D. X. Tran, T. Nguyen, H. Bui, S. T. Nguyen, Hong X. Hoang, T. C. Pham, T. de Souza-Daw
{"title":"Dual PTZ cameras approach for security face detection","authors":"D. X. Tran, T. Nguyen, H. Bui, S. T. Nguyen, Hong X. Hoang, T. C. Pham, T. de Souza-Daw","doi":"10.1109/CCE.2014.6916751","DOIUrl":null,"url":null,"abstract":"In security applications, obtaining high resolution face images is challenging since cameras are usually placed at a far distance. In this paper, we employ two approaches to detect face at a distance including using one or dual pan-tilt-zoom (PTZ) cameras to zoom in and capture proximate faces while still maintaining a large observing area. While most current approaches require a complex calibration and coordinate transformation, our system offers simple image coordinate transformation and no calibration in 3D real world coordinate. The system testing results showed high face detection accuracy, 100% zooming accuracy as well as reasonable camera moving and zooming time.","PeriodicalId":377853,"journal":{"name":"2014 IEEE Fifth International Conference on Communications and Electronics (ICCE)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2014-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2014 IEEE Fifth International Conference on Communications and Electronics (ICCE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CCE.2014.6916751","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 5
Abstract
In security applications, obtaining high resolution face images is challenging since cameras are usually placed at a far distance. In this paper, we employ two approaches to detect face at a distance including using one or dual pan-tilt-zoom (PTZ) cameras to zoom in and capture proximate faces while still maintaining a large observing area. While most current approaches require a complex calibration and coordinate transformation, our system offers simple image coordinate transformation and no calibration in 3D real world coordinate. The system testing results showed high face detection accuracy, 100% zooming accuracy as well as reasonable camera moving and zooming time.