Fingerprint identification method and device, terminal and storage medium

Abstract

The embodiment of the invention discloses a fingerprint identification method and device, a terminal and a storage medium. The method comprises the following steps: acquiring an original fingerprint image; performing image enhancement processing on the original fingerprint image to generate a first image; determining an end point and a cross point of a fingerprint ridge line on the first image; determining an inflection point on the fingerprint ridge line based on the end point and the cross point; performing feature dimension reduction on the endpoints, the cross points and the inflection points on a second image to obtain feature vectors after dimension reduction; the second image is obtained by performing smooth denoising processing on the original fingerprint image; and performing fingerprint identification based on the feature vector. According to the scheme, the inflection points of the fingerprint ridge lines are added to be combined with the end points and the cross points, the fingerprint features are enriched, the fingerprint features are more reasonable, dimensionality reduction is conducted on the inflection points, the end points and the cross points, and therefore the internal storage can be reduced, and the comparison speed during fingerprint recognition can be increased.

Description

Technical field [0001] The present invention relates to the field of fingerprint recognition technology, in particular to a fingerprint recognition method, apparatus, terminal and storage medium. Background [0002] Fingerprint recognition is mainly a process of extracting key feature information in fingerprint images, describing and registering feature information. [0003] At present, fingerprint recognition, especially small fingerprint recognition, is used in many fields, and in order to improve accuracy, the amount of information extracted by small area array fingerprint image features is usually large, which leads to a longer time. Specifically, in the current field of small area array fingerprint registration, there are mainly extreme points, ridge lines, detail points and other registration methods. Extreme point registration extracts more information, the accuracy of matching is high, but the time and memory consumption is larger, and there are certain requirements for ...

AI Technical Summary

Problems solved by technology

Extremum point registration extracts a lot of information, and the matching accuracy is high, but it consumes a lot of time and memory, and has certain requirements for equipment; the ridge feature describes the texture feature of the fingerprint, but lacks the grayscale information of the original fingerprint image; The features of the minutiae points are relatively accurate, but the number of features of the minutiae points on the small array is small, which affects the final accuracy

Images