基于BP神经网络的计算机图像智能识别方法  

Intelligent Recognition Method of Computer Image Based on BP Neural Network

作　　者：张秀良 曲红 ZHANG Xiuliang;QU Hong(Jinling Institute of Technology,Nanjing Jiangsu 211169,China;Anhui Xinhua University,Hefei Anhui 230088,China)

机构地区：[1]金陵科技学院,江苏南京211169 [2]安徽新华学院,安徽合肥230088

出　　处：《信息与电脑》2021年第15期158-161,共4页Information & Computer

摘　　要：为解决传统图像识别方法在实际应用中存在收敛速度低、无法快速降低识别误差的问题,开展基于BP神经网络的计算机图像智能识别方法设计研究。通过计算机图像数据获取与处理、基于BP神经网络构建图像识别模型以及计算机图像识别模型训练,提出一种全新的识别方法。通过对比不同图像识别方法在相同实验环境中的应用效果可知,新的识别方法收敛速度更快,可在最短的时间内降低识别误差,提高对计算机图像的识别效率。In order to solve the problem that the traditional image recognition methods have low convergence speed and can not quickly reduce the recognition error in practical application, the design and research of computer image intelligent recognition method based on BP neural network is carried out. Through the acquisition and processing of computer image data, the construction of image recognition model based on BP neural network and the training of computer image recognition model, a new recognition method is proposed. By comparing the effects of the two image recognition methods applied to the same experimental environment, it is concluded that the new recognition method converges faster, reduces the recognition error in the shortest time and improves the recognition efficiency of computer images.

关 键 词：BP神经网络 计算机 图像识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]