Please wait a minute...
浙江大学学报(工学版)
计算机科学技术     
基于抽风矢量场的深度凹陷图像分割算法
孔勇奇1, 潘志庚2
1. 浙江商业职业技术学院 信息技术学院,浙江 杭州 310053;2. 杭州师范大学 数字媒体与人机交互研究中心, 浙江 杭州 310036
Segmentation algorithm of recessed image based on vector field of suction
KONG Yong-qi1, PAN Zhi-geng2
1. School of Information Technology, Zhejiang Vocational College of Commerce, Hangzhou 310053,China; 2. Digital Media and HCI Research Center, Hangzhou Normal University, Hangzhou 310036, China
 全文: PDF(4183 KB)   HTML
摘要:

为了解决深度凹陷图像分割中的分割瓶颈问题,提出一种基于抽风矢量场的深度凹陷图像分割算法.该算法对凹陷图像进行初次图像分割,获取初次分割曲线及图像梯度矩阵;依据图像梯度矢量的分布规律,对分割曲线的控制节点进行分割瓶颈检测,并以分割瓶颈的弦线中点为中心坐标,引入算法预置的抽风矢量场;以图像梯度标识矩阵为参考,对抽风矢量场中的梯度矢量进行点积运算,并将矢量的点积结果作为分割曲线的外部图像驱动力,实现分割曲线在凹陷区域内部的继续收敛.实验数据显示;在不同凹陷程度情况下,该算法始终能够将图像分割的平均误差和覆盖比率控制在有效范围.不同算法的分割对比实验表明,该算法分割曲线对深度凹陷结构的拟合程度优于其他同类算法.

Abstract:

 An image segmentation algorithm based on the vector field of suction was presented to solve the bottleneck problem for segmenting recessed images. This method segments the recessed image and saves the image original segmentation curves and the image gradient matrix. It identifies and marks the bottlenecks on the control nodes of the segmentation curve, while the identification process is based on the distribution of gradient vector. Meanwhile, a predefined vector field of suction is introduced and the midpoint of the straight segment on the arc is selected as the coordinates of the center chord on the vector field. By referring the mark matrix on the gradient, the algorithm performs dot product on the gradient vectors in the vector field of suction, and the dot product of the vectors is specified as the drive force for splitting the curve. Thus this method can achieve segmentation curve convergence inside the boundary concavities. Experimental results show that, for varying degrees of boundary concavities. The proposed algorithm can always limit the image segmentation average error and coverage ratio within a valid range. Results of comparative experiments show that the curve obtained by the proposed algorithm is closest to the edge of the image among those obtained by the similar segmentation algorithms.

出版日期: 2015-04-01
:  TP 301.6  
基金资助:

国家自然科学基金资助项目(61170318).

通讯作者: 潘志庚,男,教授,博导.     E-mail: zgpan@hznu.edu.cn
作者简介: 孔勇奇(1973—),男,副教授,计算机图像处理与三维仿真技术研究.E-mail: kl_029@163.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  

引用本文:

孔勇奇, 潘志庚. 基于抽风矢量场的深度凹陷图像分割算法[J]. 浙江大学学报(工学版), 10.3785/j.issn.1008-973X.2014.06.009.

KONG Yong-qi, PAN Zhi-geng. Segmentation algorithm of recessed image based on vector field of suction. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 10.3785/j.issn.1008-973X.2014.06.009.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2014.06.009        http://www.zjujournals.com/eng/CN/Y2014/V48/I6/1024

[1] KASS M,WITKIN A,TERZOPOULOS D. Snakes: Active contour models [J]. International Journal of Computer Vision, 1987, 1(4): 321-331.
[2] COHEN L D,COHEN I. Finite element methods for active contour models and balloons for 2D and 3D images [J]. IEEE Trans. PAMI, 1993, 15(11): 1131-1147.
[3] 王元全,汤敏,王平安,等. Snake模型与深度凹陷区域的分割[J]. 计算机研究与发展, 2005, 42(7): 1179-1184.
WANG Yuan-quan, TANG Min, WANG Pin-gan, et al. Research on boundary concavities segmentation via snake models [J].Journal of Computer Research and Development, 2005, 42(7): 1179-1184.
[4] XU Cheng-yang,PRINCE J L. Snakes, shapes and gradient vector flow [J]. IEEE Trans. Image Processing, 1998, 7(3): 359-369.
[5] XU Cheng-yang, PRINCE J L. Generalized gradient vector flow external forces for active contours [J]. Signal Processing, 1998, 71(2): 131-139.
[6] ZHANG Fan, ZHANG Xin-hong, CAO Kui. Contour extraction of gait recognition based on improved GVF Snake model [J].Computers & Electrical Engineering, 2012, 38(4): 882-890.
[7] WANG Yuan-quan, LIU Li-xong, ZHANG Hua, et al. Image segmentation using active contours with normally biased GVF external force [J]. IEEE Signal Processing Letters, 2010, 17(10): 875-878.
[8] 周亚男,程熙,骆剑承.改进GVF的自动Snakes模型[J].中国图象图形学报,2012,17(2): 256-262.
ZHOU Yanan, CHENG Xi, LUO Jiancheng. Automatic Snakes model based on modified GVF [J]. Journal of Image and Graphics, 2012, 17(2): 256-262.
[9] 燕杨,李岩波,王云吉.一种基于改进Snake模型的边界检测方法[J].吉林大学学报: 理学版,2013,51(5): 904-907.
YAN Yang, LI Yan-bo, WANG Yun-ji. Improved snake model in the depression boundary detection [J]. Journal of Jilin University: Science Edition, 2013, 51(5): 904-907.
[10] 高向军.一种向量场卷积外力加速的GAC模型[J]. 计算机工程,2012,38(17): 192-195.
GAO Xiang-jun. A GAC model accelerated by vector field convolution external force [J]. Computer Engineering, 2012, 38(17): 192-195.
[11] SREEMATHY R,PATIL R S. Segmentation of left ventricle in cardiac MRI using Snake and GVF Snake [J]. International Journal of Engineering Science and Technology, 2011, 5(3): 4102-4107.
[12] JARJES A, WANG Kuan-quan, MOHANMMED G J. GVF snake-based method for accurate pupil contour detection [J]. Information Technology Journal, 2010, 8(9): 1653-1658.
[13] TANG Jin-shan. A multi-direction GVF snake for the segmentation of skin cancer images [J]. Pattern Recognition, 2009, 42(6): 1172-1179.
[14] 覃武星,李斌,岳小强.一种基于初始化Snake轮廓线的混合舌图像分割算法[J].中国科学技术大学学报, 2010, 40(8): 807-811.
QIN Wu-xing,LI Bin,YUN Xiao-qiang. A hybrid tongue image segmentation algorithm based on initialization of Snake contours [J]. Journal of University of Science and Technology of China, 2010, 40(8): 807-811.

[1] 董立岩, 朱琪, 李永丽. 基于最大共识的模型组合算法[J]. 浙江大学学报(工学版), 2017, 51(2): 416-421.
[2] 苗峰,谢安桓,王富安,喻峰,周华. 多阶段可替换分组并行机调度问题的求解[J]. 浙江大学学报(工学版), 2015, 49(5): 866-872.
[3] 倪广翼, 章孝灿, 苏程, 俞伟斌. 基于多染色体演化的自适应类别数聚类方法[J]. 浙江大学学报(工学版), 2014, 48(6): 980-986.
[4] 刘加海,杨茂林,雷航,廖勇. 共享资源约束下多核实时任务分配算法[J]. J4, 2014, 48(1): 113-117.
[5] 赵诗奎, 方水良, 顾新建. 柔性车间调度的新型初始机制遗传算法[J]. J4, 2013, 47(6): 1022-1030.
[6] 张俊超, 岳茂雄, 刘华锋. 结构先验约束的动态PET图像重建[J]. J4, 2012, 46(6): 961-966.
[7] 刘毅,李平,高增梁. 用于高炉铁水质量预报的改进支持向量回归[J]. J4, 2012, 46(5): 830-836.
[8] 方水良, 姚嫣菲, 赵诗奎. 柔性车间调度的改进遗传算法[J]. J4, 2012, 46(4): 629-635.
[9] 喻海清, 刘毅, 陈坤, 纪俊, 李平. 鲁棒的递推核学习建模方法在高炉过程的应用[J]. J4, 2012, 46(4): 705-711.
[10] 刘加海,杨茂林. 基于多核处理器平台的公平调度算法[J]. J4, 2011, 45(9): 1566-1570.
[11] 徐敬华, 张树有, 伊国栋, 屠立, 光耀. 面向目标变异的操作臂运动学优化设计[J]. J4, 2011, 45(2): 209-216.
[12] 倪何, 程刚, 孙丰瑞. 基于混合演化的自适应建模及其应用[J]. J4, 2010, 44(8): 1490-1495.
[13] 戴文战, 熊伟, 杨爱萍. 基于函数cot (xα)变换及背景值优化的灰色建模[J]. J4, 2010, 44(7): 1368-1372.
[14] 陈坤, 刘毅, 王海清, 等. Hammerstein系统递推辨识的自适应算法[J]. J4, 2010, 44(1): 99-103.
[15] 杨珂, 罗琼, 石教英. 图形处理器在数据库技术中的应用[J]. J4, 2009, 43(8): 1349-1360.