Utilizing Data Control and Resource Reservation for AI Image Matching with Satellite Imagery Using Long-Short Term Memory Model to Detect Flooding

RIFFAI is an advanced AI model designed to detect signs of flooding by leveraging various data sources and analytical techniques including satellite imagery databased, image matching point distribution, forecasted weather. The model continuously receives real-time satellite imagery. This imagery provides up-to-date visual information about the Earth's surface, allowing the model to monitor water bodies, land cover, and changes in terrain that may indicate flooding. The study detects flood areas using a decision fusion method and analyzes the effects on a macro scale, particularly focusing on land use and land cover. RIFFAI leverages Convolutional Neural Networks (CNN) for image classification and processing, which offers a more refined and accurate analysis compared to other competitors using physics-based (language-based) models. Additionally, our Long Short-Term Memory (LSTM) model enables us to manage and control data more effectively, preventing information overload while still making precise predictions. This approach ensures that our predictions are not just accurate but also timely, which is crucial in climate resilience and early warning systems. With the use of CNN and LSTM model to detect patterns of environment changes occur in satellite imagery. RIFFAI can detect flooding, coastal change, vegetation, and wildfire.