A computer vision anomaly detection tool using PyTorch and a fine‑tuned ResNet50 model on the MVTec AD dataset to identify defects across diverse industrial components. The system evaluates performance with accuracy and F1 scores per defect type, visualizes results using Matplotlib heatmaps and includes a Streamlit app for real‑time image uploads and detection

This project focuses on real-time object detection using the YOLOv9 deep learning model. The system captures live video through a camera, processing each frame to detect and classify various objects, such as persons, cell phones, books and others. With YOLOv9's advanced capabilities, the system ensures fast, accurate identification, making it suitable for applications in surveillance, smart systems and automated environments where immediate object recognition is required. 

This project utilizes the YOLOv8 deep learning model for fire detection in video frames. The system processes recorded video footage to identify and locate fire outbreaks accurately. Trained on a comprehensive fire dataset, YOLOv8 ensures high precision in detecting fire in various environments, making it ideal for post-event analysis and safety monitoring applications. 

This project implements a vehicle counting system in lanes using YOLOv8, a state-of-the-art object detection model. The system processes video footage to detect and track vehicles as they pass through predefined lanes. YOLOv8's high-speed and accurate detection capabilities ensure reliable vehicle counting, even in dynamic traffic conditions. This solution is ideal for traffic monitoring, smart city applications and transportation management, providing real-time insights into vehicle flow and congestion levels. 

This project utilizes YOLOv11 for vehicle speed detection through video analysis. The system captures video frames from surveillance cameras and processes them in real-time to detect and track vehicles. By analyzing the time it takes for a vehicle to pass between predefined points in the video, the system calculates the speed of each vehicle. YOLOv11's advanced object detection capabilities ensure accurate vehicle detection and tracking, making this solution ideal for traffic monitoring, law enforcement, and smart city applications to detect speeding vehicles in real-time.