Title :
Using multi-layer morphological neural network for color images retrieval
Author :
Yun, Zhang ; Ling, Zhang ; Yang Yimin
Author_Institution :
Autom. Inst., Guangdong Univ. of Technol., Guangzhou, China
Abstract :
This paper presents a new way to retrieve color images based on their shape with a multi-layer morphological (MLM) neural network. Using MLM NN based on Species Genetic Algorithm(SGA), the shape feature of the objects in color images is caught for retrieval based on shape match. The following simulation results show that the shape feature caught by SGA is more accurate than by GA, and is more insensitive to objects´ size and orientation.
Keywords :
genetic algorithms; image colour analysis; image retrieval; multilayer perceptrons; color images retrieval; multilayer morphological neural network; objects shape feature; species genetic algorithm; Color; Feedforward neural networks; Image retrieval; Mathematical model; Morphology; Multi-layer neural network; Neural networks; Neurons; Shape; Solid modeling;
Conference_Titel :
Intelligent Control and Automation, 2004. WCICA 2004. Fifth World Congress on
Print_ISBN :
0-7803-8273-0
DOI :
10.1109/WCICA.2004.1342277
