期刊文献+

基于深度全卷积神经网络的图像识别研究 被引量:1

Research on Image Recognition Based on Deep Fully Convolutional Neural Network
下载PDF
导出
摘要 将要建立多层卷积网络模型,并使用AlexNet预训练模型,在此基础上进行迁移学习,使用kaggle的猫狗数据集进一步训练,模型最终能高灵活度、高准确率的识别猫狗图像,并且不受图像中猫狗的占比大小影响。该网络模型共有6 000万参数,一共包含8个卷积层,其中某些卷积层带有归一化层和池化层,最后一层是具有两个通道的图像输出,每个通道的值分别代表图像为猫和狗的概率。整个网络模型,弃用全连接层,选用全卷积网络来代替全连接层,大大提高网络的灵活性,解决了输入图像分辨率的限制问题,并且全卷积网络的前向传播更加高效,加快了训练的速度。为了方便分析以及进一步的研究,将可视化一层卷积和二层卷积所得到的卷积核和特征图。 A multi-layer convolutional network model will be established, and the AlexNet pre-training model will be used. On this basis, migration learning will be performed. Kaggle’s cat and dog data set will be used for further training. The model will eventually be able to recognize cat and dog images with high flexibility and high accuracy without being affected by the proportion of cats and dogs in the image. The network model has a total of 60 million parameters and a total of 8 convolutional layers. Some of the convolutional layers have a normalization layer and a pooling layer. The last layer is an image output with two channels, and the value of each channel respectively represent the probability whether the image is a cat or a dog. The fully connected layer is abandoned in the entire network model, and the fully connected layer is replaced by a fully convolutional network, which greatly improves the flexibility of the network, solves the limitation of input image resolution, and the forward propagation of the fully convolutional network is more efficient speeds up training. In order to facilitate analysis and further research, this article will visualize the convolution kernel and feature maps obtained by one-layer convolution and two-layer convolution.
作者 姬壮伟 JI Zhuang-wei(Department of Computer Science,Changzhi University,Changzhi Shanxi,046011)
出处 《山西大同大学学报(自然科学版)》 2022年第2期27-29,74,共4页 Journal of Shanxi Datong University(Natural Science Edition)
关键词 深度学习 全卷积网络 卷积可视化 deep learning fully convolutional network convolutional visualization
  • 相关文献

参考文献7

二级参考文献32

  • 1Alberto Del Bimbo and Pietro Pala.Visual Image Retrieval by Elastic Matching of User Sketches.IEEE Transactions on Pattern Analysis and Machine Intelligence,1997,19(2):121~132.
  • 2Smith J R,Chang S F.Transform feature for texture classification and discrimination in large image database.In:Proc of IEEE Int'l Conf on Image Processing.Austrin,Texas,1994.
  • 3Guodong Guo,Stan Z.Li,Kap Luk Chan.Support Vector Machines Recognition.Image and Vision Computing ,19(2001):631~638.
  • 4Pierre M L Drezet,Robert F Harrison.A new method for sparsity control in support vector classification and regression.Pattern recognition,2001,34:111~125.
  • 5Vapnik V,Lerner A.Pattern Recognition Using Generalized Portrait.Automation and Remote Control.1963,24(6):774~780.
  • 6S.Amari,S Wu.Improving Support Vector Machine Classifiers by Modifying Kernel Functions.Neural Networks 12(199):783~789.
  • 7Victor L Brailovsky,Ofir Barzily,Rabin Shahave.On Global Local and Neighborhood Kernels for Support Vector Machines.Pattern Recognition Letters,20(1999):1183~1190.
  • 8Courant R,Hibert B.Methods of mathematical physics.Vol.1.New York:Widely-interscience,1953.
  • 9王亮申.[D].大连:大连理工大学,2003.
  • 10金连文,徐秉铮.基于多级神经网络结构的手写体汉字识别[J].通信学报,1997,18(5):21-27. 被引量:19

共引文献133

同被引文献3

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部