Font Size: a A A

Research On The Method Of Driver Fatigue Detection

Posted on:2013-01-08Degree:MasterType:Thesis
Country:ChinaCandidate:X Y LiFull Text:PDF
GTID:2218330371960904Subject:Signal and Information Processing
Abstract/Summary:PDF Full Text Request
With the development of the society and the improvement of the people's living standard, the automotive industry has also been developing rapidly. The traffic accidents caused by driver fatigue are increasing, so the detection of driver fatigue becomes the concerned topic of the society. How to detect the fatigue state of the driver effectively has been becoming a hot topic of the field, which is in the automotive safety technology.According to the requirement of real-time monitoring fatigue driving, this paper presents a comprehensive fatigue detection method, which includes visual detection, pulse signal detection and multi-point pressure detection. Visual detection mainly carries out feature extraction and the state detection of eyes and mouth, using the methods of machine vision and image processing. First, skin color detection is used to locate the face, then gray integral projection and curve smoothing are used to locate the eyes and mouth. Finally, the real-time state of the eyes and the mouth are detected and the value of PERCLOS is calculated to judge fatigue. Pulse signal detection collects the human pulse signal by the sensor of pulse, and sends it to the PC by the wireless communication module called nRF905, then the PC analyses the real-time data and compares with the normal pulse frequency, then judges fatigue state through the speed of the human pulse signal. The principle of multi-point pressure detection is that the body and the back will be still in a long time during fatigue state period, so the sensors are used to detect pressure points to judge fatigue. After that the system can judge the driver's fatigue state comprehensively in the method of averaging three detection results.The multi-mode detection system this paper presents has the advantages that the operation is fast in the location of the face, the mouth and the eyes, accurate and convenient to take, which can meet the requirement of real-time monitoring and detect the fatigue state of the driver accurately, so it can reduce the loss of life and property.
Keywords/Search Tags:Face Detection, Fatigue Detection, PERCLOS
PDF Full Text Request
Related items