介绍了人工神经网络技术在管道内表面质量实时检测系统中的应用，该系统采用BP神经网络作为规则检测器对被测图象进行特征提取和分类识别，并利用改进的ＢＰ算法使网络的学习过程以较快的速度收敛于全局最小点。

This paper is concerned with an application of artificial neural network to the real-time detecting system which can inspect the inner walls of pipes. As a regular detector, the BP neural network is used for extracting features of the images inspected and classifying the images in the system, and an improved BP algorithm which was named for BP′is used for making the learning procedure of the net converge to the minimum of overall situation at great rate.