Related Experiment Video
Updated: Oct 30, 2025
06:22
A Precise and Autonomous System for the Detection of Insect Emergence Patterns
Published on: January 9, 2019
5.9K
A System Using Artificial Intelligence to Detect and Scare Bird Flocks in the Protection of Ripening Fruit
Petr Marcoň1, Jiří Janoušek1, Josef Pokorný1
1Faculty of Electrical Engineering and Communication, Brno University of Technology, 61600 Brno, Czech Republic.
Sensors (Basel, Switzerland)
|July 2, 2021
Summary
This study introduces an AI-powered bird detection system using convolutional neural networks to protect fruit crops. The system triggers bird scaring devices only when flocks are detected, improving effectiveness and reducing habituation.
Area of Science:
- Agricultural Science
- Computer Science
- Artificial Intelligence
Background:
- Bird flocks cause significant damage to ripening fruit crops.
- Traditional bird scaring methods often become ineffective due to bird habituation.
Purpose of the Study:
- To develop an intelligent system for detecting bird flocks in agricultural areas.
- To trigger bird scaring actuators dynamically, only when necessary, to enhance crop protection.
Main Methods:
- Utilized videocameras and a differential algorithm to identify moving objects in vineyards.
- Trained a convolutional neural network (CNN) using labeled image data for flock detection.
- Implemented the detection algorithm on a microcomputer module with an integrated videocamera.
Main Results:
- The system accurately detects bird flocks using a trained CNN.
- The system triggers a wireless signal to activate scaring actuators upon detection.
- Algorithm performance was assessed using precision, recall, and F1 score metrics.
Conclusions:
- The developed AI system offers a more effective solution for bird-pest management in agriculture.
- Dynamic triggering of scaring devices minimizes habituation and maximizes crop protection efficiency.
