论文部分内容阅读
本文提出一种应用 BP神经网络识别工程图纸扫描图象分割后的图形符号的方法。此方法先对二值图象进行特征提取 ,并提出了改进学习算法 ,以加快收敛速度 ,从而实现图象识别。
This paper presents a method of using BP neural network to identify the graphic symbols after the engineering drawings are scanned. In this method, feature extraction is first performed on binary images, and an improved learning algorithm is proposed to speed up the convergence and achieve image recognition.