GanSystem Weed Detection MVP

GanSystem Weed Detection MVP v1.0

Initial public release of the GanSystem Weed Detection MVP, an AI-powered computer vision project designed to identify weeds and bean crops in agricultural images.

Features

• Weed and bean crop detection
• Roboflow dataset preparation
• YOLO-based computer vision workflow
• Image preprocessing and augmentation
• Model performance evaluation
• Live detection Python script
• Agricultural AI use case documentation

Dataset

• 213 agricultural images
• 2,334 annotated objects
• Classes: Bean and Weed

Model Performance

• mAP@50: 65.1%
• Precision: 42.2%
• Recall: 81.8%
• F1 Score: 55.7%

Technologies

• Python
• OpenCV
• Roboflow
• YOLO
• Computer Vision
• Machine Learning

Repository Purpose

This release showcases the MVP phase of an agricultural AI system built to support smart farming, weed detection, and automated farm monitoring workflows.