DocumentCode :
2425468
Title :
A novel shape matching method using biological sequence dynamic alignment
Author :
Zhang, Shaojie ; Ma, Kai-Kuang
Author_Institution :
Sch. of Electr. & Electron. Eng., Nanyang Technol. Univ., Singapore
Volume :
1
fYear :
2000
fDate :
2000
Firstpage :
343
Abstract :
Shape description and its corresponding matching algorithm is one of the main concerns in MPEG-7. A novel dynamic shape matching algorithm is proposed. First, the conventional 8-directional chain code is exploited to describe the binary shape. A dynamic shape matching algorithm, which is derived from the solution for the biological sequence similarity analysis based on molecular evolutionary theory, is proposed to measure the similarity distance between the shapes by computing their similarity score (SS). Based on a large set of MPEG-7 binary shapes, simulation results have shown that the proposed dynamic alignment matching algorithm is fairly promising and also robust to scale changes, small non-rigid deformation, and similarity based shape matching defined in MPEG-7 core experiments
Keywords :
biology computing; image coding; image matching; image retrieval; multimedia computing; 8-directional chain code; MPEG-7 binary shapes; MPEG-7 core experiments; MPEG-7 standard; binary shape; biological sequence dynamic alignment; biological sequence similarity analysis; dynamic alignment matching algorithm; dynamic shape matching algorithm; matching algorithm; molecular evolutionary theory; non-rigid deformation; shape description; shape matching method; similarity based shape matching; similarity distance; similarity score; simulation results; Algorithm design and analysis; Biological information theory; Biological system modeling; Biology computing; Computational modeling; Deformable models; Heuristic algorithms; MPEG 7 Standard; Robustness; Shape measurement;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Multimedia and Expo, 2000. ICME 2000. 2000 IEEE International Conference on
Conference_Location :
New York, NY
Print_ISBN :
0-7803-6536-4
Type :
conf
DOI :
10.1109/ICME.2000.869611
Filename :
869611
Link To Document :
بازگشت