基于模糊小波神经网络的交通标志识别方法研究  

Research on Road Sign Identification Method Based on Fuzzy Wavelet Network

作　　者：范必双[1] 王玉凤[2] 王英健[1] 

机构地区：[1]长沙理工大学电气与信息工程学院,湖南长沙410076 [2]长沙理工大学计算机与通信工程学院,湖南长沙410076

出　　处：《计算机仿真》2005年第9期201-204,共4页Computer Simulation

摘　　要：对交通标志进行实时、正确的识别,是车辆自动导航中一个重要方面。该文介绍了一种基于模糊小波神经网络的交通标志识别方法。该方法首先利用不变矩来提取图像特征,然后将特征向量输入模糊小波神经网络进行识别。该网络以小波函数作为模糊隶属函数,将模糊技术与神经网络相结合,利用神经网络实现模糊推理,并可对隶属函数的形状进行实时调整,从而使网络具有更强的学习和自适应能力。实验证明,该方法具有较高的识别精度和速度,在车辆自动导航中具有较高的应用价值。To identify the road signs correctly and in real time is an important part of vehicle auto navigation. A method of road sign identification based on fuzzy wavelet network is presented in this paper. In this method, the moment invariants are used to pick up the character of the images, then the eigenvector is identified with the fuzzy wavelet network. In the network, wavelet function is used as fuzzy membership function, and fuzzy technique is combined with neural network technique. The fuzzy inference is realized by neural network and the shape of membership function can be adjusted in real time. It endues the network with better capability of learning and self - adaptation. Experiments suggest that this method has higher identification precision and speed can be used to identify the road signs correctly and in real time in vehicle auto navigation. The system is simulated with MATLAB, and the result is satisfactory.

关 键 词：模糊小波神经网络 交通标志 自动导航 识别 不变矩 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]