基于改进活动轮廓模型的井筒病害识别方法
|
摘要
针对井筒图像对比度小、干扰噪声多、分割精度不高、识别效率低等问题,提出了一种基于图像增强与正则约束的改进活动轮廓模型。图像增强算子能够改善井筒图像的对比度,扩大灰度动态范围,抑制噪声干扰。正则约束因子,包括长度约束因子、面积约束因子和距离函数约束因子,能够减少初始轮廓对曲线演化的影响,实现轮廓曲线平滑快速向目标边界移动,并最终与目标边缘吻合。实验结果表明,本文模型在算法性能和分割效果上都优于C-V模型和LBF模型,能够快速准确识别井筒病害,提高井筒巡检的自动化程度。
Aiming at the problems in the images of mine shaft,such as low image contrast,too much image noises,low segmentation accuracy and image recognition efficiency,the paper proposes an improved active contour model based on image enhancement and regularization constraint. The image enhancement can improve the image contrast,expand the dynamic range of gray value,and reduce the noise effect. The regular constraints comprise the length constraint,the area constraint and the distance function constraint,which can reduce the influence of initial contour on the curve evolution,realize smooth and quick moving of the contour curve to target boundary, and end up with the edge of the target.Experimental results show that,the proposed model is superior to the C- V model and LBF model in the performance and segmentation results,which can quickly and accurately recognize the mine shaft failure, and improve the automation degree of mine shaft inspection.
Aiming at the problems in the images of mine shaft,such as low image contrast,too much image noises,low segmentation accuracy and image recognition efficiency,the paper proposes an improved active contour model based on image enhancement and regularization constraint. The image enhancement can improve the image contrast,expand the dynamic range of gray value,and reduce the noise effect. The regular constraints comprise the length constraint,the area constraint and the distance function constraint,which can reduce the influence of initial contour on the curve evolution,realize smooth and quick moving of the contour curve to target boundary, and end up with the edge of the target.Experimental results show that,the proposed model is superior to the C- V model and LBF model in the performance and segmentation results,which can quickly and accurately recognize the mine shaft failure, and improve the automation degree of mine shaft inspection.
引文
[1]刘环宇,王思敬,曾钱帮,等.基于模糊神经网络兖州矿区竖井井筒非采动破裂的判别[J].岩土工程学报,2005(27):1237-1240.
[2]周晓敏,胡启胜,马成炫,等.围岩对竖井井壁承载能力影响的对比研究[J].煤炭学报,2012(7):57-59.
[3]Kass M,Witkin A,Terzopoulos D.Snakes:Active Contour Models[J].International Journal of Computer Vision,1987:321-331.
[4]Bashir A D,Andrew H,David S.An active contour model for segmenting and measuring retinal vessels.[J].Medical Imaging IEEE Transactions on,2009,28(9):1488-97.
[5]Chan T F,Vese L A.Active contours without edges.[J].Image Processing IEEE Transactions on,2001,10(2):266-277.
[6]Chunming Li,Chiu-yen Kao,John C.Gore,et al.Implicit Active Contours Driven by Local Binary Fitting Energy[C]//Computer Vision and Pattern Recognition,2007.CVPR'07.IEEE Conference on IEEE,2007:1-7.
[7]朱晓舒,孙权森,夏德深,等.基于核函数的活动轮廓模型[J].计算机辅助设计与图形学学报,2015(27):388-393.
[8]孙传猛,曹树刚,李勇.基于LBF改进模型的煤岩细观结构研究[J].煤炭学报,2015,40(2):331-341.
[9]Abdelsamea M M,Gnecco G,Gaber M M.An efficient SelfOrganizing Active Contour model for image segmentation[J].Neurocomputing,2015,149:820-835.
[10]Hashemi S,Kiani S,Noroozi N,et al.An image contrast enhancement method based on genetic algorithm[J].Pattern Recognition Letters,2010,31(13):1816-1824.
[11]Sawant M,Deore M.A Comprehensive Review of Image Enhancement Techniques[J].International Journal of Computer Technology&Electronics Engineering,2011,1(2):39-44.
[12]Kropac O,Mucka P.Specification of Obstacles in the Longitudinal Road Profile by Median Filtering[J].Journal of Transportation Engineering,2014,137(3):214-226.
[2]周晓敏,胡启胜,马成炫,等.围岩对竖井井壁承载能力影响的对比研究[J].煤炭学报,2012(7):57-59.
[3]Kass M,Witkin A,Terzopoulos D.Snakes:Active Contour Models[J].International Journal of Computer Vision,1987:321-331.
[4]Bashir A D,Andrew H,David S.An active contour model for segmenting and measuring retinal vessels.[J].Medical Imaging IEEE Transactions on,2009,28(9):1488-97.
[5]Chan T F,Vese L A.Active contours without edges.[J].Image Processing IEEE Transactions on,2001,10(2):266-277.
[6]Chunming Li,Chiu-yen Kao,John C.Gore,et al.Implicit Active Contours Driven by Local Binary Fitting Energy[C]//Computer Vision and Pattern Recognition,2007.CVPR'07.IEEE Conference on IEEE,2007:1-7.
[7]朱晓舒,孙权森,夏德深,等.基于核函数的活动轮廓模型[J].计算机辅助设计与图形学学报,2015(27):388-393.
[8]孙传猛,曹树刚,李勇.基于LBF改进模型的煤岩细观结构研究[J].煤炭学报,2015,40(2):331-341.
[9]Abdelsamea M M,Gnecco G,Gaber M M.An efficient SelfOrganizing Active Contour model for image segmentation[J].Neurocomputing,2015,149:820-835.
[10]Hashemi S,Kiani S,Noroozi N,et al.An image contrast enhancement method based on genetic algorithm[J].Pattern Recognition Letters,2010,31(13):1816-1824.
[11]Sawant M,Deore M.A Comprehensive Review of Image Enhancement Techniques[J].International Journal of Computer Technology&Electronics Engineering,2011,1(2):39-44.
[12]Kropac O,Mucka P.Specification of Obstacles in the Longitudinal Road Profile by Median Filtering[J].Journal of Transportation Engineering,2014,137(3):214-226.