Image analysis – Histogram processing – For setting a threshold
Reexamination Certificate
1997-12-18
2001-07-31
Boudreau, Leo (Department: 2621)
Image analysis
Histogram processing
For setting a threshold
C382S176000, C382S270000, C358S462000, C358S466000
Reexamination Certificate
active
06269186
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to an image processing apparatus and method for determining, e.g., the quantization threshold of a multivalued image to perform quantization.
2. Description of the Related Art
The recent development of image processing techniques is remarkable, and image processing apparatuses capable of processing of a multivalued image such as a full-color image, and character recognition processing in the multivalued image are becoming popular. In these image processing techniques, binarization processing of a multivalued image is an indispensable technique.
In addition to a simple binarization method using a fixed threshold set in advance, conventional binarization methods include the Otsu's method (Otsu, “Automatic Threshold Selection Method Based on Discrimination and Least Square Rule”, the transactions of the Institute of Electronics and Communication Engineers of Japan, Vol. J63-D, No. 4, pp. 349-356, 1980) in which, when the histogram is divided into two classes at a certain threshold, a threshold obtained when the variance between the classes is maximized is used as a binarization threshold, and a binarization method of setting the threshold for an image having a gradation in accordance with the local density.
In the binarization methods in the conventional image processing apparatus, however, the following problems arise.
More specifically, in the simple binarization method using the fixed threshold, a proper threshold is difficult to set between the object density and background density within an image. As a result, the entire image becomes excessively dark or bright. In the Otsu's method, when the distributions of two classes are greatly different from each other, the threshold shifts to a larger class, and a binary image with many noise components is undesirably generated. In the binarization method of setting the threshold in accordance with the local density, a block skew easily occurs because the image is divided into local regions. Even if an optimal threshold can be specified, gray scale information such as the background and characters of an original image is lost by binarization.
SUMMARY OF THE INVENTION
The present invention has been made to solve the above problems, and has as its object to provide an image processing apparatus and method for setting a proper quantization threshold between the object density and background density within an input image to perform image processing.
In order to solve the above problems, it is another object of the present invention to provide an image processing apparatus and method capable of specifying an optimal quantization threshold to perform region separation without losing gray scale information such as the background and characters of an original image.
To achieve the above objects, according to the present invention, there is provided an image processing apparatus for quantizing a multivalued image to perform image processing, comprising first calculation means for calculating a luminance frequency of the multivalued image, specifying means for specifying a quantization threshold for quantization on the basis of the calculated luminance frequency, second calculation means for calculating a representative value used for quantization of the multivalued image on the basis of the specified quantization threshold and the luminance frequency, and quantization means for quantizing the multivalued image using the calculated representative value.
To achieve the above objects, according to the present invention, there is provided an image processing method of quantizing a multivalued image to perform image processing, comprising the steps of calculating a luminance frequency of the multivalued image, specifying a quantization threshold for quantization on the basis of the calculated luminance frequency, calculating a representative value used for quantization of the multivalued image on the basis of the specified quantization threshold and the luminance frequency, and quantizing the multivalued image using the calculated representative value.
Other features and advantages of the present invention will be apparent from the following description taken in conjunction with the accompanying drawings, in which like reference characters designate the same or similar parts throughout the figures thereof.
REFERENCES:
patent: 4832447 (1989-05-01), Javidi
patent: 5150433 (1992-09-01), Daly
patent: 5337373 (1994-08-01), Marandici et al.
patent: 5386508 (1995-01-01), Itonori et al.
patent: 5649024 (1997-07-01), Goldsmith
patent: 5701367 (1997-12-01), Koshi et al.
patent: 5841551 (1998-11-01), Nakatani et al.
patent: 5867593 (1999-02-01), Fukuda et al.
patent: 5872864 (1999-02-01), Imade et al.
patent: 5920655 (1999-07-01), Makita et al.
“An Automatic Threshold Selection Method Based On Discriminant and Least Squares Criteria”, by Nobuyuki Otsu, Journal of Papers of The Institute of Electronics, Information and Communication Engineers vol. J63-D, No. 4, pp. 349-356, 1980. (English abstract).
Boudreau Leo
Canon Kabushiki Kaisha
Fitzpatrick ,Cella, Harper & Scinto
Sherali Ishrat
LandOfFree
Image processing apparatus and 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 method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing apparatus and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2561801