Image analysis – Image segmentation – Distinguishing text from other regions
Reexamination Certificate
1999-10-08
2003-10-07
Boudreau, Leo (Department: 2621)
Image analysis
Image segmentation
Distinguishing text from other regions
C358S001900, C358S002100
Reexamination Certificate
active
06631210
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to an image-processing apparatus which can discriminate image areas so as to carry out an optimal image process with respect to inputted multi-value image data, in digital color copying machines and color scanners, and also concerns an image processing method for such an apparatus.
BACKGROUND OF THE INVENTION
Conventionally, in image-processing apparatuses such as digital copying machines and facsimile machines, an image process is applied to image signals obtained by reading an original document by a CCD (charge coupled device) sensor, etc. so as to improve the image quality. In other words, mesh or dot areas, silver-lead photograph areas and character areas contained in the original document are respectively discriminated, and image processes suitable for image characteristics of the respective areas are carried out so as to improve the image quality. For example, a smoothing process for reducing moire is applied to the mesh or dot areas, and an emphasizing process for clearly displaying the characters is applied to the character areas.
With respect to the image area discrimination method carried out in such image processes so as to improve the image equality, the following methods have been proposed:
Japanese Laid-Open Patent Application No. 163166/1997 “Tokukaihei 9-163166 (published on Jun. 20, 1997)” discloses a processing method for black characters. In this method, an image is divided into respective predetermined areas, and the respective areas are discriminated by using feature parameters representing the characteristics of character and mesh images. In other words, within a predetermined area including a target pixel, the average value is found so as to be set as a threshold value, and the difference between the threshold value and the target pixel value is found so that a judgement is made as to the character area based upon sizes from the results. Moreover, a black character is discriminated from 8 pixels adjacent to the target pixel, and a distinction from the mesh or dot is extracted.
Japanese Laid-Open Patent Application No. 56287/1996 “Tokukaihei 8-56287 (published on Feb. 27, 1996)” discloses a method in which discriminations are made among character/ling drawing areas, mesh areas and photograph areas based upon differences in color distribution of image data of CMY (cyan, magenta, yellow).
Japanese Laid-Open Patent Application No. 54180/1994 “Tokukaihei 6-54180 (published on Feb. 25, 1994)” discloses a method in which the average value, the maximum value, etc. within a specific area are calculated, and by comparing these with threshold values, the character area and the photograph area are discriminated.
However, the above-mentioned conventional methods fail to provide sufficient precision to discriminate the character/line drawing areas, the mesh or dot areas and the photograph areas, causing the possibility of maldiscrimination.
Moreover, the above-mentioned conventional methods can carry out discrimination on color images; however, in particular, when an edge extraction and/or emphasizing process is carried out on color characters, a problem arises in which the color of an edge portion tends to be different from the original color.
SUMMARY OF THE INVENTION
The objective of the present invention is to provide an image-processing apparatus which can carry out discrimination between character areas and mesh or dot areas as well as discrimination between black character areas and line drawing areas with high precision with respect to areas to which pixels of multi-value image data belong, and also to provide an image-processing method thereof.
In order to achieve the above-mentioned objective, an image-processing apparatus in the present invention, which carries out area discrimination on image data, is provided with: a first inspection circuit which, with respect to image data within a specific area consisting of a target pixel and pixels in a vicinity thereof, calculates the number of pixels whose density value is coincident with that of the target pixel so as to judge whether the density value of the target pixel is a maximum value or a minimum value; and a feature-amount extraction circuit for discriminating whether or not the target pixel is in a dot area.
Moreover, in order to achieve the above-mentioned objective, an image-processing method in the present invention, which carries out area discrimination on image data, is provided with a first inspection step for, with respect to image data within a specific area consisting of a target pixel and pixels in the vicinity thereof, calculating the number of pixels whose density value is coincident with that of the target pixel so as to judge whether the density value of the target pixel is a maximum value or a minimum value; and a feature-amount extraction step for discriminating whether or not the target pixel is in a dot area.
With the above-mentioned arrangement and method, the first inspection circuit and/or process make it possible to discriminate whether the density value of the target pixel is a maximum value or a minimum value within a specific area consisting of a target pixel and pixels in the vicinity thereof, with respect to inputted image data. Therefore, the feature-amount extraction circuit and/or process can discriminate which area the area including the target pixel on the image belong to, a character area, dot area or a photograph area, with high precision.
Therefore, it is possible to apply an optimal image process to the area including the target pixel on the image, and consequently to achieve high image quality.
In order to achieve the above-mentioned objective, an image-processing apparatus in the present invention is provided with: a first color detection circuit for detecting black-color portions with respect to pixels within a first area consisting of a target pixel and pixels in the vicinity thereof; a count circuit for counting the respective numbers of the black pixels and pixels having colors other than black with respect to the pixels in the first area; an edge discrimination circuit for detecting whether or not an abrupt color change occurs at the target pixel, and a black color detection circuit which, based upon the results of outputs from these color detection circuit, count circuit and edge discrimination circuit, detects which area the target pixel belongs to, the black-letter area or the line-drawing area.
In order to achieve the above-mentioned objective, the image-processing method of the present invention, which is an image-processing method for discriminating image-data areas, is provided with: a first color detection process for detecting black-color portions with respect to pixels within the first area consisting of a target pixel and pixels in the vicinity of the target pixel; a count process for counting the respective numbers of the black pixels and pixels having colors other than black with respect to the pixels in the first area; an edge discrimination process for detecting whether or not an abrupt color change occurs at the target pixel; and a black color detection process which, based upon the results of outputs from these color detection process, count process and edge discrimination process, detects which area the target pixel belongs to, the black-letter area or the line-drawing area.
With the above-mentioned arrangement and method, with respect to the first area consisting of the target pixel and pixels adjacent thereto of image data for each color component, it is possible to discriminate whether or not the target pixel on the image is black, so that an optimal image process can be applied as the black area. Moreover, with the edge discrimination circuit and/or process, with respect to image data within the first area, it is possible to discriminate whether or not the target pixel on the image belongs to a black character edge area, so that an optimal image process can be applied as an edge portion of a black character, black line drawing, etc. Then, based upon the results of these and the respective
Amagai Takayuki
Daido Takahiro
Kobayashi Setsuya
Kohashikawa Seiji
Mutoh Takeshi
Boudreau Leo
Conlin David G.
Daley, Jr. William J.
Edwards & Angell LLP
Sharp Kabushiki Kaisha
LandOfFree
Image-processing apparatus and image-processing method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image-processing apparatus and image-processing method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image-processing apparatus and image-processing method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3165732