Overview / Usage

Our project is focused on addressing the critical issue of climate change and its impact on the environment, human health, and the economy. Our first goal is to detect forest fires using machine learning and deep learning techniques. We plan to collect satellite images from NASA and use algorithms to process the data and detect climate changes and disasters. We aim to distribute the processed data to the relevant authorities to help with disaster management.

We have identified various international and national level organizations that work on disaster management, and we plan to transfer the data from our models to these organizations. We have highlighted the novelty of our project, which is its cost efficiency compared to IoT-based models for detecting wildfires, its ability to accurately predict deforestation activities, and its real-time data provision to government authorities for monitoring sea levels and detecting water body encroachments.

To turn our project into a business model, we suggest obtaining more accurate and high-resolution images from space agencies and integrating cloud services with cloud computing servers to provide 24/7 data. We also propose using highly accurate and efficient machine learning algorithms and deep learning models to predict data and convey these predictions to the relevant organizations with GPS data.

We have developed a prototype model with an accuracy of over 85% for detecting forest fires and developed a database and a local webserver to interface with the deep learning model and fetch the required results in a GUI interface. Our future scope involves designing a robust AI model to detect various calamities such as deforestation, agricultural land, drought, cyclone, volcanic eruption, landslides and seismic activity, monitoring sea levels, detecting coral reef bleaching, and crop prediction.

Methodology / Approach

The methodology for this project involves using advanced technologies such as satellite imagery, machine learning, and deep learning to detect and manage disasters caused by climate change, with a focus on detecting forest fires. We are collecting satellite images from NASA and processing them using algorithms to detect climate changes and disasters, and then distributing the data to relevant authorities for disaster management. Our development involves using frameworks such as TensorFlow and Keras for machine learning, and Python for programming, as well as adhering to standards for data privacy and security. Our techniques include data preprocessing, feature selection, and model training and evaluation.

Technologies Used

The technologies used in the development of this project include:

• Satellite imagery from NASA
• Machine learning and deep learning algorithms
• Python programming language
• TensorFlow and Keras libraries
• Cloud computing servers
• Global Positioning System (GPS)
• High-resolution images from space agencies
• Local webserver
• GUI interface
• Intel-based processors for computation and analysis.

Documents and Presentations