Cheetah Detection with YOLO

This project implements a cheetah detection system using the YOLO (You Only Look Once) object detection model. It includes scripts for training a custom YOLO model on a cheetah dataset, evaluating its performance, and running predictions on video files.

Project Structure

• train.py: Script for training the YOLO model on the cheetah_data dataset.
• test.py: Script for evaluating the trained model on a test set and running inference on video files.
• best.pt: The trained YOLO model weights, resulting from the training process.
• cheetah.yaml: Configuration file defining the dataset paths and class names for cheetah detection.
• requirements.txt: Lists all Python dependencies required to run this project.
• cheetah_data/: Contains the dataset used for training and validation. This directory should be structured according to YOLO dataset conventions.
• test_videos/: Contains video files used for testing the trained model's prediction capabilities.
• runs/: This directory will store the outputs of training (e.g., logs, checkpoints) and prediction (e.g., videos with detected bounding boxes).

Setup

To set up the project, first clone the repository and then install the required Python dependencies:

pip install -r requirements.txt

Usage

Training the Model

To train the YOLO model, run the train.py script:

python train.py

Training progress and results will be saved in the runs/train/ directory.

Evaluating and Predicting with the Model

To evaluate the trained model on the test set and run predictions on a video, execute the test.py script:

python test.py

The evaluation metrics will be printed to the console. A new video with detected cheetahs will be saved in the runs/detect/predict/ folder.

Results

• Training results: Detailed logs, metrics, and model checkpoints can be found in runs/train/.
• Prediction videos: Videos with bounding box detections are saved in runs/detect/predict/.