Python YOLO Course ~ YOLO-Based Autonomous Object Detection System
Below is a complete syllabus for 10 sessions @1.5 hours each covering the following topics:
YOLO-Based Autonomous Object Detection System for Unmanned Aerial Vehicle (UAV) Navigation
Session 1: Introduction to UAV and Automatic Object Detection
-
Introduction to UAV systems and their applications
-
Role of object detection in autonomous navigation
-
Basic concepts of YOLO (You Only Look Once)
-
Case studies of YOLO applications on drones
-
Tools: Google Colab, PyTorch, OpenCV
Session 2: YOLO Architecture and Its Variants (YOLOv4 – YOLOv8)
-
Evolution of YOLO from v1 to v8
-
Differences between YOLOv4, YOLOv5, YOLOv7, YOLOv8
-
Advantages of YOLOv8 for real-time UAV navigation
-
Installation and environment setup
Session 3: Dataset and Annotation for UAV Object Detection
-
Dataset types: aerial imagery, drone footage
-
YOLO format (txt) and bounding box annotation
-
Exercise: Image annotation with Roboflow / LabelImg
-
Introduction to object classes (humans, vehicles, obstacles)
Session 4: Training YOLOv5/YOLOv8 on Aerial Dataset
-
Data preprocessing
-
Configuration of training files and models
-
Exercise: Training model with aerial dataset
-
Evaluation: Precision, recall, mAP
Session 5: Inference and Visualization of Detection on Images and Videos
-
Load YOLO model and test on UAV images
-
Test model on drone video
-
Real-time performance analysis
-
Plot bounding boxes with OpenCV
Session 6: Integration of YOLO with UAV System (Simulation)
-
Introduction to ArduPilot, PX4, or AirSim (simulation)
-
How to feed UAV camera to YOLO detection model
-
Exercise: UAV Simulation + Object detection with AirSim + YOLO
