ORIGINAL RESEARCH article
Front. Plant Sci.
Sec. Sustainable and Intelligent Phytoprotection
Volume 16 - 2025 | doi: 10.3389/fpls.2025.1691415
This article is part of the Research TopicSmart Sensing in Plant Science: Advancing Plant-Environment Interactions for Sustainable PhytoprotectionView all articles
Intelligent Smart Sensing with ResNet-PCA and Hybrid ML–DNN for Sustainable and Accurate Plant Disease Detection
Provisionally accepted
- 1Jouf University, Sakakah, Saudi Arabia
- 2Prince Sattam bin Abdulaziz University, Al Kharj, Saudi Arabia
- 3University of Tabuk, Tabuk, Saudi Arabia
- 4Zhejiang A and F University, Hangzhou, China
- 5Jiaxing University, Jiaxing, China
- 6Lovely Professional University, Phagwara, India
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
Diseases of plants remain one of the greatest threats to sustainable agriculture, with a direct 3 adverse effect on crop productivity and threatening food security worldwide. Conventional 4 detection methods rely heavily on manual detection and laboratory analysis, which are time-5 consuming, subjective, and unsuitable for large-scale monitoring. The use of the most recent 6 progress in computer vision and artificial intelligence has opened up a prospect of automated, 7 scalable, and precise disease diagnosis. This paper introduces a feature-efficient hybrid model 8 that trains classical ML classifiers with DNNs using ResNet-based feature extraction and PCA. 9 The PlantVillage dataset with mixed crop-disease pairs is used to implement and thoroughly test 10 five hybrid models. Wide-ranging experiments proved that the LR+DNN hybrid resulted in the 11 best classification accuracy of 96.22% as compared to other models and available benchmarks. 12 Besides being able to outperform other techniques in terms of predictive power, the framework 13 displayed good training stability and robustness to class imbalance as well as a higher degree of 14 interpretability based on LIME-based analysis. The obtained results confirm the hybrid ML+DNN 15 paradigm as a safe, non-secretive, scalable recognition solution when applied to plant diseases. 16 Providing opportunities for timely and accurate disease detection, the proposed framework 17 can help with precision agriculture, where pesticide use can be reduced, consequently, and a 18 significant contribution to sustainable farming can be achieved.
Keywords: Hybrid Machine Learning–Deep Learning, Intelligent Smart Sensing, Sustainable Disease Detection, ResNet FeatureExtraction, principal component analysis (PCA)
Received: 23 Aug 2025; Accepted: 09 Oct 2025.
Copyright: © 2025 Almadhor, Alsubai, Al Hejaili and Gadekallu. 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: Ahmad Almadhor, aaalmadhor@ju.edu.sa
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.