Non-linear image filter for filtering noise

Details Non-linear image filter for filtering noise Non-linear image filter for filtering noise

2000-06-22

2002-02-12

Rogers, Scott (Department: 2624)

Image analysis

Image enhancement or restoration

Image filter

C382S265000, C382S275000

Reexamination Certificate

active

06347161

ABSTRACT:

TECHNICAL FIELD
The invention is related generally to digital signal processing and, in particular, digital signal processing of images.
BACKGROUND OF THE INVENTION
Noise is an inherent feature in any communication system, and efforts to reduce noise are continually being explored. For example, it is well known that low-pass filters with pre-set frequency ranges reduce high-frequency noise. Moreover, it is known that in many communication systems it is advantageous to reduce noise prior to displaying the signal (or image) because noise reduction improves image display quality.
Modern-day image processors also use other techniques to enhance image display. For example, image processes utilize compression (or encoding) techniques to minimize the amount of memory needed to process images. There are compression standards for still images (or pictures), color facsimile (fax) machines, medical imaging systems, video cassette records (VCRs), entertainment systems, televisions (TVs), high-definition TVs (HDTVs), World-Wide Web browsers, and for personal computer (PC) displays, for example.
The coding efficiency of an image processor can be improved if system noise is reduced prior to encoding. Adaptive filters have been applied to images for this purpose. For example, adaptive filters reduce noise by monitoring the communication system and feeding back communication system information to automatically and dynamically adjust their filtering characteristics. Generally, adaptive filters can be obtained using non-linear approaches, such as median-based filters or filters having variable behavior according to the estimation of certain system or image parameters. For example, some adaptive filters are obtained by estimating the temporal correlation of certain parameters.
These adaptive filters have limitations, however. For example, if a video signal is greatly affected by noise, these adaptive filters confuse the undesirable noise with image motion. This limitation causes the fine details of the image to be attenuated, or smoothed, which also is undesirable. Proposed solutions include variable-strength filters, i.e., filters with curved selectivity. Arbitrary curve selection, however, affects the temporal frequency of the adaptive filter, resulting in the so-called “comet effect,” which is more annoying than detail smoothing. It can be appreciated therefore that what is needed is an image enhancement technique that reduces noise without blurring fine details of the image or affecting the temporal frequency.
SUMMARY OF THE INVENTION
Presented herein is a system and method for reducing noise using recursive noise level estimation. An example embodiment includes a noise filter that uses a processing window with a target pixel and several neighboring pixels. The noise filter includes an image processor that generates a replacement value for the target pixel in the processing window based on a weighted average of the absolute values of degrees of similarity between the target pixel and the plurality of neighboring pixels. The noise filter recursively processes the noise levels to adapt to the local features of the scanned image, as well as to adapt to changing noise levels of neighboring pixels. That is the noise filter generates recursive noise level estimations whereby local feature data is fed back into the filter for calibration. The noise filter uses fuzzy logic processing to filter noise without smoothing the image.
Further features and advantages of the invention as well as the structure and operation of various embodiments are described in detail below.


REFERENCES:
patent: 5339365 (1994-08-01), Kawai et al.
patent: 5390264 (1995-02-01), Ishihara et al.
patent: 5490094 (1996-02-01), Heimburger et al.
patent: 5561724 (1996-10-01), Kido et al.
patent: 5621474 (1997-04-01), Mancuso et al.
patent: 5694492 (1997-12-01), Kim et al.
patent: 5715335 (1998-02-01), Haan et al.
patent: 5802218 (1998-09-01), Brailean
patent: 5818964 (1998-10-01), Itoh et al.
patent: 5949916 (1999-09-01), Chun
patent: 6041145 (2000-03-01), Hayashi et al.
patent: 0 698 990 (1996-02-01), None
patent: 0 878 776 (1998-11-01), None
Farbiz, F.; Menhaj, M.B., “A Modified Iterative Fuzzy Control Based Filter For Image Enhancement With Multiplicative Noise Removal Property,”IEEE J.
Farbiz, F.; Motamedi, S.A.; Menhaj, M.B., “An Iterative Method for Image Enhancement Based on Fuzzy Logic,” International Conference on Acoustics, Speech and Signal Processing.IEEE J.
Fotinos, A.; Laskaris, N.; Fotopoulos, S. “Fuzzy Color Filtering Using Potential Functions,” International Conference on Digital Signal Processing Proceedings.IEEE J.
Hsu, Yuh-Feng and Yung-Chang Chen,“A New Adaptive Separable Median Filter For Removing Blocking Effects,”IEEE Trans. On Consumer Electron., 39(3):510-513, Aug. 1993.
Jarske et al., “Post-Filtering Methods For Reducing Blocking Effects From Coded Images,”IEEE Trans. On Consumer Electron., 40(3):521-526, Aug. 1994.
Mancuso, M.; D'Alto, V.; De Luca, R.; Poluzzi, R.; Rizzotto, G.G., “Fuzzy Logic Based Image Processing in IQTV Environment,”Consumer Electronics, IEEE J.
Ramamurthi, Bhaskar and Allen Gersho, “Nonlinear Space-Variant Processing of Block Coded Images,”IEEE Trans. Acous., Seech and Signal Proc., ASSP-34(5):1258-68, Oct. 1986.
Reeve III, Howard C. and Jae S. Lim, “Reduction of Blocking Effects in Image Coding,”Opt. Eng.,23(1):034-037, Jan.-Feb. 1984.
Reid et al., “Second-Generation Image Coding: An Overview,”ACM Computing Surveys, 29(1):Mar. 3-29, 1997.

Affiliated with

Also associated with

Rating

Non-linear image filter for filtering noise does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Non-linear image filter for filtering noise, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Non-linear image filter for filtering noise will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2969421