摘要
提出了一种检测和识别交通标志的方法。该方法根据交通标志的颜色和形状,利用支持向量机的非线性分类能力将其图像区域从实景图像中检测和提取出来,然后利用具有多样性、较强容噪能力的模糊免疫网络来识别。使用不同环境下的实景图像进行实验,结果证明本方法具有平移、旋转、缩放、拉伸不变性和较强的容噪能力。
A detection and recognition method for traffic signs are presented. Traffic sign regions are detected and extracted from real world scenes on the basis of their color and shape features using non-linear classification capability of support vector machine, then recognized by fuzzy immune networks which has diversity and well tolerating noise capability. Using real road images in different environment conditions, experimental results show that the method has affine invariability of translation, rotation, scale, distortion and can well tolerating noise.
出处
《计算机工程与设计》
CSCD
北大核心
2006年第9期1542-1544,共3页
Computer Engineering and Design
基金
湖南省自然科学基金项目(03JJY3101)
湖南省教育厅科研基金项目(04C076)
关键词
交通标志
图像识别
支持向量机
模糊免疫网络
traffic signs
image recognition
support vector machine
fuzzy immune networks
