An Infrared Image Segmentation Method Based on Improved Lazy Snapping Algorithm
-
摘要: 针对红外图像含大量噪声以及对比度低等特点,提出一种结合快速模糊C均值聚类的改进Lazy Snapping分割方法。对红外图像使用快速模糊C均值聚类算法进行预分割,通过形态学骨架提取的方法在图像中标记出目标和背景种子点,将Lazy Snapping算法由全局分割转化为聚类区域分割,并构造能量函数,通过最小割算法求解能量函数的最小值并使分割效率得以提升,减少了图像存在的过分割现象,使Lazy Snapping算法由交互式算法变为非交互式算法,实现了红外图像的自动分割,提高了Lazy Snapping算法的实时性。通过对各类不同红外图像进行分割实验,再与其他分割方法进行性能评价比较,结果表明改进的算法具有良好的分割效果及较强的鲁棒性。
-
关键词:
- Lazy Snapping /
- 能量函数 /
- 图像分割 /
- 模糊C均值聚类
Abstract: Considering that infrared images contain a considerable amount of noise and are of low contrast, an improved lazy snapping(LS) segmentation method combined with fast fuzzy C-means clustering is proposed. Infrared images are pre-segmented using a fast fuzzy C-means clustering algorithm, and the target and background seed points are marked in the image by the morphological skeleton extraction method. The LS algorithm is converted from global segmentation to cluster region segmentation, and an energy function is constructed. The minimum value of the energy function is solved by the minimum cut algorithm, and the segmentation efficiency is improved. The phenomenon of over-segmentation in the image is reduced, the LS algorithm is changed from an interactive algorithm to a non-interactive algorithm. Thus, the automatic segmentation of infrared images is realized, improving the real-time nature of the LS algorithm. By performing segmentation experiments on various infrared images and then comparing the proposed method's performance with that of other segmentation methods, the results show that the improved algorithm has a good segmentation effect and strong robustness.-
Key words:
- Lazy Snapping /
- energy function /
- image segmentation /
- fuzzy C-means clustering
-
表 1 各类红外图像分割效果
Table 1. Various infrared image segmentation effects
Image type Global threshold segmentation Standard Lazy Snapping The algorithm of this paper IOU FPR IOU FPR IOU FPR Isolation switch 0.8551 0.1177 0.9553 0.0199 0.9765 0.0170 Outlet 0.6037 0.3961 0.8993 0.0935 0.9384 0.0613 Circuit breaker 0.8413 0.0361 0.9353 0.0558 0.9412 0.0531 表 2 算法运行时间比较
Table 2. Comparison of algorithm running times
Image type Standard Lazy Snapping The algorithm ofthis paper Isolation switch 1.0723 1.1319 Outlet 0.9756 1.0170 Circuit breaker 0.9862 0.9733 -
[1] Rafael C. MATLAB Implementation of Digital Image Processing[M]. Beijing: Tsinghua University Press, 2013. [2] 张锦文. 变电站电气设备红外图像分割方法研究[D]. 北京: 华北电力大学, 2018.ZHANG Jinwen. Research on Infrared Image Segmentation Method of Electrical Equipment in Substation[D]. Beijing: North China Electric Power University, 2018. [3] 徐鹏飞, 张菁, 尹腾飞, 等. 基于改进PCNN算法的电力设备图像分割研究[J]. 智能计算机与应用, 2019, 9(3): 59-62, 68. https://www.cnki.com.cn/Article/CJFDTOTAL-DLXZ201903012.htmXU Pengfei, ZHANG Jing, YIN Tengfei, et al. Research on image segmentation of power equipment based on improved PCNN algorithm[J]. Intelligent Computers and Applications, 2019, 9(3): 59-62, 68. https://www.cnki.com.cn/Article/CJFDTOTAL-DLXZ201903012.htm [4] 王智杰, 牛硕丰, 刘相兴, 等. 蝙蝠算法优化二维熵的变电设备红外图像分割应用研究[J]. 电子设计工程, 2018, 26(18): 83-87. doi: 10.3969/j.issn.1674-6236.2018.18.018WANG Zhijie, NIU Shuofeng, LIU Xiangxing, et al. Application research on infrared image segmentation of substation equipment based on bat algorithm to optimize two-dimensional entropy[J]. Electronic Design Engineering, 2018, 26(18): 83-87. doi: 10.3969/j.issn.1674-6236.2018.18.018 [5] 李鑫, 崔昊杨, 霍思佳, 等. 基于粒子群优化法的Niblack电力设备红外图像分割[J]. 红外技术, 2018, 40(8): 780-785. http://hwjs.nvir.cn/article/id/hwjs201808010LI Xin, CUI Haoyang, HUO Sijia, et al. Niblack power equipment infrared image segmentation based on particle swarm optimization method[J]. Infrared Technology, 2018, 40(8): 780-785. http://hwjs.nvir.cn/article/id/hwjs201808010 [6] 樊淑炎, 丁世飞. 基于多尺度的改进Graph cut算法[J]. 山东大学学报: 工学版, 2016, 46(1): 28-33. https://www.cnki.com.cn/Article/CJFDTOTAL-SDGY201601005.htmFAN Shuyan, DING Shifei. Improved graph cut algorithm based on multi-scale[J]. Journal of Shandong University: Engineering Edition, 2016, 46(1): 28-33. https://www.cnki.com.cn/Article/CJFDTOTAL-SDGY201601005.htm [7] Boykov Y, Funka L G. Graph cut and efficient N-D image segmentation[J]. International Journal of Computer Vision, 2006, 70(2): 109-131. doi: 10.1007/s11263-006-7934-5 [8] 郑加明, 陈昭炯. 局部颜色模型的交互式Graph-Cut分割算法[J]. 智能系统学报, 2011, 6(4): 318-323. doi: 10.3969/j.issn.1673-4785.2011.04.006ZHENG Jiaming, CHEN Zhaojiong. Interactive graph-cut segmentation algorithm of local color model[J]. Journal of Intelligent Systems, 2011, 6(4): 318-323. doi: 10.3969/j.issn.1673-4785.2011.04.006 [9] 刘松涛, 殷福亮. 基于图割的图像分割方法及其新进展[J]. 自动化学报, 2012, 38(6): 911-922. https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO201206003.htmLIU Songtao, YIN Fuliang. Image segmentation method based on graph cut and its new development[J]. Acta Automatica Sinica, 2012, 38(6): 911-922. https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO201206003.htm [10] Kohli P, Torr P. Dynamic graph cuts for efficient inference in markov random fields[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(12): 2079-2008. doi: 10.1109/TPAMI.2007.1128 [11] 周兵, 韩媛媛, 徐明亮, 等. 快速非局部均值图像去噪算法[J]. 计算机辅助设计与图形学学报, 2016, 28(8): 1260-1268. doi: 10.3969/j.issn.1003-9775.2016.08.007ZHOU Bing, HAN Yuanyuan, XU Mingliang, et al. Fast non-local mean image denoising algorithm[J]. Journal of Computer Aided Design and Graphics, 2016, 28(8): 1260-1268. doi: 10.3969/j.issn.1003-9775.2016.08.007 [12] LAN Rong, FAN Jiulun, LIU Ying, et al. Image thresholding by maximizing the similarity degree based on intuitionistic fuzzy sets[C]//Quantitative Logic and Soft Computing, Hangzhou, 2016: 631-640. [13] KAPIL S, CHAWLA M, ANSARI M D. On K-means data clustering algorithm with genetic algorithm[C]//Fourth International Conference on Parallel, Distributed and Grid Computing, 2016: 202-206. [14] LI Yin, SUN Jian, TANG Chi-Keung, et al. Lazy snapping[J]. ACM Transactions on Graphics (TOG), 2004, 23(3): 303-308. doi: 10.1145/1015706.1015719 [15] BAI X, CHEN Z, ZHANG Y, et al. Infrared ship target segmentation based on spatial information improved FCM[J]. IEEE Transactions on Cybernetics, 2016, 46(12): 3259-3271. doi: 10.1109/TCYB.2015.2501848 [16] LIN K, HUNG K, LIN C. Rule generation based on novel kernel intuitionistic fuzzy rough set model[J]. IEEE Access, 2018(6): 11953-11958. http://ieeexplore.ieee.org/document/8302495 [17] KAUSHAL M, SOLANKI R, LOHANI Q M D, et al. A novel intuitionistic fuzzy set generator with application to clustering[C]//2018IEEE International Conference on Fuzzy Systems, Piscataway, 2018: 1-8. [18] ZHAO F. FCM clustering with non local-spatial information for noisy image segmentation[J]. Frontiers of Computer Science in China, 2011, 5(1): 45-56. doi: 10.1007/s11704-010-0393-8 [19] CHEN S C, ZHANG D Q. Robust image segmentation using FCM with spatial constraints based on new kernel-induced distance measure[J]. IEEE Transactions on System, Man and Cybernetics, Part B: Cybernetics, 2004, 34(4): 1907-1916. doi: 10.1109/TSMCB.2004.831165 [20] 李云松, 冯玉东, 张国锋. 基于快速模糊C均值聚类的图像粗集分割[J]. 兰州理工大学学报, 2013, 39(1): 92-96. https://www.cnki.com.cn/Article/CJFDTOTAL-GSGY201301022.htmLI Yunsong, FENG Yudong, ZHANG Guofeng. Image rough set segmentation based on fast fuzzy C-means clustering[J]. Journal of Lanzhou University of Technology, 2013, 39(1): 92-96. https://www.cnki.com.cn/Article/CJFDTOTAL-GSGY201301022.htm