Title :
Image coding with fuzzy image segmentation
Author :
Kong, Seong-Gon ; Kosko, Bart
Author_Institution :
Univ. of Southern California, Los Angeles, CA, USA
Abstract :
A system of fuzzy rules can segment images to provide high-compression image coding. Segmentation-based image coding can achieve higher compression rates than information-theoretic imaging coding. Fuzzy image segmentation generalizes region-growing segmentation. The fuzzy rules use high-frequency information, computed with local edge-detection operators, to remove false contours in segmented images. The fuzzy segmentation-based image coding achieved faithful decoded images at more than 50:1 compression
Keywords :
edge detection; fuzzy set theory; image coding; image segmentation; fuzzy image segmentation; fuzzy rules; high-compression image coding; high-frequency information; local edge-detection operators; Decoding; Fuzzy systems; Humans; Image coding; Image processing; Image reconstruction; Image segmentation; Information theory; Pixel; Signal processing;
Conference_Titel :
Fuzzy Systems, 1992., IEEE International Conference on
Conference_Location :
San Diego, CA
Print_ISBN :
0-7803-0236-2
DOI :
10.1109/FUZZY.1992.258620
