A Deep Learning-based Pest Insect Monitoring System for Ultra-low Power Pocket-sized Drones

TL;DR

This paper presents an ultra-low power, miniaturized drone system using optimized deep learning models for pest detection, enabling efficient, real-time crop monitoring within strict hardware constraints.

Contribution

It introduces a dual-SoC system deploying quantized deep learning models for pest detection, achieving high accuracy and efficiency on ultra-low power, palm-sized drones.

Findings

Achieved 0.66 mAP with 16.1 fps on STM32H74 at 498 mW.

Achieved 0.79 mAP with 6.8 fps on GAP9 at 33 mW.

Compared to baseline, models use 15 ext% less mAP but significantly reduce memory and computation.

Abstract

Smart farming and precision agriculture represent game-changer technologies for efficient and sustainable agribusiness. Miniaturized palm-sized drones can act as flexible smart sensors inspecting crops, looking for early signs of potential pest outbreaking. However, achieving such an ambitious goal requires hardware-software codesign to develop accurate deep learning (DL) detection models while keeping memory and computational needs under an ultra-tight budget, i.e., a few MB on-chip memory and a few 100s mW power envelope. This work presents a novel vertically integrated solution featuring two ultra-low power System-on-Chips (SoCs), i.e., the dual-core STM32H74 and a multi-core GWT GAP9, running two State-of-the-Art DL models for detecting the Popillia japonica bug. We fine-tune both models for our image-based detection task, quantize them in 8-bit integers, and deploy them on the two SoCs. On the STM32H74, we deploy a FOMO-MobileNetV2 model, achieving a mean average precision (mAP) of 0.66 and running at 16.1 frame/s within 498 mW. While on the GAP9 SoC, we deploy a more complex SSDLite-MobileNetV3, which scores an mAP of 0.79 and peaks at 6.8 frame/s within 33 mW. Compared to a top-notch RetinaNet-ResNet101-FPN full-precision baseline, which requires 14.9x more memory and 300x more operations per inference, our best model drops only 15\% in mAP, paving the way toward autonomous palm-sized drones capable of lightweight and precise pest detection.