ABSTRACT
We present a multiresolution approach for blind image separation convolutely mixed. To move in transform domain, we make use of an Adaptive Quincunx Lifting Scheme based on wavelet decomposition followed by a geometric unmixing algorithm. In others words, the mixed signals are decomposed by an adaptive lifting scheme. Then, the unmixing algorithm is applied to the more relevant components. Experiments carried out on images from various origins showed that the proposed method yields better results than many widely used blind image separation algorithms.
- P. Comon, Independent component analysis, A new concept Signal Process, 36 (1994) 287--314. Google ScholarDigital Library
- S. I. Amari, A. Cichocki, H. H. Yang. A new learning algorithm for blind source separation, Advances in Neural Information Processing Systems 8, pp. 757--763, MIT Press, 1996.Google Scholar
- A. Bell, T. Sejnowski.An Information-maximization approach to blind separation and Blind Deconvolution. Neural Computation, 7:1129--1159, 1995. Google ScholarDigital Library
- P. Smaragdis. Blind separation of convolved mixtures in frequency domain. Neurocomputing, 22:21--34, 1998.Google ScholarCross Ref
- A. Ciaramella and R. Tagliaferri Separation of Convolved Mixtures in Frequency Domain ICA. International Mathematical Forum, 1, no. 16, 769--795 2006.Google ScholarCross Ref
- Hiroshi Sawada, Hirokazu Kameoka, Shoko Araki, Naonori Ueda Efficient Algorithms For Multichannel Extensions Of Itakura-Saito Nonnegative Matrix Factorization 978-1-4673-0046-9/12 ICASSP 2012.Google Scholar
- Shoko Araki, Ryo Mukai, Shoji Makino, Tsuyoki Nishikawa, Hiroshi Saruwatari. The Fundamental Limitation of Frequency Domain Blind Source Separation for Convolutive Mixtures of Speech. IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING, VOL. 11, NO. 2, MARCH 2003.Google Scholar
- Wady Naanaa, Jean-Marc Nuzillard A geometric approach to blind separation of nonnegative and dependent source signals. Signal Processing 92(11): 2775--2784 (2012). Google ScholarDigital Library
- M. Zibulevsky, B. A. Pearlmutter, P. Bofill, and P. Kisilev, Blind Source Separation by Sparse Decomposition, chapter in the book: S. J. Roberts, and R. M. Everson eds., Independent Component Analysis: Principles and Practice, Cambridge, 2001.Google ScholarCross Ref
- J. Hattay, A. Benazza-Benyahia, J.-C. Pesquet, Adaptive lifting schemes using variablesize block segmentation, Proc. of the ACIVS Conf., Brussels, Belgium, August 31-September 3, 2004.Google Scholar
- W. Sweldens, The lifting scheme: a new philosophy in biorthogonal wavelet constructions, SPIE Wavelet Applications Signal Image Processing III, vol. 2569, pp. 68--79, 1995.Google ScholarCross Ref
- Jean Christophe Feauveau Analyse multirésolution pour les images avec un facteur de résolution √2. Traitement de Signal, 7(2) 1990.Google Scholar
- J. Hattay, A. Benazza-Benyahia, J.-C. Pesquet, Adaptive lifting for multicomponent image coding through quadtree partitioning, Proc. of the IEEE Int. Conf. on Acoustics, Speech, Signal Processing, Philadelphia, USA, March 2005.Google Scholar
- A. Benazza-Benyahia, J.-C. Pesquet, J. Hattay, H. Masmoudi, Blockbased adaptive vector lifting schemes for multichannel image coding, Eurasip International Journal of Image and Video Processing (IJIVP) 2007(1) (2007) article ID 13421, 10 pp. Google ScholarDigital Library
- T. Motzkin, H. Raiffa, G. Thompson, R. J. Thrall, The Double Description Method. Annals of Math Studies, 8, Princeton University Press, 1953, pp. 51--73.Google Scholar
- D. Cochran, H. Gish, D. Sinno, A geometric approach to multiple-channel signal detection, IEEE Trans. on Signal Process., 43(9) (1995), 2049--2057. Google ScholarDigital Library
- Ilkka Virtanen Entropy correlation coefficient, a measure of statistical dependence for categorized data. Paper presented at the 15th European Meeting of Statisticians, Palermo, Italy 1982.Google Scholar
- C. R. Rao and T. K. Nayak Cross entropy, dissimilarity measures, and characterizations of quadratic entropy IEEE Transactions on Information Theory vol. IT-31 no. 5 pp. 589--593 1985 Google ScholarDigital Library
- D. Donoho, Denoising by soft thresholding, IEEE Transactions on Information Theory, vol. 41, no. 3, pp. 613--627, May 1995. Google ScholarDigital Library
- http://vasc.ri.cmu.edu/idb/html/face/frontalimages.Google Scholar
- http://sun.iinf.polsl.gliwice.pl/rstaros/mednat/index.Google Scholar
Index Terms
- Multiscale convolutive blind source separation in wavelet transform domain
Recommendations
Adaptive vectorial lifting concept for convolutive blind source separation
This paper describes a new multi-resolution approach for the blind separation of convolutive image mixtures in transform domain. The proposed method uses an Adaptive Vectorial case of Quincunx Lifting Scheme (AVQLS), based on wavelet decomposition, and ...
Blind Source Separation Algorithm Based on Wavelet Denoising
PCSPA '10: Proceedings of the 2010 First International Conference on Pervasive Computing, Signal Processing and ApplicationsIn this paper, a method of blind source separation (BSS) is proposed based on wavelet denoising. This method firstly makes wavelet transform (WT) to the observation signals, and then adopts a uniform method which combines signal whitening and maximum ...
Convolutive blind source separation by fourth-order statistics
ICICS'09: Proceedings of the 7th international conference on Information, communications and signal processingIn this paper, we present a new method using fourthorder statistics for convolutive blind source separation. After a whitening process of the observed data, the separation of convolutive mixtures are transformed to find a semi-unitary demixing matrix. ...
Comments