基于模糊综合评判的在线图像分割
|
摘要
为使矿岩块度在线检测能精确和快速地进行,提出了一种自适应阈值分割算法:首先选取全局自动阈值分割算法粗分割图像,然后针对粗分割效果较差的目标区域,重复利用阈值分割算法直到无区域可分离。在重复分割前,利用模糊综合评判方法对每一个目标物体的参数(尺寸,形状及梯度等)进行综合评价,若该目标为非正常块度,对其进行重复分割。通过对一次矿山爆破的300多幅矿岩块度图像进行测试表明:在线图像分割算法与新近的聚类分析及图论最小割算法相比,不仅简单而且具有较好的分割效果,在保证在线处理速度的要求下,分割准确率达95%以上,可有效地扩展应用在其它颗粒在线检测系统中。
In this paper,a monitoring system grabed the fragment images at the end of the conveyor belt and did the image processing in time,and the key task and the problem for the system was to deal with the touching fragments quickly and accurately.To resolve the problem,an adaptive image segmentation algorithm was proposed based on the analyses of fragment size,shape and gradient etc.Firstly a fragment image was roughly segmented by a recursive thresholding algorithm,and then the treatment was repeated for the regions of touching fragments.The judgment for touching fragments applied fuzzy comprehensive evaluation for rock fragment size,shape and gradient etc.In one Mine,300 fragment images were took to do testing,the experimental results indicated that as compared to clustering analysis and graph based algorithms,the new algorithm can successfully make fragment image segmentation online,the accuracy was about 95%.The new algorithm can also be expanded into other complicated particle online monitoring systems.
In this paper,a monitoring system grabed the fragment images at the end of the conveyor belt and did the image processing in time,and the key task and the problem for the system was to deal with the touching fragments quickly and accurately.To resolve the problem,an adaptive image segmentation algorithm was proposed based on the analyses of fragment size,shape and gradient etc.Firstly a fragment image was roughly segmented by a recursive thresholding algorithm,and then the treatment was repeated for the regions of touching fragments.The judgment for touching fragments applied fuzzy comprehensive evaluation for rock fragment size,shape and gradient etc.In one Mine,300 fragment images were took to do testing,the experimental results indicated that as compared to clustering analysis and graph based algorithms,the new algorithm can successfully make fragment image segmentation online,the accuracy was about 95%.The new algorithm can also be expanded into other complicated particle online monitoring systems.
引文
[1] RAINA A K.A history of digital image analysis technique for blast fragmentation assessment and some Indian contributions[C]//Measurement and Analysis of Blast Fragmentation,Sanchidrián&Singh(Eds),2013:3-11.
[2] WANG Weixing,WANG Fengping,HUANG Xiaojun,et al.Rock fracture image acquisition using two kinds of lighting and fusion on a wavelet transform[J].Bulletin of Engineering Geology and the Environment,2016,75(1):311-324.
[3] MATTHEW J T.Automated online measurement of limestone particle size distributions using 3Drange data[J].Journal of Process Control,2011,21:254-262.
[4] HAMZELOO E,MASINAEI M,MEHRSHA N.Estimation of particle size distribution on a conveyor belt using image analysis and neural network[C]//International Mineral Processing Congress,2012:3298-3305.
[5] PRESLES Benoit,DEBAYLE Johan,FéVOTTE Gilles,et al.Novel image analysis method for in situ monitoring the particle size distribution of batch crystallization processes[J].Electron Imaging,2010,19(3):031207.
[6] ARASAN Seracettin A,HASILOGLU Samet,AKBULUT Suat.Shape properties of natural and crushed aggregate using image analysis[J].International Journal of Civil and Structural Engineering,2010,1(2):221-233.
[7] OTSU N. A threshold selection method from gray-level histograms[J].IEEE Trans.System Man Cybernetics,SMC,1979,9(1):62-66.
[8] TSAI W H.Moment-preserving thresholding:a new Approach[J].Computer Vision Graphics Image Process,1985,29:377-393.
[9]KAPUR J N,SAHOO P K,WONG A K C.A new method for gray-level picture thresholding using the entropy of the histogram[J]. Computer Vision, Graphics and Image Processing,1985,29(3):273-285.
[10] WANG Weixing,ZHOU Weigang and ZHAO Xiangmo.airplane extraction and ldentification by improved PCNN with wavelet transform and modified zernike moments[J].The Imaging Science Journal,2014,62(1):27-34.
[11]KINNUNEN I,MAKYNEN A.image based size distribution measurement of gravel particles[C]//Instrumentation and Measurement Technology Conference,2011:1-6.
[12] ANDERSSON T,THURLEY M J.visibility classification of rocks in piles[J].Digital Image Computing:Techniques and Applications,2008,85:207-213.
[13]WANG Weixing.Image analysis of particles by modified ferret method-best-fit rectangle[J].Powder Technology,2006,165(1):1-10.
[14] ZHANG Juan,HUANG Xinyuan,WU Jian,et al.The application of digital image processing technology in the quantitativestudyofthecoarseaggregateshape characteristics[C]//Information Science and Engineering,2009,1st International Conference on Digital Object Identifier,2009:1471-1475.
[15]WANG Weixing,LIU Sheng.Online burning material pile detection on color clustering and quaternion based edge detection in boiler[J].KSII Transactions on Internet and Information Systems,2015,9(1):190-207.
[16] ZHONG Qu,MA Qingwei.Research on evaluation of image segmentation based on measurement method of particle's parameters[C]//IEEE.Computer Society,The Second International Conference on Intelligent Human-machine Systems&Cybernetics,2010,158:231-235.
[17] LIAO C W,TARNG Y S.On-line automatic optical inspection system for coarse particle size distribution[J].Powder Technology,2009,189:508-513.
[18]WANG Weixing,CAO Ting,LIU Sheng,et al.Remote sensing image automatic registration on multi-scale HarrisLaplacian[J].Journal of the Indian Society of Remote Sensing,2015,43(3):501-511.
[19] WANG Weixing,ZHANG Xin,CAO Ting,et al.Fuzzy and touching cell extraction on modified graph MST and skeleton distance mapping histogram[J].Journal of Medical Imaging and Health Informatics,2014,4(3):350-357.
[20] KRINIDIS Stelios,KRINIDIS Michail and CHATZIS Vassilios.An unsupervised image clustering method based on EEMD,image histogram[J].Journal of Information Hiding and Multimedia Signal Processing,2012,3(2):151-163.
[2] WANG Weixing,WANG Fengping,HUANG Xiaojun,et al.Rock fracture image acquisition using two kinds of lighting and fusion on a wavelet transform[J].Bulletin of Engineering Geology and the Environment,2016,75(1):311-324.
[3] MATTHEW J T.Automated online measurement of limestone particle size distributions using 3Drange data[J].Journal of Process Control,2011,21:254-262.
[4] HAMZELOO E,MASINAEI M,MEHRSHA N.Estimation of particle size distribution on a conveyor belt using image analysis and neural network[C]//International Mineral Processing Congress,2012:3298-3305.
[5] PRESLES Benoit,DEBAYLE Johan,FéVOTTE Gilles,et al.Novel image analysis method for in situ monitoring the particle size distribution of batch crystallization processes[J].Electron Imaging,2010,19(3):031207.
[6] ARASAN Seracettin A,HASILOGLU Samet,AKBULUT Suat.Shape properties of natural and crushed aggregate using image analysis[J].International Journal of Civil and Structural Engineering,2010,1(2):221-233.
[7] OTSU N. A threshold selection method from gray-level histograms[J].IEEE Trans.System Man Cybernetics,SMC,1979,9(1):62-66.
[8] TSAI W H.Moment-preserving thresholding:a new Approach[J].Computer Vision Graphics Image Process,1985,29:377-393.
[9]KAPUR J N,SAHOO P K,WONG A K C.A new method for gray-level picture thresholding using the entropy of the histogram[J]. Computer Vision, Graphics and Image Processing,1985,29(3):273-285.
[10] WANG Weixing,ZHOU Weigang and ZHAO Xiangmo.airplane extraction and ldentification by improved PCNN with wavelet transform and modified zernike moments[J].The Imaging Science Journal,2014,62(1):27-34.
[11]KINNUNEN I,MAKYNEN A.image based size distribution measurement of gravel particles[C]//Instrumentation and Measurement Technology Conference,2011:1-6.
[12] ANDERSSON T,THURLEY M J.visibility classification of rocks in piles[J].Digital Image Computing:Techniques and Applications,2008,85:207-213.
[13]WANG Weixing.Image analysis of particles by modified ferret method-best-fit rectangle[J].Powder Technology,2006,165(1):1-10.
[14] ZHANG Juan,HUANG Xinyuan,WU Jian,et al.The application of digital image processing technology in the quantitativestudyofthecoarseaggregateshape characteristics[C]//Information Science and Engineering,2009,1st International Conference on Digital Object Identifier,2009:1471-1475.
[15]WANG Weixing,LIU Sheng.Online burning material pile detection on color clustering and quaternion based edge detection in boiler[J].KSII Transactions on Internet and Information Systems,2015,9(1):190-207.
[16] ZHONG Qu,MA Qingwei.Research on evaluation of image segmentation based on measurement method of particle's parameters[C]//IEEE.Computer Society,The Second International Conference on Intelligent Human-machine Systems&Cybernetics,2010,158:231-235.
[17] LIAO C W,TARNG Y S.On-line automatic optical inspection system for coarse particle size distribution[J].Powder Technology,2009,189:508-513.
[18]WANG Weixing,CAO Ting,LIU Sheng,et al.Remote sensing image automatic registration on multi-scale HarrisLaplacian[J].Journal of the Indian Society of Remote Sensing,2015,43(3):501-511.
[19] WANG Weixing,ZHANG Xin,CAO Ting,et al.Fuzzy and touching cell extraction on modified graph MST and skeleton distance mapping histogram[J].Journal of Medical Imaging and Health Informatics,2014,4(3):350-357.
[20] KRINIDIS Stelios,KRINIDIS Michail and CHATZIS Vassilios.An unsupervised image clustering method based on EEMD,image histogram[J].Journal of Information Hiding and Multimedia Signal Processing,2012,3(2):151-163.