Title :
Improved Morphological TOP-HAT Filter Optimized with Genetic Algorithm
Author :
Wang, Jiangang ; Gao, Deyuan
Author_Institution :
Coll. of Comput., Northwestern Poly Tech. Univ., Xi´´an, China
Abstract :
The Top-Hat morphological filters are a class of nonlinear signal processing algorithms, which have been applied extensively to computer vision, image processing, and more recently target detection. In this paper a novel method for optimal learning of morphological filtering parameters for spot target detection is presented. We show how the genetic algorithms can be used for an automatic optimization of structuring elements. Experimental results show that the identified probability to the image of SNR 2 can reach more than 98% by this method.
Keywords :
genetic algorithms; image segmentation; infrared imaging; mathematical morphology; nonlinear filters; object detection; TOP-HAT morphological filter; automatic infrared image target detection; genetic algorithm; nonlinear signal processing algorithm; spot target detection; Bismuth; Computer vision; Filtering; Filters; Genetic algorithms; Image processing; Morphology; Object detection; Shape measurement; Signal processing algorithms;
Conference_Titel :
Image and Signal Processing, 2009. CISP '09. 2nd International Congress on
Conference_Location :
Tianjin
Print_ISBN :
978-1-4244-4129-7
Electronic_ISBN :
978-1-4244-4131-0
DOI :
10.1109/CISP.2009.5303727
