Research of Real-time Driver Fatigue Detection System Based on ADSP-BF533
|School||Xi'an University of Electronic Science and Technology|
|Course||Circuits and Systems|
|Keywords||ADSP-BF533 PERCLOS Kalman Face Detection Eye Location|
Driver fatigue driving detection system is an important part of the active safety warning systems . Study how effective driver fatigue detection to reduce traffic accidents has a very important practical significance . With the rapid development of DSP technology , face recognition technology has been widely used , and the papers of face recognition theory based on a driver's eye condition monitoring system based on the ADSP-BF533 , the system according to PERCLOS Physiologica judging principle, to determine the fatigue state of the driver . This paper describes the status quo of research on driver fatigue at home and abroad , focus on eyes closed fatigue evaluation methods - PERCLOS . Comprehensive analysis of the face detection algorithm based on face detection method based on skin color segmentation and positioning . Taking into account the complexity of the human face detection , face region tracking forecast choice of the Kalman filter . Eye positioning , gray integral projection method to achieve eye location . On the hardware design , the choice of the ADI's ADSP-BF533 chip build the hardware platform , including video capture module , LCD module , video storage module , keyboard , and voice warning module . The experiment proved that the system is capable of more accurate real-time detection of driver fatigue . The system selects the ADSP-BF533 fully embody the core processing chip miniaturization, high efficiency , integrated design requirements . Optimize the design of face tracking and detection , DSP program has made some improvements and further research direction .