Guava Disease Detection and Classification: A Systematic Literature Review

Guavas (Psidium guajava) are nutrient-rich fruits that provide significant health benefits. However, guava cultivation faces persistent threats from various diseases affecting both leaves and fruits, leading to substantial yield and quality losses. The early and accurate detection of these diseases is crucial but remains challenging due to economic constraints and limited infrastructure. While plant pathologists employ various diagnostic methods, these approaches are often time-consuming, costly, and sometimes inconsistent. Recent advancements in deep learning (DL) and machine learning (ML) have introduced innovative techniques for guava disease identification. This study conducts a Systematic Literature Review (SLR) to evaluate the existing research on guava leaf and fruit disease detection, focusing on dataset sources, identified disease categories, preprocessing and augmentation techniques, applied algorithms, and reported evaluation metrics. A comprehensive search was conducted across multiple databases, covering publications from 2017 to 2023, leading to the identification of 47 relevant studies. After applying exclusion criteria, 16 studies were selected for in-depth analysis. The findings highlight the most commonly used datasets, the predominant classification techniques, and the effectiveness of various deep learning models based on multiple performance metrics, providing insights into current research trends, existing limitations, and potential directions for future studies. This review serves as a valuable reference for researchers aiming to enhance the accuracy and efficiency of guava leaf and fruit disease diagnosis through data-driven approaches.