论文部分内容阅读
随着计算机技术、图像处理、人工智能等理论的迅速发展,神经网络图像识别作为一种新型图像识别技术也迅速发展起来并在各个领域广泛应用。文章主要内容是利用人工神经网络理论知识研究在图像识别中的应用的图像识别技术,同时对BP神经网络进行了分析,介绍隐含层节点个数的选择的公式,并针对BP算法存在的问题提出增加动量因子的BP算法等方面的改进,从而加快了训练速度从而改善了BP网络的学习效果。
With the rapid development of computer technology, image processing, artificial intelligence and other theories, neural network image recognition as a new image recognition technology is rapidly developed and widely used in various fields. The main content of this article is to use the theory of artificial neural network to study the image recognition technology used in image recognition. At the same time, the BP neural network is analyzed, the selection formula of the number of nodes in the hidden layer is introduced, and the problems existing in the BP algorithm Proposed to increase the momentum factor of the BP algorithm and other improvements, thus accelerating the training speed and thus improve the learning effect of BP network.