Font Size: a A A

The Research Of Driver Fatigue System In Low Illuminance Based On DSP/BIOS

Posted on:2008-12-04Degree:MasterType:Thesis
Country:ChinaCandidate:C WangFull Text:PDF
GTID:2178360212479691Subject:Signal and Information Processing
Abstract/Summary:PDF Full Text Request
Driving in case of drowse is a most important factor of traffic accidents. Now many countries are active in studying drowsy monitor.This article design and implements a set of driver fatigue detection system under low degree of illumination condition according to the study of physiological changes about the fatigue based on the platform of SEED-VPM642, the system uses the infrared camera gathering picture information and examines the state of eyes through the method of picture processing and judges the fatigue degree of driver without affecting the normal driving situation.DSP/BIOS is a simple real time OS. Procedure is different from development DSP of tradition, the developer does not need to control DSP completely, operation relevant to the hardware all can draw support from the chip that DSP/BIOS provides supporting a warehouse to be completed. Consumer's application is built above DSP/BIOS basis, preferential step being interrupted under the controller, according to the mission in whose, lines up to wait to carry out. This article finished the instauration of project, configuration of CDB files, communications between tasks and achieved the driver of capturing and displaying of video. The program can be functioned in the SEED-VPM642 breaking away from PC.How to locate the eyes is a linchpin to the system. This system has adopt the effect fixing position well, and calculate the allocation algorithm that amounts realize relatively less and easily. Firstly, the article use frames difference and binarization to abstract the motion human body and background border. And then make use of part statistics filtering to denoising and takeoff background information which is motionless object. Using edge detection to enhance the boundary of moving object. Locate face position using region search. In face area, the system track the eyes area using the way of binarization,erosion,dilation and integration projection.Compare with the threshold which is decided after lots of experiments, the state oftwo eyes can be acquired. Judge the driver state according PERCLOS, alarm when he or she is fatigue.Many tests show that the system can detect blinking times and visible pupil area in real time. So it can detect the degree of fatigue efficaciously.
Keywords/Search Tags:fatigue driving, part statistic filter, SEED-VPM642, PERCLOS
PDF Full Text Request
Related items