Detection of Common Tomato Leaf Diseases Using Fluorescence Spectroscopy Technique

Tomato leaf diseases pose significant challenges to the agricultural sector in Ethiopia, leading to substantial productivity losses. To address this issue, accurate disease detection and timely intervention are critical for effective management. In this study, we collected leaf samples from diverse growth conditions, including lighthouses and farm fields, at the Melkassa Agricultural Research Centre (MARC). In this research, the fluorescence spectrophotometer equipped with a xenon lamp as a source of light was employed for the detection of tomato leaf disease (bacterial leaf spot, powdery mildew, and early blight) based on unique fluorescence emission spectra characteristics. The fluorescence emission spectra were recorded at varying excitation wavelengths of 250, 350, 450, and 550 nm. The acquired emission spectra patterns served as distinct fingerprints, enabling the differentiation of healthy and diseased tomato leaf samples. The spectral properties of diseased tomato leaves show broadened, flattened, and low fluorescence intensity features. On the other hand, healthy leaves exhibit sharp and high fluorescence intensity. Principal component analysis (PCA) was employed to analyze the fluorescence emission spectra of healthy and diseased tomato leaves using the loading plots. PCA effectively distinguished between them, particularly at excitation wavelengths of 250 and 550 nm, achieving a classification accuracy of 90%. However, excitation wavelengths of 350 and 450 nm failed to accurately classify the tomato leaf samples using the first two principal components (PC1 and PC2).