《计算机应用研究》|Application Research of Computers

利用卷积神经网络改进迭代深度学习算法的图像识别方法研究

Research of image recognition based on improved iterative depth learning algorithm using convolutional neural network

免费全文下载 (已被下载 次)  
获取PDF全文
作者 孙平安,祁俊,谭秋月
机构 1.武夷学院 a.实验室管理中心;b.数学与计算机学院,福建 武夷山 354300;2.青海大学 计算机技术与应用系,西宁 810016
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)07-064-2223-05
DOI 10.19734/j.issn.1001-3695.2018.01.0071
摘要 针对图像识别算法中图像集上几何曲面的特定分类会导致判别信息丢失的问题,提出一种融合卷积神经网络的改进型迭代深度学习算法(IIDLA)。该算法采用混合卷积网进行底层的平移不变特征学习,以层次化的方式迭代应用卷积神经网络对输入图像集的不同非线性特征进行学习。算法的图库和查询实例中包括了不同视角、背景、面部表情、解析度和照明度的人脸或物体图像集。采用数据集将提出的算法与其他算法进行评估对比,实验结果表明提出的算法在被测数据集上的性能最优。
关键词 深度学习; 卷积神经网络(CNN); 自适应; 图像识别; 层次化迭代
基金项目 国家自然科学基金资助项目(61762086)
福建省教育厅信息化类项目(JAT160524)
福建省教育厅科技A类项目(JA15513)
本文URL http://www.arocmag.com/article/01-2019-07-064.html
英文标题 Research of image recognition based on improved iterative depth learning algorithm using convolutional neural network
作者英文名 Sun Ping'an, Qi Jun, Tan Qiuyue
机构英文名 1.a.Laboratory Management Center,b.College of Mathematics & Computer,Wuyi University,Wuyishan Fujian 354300,China;2.Dept. of Computer Technology & Application,Qinghai University,Xining 810016,China
英文摘要 Aiming at the problem that the specific classification of geometric surfaces on the set of image recognition algorithms could lead to the loss of discriminative information, this paper proposed an improved iterative depth learning algorithm(IIDLA) based on convolution neural network(CNN). The algorithm used the hybrid convolution network to perform the translation invariant feature learning at the bottom, and iteratively used CNN to learn different nonlinear characteristics of the input image set in a hierarchical way. The graphic library and query examples of the algorithm included different perspectives, backgrounds, facial expressions, resolution and illumination, and the image sets of objects or objects. This paper compared the proposed algorithm with other algorithms on a data set. The experimental results show that the proposed algorithm has the best performance on the measured data set.
英文关键词 depth learning; convolutional neural network(CNN); adaptive; image recognition; layering iterative
参考文献 查看稿件参考文献
 
收稿日期 2018/1/25
修回日期 2018/4/2
页码 2223-2227
中图分类号 TP391.41
文献标志码 A