Hybrid Change Detection Technique with Particle Swarm Optimization for Land Use Land Cover Using Remote-Sensed Data

The process of identifying and analyzing the changes occurring over a period using remote-sensed data is change detection and has various application areas such as land use land cover, resource planning, urbanization and many more. The detection of changes is required for better decision-making and for understanding the impact of changes occurring at local and global levels. This research presents an implementation of two change detection techniques: image differencing and image ratioing on a set of ten remote-sensed data. The output images obtained are further segmented using artificial intelligence-based particle swarm optimization and conventional techniques. The output images are compared and validated though the use of entropy and piqe. It has been observed that image differencing followed by PSO gives better and superior image quality in comparison to other implemented techniques.