A Novel Method for Efficient Indoor–Outdoor Image Classification
- 576 Downloads
Most traditional indoor–outdoor scene classification approaches utilize the simple statistics of the low-level features, such as colors, edges, and textures. However, the existence of colors similar to sky or grass often yields the false positives. To cope with this deficiency, we focus on the orientation of low-level features in this paper. First, the image is partitioned into five block regions, whose features are differently weighted in the following classification stage according to the block positions. The edge and color orientation histogram (ECOH) descriptors are defined to represent each block efficiently. Finally, all ECOH values are concatenated to generate the feature vector and fed into the SVM classifier for the indoor–outdoor scene classification. To justify the efficiency and robustness of the proposed method, the evaluation is conducted over 1200 images.
KeywordsIndoor–outdoor classification Low-level features Edge and color orientation histogram SVM classifier
This research was supported by the MKE(The Ministry of Knowledge Economy), Korea, under the ITRC(Information Technology Research Center) support program supervised by the NIPA(National IT Industry Promotion Agency NIPA-2010-(C1090-0902-0017)).
- 1.Zhang, L., Mingjing, L., Hong-Jing, Z. (2002). Boosting image orientation detection with indoor vs. outdoor classification. IEEE Workshop on Applications of Computer Vision (WACV 2002), Orlando, FL, pp. 95–99.Google Scholar
- 3.Szummer, M., Picard, R., W. (1998). Indoor–outdoor image classification. IEEE Workshop on Content-Based Access of Image and Video Database, Bombay, India, pp. 42–51.Google Scholar
- 5.Serrano, N., Savakis, A., Luo, J. (2002). A computationally efficient approach to indoor/outdoor scene classification. International Conference on Pattern Recognition (ICPR), QC, Canada, 4, pp. 146–149.Google Scholar
- 6.Gupta, L., et al. (2007). Indoor versus outdoor scene classification using probabilistic neural network. EURASIP Journal on Advances in Signal Processing, 1, 123–133.Google Scholar
- 8.Luo, J., Savakis, A. (2001). Indoor vs. outdoor classification of consumer photographs using low-level and semantic features. International Conference on Image Processing (ICIP), Thessaloniki, Greece, vol. 2, pp. 745–748.Google Scholar
- 9.Boutell, M., Luo, J. (2004). Bayesian fusion of camera metadata cues in semantic scene classification. Computer Vision and Pattern Recognition (CVPR), Washington, D.C., vol. 2, pp. 623–630.Google Scholar
- 10.Boutell, M., Choudhury, A., Luo, J., Brown, C., M. (2006). Using semantic features for scene classification: how good do they need to be ?. International Conference on Multimedia and Expo (ICME), Toronto, Canada, pp. 785–788.Google Scholar
- 12.Spyrou, E., Mylonas, P., Avrithis, Y. (2008). Using region semantics and visual context for scene classification. International Conference on Image Processing (ICIP), San Diego, CA, pp. 53–56.Google Scholar
- 13.Kim, W., & Kim, C. (2009). An efficient indoor-outdoor scene classification method. Journal of the Institute of Electronics Engineers of Korea, 46-SP(5), 48–55.Google Scholar
- 16.Dalal, N., Triggs, B. (2005). Histograms of oriented gradients for human detection. Computer Vision and Pattern Recognition (CVPR), San Diego, CA, vol. 1, pp. 886–893.Google Scholar
- 17.Gonzalez, R. C., & Woods, R. E. (2002). Digital image processing. Upper Saddle River: Prentice-Hall.Google Scholar