摘要
智能交通系统是近年来国内外广泛关注的研究课题，在基于计算机视觉系统的交通标志识别中，关键步骤之一是如何快速有效地检测并提取交通标志。该文提出了基于色彩的指示标志检测与提取的快速有效方法。它包括：对实景图像进行色调不变的彩色增强，彩色聚类，形态滤波，模板匹配，特征点判别等步骤。最后，由搜索到的特征点从增强后的彩色图像中抽取指示标志。实验显示该方法能快速有效地从实景图像中检测指示标志。
In recent years, intelligent transportation system becomes a topic being widely interested.How to detect traffic sign from scene is a key step in traffic sign recognition system based on computer vision.This paper proposes an effective method for detecting indicatory signs in color image.The method includes color enhancing with. hue-invariance,color clustering, morphological filtering,pattern matching,feature judging. Finally,the traffic sign is extracted from the enhanced color image.Experimental results show that the proposed method can detect indicatory signs quickly and efficiently.
出处
《计算机工程与应用》
CSCD
北大核心
2000年第4期169-172,共4页
Computer Engineering and Applications
基金
浙江省的自然科学基金资助!(项目批准号:No.698064)
关键词
智能交换系统
交通标志
计算机
图像处理
intelligent transportation System,traffic sign,computer image processing
