Level Crossing oject classification using YOLO

The project involves developing an object classification system for railway level crossings using YOLO (You Only Look Once) implemented in C++. The system processes pre-recorded video or image datasets to detect and classify objects such as vehicles, pedestrians, and barriers near level crossings. The implementation utilizes the YOLO model for inference, with input frames preprocessed and analyzed sequentially. The results include bounding boxes and classification labels drawn on detected objects, which are saved for further review. The focus is on accuracy and robustness rather than real-time performance, making it suitable for offline analysis and safety audits. The system will be tested under various environmental conditions to ensure reliability and scalability.