Investigation on agent based models for image classification of land use and land cover maps

Rapid rises in population and urbanization can cause crucial and expeditious changes on land cover and land use. For this reason, monitoring in frequent periods of changes in the environment and heterogeneous areas is needed for strategic planning of sustainable applications and optimized management issues. As known, various regional and global initiations co-operates on Earth Observation services for environmental monitoring. Some main topics for these services are land management projects, cadastre, forestry, agriculture, rural and urban planning, environmental monitoring and so on. The improving technology and studies enable to use and analyze many different data sources efficiently and develop new methods for image interpretation, geo-information extraction, and processing. Today, geospatial intelligence influences all spatial and geographical sciences, as well image analysis. Widespread usage of remote sensing images for concerning both Earth's physical features and increase of man-made environmental changes bring semi-automatic and automatic analysis classification methods by side. Especially, VHS resolution imagery is started to be used as object-based image analyses methods, rule-based classification methods. Developing countries are started to standardize the land use and land cover (LULC) classification systems and nomenclature managing information more effectively and rapidly at national or regional levels years ago. In this study, it is searched of advantages and requirements of ABM and agent based image analysis as an additional method to other image analysis methods for regional monitoring programs. In additionally, ABM stages of the classification algorithm and several ABM approaches (based on probability, Bayesian, Neural Network, and Genetic Algorithms) are investigated and described with examples on remotely sensed data. In the result of the models; the different land cover and land use map products are compared and interpreted in a scientific explication.