受损图片分类的双并行交叉降噪卷积神经网络  

Double parallel cross denoising convolution neural network for damaged image classification

作　　者：王达 周雪峰 徐智浩 蔡奕松 陈洲 WANG Da;ZHOU Xuefeng;XU Zhihao;CAI Yisong;CHEN Zhou(School of Information Science and Engineering,Shenyang University of Technology,Shenyang 110870,China;Guangdong Key Laboratory of Modern Control Technology,Guangdong Institute of Intelligent Manufacturing,Guangzhou 510070,China)

机构地区：[1]沈阳工业大学信息科学与工程学院,沈阳110870 [2]广东省智能制造研究所广东省现代控制技术重点实验室,广州510070

出　　处：《计算机工程与应用》2018年第18期147-153,共7页Computer Engineering and Applications

基　　金：国家自然科学基金(No.51405091);广东省科学院优秀青年科技人才基金(No.rcjj201504);广州市珠江科技新星专项(No.201610010054)

摘　　要：应对大量受干扰图像的分类问题,提出了一种双并行交叉降噪卷积模型,该模型由两部分并行交叉网络结构组成,分别对应改进的自编码方式和并行交叉卷积神经网络,同时在该模型训练的过程中,使用批量正则化和改进激活函数的方法。经实验验证,与同类模型相比,该模型首先具有降噪能力强、鲁棒性好、泛化能力强和准确率高的特点,其次避免过拟合,加快收敛速度。在图片不同程度受损的情况下,它也可较好地完成图像目标识别分类任务。In order to deal with the classification problem of a large number of disturbed images,a double parallel crossover noise reduction convolution model is proposed.The model consists of two parts of parallel cross-network structure,corresponding to improved self-coding and parallel cross-convolutional neural network.At the same time,the model uses the method of batch normalization and improvement of traditional activation function in the model training.It is proved by experiments that the model has the characteristics of strong noise reduction,good robustness,strong generalization ability and high accuracy in comparison with the same model.Secondly,it avoids overfitting and speeds up convergence.In the case of different degrees of picture damage,it can also accomplish the task of image target recognition classification well.

关 键 词：图像分类 自编码器 卷积神经网络 双并行交叉 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP389.1[自动化与计算机技术—控制科学与工程]