Areca-CEDNet: Arecanut Plant Disease Classification Using Convolutional Encoder–decoder Network
Plant disease classification is crucial in agriculture for ensuring food security and sustainability.Traditional methods require expert knowledge, are time-consuming, and does not appropriate for large-scale analysis.Deep Learning (DL)-assisted image classification techniques offer effective performance in identifying plant diseases.This research proposes the arecanut leaf disease classification approach called Areca-Convolutional Encoder-Decoder Network (Areca-CEDNet) architecture, which builds on and extends the concepts of the CED-Net and the Temporal Convolutional Network (TCN).The study uses two datasets: a private Arecanut and the publicly available PlantVillage dataset, for implementing the proposed approach.A Visual Geometric Group-16 (VGG16) architecture extracts both low-and high-level features.Experimental results demonstrate that the proposed Areca-CEDNet method achieves an impressive accuracy of 99.93% on Arecanut dataset and 99.65% on PlantVillage dataset.This approach makes a significant contribution to sustainable Artificial Intelligence (AI) for plant disease detection by improving model generalization and allowing precise classification over diverse classes with smaller image sizes in DL-based agricultural applications.
Authors
- Raviprakash Madenur Lingaraju (ORCID: https://orcid.org/0000-0001-6149-7776)
- Balaji Prabhu Baluvaneralu Veeranna
- Niranjan Shadaksharappa Jayanna (ORCID: https://orcid.org/0009-0003-7736-1736)
- Pramod Kumar Palthady Monappa
- Kavitha Magadi Gopalakrishna
Institutions
- Visvesvaraya Technological University (IN)
Publication Details
- Journal
- International journal of intelligent engineering and systems
- Published
- 2026-11-17
- DOI
- https://doi.org/10.22266/ijies2026.1231.42
- Primary Topic
- Smart Agriculture and AI
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Visvesvaraya Technological University