Adaptive Network-Based Fuzzy Inference System for Driving Status Analysis

• Main Authors: Chi-Ming Chen, 陳啟銘
• Other Authors: Jiann-Der Lee
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/49789636410910847334

碩士 === 長庚大學 === 電機工程研究所 === 95 === This thesis proposes an intelligent system which analyzes driving status. The system infrastructure is utilized two fixed cameras on the host vehicle. One is used to capture driver’s image in order to analyze driver’s sight line, and the other is used to capture image of road ahead for analyzing driving pattern. In the section of driver’s image, it’s necessary to utilize AdaBoost algorism to recognize face and then get the positions of eyes, nose and lips to diagnose the angles of driver’s head and driver’s sight line. In the section of road image, it’s used edge detection, inverse perspective transform and Hough transform to get lane line. Next is to analyze the mid-line of lane by adaptive network-based fuzzy inference system (ANFIS) for driving pattern recognition. The last step is to combine both driver’s sight line and driving pattern to judge if it’s safe driving.