融合显著性特征的自适应主动轮廓模型

doi:10.3778/j.issn.1002-8331.2109-0495

摘要/Abstract

摘要： 主动轮廓模型存在演化速度慢、对初始轮廓和噪声敏感、弱边缘泄漏及目标过分割等问题。对以上问题进行了研究，提出了融合显著性特征的自适应主动轮廓模型。提出基于去雾算法的显著性映射作为正则项提升模型对初始轮廓位置的鲁棒性，防止轮廓演化过程过早陷入局部最优解，同时缩短轮廓演化时间。为了防止模型在演化过程中出现弱边界泄漏，模型中引入边缘检测函数作为能量泛函的权重。该模型利用最大面积稀疏约束，提出自适应目标提取方法来消除目标过分割影响。与多种主动轮廓模型在数据集MRSA500（500张）上进行实验对比，表明了提出的模型对初始轮廓和噪声的鲁棒性，而且提出模型的平均分割效率提升约5.6倍，平均Jaccard相似度系数提升约22%。

关键词: 主动轮廓模型, 图像分割, 图像增强, 显著性, 稀疏约束

Abstract: Active contour model exits some problems including slow evolution, sensitivity to initial contour and noise, weak edge leakage and target over-segmentation. The above problems are studied, an adaptive target extraction active contour model combined with image salient features（ASF） is proposed. Firstly, a saliency detection based on the defogging algorithm is used as a regularization term to improve the robustness of the model to the initial contour position to prevent the contour evolution process from falling into the local optimal solution prematurity meanwhile shorten the contour evolution time. Secondly, in order to prevent weak boundary leakage during the evolution of the model, the edge detection function is introduced into the model as the weight of the energy functional. Finally, oriented on the maximum area sparse constraint, an adaptive target extraction method is proposed to eliminate the over-segmentation effect of the target. Experimental comparison with various active contour models on MRSA500 data set shows that the proposed model is robust to the initial contour and noise, the average segmentation efficiency of the proposed model is improved by about 5.6 times, and the average Jaccard similarity coefficient is improved by about 22%.

Key words: active contour model, image segmentation, image enhancement, saliency, sparse constrain

刘国奇, 蒋优, 常宝方, 茹琳媛, 宋一帆, 李旭升. 融合显著性特征的自适应主动轮廓模型[J]. 计算机工程与应用, 2023, 59(5): 312-320.

LIU Guoqi, JIANG You, CHANG Baofang, RU Linyuan, SONG Yifan, LI Xusheng. Adaptive Active Contour Model Combined with Salient Features[J]. Computer Engineering and Applications, 2023, 59(5): 312-320.

参考文献

[1] NIAZ A，RANA K，JOSHI A，et al.Hybrid active contour based on local and global statistics parameterized by weight coefficients for inhomogeneous image segmentation[J].IEEE Access，2020，8：57348-57362.
[2] LI C，XU C，GUI C，et al.Distance regularized level set evolution and its application to image segmentation[J].IEEE Transactions on Image Processing，2010，19（12）：3243-3254.
[3] LI C，XU C，GUI C，et al.Level set evolution without re-initialization：a new variational formulation[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition（CVPR’05），2005：430-436.
[4] HUANG C，ZENG L.Level set evolution model for image segmentation based on variable exponent p-Laplace equation[J].Applied Mathematical Modelling，2016，40（17/18）：7739-7750.
[5] MIAO J，HUANG T Z，ZHOU X，et al.Image segmentation based on an active contour model of partial image restoration with local cosine fitting energy[J].Information Sciences，2018，447：52-71.
[6] MUNIR A，SOOMRO S，LEE C H，et al.Adaptive active contours based on variable kernel with constant initiali-zation[J].IET Image Processing，2018，12（7）：1117-1123.
[7] LI C，KAO C Y，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.
[8] 刘国奇，董一飞.基于超像素/像素协同约束和稀疏分解活动轮廓模型[J].计算机应用研究，2021，38（3）：946-951.
LIU G Q，DONG Y F，et al.Active contour model based on joint constraint with superpixel/pixel and sparse decomposition[J].Application Research of Computer，2021，38（3）：946-951.
[9] ROUHI R，JAFARI M.Classification of benign and malignant breast tumors based on hybrid level set segmentation[J].Expert Systems with Applications，2016，46：45-59.
[10] CHAN T F，VESE L A.Active contours without edges[J].IEEE Transactions on Image Processing，2001，10（2）：266-277.
[11] MUMFORD D B，SHAH J.Optimal approximations by piecewise smooth functions and associated variational problems[J].Communications on Pure and Applied Mathe-matics，1989，42（5）：577-685.
[12] MIN H，LU J，JIA W，et al.An effective local regional model based on salient fitting for image segmentation[J].Neurocomputing，2018，311：245-259.
[13] ZHI X H，SHEN H B.Saliency driven region-edge-based top down level set evolution reveals the asynchronous focus in image segmentation[J].Pattern Recognition，2018，80：241-255.
[14] LI C，HUANG R，DING Z，et al.A level set method for image segmentation in the presence of intensity inhomogeneities with application to MRI[J].IEEE Transactions on Image Processing，2011，20（7）：2007-2016.
[15] HUANG G，JI H，ZHANG W.A fast level set method for inhomogeneous image segmentation with adaptive scale parameter[J].Magnetic Resonance Imaging，2018，52：33-45.
[16] ZHANG Y，FOSTER R，SUN X，et al.Restraint stress in duces lymphocyte reduction through P53 and PI3K/NF-KB pathways[J].Journal of Neuroimmunology，2008，200（1/2）：71-76.
[17] LIU G，ZOU J.Level set evolution with sparsity constraint for object extraction[J].IET Image Processing，2018，12（8）：1413-1422.
[18] 潘沛鑫，潘中良.结合显著性的主动轮廓图像分割[J].计算机工程与应用，2021，57（8）：225-230.
PAN P X，PAN Z L.Active contour image segmentation combined with saliency[J].Computer Engineering and Applications，2021，57（8）：225-230.
[19] CHEN C，WEI J，PENG C，et al.Improved saliency detection in RGB?D images using two-phase depth estimation and selective deep fusion[J].IEEE Transactions on Image Processing，2020，29：4296-4307.
[20] HE K，SUN J.TANG X.Single image haze removal using dark channel prior[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2010，33（12）：2341-2353.
[21] LIU H，FANG J，ZHANG Z，et al.A novel active contour model guided by global and local signed energy-based pressure force[J].IEEE Access，2020，8：59412-59426.
[22] CHENG M M，MITRA N J，HUANG X，et al.Global contrast based salient region detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2014，37（3）：569-582.