基于BP神经网络管道裂缝图像分割

摘要

BP neural network method for segmentation the pipeline crack image was used. Select 100 pixel RGB values of the image as training samples of a network. The improved BP algorithm was used to train the parameter of the network. The effective parameter was a-chieved after 226 times training. The error was le- 5, expected segmentation result could be achieved. The result of this experiment showed that the segmentation result was better than the traditional method of segmentation.%采用BP神经网络方法对管道裂缝图像进行分割,选取图像中100个像素点的RGB值做为网络的训练样本,用改进BP算法对神经网络权值进行训练,经过226次循环后,误差将为0.00001,获取了有效的网络权值,实现了裂缝图像与背景图像的分割；实验证明、该方法分割的结果优于传统的分割方法.