SONG Zhengdan,JIN Xiaofeng*.Semantic image classification method based on locality-constrained liner sparse coding in multi-scale space[J].Journal of Yanbian University,2018,44(02):134-138.
基于多尺度空间LLC的图像语义分类方法
- Title:
- Semantic image classification method based on locality-constrained liner sparse coding in multi-scale space
- Keywords:
- locality-constrained linear coding(LLC); image semantic; spatial pyramid; probabilistic latent semantic analysis(PLSA); multi-scale space
- 分类号:
- TP391.41
- 文献标志码:
- A
- 摘要:
- 为了提高图像的空间分布和语义信息的有效利用,采用金字塔模型提出一种将多尺度空间、LLC和图像语义分析相融合的图像语义分类方法.首先对图像空间域金字塔划分的各个层次的局部块分别进行线性局部稀疏编码,并对不同层次上的量化编码进行串接生成共生矩; 其次使用概率潜在语义模型对图像进行语义分析以获得最终的图像表示; 最后采用线性多类别SVM对图像进行分类.实验结果表明,本文提出的算法生成的图像特征具有较高的分类性能和效率.
- Abstract:
- Adopting pyramid model, in order to make up for these problems, this paper proposes an image semantic classification method by fusing locality-constrained linear coding(LLC)based on multi-scale space and mage semantic analysis. Firstly, locality-constrained linear coding is adopted to quantify local features of each block by using spatial pyramid approach, in order to produce the co-occurrence matrix by concatenating all local block. Secondly, the probabilistic latent semantic analysis(PLSA)is used to extract the latent semantic information to obtain the final image representation. Finally, the linear support vector machine(SVM)classifier is adopted in to improve the classification performance. Experimental results demonstrate that our approach has high classification performance and time efficiency.
参考文献/References:
[1] Lowe D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004,60(2):91-110.
[2] Lazebnik S, Schmid C, Ponce J. Beyond bags of features: spatial pyramid matching for recognizing natural scene categories[C]//IEEE Computer Society Conference on Computer Vision & Pattern Recognition. New York: IEEE Computer Society, 2006:2169-2178.
[3] Yang J, Yu K, Gong Y, et al. Linear spatial pyramid matching using sparse coding for image classification[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Miami:IEEE Computer Society, 2009:1794-1801.
[4] Leborgne H, Shabou A. Locality-constrained and spatially regularized coding for scene categorization[C]//IEEE Conference on Computer Vision and Pattern Recognition. Rhode Island: IEEE, 2012:3618-3625.
[5] Wang J, Yang J, Yu K, et al. Locality-constrained linear coding for image classification[C]//IEEE Computer Vision and Pattern Recognition. San Francisco: IEEE, 2010:3360-3367.
[6] Li F F, Fergus R, Perona P. Learning generative visual models from few training examples: an incremental bayesian approach tested on 101 object categories[J]. Computer Vision and Image Understanding, 2007,106(1):178.
[7] 焦李成,赵进,杨淑媛,等.稀疏认知学习、计算与识别的研究进展[J].计算机学报,2016,39(4):835-852.
[8] Yu K, Zhang T, Gong Y. Nonlinear learning using local coordinate coding[C]//International Conference on Neural Information Processing Systems. Vancouver: Curran Associates Inc, 2009:2223-2231.
[9] Hofmann T. Unsupervised learning by probabilistic latent semantic analysis[J]. Machine Learning, 2001,42(1/2):177-196.
[10] 李志欣,施智平,李志清,等.融合语义主题的图像自动标注[J].软件学报,2011,22(4):801-812.
[11] 赵仲秋,季海峰,高隽,等.基于稀疏编码多尺度空间潜在语义分析的图像分类[J].计算机学报,2014,37(6):1251-1260.
[12] Yu H F, Hsieh C J, Chang K W, et al. Large linear classification when data cannot fit in memory[J]. Acm Transactions on Knowledge Discovery from Data, 2012,5(4):23.
备注/Memo
*通信作者: 金小峰(1970—),男,教授,研究方向为音视频处理、模式识别、智能计算.