Systematic analysis of satellite image-based land cover classification techniques: literature review and challenges

Abstract

ABSTRACT As the land cover is a basic and important factor, affecting and connecting various parts of the human and physical environment, the classification of land cover plays a major role in the recent research. Hence, accurate and effective techniques are required for the classification to provide meaningful information regarding climate change, bio-diversity variation, and so on. Remote Sensed (RS) data obtained from the remote sensors are capable of providing easily accessible data that is used in different earth observation applications. Satellite image-based land cover classification is one of the interesting research areas. In this paper, 50 research papers that are based on the land cover classification using satellite images are surveyed. The research papers are categorized based on different classification techniques, such as fuzzy, Support Vector Machine (SVM), Neural Network (NN), Bayesian model, Decision Tree (DT) and so on. Finally, review and analysis are done based on the datasets, the number of bands considered, evaluation metrics, simulation platform, sensors, and the performance attained. Furthermore, the review suggests some major future scope to the researches based on the challenges and the research gaps in the reviewed papers.