Title :
Feature oriented image sequence processing and 3D adaptive morphology-formulation and properties
Author :
Cheng, F. ; Venetsanopoulos, A.N.
Author_Institution :
Dept. of Electr. Eng., Toronto Univ., Ont., Canada
Abstract :
The authors propose a type of feature oriented image sequence processing based on the 3D geometrical structures of image sequences, 3D adaptive morphological operators are developed for the purpose. The operators can adapt the shapes of their operational windows according to the 3D local geometrical structures of image sequences. Thus, not only can the spatial correlation in each frame be utilized to preserve details, but also the temporal correlation between frames can be utilized to trace motion in a simple way. A number of properties of the operators developed are proved. These properties lead to the geometrical explanation of the performance of the operators, and the development of fast algorithms
Keywords :
image sequences; 3D adaptive morphology; 3D geometrical structures; feature oriented processing; image sequence processing; operational windows; spatial correlation; temporal correlation; Entropy; Filters; Image analysis; Image processing; Image sequences; Morphology; Psychology; Shape; Statistical analysis; Statistics;
Conference_Titel :
Circuits and Systems, 1992. ISCAS '92. Proceedings., 1992 IEEE International Symposium on
Conference_Location :
San Diego, CA
Print_ISBN :
0-7803-0593-0
DOI :
10.1109/ISCAS.1992.230680