An Intelligent Framework of Plant Disease Segmentation and Classification using Heuristic Approach-aided K-Means Clustering and Attention with Multi-Dilated DenseNet

M. Vani Pujitha and K. V. D. Kiran ()
Additional contact information
M. Vani Pujitha: Department of CSE, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh, India†Department of CSE, Velagapudi Ramakrishna Siddhartha Engineering College, Vijayawada, India
K. V. D. Kiran: Department of CSE, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh, India

Journal of Information & Knowledge Management (JIKM), 2025, vol. 24, issue 05, 1-50

Abstract: Accurately recognising and forecasting plant disease plays a significant role in the agricultural sector at an earlier stage. Thus, it enables timely intervention to avoid the propagation of disease, reduce crop losses and enhance productivity. Several deep learning techniques have emerged for segmenting and classifying plant diseases. However, the presence of noise in the images is highly affecting the image quality, leading to poor outcomes. To address these challenges, a deep learning-aided plant disease classification method is introduced. The key objective of the work is to classify the disease and reduce damage caused by plant diseases. In the initial phase, the images are garnered from the standard sources and provided to the segmentation phase. Here, the affected regions of the images are segmented through Optimised K-Means Clustering (OKMC). Here, the segmentation process can effectively handle a large number of datasets and also it can segment the Region Of Interest (ROI) from the complex background. During this phase, the parameters of the OKMC are optimally tuned using an Iterative-based Upgrading in Pine Cone Optimisation (IUPCO). IUPCO can effectively solve complex optimisation problems and help the developed technique achieve better performance in the classification stage. In this stage, the classification is carried out using an Attention-based Multi-Dilated DenseNet (AMD-DNet). This model aims to achieve precise disease classification outcomes. The attention mechanism in the proposed approach aids in focusing on the relevant features in the images. Further, the multiscale operation can capture the features at different scales and make the classification process more robust. Evaluation is conducted on the developed approach to validate its effectiveness in plant disease classification. The outcomes showcase the developed approach’s potential as a robust result in the field.

Keywords: Plant disease classification; abnormality segmentation; optimised K-means clustering; iterative-based upgrading in pine cone optimisation; attention-based multi-dilated DenseNet; parameter optimisation (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.worldscientific.com/doi/abs/10.1142/S0219649225500534
Access to full text is restricted to subscribers

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:wsi:jikmxx:v:24:y:2025:i:05:n:s0219649225500534

Ordering information: This journal article can be ordered from

DOI: 10.1142/S0219649225500534

Access Statistics for this article

Journal of Information & Knowledge Management (JIKM) is currently edited by Professor Suliman Hawamdeh

More articles in Journal of Information & Knowledge Management (JIKM) from World Scientific Publishing Co. Pte. Ltd.
Bibliographic data for series maintained by Tai Tone Lim ().