Weed Detection using YOLOv8

This project focuses on detecting weeds in agricultural fields using the YOLOv8 deep learning model. It consists of two versions:

• Version 6 (v6): Detects weeds vs crops with bounding boxes.
• Version 11 (v11): Detects 8 different types of weeds using a more detailed dataset.

---

Project Overview

Weed detection is crucial for precision agriculture, helping farmers to identify and selectively treat weeds, reducing herbicide usage and increasing crop yield. This project leverages YOLOv8 for fast and accurate object detection of weeds.

---

Data

• Version 6 Data: Crop and weed detection with bounding boxes.
  Source: Kaggle Dataset

• Version 11 Data: Dataset containing 8 weed types for multi-class weed detection.
  Source: Mendeley Data

---

Features

• Detects weeds and crops with bounding boxes (v6).
• Multi-class weed detection with 8 different weed types (v11).
• Trained and tested with YOLOv8 for state-of-the-art performance.
• Visualization of detection results with annotated bounding boxes.