A Survey of Particle Swarm Optimization and Random Forest based Land Cover Classification

The main objective of the survey is to Study the emerging classifiers like Random Forest (RF) and Particle Swarm Optimization (PSO) and its application to Satellite Imageries to achieve enhanced and highly accurate Land Cover Classification Model. RF is an ensemble type voting based machine learning algorithm. RF algorithm considers single pixels for Classification instead of sets of pixels as in the case of object-oriented algorithms. Particle Swarm Optimization is an advanced but simple search strategy for optimizing complex numerical functions based on group response of bird flocking. It is a population based process of Swarm Intelligence. This paper is a literature survey that examines the performances of PSO and RF algorithm.