Image analysis – Image segmentation
Reexamination Certificate
1999-09-24
2002-10-08
Johnson, Timothy M. (Department: 2623)
Image analysis
Image segmentation
C382S243000
Reexamination Certificate
active
06463174
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a macroblock-based segmentation and background mosaicking method for object-oriented coding of image sequence having a stationary background and more particularly to a segmentation method for segmenting an image sequence into moving and changing object regions and stationary background regions having rare change and constructing a background mosaic consisting of only macroblocks corresponding to a region which rarely has temporal change with respect to the background regions in an image so as to minimize a size of a region to be coded, thereby maximizing coding efficiency. The present invention also adopts a macroblock of 16×16 pixels as the unit of image segmentation, thereby maximizing object shape coding efficiency with very low complexity.
2. Description of Related Art
Typically, a videophone system using moving pictures employs a communication channel of extremely limited bandwidth, so compressing images before transmission is essential for the purpose of reducing the volume of information to be transmitted.
One of the representative video coding methods is an object-based coding method segmenting an image into background regions and object regions before compression.
Specifically, according to this method, the image is segmented into the stationary background regions and the moving and changing object regions. Each segmented region is coded by usual block-based video coding techniques, such as discrete cosine transform and quantization.
Furthermore, shape information of each segmented region is separately coded and transmitted for reconstruction of the image during decoding.
Efficient encoding of object shape is very important to maximize compression efficiency in the object-based video coding method. According to conventional segmentation techniques, an image is segmented into regions based upon pixels using temporal or spatial change information of the image.
In other words, to segment the image into several regions, a change of brightness is calculated for each pixel and the corresponding pixel is determined as the background region or the object region according to the results of evaluation. This is very complicated.
As illustrated, since complicated processes are carried out to segment an image based upon pixels in the conventional segmentation methods, it is very difficult to have real-time implementation
Moreover, large amount of bits are required to encode the shape information of regions segmented based upon pixels for transmission. This limits maximization of coding efficiency. Even if an overall coding rate decreases, a coding rate of the shape information cannot be lowered below a certain level. Therefore, the object-based coding method is not advantageous in case of very low bit rate communication.
In addition, the background region is always determined according to a difference between a current image and a preceding image. This disturbs realization of an advantage of the object-based coding method for enhancing the coding efficiency by segmenting stationary background regions against moving object regions.
SUMMARY OF THE INVENTION
Accordingly, the present invention is directed to a macroblock-based segmentation and background mosaicking method that substantially obviates one or more of the limitations and disadvantages of the related art.
An objective of the present invention is to provide a segmentation method for segmenting an image into moving and changing object regions and rarely changing stationary background regions and constructing a background mosaic with respect to the background regions so as to minimize a size of an area to be coded, thereby maximizing coding efficiency.
Another objective of the present invention is to provide a segmentation method of adopting a macroblock of 16×16 pixels as the unit of image segmentation, thereby maximizing object shape coding efficiency with very low complexity.
Additional features and advantages of the invention will be set forth in the following description, and in part will be apparent from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure as illustrated in the written description and claims hereof, as well as the appended drawings.
To achieve these and other advantages, and in accordance with the purpose of the present invention as embodied and broadly described, a macroblock-based segmentation and background mosaicking method comprises the steps of: receiving consecutive two frames; calculating validation statistics from the received frames to detect change/changelessness of respective macroblocks in a current frame; determining change/changelessness of respective macroblocks using the calculated validation statistics and generating a change detection mask; and judging respective macroblocks as background regions or object regions, using the change detection mask of the current frame, a change detection mask of a preceding frame, the current frame, and a background mosaic.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory and are intended to provide further explanation of the invention as claimed.
REFERENCES:
patent: 5034986 (1991-07-01), Karmann et al.
patent: 5631975 (1997-05-01), Riglet et al.
patent: 5896176 (1999-04-01), Das et al.
patent: 6208693 (2001-03-01), Chen et al.
Background Mosaicking for Low Bit Rate Video Coding, by Frédéric Dufaux et al, ICIP New York, Sep. 16-18, 1996, vol. 3, pp. 673-676, 0-7803-3258-X/96 © 1996 IEEE.
“Foreground/Background Video Coding Scheme”, by Douglas Chai et al, appearing in 1997 IEEE International Symposium on Circuits and Systems, Jun. 9-12, Hong Kong, pp. 1448-1451, 0-7803-3583-X/97 ©1997 IEEE.
“Hierarchical Morphological Segmentation for Image Sequence Coding”, by Philippe Salembier et al, appearing in IEEE Transactions on Image Proceeding, vol. 3, No. 5, Sep. 1994, pp. 639-651, 1057-7149/94, © 1994 IEEE.
Ahn Chie Teuk
Lim Young Kwon
Park Sang Gyu
Antonelli Terry Stout & Kraus LLP
Electronics and Telecommunications Research Institute
Johnson Timothy M.
LandOfFree
Macroblock-based segmentation and background mosaicking 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 Macroblock-based segmentation and background mosaicking method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Macroblock-based segmentation and background mosaicking method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2950838