{"title":"Image sequence segmentation using the gradient structure tensor method and self-organizing map","authors":"Tin Mon Mon Swe, T. Kondo, W. Kongprawechnon","doi":"10.1109/ECTICON.2008.4600462","DOIUrl":null,"url":null,"abstract":"This paper presents a technique for segmenting image sequences using the gradient structure tensor method (GSTM) and the self-organizing feature map neural network technique (SOM). GSTM accurately and robustly estimates motion vectors in an image sequence, while SOM classifies the estimated motion vectors in an unsupervised manner. Consequently, the segmentation of an image sequence is achieved. Simulation results show that the combination of the two techniques is successful for both synthetic and real image sequences.","PeriodicalId":176588,"journal":{"name":"2008 5th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology","volume":"60 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2008 5th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ECTICON.2008.4600462","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
This paper presents a technique for segmenting image sequences using the gradient structure tensor method (GSTM) and the self-organizing feature map neural network technique (SOM). GSTM accurately and robustly estimates motion vectors in an image sequence, while SOM classifies the estimated motion vectors in an unsupervised manner. Consequently, the segmentation of an image sequence is achieved. Simulation results show that the combination of the two techniques is successful for both synthetic and real image sequences.