Deep Learning-Based Approach for Identification of Potato Leaf Diseases Using Wrapper Feature Selection and Feature Concatenation

2025-02-05

AI Summary

Addressing the challenges of early identification of potato early and late blight, as well as poor model robustness under limited training samples, this paper proposes an end-to-end deep learning diagnostic framework. First, histogram equalization is applied to enhance leaf image quality. Subsequently, a multi-layer convolutional neural network (CNN) extracts hierarchical features, which are concatenated into a comprehensive feature representation. Crucially, a wrapper-based feature selection strategy—specifically recursive feature elimination (RFE)—is integrated to jointly optimize and refine the deep feature subset. Finally, a support vector machine (SVM) performs classification. To the best of our knowledge, this is the first work to synergistically combine wrapper-based feature selection with deep CNN feature concatenation for potato disease recognition. Evaluated on a standard benchmark dataset, the method achieves 99% classification accuracy using only 550 selected features—substantially outperforming conventional approaches—and demonstrates markedly improved discriminative stability and generalization capability in low-data regimes.

Technology Category

• Machine Learning: Dimensionality Reduction/Feature Selection
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

The potato is a widely grown crop in many regions of the world. In recent decades, potato farming has gained incredible traction in the world. Potatoes are susceptible to several illnesses that stunt their development. This plant seems to have significant leaf disease. Early Blight and Late Blight are two prevalent leaf diseases that affect potato plants. The early detection of these diseases would be beneficial for enhancing the yield of this crop. The ideal solution is to use image processing to identify and analyze these disorders. Here, we present an autonomous method based on image processing and machine learning to detect late blight disease affecting potato leaves. The proposed method comprises four different phases: (1) Histogram Equalization is used to improve the quality of the input image; (2) feature extraction is performed using a Deep CNN model, then these extracted features are concatenated; (3) feature selection is performed using wrapper-based feature selection; (4) classification is performed using an SVM classifier and its variants. This proposed method achieves the highest accuracy of 99% using SVM by selecting 550 features.

Problem

• Identify potato leaf diseases
• Early detection to enhance yield
• Use image processing and machine learning

Innovation

• Deep CNN for feature extraction
• Wrapper-based feature selection
• SVM classifier for disease identification