Fingerprint minutiae detection method, device, apparatus, system and storage medium

Abstract

The invention discloses a fingerprint minutiae detection method, which comprises the following steps of: a collected fingerprint image is received; the fingerprint image is input to the pre-trained fingerprint minutiae detection model; a fingerprint minutiae detection model is a full convolution neural network based on convolution neural network training for fingerprint sample image; the minutiaefeature of fingerprint image is extracted by the downsampling unit in the minutiae detection model; the position information of minutiae is reconstructed by the upsampling unit in the minutiae detection model, and the minutiae position image is obtained. The method can detect and locate fingerprint minutiae with high precision, anti-jamming ability and strong robustness through full convolution neural network. The invention also discloses a fingerprint minutiae detection device, a system and a readable storage medium, which have the beneficial effects.

Description

technical field [0001] The invention relates to the field of biometric identification, in particular to a fingerprint minutiae detection method, device, equipment, system and a readable storage medium. Background technique [0002] In recent years, biometric technology based on fingerprint recognition has become more and more widely used, including fingerprint access control, fingerprint unlocking, etc. Excellent fingerprint recognition and matching algorithms are very important for information security and authentication. [0003] Fingerprint minutiae detection is the basis of fingerprint identification and fingerprint matching algorithms commonly used in the market, and it is also a basic premise of fingerprint structural matching algorithms. Currently commonly used fingerprint minutiae detection algorithms are based on traditional image processing algorithms (including image preprocessing, binarization, thinning, and feature minutiae extraction) on the collected fingerpri...

AI Technical Summary

Problems solved by technology

[0004] The processing process of this algorithm is complex, and the extraction results are easily affected by various factors such as fingerprint wrinkles, scratches, dry fingers or wet fingers, and image blur during the fingerprint image collection process. Under adverse factors, the anti-interference ability is weak, and the detected fingerprint details Point features have low accuracy

Images