change detection based on integration of remote sensing and multi-criteria analysis

What is it about?

Based on the literature, the aforementioned methods integrated the transformed features or the spectral features. Thus, it is necessary to propose a framework in order for integrating all features including the spectral, textural and transformed ones. Based on the literature, fuzzy systems are a powerful tool in order for the integration of features because of their ability to model uncertainties. Moreover, the importance of each feature (i.e., the spectral, textural and transformed features) may change due to various factors, such as the properties of the study area. Hence, a suitable framework should be proposed for CD which must be able to adapt its parameters and modelling uncertainties. The Adaptive-Network-Based Fuzzy Inference System (ANFIS) is used as a MCDM method to integrate CD methods (Janalipour and Mohammadzadeh 2015). In addition to the mentioned goals, to achieve robust results, a comprehensive sensitivity analysis is proposed to obtain the best ANFIS parameters leading to the best result and accuracy.

Featured Image