A Multi-Scale Feature Extraction and Fusion Deep Learning Method for Classification of Wheat Diseases

2025-01-01

Journal of Computer Science

AI Summary

Accurate discrimination of morphologically similar wheat diseases—such as loose smut, leaf rust, and crown rot—remains challenging in field conditions. To address this, we propose a deep learning ensemble framework integrating multi-scale feature extraction with semantic segmentation. Our approach introduces a novel collaborative architecture combining Xception, InceptionV3, and ResNet50, augmented by pixel-level segmentation to enhance lesion localization. Furthermore, we design a dual-path ensemble strategy comprising both majority voting and stacking mechanisms. Evaluated on the publicly available 2020 wheat disease dataset, the framework achieves an overall classification accuracy of 99.75%, with the Xception submodel attaining state-of-the-art performance—surpassing existing methods significantly. This work delivers a highly robust, interpretable, and fine-grained diagnostic solution for in-field wheat disease identification.

Technology Category

• Machine Learning: Ensemble Methods
• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Wheat is an important source of dietary fiber and protein that is negatively impacted by a number of risks to its growth. The difficulty of identifying and classifying wheat diseases is discussed with an emphasis on wheat loose smut, leaf rust, and crown and root rot. Addressing conditions like crown and root rot, this study introduces an innovative approach that integrates multi-scale feature extraction with advanced image segmentation techniques to enhance classification accuracy. The proposed method uses neural network models Xception, Inception V3, and ResNet 50 to train on a large wheat disease classification dataset 2020 in conjunction with an ensemble of machine vision classifiers, including voting and stacking. The study shows that the suggested methodology has a superior accuracy of 99.75% in the classification of wheat diseases when compared to current state-of-the-art approaches. A deep learning ensemble model Xception showed the highest accuracy.

Problem

• Wheat Disease Identification
• Accuracy Improvement
• Plant Health Management

Innovation

• Deep Learning Ensembles
• Wheat Disease Recognition
• Advanced Image Analysis