Track and Field Image Segmentation Technology of Automatic Multi-Threshold Block Sampling Based on Matlab

Abstract:

Based on block sampling and genetic algorithm we improve the image segmentation technology, and obtain the new image processing method based on image pixel block cross reconstruction, and apply this algorithm to the block and the reconstruction calculation of track and field image. In order to verify the effectiveness and reliability of block sampling and genetic algorithm, we use MATLAB to test the algorithm. Through calculation, the number of 2D block is 20, the minimum number of 3D pixel array is 1259, and the calculation residual reaches the highest, with 0.0122. When the number of 2D block is 40, the maximum number of 3D pixel array is 3875, and the calculation residual reaches the lowest, with 0.0013. Therefore, according to the needs of computer hardware conditions, we can increase the 2D block number, improve the segmentation number of 3D pixel array for track and field image, which can reduce the residual and improve the precision of calculation.

• Periodical: Advanced Materials Research (Volumes 989-994)
• Pages: 3856-3860
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.989-994.3856
• Online since: July 2014
• Authors: A Li Yu*
• Keywords: Block Sampling, Cross Compiler, Genetic Algorithm (GA), Image Segmentation, MATLAB Programming, Multi Threshold

References

[1] Lin Zhengchun, Wang Zhiyan, Zhang Yanqing. Threshold segmentation of image optimization [J]. Journal of computer aided design and graphics, 2012, 22(7): 1201-1206.

[2] Qiao Lishan, Chen Songcan, Wang Min. The image threshold technology based on relevance vector machine [J]. Journal of computer research and development, 2010, 47(8): 1329-1337.

[3] He Xiaojuan. Distribution Estimation algorithm and its application in production scheduling [D]. Lanzhou University of Technology, 2011: 2-14.

[4] Cao Song, Sun Fuchun, Hu Laihong. Distribution estimation algorithm based on flight optimization [J]. Journal of Tsinghua University (NATURAL SCIENCE EDITION), 2012, 52(1): 66-71.

[5] Zhuang Jian, Yang Qingyu, Du Haifeng, Yu Dehong. A higher efficient and complex system genetic algorithm [J]. Software, 2011, 3(13): 45-49.

[6] Chen Hao, Cui Duwu. Population evolutionary algorithm [J]. Journal of software, 2010, 21(5): 978-990.

[7] Yue Qian, Feng Shan. Calculation performance analysis of the genetic algorithm [J]. Journal of computer, 2011, 32(12): 2389-2392.

[8] Li Junhua, Li Ming, Yuan Lihua. The pseudo parallel genetic algorithm based on clustering [J]. Pattern recognition and artificial intelligence, 2011, 22(2): 188-194.

[9] Chen Shouwen, Li Mingdong. An improved genetic algorithm and its simulation [J]. Computer applications and software, 2012, 27(9): 100-102.