Deep Neural Network Classifier for Detecting Cotton Plant Diseases Using Antlion Optimization

Abstract

Image processing techniques are used to autonomously identify agricultural plant diseases can significantly reduce reliance on farmers for safeguarding crop yields. The classification of cotton leaf diseases presents a formidable challenge. This study introduces a novel approach for classifying Cotton Leaf Diseases, employing an Antlion Optimization (ALO)-enhanced Deep Neural Network (DNN) classifier. The dataset comprises 10,000 images, a combination of directly captured farm field images and downloaded samples, encompassing normal leaves, bacterial blight, Anthracnose, Cercospora leaf spot, and Alternaria diseases. Preprocessing incorporates Wiener filtering to eliminate image noise, while Fuzzy Rough C-Means (FRCM) clustering is employed for diseased and normal portion segmentation. The ALO-augmented DNN achieves an impressive 93.37% accuracy in classifying cotton leaf diseases.

Plant phenotyping relevance

綿花葉の病徴を画像から分類する手法の開発・評価が研究の中心であり、植物の病害状態を直接推定するため、植物フェノタイピング手法として採用。

Code and data availability

The paper describes a cotton leaf disease image dataset (10,000 images) and an ALO-DNN pipeline, but contains no availability statement, deposit, or public URL for the dataset, images, code, or trained model. No paper-specific public asset is actionable.

No evidence-backed public reproduction asset is currently recorded.