Title :
A novel approach of road recognition based on deformable template and genetic algorithm
Author :
Liu, Tie ; Zheng, Nanning ; Cheng, Hong ; Xing, Zhengbei
Author_Institution :
Inst. of Intelligence & Robotics, Xi´´an Jiaotong Univ., China
Abstract :
Road recognition based on vision navigation is an important task in intelligent vehicle research. Road image is usually influenced by shadows, noise and discontinuity of road contour, and this induces the traditional edge-based algorithm´s robustness decrease greatly. To avoid negative influence from threshold selection, this paper presents a deformable template and genetic algorithm based road recognition algorithm. Firstly, preprocess road image with edge operator to get the edge information, then construct a deformable template model of road contour and its likelihood function which define the fitting degree for a given template deformable parameter, finally genetic algorithm is used to search the global maximal value of the likelihood function to get the optimal parameter of the deformable template model of road contour. Experimental results indicate that the algorithm has strong robustness for shadows, noise and discontinuity of road contour.
Keywords :
computer vision; genetic algorithms; image recognition; maximum likelihood estimation; road traffic; deformable template; genetic algorithm; global maximal value; intelligent vehicle research; likelihood function; road image; road recognition; traditional edge-based algorithm; vision navigation; Cameras; Deformable models; Genetic algorithms; Image edge detection; Intelligent robots; Intelligent vehicles; Navigation; Noise robustness; Remotely operated vehicles; Road vehicles;
Conference_Titel :
Intelligent Transportation Systems, 2003. Proceedings. 2003 IEEE
Print_ISBN :
0-7803-8125-4
DOI :
10.1109/ITSC.2003.1252684