Title :
Traffic measuring and authentication on digital video
Author :
Xuan, Guorong ; Xiao, Yang ; Yang, Xiaoguang ; Chen, Jidong ; Yang, Chengyun ; Zhang, Ruhua
Author_Institution :
Dept. of Comput. Sci., Tongji Univ., Shanghai, China
Abstract :
Several schemes for traffic monitor on video are developed. The first is the dynamic gap technique by means of digital video processing instead of subsurface electromagnetic detectors. The dynamic background refreshing and threshold selection are obtained by statistical methods. The second is density and queue length measurement in a traffic detection region. The third is interpretation and identification of traffic activities by hidden Markov models and entropy minimization to discover the relationships between hidden causes and observed scenes. The learning system learns a concise model of scene behavior directly from optical flow. The fourth is to authenticate traffic images with robust digital watermarking technique and hiding data in image losslessly.
Keywords :
computerised monitoring; data encapsulation; hidden Markov models; identification; image sequences; minimum entropy methods; statistical analysis; traffic control; video signal processing; watermarking; data hiding; density measurement; digital video processing; dynamic background refreshing; dynamic gap technique; entropy minimization; hidden Markov model; identification; interpretation; learning system; optical flow; queue length measurement; robust digital watermarking technique; statistical methods; subsurface electromagnetic detectors; traffic authentication; traffic detection region; traffic image authentication; traffic measuring system; traffic monitor; watermarking; Authentication; Detectors; Electromagnetic measurements; Entropy; Hidden Markov models; Layout; Length measurement; Monitoring; Statistical analysis; Traffic control;
Conference_Titel :
Intelligent Transportation Systems, 2003. Proceedings. 2003 IEEE
Print_ISBN :
0-7803-8125-4
DOI :
10.1109/ITSC.2003.1252046
