Image analysis – Image segmentation
Reexamination Certificate
1995-09-06
2001-08-28
Patel, Jay (Department: 2623)
Image analysis
Image segmentation
C358S453000, C358S538000
Reexamination Certificate
active
06282314
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to image processing method and apparatus for analyzing characteristics of image data of each pixel.
2. Related Background Art
Conventional methods of forming retrieval keywords in an electronic file system consist of mainly the following two methods.
(I) Method in which image data is read from an image input apparatus such as a scanner or the like and the operator manually inputs a retrieval keyword.
(II) Method of reading the image data from an image input apparatus such as a scanner or the like, recognizing a character image in an original image, and automatically adding a retrieval keyword.
Although the method of (I) is a certain method and the retrieval keyword can be set independent of the input image, the operator needs to add the keyword every time and it is troublesome from a viewpoint of the operation.
According to the method of (II), since the characters in the image are recognized and are automatically added for retrieval, the operation is simple.
The method of (II), however, has the following problems.
That is, an image in which only character portions are extracted, namely, which is binarized by what is called an optimal threshold process is preferable as an image for the character recognition, so that it is undesirable to execute a halftone process even to an image in which original image includes a halftone image. Therefore, a case such that the portion of the halftone image perfectly becomes black or white and the contents of the image in such an area cannot be recognized occurs.
On the other hand, an image processed so as to relatively accurately conform to the original image is requested as an image to be displayed or recorded.
Therefore, the image binarized by the optimal fixed threshold is necessary for character recognition and the image which was halftone processed is necessary to display/record. The image suitable for the character recognition isn't always suitable for the image to display/record, and vice versa.
SUMMARY OF THE INVENTION
The invention is made in consideration of the above problems and an object of the invention is to convert multivalue image data to data in a form which can be easily handled.
According to the invention, such an object is accomplished by an image processing method comprising the steps of: inputting multivalue image data of each pixel; and sequentially converting an image which is expressed by the multivalue image data into data which is hierarchically structured to a plurality of layers.
Another object of the invention is to efficiently execute an area division in accordance with the kind of image.
According to the invention, such an object is accomplished by an image processing method comprising the steps of: inputting image data representing an image; determining a first threshold based on the image data; binarizing the image data by comparing the image data with the first threshold and generating first binary data; dividing the image into a plurality of partial areas in accordance with the first binary data; determining a second threshold based on the image data in one of the partial areas; binarizing the image data in the partial area by comparing the image data with the second threshold and generating second binary data; and dividing the partial area into a plurality of sub-partial areas in accordance with the second binary data.
Still another object of the invention is to easily provide an image format which is adaptive to each function of an electronic file.
Further another object of the invention is to provide a data structure in which image data doesn't excessively increase for various requests regarding image processes.
The above and other objects and features of the present invention will become apparent from the following detailed description and the appended claims with reference to the accompanying drawings.
REFERENCES:
patent: 4893188 (1990-01-01), Murakami et al.
patent: 5187592 (1993-02-01), Sugiyama et al.
patent: 5280547 (1994-01-01), Mahoney
patent: 5282059 (1994-01-01), Fukushima et al.
patent: 5297219 (1994-03-01), Weldy
patent: 5371610 (1994-12-01), Sugawa
patent: 5387983 (1995-02-01), Sugiura et al.
patent: 5465164 (1995-11-01), Sugiura et al.
patent: 5475496 (1995-12-01), Kumada
patent: 5521990 (1996-05-01), Ishizawa et al.
patent: 5533148 (1996-07-01), Sayah et al.
patent: 5557689 (1996-09-01), Huttenlocher et al.
patent: 5568571 (1996-10-01), Willis et al.
patent: 5644648 (1997-07-01), Bose et al.
patent: 5748794 (1998-05-01), Maeda et al.
patent: 5-233873 (1993-09-01), None
Maeda Mitsuru
Makita Takeshi
Mori Hiroshi
Sugiura Susumu
Yamada Osamu
Canon Kabushiki Kaisha
Fitzpatrick ,Cella, Harper & Scinto
Patel Jay
LandOfFree
Image processing method and apparatus which iteratively... 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 method and apparatus which iteratively..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing method and apparatus which iteratively... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2540706