Luminescent imaging techniques, utilizing optical chemical sensors based on fluorescence, have been the subject of much attention in recent years. In aerodynamic research, point measurements of a given quantity over the surface of a test article are widely utilized as the most straightforward and conventional approach. Although point measurements have several advantages such as sensitivity, linearity, and mechanical robustness, the invasiveness of many transducers creates additional disturbances in the observed flow. By contrast, luminescent imaging, which allows for high spatial resolution, non-intrusiveness, and remote sensing capabilities, requires an improvement in sensitivity to capture minute signal fluctuations.A differential luminescent imaging method is developed to detect and resolve small fluctuations in an unsteady measurement. The target resolution is fluctuations less than 1.0 % of the mean component of the measurement quantity. With existing luminescent imaging techniques, it is still challenging to capture such fluctuations because the mean value occupies most of the dynamic range of the measurement output. An interdisciplinary approach using photochemistry, electrical engineering, and optical diagnostics has been used to accomplish this measurement goal.A photophysical model is derived in this study to describe the developed luminescent imaging method. A point-based measurement system is developed to validate the static and dynamic characteristics of the model. As an illustrative application of the differential luminescent imaging, an acoustic field in a resonance box, which creates spatially distributed minute pressure fluctuations, is used. The developed point-based measurement system captures pressure information by traversing over the acoustic field and capturing consecutive point measurements. A fluctuation of 0.4 % of the mean pressure is resolved with a spatial resolution of 1.5 mm in diameter and temporal resolution of 100 kHz. Uncertainty analysis of the differential luminescent imaging method is given based on a linear regression, photodegradation, and statistical propagation of noise.