Strawberry Plant Disease Classification for Smart Agriculture Using MobileNetV2

Plant disease detection and classification play a crucial role in maintaining agricultural productivity by enhancing crop yields and mitigating the impact of diseases. Strawberry disease poses a significant threat to the agricultural industry, often leading to substantial reductions in both crop quality and yield. Traditional detection methods, which rely on manual inspection, are limited by subjectivity, inconsistency, and a dependence on expert knowledge. To overcome these limitations, this work explores the potential of MobileNetV2, a lightweight convolutional neural network, in achieving its intended tasks for automated classification of strawberry plant diseases. MobileNetV2 is optimized for classification tasks, making it well-suited for real-time applications and deployment in resource-constrained environments. The model was trained and evaluated on a dataset comprising six disease classes, with data augmentation techniques employed to enhance generalization. Its performance was assessed in terms of classification accuracy, inference time, and computational efficiency. The results indicate that MobileNetV2 achieved a classification accuracy of 98.3%, demonstrating its strong potential for accurately identifying strawberry plant diseases.