Overview:
This project aims to detect cars and other vehicles using YOLOv8m, a real-time object detection model. The model processes 240 images and identifies up to 10 vehicle classes, including cars and trucks. With 50 training epochs and a 70/20/10 train/validation/test split, the model demonstrates effective detection across a range of vehicles.
Key Features:
1. YOLOv8m for Vehicle Detection:
The model is built to detect up to 10 different vehicle classes, including cars and trucks, offering versatility in various traffic environments.
2. Dataset of 240 Images:
The model is trained on a dataset of 240 images with a resolution of 640×640, ensuring detailed object detection for each vehicle.
3. No Augmentation:
The model was trained without any data augmentation, focusing purely on the provided dataset to measure its baseline performance.
4. Performance Metrics:
The model achieved a mean average precision (mAP) of 0.764, with precision at 0.757 and recall at 0.687, showing solid detection capabilities.
5. Training Process:
The data was split into 70% for training, 20% for validation, and 10% for testing over 50 epochs, which helped optimize model performance.
Category:
Tags:
Links
Portfolio 
Leave a comment