A portable plant physiological feature image processing technique for groundnuts rosette disease diagnosis

Abstract

The monitoring and early detection of groundnut diseases are crucial for effective crop management and disease control. However, existing methods such as DNA-based and serological tests suffer from time-consuming processes and expensive laboratory setups, making them impractical for remote site testing. Additionally, recognizing groundnut rosette disease based on visual characteristics alone is challenging and unreliable. To address these issues, this study aimed to develop a model for Groundnuts Rosette Disease Diagnosis Using Plant Physiological Feature yolov8 Image Processing Technique. The research objectives were three-fold. First, the study aimed to establish the requirements for a groundnut rosette disease detection model using plant physiological feature image recognition technique. Secondly, to design the groundnut rosette disease diagnosis model using plant physiological feature image recognition technique. Third, the performance of the model was evaluated in terms of accuracy, precision, recall, and F1-score, and a comparison was made with existing methods for groundnut rosette disease detection. The study contributed to the development of a mobile application that assists farmers in making quick decisions regarding Groundnut Rosette Disease management. This app leverages the YOLOv8 model to enable farmers to rapidly and accurately identify the disease based on images of their crops. By providing real-time diagnostic capabilities, the application empowers farmers to implement timely interventions, improving overall crop health and yield. Additionally, the app includes resources on best management practices, helping farmers understand how to mitigate the effects of the disease. User-friendly features make it accessible to farmers with varying levels of technical expertise. Ultimately, this tool enhances efficient crop management practices and supports sustainable agriculture by reducing losses associated with Groundnut Rosette Disease. By promoting informed decision-making, the app aims to improve food security and farmers' livelihoods. This study addressed challenges in diagnosing Groundnut Rosette Disease (GRD) by developing a portable YOLOv8 model using plant physiological feature image recognition. The research followed three key objectives: first, identifying critical physiological features— such as changes in leaf morphology, color, and texture—to establish model requirements; second, collecting diverse data across various groundnut growth stages to train the model to recognize disease-related variations; and third, evaluating the model’s performance using accuracy, precision, recall, and F1-score metrics. The model achieved perfect precision (100%) at a confidence threshold of 0.964, significantly surpassing existing methods reporting precision between 0.75 and 0.90. An F1-score of 0.80 at a confidence threshold of 0.454 demonstrated balanced and reliable disease detection. These results are vital for early and accurate diagnosis, enabling timely interventions that reduce yield losses. The portable YOLOv8 model enhances real-time disease diagnosis and empowers farmers to make informed decisions, contributing to better disease management and improved food security in affected regions. This study introduces a novel image processing technique for diagnosing Groundnut Rosette Disease using plant physiological features. The developed model provides a fast, accurate, and reliable tool to assist farmers in managing groundnut crops. The resulting mobile application empowers farmers to make informed decisions, enhancing crop productivity and disease control. Future work will focus on refining the model, incorporating additional features, and integrating with other agricultural technologies.