On the precise number of (0,1)-matrices in U(R,S)
Discrete Mathematics
On the computational complexity of reconstructing lattice sets from their x-rays
Discrete Mathematics
New ideas in optimization
Principles of computerized tomographic imaging
Principles of computerized tomographic imaging
A massively parallel architecture for distributed genetic algorithms
Parallel Computing - Special issue: Parallel and nature-inspired computational paradigms and applications
An introduction to periodical discrete sets from a tomographical perspective
Theoretical Computer Science
A genetic algorithm for discrete tomography reconstruction
Genetic Programming and Evolvable Machines
A distributed genetic algorithm for restoration of vertical line scratches
Parallel Computing
An Evolutionary Approach for Object-Based Image Reconstruction Using Learnt Priors
SCIA '09 Proceedings of the 16th Scandinavian Conference on Image Analysis
An evolutionary algorithm for discrete tomography
Discrete Applied Mathematics - Special issue: IWCIA 2003 - Ninth international workshop on combinatorial image analysis
Reconstruction of 8-connected but not 4-connected hv-convex discrete sets
Discrete Applied Mathematics - Special issue: Advances in discrete geometry and topology (DGCI 2003)
A memetic approach to discrete tomography from noisy projections
Pattern Recognition
An island strategy for memetic discrete tomography reconstruction
Information Sciences: an International Journal
| Hi-index | 0.00 |
Soft computing is a term indicating a coalition of methodologies, and its basic dogma is that, in general, better results can be obtained through the use of constituent methodologies in combination, rather than in a stand alone mode. Evolutionary computing belongs to this coalition, and thus memetic algorithms. Here, we present a combination of several instances of a recently proposed memetic algorithm for discrete tomography reconstruction, based on the island model parallel implementation. The combination is motivated by the fact that, even though the results of the recently proposed approach are finally better and more robust compared to other approaches, we advised that its major drawback was the computational time. The underlying combination strategy consists in separated populations of agents evolving by means of different processes which share some individuals, from time to time. Experiments were performed to test the benefits of this paradigm in terms of computational time and correctness of the solutions.