基于三分量色差法的交通标志分割*

作者：胡牡丹，杨立敬，朱双东 日期：2009-11-16/span> 浏览：2997 查看PDF文档

基于三分量色差法的交通标志分割*

胡牡丹，杨立敬，朱双东
（宁波大学 信息科学与工程学院，浙江 宁波 315211）

摘要：为了改善交通标志在RGB模型下的颜色分割效果，提出了一种基于三分量色差法的动态阈值分割改进方法。该方法首先提取了RGB彩图的R、G、B三分量，并根据要提取的交通标志颜色选择了合适的特征算子对三分量作代数运算，得到了色差灰度图像；然后对该色差灰度图像使用了Otsu法动态选取阈值进行分割。实验结果表明，对不同自然光照条件下的交通标志场景图，该改进方法都能较好地实现交通标志和图像背景的分割，提高了交通标志形状检测步骤的准确度和速度。
关键词：交通标志；RGB模型；色差；图像分割；Otsu算法
中图分类号：TP391文献标识码：A文章编号：1001-4551(2009)10-0023-04

Traffic sign segment based on chromatic aberration of three colorcomponents
HU Mudan, YANG Lijing, ZHU Shuangdong
（College of Information Science and Engineering, Ningbo University, Ningbo 315211, China）
Abstract: To improving color segmentation effect of traffic sign under the RGB model, a method for image segmentation based on chromatic aberration was presented. Starting with separation of the RGB colorcomponents and selection of characteristic operator on the basis of the chromafeatures of traffic sign, algebra operation was performed with the selected operator on three RGB quantities, the grayscaled image of chromatic aberration was obtained, the Otsus threshold algorithm was subsequently applied to output the segmented image. The experimental result shows that the proposed algorithm can extract the traffic sign from the background well up to the technical standards under various natural illumination, and it can improve the performance in shape checking.
Key words: traffic sign; RGB model; chromatic aberration; image segmentation; Otsus algorithm
参考文献(References):
［1］GAVRILA D M. FRANKE U, WOHLER C, et al. Realtime vision for intelligent vehicles［J］. IEEE Instrumentation & Measurement Magazine,2001,4(2):22-27.
［2］朱双东,刘兰兰,陆晓峰.一种用于道路交通标志识别的颜色几何模型［J］.仪器仪表学报,2007,28(5):956-960.
［3］FANG C Y, FUH C S, YEN P S. An automatic road sign recognition system based on a computational model of human recognition processing［J］. Computer Vision and Image Understanding,2004,63(3):237-268.
［4］GONZALEZ R C, WOODS R E.数字图像处理［M］.阮秋琦，阮宇智,等，译.2版.北京:电子工业出版社,2006.
［5］SOETEDJO A, YAMADA K. An efficient algorithm for traffic sign detection［J］. Journal of Advanced Computational Intelligence and Intelligent Informatic,2006,10(3):409-417.
［6］黄志勇,孙光民,李芳.基于RGB视觉模型的交通标志分割［J］.微电子学与计算机,2004,21(10):147-148.
［7］OTSU N. A threshold selection method from graylevel histograms［J］. IEEE Transactions on System Man and Cybernetics,1979,9(1):62-69.