Views
No views yet
pip install ultralytics opencv-python numpy1from ultralytics import YOLO
2
3# Load the trained model
4model = YOLO('models/yolov8m_stage2_improved_best.pt')
5
6# Run inference on an image
7results = model('path/to/image.jpg')
8results[0].show()
9
10# Process video
11results = model('path/to/video.mp4', save=True)python main.pymodels/yolov8m_stage2_improved_best.pt - Final model (recommended)models/yolov8m_stage1_smart_best.pt - Stage 1 model (for comparison)main.py - Complete vehicle detection and counting applicationexample_usage.py - Simple usage examplesrequirements.txt - Python dependenciestest_improved_model.bat - Windows testing scriptfinetune_dataset/images/ - 92 fine-tuning imagesfinetune_dataset/labels/ - Corresponding annotation filesfinetune_dataset/README.dataset.txt - Dataset informationfinetune_dataset/README.roboflow.txt - Roboflow export infodataset_configs/main_data.yaml - Main dataset configuration (8 classes)dataset_configs/finetune_data.yaml - Fine-tuning dataset configurationtraining_logs/stage2_results.png - Training results visualizationtraining_logs/stage2_confusion_matrix.png - Confusion matrixtraining_logs/stage2_results.csv - Detailed training metricstraining_logs/stage2_val_batch0_pred.jpg - Sample validation predictionstraining_runs/stage1_smart/ - Stage 1 training configuration and weights
args.yaml - Training argumentsweights/last.pt - Last epoch weightstraining_runs/stage2_improved/ - Stage 2 training configuration and weights
args.yaml - Training argumentsweights/last.pt - Last epoch weightsBoxF1_curve.png - F1 score curveBoxPR_curve.png - Precision-Recall curvelabels.jpg - Label distribution visualizationPROJECT_REPORT.md - Complete project documentationREADME.md - This file1from ultralytics import YOLO
2import cv2
3
4# Load the final model
5model = YOLO('models/yolov8m_stage2_improved_best.pt')
6
7# Detect vehicles in image
8results = model('highway_image.jpg')
9
10# Process results
11for result in results:
12 boxes = result.boxes
13 for box in boxes:
14 x1, y1, x2, y2 = box.xyxy[0]
15 conf = box.conf[0]
16 cls = int(box.cls[0])
17 class_name = model.names[cls]
18 print(f"Detected: {class_name} (confidence: {conf:.2f})")1# Process video with vehicle counting
2results = model('traffic_video.mp4', save=True, save_txt=True)
3
4# The main.py script provides advanced counting and tracking features1# Run the full application with counting and visualization
2from main import VehicleCounter
3
4counter = VehicleCounter()
5counter.process_video('input_video.mp4', 'output_video.mp4')1@misc{highway-vehicle-detection-code,
2 title={Highway Vehicle Detection - Code \& Models},
3 author={Nguyen Quoc Viet},
4 year={2025},
5 url={https://huggingface.co/bichuche0705/highway-vehicle-detection-code}
6}