)” (IEICE Transactions Vol.J75-DIINo.1
pp.39-47 Jan. 1992) and Japanese Patent No. 3004255 introduce a mechanism that detects a halftone dot region in addition to detecting an edge region and does not judge a halftone dot pattern as a character edge. are doing. Further, in the techniques disclosed in JP-A-4-14378 and JP-A-7-203198, by limiting the detected characters to only thin characters, it is possible to reduce erroneous detection as a character area in a picture. I'm sorry.

[0003]

Generally, when an image input by a color copier is reproduced in color or monochrome, when "character / photo mode" is selected,
In particular, regarding the character edge determination, the detection determination is performed using the same architecture and the same determination parameters. Conventionally, in detecting a character edge, a part having a large contrast is locally detected, and depending on the reading accuracy of a scanner, a part of a dark character edge cannot be detected as a character edge due to deterioration of MTF or the like. In some cases, the character is not detected as a character edge when the character is originally thin, as represented by. As a result, the image quality of the characters often deteriorates, or the characters become difficult to read. For this reason, when performing monochrome reproduction with a color copying machine, no special measures have been taken even though there are many users who attach particular importance to characters.

An object of the present invention is to solve the above-mentioned problem. The image area separation apparatus performs character-oriented processing at the time of monochrome reproduction, but does not degrade the reproduction image quality even in the case of pattern processing, particularly halftone processing. A first object is to provide an image area separation method and a recording medium, an image processing device, an image processing method, and a recording medium. It is a second object of the present invention to provide an inexpensive image area separating apparatus, an image area separating method and a recording medium, an image processing apparatus, an image processing method and a recording medium with a reduced hardware scale. In color reproduction, the accuracy of character detection can be somewhat improved by black determination as disclosed in Japanese Patent Publication No. 7-8007, but in monochrome reproduction, only a part of color information or brightness Only information has been used. Behind such a configuration, the spread of color reproduction machines has led to the separation of "color of image quality" and "monochrome of speed". Also, adopting the same processing system will reduce costs. This is because they also had a purpose. For this reason, when performing monochrome reproduction with a color copying machine, this is mainly for character documents, while when reproducing photographic images in monochrome, important color information and gradation characteristics are lost due to monochrome conversion. Because
Dare to play in black and white has become extremely rare. However, for monochrome playback, it is necessary to secure processing speed and reduce costs.
It is becoming very difficult to separately provide a special processing system dedicated to monochrome. Therefore, during monochrome playback,
Despite the fact that most are text documents, the default setting of “Text / Photo mode” is a “Color reproduction process” that is adjusted to obtain average image quality for both photos and text. It has been diverted. Along with the above-mentioned reason, the incompleteness of the conventional character edge detection is a factor that greatly reduces the character quality during monochrome reproduction. The present invention has been made to solve the above-described problem, and does not cause a reduction in processing speed by using complicated processing or an increase in cost due to the addition of a special processing system. A third object is to provide an image processing device, an image processing method, and a recording medium that improve image quality, that is, character quality during monochrome reproduction.

[0005]

According to a first aspect of the present invention, there is provided an image area separating apparatus for detecting a character area from input image data. Edge area detection that includes a discrimination performance control unit that controls the judgment performance based on the type of image data, and at least one judgment processing unit that judges an image feature for an area in the input image data, and detects an edge area as the judgment processing unit Wherein the discrimination performance control section controls a judgment parameter in the edge area detection section in a direction to detect an edge more when an input image is reproduced as a monochrome image. Further, the image area separating apparatus according to claim 2 of the present invention is an image area separating apparatus that detects a character area from input image data, wherein a discrimination performance control unit that controls a judgment performance based on a type of a reproduced image; The image processing apparatus includes at least two determination processing units that determine an image feature of an area in the input image data, and includes an edge area detection unit that detects an edge area and a halftone dot area detection unit that detects a halftone dot area as the determination processing unit. When the input image is reproduced as a monochrome image, the discrimination performance control unit does not pass through the edge region detection unit, and controls a determination parameter in the halftone dot region detection unit that detects a halftone dot. It is characterized by the following. Further, the image area separating method according to claim 3 of the present invention,
In an image area separation method for detecting a character area from input image data, a determination performance control step of controlling determination performance based on a type of a reproduced image, and an edge area detection step of detecting an edge area in the input image data. Having a halftone dot detection step of detecting a halftone dot area in the input image data, wherein the determination performance control step, when reproducing the input image as a monochrome image, without passing through the edge area detection step, A decision parameter in the halftone dot detection step for detecting a halftone dot is controlled, or a decision parameter in the edge area detection step is controlled in a direction to detect an edge more.

According to a fourth aspect of the present invention, there is provided a computer-readable recording medium storing a program for causing a computer to function as an image area separating apparatus for detecting a character area from input image data. A determination performance control unit that controls determination performance based on a type of a reproduced image; and at least two determination processing units that determine an image feature for an area in the input image data, wherein an edge area is used as the determination processing unit. Having an edge area detection unit for detecting the halftone dot area and a halftone dot area detection unit for detecting the halftone area, the discrimination performance control unit, when reproducing the input image as a monochrome image, without passing through the edge area detection unit Controlling the determination parameter in the halftone dot area detecting section for detecting a halftone dot, or controlling the edge area detecting section in a direction for detecting an edge more. It was recorded image area separation processing program so as to control the decision parameter. An image processing apparatus according to claim 5 of the present invention is an image processing apparatus that reproduces input image data as a color image or a monochrome image. When an input image is reproduced as a monochrome image, an edge area is not detected. An image area separating section for detecting a halftone dot area, and a character correcting section for performing a correction for enhancing the contrast of the non-halftone area detected by the image area separating section are provided. In the image processing apparatus according to a sixth aspect of the present invention, in the image processing apparatus according to the fifth aspect, the character correction section highlights a non-dot area separated by the image area separation section. It is characterized in that gamma correction with a steep slope is performed so that a portion and a shadow portion are emphasized. An image processing apparatus according to a seventh aspect of the present invention is an image processing apparatus that reproduces input image data as a color image or a monochrome image. An image area separation unit that detects a halftone area, and for a non-halftone area detected by the image area separation unit, a lower threshold is set when performing conversion from multi-valued data to small value data. And a halftone processing unit that performs threshold control so as to increase the contrast. An image processing apparatus according to an eighth aspect of the present invention is the image processing apparatus according to the seventh aspect, wherein the halftone processing unit uses an error diffusion processing method.

According to a ninth aspect of the present invention, in the image processing apparatus of the eighth aspect, the halftone processing section includes a halftone dot area separated by the image area separation section and a non-halftone area. It is characterized in that different error correction values are used for the point region. In the image processing method according to claim 10 of the present invention, in the image processing method for reproducing the input image data as a color image or a monochrome image, when the input image is reproduced as a monochrome image, the edge area is not detected. The image processing apparatus further comprises an image area separating step of detecting a halftone dot area, and a character correcting step of performing correction for enhancing the contrast of the non-dot area detected in the image area separating step. In the image processing method according to claim 11 of the present invention, in the image processing method for reproducing the input image data as a color image or a monochrome image, the edge area is not detected when the input image is reproduced as a monochrome image. An image area separation step of detecting a halftone dot area, and for a non-halftone area detected in the image area separation step, a lower threshold value is set when performing conversion from multi-valued data to small value data. And a halftone processing step of performing threshold control so as to increase the contrast. A recording medium according to a twelfth aspect of the present invention is a computer-readable recording medium that stores a program for causing a computer to function as an image processing apparatus that reproduces input image data as a color image or a monochrome image. ,
When the input image is reproduced as a monochrome image, the edge area is not detected, and the image area separation unit that detects the halftone area, and the non-halftone area detected by the image area separation unit,
An image processing program functioning as a character correction unit for performing correction for enhancing contrast was recorded. A recording medium according to a thirteenth aspect of the present invention is a computer-readable recording medium storing a program for causing a computer to function as an image processing apparatus that reproduces input image data as a color image or a monochrome image. ,
When the input image is reproduced as a monochrome image, the edge area is not detected, and the image area separation unit that detects the halftone area, and the non-halftone area detected by the image area separation unit,
An image processing program functioning as a halftone processing unit for setting a threshold value for performing conversion from multi-valued data to small-valued data lower and controlling the threshold value so as to increase contrast was recorded.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The construction and operation of an embodiment of the present invention will be described below in detail with reference to the drawings. <First Embodiment (Image Area Separation Apparatus)> FIG. 1 is a block diagram showing a configuration of a part of an image area separation processing circuit in an image processing apparatus (for example, a color copying machine). In FIG. 1, the image area separating apparatus according to the first embodiment determines a determination parameter for edge detection from a type (a color image or a monochrome image) when an image of an input signal is reproduced by an image processing apparatus. The control unit 10, an edge region detection unit 20 that detects an edge region in an image of an input signal using a discrimination parameter from the discrimination performance control unit 10, and a halftone dot that detects a halftone region in an image of an input signal. It is composed of an area detection section 30 and a comprehensive determination section 40 that comprehensively determines whether the area is a character area based on the results of the edge area detection section 20 and the halftone dot area detection section 30. Further, the edge region detection unit 20 includes an edge enhancement unit 22 that is a pre-process of the determination process on the input image signal, a black pixel continuity detection unit 24 that detects local black pixel continuity, A white pixel continuity detecting unit 26 for detecting the continuity of white pixels, and a judging unit 28 for judging whether an area is an edge area based on the results of the black pixel continuity detecting unit 24 and the white pixel continuity detecting unit 26. Have been. The operation of determining the character area in the image area separating apparatus configured as described above will be described.

The input image signal is data obtained from, for example, a color scanner using a CCD element or the like. The input image signal is read at a resolution of 600 dpi and subjected to A / D conversion processing.
A 6-bit, substantially linear image signal is used for each pixel. Here, a green signal of the three primary colors is used, but a predetermined signal, for example, ma
x (R, G, B) or the like may be used as an input signal. It is clear that the response to the color character is improved with the max (R, G, B) signal. Also, here, for the sake of explanation, it is assumed that "0" represents white of the document and "63" represents black of the document as the meaning of the signal. The input signal may be image data obtained through a network. First, an operation when the edge area detection unit 20 detects whether the input image data is an edge area will be described. The edge emphasizing unit 22 uses, for example, a differential filter having coefficients as shown in FIG. Performs filtering to sharpen the image. For the image data processed by the edge enhancement unit 22, the black pixel continuity detection unit 24 detects a black pixel at the threshold set by the discrimination performance control unit 10, and then sets, for example, the target pixel as the center pixel. When the black / non-black pattern in the 3 × 3 matrix matches one of the patterns as shown in FIG. 3, the target pixel is assumed to be black continuous. The discrimination performance control unit 10 sets a threshold (th) when the input image is reproduced by the image processing apparatus when the image is color.
B_color) and the threshold value in the case of monochrome (thB_mon)
o) is set to have the following relationship. thB_color> thB_mono When a monochrome image is reproduced as described above, by setting a small threshold value, a signal having a lighter density can be detected as a black pixel. Further, at the time of color reproduction, since color reproducibility and gradation reproducibility are emphasized, there is a limit in setting a low threshold value. However, during monochrome playback,
In particular, when emphasis is placed on character reproducibility, even for a low-contrast character such as a pencil, it must be given priority to be reliably determined as a character. In addition, the reproducibility of low-contrast characters during monochrome reproduction can be improved. For the image data processed by the edge enhancement unit 22, the white pixel continuity detection unit 26 detects a white pixel with the threshold set by the discrimination performance control unit 10, and then sets, for example, the target pixel as the center pixel. When the white / non-white pattern in the 3 × 3 matrix matches one of the patterns as shown in FIG. 3, the target pixel is assumed to be white continuous. The determination unit 28 performs pattern matching between the black pixel continuity detection unit 24 and the white pixel continuity detection unit 26 within a block of 5 × 5 pixels in order to detect several pixels from the boundary between the character and the background as an edge region. If there is at least one detected black continuous pixel and white continuous pixel, the block of interest is determined to be an edge region, otherwise, it is determined to be a non-edge region.

FIG. 4 is a part of a flowchart showing the flow of the processing of the edge area detecting section 20. When a threshold value is determined for a black pixel and a white pixel in a later process with respect to an input image signal (for example, data obtained from a color scanner using a CCD device or image data obtained through a network), a change in the scanner may occur. For example, in order to obtain an effect that does not fluctuate, a filtering process is performed with a differential filter of coefficients as shown in FIG. 2 to sharpen an image (step S100). It is determined whether the image when the input image signal is reproduced by the image processing apparatus is color or monochrome (step S110). At the time of color image reproduction, a threshold value for color reproduction (thB_col
or) is set (step S120). At the time of monochrome image reproduction, a threshold value (thB_mono) for monochrome reproduction is set (step S130). It is assumed that these thresholds have the following relationship. thB_color> thB_mono Step S is performed on the image data on which the edge enhancement process has been performed.
After detecting a black pixel with the threshold value set in step S120 or step S130, for example, 3
When the black / non-black pattern in the × 3 matrix matches any one of the patterns as shown in FIG. 3, it is assumed that the target pixel is continuous black (step S140). Step S120 is performed on the edge-enhanced image data.
Alternatively, after detecting a white pixel with the threshold value set in step S130, for example, a white / non-white pattern in a 3 × 3 matrix with the target pixel as the center pixel matches one of the patterns shown in FIG. At this time, the target pixel is assumed to be white continuous (step S150).

In order to detect several pixels from the boundary between the character and the background as an edge area, in the block of 5 × 5 pixels, both the black continuous pixels and the white continuous pixels detected by the pattern matching in steps S140 and S150 are used. It is determined whether there is at least one of them (step S
160). When there is at least one block, the block of interest is set as an edge area (step S180), otherwise, it is determined as a non-edge area (step S170).
Further, the halftone dot area detection unit 30 operates in parallel with the processing of the edge area detection unit 20, and detects whether or not the input image data is a pixel block belonging to the halftone dot area. As a method of detecting the halftone dot area, Japanese Patent Laid-Open No. 2-115988
Halftone dot separation method using "extreme pixel" described in
Alternatively, a paper “An Image Area Separation Method for Mixed Images of Text / Pattern (dots, photos)” proposed by the present applicant (IEICE Transactions Vol.J75-DIINo.1 pp.39-47 1992 The technology of halftone dot detection by “peak pixel” detection described in January) is used. Finally, when the output from the edge area detecting section 20 is an edge area and the output from the halftone area detecting section 30 is a non-halftone area, the comprehensive determining section 40 determines that the area is a character area. Is output as a picture area. In general, the user's preference is that the image quality in the image is strict in color reproduction (in other words, the character image quality is low), and that the image quality is strict in monochrome reproduction (in other words, the image quality in the image). Hope to treat). This tendency is stronger when the original is color. Based on the background described above, by performing control such that image area separation processing often used when performing adaptive processing is determined based on a pattern during color reproduction and based on characters during monochrome reproduction, as a result, It is possible to reflect the user's preference on the reproduced image.

<Modification of First Embodiment (Image Area Separation Device)> In a modification of the first embodiment, when an input image is reproduced as a monochrome image, the edge area detection unit 20 is set to the through state (not operated). , The entire area is regarded as an edge area, and the character / non-character determination is performed only by the result of the halftone dot area detection unit 30. The advantage of this configuration is that the low-contrast character can be determined as a character, as well as the inside of the character, which has conventionally been determined to be a pattern, which may cause a thinning failure (see FIG. 5). That is, it can be determined. On the other hand, a super-highlight portion or a super-dark portion where the undulation of the halftone dot is small is not detected as a halftone dot, and may be determined as a result character region.
Therefore, when the edge area detection unit 20 is set to the through state, control is performed so that the halftone dot detection unit 30 lowers the threshold for detecting halftone dots. For example, in the above-mentioned paper “Image area separation method for mixed images of characters / patterns (dots and photographs)”, (A) the density level (L) of the central pixel is higher than the density levels of all surrounding pixels, or Low. (B) Density levels of diagonal pixels (a,
b) has the following relationship with all four pairs. | 2 × Lab−> TH When both (A) and (B) are satisfied at the same time, the center pixel is detected as the peak pixel. In the case of the modification of the first embodiment, the discrimination performance control unit 10 sets the threshold TH at this time to be low. As a result, since the halftone dot region detection process is always performed, the image quality of the halftone dot document (for example, occurrence of moiré) does not occur.

<Second Embodiment (Image Processing Apparatus)> Conventionally, since a photographic image belonging to a non-dot image is also considered at the same time, an edge is detected in a very narrow range and an edge is erroneously detected in the photographic image. To prevent them from doing so. for that reason,
In the case of a large character or a bold line used for a heading or the like, the center part leaks from the edge area and is processed as a picture, so that only the outline of the character is detected as an edge as shown in FIG. There is a problem that the central portion is detected as a picture. Further, in order to secure the gradation in a photographic image, the threshold value (thB_mono) for detecting the black pixel described above cannot be set extremely low, and many low-contrast characters are lost. Therefore, the image processing apparatus according to the second embodiment, as shown in FIG.
An image area separation unit 100 that performs the image area separation processing described in the first embodiment on the input image signal, a character correction unit 110 that performs a character correction on the non-halftone area that has been subjected to the image area separation, The above-mentioned problem is solved by comprising a halftone processing unit 120 that performs halftone processing on the image data and outputs an output image signal. In such a configuration, when reproducing the input signal as a monochrome image, the image area separation unit 100 does not operate the edge area detection unit 20 for separating the character and the picture (see a modification of the first embodiment). ), Only the halftone dot area detection section 30 is operated. That is, when the non-dot area is reproduced as a monochrome image, the entire area is regarded as a character area, and the correction for characters is applied to the entire non-dot area, thereby preventing a deterioration in character quality due to erroneous detection as shown in FIG. In addition, it is possible to speed up the processing. The character correction unit 110 enhances the contrast of the non-dot area detected by the image area separation unit 100 during monochrome reproduction, thereby improving the visibility of the entire character including low-contrast characters. . This is because, in the γ correction performed before the halftone processing, a plurality of γ correction tables or γ determination means are prepared, and a steeper slope is selected so that a highlight portion and a shadow portion are emphasized from among them. This can be achieved by choosing one with characteristics. Normally, in consideration of the MTF characteristics of the scanner unit and the output unit, FIG. 8 (an example of a γ characteristic curve)
Is selected, but by selecting a high γ curve (or a high γ straight line) as shown in FIG. 8, character reproducibility can be improved.

Similarly, the halftone processing section 120 prepares a plurality of threshold values in the halftone processing at the time of monochrome reproduction, and converts the non-halftone area detected by the image area separating section 100 from the multivalued data to the small value. A similar effect can be obtained by setting a lower threshold value when performing conversion into data (or binary data). If the error diffusion method is used for the halftone processing, the contrast can be adjusted not only by the threshold value and the γ correction but also by the error correction value. That is, in the error diffusion process, in monochrome reproduction, the error value when the dot is ON is ERORRxy, the input data is DATAxy, and the error correction value VALUE when ERORRxy = DATAxy-VALUE is the same as the normal dot area. Therefore, a MAX value of a reproducible gradation level (255 for 256 gradations) is used, but by setting a smaller value for a non-halftone dot area, the next dot is easily generated, and as a result, As a result, the contrast can be improved. As described above, it is stated that attention is paid to character quality from the usage pattern when reproducing a color image as a monochrome image, and the reproduction of a picture is not particularly mentioned.
Here, as shown in FIG. 9, if the result of performing image area separation on a document belongs to a non-dot image, it can be processed as a character / line image or reproduced as a monochrome image. It is considered that the quality is greatly reduced due to this. However, since there is an absolute difference in the amount of information that can be handled between color and monochrome, it is impossible to prevent a decrease in quality, and a means that can faithfully reproduce the original as a user's psychology (in this case, color reproduction) Means that the quality is significantly reduced while there is (monochrome playback)
It is considered that there is no particular problem because it is considered very rare. Conversely, characters and line drawings are rarely recognized as halftone images, and general monochrome prints, such as photographs of documents, are reproduced with halftone dots.
By making the correction for the halftone dot image portion (any of the above-mentioned corrections can be diverted) with emphasis on gradation, it is possible to sufficiently guarantee the quality even for a picture. Furthermore, since the above-described correction methods can be realized by means that does not impose a burden on the processing system such as switching tables, it is possible to improve image quality without lowering the processing speed.

<Examples by Computer> Further, the present invention is not limited to the above embodiments.
Each function constituting the image area separation device or the image processing device that realizes the function of the above-described embodiment is programmed,
The program is previously written on a recording medium such as a CD-ROM, and the CD-ROM is mounted on a computer equipped with a medium drive device such as a CD-ROM drive, and these programs are stored in a memory or a storage device of each computer. By storing and executing it, the object of the present invention can be achieved. The recording medium may be any of a semiconductor medium (for example, ROM, IC memory card, etc.), an optical medium (for example, DVD, MO, MD, CD-R, etc.) and a magnetic medium (for example, magnetic tape, flexible disk, etc.). It may be. When the program for realizing the above-described embodiment is a semiconductor recording medium such as a ROM or the like, the program is directly loaded into the memory and executed, not from the medium driving device. Further, not only the functions of the above-described embodiments are realized by executing the loaded program, but also the operating system or the like performs part or all of the actual processing based on the instructions of the program, and the processing performs The case where the functions of the above-described embodiments are realized is also included. Further, a program for realizing the functions of the above-described embodiment is loaded into a memory provided in a function expansion board or a function expansion unit,
Based on the instructions of the program, the CPU or the like provided 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. .

<Operation in Network Environment of the Present Invention> A mode in which the present invention is connected to a network and operated may be adopted. For example, a server computer holding an image area separation processing program or an image processing program and a user input an image and output a result (printer, display,
(Including output to a storage device and transmission to another terminal). This network is a transmission path for coupling a server computer and a user terminal, and is generally realized by a cable, and TCP / IP is used as a communication protocol. However, the transmission path is not limited to the cable, but may be any of wireless, wired and broadcast waves as long as the communication protocol between them is the same. For example, LAN (Local Area Net)
work), WAN (Wide Area Network), Internet, analog telephone network, digital telephone network (ISDN: Inte
gral Service Digital Network), PHS (Personal Handy Phone System), mobile phone network, satellite communication network, etc. can be used. The user terminal receives the image area separation processing program and the image processing program from the server computer, inputs an image, executes the received program, and outputs the execution result.
This has the advantage that the program is always up to date. When the server computer and the user's terminal are connected via a network, a program for realizing the functions of the present invention is stored in a storage device such as a magnetic disk of the server computer and distributed to the terminal in a form such as download. It is also possible. Further, the program for realizing the functions of the present invention may be provided by distributing it by broadcast waves.

[0017]

As described above, according to the present invention,
When performing the image area separation process on the input color image, it is determined that the pattern is emphasized when reproducing the color image, and the character is emphasized when reproducing the monochrome image. It is possible to obtain a reproduction quality in which the user's preference is reflected in the image. Also, "character"
By turning off (through) the edge area detection processing for judging / correcting whether the character is a “picture” or not, it is possible to improve the processing speed and to prevent the deterioration of the character quality due to erroneous detection of the character edge. It has become possible.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a first exemplary embodiment of the present invention.

FIG. 2 is a diagram illustrating an example of coefficients of an edge enhancement filter.

FIG. 3 is a diagram illustrating an example of a 3 × 3 pattern matrix for determining pixel connectivity;

FIG. 4 is a flowchart illustrating a process performed by an edge area detection unit according to the first embodiment.

FIG. 5 is a diagram illustrating a change in an output image according to a change in edge processing.

FIG. 6 is a diagram illustrating an example of a case where edge detection is performed incompletely.

FIG. 7 is a block diagram illustrating a configuration of a second exemplary embodiment of the present invention.

FIG. 8 is a diagram illustrating an example of a γ characteristic curve.

FIG. 9 is a diagram for explaining a relationship between a document to be input and characteristics important for processing the document;

[Explanation of symbols]

 Reference Signs List 10 discrimination performance control unit 20 edge region detection unit 22 edge enhancement unit 24 black pixel continuity detection unit 26 white pixel continuity detection unit 28 judgment unit 30 halftone dot region detection unit 40 comprehensive judgment unit 100 image area separation unit 110 character correction Unit 120 Halftone processing unit

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI theme coat ゛ (Reference) H04N 1/60 H04N 1/40 D 1/46 1/46 CF term (Reference) 2C362 CA05 CA06 CA14 CB23 CB24 CB25 CB29 CB62 5B057 AA11 BA02 CA01 CA08 CA12 CA16 CB01 CB02 CB08 CB12 CB16 CC01 CE03 CE17 CH08 5C077 LL19 MP02 MP05 MP06 MP07 MP08 NN11 PP03 PP27 PP28 PP43 PP47 PP55 PP65 PQ08 RR06 RR08 LA11 NA11 LA11 RR16 LA11 NA11 5L096 AA02 AA06 BA07 DA01 EA17 FA06 FA43 FA44 GA17

Claims (13)

[Claims] 1. An image area separation apparatus for detecting a character area from input image data, comprising: a discrimination performance control unit for controlling a judgment performance based on a type of a reproduced image; and an image feature for the area in the input image data. The apparatus includes at least one determination processing unit, and includes an edge region detection unit that detects an edge region as the determination processing unit. The determination performance control unit detects more edges when the input image is reproduced as a monochrome image. An image area separation device, wherein a determination parameter in the edge area detection unit is controlled in a direction in which the edge area is detected. 2. An image area separating apparatus for detecting a character area from input image data, comprising: a discrimination performance control unit for controlling a judgment performance based on a type of a reproduced image; and an image feature for the area in the input image data. The discrimination performance control unit includes at least two determination processing units, and includes an edge region detection unit that detects an edge region and a halftone dot region detection unit that detects a halftone dot region as the determination processing unit. An image area separating apparatus, wherein when an image is reproduced as a monochrome image, a determination parameter in the halftone dot area detection unit for detecting a halftone dot is controlled without passing through the edge area detection unit. 3. An image area separation method for detecting a character area from input image data, comprising: a determination performance control step of controlling determination performance based on a type of a reproduced image; and detecting an edge area in the input image data. An edge area detecting step, comprising: a halftone dot area detecting step of detecting a halftone dot area in the input image data, wherein the determining performance control step comprises: when reproducing the input image as a monochrome image, the edge area detecting step Without passing through
An image characterized by controlling a determination parameter in the halftone dot detection step for detecting a halftone dot, or controlling a determination parameter in the edge area detection step in a direction to detect an edge more. Area separation method. 4. A computer-readable recording medium in which a program for causing a computer to function as an image area separating device for detecting a character area from input image data is recorded, wherein the determination is made based on the type of reproduced image. A discrimination performance control unit for controlling performance; and at least two judgment processing units for judging an image feature for an area in the input image data; an edge area detection unit for detecting an edge area as the judgment processing unit; The discrimination performance control unit, when reproducing the input image as a monochrome image, without passing through the edge region detection unit, the halftone dot detection to detect a halftone dot Control the determination parameter in the edge area detection unit in the direction to detect the edge more. And a computer-readable recording medium having recorded thereon an image area separation processing program. 5. An image processing apparatus for reproducing input image data as a color image or a monochrome image, wherein when an input image is reproduced as a monochrome image, an image area separation for detecting a halftone dot area without detecting an edge area. An image processing apparatus comprising: a unit; and a character correction unit configured to perform correction for enhancing contrast on a non-halftone dot region detected by the image area separation unit. 6. The image processing apparatus according to claim 5, wherein the character correcting unit emphasizes a highlight portion and a shadow portion with respect to the non-dot region separated by the image area separating unit. An image processing apparatus characterized in that gamma correction with a steep slope is performed. 7. An image processing apparatus for reproducing input image data as a color image or a monochrome image, wherein when an input image is reproduced as a monochrome image, an image area separation for detecting a halftone dot area without detecting an edge area. And a threshold control for setting a lower threshold for performing conversion from multi-valued data to small-valued data for a non-halftone dot region detected by the image area separation unit, so as to increase the contrast. An image processing apparatus comprising: a halftone processing unit. 8. The image processing apparatus according to claim 7, wherein the halftone processing unit uses an error diffusion processing method. 9. The image processing apparatus according to claim 8, wherein the halftone processing unit uses a different error correction value between a halftone area and a non-halftone area separated by the image area separation unit. Characteristic image processing device. 10. An image processing method for reproducing input image data as a color image or a monochrome image. In reproducing an input image as a monochrome image, image area separation for detecting a halftone dot region without detecting an edge region. An image processing method comprising: a step; and a character correction step of performing a correction for enhancing contrast on a non-halftone dot area detected in the image area separation step. 11. An image processing method for reproducing input image data as a color image or a monochrome image. In reproducing an input image as a monochrome image, image area separation for detecting a halftone dot region without detecting an edge region. And controlling the threshold value when converting the multi-valued data to the small-valued data for the non-dot area detected in the image area separation step, so that the threshold value is controlled so as to increase the contrast. An image processing method comprising: a halftone processing step. 12. A computer-readable recording medium storing a program for causing a computer to function as an image processing device for reproducing input image data as a color image or a monochrome image, wherein the input image is reproduced as a monochrome image. In this case, an image area separating unit that detects a halftone area without detecting an edge area, and a character correction that performs a correction that emphasizes contrast on a non-halftone area detected by the image area separating unit A computer-readable recording medium on which an image processing program functioning as a unit is recorded. 13. A computer-readable recording medium on which a program for causing a computer to function as an image processing device for reproducing input image data as a color image or a monochrome image is provided, wherein the input image is reproduced as a monochrome image. In this case, the edge area is not detected, and the image area separating unit that detects the halftone area, and the non-halftone area detected by the image area separating unit is converted from multi-valued data to small value data. A computer-readable recording medium that stores an image processing program that functions as a halftone processing unit that performs threshold control so as to increase the threshold by setting a lower threshold when performing image processing.