This paper introduces a method for quantization of image-sequences which takes into account the human sensitivities in both space and time. A weighted clustering approach is used for quantization which allows flexibility in the choice of weights. Assigning weights proportional to the space gradients and the time gradients is shown to produce better quantization of color image sequences.
|Title of host publication||Image Analysis Applications and Computer Graphics - 3rd International Computer Science Conference ICSC 1995, Proceedings|
|Editors||Roland T. Chin, Ting-Chuen Pong, Avi C. Naiman, Horace H.S. Ip|
|Number of pages||8|
|State||Published - 1995|
|Event||3rd International Computer Science Conference, ICSC 1995 - Hong Kong, Hong Kong|
Duration: 11 Dec 1995 → 13 Dec 1995
|Name||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Conference||3rd International Computer Science Conference, ICSC 1995|
|Period||11/12/95 → 13/12/95|
Bibliographical notePublisher Copyright:
© Springer-Verlag Berlin Heidelberg 1995.
ASJC Scopus subject areas
- Theoretical Computer Science
- Computer Science (all)