<![CDATA[The decline in productivity of paprika plants (Capsicum annuum) is often closely associated with leaf health disorders, particularly bacterial spot disease, which is difficult to identify at an early stage through conventional visual inspection. Early detection of this disease is crucial, as delayed diagnosis can lead to significant yield losses and reduced crop quality. However, manual monitoring relies heavily on expert knowledge and is prone to subjectivity and human error, especially in large-scale cultivation systems. To address this challenge, this study proposes a deep learning–based image classification approach for identifying diseases in paprika leaves by leveraging transfer learning with the MobileNetV2 architecture. MobileNetV2 was selected due to its lightweight structure and computational efficiency, making it suitable for practical and real-time agricultural applications. The dataset used in this research was obtained from the PlantVillage database and consists of two classes: healthy paprika leaves and leaves infected with bacterial spot disease. To enhance the robustness and generalization capability of the proposed model, the training data were enriched using various data augmentation techniques, including rotation, flipping, scaling, and brightness adjustment. These techniques help mitigate overfitting and improve the model’s ability to recognize disease patterns under diverse imaging conditions. Experimental results demonstrate that the proposed model achieves stable and reliable classification performance, with an overall accuracy of 96%, accompanied by balanced precision and sensitivity values across both classes. These results indicate that MobileNetV2 is highly effective for paprika leaf disease classification. Furthermore, the findings suggest strong potential for implementing the proposed approach as an image-based plant disease detection system, supporting precision agriculture and enabling early intervention to improve crop productivity and sustainability.]]>
