基于主动轮廓模型的行人边缘检测应用研究  

Research on Pedestrian Edge Detection Application Based on Active Contour Models

作　　者：陈康[1] 黄影平[1] 张仁杰[1] 

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093

出　　处：《计算机与现代化》2012年第12期66-68,72,共4页Computer and Modernization

摘　　要：应用主动轮廓模型对道路现场的行人轮廓和障碍物进行精确轮廓的提取,为基于视觉的辅助驾驶系统中的潜在障碍物分类提供一个有效的手段。首先讨论主动轮廓模型,然后分别应用GVF-Snake模型和level set模型对行人轮廓边缘进行提取实验,并对实验结果进行分析。This paper uses the active contour models to extract the outline of a pedestrian and obstacles in a road scene accurately, provides an effective means for potential obstacles classification which based on visual auxiliary driving system. First, the paper discusses the active contour models, then uses GVF - snake and level set model to make extraction experiment for pedestrian contour edge, and the experimental results are analyzed.

关 键 词：图像边缘检测 主动轮廓模型 GVF-SNAKE LEVEL SET 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]