DocumentCode :
3363035
Title :
Real-time occluded object matching using the DWT on a BDFA
Author :
Wilburn, V.C. ; Yoon, S.H. ; Alexander, W.E.
Author_Institution :
Dept. of Electr. & Comput. Eng., North Carolina State Univ., Raleigh, NC, USA
fYear :
1994
fDate :
25-28 Oct 1994
Firstpage :
170
Lastpage :
173
Abstract :
Matching of occluded objects is a difficult problem. Moreover, the real-time matching problem is considerable. Real-time implementation is very important in industrial automation. In the paper, the authors propose using the discrete wavelet transform (DWT) of a boundary to obtain a reduced resolution representation in real-time with the BDFA paradigm. The existing image representation approaches are not effective for real-time occluded object matching since they use a large number of data points. The authors propose a new boundary representation which is generated by locating zero-crossings of the curvature function of boundaries at only the maximum scales where the number of zero-crossings is consistent with the original scale. The proposed representation reduces the number of points in the boundary representation and is appropriate for real-time matching in the presence of occlusion
Keywords :
edge detection; image matching; image representation; image resolution; real-time systems; transforms; wavelet transforms; BDFA; boundary representation; curvature function; discrete wavelet transform; industrial automation; real-time implementation; real-time occluded object matching; reduced resolution representation; zero-crossings; Automation; Computer industry; Computer vision; Discrete wavelet transforms; Image generation; Multiresolution analysis; Pattern recognition; Signal resolution; Spline; Wavelet analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Time-Frequency and Time-Scale Analysis, 1994., Proceedings of the IEEE-SP International Symposium on
Conference_Location :
Philadelphia, PA
Print_ISBN :
0-7803-2127-8
Type :
conf
DOI :
10.1109/TFSA.1994.467266
Filename :
467266
Link To Document :
بازگشت