Summary of A Deep Learning-based Pest Insect Monitoring System For Ultra-low Power Pocket-sized Drones, by Luca Crupi et al.
A Deep Learning-based Pest Insect Monitoring System for Ultra-low Power Pocket-sized Drones
by Luca Crupi, Luca Butera, Alberto Ferrante, Daniele Palossi
First submitted to arxiv on: 2 Apr 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Robotics (cs.RO)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper presents a novel solution for detecting pests using miniaturized palm-sized drones in precision agriculture. The authors develop two ultra-low power System-on-Chips (SoCs) to run deep learning (DL) models, achieving accurate results while keeping memory and computational needs under an ultra-tight budget. They fine-tune and quantize DL models in 8-bit integers and deploy them on the SoCs. The FOMO-MobileNetV2 model runs at 16.1 frame/s within 498 mW on the STM32H74, while the SSDLite-MobileNetV3 model peaks at 6.8 frame/s within 33 mW on the GAP9 SoC. Compared to a full-precision baseline, the best model drops only 15% in mean average precision (mAP), paving the way for autonomous palm-sized drones capable of lightweight and precise pest detection. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about using tiny drones to help farmers detect pests quickly and efficiently. The drones can fly over crops to look for signs of pests, which helps prevent damage and saves time. To make this work, the authors developed special chips that can run complex computer programs while still being very small and energy-efficient. They tested these chips with two different models and found that they could detect pests just as well as more powerful systems, but using much less power. This technology could be used in the future to create autonomous drones that can help farmers protect their crops without needing human intervention. |
Keywords
» Artificial intelligence » Deep learning » Mean average precision » Palm » Precision
