Accessible area detection based on fusion of multi-channel texture and spatial information

被引：0

作者：

Cheng, Xintian ^{[1
]}

Tang, Zhenmin ^{[1
]}

Huang, Pu ^{[1
]}

机构：

[1] Nanjing University of Science and Technology

来源：

Journal of Computational Information Systems | 2014年 / 10卷 / 04期

关键词：

Environment understand; Multi-channel; Texture; Unstructured;

D O I：

10.12733/jcis9446

中图分类号：

学科分类号：

摘要：

It is a precondition for intelligent mobile robots working in unstructured outdoor environment. This paper proposes a new method of feature extraction for accessible area detection. In this paper, first we extracte sub-block images' multi-channel texture features based on discrete cosine transform from three channel of HSV color space, and then obtain their color covariance features, finally fuse the spatial information with them. In this way, we can obtain the features that contain both local information and global information. At last we classify them by the K-nearest Neighbor classifier on the experimental test dataset. The experimental results show that, this method has better classification ability and possess good robustness in dealing with the light, shadows and other noises comparing with other related methods. © 2014 Binary Information Press.

引用

页码：1589 / 1595

页数：6

共 10 条

[1]

Manduchi R., Obstacle detection and terrain classification for autonomous off-road navigation, Autonomous Robots, 18, 1, pp. 81-102, (2005)

[2]

Castano R., Manduchi R., Fox J., Classification experiments on real-world texture, Workshop on Empirical Evaluation in Computer Vision, (2001)

[3]

Shi X.J., Manduchi R., On the Bayes fusion of visual features, Image And Vision Computing, 25, pp. 1748-1758, (2007)

[4]

Manduchi R., Bayesian fusion of color and texture segmentations, Computer Vision, 1999.The Proceedings of the Seventh IEEE International Conference on, 2, pp. 956-962, (1999)

[5]

Kim S.C., Kang T.J., Texture classification and segmentation using wavelet packet frame and Gaussian mixture model, Pattern Recognition, 40, pp. 1207-1221, (2007)

[6]

Yu C., Wang X., Human action classification based on combinational features from video, Journal of Computational Information Systems, 8, 12, pp. 5245-5254, (2012)

[7]

Li J., Gray R., Olshen R., Multi-resolution image classification by hierarchical modeling with twodimensional hidden Markov models, IEEE Transactions on Information Theory, 46, 5, pp. 1826-1814, (2000)

[8]

Zhuang Y., Chen D., Wang W., Status and development of natural scene understanding for vision-based outdoor moblie robot, Acta Automatica Sinica, 36, 1, pp. 774-784, (2010)

[9]

Hu B., He K.Z., Applications of computer vision to outdoor mobile robot, Acta Automatica Sinica, 32, 5, pp. 774-784, (2006)

[10]

Zhu S., Yang J., Using sub-block energy for image classification based on color correlogram, Journal of Computational Information Systems, 7, 16, pp. 5822-5828, (2011)

← 1 →