GANet: A Lightweight Group Attention Network for Plant Disease Identification on Embedded Systems

Title: GANet: A Lightweight Group Attention Network for Plant Disease Identification on Embedded Systems

标题：GANet：一种用于嵌入式系统植物病害识别的轻量级群注意力网络

Abstract: Global warming and climate change are major catalysts for plant diseases, significantly affecting food production worldwide. Addressing this issue requires continuous monitoring and early detection of plant diseases.

摘要：全球变暖和气候变化是植物病害的主要诱因，严重影响了全球粮食生产。解决这一问题需要对植物病害进行持续监测和早期发现。

While recent advancements in deep learning offer potential solutions, most models demand high processing power and struggle to perform efficiently on low-end electronic hardware. Furthermore, many existing works focus on specific crop diseases, limiting their broader applicability.

尽管深度学习的最新进展提供了潜在的解决方案，但大多数模型需要高处理能力，且难以在低端电子硬件上高效运行。此外，许多现有研究仅关注特定的作物病害，限制了其更广泛的适用性。

This paper presents a lightweight Group Equivariance Attention Network (GANet) designed for the prediction and monitoring of 38 types of plant diseases. GANet introduces a novel Group Attention block and Group-Wise Feature Fusion block, making it both sparse and efficient.

本文提出了一种轻量级群等变注意力网络（GANet），旨在预测和监测 38 种植物病害。GANet 引入了新颖的群注意力模块和群组特征融合模块，使其既稀疏又高效。

With only 0.001M parameters, the model achieves an impressive 99.99% accuracy on the PlantVillage dataset. Additionally, GANet is optimized for deployment on edge devices such as the Raspberry Pi 4b and Jetson Nano, where it reaches 5 FPS and 11 FPS, respectively, demonstrating its potential for real-time plant disease monitoring on low-power devices.

该模型仅有 0.001M 个参数，在 PlantVillage 数据集上实现了 99.99% 的惊人准确率。此外，GANet 针对 Raspberry Pi 4b 和 Jetson Nano 等边缘设备进行了部署优化，分别达到了 5 FPS 和 11 FPS 的处理速度，展示了其在低功耗设备上进行实时植物病害监测的潜力。