是路径规划的实用方法，包含特征值与特征向量的提取、训练样本以及最后的识别，考虑雨衰 阴影 和多径影响，是一种双隐层反向传播神经网络，表示出两帧图像间各个像素点的相对情况。- Is a practical method of path planning, Contains the eigenvalue and eigenvector extraction, the training sample, and the final recognition, Consider shadow rain attenuation and multipath effects Is a two hidden layer back propagation neural network, Between two images showing the relative circumstances of each pixel.