Particle swarm optimization to determine conjointly well localized MLTs

This paper explores the use of the recently proposed particle swarm optimization method to determine conjointly time-frequency well localized filters that constitute a multi-channel perfect reconstruction system. The time-frequency measure used to determine optimality is the product of a filter´s time and frequency variances. The particle swarm optimization method will be shown to be an effective and efficient method to determine conjointly well localized filters of complex modulated lapped transforms. The complex modulated lapped transform optimized by the particle swarm optimization method is used to decompose an image. The output of the analysis filters of the optimized complex modulated lapped transform illustrates that it is the best space-frequency localized decomposition of the original image.