A Natural Image Compression Approach Based on ICA and Visual Saliency Detection

Abstract:

In this paper, a natural image compression method is proposed based on independent component analysis (ICA) and visual saliency detection. The proposed compression method learns basis functions trained from data using ICA to transform the image at first; and then sets percentage of the zero coefficient number in the total transforming coefficients. After that, transforming coefficients are sparser which indicates further improving of compression ratio. Next, the compression method performance is compared with the discrete cosine transform (DCT). Evaluation through both the usual PSNR and Structural Similarity Index (SSIM) measurements showed that proposed compression method is more robust to DCT. And finally, we proposed a visual saliency detection method to detect automatically the important region of image which is not or low compressed while the other regions are highly compressed. Experiment shows that the method can guarantee the quality of important region effectively.

• Periodical: Applied Mechanics and Materials (Volumes 128-129)
• Pages: 457-460
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.128-129.457
• Online since: October 2011
• Authors: Li Juan Duan, Chun Xia Ke, Chun Peng Wu, Zhen Yang, Jun Miao
• Keywords: Compression, DCT, ICA, Visual Saliency Detection

References

[1] Hyvarinen A, Cristescu R and Oja E, in: Independent Component Analysis [M]. New York: Wiley Interscience Publishing, (2001).

[2] Ferreira A. J and Figueiredo M.A. T, in: On the use of independent component analysis for image compression, Signal Processing: Image Communication, 2006, 21(5): 378-389.

DOI: 10.1016/j.image.2006.01.002

[3] Shahid Z, Dupont F and Baskurt A, in: A Novel Efficient Image Compression System Based on Independent Component Analysis, WAIP'09, Proceedings of the SPIE, 2009, 7248(08): 1-9.

DOI: 10.1117/12.806159

[4] Hyvarinen A and Oja E, in: A fast fixed-point algorithm for independent components analysis, Neural Computation, 1997, 9 (7): 1483-1492.

DOI: 10.1162/neco.1997.9.7.1483

[5] Shang li and Huai wenjun, in: The application of sparse coding algorithm based on kurtosis criterion in natural image compression, Pattern Recognition, 2009: 304-308.

DOI: 10.1109/ccpr.2009.5344107

[6] Duan lijuan, Wu chunpeng etc, in: Visual Saliency Detection by Spatially Weighted Dissimilarity, CVPR'11, Colorado Springs, USA, Jun. 21-23, (2011).

DOI: 10.1109/cvpr.2011.5995676

[7] Wang Z, Bovik A. C etc in: Image quality assessment: From error visibility to structural similarity. IEEE Transactions on Image Processing, 2004, 13(4): 600-612.

DOI: 10.1109/tip.2003.819861