Title :
Fast texture segmentation using genetic programming
Author :
Song, Andy ; Ciesielski, Vic
Author_Institution :
Sch. of Comput. Sci. & Inf. Technol., RMIT Univ., Melbourne, Vic., Australia
Abstract :
This paper presents a method which extends the use of genetic programming (GP) to a complex domain, texture segmentation. By this method, segmentation tasks are performed by texture classifiers which are evolved by the GP. Small cutouts sampled from images of various textures are used for the evolution. The generated classifiers directly use pixel values as input. Based on these classifiers an algorithm which uses a voting strategy to partition texture regions is developed. The results of the investigation indicate that the proposed method is able to accurately identify the boundaries between different texture regions, even if the boundaries are not regular. The method can segment two textures as well as multiple textures. Furthermore, fast segmentation can be achieved. The speed of the proposed texture segmentation method can be a hundred times faster than conventional methods.
Keywords :
genetic algorithms; image segmentation; image texture; complex domain; fast segmentation; genetic programming; pixel values; texture classifiers; texture region partitioning; texture segmentation; voting strategy; Australia; Computer science; Dynamic range; Feature extraction; Genetic programming; Image segmentation; Image texture analysis; Information technology; Partitioning algorithms; Voting;
Conference_Titel :
Evolutionary Computation, 2003. CEC '03. The 2003 Congress on
Print_ISBN :
0-7803-7804-0
DOI :
10.1109/CEC.2003.1299935