As the optical fingerprint that reflects the physiological states of skin, the accurate acquisition of skin spectral reflectance iscrucial for pathological diagnosis, ensuring digital imaging fidelity, and enabling the quantitative evaluation of cosmetic efficacy.This paper surveys the advancements in skin spectral reflectance measurement techniques and their applications. Regardingmeasurement methods, spectrophotometry (250–2500 nm) delivers high-precision single-point data, while hyperspectral im-aging (400–1000 nm) enables spatially resolved spectral analysis for applications like skin cancer detection, albeit with highercosts and lower imaging efficiency. Meanwhile, computational spectral imaging leverages RGB reconstruction to achieve highspatial resolution, though metamerism and generalizability limitations remain unresolved. In the field of skin reflectance appli-cations, current efforts focus on three domains of medical diagnostics (e.g., dermatitis grading and blood oxygen monitoring),digital imaging (e.g., skin tone correction), and cosmetic technology (e.g., dark circle assessment, efficacy testing of cosmetics,and beauty device development). Despite these innovations, limitations such as measurement errors on uneven skin, high equip-ment costs, hardware constraints, and operational inconsistencies in skin tone calibration continue to pose significant barriers.To overcome these limitations, we proposed leveraging compressed sensing for efficient hardware-level data acquisition andphysics-constrained meta-learning algorithms to enhance generalizability. This survey offers insights for advancing skin spec-tral measurement technologies and their analytical applications.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)