DocumentCode
2084930
Title
Self-similarity coding of random texture with applications to design and detection of complicated patterns
Author
Kamejima, Kohji
Author_Institution
Fac. of Eng., Osaka Inst. of Technol., Japan
fYear
1996
fDate
11-14 Nov 1996
Firstpage
370
Lastpage
375
Abstract
A dynamic scheme is presented for identifying generation processes of complicated patterns. Patterns with random textures are filtered to yield a discrete feature pattern. This feature pattern is successively expanded for estimating self-similar partition. A set of reduced affine mappings is specified through joint statistic-geometry analysis of the partition. The coding method is applied to design and detection of complicated patterns
Keywords
computational geometry; fractals; image coding; image texture; pattern recognition; probability; statistical analysis; coding method; complicated pattern recognition; discrete feature pattern; fractal; random texture; reduced affine mappings; self-similar partition; self-similarity coding; statistic-geometry analysis; Design engineering; Fractals; Image analysis; Image converters; Image generation; Layout; Pattern analysis; Surface texture; Testing; Visualization;
fLanguage
English
Publisher
ieee
Conference_Titel
Robot and Human Communication, 1996., 5th IEEE International Workshop on
Conference_Location
Tsukuba
Print_ISBN
0-7803-3253-9
Type
conf
DOI
10.1109/ROMAN.1996.568865
Filename
568865
Link To Document