Finger vein feature comparison method, device, storage medium and processor

Abstract

The invention discloses a finger vein feature comparison method, device, storage medium and processor. The method includes: preprocessing an input finger vein image; using a Gabor function to process the preprocessed image to obtain a Gabor vector feature ; Perform dimensionality reduction processing on the obtained Gabor vector features; perform binary Gabor gradient feature extraction on the Gabor vector features after dimensionality reduction: by calculating the gradient of the Gabor coefficient at the corresponding direction under different frequency parameters, and according to the positive and negative of the gradient value The symbol binarizes the gradient value into "0" and "1" values, and then places the corresponding value in a new vector corresponding to the component position indexed by the block coordinates, frequency, and direction parameters to generate a new vector feature of binary expression ; Perform feature comparison on the new vector features expressed in binary and output the comparison result. The invention obtains stable expression features and improves the performance of the finger vein recognition algorithm.

Description

technical field [0001] The present invention relates to the field of finger vein recognition, in particular to a finger vein feature comparison method, device, storage medium and processor. Background technique [0002] Finger vein recognition is a new biometric technology developed based on the characteristics of hemoglobin in blood that absorbs near-infrared rays. [0003] In the current biometric technology: Although the fingerprint technology is very mature, the fingerprint features are easy to be imitated, and the face is also easy to be imitated, so their security performance is lacking and needs to be improved, and the face recognition technology still exists The risk of unreliable results under large posture and large expression changes; iris recognition technology is stable, but its contact operation method is not accepted by a wide range of people and is not suitable for large-scale promotion; although voiceprint recognition has advantages in collection and applica...

AI Technical Summary

Problems solved by technology

[0003] In the current biometric technology: Although the fingerprint technology is very mature, the fingerprint features are easy to be imitated, and the face is also easy to be imitated, so their security performance is lacking and needs to be improved, and the face recognition technology still exists The risk of unreliable results under large posture and large expression changes; iris recognition technology is stable, but its contact operation method is not accepted by a wide range of people and is not suitable for large-scale promotion; although voiceprint recognition has advantages in collection and application, But first of all, the characteristics of individual voiceprints will change with age, emotion, and physical condition. Secondly, the quality of the collected voiceprint signals is closely related to the collection equipment. What's more, noise signals such as the background environment will seriously interfere with the voiceprint signals; the above-mentioned biological Although identification technologies have their own advantages, the limitations of their use should not be underestimated

Images