PCDRN: A Light Patch Based Attention U-Net for Plant Disease Detection in Modern Agriculture using Deep Learning

DOI

10.2991/978-94-6239-616-6_69How to use a DOI?

Keywords

PCDRN; CEAU; CNN; U-Net

Abstract

Detecting plant diseases quickly is crucial for maintaining good crop yields and sustainable farming. Current approaches often depend on manual inspection or heavy convolutional neural networks that are computationally expensive and too large to run on IoT devices. To address these issues, this proposed work introduces PCDRN (Patch-based Channel Depth Refinement Network), an innovative lightweight enhanced Attention U-Net framework for plant disease detection in modern agriculture. This proposed work includes a Channel Excitation and Attention Unit (CEAU) to highlight disease-specific channels, Depth wise Separable Convolutions for efficient feature extraction, and a Patch sampling method for recognizing discriminatively miserable patches in leaf images. Multi-scale feature fusion has been rendered easier through the encoder-decoder architecture, which also ensures a low number of parameters with high inference efficiency. Findings from a dataset encompassing 19 distinct crop diseases reveal that PCDRN surpasses existing techniques with its efficient network architecture, attaining a test accuracy of 95.79% and a significantly reduced parameter size (0.97MB).

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - V. Asha Merlin
AU  - P. L. Chithra
PY  - 2026
DA  - 2026/03/31
TI  - PCDRN: A Light Patch Based Attention U-Net for Plant Disease Detection in Modern Agriculture using Deep Learning
BT  - Proceedings of the International Conference on Artificial Intelligence and Secure Data Analytics (ICAISDA 2025)
PB  - Atlantis Press
SP  - 927
EP  - 938
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-616-6_69
DO  - 10.2991/978-94-6239-616-6_69
ID  - Merlin2026
ER  -