AI-Powered Port Logistics Analysis: Container Detection from Satellite Images

This project utilizes deep learning-based object detection to analyze port logistics and container activity from satellite imagery. By leveraging Faster R-CNN or SSD models, it detects and tracks shipping containers, vessels, and port capacity, providing insights into supply chain efficiency and economic trends.

As a Research Assistant at the University of South Florida, I am actively involved in the AI-powered port logistics analysis research project under the guidance of Professor Seckin Ozkul and Professor Ehsan Sheybani. My contributions focus on developing and implementing deep learning models for container detection and port activity analysis using satellite imagery. 

Working alongside fellow researchers Manvith and Premal, I play a key role in:

• Preprocessing geospatial data to ensure model accuracy and efficiency.

• Implementing deep learning models (such as Faster R-CNN and SSD) for object detection.

• Running inference and analyzing detection outputs to estimate port capacity and supply chain efficiency.

• Enhancing visualization techniques using OpenCV and TensorFlow to improve interpretability of results.

• Optimizing model performance and fine-tuning hyperparameters for real-world applications.

This research contributes to logistics optimization, economic impact assessment, and supply chain management by providing data-driven insights into container movement and port congestion trends. Through this collaborative effort, we aim to enhance decision-making for policymakers and logistics professionals using AI-driven geospatial analysis.