Research of the Fatigue Driving Real-time Monitoring Based on DSP
This research proposes a fatigue driving real-time monitoring system based on DSP.Firstly,a facial skin-color model is built based on facial sample database,which is then used in a motion and skin detecting algorithm to locate face;Secondly,hybrid projection method is proposed to locate eyes by combining the methods of the traditional integral projection and differential projection;after that,the blink times are determined by calculating the change of distance between the upper and lower eyelids.Here the real-time analysis is realized by using SWI and PRD in the DSP/BIOS provided by CCS2.Experiment results show that this system can not only locate eyes but also calculate the eyelid blink times accurately,so as to reflect the driversfatigue degree.
Yu Liu Jie Yang Peng Wu Jing-wei Zhang
The School of Information Engineering,Wuhan University of Technology Wuhan,Hubei,P.R.China 430063
国际会议
9th International Conference on Signal Processing(第九届国际信号处理学术会议)(ICSP08)
北京
英文
2008-10-26(万方平台首次上网日期,不代表论文的发表时间)