• Research interests
• Publications
• Software
• Data sets

Research interests

• I've always had a strong interest on Physics. I've had, for a long time, some ideas on General Relativity that I feel may be interesting, and I've recently decided to develop them further. I'm presently writing them down in a (somewhat long) paper. This may take a while to complete beacuse, as I write them down, I sometimes find aspects that can still be perfected, and do some more work on them.

• "Really blind" deblurring of natural images: deblurring of natural images degraded by blurs about which we know very little. Basically, we assume that the blur's point spread function has a limited support. This includes focus blurs, motion blurs, combined focus + motion blurs, as well as many other kinds of blurs about which we know very little.
   
  We have developed a method that can perform a rather good-quality deblurring in these kinds of situations. See the IEEE Trans. and ICASSP 2008 papers below.
   
• One of the directions in which we're currently working is the deblurring of images in which different parts have different blurs, such as an image with a sharp background and with a fast-moving car which is affected by motion blur. See the ICIP 2009 paper below.

Independent Component Analysis / Blind Source Separation

• Independent component analysis (ICA) and blind source separation (BSS) algorithms. Especially, blind separation of nonlinear mixtures and of non-independent sources.
  
  MISEP is a method for linear and nonlinear ICA / BSS, based on the minimization of the mutual information of the extracted components.
  
  It is an extension of the well known INFOMAX algorithm of Bell and Sejnowski. It extends INFOMAX in two aspects:
• It is able to perform both linear and nonlinear ICA.
• It learns the output nonlinearities, instead of having them chosen a priori.
  For nonlinear ICA it is important that the output nonlinearities be closely matched to the independent components' cumulative distribution functions. Otherwise the method would not be minimizing the mutual information of the extracted components, and therefore would not extract independent components.
  
  
• Get the MISEP Linear and Nonlinear Independent Component Analysis Toolbox here.
  
  
  

• Neural networks (learning algorithms, statistical aspects).
  
• Signal processing (fractional Fourier transform, time-frequency analysis).
• Speech processing.

Publications

For the most recent publications, see here.

Blind image deblurring
 

Papers

M. S. C. Almeida and L. B. Almeida, "Blind and Semi-Blind Deblurring of Natural Images ", IEEE Trans. on Image Processing, Vol. 19, No. 1, pp. 36 - 52 , January, 2010. (download)

M. S. C. Almeida and L. B. Almeida, "Blind deblurring of foreground-background images", Proc IEEE International Conf. on Image Processing - ICIP, Cairo, Egypt, November, 2009. (download)

M. S. C. Almeida and L. B. Almeida, "Blind Deblurring of Natural Images", Proc. 2008 ICASSP (Int. Conf. Acoustics, Speech and Signal Processing), Las Vegas, NV, U.S.A., March-April 2008, pp. 1261-1264. (download)

M. S. C. Almeida, J. Sanches and Luis B. Almeida, Blind Adaptive Deblurring of Chromosome images, FPBC-2008 - First Portuguese Forum on Computational Biology, Oeiras, Portugal, July 10-12, 2008.

Nonlinear ICA / BSS
 

Book

L. B. Almeida, Nonlinear Source Separation, Synthesis Lectures on Signal Processing, Morgan and Claypool Publishers, 2006.
doi:10.2200/S00016ED1V01Y200602SPR002
Available in electronic and printed form. Abstract.

Papers

* Papers on the MISEP method of linear and nonlinear ICA are marked with an asterisk.
 

* M. S. C. Almeida and L. B. Almeida , "Nonlinear separation of show-through image mixtures using a physical model trained with ICA", Signal Processing, Vol. 92, No. 4, pp. 872-884, April 2012.

M. S. C. Almeida and L. B. Almeida, "Wavelet-based nonlinear separation of nonlinear show-through and bleed-through image mixtures", Neurocomputing, vol. 72, pp. 57-70, 2008. (download)

* M. S. C. Almeida and L. B. Almeida, "Separating Nonlinear Image Mixtures Using a Physical Model Trained with ICA", Proc. 2006 IEEE Int. Worksh. Machine Learning for Signal Processing, Maynooth, Ireland, September 2006. (download)

M. S. C. Almeida and L. B. Almeida, "Wavelet Based Nonlinear Separation of Images",  Proc. 2006 IEEE Int. Worksh. Machine Learning for Signal Processing, Maynooth, Ireland, September 2006. (download)

* L. B. Almeida, “Separating a Real-Life Nonlinear Image Mixture”, Journal of Machine Learning Research, Vol. 6, pp. 1199-1232, 2005.
The pdf version of this paper has some slight artifacts in the images, due to the JPEG encoding. The postscript version is larger, but has no artifacts.
Download: pdf (~ 7 MB); postscript (~ 14 MB).
Note: If you've downloaded before August 16, 2005, please download again. There were some errors in the values of Q2 in Tables 1 and 2, which have been corrected.

* L. B. Almeida and Miguel Faria, “Separating a Real-Life Nonlinear Mixture of Images”, Independent Component Analysis and Blind Signal Separation (ICA 2004), Granada, Spain, September 2004, Springer-Verlag, Lecture Notes in Computer Science, vol. LNCS 3195. (download)

* L. B. Almeida, “Linear and Nonlinear ICA Based on Mutual Information – the MISEP Method”, Signal Processing, Vol. 84, No. 2, February 2004, pp. 231-245. (download)

* L. B. Almeida, “MISEP – linear and nonlinear ICA based on mutual information”, Journal of Machine Learning Research, Vol. 4, December 2003, pp. 1297-1318. (download from JMLR site: abstract pdf ps.gz ps; download from my site: pdf)

* L. B. Almeida, “Faster training in nonlinear ICA using MISEP”, in Proc. Fourth Int. Symp. on Independent Component Analysis and
Blind Signal Separation, Nara, Japan, 2003. (download paper) (download poster presented at the conference)

* L. B. Almeida, “MISEP – an ICA method for linear and nonlinear mixtures, based on mutual information”, in Proc. 2002 Int. Joint
Conf. on Neural Networks, Honolulu, Hawaii, 2002. (download)

* L. B. Almeida, “ICA of linear and nonlinear mixtures based on mutual information”, in Proc. 2001 Int. Joint Conf. on Neural Networks, Washington, D.C., 2001. (download)

* L. B. Almeida, “Linear and nonlinear ICA based on mutual information”, in Proc. Symp. 2000 on Adapt. Sys. for Sig. Proc., Commun.
and Control, Lake Louise, Alberta, Canada, 2000. (download)

L. B. Almeida, “Simultaneous MI-based estimation of independent components and of their distributions”, in Proc. Second Int. Worksh. Independent Component Analysis and Blind Signal Separation, Helsinki, Finland, 2000, pp. 169–174. (download)

L. B. Almeida, "Blind Source Separation by Pattern Repulsion'', in Proc International Work-Conference on Artificial and Natural Neural Networks, IWANN'99, Jun.1999, Alicante, Spain.

G. C. Marques and L. B. Almeida, “Separation of nonlinear mixtures using pattern repulsion”, in Proc. First Int. Worksh. Independent Component Analysis and Signal Separation, J. F. Cardoso, C. Jutten, and P. Loubaton, Eds., Aussois, France, 1999, pp. 277–282. (download)

G. C. Marques and L. B. Almeida, “An objective function for independence”, in Proc. International Conference on Neural Networks, Washington DC, 1996, pp. 453–457.

Adaptive decorrelation

L. B. Almeida and F. M. Silva, "Adaptive Decorrelation", in I. Aleksander and J. Taylor (eds.), Artificial Neural Networks, 2, Elsevier Science Publishers, 1992.

F. M. Silva and Luís B. Almeida, "A Distributed Decorrelation Algorithm", in Erol Gelenbe (ed.), Neural Networks - Advances and Applications, North-Holland, 1991.

F. M. Silva and L. B. Almeida, "A Distributed Solution for Data Orthonormalization", Proceedings of the International Conference on Artificial Neural Networks (ICANN'91), Espoo, Finland, 1991.

Fast training of multilayer perceptrons, and fast gradient-based optimization in general

L. B. Almeida, T. Langlois, J. D. Amaral and A. Plakhov, “Parameter Adaptation in Stochastic Optimization”, in D. Saad (ed.), On-Line Learning in Neural Networks, Cambridge University Press, 1998.  (download)

L. B. Almeida, “Multilayer perceptrons”, in Handbook of Neural Computation, E. Fiesler and R. Beale, Eds. Institute of Physics, 1997, Oxford University Press. (download pdf or zipped postscript).

Recurrent backpropagation

L. B. Almeida, "Backpropagation in Nonfeedforward Networks", in Igor Aleksander (ed.), Neural Computing, Kogan Page, London, 1989.

L. B. Almeida, "Backpropagation in Perceptrons with Feedback", in R. Eckmiller and C. v.d. Malsburg (eds.), Neural Computers Springer-Verlag, Heidelberg, 1988.

L. B. Almeida, "A Learning Rule for Asynchronous Perceptrons with Feedback in a Combinatorial Environment", Proceedings of the IEEE First International Conference on Neural Networks (ICNN'87), S. Diego, California, June 1987.

Fractional Fourier transform

L. B. Almeida, “Product and Convolution Theorems for the Fractional Fourier Transform, IEEE Signal Processing Letters, vol. 4, no.1, Jan. 1997.

T. Alieva, V. López, G. Agullo-López and L. B. Almeida, "The Fractional Fourier Transform in Optical Propagation Problems", J. Modern Optics, vol. 41, no. 5, 1994.

L. B. Almeida, "The Fractional Fourier Transform and Time-Frequency Representations", IEEE Transactions on Signal Processing, November 1994.

L. B. Almeida, "An Introduction to the Angular Fourier Transform", Proceedings of the 1993 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP'93), Minneapolis, E.U.A., 1993.

Sinusoidal modelling of speech

J. S. Marques and L. B. Almeida, "Sinusoidal Modelling of Voiced and Unvoiced Speech", Proceedings of the European Conference on Speech Communication and Technology (EUROSPEECH'89), Paris, France, September 1989.

J. S. Marques and L. B. Almeida, "Frequency Varying Sinusoidal Modelling of Speech", IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-37, May 1989.

J. S. Marques and L. B. Almeida, "Sinusoidal Modelling of Speech: Representation of Unvoiced Sounds with Narrow-Band Basis Functions", J. Lacoume, A. Chehikian, N. Martin e J. Malbos (eds.), Signal Processing IV: Theories and Applications, Elsevier Science Publishers, September 1988.

J. S. Marques and L. B. Almeida, "New Basis Functions for Sinusoidal Decompositions", Proceedings of the 8th European Conference in Electrotechnics (EUROCON'88), Stockholm, Sweden, June 1988.

J. S. Marques and Luís B. Almeida, "Quasi-Optimal Analysis for Sinusoidal Representation of Speech", Proceedings of the GRETSI'87, Nice, France, 1987.

J. S. Marques and L. B. Almeida, "A Background for Sinusoid-Based Representation of Voiced Speech", Proceedings of the 1986 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP'86), Tokyo, Japan, April 1986.

L. B. Almeida and F. M. Silva, "Variable-Frequency Synthesis: An Improved Harmonic Coding Scheme", Proceedings of the 1984 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP'84), S. Diego, California, March 1984.

L. B. Almeida and F. M. Silva, "Harmonic Coding with Variable-Frequency Synthesis", Proceedings of the 1983 Spain Workshop on Signal Processing and its Applications (WSPA'83), Sitges, Spain, September 1983.

Other topics
 
C. Carreiras, L. B. Almeida and J. M. Sanches, ERD/ERS Event Detection from Phase Desynchronization Measurements in BCI, 17th Portuguese Conference on Pattern Recognition, RecPad, October 28,  Porto 2011.