非负局部约束线性编码图像分类算法

刘培娜; 刘国军; 郭茂祖; 刘扬; 李盼

doi:10.16383/j.aas.2015.c140753

非负局部约束线性编码图像分类算法

doi: 10.16383/j.aas.2015.c140753

1.
哈尔滨工业大学计算机科学与技术学院哈尔滨 150001

基金项目:

国家自然科学基金 (61171185, 61271346), 黑龙江省青年科学基金 (QC2014C071)资助

详细信息

作者简介:
刘培娜哈尔滨工业大学计算机科学与技术学院硕士研究生. 2013 年获得内蒙古大学计算机学院学士学位. 主要研究方向为计算机视觉与机器学习.E-mail: liupeina@hit.edu.cn

计量
- 文章访问数: 2373
- HTML全文浏览量: 101
- PDF下载量: 1791
- 被引次数: 33
出版历程
- 收稿日期: 2014-10-28
- 修回日期: 2015-03-03
- 刊出日期: 2015-07-20

Image Classification Based on Non-negative Locality-constrained Linear Coding

1.
School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001

Funds:

Supported by National Natural Science Foundation of China (61171185, 61271346) and Heilongjiang Province Science Foundation for Youths (QC2014C071)

摘要

摘要: 基于特征提取的图像分类算法的核心问题是如何对特征进行有效编码. 局部约束线性编码(Locality-constrained linear coding, LLC) 因其良好的特征重构性与局部平滑稀疏性, 已取得了很好的分类性能. 然而, LLC编码的分类性能对编码过程中的近邻数k的大小比较敏感, 随着k的增大, 编码中的某些负值元素与正值元素的差值绝对值也可能增大, 这使得LLC越来越不稳定. 本文通过在LLC优化模型的目标方程中引入非负约束, 提出了一种新型编码方式, 称为非负局部约束线性编码(Non-negative locality-constrained linear coding, NNLLC). 该模型一般采取迭代优化算法进行求解, 但其计算复杂度较大. 因此, 本文提出两种近似非负编码算法, 其编码速度与LLC一样快速. 实验结果表明, 在多个广泛使用的图像数据集上, 相比于LLC, NNLLC编码方式不仅在分类精确率上提高了近1%～4%, 而且对k的选取具有更强的鲁棒性.
- 局部约束线性编码 /
- 非负约束 /
- 空间金字塔匹配 /
- 图像分类
Abstract: The most important issue of image classification algorithm based on feature extraction is how to efficiently encode features. Locality-constrained linear coding (LLC) has achieved the state-of-the-art performance on several benchmarks, due to its underlying properties of better construction and local smooth sparsity. However, the performance of LLC on image classification is sensitive to the number of neighbors, i.e., the value of k. With the increase of k, the absolute difference of some negative and positive elements may likely become larger and larger. This will make LLC more unstable. In this paper, a new coding scheme called non-negative locality-constrained linear coding (NNLLC) is proposed. It adds an extra non-negative constraint to the objective function of LLC. Generally, this new model can be solved by iterative optimization methods, however, such solutions are quite impractical due to high computational cost. Therefore, two fast approximation algorithms are proposed; more importantly, they and LLC have a similar computational complexity. To compare with LLC, the experiment results on several widely used image datasets demonstrate that NNLLC not only can improve the classification accuracy by nearly 1%～4%, but also is more robust on the selection of k.
- Locality-constrained linear coding (LLC) /
- non-negative constraint /
- spatial pyramid matching (SPM) /
- image classification

HTML全文

参考文献(20)

[1]	Sivic J, Zisserman A. Video google: a text retrieval approach to object matching in videos. In: Proceedings of the 9th IEEE International Conference. Nice, France: IEEE, 2003. 1470-1477
[2]	Csurka G, Dance C R, Fan L X, Willamowski J, Bray C. Visual categorization with bags of keypoints. In: Proceedings of the 2004 ECCV International Workshop on Statistical Learning in Computer Vision. Grenoble, France: ECCV, 2004. 1-22
[3]	Zhao Zhong-Qiu, Ji Hai-Feng, Gao Jun, Hu Dong-Hui, Wu Xin-Dong. Sparse coding based multi-scale spatial latent semantic analysis for image classification. Chinese Journal of Computers, 2014, 37(6): 1251-1260(赵仲秋, 季海峰, 高隽, 胡东辉, 吴信东. 基于稀疏编码多尺度空间潜在语义分析的图像分类. 计算机学报, 2014, 37(6): 1251-1260)
[4]	Li Qian-Qian. Image Classification Research of Improved Non-negative Sparse Coding [Master dissertation], Nanjing University of Science and Technology, China, 2014.(李钱钱. 基于改进的非负稀疏编码图像分类研究[硕士学位论文], 南京理工大学, 中国, 2014.)
[5]	Lazebnik S, Schmid C, Ponce J. Beyond bags of features: spatial pyramid matching for recognizing natural scene categories. In: Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, USA: IEEE, 2006. 2169-2178
[6]	Yang J C, Yu K, Gong Y H, Huang T. Linear spatial pyramid matching using sparse coding for image classification. In: Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami, FL: IEEE, 2009. 1794-1801
[7]	Wang J J, Yang J C, Yu K, Lv F J, Huang T S, Gong Y H. Locality-constrained linear coding for image classification. In: Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition. San Francisco, USA: IEEE, 2010. 3360-3367
[8]	Yang J C, Wang J P, Huang T. Learning the sparse representation for classification. In: Proceedings of the 2011 IEEE International Conference on Multimedia and Expo (ICME). Barcelona, Spanish: IEEE, 2011. 1-6
[9]	Yu K, Zhang T, Gong Y H. Nonlinear learning using local coordinate coding. In: Proceedings of the 2009 Advances in Neural Information Processing Systems. Vancouver, Canada: NIPS, 2009. 2223-2231
[10]	Hoyer P O. Non-negative sparse coding. In: Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing. Falmouth, USA: IEEE, 2002. 557-565
[11]	Zhang C J, Liu J, Tian Q, Xu C S, Lu H Q, Ma S D. Image classification by non-negative sparse coding, low-rank and sparse decomposition. In: Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition. Colorado Springs, USA: IEEE, 2011. 1673-1680
[12]	Lin T H, Kung H T. Stable and efficient representation learning with non-negativity constraints. In: Proceedings of the 31st International Conference on Machine Learning. Beijing, China: JMLR W&CP, 2014. 1323-1331
[13]	Roweis S T, Saul L K. Nonlinear dimensionality reduction by locally linear embedding. Science, 2000, 290(5500): 2323-2326
[14]	Goldfarb D, Idnani A. A numerically stable dual method for solving strictly convex quadratic programs. Mathematical Programming, 1983, 27(1): 1-33
[15]	Fan R E, Chang K W, Hsieh C J, Wang X R, Lin C J. Liblinear: a library for large linear classification. The Journal of Machine Learning Research, 2008, 9: 1871-1874
[16]	Li F F, Fergus R, Perona P. Learning generative visual models from few training examples: an incremental Bayesian approach tested on 101 object categories. Computer Vision and Image Understanding, 2007, 106(1): 59-70
[17]	Zhang H, Berg A C, Maire M, Malik J. SVM-KNN: discriminative nearest neighbor classification for visual category recognition. In: Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, USA: IEEE, 2006. 2126-2136
[18]	Boiman O, Shechtman E, Irani M. In defense of nearest-neighbor based image classification. In: Proceedings of the 2008 IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, Alaska: IEEE, 2008. 1-8
[19]	Balasubramanian K, Yu K, Lebanon G. Smooth sparse coding via marginal regression for learning sparse representations. In: Proceedings of the 30th International Conference on Machine Learning. Edinburgh, Scotland: ICML, 2012. 1326-1334
[20]	van Gemert J C, Geusebroek J M, Veenman C J, Smeulders A W M. Kernel codebooks for scene categorization. In: Proceedings of the 10th European Conference on Computer Vision. Marseille, France: ECCV, 2008. 696-709

施引文献

期刊类型引用(16)

1.	巨志勇，张泽晨. 一种局部线性编码的深度学习果蔬分类算法. 小型微型计算机系统. 2020(04): 741-745 . 百度学术
2.	杨建军，常丽萍，李胜，朱霆威，何熊熊. 基于新型特征和特征袋模型的内窥镜大肠病变辅助诊断. 中国生物医学工程学报. 2020(04): 404-412 . 百度学术
3.	万源，张景会，陈治平，孟晓静. 基于弹性网和直方图相交的非负局部稀疏编码. 计算机应用. 2019(03): 706-711 . 百度学术
4.	王仁芳，刘云鹏，孙德超，张亮. 基于多核融合与局部约束的协同表示目标跟踪. 光电子·激光. 2019(01): 70-78 . 百度学术
5.	万源，张景会，吴克风，孟晓静. 基于多层非负局部Laplacian稀疏编码的图像分类. 计算机应用. 2018(09): 2489-2494+2499 . 百度学术
6.	张睿萍，马宗梅. 基于Hadoop平台的大数据图像分类机制. 吉林大学学报(理学版). 2018(05): 1206-1212 . 百度学术
7.	王倩，张新明，蔡强，祝晓斌. 基于改进VLAD算法的图像分类. 计算机应用研究. 2018(10): 3151-3154 . 百度学术
8.	张勇，张阳阳，程洪，张艳霞. 基于非负弹性网稀疏编码算法的图像分类方法. 计算机工程. 2017(07): 239-243+249 . 百度学术
9.	杨赛，赵春霞，刘凡，陈峰. 一种基于多种特征融合的人脸识别算法. 计算机辅助设计与图形学学报. 2017(09): 1667-1672 . 百度学术
10.	刘盛清，孙季丰，余家林，宋治国. 基于多尺度特征融合Hessian稀疏编码的图像分类算法. 计算机应用. 2017(12): 3517-3522 . 百度学术
11.	袁静，章毓晋. 融合梯度差信息的稀疏去噪自编码网络在异常行为检测中的应用. 自动化学报. 2017(04): 604-610 . 本站查看
12.	万源，史莹，吴克风，陈晓丽. 融合局部性和非负性的Laplacian稀疏编码的图像分类. 模式识别与人工智能. 2017(06): 481-488 . 百度学术
13.	万源，史莹，陈晓丽. 非负局部Laplacian稀疏编码和上下文信息的图像分类. 中国图象图形学报. 2017(06): 731-740 . 百度学术
14.	许允喜，陈方. 特征联合和旋转不变空间分割联合的局部图像描述符. 自动化学报. 2016(04): 617-630 . 本站查看
15.	杨赛，赵春霞，徐威. 一种基于词袋模型的新的显著性目标检测方法. 自动化学报. 2016(08): 1259-1273 . 本站查看
16.	张运超，陈靖，王涌天. 一种融合重力信息的快速海量图像检索方法. 自动化学报. 2016(10): 1501-1511 . 本站查看