{"title":"Study on damaged region segmentation model of image","authors":"Huaming Liu, Yun Chen, Xuehui Bi","doi":"10.1109/ICICISYS.2010.5658284","DOIUrl":null,"url":null,"abstract":"After studying the damaged region classification of Thangka image, and in light of the specific damaged situation, as for damaged region can be segmented accurately and the advantages of different image segmentation algorithms can be full played, it is proposed a damaged region segmentation model of image. Model integrates different image segmentation algorithms, the system can segment damaged regions using different algorithms according to the feature of the image damaged regions, so it can avoid looking for “universal” algorithms for segmenting image damaged regions. For the image segmentation results, it is needed to evaluate through by segmentation evaluation, at the same time, considering the error segmented region condition, here subjective evaluation is introduced in the model. The system selects algorithm, and the results of the evaluation and so on, these features are stored in the information database of image segmentation, decision-making module analysis and learning continually, and will optimize all types of segmentation algorithm in the information database. Decision-making module can make use of the historical information to guide the image damaged region segmentation, increase system segmentation efficiency and accuracy.","PeriodicalId":339711,"journal":{"name":"2010 IEEE International Conference on Intelligent Computing and Intelligent Systems","volume":"30 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2010-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2010 IEEE International Conference on Intelligent Computing and Intelligent Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICICISYS.2010.5658284","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
After studying the damaged region classification of Thangka image, and in light of the specific damaged situation, as for damaged region can be segmented accurately and the advantages of different image segmentation algorithms can be full played, it is proposed a damaged region segmentation model of image. Model integrates different image segmentation algorithms, the system can segment damaged regions using different algorithms according to the feature of the image damaged regions, so it can avoid looking for “universal” algorithms for segmenting image damaged regions. For the image segmentation results, it is needed to evaluate through by segmentation evaluation, at the same time, considering the error segmented region condition, here subjective evaluation is introduced in the model. The system selects algorithm, and the results of the evaluation and so on, these features are stored in the information database of image segmentation, decision-making module analysis and learning continually, and will optimize all types of segmentation algorithm in the information database. Decision-making module can make use of the historical information to guide the image damaged region segmentation, increase system segmentation efficiency and accuracy.