Authors:
Ho Suk Lee
and
Jun Ki Kim
Affiliation:
Hoseo University, Korea, Republic of
Keyword(s):
Moving object edge, Moving object boundary, Boundary linking, Background, Moving object segmentation.
Related
Ontology
Subjects/Areas/Topics:
Computer Vision, Visualization and Computer Graphics
;
Image and Video Coding and Compression
;
Image and Video Processing, Compression and Segmentation
;
Image Formation and Preprocessing
;
Multimedia
;
Multimedia Signal Processing
;
Telecommunications
Abstract:
Moving object boundary is very important for moving object segmentation. We extract the moving object boundary from the moving object edge. But the object boundary shows broken boundary and we use a boundary linking to link the broken boundary. The boundary linking algorithm forms a quadrant around the terminating pixel in the broken boundary and searches forward other terminating pixel to link within a radius. The linking algorithm guarantees shortest distance linking. We register the background from image sequence. We construct two object masks, one from boundary linking and the other from the background, and use these two complementary object masks for moving object segmentation. We also filter out the moving cast shadow using gradient operator. The major characteristics of the proposed algorithms are accurate moving object segmentation, multiple moving objects segmentation, and the segmentation of an object which has holes in its region using these two object masks. We experiment
the algorithms using the standard MPEG-4 test sequences and real video sequence. The proposed algorithms are very efficient and can process QCIF image more than 48 fps and CIF image more than 19 fps in a 2GHz Pentium-IV computer.
(More)