Journal of Software, Vol 6, No 7 (2011), 1313-1320, Jul 2011
doi:10.4304/jsw.6.7.1313-1320

Non-Parameter Local Transformation of Low Frequency Wavelet Coefficients Applied in Aerial Texture Retrieval

Xubing Zhang, Cai Bo, Xinrong Hu, Min Li

Abstract


Low frequency wavelet coefficients include much important vision information of the image, and are useful to image recognition and understanding. While at present, the applications of the low frequency wavelet coefficients are limited in the researches of image analysis.  In this paper, the authors extracted the BFV (Binary Feature Vector, BFV) and TFV (Ternary Feature Vector, TFV) of low frequency wavelet coefficients based on non-parameter local transformation, which adopts the comparison results of the coefficient amplitudes in the neighborhood with the center coefficient to extract the feature rapidly. The TFV describes the texture more accuracy than BFV by adopting the two adaptive thresholds, and the by adjusting the parameter f TFV can adapt itself to the different texture data. The authors apply the BFV and TFV in aerial textures retrieval. In the experiments, our method is compared with the GLCM, Markov and Fractal algorithms, and the results prove that our methods behave well in the retrieval rate, especially the rapid processing speed.


Keywords


Non-parameter local transformation, low frequency, wavelet coefficient, texture retrieval

References


[1] Fei. Yuanyuan, Sun Jinguang, and Tao Zhiyong, “Textre image retirval based on wavelet decomposition and gray level co-occurrence matrix,” Modern Computer, no. 269, pp. 58–59, October 2007.

[2] Ming-xin Cai, “License-plate recognition using wavelet transform and neural network,” Chaoyan university of science and technology, Master thesis, June 2004.

[3] He Jiazhong, Du Minghui, “Principal component analysis combined with wavelet low-frequency band,” Journal of South China University of Technology (Natural Science Edition), vol. 35, no. 1, pp. 44–48, January 2007.

[4] Li Yan, Peng Jiaxiong, “Wavelet transform based multiscale hurst parameter texture features and its application,” Acta Electronica Sinica, vol. 30, no. 7, pp. 1041–1043, July 2007.

[5] Guo Zhiqiang, “Wavelet transform image fusion based on regional features,” Journal of Wuhan University of Technology, vol. 27, no. 2, pp. 65–71, February 2005.

[6] Wang Ronghui, Liu Gang, “Detecting of targets in natural texture background based on wavelet energy,” Journal of Changchun University of Science and Technology, vol. 28, no. 3, pp. 70–72, September, 2005.

[7] Wu Yan, “The study on texture extration of SAR image in wavelet domain,” Xidian University, Master thesis, June, 2007.

[8] Bow, S. T., Pattern Recognition and Image Preprocessing. New York, marcel dekker, 1992.

[9] Zhong, D. and I. Defee, “Study of image retrieval based on feature vectors in compressed domain,” Signal Processing Symposium, NORSIG 2006, pp. 202–205, 2006.

[10] Au, K. M., L. N. F and S. W. C. “Unified feature analysis in different compressed domains,” the 4th International Conference on Information, Communications and Signal Processing, 2003.

[11] Zabih, R. and W. Johns, “Non-parametric local transforms for computing visual correspondence,” Lecture Notes In Computer Science, vol. 801, pp. 151–158, 1994.
http://dx.doi.org/10.1007/BFb0028345

[12] Haralick R M, Shamugam K, “Textural features for image classification”, IEEE Trans. on SMC, vol. 6, pp. 610-612, March 1973.

[13] Zheng Zhaobao, Markov Random Field Method of Image Analyses. Wuhan, Wuhan technical university of surveying and mapping, 2000.

[14] Huang Guilan, Zheng Zhaobao, “The analysis and experiment on aerial image texture classification using three kinds of methods”, WTUSM Bulletin of Science and Technology, no. 3, pp. 12-15, 1996.


Full Text: PDF


Journal of Software (JSW, ISSN 1796-217X)

Copyright @ 2006-2013 by ACADEMY PUBLISHER – All rights reserved.