YOLOv8 & YOLO11: Phát hiện Đối tượng Tùy chỉnh & Ứng dụng Web 2025

YOLOv8 and YOLO11: Cutting-Edge Object Detection Models

YOLOv8 and YOLO11 are the latest state-of-the-art object detection models from Ultralytics, surpassing previous versions in both speed and accuracy. These models build upon the advancements of earlier YOLO versions, introducing significant architectural and training improvements, making them versatile tools for a variety of computer vision tasks.

The YOLOv8 and YOLO11 models support a wide range of applications, including object detection, instance segmentation, image classification, pose estimation, and oriented object detection (OBB).

Course Structure

This course is divided into two parts:

Part 1: YOLOv8

1. Introduction to YOLO

   • Overview of CNN, RCNN, Fast RCNN, Faster RCNN, Mask R-CNN

   • Introduction to YOLOv8

   • Comparison of YOLOv8 and YOLOv7 with a focus on License Plate Detection

2. Running YOLOv8

   • Setting up YOLOv8 on Windows

   • Using YOLOv8 in Google Colab

3. Dataset Preparation

   • How to find datasets

   • Data annotation, labeling, and automatic dataset splitting

4. Training YOLOv8

   • Train/ Fine-Tune YOLOv8 Model on a Custom Dataset

   • Custom Projects:

     • Potholes Detection

     • Personal Protective Equipment (PPE) Detection

     • Pen and Book Detection

5. Multi-Object Tracking

   • Introduction to Multi-Object Tracking

   • Implementing YOLOv8 with the DeepSORT algorithm

     • Running on Google Colab

   • Traffic Analysis:

     • Vehicle counting and car velocity calculation

     • Vehicle entry and exit counting

   • YOLOv8 Segmentation with Tracking

6. Advanced Applications

   • Potholes Segmentation

   • Traffic Lights Detection and Color Recognition

   • Cracks Segmentation

   • Helmet Detection and Segmentation

   • Automated Vehicle Direction Detection and Counting

   • Face Detection, Gender Classification, Crowd Counting, and Tracking

   • License Plate Detection and Recognition with EasyOCR

   • Object Blurring with Object Tracking

   • Vehicle Segmentation, Counting, and Speed Estimation

7. Web Integration

   • Integrating YOLOv8 with Flask

   • Creating a Complete Web App for PPE Detection

Part 2: YOLO11

1. What's New in YOLO11

   • Key updates and features in Ultralytics YOLO11

2. Using YOLO11 for Various Tasks

   • Object Detection, Instance Segmentation, Pose Estimation, and Image Classification on Windows/Linux.

3. Model Performance Evaluation

   • Testing and analyzing YOLO11 performance

4. Training YOLO11

   • Train/ Fine-Tune YOLO11 Object Detection Model on a Custom Dataset for Personal Protective Equipment Detection.

   • Train/ Fine-Tune YOLO11 Instance Segmentation Model on a Custom Dataset for Potholes Detection.

   • Fine-Tune YOLO11 Pose Estimation Model on a Custom Dataset for Human Activity Recognition.

   • Train/ Fine-Tune YOLO11 Image Classification Model on a Custom Dataset for Plant Classification.

5. Advanced Multi-Object Tracking

   • Implementing Multi-Object tracking with Bot-SORT and ByteTrack algorithms

6. Specialized Projects

   • License Plate Detection & Recognition with YOLO11 and EasyOCR

   • Car and License Plate Detection & Recognition with YOLO11 and PaddleOCR

7. Web Integration with YOLO11

   • Integrating YOLO11 with Flask to build a web app

   • Creating a Streamlit Web App for object detection