{"title":"A projection twin SVM-based active contour model for image segmentation","authors":"Xiaomin Xie, Tingting Wang","doi":"10.1109/M2VIP.2016.7827264","DOIUrl":"https://doi.org/10.1109/M2VIP.2016.7827264","url":null,"abstract":"This paper presents an alternative criterion derived from the least squares projection twin support vector machine (LSPTSVM) for image segmentation. The proposed model treats image segmentation as pattern classification problem, and hence tries to seek the projected axis and center for the foreground and background intensities respectively. With level set representation, the discriminative function of LSTSVM is incorporated into the energy function of the active contour model (ACM), and drives the contour evolution accordingly. Experiment results demonstrate that our model holds the higher segmentation accuracy and more noise robustness, compared with the stand-alone CV and LSPTSVM models.","PeriodicalId":125468,"journal":{"name":"2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP)","volume":"166 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126936093","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Comparison of OpenCV's feature detectors and feature matchers","authors":"Frazer K. Noble","doi":"10.1109/M2VIP.2016.7827292","DOIUrl":"https://doi.org/10.1109/M2VIP.2016.7827292","url":null,"abstract":"There exists a range of feature detecting and feature matching algorithms; many of which have been included in the Open Computer Vision (OpenCV) library. However, given these different tools, which one should be used? This paper discusses the implementation and comparison of a range of the library's feature detectors and feature matchers. It shows that the Speeded-Up Robust Features (SURF) detector found the greatest number of features in an image, and that the Brute Force (BF) matcher matched the greatest number of detected features in an image pair. Given a benchmark image set, OpenCV's SURF detector found, on average, 1907.20 features in 1538.61 ms, and OpenCV's BF matcher, on average, matched features in 160.24 ms. The combination of the Binary Robust Invariant Scalable Key-points (BRISK) detector and BF matcher was found to be the highest ranked combination of OpenCV's feature detectors and feature matchers; on average, detecting and matching 1132.00 and 80.20 features, respectively, in 265.67 ms. It was concluded that if the number of features detected is important, the SURF detector should be used; else, if the number of features matched is important, the BF matcher should be used; otherwise, the combination of the OpenCV's BRISK feature detector and BF feature matcher should be used.","PeriodicalId":125468,"journal":{"name":"2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125197119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Modeling and optimization for circumferential uniformity of MOCVD showerhead hole arrangement","authors":"Jia Fang, X. Fu, J. Geng","doi":"10.1109/M2VIP.2016.7827325","DOIUrl":"https://doi.org/10.1109/M2VIP.2016.7827325","url":null,"abstract":"This paper presents a method for modeling and analyzing the circumferential uniformity of MOCVD showerhead hole arrangement. The coordinate expressions of MOCVD showerhead hole are derived from the initial position of the radial hole on the helix curve. Through rotating the strip observation window around the center of the showerhead face plate at an equal angle, a data set is obtained by computing the total area of holes in the strip observation window. The circumferential uniformity of hole layout is measured according to the standard deviation of this dataset, and then an optimal layout model of the holes is established. The result indicates that the sampling rotating angle has limited impact on the ultimate hole arrangement, which verifies the robustness of the proposed method. Finally, optimal parameter of the helix curve is built using the genetic algorithm (GA), and the optimal layout of showerhead holes is illustrated correspondingly.","PeriodicalId":125468,"journal":{"name":"2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP)","volume":"4 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134130495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Integer U transform and its application in image coding","authors":"Guo Fenhong, Xie Liyan, Xiong Changzhen","doi":"10.1109/M2VIP.2016.7827305","DOIUrl":"https://doi.org/10.1109/M2VIP.2016.7827305","url":null,"abstract":"In the image compression field, novel transform coding approaches are always necessary to obtain lower computational complexity and better reconstruction. Therefore, integer U-orthogonal transform and image coding algorithm is constructed in this paper. First, an 8×8 integer U-orthogonal transform through the discrete orthogonal U-system is constructed, and its fast algorithm is derived according to its symmetric and recursive property. Then, a quantization table meeting sequency characteristics of U-transform matrix is provided. Finally, the integer U transform and the new quantization table are applied to image coding. The comparative experimental results show that the coding effect of U-transform method is better than JPEG encoding method. Therefore, the integer U-transform image coding method is an alternative scheme.","PeriodicalId":125468,"journal":{"name":"2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP)","volume":"352 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131230328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Development and performance analysis of the flexible pneumatic artificial muscle","authors":"Yongqiang Gong, Chenxi Ren, Xingsong Wang, Bing Zhang","doi":"10.1109/M2VIP.2016.7827280","DOIUrl":"https://doi.org/10.1109/M2VIP.2016.7827280","url":null,"abstract":"The excellent performance, such as lightweight, consistent, high power output, of pneumatic artificial muscles makes them attractive for use in mobile robots and prosthetic appliances. In this paper, we make a kind of artificial muscle only by using Kevlar fibers and silicon rubber, which is called flexible pneumatic artificial muscle (FPAM). Their flexible nature permits them to be used without rigid joints. Without these joints, however, sealing FPAM is challenging. We make the reinforcing braid of FPAM soft by weaving it from Kevlar fibers. The reinforcing fibers and the soft silicone rubber can be used to seal FPAM with the mutual cooperation between them, whereas other proposed self-sealing techniques rely on the addition of special rigid component, the technique presented in this paper can be implemented without additions of this kind. A self-sealing silicone tube is used in this new structure. The Pressure characteristics of FPAM and were compared with McKibben artificial muscles ones. The result shows that the contraction force of the new artificial muscle is slightly smaller than the McKibben artificial muscles. We also test the basic properties of silicone rubber. A set of experimental frequency response characteristics are also given in this paper. Due to there is no metal joint, the artificial muscle can be applied in a wider range of areas, such as wearable medical devices.","PeriodicalId":125468,"journal":{"name":"2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121819029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"An algorithm of filling the mutational structure for FDM prototyping","authors":"Yubo Hu, Min Dai, Zhisheng Zhang, Junqing Zhang","doi":"10.1109/M2VIP.2016.7827326","DOIUrl":"https://doi.org/10.1109/M2VIP.2016.7827326","url":null,"abstract":"Fused Deposition Modeling (FDM) is an additive manufacturing process that can quickly produce geometrically complex parts through the melting, depositing and solidifying of thermoplastics, layer by layer. This paper proposes an algorithm of the filling mutational structure which is used to fill the mutational structure with a high density and the other regions with a low filling density. Through analyzing the area classification and geometric feature of the printing model, two-dimensional filling algorithm and three-dimensional filling algorithm are proposed respectively. By experimental verification, this filling algorithm not only improves the strength of the mutational structure but also saves the printing time and enhances the working efficiency.","PeriodicalId":125468,"journal":{"name":"2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131082311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research and development of mecanum-wheeled omnidirectional mobile robot implemented by multiple control methods","authors":"Qian Jia, Mulan Wang, Shuqing Liu, Jianjing Ge, Chen Gu","doi":"10.1109/M2VIP.2016.7827337","DOIUrl":"https://doi.org/10.1109/M2VIP.2016.7827337","url":null,"abstract":"A mecanum-wheeled robot is a kind of popular and representative omnidirectional mobile robot, which can move in all directions on the work plane, e.g. forth and back, sideway and spin. In this paper, a kinematics model of a mecanum-wheeled robot is analyzed and discussed. Based on the model, two omnidirectional mobile robots are designed and developed. Because robots implemented by multiple control methods have more flexible and adaptable in different occasions, three control methods: speech recognition, Bluetooth and infrared remote control are applied to the two robots. Then, motion of the two robots controlled by three modes are compared and concluded.","PeriodicalId":125468,"journal":{"name":"2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131133832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Intelligent selection and optimization of measurement poses for a comprehensive error model identification of 6-DOF serial robot","authors":"Xiaoyan Chen, Qiuju Zhang, Yilin Sun","doi":"10.1109/M2VIP.2016.7827279","DOIUrl":"https://doi.org/10.1109/M2VIP.2016.7827279","url":null,"abstract":"Non-geometric errors mainly caused by the joint compliance should be identified and compensated as well as geometric errors to improve the accuracy. This paper presents a new comprehensive error model consisting of both geometric and compliance parameters. A new approach is proposed for intelligent selection and optimization of measurement poses based on interference detection method and linearly decreasing weight particle swarm optimization (LinWPSO) algorithm. Simulation results on a 6-DOF serial industrial robot demonstrate that using the optimal measurement poses can significantly improve the calibration accuracy and measurement efficiency.","PeriodicalId":125468,"journal":{"name":"2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125641490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Optimized online guidance algorithm for the fixed-wing flying robot","authors":"S. Shan, Z. Hou, Yue Li","doi":"10.1109/M2VIP.2016.7827271","DOIUrl":"https://doi.org/10.1109/M2VIP.2016.7827271","url":null,"abstract":"The classical nonlinear guidance algorithm for a fixed-wing flying robot has the shortcoming of the unchangeable guidance length. To improve the algorithm, a novel guidance method is proposed based on the guidance point optimized online. The kinetic equations and the classical nonlinear guidance algorithm are introduced. The effects of the guidance length to the tracking performance are analyzed. A modified guidance algorithm is designed, and the procedure is described. The simulation experiments show that the proposed method can track complex trajectories excellently because of the compromise of the stability and the precision.","PeriodicalId":125468,"journal":{"name":"2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP)","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127821019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Target accurate positioning based on the point cloud created by stereo vision","authors":"Mo Yuda, Zou Xiangjun, Situ Weiming, Luo Shaofeng","doi":"10.1109/M2VIP.2016.7827268","DOIUrl":"https://doi.org/10.1109/M2VIP.2016.7827268","url":null,"abstract":"To solve the problem of workpiece target accurate positioning in industrial environment, we proposed a method that used object's surface point cloud which created by stereo matching of binocular vision. Firstly, found out the ROI(Region Of Interest) of the stereo matching point cloud by template matching, and used an algorithm of noise reduction to gain a clean ROI point cloud. And then, extracted the perfect point cloud of object's surface from object's 3d-model, and used improved iterative closest point algorithm to do point cloud registration that can gain the accurate pose of object. Experiment shows that positioning accuracy is less than 1.5 mm(Euclidean Distance) which can meet the needs of industrial robot doing sorting or accurate grabing.","PeriodicalId":125468,"journal":{"name":"2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128032265","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}