SIP12

Identifying Insect Pests in Leafy Green Vegetables Using AI Image Recognition (BugFinderAI)

Partners involved: Smart Farm Robotix (ADS developer) / Bio Contact OOD – Sofina Organic Farm (End-user)

Sector: Vegetable Production / Pest and Disease Detection

Country of the Pilot: Bulgaria

About the SIP

BugFinderAI aims to revolutionize pest detection in leafy green vegetables through AI-powered image recognition and autonomous field robotics. The pilot, implemented at Sofina Organic Farm near Sofia, focuses on early and precise identification of cabbage aphids (Brevicoryne brassicae), one of Europe’s most destructive pests affecting crops like cabbage, broccoli, and kale.

The SIP builds on Smart Farm Robotix’s high-TRL robotic platform, RoboAiWeeder, enhancing it with pest-recognition capabilities based on deep learning. Using advanced neural networks, the system analyzes high-resolution images of plant leaves to detect early infestation signs. This approach enables preventive, rather than reactive, pest management and supports small and medium-sized farms in transitioning to more data-driven, sustainable farming practices.

 

The Problem

Aphid infestations cause annual crop losses worth hundreds of millions of euros across Europe. Conventional monitoring relies on manual inspection, time-consuming, subjective, and often too late to prevent widespread damage.

Small farmers, particularly in organic production, need reliable and affordable digital tools to detect pest outbreaks early and target treatment only where needed. However, connectivity issues and high costs of drone or laboratory-based systems limit adoption in rural areas.

The Solution

BugFinderAI delivers an integrated robotic pest-detection system that operates in both cloud-based and mixed (edge/cloud) modes, combining robotics, AI, and open-source services:

  • Cloud-based ADS – The autonomous robot scans crops using RGB and multispectral cameras mounted on a robotic arm. Images are uploaded to the cloud, where a convolutional neural network (CNN) identifies aphid presence. Data from OpenAgri’s Weather Data and Pest & Disease Management Services are integrated to generate a pest-risk index, visualized through automated field heat maps and reports.

  • Mixed/Edge-based ADS – In connectivity-limited conditions, the Jetson Nano onboard computer performs local AI inference, sending only summarized pest-risk values to the cloud via the OpenAgri Gatekeeper once a connection is available. This setup ensures real-time results and significantly reduces data transmission and energy consumption.

The ADS integrates five OpenAgri OS ServicesSemantic Model, Gatekeeper, Pest & Disease Management, Weather Data, and Reporting Service—to ensure interoperability, secure data flow, and actionable insights.

By enabling early, accurate, and automated pest detection, BugFinderAI reduces pesticide use, prevents production losses, and saves valuable labor time, helping organic and conventional farmers alike make smarter, more sustainable decisions.

 

© 2026 OpenAgri

Project Coordination:

Prof. Christopher Brewster
Maastricht University

Minderbroedersberg 4-6,
6211 LK Maastricht,
Netherlands

christopher.brewster@

maastrichtuniversity.nl

Project Communication:

Maja Radisic
Foodscale Hub
Trg Dositeja Obradovića 8
21000 Novi Sad,
SERBIA
maja@foodscalehub.com
 
foodscalehub.com

Follow Us:

Subscribe:

OpenAgri — Where Inclusive Innovation Meets Agriculture!

OpenAgri has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101134083. This output reflects only the author’s view and the European Commission cannot be held responsible for any use that may be made of the information contained therein.
Fill the form

Drop us a line

Fill in this form or send us an e-mail with your inquiry.

Or come visit us at:

301 Howard St. #600
San Francisco, CA 94105