Image analysis – Image transformation or preprocessing
Reexamination Certificate
1997-07-15
2001-01-09
Mehta, Bhavesh (Department: 2721)
Image analysis
Image transformation or preprocessing
C382S128000, C382S232000
Reexamination Certificate
active
06173086
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to an image processing method. This invention particularly relates to characteristics of a filter used in a wavelet transform carried out on an image signal representing an image, in which periodical structure patterns are embedded.
2. Description of the Prior Art
A wavelet transform is one of techniques for signal frequency analysis. The wavelet transform is advantageous over a Fourier transform, which has heretofore been used widely as the frequency analysis technique, in that a local change in a signal can be detected easily. Therefore, recently, the wavelet transform has attracted particular attention in the field of signal processing.
In the field of signal processing, processing is often carried out on an image signal such that different processes may be carried out for different frequency bands. In such cases, the wavelet transform is employed as means for classifying the image signal into signals falling within different frequency bands. Specifically, for example, the processes include a process for eliminating noise by separating high frequency components and a process for compressing an image signal by reducing the information of a frequency band, in which much noise is contained. The applicant proposed various image signal compressing methods, which are based upon a technique proposed in “Image Coding Using Wavelet Transform” by Marc Antonini, et al., IEEE Transactions on Image Processing, Vol. 1, No. 2, pp. 205-220, April 1992. In the proposed image signal compressing methods, the wavelet transform is carried out in order for different compressing processes to be carried out for different frequency bands. (The image signal compressing methods are proposed in, for example, U.S. Ser. No. 08/253,857; U.S. Ser. No. 08/303,198, and Japanese Unexamined Patent Publication Nos. 6(1994)-350990, 7(1995)-23228 and 7(1995)-23229. In such image processing, the signals having been obtained from the wavelet transform are subjected to different processes for different frequency bands, and the processed signals are then subjected to inverse wavelet transform. An image obtained from the inverse wavelet transform is utilized as the product of the image processing.
In cases where an X-ray image of an object is recorded on a recording medium, such that the image quality may be prevented from becoming bad due to scattered X-rays impinging upon the recording medium, a technique is often employed wherein a stationary grid having a predetermined grid pitch is located between the object and the recording medium and wherein the X-ray image of the object is thereby recorded on the recording medium. Heretofore, the wavelet transform has also been carried out on a radiation image, which has been recorded in such a manner, in order for the image quality to be kept high or in order for the image signal to be compressed for the purposes of signal transmission or signal storage. In such cases, heretofore, an image obtained ultimately from the inverse wavelet transform, which image has the same size as the size of the original image, has been reproduced, primarily on photographic film, and viewed.
An image signal falling within a low frequency band, which image signal is obtained from the wavelet transform carried out on an original image signal, represents an image having a reduced size and having a resolution lower than the resolution of the original image. In the conventional techniques for processing radiation images, the image to be viewed is the one which is obtained from the inverse wavelet transform, and the reduced image, which is obtained as an intermediate product from the wavelet transform, is not utilized. Therefore, no problem has occurred with regard to the image quality of the reduced image.
In image processing systems, it is often desired that an image subjected to processing can be checked on a cathode ray tube (CRT) display device, or the like. In such cases, ordinarily, the original image signal is subjected to a filtering process, and the level of resolution of the image is thereby decreased to a level appropriate for the level of resolution on the CRT display device, or the like. Therefore, in cases where the image processing systems have the characteristics such that the wavelet transform may be carried out on the image signal, the reduced image obtained as the intermediate product in the wavelet transform can be utilized as the image to be displayed on the CRT display device, or the like.
However, the reduced image having a low resolution, which is obtained from the wavelet transform, is the image obtained by carrying out sub-sampling at predetermined sampling intervals on the original image signal. Therefore, in cases where the reduced image is an image in which periodical structure patterns, such as the patterns of the stationary grid, are embedded, and the frequency corresponding to the periodical structure patterns is higher than the Nyquist frequency in the sub-sampling, moire due to the sub-sampling occurs in the displayed image. As a result, the image quality of the image displayed on the CRT display device cannot be kept good.
SUMMARY OF THE INVENTION
The primary object of the present invention is to provide an image processing method, wherein a reduced image free from moire and having good image quality is obtained even from a wavelet transform carried out on an original image signal representing an original image, in which periodical structure patterns are embedded.
Another object of the present invention is to provide an image processing method, which enables efficient utilization of a reduced image obtained from a wavelet transform.
The present invention provides an image processing method for carrying out a wavelet transform on an image signal, and thereby obtaining a reduced image to be viewed, which is represented by low frequency components of the image signal, the low frequency components being obtained from sub-sampling, which is carried out on the image signal at predetermined sub-sampling intervals, wherein the wavelet transform is carried out on an image signal representing an image, in which periodical structure patterns corresponding to a frequency higher than a Nyquist frequency in the sub-sampling are embedded, by using at least one kind of low pass filter capable of decreasing the frequency components of the image signal, which correspond to the periodical structure patterns.
Examples of the image, in which the periodical structure patterns are embedded, include a radiation image having been recorded by locating a stationary grid between an object and a recording medium, a photograph taken via wire netting, and a photograph of an object having a stripe pattern.
In cases where the frequency of the sub-sampling is represented by N (i.e., the predetermined sub-sampling intervals are equal to 1/N), the Nyquist frequency is equal to N/2. The Nyquist frequency is defined by the sampling theorem. The term “sub-sampling” as used herein means the sampling of a digital signal, which has been sampled from an analog signal.
In cases where only one kind of low pass filter is used in the filtering for obtaining the reduced image, the term “at least one kind of lowpass filter” as used herein means the low pass filter used in the filtering. In cases where the reduced image is obtained by carrying the wavelet transform in several steps and by using several kinds of low pass filters, the term “at least one kind of low pass filter” as used herein means at least one kind of low pass filter among the several kinds of low pass filters.
Further, the term “image processing for obtaining a reduced image to be viewed” as used herein is not limited to the image processing, which aims at obtaining a reduced image to be viewed, and means every kind of image processing, in which it is possible that a reduced image obtained during the processing will be viewed. Therefore, for example, the image processing for obtaining a reduced image to be viewed may be the processing for classifying an image signal int
Fuji Photo Film Co. , Ltd.
Mehta Bhavesh
Patel Kanji
Sughrue Mion Zinn Macpeak & Seas, PLLC
LandOfFree
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 method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2476637