摘要
为了提高道路环境图像识别中抗干扰性和鲁棒性,提出了一种基于粒子群优化的道路检测算法。该方法针对道路几何特征引入直线变形模型描述道路结构。利用似然估计概率评价道路图像与模型的匹配程度。在先验知识的条件下结合粒子群优化算法搜索参数空间的全局理想值。同时,在模型中定义友谊区域减少运算复杂度与干扰因素。通过对实际大量路面的检测,证明了该方法在阴影、光照等干扰环境下对道路检测的有效性和实用性。
A new detection algorithm based on particle swarm optimization（PSO） is proposed to enhance the robustness and the interference immunity of the recognition in road circumstances.The algorithm introduces a deformable line model with road geometry characteristics and describes the road structure,and utilizes likelihood density probability for evaluating how correctly road images match the model.Then,PSO is incorporated into searching global ideal value in parameter space with the prior knowledge.Besides,a friendly region associated with images is defined to decrease computation difficulties and other interfering factors.Experiments with road images from Canegie Mellon University（CMU） lab and the system prove the effectiveness and the applicability of the algorithm.
出处
《数据采集与处理》
CSCD
北大核心
2010年第3期384-388,共5页
Journal of Data Acquisition and Processing
基金
国家自然科学基金(60705020
60873151)资助项目
关键词
智能车辆系统
直线变形模型
似然概率
粒子群优化算法
intelligent vehicle systems
deformable line model
likelihood density probability
particle swarm optimization
