In recent years, the development of deep learning technology has significantly benefited agriculture in domains such as smart and precision farming. Deep learning models require a large amount of high-quality training data. However, collecting and managing large amounts of guaranteed-quality data is a critical issue. To meet these requirements, this study proposes a scalable plant disease information collection and management system (PlantInfoCMS). The proposed PlantInfoCMS consists of data collection, annotation, data inspection, and dashboard modules to generate accurate and high-quality pest and disease image datasets for learning purposes. Additionally, the system provides various statistical functions allowing users to easily check the progress of each task, making management highly efficient. Currently, PlantInfoCMS handles data on 32 types of crops and 185 types of pests and diseases, and stores and manages 301,667 original and 195,124 labeled images. The PlantInfoCMS proposed in this study is expected to significantly contribute to the diagnosis of crop pests and diseases by providing high-quality AI images for learning about and facilitating the management of crop pests and diseases.

Quick disease diagnosis is crucial in agriculture to reduce crop yield losses. Deep learning techniques have already been successfully applied to several tasks in agriculture, but it requires sufficient datasets. Obtaining sufficient plant disease images is challenging due to seasonality and a shortage of experts. Generative Adversarial Networks (GANs) can address this issue by generating synthetic images, but existing methods struggle with complex diseases like fire blight, where symptoms are localized. To overcome this problem, we propose LITP-GAN, a novel image-to-image translation pipeline considering the lesion area. LITP-GAN includes three stages: (Stage 1) Lesion area detection using DEtection TRansformer (DETR) to detect lesion areas, (Stage 2) Image translation via Global and Local Alignment Networks (GLA-Net) to generate diseased leaf images, and (Stage 3) Post-Processing using Seamless Cloning method for generating realistic images by minimizing visual inconsistencies. We tested 400 healthy leaf images, and 236 generated images were identified as realistic diseased images. Applying these to a real field plant disease detection model showed significant improvements: 7.57% in precision, 4% in recall, and 3.5% in mAP compared to models trained on the original datasets. These findings suggest that LITP-GAN-augmented datasets can improve real field plant disease diagnosis, providing a promising tool for agricultural applications.