Title :
Multi-metric video segmentation using isomorphic transformations
Author :
Agbinya, Johnson I. ; Rees, David
Author_Institution :
Telecommun. & Ind. Phys., Imaging & Signal Process., CSIRO, Epping, NSW, Australia
Abstract :
A multi-metric video segmentation is used for scene change detection. Isomorphic transforms are used to define three related metrics called self-contrast, resemblance, and direct-contrast. We show that they form a robust set of metrics for scene change detection. Thresholds based on variances are set. A unity change in variance in colour bands is enough to detect scene change
Keywords :
image colour analysis; image segmentation; image sequences; video signal processing; colour bands; direct-contrast; isomorphic transformations; multi-metric video segmentation; resemblance; scene change detection; self-contrast; thresholds; variances; Cameras; Communication industry; Image edge detection; Image processing; Image segmentation; Layout; Physics; Space technology; Video sequences; Video signal processing;
Conference_Titel :
TENCON '97. IEEE Region 10 Annual Conference. Speech and Image Technologies for Computing and Telecommunications., Proceedings of IEEE
Conference_Location :
Brisbane, Qld.
Print_ISBN :
0-7803-4365-4
DOI :
10.1109/TENCON.1997.648537
