基于图像矩特征的工件摆放类型Parzen窗估计
|
摘要
工件的视觉判别是实现机械手抓取工件的重要环节,如何对姿态变换后的工件进行视觉识别及分类,具有重要的理论及工程研究价值。设计了工件姿态变换实验平台,实验研究了工件姿态变换的视觉识别问题。首先,采用分水岭图像分割方法,实现了工件目标图像的分割。其次,针对工件在水平面旋转姿态变换及在三维空间进行姿态变换两种情况,分别选取单个工件、部分遮挡工件及部分重叠工件构建图像样本,并实现了七个特征不变矩的提取。最后,将工件训练样本的五阶矩、七阶矩的平均值作为工件种类分类目标,采用Parzen窗对待识别工件类型进行概率密度估计,并用带惯性的粒子群算法对Parzen窗估计结果进行优化,克服Parzen窗估计的多峰值问题,实现了对工件摆放类型的准确判别。
Visual recognition of workpiece is an important part for robot to grasp the workpiece. How to recognize and classify the workpiece after attitude transformation has an important theoretical and engineering value. In this paper, the experiment platform of workpiece attitude transformation is designed, and researched the visual recognition of workpiece under pose transformation experimentally. First of all, realizes the target image segmentation of workpiece with watershed algorithm. Secondly, for the two cases as rotation pose transformation at the horizontal plane and pose transformation in three dimensional spaces, single work-piece, partial occlusion workpiece and partial overlap workpiece were selected to build image samples, and seven feature invariant moments of workpiece were extracted. Thirdly, averages of fifth-order moments and seventh-order moments of workpiece in training sample as classification target of workpiece type, the probability density of type of workpiece to be identified is estimated based on Parzen window. Used particle swarm optimization(PSO) algorithm with dynamic inertia to optimize the estimated result by Parzen window, overcome the multi peak problem of Parzen window estimation, and the accurate identification for the placed type of workpiece is realized.
Visual recognition of workpiece is an important part for robot to grasp the workpiece. How to recognize and classify the workpiece after attitude transformation has an important theoretical and engineering value. In this paper, the experiment platform of workpiece attitude transformation is designed, and researched the visual recognition of workpiece under pose transformation experimentally. First of all, realizes the target image segmentation of workpiece with watershed algorithm. Secondly, for the two cases as rotation pose transformation at the horizontal plane and pose transformation in three dimensional spaces, single work-piece, partial occlusion workpiece and partial overlap workpiece were selected to build image samples, and seven feature invariant moments of workpiece were extracted. Thirdly, averages of fifth-order moments and seventh-order moments of workpiece in training sample as classification target of workpiece type, the probability density of type of workpiece to be identified is estimated based on Parzen window. Used particle swarm optimization(PSO) algorithm with dynamic inertia to optimize the estimated result by Parzen window, overcome the multi peak problem of Parzen window estimation, and the accurate identification for the placed type of workpiece is realized.
引文
[1] 曲道奎.中国机器人产业发展现状与展望[J].中国科学院院刊,2015,30(3):342—345.Qu Daokui.Development situation and prospects of chinese robot industry[J].Bulletin of Chinese Academy of Sciences,2015,30(3):342—345.
[2] 朱海波.基于视觉引导的工业机器人工件搬运技术研究[D].沈阳:沈阳工业大学,2013:20—35.Zhu Haibo.Research on workpiece handling technology with industrial robot based on vision guided[D].Shenyang:Shenyang University of Technology,2013:20—35.
[3] 刘正琼,万鹏,凌琳,等.基于机器视觉的超视场工件识别抓取系统[J].机器人,2018,40(3):294—300.Liu Zhengqiong,Wan Peng,Ling Lin,et al.Recognition and grabbing system for workpieces exceeding the visual field based on machine vision[J].Robot,2018,40(3):294—300.
[4] 兰佳,叶文华.基于立体视觉的工件识别与定位系统研究[J].机械制造与自动化,2017,46(5):188—193.Lan Jia,Ye Wenhua.Research on Parts' recognition and orientation algorithm based on stereo vision[J].Machine Building & Automation,2017,46(5):188—193.
[5] Lowe D G.Distinctive image features from scale-invariant key points [J].International Journal of Computer Vision,2004,60(2):91—110.
[6] 罗四,李少洪.基于相关成分分析法的雷达目标识别仿真研究[J].系统仿真学报,2009,21(2):594—598.Luo Si,Li Shaohong.Simulation research of radar target recognition based on RCA[J].Journal Of System Simulation,2009,21(2):594—598.
[7] 李俊威.基于不变矩的人形识别技术研究[D].西安科技大学,2012:33—38.Li junwei.Research on moment invariant based human recognition technology[D].Xi'an University of Science and Technology,2012:33—38.
[8] 庞菲菲,张群飞,史文涛,等.基于Parzen窗的水下无线传感器网络目标定位方法[J].电子与信息学报,2017,39(1):45—50.Pang Feifei,Zhang Qunfei,Shi Wentao,et al.Target localization method based on parzen window in underwater wireless sensor network[J].Journal of Electronics and Information Technology,2017,39(1):45—50.
[9] Takeuchi Y,Shimizu H,Mukai I,et al.Automatic measurement of 3-dimensional coordinate measuring machine by means of CAD and image data[J].Journal of the Japan Society for Precision Engineering,1990,39(1):565—568.
[10] S H Darwish,L N Abdullah,F Khalid,et al.texture-based classification of workpiece surface images using the support vector machine[J].International Journal of Software Engineering,2015,9(10):147—160.
[11] Cristina Santos.Computer vision and fuzzy rules applied to an industrial desktop robot[J].IAssembly Automation,2006,26(1):59—68.
[12] R M R,Y W,Rangayyan R M,et al.Modeling and classification of knee-joint vibroarthrographic signals using probability density functions estimated with Parzen windows [C]// Vancouver,British Columbia,Canada,International Conference of the IEEE Engineering in Medicine & Biology Society:IEEE,2008:2099—2102.
[13] Lázaro M,Hayes M H,Figueiras-Vidal A R.Training neural network classifiers through Bayes risk minimization applying unidimensional Parzen windows[J].Pattern Recognition,2018,77(5):204—215.
[14] Vesal S,Diazpinto A,Ravikumar N,et al.Semi-automatic algorithm for breast mri lesion segmentation using marker-controlled watershed transformation[C]//Nuclear Science Symposium and Medical Imaging Conference.Zeiss Meditec,Conference:IEEE 2017:1—3.
[15] Kumar A.Deblurring of motion blurred images using histogram of oriented gradients and geometric moments[J].Signal Processing Image Communication,2017,55(7):55—65.
[16] Yeung D Y,Chow C.Parzen-window network intrusion detectors[C]// 16th International Conference on.Quebec,Canada:IEEE,2002:385—388.
[17] Babich G A,Camps O I.Weighted Parzen windows for pattern classification[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1996,18(5):567—570.
[18] Panahi N,Shayesteh M G,Mihandoost S,et al.Recognition of different datasets using PCA,LDA,and various classifiers[C]// Application of Information and Communication Technologies (AICT2011 ).Baku,Azerbaijan:IEEE,2011:1—5.
[19] 郑运平,常宜斌.一种新的灰度图像的快速矩计算算法[J].计算机学报,2017(11):2619—2634.Zheng Yunping,Chang Yibin.A novel fast calculation algorithm of geometric moments for gray images[J].Chinese Journal of Computers,2017(11):2619—2634.
[20] Pluhacek M,Senkerik R,Davendra D,et al.On the behavior and performance of chaos driven PSO algorithm with inertia weight[J].Computers & Mathematics with Applications,2013,66(2):122—134.
[21] 谭叶青,李萍,李潇,等.基于BreedPSO算法及交叠法的紧凑型大角度双自由曲面透镜设计[J].光学技术,2018,44(1):123—128.Tan Yeqing,Li Ping,Li Xiao,et al.Optimization of double freeform-surface lens with large view angle based on BreedPSO algorithm and overlapping method[J].Optical Technique,2018,44(1):123—128.
[22] 张学典,王瑞.优化算法用于变焦耳镜的光学系统设计[J].光学技术,2016,42(5):396—399.Zhang Xuedian,Wang Rui.Design of a zoom lens system of an otoscope with optimization algorithm [J].Optical Technique,2016,42(5):396—399.
[23] Ruofei D.Research on the optimization model of the supply chain with concurrent negotiation particle swarm optimization[J].International Journal of u-and e-Service,Science and Technology,2016,9(8):307—318.
[2] 朱海波.基于视觉引导的工业机器人工件搬运技术研究[D].沈阳:沈阳工业大学,2013:20—35.Zhu Haibo.Research on workpiece handling technology with industrial robot based on vision guided[D].Shenyang:Shenyang University of Technology,2013:20—35.
[3] 刘正琼,万鹏,凌琳,等.基于机器视觉的超视场工件识别抓取系统[J].机器人,2018,40(3):294—300.Liu Zhengqiong,Wan Peng,Ling Lin,et al.Recognition and grabbing system for workpieces exceeding the visual field based on machine vision[J].Robot,2018,40(3):294—300.
[4] 兰佳,叶文华.基于立体视觉的工件识别与定位系统研究[J].机械制造与自动化,2017,46(5):188—193.Lan Jia,Ye Wenhua.Research on Parts' recognition and orientation algorithm based on stereo vision[J].Machine Building & Automation,2017,46(5):188—193.
[5] Lowe D G.Distinctive image features from scale-invariant key points [J].International Journal of Computer Vision,2004,60(2):91—110.
[6] 罗四,李少洪.基于相关成分分析法的雷达目标识别仿真研究[J].系统仿真学报,2009,21(2):594—598.Luo Si,Li Shaohong.Simulation research of radar target recognition based on RCA[J].Journal Of System Simulation,2009,21(2):594—598.
[7] 李俊威.基于不变矩的人形识别技术研究[D].西安科技大学,2012:33—38.Li junwei.Research on moment invariant based human recognition technology[D].Xi'an University of Science and Technology,2012:33—38.
[8] 庞菲菲,张群飞,史文涛,等.基于Parzen窗的水下无线传感器网络目标定位方法[J].电子与信息学报,2017,39(1):45—50.Pang Feifei,Zhang Qunfei,Shi Wentao,et al.Target localization method based on parzen window in underwater wireless sensor network[J].Journal of Electronics and Information Technology,2017,39(1):45—50.
[9] Takeuchi Y,Shimizu H,Mukai I,et al.Automatic measurement of 3-dimensional coordinate measuring machine by means of CAD and image data[J].Journal of the Japan Society for Precision Engineering,1990,39(1):565—568.
[10] S H Darwish,L N Abdullah,F Khalid,et al.texture-based classification of workpiece surface images using the support vector machine[J].International Journal of Software Engineering,2015,9(10):147—160.
[11] Cristina Santos.Computer vision and fuzzy rules applied to an industrial desktop robot[J].IAssembly Automation,2006,26(1):59—68.
[12] R M R,Y W,Rangayyan R M,et al.Modeling and classification of knee-joint vibroarthrographic signals using probability density functions estimated with Parzen windows [C]// Vancouver,British Columbia,Canada,International Conference of the IEEE Engineering in Medicine & Biology Society:IEEE,2008:2099—2102.
[13] Lázaro M,Hayes M H,Figueiras-Vidal A R.Training neural network classifiers through Bayes risk minimization applying unidimensional Parzen windows[J].Pattern Recognition,2018,77(5):204—215.
[14] Vesal S,Diazpinto A,Ravikumar N,et al.Semi-automatic algorithm for breast mri lesion segmentation using marker-controlled watershed transformation[C]//Nuclear Science Symposium and Medical Imaging Conference.Zeiss Meditec,Conference:IEEE 2017:1—3.
[15] Kumar A.Deblurring of motion blurred images using histogram of oriented gradients and geometric moments[J].Signal Processing Image Communication,2017,55(7):55—65.
[16] Yeung D Y,Chow C.Parzen-window network intrusion detectors[C]// 16th International Conference on.Quebec,Canada:IEEE,2002:385—388.
[17] Babich G A,Camps O I.Weighted Parzen windows for pattern classification[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1996,18(5):567—570.
[18] Panahi N,Shayesteh M G,Mihandoost S,et al.Recognition of different datasets using PCA,LDA,and various classifiers[C]// Application of Information and Communication Technologies (AICT2011 ).Baku,Azerbaijan:IEEE,2011:1—5.
[19] 郑运平,常宜斌.一种新的灰度图像的快速矩计算算法[J].计算机学报,2017(11):2619—2634.Zheng Yunping,Chang Yibin.A novel fast calculation algorithm of geometric moments for gray images[J].Chinese Journal of Computers,2017(11):2619—2634.
[20] Pluhacek M,Senkerik R,Davendra D,et al.On the behavior and performance of chaos driven PSO algorithm with inertia weight[J].Computers & Mathematics with Applications,2013,66(2):122—134.
[21] 谭叶青,李萍,李潇,等.基于BreedPSO算法及交叠法的紧凑型大角度双自由曲面透镜设计[J].光学技术,2018,44(1):123—128.Tan Yeqing,Li Ping,Li Xiao,et al.Optimization of double freeform-surface lens with large view angle based on BreedPSO algorithm and overlapping method[J].Optical Technique,2018,44(1):123—128.
[22] 张学典,王瑞.优化算法用于变焦耳镜的光学系统设计[J].光学技术,2016,42(5):396—399.Zhang Xuedian,Wang Rui.Design of a zoom lens system of an otoscope with optimization algorithm [J].Optical Technique,2016,42(5):396—399.
[23] Ruofei D.Research on the optimization model of the supply chain with concurrent negotiation particle swarm optimization[J].International Journal of u-and e-Service,Science and Technology,2016,9(8):307—318.