結(jié)合局部能量與改進的符號距離正則項的圖像目標分割算法
doi: 10.11999/JEIT141473
-
2.
(石家莊學(xué)院計算機系 石家莊 050035) ②(燕山大學(xué)信息科學(xué)與工程學(xué)院河北省計算機虛擬技術(shù)與系統(tǒng)集成重點實驗室 秦皇島 066004) ③(河北科技大學(xué)信息科學(xué)與工程學(xué)院 石家莊 050018)
基金項目:
河北省自然科學(xué)基金(F2012208004),河北省教育廳高等學(xué)??茖W(xué)研究計劃自然科學(xué)重點項目(ZD20132013)和河北省科技支撐計劃項目(14210302D)
Local Energy Information Combined with Improved Signed Distance Regularization Term for Image Target Segmentation Algorithm
-
摘要: 針對傳統(tǒng)C-V模型對顏色不均勻圖像分割失敗并且對初始輪廓和位置敏感問題,以及現(xiàn)有符號距離正則項存在周期性振蕩和局部極值問題。該文提出結(jié)合局部能量信息和改進的符號距離正則項的圖像目標分割算法。首先,將全局圖像信息擴展到HSV空間,并使用局部能量項信息分析每個像素及其領(lǐng)域內(nèi)的統(tǒng)計特性,從而在較少的迭代次數(shù)內(nèi)有效分割顏色分布不均勻圖像。其次,改進現(xiàn)有符號距離正則項,改進后的符號距離正則項在避免水平集函數(shù)的重新初始化的同時,提高了計算效率,保證了水平集函數(shù)演化過程的穩(wěn)定性。然后,定義閾值判斷法的水平集函數(shù)演化的終止準則,使曲線準確演化到目標輪廓。該算法與同類模型的對比實驗表明該模型具有較高的分割精度和對初始輪廓的魯棒性。Abstract: The uneven color image can not be segmented successfully with the traditional C-V model, and the C-V model is sensitive to the initial contour and the location. The existing signed distance regularization term has disadvantages, such as the periodic oscillation and the local extremum. This paper proposes the target segmentation algorithm, which combines the local energy information with improved signed distance regularization term. Firstly, the global image information can be expanded to the HSV space, and each pixels and its statistical properties are analyzed with the local energy information within the neighborhood, which can effectively realize the uneven distribution of color image segmentation in less iteration. Secondly, the improved signed distance regularization term avoids re-initialization of level set function, improving the computational efficiency, and maintains stability in the level set function evolution process. Finally, the termination criterion of threshold evaluation method for the level set function evolution is defined, in order to make the curve accurately evolution to the target contour. The experimental results show that the proposed algorithm has higher segmentation accuracy and robust than other similar models.
-
Key words:
- Image processing /
- Local energy /
- Signed distance regularization term /
- Level set evolution /
- C-V model
-
Shi Y and Karl W C. Real-time tracking using level sets[C]. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Diego, USA, 2005: 34-41. Kiran Thapaliyaa, Jae-Young Pyuna, Chun-Su Parkb, et al.. Level set method with automatic selective local statistics for brain tumor segmentation in MR images[J]. Computerized Medical Imaging and Graphics, 2013, 37(1): 522-537. Jiang Xin, Zhang Ren-jie, and Nie Sheng-dong. Image segmentation based on level set method[J]. Physics Procedia, 2012, 33(6): 840-845. Vese L and Chan T. A multiphase level set frame work for image segmentation using the mumford and shah model[J]. Journal of Computer Vision, 2002, 50(3): 271-293. Wang Hui and Huang Ting-zhu. An adaptive weighting parameter estimation between local and global intensity fitting energy for image segmentation[J]. Communication Nonlinear Science Numeral Simulation, 2014, 19(2): 3098-3105. 戚世樂, 王美清. 結(jié)合全局和局部信息的兩階段活動輪廓模型[J]. 中國圖象圖形學(xué)報, 2014, 19(3): 421-427. Qi Shi-le and Wang Mei-qing. Two-stage active contour model driven by local and global information[J]. Journal of Image and Graphics, 2014, 19(3): 421-427. Li Chun-ming, Kao Chiu-Yen, Gore J C, et al.. Minimization of region-scalable fitting energy for image segmentation[J]. IEEE Transactions on Image Processing, 2008, 17(10): 1940-1949. Mumford D and Shah J. Optimal approximations by piecewise smooth functions and associated variational problems[J]. Communications on Pure and Applied Mathematics, 1989, 42(5): 577-585. Zhao Min-rong, Zhang Xi-wen, and Jiang Juan-na. Topography image segmentation based on improved Chan-Vese model[J]. Computer Aided Drafting, Design and Manufacturing, 2013, 23(2): 13-17. 葛琦, 韋志輝, 肖亮, 等. 基于局部特征的自適應(yīng)快速圖像分割模型[J]. 計算機研究與發(fā)展, 2013, 50(4): 815-822. Ge Qi, Wei Zhi-hui, Xiao Liang, et al.. Adaptive fast image segmentation model based on local feature[J]. Journal of Computer Research and Development, 2013, 50(4): 815-822. 劉存良, 潘振寬, 鄭永果, 等. 兩種保持符號距離函數(shù)的水平集分割方法[J].吉林大學(xué)學(xué)報(工學(xué)版), 2013, 43(增刊): 115-120. Liu Cun-liang, Pan Zhen-kuan, Zheng Yong-guo, et al.. Two algorithms for level set method preserving signed distance functions[J]. Journal of Jilin University (Engineering and Technology Edition), 2013, 43(suppl.): 115-120. Ahmed D, Kamal H, and Moussa D. Fast multilevel thresholding for image segmentation through a multiphase level set method[J]. Signal Processing, 2013, 93(7): 139-153. 王青平, 趙宏宇, 吳微微, 等. 融合局部和非局部信息的自適應(yīng)貝葉斯分割方法[J].電子與信息學(xué)報, 2014, 36(4): 1003-1007. Wang Qing-ping, Zhao Hong-yu, Wu Wei-wei, et al.. An adaptive Bayesian segmentation method fused of local and non-local information[J]. Journal of Electronics Information Technology, 2014, 36(4): 1003-1007. Wang L, Li C M, Sun Q S, et al.. Active contours driven by local and global intensity fitting energy with application to brain MR image segmentation[J]. Computerized Medical Imaging and Graphics, 2009, 33(7): 520-531. Annupan R and Stanislav S M. Multi-feature gradient vector ow snakes for adaptive segmentation of the ultrasound images of breast cancer[J]. Journal of Visual Communication and Image Representation, 2013, 24(4): 1414-1430. -
計量
- 文章訪問數(shù): 1398
- HTML全文瀏覽量: 129
- PDF下載量: 521
- 被引次數(shù): 0