基于卷积神经网络的图像特征识别研究  

Research on Image Feature Recognition based on Convolution Neural Network

作　　者：杨念聪 任琼[1] 张成喆 周子煜[1] 李倩 邱兰 Yang Congcong;Ren Qiong;Zhang Chengzhe;Zhou Ziyu;Li Qian;Qiu Lan(School of Mathematics and Computer Science,Jianghan University,Wuhan Hubei 430056,China)

机构地区：[1]江汉大学数学与计算机科学学院,湖北武汉430056

出　　处：《信息与电脑》2017年第14期62-64,共3页Information & Computer

摘　　要：在深度学习领域,基于卷积神经网络模型的图像特征识别对人工智能的发展有着重要的意义。这几年来,将卷积神经网络模型运用于图像特征识别已成为研究热点,并且取得可观的成果,对人工智能有着重大的意义。笔者基于深度学习的相关理论,分析并实现了卷积神经网络模型,研究了影响其精度的参数,并在MNIST数据集上进行了实验论证。具体工作如下:分析卷积神经网络的基本原理,设计卷积神经网络模型,通过使用MNIST数据集对模型进行训练,研究参数对卷积神经网络的影响。In the field of deep learning,image feature recognition based on convolution neural network model is of great significance to the development of artificial intelligence.In the past few years,the application of convolution neural network model to image feature recognition has become a hotspot,and the results achieved are significant,which is of great significance to artificial intelligence.Based on the theory of deep learning,the author analyzes and studies the convolution neural network model,studies the parameters that affect its accuracy,and makes an experimental demonstration on the MNIST dataset.Specific work is as follows:analyzing the basic principle of convolution neural network,designing the convolution neural network model,and training the model by using MNIST data set to study the influence of parameters on convolution neural network.

关 键 词：深度学习 图像特征识别 卷积神经网络 过拟合 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]