ORIGINAL RESEARCH article
Front. Plant Sci.
Sec. Sustainable and Intelligent Phytoprotection
This article is part of the Research TopicIntegrating Visual Sensing and Machine Learning for Advancements in Plant Phenotyping and Precision AgricultureView all 9 articles
LMP-PM:a lightweight multi-path pruning method for plant leaf disease recognition
Provisionally accepted
- Jiangxi Agricultural University, 南昌市, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Plant leaf diseases pose a significant threat to plant growth and productivity, necessitating accurate and timely identification. In addressing the task of identifying plant leaf diseases, we have developed a novel lightweight approach named the Lightweight Multi-Path Pruning Method (LMP-PM). LMP-PM effectively reduces the complexity of the original model while enhancing its performance. LMP-PM offers flexible lightweight optimization, configurable via pruning parameters and path expansion ratios, to meet diverse task requirements. This enables users to balance significant reductions in model parameters and FLOPs against potential inference time increases, thereby tailoring model size, performance, and real-time needs to specific application scenarios. Specifically, we first constructed an original model (OMNet) that features high performance and complexity, incorporating various structures and a three-branch parallel module (TBP block) to optimize its performance. We then applied LMP-PM to perform lightweight processing on OMNet, resulting in several lightweight models. Through extensive experimentation, we identified the optimal model that balances performance and complexity, which we named LMNet (Lightweight Multi-Path Network). LMNet has only 5.69% of the parameters and 3.80% of the FLOPs of OMNet, achieving an accuracy of 99.23% on the Plant Village dataset, an improvement of 0.58% over OMNet; on the AI 2018 Challenger dataset, LMNet attained an accuracy of 87.27%, surpassing OMNet by 1.91%.
Keywords: Convolutional Neural Network, deep learning, Lightweigh, LMP-PM, Plant disease identification
Received: 01 Nov 2025; Accepted: 11 Feb 2026.
Copyright: © 2026 Hua, Zou, Zhu, Deng and He. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Jing Hua
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.