集成边缘和区域信息的格子波尔兹曼模型图像分割

温军玲; 严壮志; 孙玉彪; 林笑曼

doi:10.3969/j.issn.1671-7775.2013.06.012

PDF(2170 KB)

江苏大学学报（自然科学版） ›› 2013, Vol. 34 ›› Issue (6) : 687-692. DOI: 10.3969/j.issn.1671-7775.2013.06.012

论文

集成边缘和区域信息的格子波尔兹曼模型图像分割

作者信息 +

Image segmentation of integrated edge and region information by lattice Boltzmann model

Author information +

文章历史 +

摘要

针对传统水平集方法计算耗时、需要重新初始化等问题,提出一种基于格子波尔兹曼(LB)模型的图像分割方法．该方法引入图像的边缘信息作为LB演化方程的松弛因子,引入区域统计信息作为LB的有源项,二者作为内力和外力共同控制活动轮廓的演化．为了验证该模型,分别对合成图像和真实图像进行了分割试验,并与活动轮廓模型的GAC和CV模型进行了比较,采用面积交迭度和误分率两种指标进行客观评价．结果表明,所提出的算法具有编程简单、计算快速、无需重新初始化的优点,且能实现可选择的全局或局部分割．

Abstract

To avoid the problems of heavy calculating work and needing re-initialization of traditional Level Set method, a novel lattice Boltzmann (LB) method was proposed for image segmentation. The edge information and the region statistical information were introduced respectively as relaxing factor and source term of LB equation to control the evolution of active contour as both internal and external forces. Synthetic images and real images were experimented to verify the model. The proposed model was compared with geodesic active contour(GAC) model and chan-vese(CV) model. Area overlap measure and misclassified error were used as evaluation indexes to discriminate the results. The experimental results show that the algorithm is simple and fast with high efficiency, and can avoid re-initialization. The global or local image segmentation can be selectively realized in the proposed method.

导出引用

温军玲, 严壮志, 孙玉彪, 等. 集成边缘和区域信息的格子波尔兹曼模型图像分割[J]. 江苏大学学报（自然科学版）, 2013, 34(6): 687-692 https://doi.org/10.3969/j.issn.1671-7775.2013.06.012

WEN Jun-Ling, YAN Zhuang-Zhi, SUN Yu-Biao, et al. Image segmentation of integrated edge and region information by lattice Boltzmann model[J]. Journal of Jiangsu University(Natural Science Edition), 2013, 34(6): 687-692 https://doi.org/10.3969/j.issn.1671-7775.2013.06.012

参考文献

［1］Chan T F, Vese L A. Active contours without edges ［J］. IEEE Transactions on Image Processing, 2001,10(2): 266-277.
［2］王大凯,候榆青,彭进业.图像处理的偏微分方程方法［M］.北京:科学出版社,2008.
［3］Zhao Shuying, Zhang Dan, Song Xiangman, et al. Segmentation of hippocampus in MRI images based on the improved level set［C］∥Proceedings of 2011 4th International Symposium on Computational Intelligence and Design. Hangzhou, China:IEEE Computer Society, 2011:123-126.
［4］Balla-Arabé S, Wang Bin, Gao Xinbo. Level set region based image segmentation using lattice boltzmann me-thod［C］∥Proceedings of 2011 Seventh International Conference on Computational Intelligence and Security. Sanya,China:IEEE Computer Society, 2011:1159-1163.
［5］Balla-Arabé S, Gao Xinbo. Image multi-thresholding by combining the lattice Boltzmann model and a localized level set algorithm［J］. Neurocomputing, 2012,93(9):106-114.
［6］Wang Zhiqiang, Yan Zhuangzhi, Chen George. Lattice Boltzmann method of active contour for image segmentation［C］∥Proceedings of Sixth International Conference on Image and Graphics. Hefei, China:IEEE Computer Society, 2011:338-343.
［7］Chen Yu, Yan Zhuangzhi, Chu Yungao. Cellular auto-mata based level set method for image segmentation［C］∥Proceedings of 2007 IEEE／ICME International Conference on Complex Medical Engineering. Beijing, China:IEEE, 2007:171-174.
［8］Zhang Kaihua, Zhang Lei, Song Huihui, et al. Active contours with selective local or global segmentation: a new formulation and level set method ［J］. Image and Vision Computing, 2010,28(4): 668-676.
［9］Rosen J B. The gradient projection method for nonlinear programming. part II. nonlinear constraints ［J］. Journal of the Society for Industrial and Applied Mathema-tics,1961,9(4):514-532.