Khadidja Sadeddine, R. Djeradi, F. Chelali, A. Djeradi
{"title":"Recognition of Static Hand Gesture","authors":"Khadidja Sadeddine, R. Djeradi, F. Chelali, A. Djeradi","doi":"10.1109/ICMCS.2018.8525908","DOIUrl":null,"url":null,"abstract":"Human-Human (deaf people-ordinary people) and Human-Machine communication have become an interesting area of research requiring robust recognition systems. The paper proposes an implementation of hand posture recognition using three databases (Arabic Sign Language ArSL, American Sign Language ASL, and NUS hand posture) under uniform background. For that Hu's invariant moments descriptor, Local Binary Pattern (LBP) descriptor, Zernike moments descriptor, and Generic Fourier descriptor (GFD) are employed for the image characterization. Classification task is based on neural networks. The paper implements the fusion of the descriptors in order to increase the performance. Best recognition rates are reached for American Language with 93.33% for LBPD and same accuracy for NUS dataset with GFD.","PeriodicalId":272255,"journal":{"name":"2018 6th International Conference on Multimedia Computing and Systems (ICMCS)","volume":"101 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 6th International Conference on Multimedia Computing and Systems (ICMCS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICMCS.2018.8525908","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 8
Abstract
Human-Human (deaf people-ordinary people) and Human-Machine communication have become an interesting area of research requiring robust recognition systems. The paper proposes an implementation of hand posture recognition using three databases (Arabic Sign Language ArSL, American Sign Language ASL, and NUS hand posture) under uniform background. For that Hu's invariant moments descriptor, Local Binary Pattern (LBP) descriptor, Zernike moments descriptor, and Generic Fourier descriptor (GFD) are employed for the image characterization. Classification task is based on neural networks. The paper implements the fusion of the descriptors in order to increase the performance. Best recognition rates are reached for American Language with 93.33% for LBPD and same accuracy for NUS dataset with GFD.