Digital image processing in determining the degree of turnip yellows virus infection in oilseed rape

In agricultural crops, leaves play crucial roles in plant physiology, providing information on both the quantity and quality of future production. They also serve as indicators for diagnosing foliar diseases. Detecting plant diseases through image reading and computer vision in a specific format allows for both quantitative and qualitative studies, which is a key step in diagnosing diseases using images of leaves exhibiting symptoms specific to a pathogen. In this study, we present an approach that integrates digital image processing to assess the extent of turnip yellows virus (genus Polerovirus, family Solemoviridae) infection in oilseed rape. This study focuses on the early detection and classification of the pathogen, as well as the identification of leaf surfaces affected by the turnip yellows virus. This allows for the evaluation of the severity of the infection and its impact on plant productivity together with the development of a pathogen control strategy that will benefit agricultural farmers. Thus, this experiment serves as a benchmark for future research in this field. The results show that the proposed method, based on color parameters in the L*a*b* color space and Lp-norms, accurately classifies the severity of viral attacks, enabling the precise evaluation of crop health. This method can be applied in real-time crop health monitoring, scientific research to assess the effectiveness of phytosanitary treatments, the testing of disease resistance, and targeted intervention strategies. It will help reduce crop losses and support sustainable agricultural practices.