采用概率混合模型的圆周曲线识别方法
|
摘要
曲线识别是图像识别和机器视觉的一个重要研究课题。建立圆周曲线的概率混合模型,并分别利用EM算法和贝叶斯阴阳和谐学习(BYY)算法,在曲线条数已知和未知的情况下实现模型选择和参数估计,从而完成对圆周曲线的识别以及数据点的聚类。试验结果表明:用这两种算法处理平面曲线的混合模型可以准确地估计出曲线条数并同时完成参数估计,较好地完成曲线识别。
Curve recognition is an important research subjects among image recognition and machine vision. This paper establishes the probability mixture model of circular curve. And model selection and parameter estimation are achieved with expectation maximization( EM) algorithm and Bayesian Ying-Yang harmony learning( BYY),under the circumstance that the number of curves is known and unknown,respectively. Thus we complete the recognition of circular curve as well as the clustering of data points. Experimental results demonstrate that,these two algorithms not only can serve better in the processing of mixture model of circular curves,accurately estimate the number of straight lines,but also can accomplish parameter estimation simultaneously. Therefore,the algorithms can well realize line recognition.
Curve recognition is an important research subjects among image recognition and machine vision. This paper establishes the probability mixture model of circular curve. And model selection and parameter estimation are achieved with expectation maximization( EM) algorithm and Bayesian Ying-Yang harmony learning( BYY),under the circumstance that the number of curves is known and unknown,respectively. Thus we complete the recognition of circular curve as well as the clustering of data points. Experimental results demonstrate that,these two algorithms not only can serve better in the processing of mixture model of circular curves,accurately estimate the number of straight lines,but also can accomplish parameter estimation simultaneously. Therefore,the algorithms can well realize line recognition.
引文
[1]XU Z,SHIN B S,KLETTE R.Accurate and robust line segment extraction using minimum entropy with hough transform[J].IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society,2015,24(3):813-822.
[2]NG C S.Intelligent hough transform for object detection and tracking[J].Innovative Food Science&Emerging Technologies,2015,5(3):307-316.
[3]MUKHOPADHYAY P,CHAUDHURI B B.A survey of hough transform[J].Pattern Recognition,2015,48(3):993-1010.
[4]陈小艳,王强,李柏林.改进的Hough变换检测圆方法[J].计算机系统应用,2015,24(8):197-201.
[5]XU G,YANG Y Q,LIU B B,et al.An efficient hybrid multi-objective particle swarm optimization with a multi-objective dichotomy line search[J].Journal of Computational&Applied Mathematics,2015,280(C):310-326.
[6]LI Jin-long,MA Hong-feng,ZHANG W Z,et al.Research on edge detection of track fastener based on machine vision[J].Journal of Northwest Normal University,2017(7):96-101.
[7]SOMKANTHA K,THEERAUMPON N,AUEPHANWIRIYAKUL S.Boundary detection in medical images using edge following algorithm based on intensity gradient and texture gradient features.[J].IEEE transactions on bio-medical engineering,2011,58(3):567.
[8]石磊,孙凯明,王刚,等.基于模糊聚类的标识点图形识别方法[J].自动化技术与应用,2015,34(12):90-92.
[9]FANG C,MA J.A Fixed-point EM algorithm for straight line detection[M].Berlin:Springer Berlin Heidelberg,2011:136-143.
[10]LONG T,JIAO W,HE G,et al.Automatic line segment registration using gaussian mixture model and expectation-maximization algorithm[J].IEEE Journal of Selected Topics in Applied Earth Observations&Remote Sensing,2014,7(5):1688-1699.
[11]ARELLANO C,DAHYOT R.Robust ellipse detection with Gaussian mixture models[J].Pattern Recognition,2016,58(C):12-26.
[12]XU L.Best harmony,unified RPCL and automated model selection for unsupervised and supervised learning on Gaussian mixtures,three-layer nets and ME-RBF-SVM models.[J].International Journal of Neural Systems,2001,11(1):43-69.
[13]MA J,WANG T,XU L.A gradient BYY harmony learning rule on Gaussian mixture with automated model selection[J].Neurocomputing,2004,56:481-487.
[14]JIA Y,YANG Z,SONG Q.Experimental study of random dynamic loads identification based on weighted regularization method[J].Journal of Sound&Vibration,2015,342:113-123.
[15]吕琪.不适定问题的迭代正则化方法研究[D].武汉:武汉理工大学,2012.
[2]NG C S.Intelligent hough transform for object detection and tracking[J].Innovative Food Science&Emerging Technologies,2015,5(3):307-316.
[3]MUKHOPADHYAY P,CHAUDHURI B B.A survey of hough transform[J].Pattern Recognition,2015,48(3):993-1010.
[4]陈小艳,王强,李柏林.改进的Hough变换检测圆方法[J].计算机系统应用,2015,24(8):197-201.
[5]XU G,YANG Y Q,LIU B B,et al.An efficient hybrid multi-objective particle swarm optimization with a multi-objective dichotomy line search[J].Journal of Computational&Applied Mathematics,2015,280(C):310-326.
[6]LI Jin-long,MA Hong-feng,ZHANG W Z,et al.Research on edge detection of track fastener based on machine vision[J].Journal of Northwest Normal University,2017(7):96-101.
[7]SOMKANTHA K,THEERAUMPON N,AUEPHANWIRIYAKUL S.Boundary detection in medical images using edge following algorithm based on intensity gradient and texture gradient features.[J].IEEE transactions on bio-medical engineering,2011,58(3):567.
[8]石磊,孙凯明,王刚,等.基于模糊聚类的标识点图形识别方法[J].自动化技术与应用,2015,34(12):90-92.
[9]FANG C,MA J.A Fixed-point EM algorithm for straight line detection[M].Berlin:Springer Berlin Heidelberg,2011:136-143.
[10]LONG T,JIAO W,HE G,et al.Automatic line segment registration using gaussian mixture model and expectation-maximization algorithm[J].IEEE Journal of Selected Topics in Applied Earth Observations&Remote Sensing,2014,7(5):1688-1699.
[11]ARELLANO C,DAHYOT R.Robust ellipse detection with Gaussian mixture models[J].Pattern Recognition,2016,58(C):12-26.
[12]XU L.Best harmony,unified RPCL and automated model selection for unsupervised and supervised learning on Gaussian mixtures,three-layer nets and ME-RBF-SVM models.[J].International Journal of Neural Systems,2001,11(1):43-69.
[13]MA J,WANG T,XU L.A gradient BYY harmony learning rule on Gaussian mixture with automated model selection[J].Neurocomputing,2004,56:481-487.
[14]JIA Y,YANG Z,SONG Q.Experimental study of random dynamic loads identification based on weighted regularization method[J].Journal of Sound&Vibration,2015,342:113-123.
[15]吕琪.不适定问题的迭代正则化方法研究[D].武汉:武汉理工大学,2012.