基于双自适应遗传算法的Otsu图像分割研究
|
摘要
为了克服标准遗传算法在图像分割过程中存在易早熟、易陷入局部最优的缺点,同时针对类间最大方差法仅对灰度直方图分布呈双峰的图像效果显著的不足,提出了一种基于双自适应遗传算法的改进Otsu图像分割算法。该方法将传统的自适应遗传算法与新算法进行融合形成双自适应遗传算法,同时考虑目标背景可变的Otsu图像分割算法,使得对个体的评价更为合理,改善种群的全局搜索能力。实验结果表明:与传统Otsu图像分割法及基于遗传算法的图像分割方法相比,该算法求得出的阈值范围更加稳定,且可以获得更优的图像分割效果,有利于计算机视觉的后续处理。
tandard genetic algorithm in image segmentation has the shortcomings,such as easy prematurting and falling intolocal optimization. At the same time,the disadvantage of Otsu is that the gray histogram distribution is bimodal. In order to overcomethese an improved Otsu image segmentation algorithm based on double adaptive genetic algorithm is proposed in this paper. Themethod combines the traditional AGA with the new algorithm,considering Otsu image segmentation algorithm with target back-ground variation. And then,individual evaluation is more rational. Not only it can improve the global search ability of the popula-tion. Results show that,threshold range keeps more stable and the better image segmentation effect when it is compared with Otsuand the standard GA. It is beneficial to the subsequent processing of computer vision.
tandard genetic algorithm in image segmentation has the shortcomings,such as easy prematurting and falling intolocal optimization. At the same time,the disadvantage of Otsu is that the gray histogram distribution is bimodal. In order to overcomethese an improved Otsu image segmentation algorithm based on double adaptive genetic algorithm is proposed in this paper. Themethod combines the traditional AGA with the new algorithm,considering Otsu image segmentation algorithm with target back-ground variation. And then,individual evaluation is more rational. Not only it can improve the global search ability of the popula-tion. Results show that,threshold range keeps more stable and the better image segmentation effect when it is compared with Otsuand the standard GA. It is beneficial to the subsequent processing of computer vision.
引文
[1]谭优,王泽勇.图像阈值分割算法实用技术研究与比较[J].微计算机信息,2007,23(24):298-299,233.TAN You,Wang Zeyong.Study on applied technology arith-metic of image threshold segmentation[J].MicrocomputerInformation,2007,23(24):298-299,233.
[2]S.Sharma,D.Shah.A Practical Animal Detection andCollision Avoidance System Using Computer Vision Tech-nique[J].IEEE Access,2016:1-1.
[3]刘紫燕,吴俊熊,毛攀,等.基于遗传模拟退火算法的Otsu图像分割研究[J].电视技术,2016,40(8):15-18.LIU Zhiyan,WU Junxiong,MAO Pan,et al.Image segmen-tation on genetic simulated annealing algorithm[J].Videoengineering,2016,40(8):15-18.
[4]师文,朱学芳,朱光.基于形态学的MRI图像自适应边缘检测算法[J].仪器仪表学报,2013,34(2):408-414.SHI Wen,ZHU Xuefang,ZHU Guang.A daptive edge de-tection algorithm of MRI image based on morphology[J].Chinese Journal of Scientific Instrument,2013,34(2):408-414.
[5]Qin A K,Claus D A.Multivariate Image Segmentation Us-ing Semantic Region Growing With Adaptive Edge PenaltyImage Processing[J].IEEE Transactions on 2010,19(8):2167-2169.
[6]OTSU N.A threshold selection method from gray lev-el histograms[J].IEEE Trans on SMC,1979,9(1):62-69.
[7]WANG F J,LI J L,LIU S W,et al.An improved adap-tive genetic algorithm for image segmentation and visionalignment used in microelectronic bonding[J].IEEE trans-actions on mechatronics,2014,19(3):916-923.
[8]TAO W B,JIN H.Image thresholding using graph cuts[J].IEEE Transactions on Systems,Man,and Cybernet-ics,2008,38(5):1181-1194.
[9]刘秋生,楚来国,杨继昌.基于遗传优化的阈值选取方法[J].信号处理,2002,18(4):374-377.LIU Qiusheng,CHU Laiguo,YANG Jichang.Seleotion oflmage Threshold on the Basis of Genetie Algorithms[J].Signal Processing,2002,18(4):374-377.
[10]Zhang L,Chang H,Xu R.Equal-width partitioning rou-lette wheel selection in genetic algorithm[C]//Technolo-gies and Applications of Artificial Intelligence(TAAI),2012 Conference on IEEE,2012:62-67.
[11]邝航宇,金晶,苏勇.自适应遗传算法交叉变异算子的改进[J].计算机工程与应用,2006,42(12):93-96,99.KUANG Hangyu,JIN Jing,SU Yong.Improving Cross-over and Mutation for Adaptive Genetic Algorithm[J].Computer Enginerring and Applications,2006,42(12):93-96,99.
[12]何春华,胡迎春.基于改进遗传算法的自动阈值图像分割方法[J].计算机仿真,2011(2):312-315.HE Chunhua,HU Yingchun.Automatic Threshold ImageSegmentation Approach Baced on Improved Genetic Al-gorithm[J].Computer Simulation,2011(2):312-315.
[13]Haralick RM,Shapiro LG,Image Segmentation tech-niques[J].CVGIP,1985,29(37):100-132.
[14]王爽,黄友锐,李冬.基于蚁群算法的改进Otsu理论的图像多阈值分割[J].微计算机应用,2008,29(4):25-28.WANG Shuang,HUANG Yourui,LI Dong.MultilevelThresholging Methods for Image Segmentation with Im-proved Otsu Based on Ant Colony Algorithm[J].Micro-computer Applications,2008,29(4):25-28.
[15]周云燕,杨坤涛,黄鹰.基于最小类内离散度的改进Otsu分割方法的研究[J].华中科技大学学报(自然科学版),2007,35(2):101-103.ZHOU Yunyan,YANG Kuntao,HUANG Ying.ImprovedOtsu Thresholding Based on Minimum Inner-clusterVariance[J].J.Huazhong Univ.of Sci.&Tech.(NatureScience Edition),2007,35(2):101-103.
[2]S.Sharma,D.Shah.A Practical Animal Detection andCollision Avoidance System Using Computer Vision Tech-nique[J].IEEE Access,2016:1-1.
[3]刘紫燕,吴俊熊,毛攀,等.基于遗传模拟退火算法的Otsu图像分割研究[J].电视技术,2016,40(8):15-18.LIU Zhiyan,WU Junxiong,MAO Pan,et al.Image segmen-tation on genetic simulated annealing algorithm[J].Videoengineering,2016,40(8):15-18.
[4]师文,朱学芳,朱光.基于形态学的MRI图像自适应边缘检测算法[J].仪器仪表学报,2013,34(2):408-414.SHI Wen,ZHU Xuefang,ZHU Guang.A daptive edge de-tection algorithm of MRI image based on morphology[J].Chinese Journal of Scientific Instrument,2013,34(2):408-414.
[5]Qin A K,Claus D A.Multivariate Image Segmentation Us-ing Semantic Region Growing With Adaptive Edge PenaltyImage Processing[J].IEEE Transactions on 2010,19(8):2167-2169.
[6]OTSU N.A threshold selection method from gray lev-el histograms[J].IEEE Trans on SMC,1979,9(1):62-69.
[7]WANG F J,LI J L,LIU S W,et al.An improved adap-tive genetic algorithm for image segmentation and visionalignment used in microelectronic bonding[J].IEEE trans-actions on mechatronics,2014,19(3):916-923.
[8]TAO W B,JIN H.Image thresholding using graph cuts[J].IEEE Transactions on Systems,Man,and Cybernet-ics,2008,38(5):1181-1194.
[9]刘秋生,楚来国,杨继昌.基于遗传优化的阈值选取方法[J].信号处理,2002,18(4):374-377.LIU Qiusheng,CHU Laiguo,YANG Jichang.Seleotion oflmage Threshold on the Basis of Genetie Algorithms[J].Signal Processing,2002,18(4):374-377.
[10]Zhang L,Chang H,Xu R.Equal-width partitioning rou-lette wheel selection in genetic algorithm[C]//Technolo-gies and Applications of Artificial Intelligence(TAAI),2012 Conference on IEEE,2012:62-67.
[11]邝航宇,金晶,苏勇.自适应遗传算法交叉变异算子的改进[J].计算机工程与应用,2006,42(12):93-96,99.KUANG Hangyu,JIN Jing,SU Yong.Improving Cross-over and Mutation for Adaptive Genetic Algorithm[J].Computer Enginerring and Applications,2006,42(12):93-96,99.
[12]何春华,胡迎春.基于改进遗传算法的自动阈值图像分割方法[J].计算机仿真,2011(2):312-315.HE Chunhua,HU Yingchun.Automatic Threshold ImageSegmentation Approach Baced on Improved Genetic Al-gorithm[J].Computer Simulation,2011(2):312-315.
[13]Haralick RM,Shapiro LG,Image Segmentation tech-niques[J].CVGIP,1985,29(37):100-132.
[14]王爽,黄友锐,李冬.基于蚁群算法的改进Otsu理论的图像多阈值分割[J].微计算机应用,2008,29(4):25-28.WANG Shuang,HUANG Yourui,LI Dong.MultilevelThresholging Methods for Image Segmentation with Im-proved Otsu Based on Ant Colony Algorithm[J].Micro-computer Applications,2008,29(4):25-28.
[15]周云燕,杨坤涛,黄鹰.基于最小类内离散度的改进Otsu分割方法的研究[J].华中科技大学学报(自然科学版),2007,35(2):101-103.ZHOU Yunyan,YANG Kuntao,HUANG Ying.ImprovedOtsu Thresholding Based on Minimum Inner-clusterVariance[J].J.Huazhong Univ.of Sci.&Tech.(NatureScience Edition),2007,35(2):101-103.