Since September 2010, I have been conducting research in the field of image reconstruction at Ghent University, Faculty of Engineering, Department of Telecommunications and Information Processing, Image Processing and Interpretation group.
Research Interests:My main research interests are on inverse problems, medical imaging (computing), structured sparsity and other related topics in image reconstruction or processing field. In particular, I would concentrate on the regularization of an optimization problem.
On the inverse problem aspect, I have proposed a class of Weakly Convex Discontinuity Adaptive (WCDA) models as the regularization in microwave imaging (microwave tomography). This proposed models helped to improve the imaging results in electromagnetic field significantly. The preliminary results from our method about solving this inverse problem has already published in the journal IEEE Transactions on antenna and propagation, where a much sparser 2D Fresnel experimental data was verified. Furthermore, we have explored the models deeply and tested the proposed models with different subsampling strategies at 3D Fresnel experimental data, which has been accepted by the journal Inverse Problems. With far less complex data, the 3D imaging speed is improved greatly.
Some of the results on 3D quantitative microwave imaging in biomedical application (Breast phantoms with artificial tumour) for early breast cancer detection was presented in the 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC’13).
Structured sparsity is another main topic I am interested in. Specifically, I am working on graph sparsity using Ising model from Markov Random Field (MRF) as a spatial prior in the optimization. Magnetic Resonance Imaging (MRI) is one of typical applications from this topic.
Structured sparsity (MRI application)