Một mô hình suy diễn mờ phức không gian mới cho bài toán

Spatial and temporal changing detection in dataset of remote sensing images is very important to support in detect effectively the changing of natural events such as weather, climate, flood,... In this paper, we propose a novel spatial complex fuzzy inference system denoted as Spatial CFIS to support in detecting changes in remote sensing images. Our proposed model generates complex fuzzy rules using fuzzy clustering and provides predicted images using triangular spatial complex fuzzy rule base. To increase the performance, Spatial CFIS uses ADAM algorithm in order to optimize the weight set of rule base. The proposed model is experimented on a dataset from weather image data warehouse of USA Navy and evaluated by comparing with other models related to changing detection of sensing images. The implemental results show that Spatial CFIS is the best model in term of the value of RMSE.