Image analysis – Image enhancement or restoration – Edge or contour enhancement
Reexamination Certificate
2000-06-29
2003-10-21
Mehta, Bhavesh M. (Department: 2724)
Image analysis
Image enhancement or restoration
Edge or contour enhancement
C358S426010, C358S003050, C375S240290, C382S162000, C382S260000, C382S165000, C382S261000
Reexamination Certificate
active
06636645
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to the field of digital image processing, and more particularly to methods for reducing noise and blocking artifacts in a digital image.
BACKGROUND OF THE INVENTION
Digital images are widely used in image communication. Due to the various components in the image communication chain, along with various image processing methods applied to the digital images, these digital images may contain certain artifacts, such as imaging noise and JPEG blocking artifacts (if the image is highly compressed in a JPEG format). Thus, effective image enhancement techniques are required to process these digital images in order to achieve better image quality.
A typical approach known in the prior art for such image enhancement may comprise a series of sequential steps, such as the following:
1. noise reduction for a luminance channel through image filtering (e.g., as described in commonly assigned U.S. patent application Ser. No. 09/522,742, entitled “Noise reduction method utilizing statistical weighting, apparatus, and program for digital image processing”, filed Mar. 10, 2000 in the names of E. B. Gindele and J. Luo);
2. noise reduction for a chrominance channel through image filtering (e.g., as described in commonly assigned U.S. patent application Ser. No. 09/415,374, entitled “Removing chroma noise from digital images by variable shape pixel neighborhood regions” filed Oct. 8, 1999 in the names of J. E. Adams, Jr. and J. F. Hamilton, Jr.); and
3. JPEG de-blocking through image filtering (e.g., as described in H. C. Reeve III and J. S. Lim, “Reduction of blocking effect in image coding,”
ICASSP
, pp. 1212-1215, 1983).
The main drawbacks of such sequential processing are the following:
1. output images are soft with severe loss of details due to consecutive filtering operations (i.e., smoothing in nature); and
2. execution speed is slow, and thus may not be suitable for time-critical applications.
In J. Luo et al, “Artifact Reduction in Low Bit Rate DCT-based Image Compression”,
IEEE Transaction on Image Processing
, vol. 5, No. 9, September, 1996, Luo et al teaches a method for artifact reduction for JPEG compressed images by using a Huber-Markov random-field model-based filtering technique, which differentiates artifacts (e.g., noise and JPEG block boundaries) from image details and applies filtering that treats block boundaries and non-block boundaries differently.
This method for artifact reduction taught by Luo et al is based on a number of assumptions, such as:
1. image noise is not intensity dependent;
2. JPEG block boundary locations are known in advance; and
3. the input signal is a gray scale digital image.
These assumptions, however, may not hold for certain image applications. For example, images from most digital cameras are color images that have three channels (R, G, B). Moreover, for most digital cameras, noise level is highly intensity dependent. In addition, there will be no prior knowledge as to where the block boundaries are located if a cropping operation has been performed on the JPEG compressed images.
Therefore, it is highly desirable to develop an efficient algorithm that adaptively removes noise and JPEG blocking artifacts at the same time. Thus there is a need for an improved, efficient method for processing an image in order to reduce noise and blocking artifacts.
SUMMARY OF THE INVENTION
The present invention is directed to overcoming one or more of the problems set forth above. Briefly summarized, according to one aspect of the present invention, a digital image processing method reduces noise and blocking artifacts in a digital image having pixels representing RGB values by converting the digital image pixels to X
1
, X
2
and X
3
components; detecting the block boundaries in the X
1
, X
2
and X
3
image components; estimating the noise in the X
1
, X
2
and X
3
image components; constructing one or more noise tables for the X
1
, X
2
and X
3
image components; applying an adaptive Huber-Markov random-field model-based filter (HMRF) to the X
1
, X
2
and X
3
image components, where the HMRF employing the detected block boundaries and the noise tables to produce filtered X
1
, X
2
and X
3
image components; and converting the filtered X
1
, X
2
and X
3
image components to RGB components.
ADVANTAGES
The present invention has the advantage that the amount of noise reduction applied is adapted to the local intensity and noise statistics of the digital image, resulting in a consistent-appearing image. The present invention also has the advantage that it does not require prior knowledge as to where the JPEG block boundaries are located.
REFERENCES:
patent: 5452104 (1995-09-01), Lee
patent: 5856848 (1999-01-01), Fujiwara et al.
patent: 5966465 (1999-10-01), Keith et al.
patent: 6040876 (2000-03-01), Pettitt et al.
patent: 6151420 (2000-11-01), Wober et al.
patent: 6226410 (2001-05-01), O'Rourke
patent: 6434275 (2002-08-01), Fukuda et al.
patent: 6463182 (2002-10-01), Onishi et al.
patent: 6510242 (2003-01-01), Westerman
patent: 2001/0002214 (2001-05-01), Yeo et al.
patent: 2001/0022857 (2001-09-01), Kondo et al.
patent: 2001/0031097 (2001-10-01), Mancuso et al.
patent: 2002/0196355 (2002-12-01), Hiroshige et al.
patent: 2003/0020835 (2003-01-01), Petrescu
U.S. patent application Ser. No. 09/415,374, Adams, Jr. et al., filed Oct. 8, 1999.
U.S. patent application Ser. No. 09/522,742, Gindele et al., filed Mar. 10, 2000.
“Reduction of Blocking Effect in Image Coding” by Howard C. Reeve III and Jae S. Lim. ICASSP, pp. 1212-1215, 1983.
“Artifact Reduction in Low Bit Rate DCT-Based Image Compresion” by Jiebo Luo, Chang Wen Chen, Kevin J. Parker, and Thomas S. Huang.
Joshi Rajan L.
Luo Jiebo
Yu Qing
Choobin Barry
Eastman Kodak Company
Mehta Bhavesh M.
Woods David M.
LandOfFree
Image processing method for reducing noise and blocking... 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 for reducing noise and blocking..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing method for reducing noise and blocking... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3121569