Segmentation of Very High Resolution Remote Sensing Imagery of Urban Areas Using Particle Swarm Optimization Algorithm

Abstract

As the improvement of the resolution of aerial and satellite remote sensed images, the semantic richness of the image increases which makes image analysis more difficult. Dense urban environment sensed by very high-resolution (VHR) optical sensors is even more challenging. Occlusions and shadows due to buildings and trees hide some objects of the scene. Fast and efficient segmentation of such noisy images (which is essential for their further analysis) has remained a challenging problem for years. It is difficult for traditional methods to deal with such noisy and large volume data. Clustering-based segmentation with swarm-based algorithms is emerging as an alternative to more conventional clustering methods, such as hierarchical clustering and k-means. In this paper, we introduce the use of Particle Swarm Optimization (PSO) clustering algorithm segmenting high resolution remote sensing images. Contrary to the localized searching of the K-means algorithm, the PSO clustering algorithm performs a globalized search in the entire solution space. We applied the PSO and K-means clustering algorithm on thirty images cropped from color aerial images. The results illustrate that PSO algorithm can generate more compact clustering results than the K-means algorithm.