Fingerprints have long been regarded as the gold standard for individual identification due to their uniqueness and stability. However, the increasing complexity of crime scenes and the advancement of anti-forensic techniques have significantly raised the difficulty of extracting complete and clear fingerprints, posing challenges to traditional identification methods relying on minutiae features. Starting from the limitations of existing fingerprint features (Level 1, Level 2, and Level 3), this paper discusses their shortcomings in donor attribution, individual identification, and donor analysis, proposing ridge curvature features as a supplementary method. Subsequently, it reviews the mathematical definition and computational models (geometric methods, IPAN99 algorithm) of ridge curvature, as well as research progress in computational science and forensic science. The potential of ridge curvature as an individualizing feature is demonstrated from three aspects: feature specificity, stability, and discriminative capability. Future research directions are suggested, including optimization of curvature calculation algorithms, stability analysis of distorted fingerprint features, and fidelity preservation under different development conditions, to facilitate the transition of this feature from theoretical exploration to practical forensic applications.