This final computer vision project utilizes the Real-ESRGAN model to enhance an image dataset obtained from the open-source website Roboflow. The enhanced images are then used to train a Bike Helmet Detection model using the YOLOv8 framework.
-
Image Enhancement
- Model Used: Real-ESRGAN
- Purpose: To enhance the quality of the image dataset obtained from Roboflow, improving the clarity and detail of the images.
-
Model Training
- Model Used: YOLOv8
- Purpose: To train a Bike Helmet Detection model using the enhanced image dataset.
The project demonstrates the application of advanced image enhancement techniques and state-of-the-art object detection models to create an effective Bike Helmet Detection system. By leveraging the Real-ESRGAN model for image enhancement and YOLOv8 for object detection, the project aims to improve the accuracy and reliability of helmet detection in various images.