P.-Y. Yin (Taiwan)
Polygonal approximation, particle swarm optimization,genetic algorithm, local optimum, global optimum
Polygonal approximation of digital curves is one of the crucial steps prior to many image analysis tasks. This paper presents a new polygonal approximation approach based on the particle swarm optimization (PSO) algorithm. Each particle represented as a binary vector corresponds to a candidate solution. A swarm of particles are initiated and fly through the solution space for targeting the optimal solution. We also propose hybrid versions of PSO embedding a local optimizer to enhance the performance. The experimental results manifest that the proposed discrete PSO is comparable to the genetic algorithm, and it outperforms another discrete implementation of PSO in the literature. The proposed hybrid versions of PSO can significantly improve the approximation results in terms of the compression ratio, and the results obtained in different runs are more consistent.
Important Links:
Go Back
