Machine Learning Models for the Analysis of MultiTemporal Satellite Images for Change Identification and Predictions

Satellite data are used to monitor certain fields such as deforestation, polar ice loss, and climate change. The rate and extent of forest loss and ice melt, as well as global climate change trends, may all be tracked. Multi-temporal images are satellite photos or aerial photographs acquired by a sensor at different times but corresponding to the same location or area covered by that sensor. Change detection and prediction is essential for detecting changes in aerial images, environmental changes, and determining land utilization and coverage. Neural networks are robust Machine Learning (ML) models that can learn from data and perform many tasks, including image identification, natural language processing, and speech synthesis. Most neural networks lack the ability to adapt to new data inputs that change over time. The proposed study employs an Artificial Neural Networks (ANN) that adapts to the dynamic and unpredictable satellite image detection and prediction.