Image analysis – Image enhancement or restoration – Object boundary expansion or contraction
Reexamination Certificate
1999-07-02
2004-08-03
Lee, Thomas D. (Department: 2724)
Image analysis
Image enhancement or restoration
Object boundary expansion or contraction
C382S260000, C382S308000, C382S173000
Reexamination Certificate
active
06771834
ABSTRACT:
BACKGROUND
1. Field
This disclosure relates to processing digital images and, more particularly, to segmenting a digital image.
2. Background Information
In digital image processing systems, a variety of forms of digital image processing operations are performed. At times, the entire image processing system, as well as its individual component operations, depends on the particular application. Nonetheless, a challenge in digital image processing systems at large is to successfully segment a digital image, e.g., to partition the image into a set of smaller and meaningful regions. This is desirable in a plethora of situations. For example, it may be desirable to use different methods to process, transmit or store the different regions of the image depending on the region contents. In video and image communications, if limited bandwidth and/or memory are available it may be interesting or desirable to dedicate more bits to represent the most relevant regions of the image according to some application specific interest metric. A commonly used interest metric in segmentation of such a digital image partitions the image so that regions corresponding to moving objects in the scene, referred to in this context as “foreground objects,” are segmented or separated from the static “background” of the scene being imaged. Segmenting the image in this manner, also known as background/foreground segmentation, may be desirable in a host of possible applications. These include, but are not limited to, automated target acquisition and tracking, surveillance applications, navigation applications, gesture interpretation, video-based command and control, and computer games. A computer or similar computing platform may be used to execute such a segmentation method. Well-known segmentation methods based on edge detection generally fail to use the motion information that separates the foreground from the background. Methods that are based on computing a difference between a known shot of the background and the current video frame may produce artifacts known as false negatives (e.g., holes in the moving objects) and false positives (e.g., pieces of background merged with the foreground objects). Processes that are based on elaborate physical models for object dynamics may deliver better segmentation results, but often are too computationally intensive to perform in real-time. A need, therefore, exists for a method or technique for segmenting a digital image that improves upon the state of the art technology for real-time image segmentation.
SUMMARY
Briefly, in accordance with one embodiment of the invention, a method of segmenting an initial digital image includes the following. The initial digital image is processed to produce a first digital image with defined edges corresponding to the initial digital image and to produce a second digital image with at least two dominant contiguous regions corresponding to the initial digital image. Distinct non-overlapping regions of the first digital image formed by the defined edges are identified. The distinct non-overlapping regions of the first digital are combined based, at least in part, on a correspondence with the at least two dominant contiguous regions in the second digital image. Based, at least in part, on the remaining regions after combining the distinct non-overlapping regions of the first digital image, the initial digital image is segmented.
Briefly, in accordance with another embodiment of the invention, a method of removing at least some noisy pixels internal to one or more objects in a digital image includes the following. A sequence of one or more morphological openings and/or closings are applied in gradually increasing size. Then, a morphological closing with reduced erosion is applied, the erosion having a size larger than any of the one or more morphological openings and/or closings in the sequence.
REFERENCES:
patent: 5764306 (1998-06-01), Steffano
patent: 5812787 (1998-09-01), Astle
patent: 6301386 (2001-10-01), Zhu et al.
patent: 6396949 (2002-05-01), Nichani
patent: 6453069 (2002-09-01), Matsugu et al.
U.S. patent application Ser. No. 09/296,451, filed Apr. 21, 1999.
U.S. patent application Ser. No. 09/205,638, filed Dec. 3, 1998.
U.S. patent application Ser. No. 09/216,184, filed Dec. 18, 1998.
U.S. patent application Ser. No. 09/156,571, filed Sep. 17, 1998.
U.S. patent application Ser. No. 09/222,232, filed Dec. 29, 1998.
U.S. patent application Ser. No. 09/205,638, filed Dec. 3, 1998.
U.S. patent application Ser. No. 09/179,491, filed Oct. 26, 1998.
U.S. patent application Ser. No. 09/206,791, filed Dec. 7, 1998.
U.S. patent application Ser. No. 09/216,184, filed Dec. 18, 1998.
U.S. patent application Ser. No. 09/291,799, filed Apr. 14, 1999.
U.S. patent application Ser. No. 09/221,666, filed Dec. 23, 1998.
U.S. patent application Ser. No. 09/074,556, filed May 7, 1998.
U.S. patent application Ser. No. 09/502,493, filed Mar. 30, 1998.
U.S. patent application Ser. No. 09/109,571, filed Jul. 2, 1998.
U.S. patent application Ser. No. 09/151,555, filed Sep. 11, 1998.
U.S. patent application Ser. No. 09/767,362 filed Dec. 18, 1996.
Wang,Demin; “A Multiscale Gradient Algorithm for Image Segmentation Using Watersheds,” Communications Research Centre, 1997, pp. 2043-2052.
Wang, Demin and Labit, Claude, “Morphological Spatio-Temporal Simplificatin for Video Image Simplification,” Signal Processing Image Communication, 1997, pp. 161-170.
Wang, Demin, “Unsupervised Video Segmentation Based on Watersheds and Temporal Tracking,” IEEE Transactions on Circuits and Systems For Video Technology, vol. 8, No. 5, Sep. 1998.
Hazra Rajeeb
Martins Fernando C. M.
Intel Corporation
Lee Thomas D.
Wong Sharon
LandOfFree
Method for segmenting a digital image does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for segmenting a digital image, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for segmenting a digital image will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3282609