97hackbrian/Seeker_ComputerVision

We're building an AI vision system for a minesweeper robot competing in Minesweeper 2024, Egypt. Our approach includes machine learning, deep learning, and a Convolutional Neural Network (CNN) for mine detection, using YOLOv8 for efficient mine location. # Seeker_ComputerVision Our project aims to implement an artificial vision system for a minesweeper robot, which will compete in the Minesweeper 2024 competition in Egypt. The artificial vision system leverages a combination of machine learning and deep learning techniques, encompassing data acquisition, image processing, and compression methods. Additionally, a Convolutional Neural Network (CNN) is employed for mine detection and classification. To approach the mines effectively, we employ techniques based on YOLOv8. ## Badges # Table of contents 1. Introduction 2. Problem 3. Solution 1. Block_diagram # Introduction Landmines, concealed beneath the Earth's surface, pose a grave threat, claiming lives and causing grievous injuries, particularly among innocent civilians. These insidious devices have left a lasting legacy in conflict-ridden regions, hindering economic progress and devastating communities dependent on affected land. Current manual demining efforts are both perilous and time-consuming, requiring individuals to risk their lives dismantling these hidden killers. This project addresses the urgent need for advanced solutions to this humanitarian crisis by leveraging cutting-edge technology. Robots and autonomous systems offer a ray of hope, promising safe detection and efficient elimination of landmines. Initiatives like the "Minesweepers" competition and the "SEEKER" project champion technological innovation as the key to solving this global issue.The project employs computer vision methods to enhance mine detection capabilities. The Region Proposing Convolution Network (RPN) is integral in proposing regions of interest within images. Furthermore, end-to-end training streamlines the model, improving accuracy, reducing training time, and simplifying the process. The YOLO algorithm, incorporated in the SEKEER mine finder, is employed for metal mine detection, with the ability to predict bounding boxes and class probabilities for potential metal mines i …