Enhanced-RICAP: A Novel Data Augmentation Strategy for Improved Deep Learning-Based Plant Disease Identification and Mobile Diagnosis

Mamadou Bailo Diallo¹Yue Li^1*Okafor Sylevester Chukwuka¹Solomon Boamah¹Yuhong Gao¹Mohamed Meyer Kana Kone²Gelebo Rocho¹

• ¹Gansu Agriculture University, Lanzhou, China
• ²Nanchang University, Nanchang, China

Plant diseases pose significant threats to food security and agricultural productivity worldwide. Accurate and timely disease identification is crucial for effective crop management and minimizing economic losses. This study introduces Enhanced-RICAP, a novel data augmentation technique that improves the identification accuracy of deep learning models for plant disease identification. Enhanced-RICAP leverages class activation maps to guide the augmentation process, ensuring the preservation of discriminative regions while introducing sufficient variability. The proposed method was evaluated using several benchmark deep learning architectures, including RNeT18, RNeT34, RNeT50, EfficientNetb, and Xception, on two datasets: the cassava leaf disease dataset and the PlantVillage tomato leaf disease. Experimental results demonstrate that Enhanced-RICAP consistently outperforms existing augmentation techniques such as CutMix, MixUp, CutOut, Hide-and-Seek, and RICAP across key evaluation metrics, including accuracy, precision, recall, and F1-score. The RNeT18+Enhanced-RICAP configuration achieved an accuracy of 99.86% on the tomato leaf disease dataset, while the Xception+Enhanced-RICAP model attained 96.64% accuracy in classifying four cassava leaf disease categories. To facilitate practical application, the RNeT18+Enhanced-RICAP model was integrated into a mobile application called PlantDisease, which enables real-time, on-site disease identification and provides prevention and treatment recommendations. Sample et al. Running Title By empowering farmers and agricultural professionals with accessible and accurate diagnostic tools, this study contributes to sustainable crop management practices and strengthens food security.