摘要

In order to ensure the security and integrity of the authenticated user';s fingerprint data transmitted via Web and increasing the credibility of the automatic fingerprint identification system (APIS), an adaptive embedding algorithm is introduced that can hide the fingerprint minutiae data in a synthetic fingerprint image. Based on characteristics of human visual system (HVS), including texture/contract masking property and sensitivity of frequency, and combining with wavelet multi-resolution theory, a novel just noticeable difference (JND) model that can embed watermark adaptively is proposed. Experiment results show this algorithm is invisible and has good robustness.

全文