Discriminate Dictionary Training Algorithm  Based on Adaptive Steepest Descent

doi:10.3969/j.issn.1000-3428.2013.05.050

Computer Engineering ›› 2013, Vol. 39 ›› Issue (5): 225-229. doi: 10.3969/j.issn.1000-3428.2013.05.050

• Networks and Communications • Previous Articles Next Articles

Discriminate Dictionary Training Algorithm Based on Adaptive Steepest Descent

XU Jian ^1,2, CHANG Zhi-guo ³, ZHAO Xiao-qiang ¹, MA Xiang³

(1. School of Communication and Information Engineering, Xi’an University of Posts & Telecommunications, Xi’an 710121, China; 2. School of Electronic and Information Engineering, Xi’an Jiaotong University, Xi’an 710049, China; 3. School of Information Engineering, Chang’an University, Xi’an 710064, China)

Received:2012-05-24 Online:2013-05-15 Published:2013-05-14

基于自适应梯度最速下降的分类字典训练算法

徐健^1,2，常志国³，赵小强¹，马祥³

(1. 西安邮电学院通信与信息工程学院，西安 710121；2. 西安交通大学电子与信息工程学院，西安 710049； 3. 长安大学信息工程学院，西安 710064)

作者简介:徐健(1981－)，女，讲师、硕士，主研方向：模式识别，数字图像处理；常志国，讲师、博士；赵小强，副教授、硕士；马祥，讲师、博士
基金资助:
国家自然科学基金资助项目(61101215)；中央高校基本科研业务费专项基金资助项目(CHD2012JC012)；陕西省社会发展科技攻关基金资助项目(2010K11-02-11)；陕西省教育厅科学研究计划基金资助项目(11JK0994)

Abstract

Abstract: To improve the correct rate of image classification using sparse representation dictionaries, a discriminate dictionary training algorithm based on adaptive steepest descent is proposed. This algorithm uses alternating gradient descent for bi-objective optimization model of discriminate dictionary training. To guarantee the convergence of sparse representation residual and the incoherence of atoms from dictionaries corresponding to different classes and increasing convergence rate, adaptive step size is adopted and the method of calculating the step size is proved. The steps of the algorithm are as following: Fix dictionaries and use current dictionaries and data samples to calculate the steepest descent directions and adaptive step sizes of the sparse representation coefficients. Small coefficients are set to zero according to sparseness constraints. Fix sparse representation coefficients, and use sparse representation coefficients and data samples to calculate the steepest descent direction and adaptive step size of dictionaries. Experimental results of hand-writing recognition show that the correct rate is 96.51%.

Key words: image classification, dictionary training, sparse representation, hand-written character recognition, bi-variable optimization, adaptive step size

摘要： 为提高稀疏表示字典用于图像分类时的正确率，提出一种基于自适应梯度最速下降的分类字典训练算法。该算法采用交替梯度下降法解决分类字典训练的双变量优化模型。为提高收敛速度，并保证稀疏表示残差和不同类别对应字典原子间的不相关性同时收敛，采用自适应步长，推导证明自适应步长的计算方法。通过固定字典，运用当前字典和训练样本计算出稀疏表示系数的下降方向和自适应步长，按照稀疏度约束将小系数置零，固定稀疏表示系数，利用稀疏表示系数和样本找到字典的下降方向和自适应步长。实验结果表明，该算法在手写字符识别中正确率能达到96.51%。

关键词: 图像分类, 字典训练, 稀疏表示, 手写字符识别, 双变量优化, 自适应步长

CLC Number:

TP301.6

XU Jian, CHANG Zhi-Guo, DIAO Xiao-Jiang, MA Xiang. Discriminate Dictionary Training Algorithm Based on Adaptive Steepest Descent[J]. Computer Engineering, 2013, 39(5): 225-229.

徐健, 常志国, 赵小强, 马祥. 基于自适应梯度最速下降的分类字典训练算法[J]. 计算机工程, 2013, 39(5): 225-229.

/ / Recommend / Download Citations

URL: http://www.ecice06.com/EN/10.3969/j.issn.1000-3428.2013.05.050

http://www.ecice06.com/EN/Y2013/V39/I5/225

References

[1] Cherian A, Sra S, Papanikolopoulos N. Denoising Sparse Noise via Online Dictionary Learning[C]//Proc. of IEEE International Conference on Acoustics, Speech, and Signal Processing. [S. l.]: IEEE Press, 2011.
[2] Elad M, Aharon M. Image Denoising via Sparse and Redundant Representations over Learned Dictionaries[J]. IEEE Transactions on Image Processing, 2006, 15(12): 3736-3745.
[3] Xu Zongben, Sun Jian. Image in Painting by Patch Propagation Using Patch Sparsity[J]. IEEE Transactions on Image Processing, 2010, 19(5): 1153-1165.
[4] Zhang Haichao, Zhang Yanning, Huang T S. Efficient Sparse Representation Based Image Super Resolution via Dual Dictionary Learning[C]//Proc. of IEEE International Conference on Multimedia and Expo. [S. l.]: IEEE Press,

2011.
[5] Mairal J, Bach F, Ponce J, et al. Discriminative Learned Dictionaries for Local Image Analysis[C]//Proc. of IEEE Conference on Computer Vision and Pattern Recognition. [S. l.]: IEEE Press, 2008.
[6] Wright J, Ma Yi, Mairal J, et al. Sparse Representation for Computer Vision and Pattern Recognition[J]. Proceedings of the IEEE, 2010, 98(6): 1031-1044.
[7] Mairal J, Bach F, Ponce J. Task-driven Dictionary Learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(4): 791-804.
[8] Wright J, Yang A Y, Ganesh A, et al. Robust Face Re- cognition via Sparse Representation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(2): 210-227.
[9] Gangeh M J, Ali G, Kamel MS. Dictionary Learning in Texture Classification[C]//Proc. of International Conference on Image Analysis and Recognition. Burnaby, Canada: [s. n.], 2011.
[10] Mairal J, Bach F, Ponce J, et al. Supervised Dictionary Learning[C]//Proc. of Annual Conference on Neural Information Processing Systems. Vancouver, Canada: [s. n.], 2008.
[11] Ramirez I, Sprechmann P, Sapiro G. Classification and Clustering via Dictionary Learning with Structured Incoherence and Shared Features[C]//Proc. of IEEE Conference on Computer Vision and Pattern Recognition. [S. l.]: IEEE Press, 2010.
[12] Yang Jianchao, Yu Kai, Gong Yihong, et al. Linear Spatial Pyramid Matching Using Sparse Coding for Image Classification[C]//Proc. of IEEE Conference on Computer Vision and Pattern Recognition. [S. l.]: IEEE Press, 2009.
[13] Wang Haoran, Yuan Chunfeng, Hu Weiming, et al. Supervised Class-specific Dictionary Learning for Sparse Modeling in Action Recognition[J]. Pattern Recognition, 2012, 45(11): 3902-3911.
[14] 徐健, 常志国. 基于聚类的自适应图像稀疏表示算法及其应用[J]. 光子学报, 2011, 40(2): 316-320.
[15] Nagesh P, Gowda R, Li B. Fast GPU Implementation of Large Scale Dictionary and Sparse Representation Based Vision Problems[C]//Proc. of IEEE International Conference on Acoustics, Speech, and Signal Processing. [S. l.]: IEEE Press, 2010.
编辑顾逸斐

Please choose a citation manager

Content to export

Discriminate Dictionary Training Algorithm Based on Adaptive Steepest Descent

基于自适应梯度最速下降的分类字典训练算法

PDF

Knowledge

Cited

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics

Comments

[1]	Hong XIE, Wengang JIANG. RRA-InceptionV3 Combined Robust Sparse Representation Method for Expression Recognition [J]. Computer Engineering, 2023, 49(7): 196-203.
[2]	Ping CAO, Huaizhi YANG, Yijun BO, Jia YOU, Chunjie ZHANG, Danyong LI. Low-quality Crack Image Classification with Multi-Knowledge Distillation [J]. Computer Engineering, 2023, 49(7): 204-213.
[3]	HAN Yanling, SHEN Siyang, XU Lijun, WANG Jing, ZHANG Yun, ZHOU Ruyan. GPU Parallel Method for Deep Learning Image Classification [J]. Computer Engineering, 2023, 49(1): 191-200.
[4]	PENG Hu, LI Yuanhan, DENG Changshou, WU Zhijian. Multi-Strategy Reconciled Cuckoo Search Algorithm [J]. Computer Engineering, 2022, 48(8): 85-97.
[5]	YUAN Danfei, CHEN Cifa, DONG Fangmin. Research on Residual Network of Image Recognition Based on Multiscale Split [J]. Computer Engineering, 2022, 48(5): 258-262,271.
[6]	CHEN Shan, SUN Rencheng, SHAO Fengjing, SUI Yi. Research and Improvement of Dynamic Routing Based on Capsule Network [J]. Computer Engineering, 2022, 48(5): 208-214.
[7]	HU Zongcheng, ZHOU Yatong, SHI Baojun, HE Hao. Static Gesture Recognition Algorithm Based on Attention Mechanism and Feature Fusion [J]. Computer Engineering, 2022, 48(4): 240-246.
[8]	TAN Run, YE Wujian, LIU Yijun. Fine-Grained Image Classification Combining Dual Semantic Data Augmentation and Target Location [J]. Computer Engineering, 2022, 48(2): 237-242,249.
[9]	LIU Pengfei, ZHU Jianchen, WAN Liangyi, JIANG Bo. Research on Hyperspectral Remote Sensing Image Classification Using Low-Power Heterogeneous Computing Architecture [J]. Computer Engineering, 2022, 48(12): 9-15,23.
[10]	LI Wei, HUANG Heming, WU Fengying, ZHANG Huiyun. Adaptive CNN-based Image Classification Algorithm Based on Deep Fusion of Multi-Feature [J]. Computer Engineering, 2021, 47(9): 235-239,251.
[11]	TAO Yang, BAO Linglang, HU Hao. Dimensionality Reduction Method Combining Representation Learning and Embedded Subspace Learning [J]. Computer Engineering, 2021, 47(6): 83-87,97.
[12]	TAO Yang, BAO Linglang, HU Hao. Structure-Constrained Symmetric Low-Rank Representation Algorithm for Subspace Clustering [J]. Computer Engineering, 2021, 47(4): 56-61,67.
[13]	LAI Yanling, SHI Junfeng, CHEN Jixin, BAI Hanli, TANG Xiaolan, DENG Biying, ZHENG Desheng. Adversarial Example Defense Model Based on U-Net [J]. Computer Engineering, 2021, 47(12): 163-170.
[14]	CHEN Xinhua, QIAN Xuezhong, SONG Wei. Image Classification Algorithm Based on Lightweight Feature Fusion Convolutional Network [J]. Computer Engineering, 2021, 47(11): 268-275.
[15]	SUN Dengdi, LING Yuan, DING Zhuanlian, LUO Bin. Multi-Layer Network Community Detection Based on Sparse Subspace Clustering [J]. Computer Engineering, 2021, 47(10): 52-60.

模态框（Modal）标题

Please choose a citation manager

Content to export

Discriminate Dictionary Training Algorithm Based on Adaptive Steepest Descent

基于自适应梯度最速下降的分类字典训练算法

PDF

Knowledge

Cited

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics

Comments