Analyzing Remote Sensing Images Based on Climate Prediction Model for Water Resources Using Deep Learning

In recent years, satellite imagery can provide key information about the properties of land, water, and natural resources. Climate change threaten the environment and the satellite image processing has shown its potential to support future prediction of natural resources. The aim of this research is to proposed a model to analyze the extracted information from the environmental image for water area and follow the changes that may happen over time. The proposed model is (Image Time Series Tracking with LSTM Deep Learning Method) that has two stages: the first stage is analysis remote sensing images of water area and track its change over years and second stage is to predicate the future change by using LSTM deep learning method. In first stage a data set is collected for Aral Sea from 2003-2018 and in the second stage Long Short-Term Memory (LSTM) deep learning method is implemented to detect the future change in environment. The results show a promising prediction especially when the input that we used is real data set for Aral Sea sensing images for water area over long years in LSTM deep learning methods.