Scene Classification Based on Improved Spatial Partition Model


Article Preview

In order to make full use of the spatial information of images in the classification of natural scene, we use the spatial partition model. But mechanically space division caused the abuse of spatial information. So spatial partition model must be properly improved to make the different categories of images were more diversity, so that the classification performance is improved. In addition, to further improve the performance, we use FAN-SIFT as local image features. Experiments made on 8 scenes image dataset and Caltech101 dataset show that the improved model can obtain better classification performance.



Edited by:

Abdel Hamid Ismail Mourad and József Kázmér Tar




Z. Y. Liu et al., "Scene Classification Based on Improved Spatial Partition Model", Applied Mechanics and Materials, Vol. 527, pp. 339-342, 2014

Online since:

February 2014




* - Corresponding Author

[1] Yincheng Liang, Chanjuan Liu. Scene Image Classification Based on the Feature of Fan-SIFT[J]. Journal of Information & Computational Science 10: 1 (2013) 289–295.

[2] Changxin Gao, Nong Sang, Space pyramid matching model combined local feature and filter feature, Electronic Journal, 9(39), 2011, 2034-(2038).

[3] Chunhui Cui, King Ngi Ngan, Scale and affine invariant fan feature, IEEE Transaction on Image Process, 20(6), 2011, 1627-1640.


[4] J. Yang, Y. G. Jiang, A. G. Hauptmann, and C. W. Ngo, Evaluating bag-of-visual-words representations in scene classification, in ACM MIR, (2007).


[5] S. Lazebnik, C. Schmid, J. Ponce, Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2006, 169-178.


[6] D. G. Lowe, Distinctive image features from scale-invariant key points, International Journal of Computer Vision, 60 (2004), 91-110.


[7] F. -F. Li and P. Perona. A bayesian hierarchical model for learning natural scene categories. InProc. of the2005 IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, pages 524-531, (2005).