Study of Relative Entropy Coefficients for Image Segmentation Based on Particle Swarm Optimization

Abstract:

Article Preview

Image segmentation is the basis of image analysis, and because of its simplicity, rapidity and stability, the threshold method is the important one, applying in the image processing and recognition widely. In this paper, a new method is proposed, which based on relative entropy coefficients between random variables. It maximizes the target and background, which is the relative entropy coefficient in probability distribution, and gets the optimal threshold of image segmentation, and then optimizes it using particle swarm algorithm which is an evolutionary computation algorithm. The result of relative entropy coefficients for image segmentation proves its feasibility and better effect.

Info:

Periodical:

Edited by:

Qi Luo

Pages:

1056-1060

DOI:

10.4028/www.scientific.net/AMM.58-60.1056

Citation:

Y. R. Huang and L. G. Qu, "Study of Relative Entropy Coefficients for Image Segmentation Based on Particle Swarm Optimization", Applied Mechanics and Materials, Vols. 58-60, pp. 1056-1060, 2011

Online since:

June 2011

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.