THE OBJECTIVE OF THIS STUDY WAS TO DETERMINE HOW ENVIRONMENTAL VARIABILITY INFLUENCES THE DETECTION OF DISEASES OF GRAPE

Abstract

A variety of fungal, bacterial and viral diseases are very common in grape production. In recent times, image-based Artificial Intelligence (AI) and Machine Learning (ML) have emerged as potent tools for speedy and automat- ed disease detection. Unfortunately, such models are highly sensitive to envi- ronmental variation, and the accuracy and reliability of models is heavily de- pendent on environmental differences. Detection performance is influenced by visual symptom, disease progression, image quality, which is affected by fac- tors like temperature, humidity, light intensity, rainfall and seasonal changes. This review aims to comprehensively summarize the impact of environmental factors on grape disease development and the impact on conventional and deep learning–based detection methods. It also highlights progress in sensor-enables systems, multi-modal models, and environment-aware AI frameworks. The study points out difficulties such as low-light imaging, fluctuating humidity, and symptom variation due to climatic factors. Lastly, it outlines the research gaps and suggests future research directions for building the more robust and environment-adaptive grape disease detection systems.