Spatial models for fuzzy clustering
Computer Vision and Image Understanding
Interactive Organ Segmentation Using Graph Cuts
MICCAI '00 Proceedings of the Third International Conference on Medical Image Computing and Computer-Assisted Intervention
A non-local fuzzy segmentation method: Application to brain MRI
Pattern Recognition
Automatic FFT Performance Tuning on OpenCL GPUs
ICPADS '11 Proceedings of the 2011 IEEE 17th International Conference on Parallel and Distributed Systems
Hi-index | 0.00 |
Image segmentation is one of the most difficult tasks in image processing and plays a critical role in the analysis of medical images used for diagnosis and treatment. With the decreased hardware costs and improvements in computing power of many-core architectures, there is an opportunity to both improve upon image segmentation algorithms and to make this technology more accessible. This paper describes our on-going research efforts to implement efficient image segmentation algorithms on graphical processing units (GPUs). A focused case study was performed with a suitable algorithm based on Cellular Automata, a parallel computational technique. Preliminary segmentation results are shown to validate our approach. Plans to improve the algorithm by making it more robust to noise and more efficient on GPU architectures are discussed. Our use of graph theoretic techniques and their implementation on GPUs will have broad application to other areas requiring computationally intensive calculations, as found in many problems involving modeling and simulation.