Birdy - Final Year Project

Mango Orchards Monitoring Using Remote Sensing Data

In my Final Year Project I dedicated my efforts to revolutionize the agricultural sector in Pakistan through the integration of cutting-edge technology. The project addressed challenges in traditional tree counting methods using Drone Imagery and an AI-powered system for automated tree detection, geo-tagging, and classification. Boosting efficiency and accuracy. Following are some of the highlights of the project:

• Detected, located, and classified Mango Trees using raw drone imagery.
• Processed raw drone images using the NodeODM API.
• Annotated images, totalling approximately 5000 labels.
• Trained and Tested different Computer Vision algorithms.
• Achieved a high mAP of 0.94 for accurate localization and identification of mango trees.
• Attained 97.8% accuracy in health classification and 70% in variety classification.
• Designed a user friendly interface for seamless interaction.

Technologies Used: React, Flask, Postgres, Docker, OpenDroneMap, Google Colab, Makesense, Roboflow.


• 1st Position in Final Year Project among 100+ projects.  🏆
• 3rd Position for Poster Design for my Final Year Project.

View Demo
System Architecture