Cucumber mosaic virus (CMV) is one of the most destructive plant viruses, causing considerable reductions in tomato growth and yield. Early detection of CMV infection before the appearance of visible symptoms is essential for effective disease management. This study evaluated the capability of hyperspectral imaging for the early detection of CMV infection in tomato plants using spectral reflectance and vegetation indices. A completely randomized experiment with 12 healthy and 12 infected plants was conducted. Hyperspectral images were acquired daily for 14 days after inoculation within the 400–950 nm spectral range. Spectral reflectance and vegetation indices, including the NDVI, PRI, SIPI, REP, SAVI, PSND-CHA, PSND-CHB, and PSND-CAR, ARI, MTVI1 and MTVI2, and VI-GREEN were analyzed. The results showed that CMV infection caused clear changes in spectral reflectance, particularly in the red-edge and near-infrared regions. Differences between healthy and infected plants became detectable from the eighth day after inoculation, before visible symptoms appeared. Most chlorophyll-related indices decreased, whereas PSND-CAR increased in infected plants, reflecting physiological responses to viral stress. These findings demonstrate that hyperspectral imaging is a rapid, non-destructive, and effective technique for the early detection of CMV infection and has strong potential for precision agriculture and plant health monitoring.