Dr James Nelson

Position Senior Lecturer
Personal webpage http://www.homepages.ucl.ac.uk/~ucakjdb/
Themes Stochastic Modelling and Time Series

Biographical Details

James Nelson

James Nelson joined the Department of Statistical Science as a Lecturer in 2010 and became Senior Lecturer in 2013. After a PhD in applied harmonic analysis from the Mathematics Department at Anglia Polytechnic University (1998-2001), he held post-doc positions in: the Applied Mathematics and Computing Group at the University of Cranfield (2001-2004); the Information: Signals, Images, and Systems Research Group at the University of Southampton (2004-2006); and the Signal Processing and Communications Laboratory at the University of Cambridge (2006-2010).

Research Interests

Multiresolution analysis; random fields; spatial models; statistical machine learning; statistical signal and image processing. Examples: locally stationary wavelet processes, Markov random fields; wavelet/Riesz basis and packet construction and pursuit/optimisation; sparsity and other regularisation with applications to detection and classification; weak self-similar random fields; support vector machine kernel construction and hyper-parameter estimation; generalised cardinal series and sampling theory; robust, regularised scaling exponent estimation.


  • Hojjat Akhondi-Asl (time series and regularisation)
  • Chunli Guo [now at Winton Capital] (compressive sensing for beamforming)
  • Alfredo Kalaitzis [now at Tata Consultancy Services] sparse methods for anomaly detection
  • Vladimir Krylov [now at University of Genoa]

    • stochastic geometry for image analysis
    • semi-supervised anomaly detection
  • Diego Tomassi [funded visit from National University of the Littoral] (wavelet shrinkage using adaptive structured sparsity constraints)

Research Students

  • Alex Gibberd (sparsity on graphs and statistical machine learning on networks)
  • Maria Toomik (multiresolution analysis and regularisation approaches to highly structured data)
  • Jean-Baptiste Regli (computational statistics and spatial models)
  • Nikolaos Tsipinakis (regularity and information filtering approaches to uncertainty management)
  • Johannes Happenhofer (the lasso and multiresolution analysis)
  • Christian Niedworok (automated cell detection algorithms in 2-photon microscopy of whole brain images; 1st supervisor Prof. Troy Margrie, MRC National Institute for Medical Research)

Selected publications

  • Nelson, J. D. B., Nafornita, C., and Isar, A. (2016) "Semi-local scaling exponent estimation with box-penalty constraints and total-variation regularization. IEEE Transactions on Image Processing, 25(7): 3167 - 3181
  • Peng, C., Nelson, J. D. B., and Tourneret, J.-Y. (2016) "Towards a sparse Bayesian Markov random field approach to hyperspectral unmixing and classification". IEEE Transactions on Image Processing (accepted) 
  • Nelson, J. D. B. (2015) "Enhanced B-wavelets via mixed, composite packets" IEEE Transactions on Signal Processing, 63(12):3191-3203
  • Tomassi, D. R., Milone, D. H., and Nelson, J. D. B. (2015) "Wavelet shrinkage using adaptive structured sparsity constraints". Signal Processing, 106:73-87
  • Nelson, J. D. B. (2014) "On the equivalence between a minimal codomain cardinality Riesz basis, a system of Hadamard-Sylvester operators, and a class of sparse, binary optimisation problems". IEEE Transactions on Signal Processing, 62(20):5270-5281
  • Krylov, V. and Nelson, J. D. B. (2014) "Stochastic extraction of elongated curvilinear structures with applications" IEEE Transactions on Image Processing, 23(12):5360--5373
  • Nelson, J. D. B. (2013) "Fused Lasso and rotation invariant autoregressive models for texture classification". Pattern Recognition Letters 34(16):2166-2172 

More publications, with preprints, code, and opportunities, etc can be found here

See also my Google Scholar and Google Sites pages and the Centre for Computational Statistics and Machine Learning site.