Affiliation:
1. South China University of Technology
Abstract
In order to improve the real-time performance and accuracy of the traditional SRG(Seeded Region Growing) algorithm in image processing, this paper proposes a intellective and rapid image segmentation by imitating the process of the virus infection in nature, and then implement it on vc++6 platform. On one hand , the algorithm can detecting automatically detect the seeds in image region and can be adapt for uneven-light image by adjusting the parameters based on the brightness of the background; On the other hand, only by one of the image scanning, it can segment and mark the objects from the background. The experimental results show that compared with the traditional SRG algorithm, this algorithm can improve the segmentation speed in different background with higher accuracy.
Publisher
Trans Tech Publications, Ltd.
Subject
Mechanical Engineering,Mechanics of Materials,General Materials Science
Reference7 articles.
1. Frank Y. Shih and S.X. Cheng: Automatic seeded region growing for color images segmentation [J]. Image and Vision Computing, 2005, (10), p, 877-886.
2. J.P. Fan, G.H. Zeng and Body Mathurin: Seeded Region Growing: an extensive and comparative Study [J]. Pattern Recognition Letters, 2005, (8), p, 113 9-1156.
3. Takanashi Tetsuya and Shin Jungpil: Color image segmentation based on region growing algorithm [J]. Journal of Convergence Information Technology, 2012, (16), p, 152-160.
4. J.N. Wang, J. Kong and Y.H. Lu: A region-based SRG algorithm for color image segmentation [J]. Proceedings of the Sixth International Conference on Machine Learning and Cybernetics, 2007, (3), p, 1524-1547.
5. H. Lu and Y. Wen: Region growing algorithm in PCB element segmentation [J]. Journal of Chinese Computer Systems, 2007, (8), p, 1489-1491.