CN-114529950-B - Finger vein recognition method, device, computer readable storage medium and apparatus

CN114529950BCN 114529950 BCN114529950 BCN 114529950BCN-114529950-B

Abstract

The invention discloses a finger vein recognition method, a finger vein recognition device, a computer readable storage medium and equipment, and belongs to the field of biological recognition. The finger vein feature extraction method comprises the steps of obtaining a finger vein image, calculating gradient information of the finger vein image, searching for a finger boundary by utilizing a greedy strategy according to the gradient information, detecting a finger joint on the finger vein image, extracting an effective region of a finger vein according to the finger joint and the finger boundary, extracting features of the effective region through a Gabor filter to obtain finger vein features, setting a sliding window on the finger vein image, comparing the similarity between the finger vein features of the finger vein image in the sliding window and corresponding positions of a finger vein feature template, finding out the finger vein image in the sliding window with the highest similarity as a maximum matching region, and taking the similarity between the maximum matching region and the corresponding positions of the finger vein feature template as a matching score. The finger vein identification method and the finger vein identification device can realize stable and efficient finger vein identification aiming at different equipment and different acquisition environments.

Inventors

ZHOU JUN
XU MENG

Assignees

北京眼神智能科技有限公司
北京眼神科技有限公司

Dates

Publication Date: 20260508
Application Date: 20201030

Claims (9)

1. A method of finger vein identification, the method comprising: acquiring a finger vein image to be detected; Calculating gradient information of the finger vein image, and searching for a finger boundary by using a greedy strategy according to the gradient information; detecting finger joints on a finger vein image, and extracting an effective area of the finger vein according to the finger joints and finger boundaries; Extracting features of the effective area through a Gabor filter to obtain finger vein features; setting a sliding window on the finger vein image, comparing the finger vein features of the finger vein image in the sliding window with the corresponding positions of the finger vein feature template, and finding out the finger vein image in the sliding window with the highest similarity as a maximum matching area; the similarity between the finger vein features of the finger vein image in the maximum matching area and the corresponding positions of the finger vein feature template is used as a matching score; calculating gradient information of the finger vein image, and searching for a finger boundary by using a greedy strategy according to the gradient information, wherein the calculating comprises the following steps: Performing scale reduction on the finger vein image to obtain a zoom image, calculating gradient information of the zoom image, and finding out a finger rough positioning boundary by using a greedy strategy according to the gradient information of the zoom image; performing rotation correction on the finger vein image according to the finger rough positioning boundary; and finding the finger boundary by using a greedy strategy according to gradient information of the finger vein image after rotation correction.
2. The method for recognizing finger veins according to claim 1, wherein said scaling down the finger vein image to obtain a scaled map, calculating gradient information of the scaled map, finding a finger rough positioning boundary by using a greedy strategy according to the gradient information of the scaled map, comprises: downsampling the finger vein image according to a scaling factor to obtain the scaling map; Projecting the finger vein image in the horizontal direction, and selecting a row with the largest projection value as a dividing line; Calculating gradient information of the zoom map, and dividing the gradient information of the zoom map into an upper half and a lower half by the dividing line; for the upper half part of gradient information of the zoom map, reserving pixel points with gradients in the vertical direction larger than a first screening threshold value, and setting the gradients in the vertical direction larger than a first cut-off threshold value as the first cut-off threshold value to obtain an upper boundary coarse positioning pixel point set; For the lower half part of gradient information of the zoom map, reserving pixel points with gradient in the vertical direction smaller than a second screening threshold value, and setting the gradient in the vertical direction smaller than a second cutoff threshold value as a second cutoff threshold value to obtain a lower boundary coarse positioning pixel point set; And searching a line segment with the largest accumulated gradient in one direction from the other direction in the upper boundary coarse positioning pixel point set and the lower boundary coarse positioning pixel point set by utilizing a greedy strategy respectively to obtain a coarse positioning upper boundary and a coarse positioning lower boundary which are used as the finger coarse positioning boundary.
3. The method of finger vein recognition according to claim 2, wherein the performing rotation correction on the finger vein image according to a finger rough positioning boundary comprises: determining a center point set according to the coarse positioning upper boundary and the coarse positioning lower boundary; fitting according to the center point set to obtain a center straight line, and calculating an included angle theta between the center straight line and the horizontal direction; and rotating the finger vein image anticlockwise by an angle theta so that the central straight line coincides with the horizontal direction, and restoring the finger rough positioning boundary of the zoom map onto the finger vein image.
4. A finger vein recognition method according to claim 3, wherein said finding said finger boundary using a greedy strategy based on gradient information of a rotation corrected finger vein image comprises: Respectively intercepting an upper sub-image containing a coarse positioning upper boundary and a lower sub-image containing a coarse positioning lower boundary on the finger vein image after rotation correction; gradient information of the upper sub-image and the lower sub-image is calculated respectively; for gradient information of the upper sub-image, reserving pixel points with gradients in the vertical direction larger than a third screening threshold, and setting the gradients in the vertical direction larger than a third cutting threshold as the third cutting threshold to obtain an upper boundary fine positioning pixel point set; for gradient information of the lower sub-image, reserving pixel points with gradients in the vertical direction smaller than a fourth screening threshold, and setting the gradients in the vertical direction smaller than a fourth cutoff threshold as a fourth cutoff threshold to obtain a lower boundary fine positioning pixel point set; And searching a line segment with the largest accumulated gradient in one direction from the other direction in the upper boundary fine positioning pixel point set and the lower boundary fine positioning pixel point set by utilizing a greedy strategy respectively to obtain an upper boundary and a lower boundary of the finger as the finger boundary.
5. The method of claim 4, wherein detecting a finger joint on the finger vein image and extracting an effective area of the finger vein based on the detected finger joint and the finger boundary comprises: Determining a finger inscription area according to the upper boundary and the lower boundary of the finger; Detecting finger joints on the finger inscribed region, and determining the horizontal range of the effective region according to the number and the positions of the finger joints, wherein the horizontal range comprises all the finger joints; Determining the ordinate of the finger midline according to the upper boundary and the lower boundary of the finger, and determining the vertical range of the effective area according to the ordinate of the finger midline and the vertical width of the finger at the finger joint; the region consisting of the horizontal range and the vertical range of the effective region is the effective region.
6. The method for identifying a finger vein according to any one of claims 1 to 5, wherein the feature extraction is performed on the effective area by a Gabor filter to obtain a finger vein feature, further comprising: image enhancement is carried out on the effective area through the following formula: Wherein, the For the enhanced image, k is the scale range of the enhanced image; I (x, y) is an image before enhancement, I (x, y) =l (x, y) x R (x, y), L (x, y) is an ambient light irradiation component in I (x, y), R (x, y) is a target object reflection component in I (x, y), I (x, y) is estimated by L (x, y) =i (x, y) x F (x, y), and R (x, y) is estimated by I (x, y) =l (x, y) x R (x, y); d is a set coefficient.
7. A finger vein recognition device, the device comprising: the finger vein image acquisition module is used for acquiring finger vein images to be detected; the finger boundary positioning module is used for calculating gradient information of the finger vein image and searching for a finger boundary by utilizing a greedy strategy according to the gradient information; The effective area extraction module is used for detecting finger joints on the finger vein image and extracting an effective area of the finger vein according to the finger joints and the finger boundaries; the feature extraction module is used for extracting features of the effective area through a Gabor filter to obtain finger vein features; The sliding comparison module is used for setting a sliding window on the finger vein image, comparing the similarity between the finger vein features of the finger vein image in the sliding window and the corresponding positions of the finger vein feature template, and finding out the finger vein image in the sliding window with the highest similarity as a maximum matching area; the matching score determining module is used for taking the similarity between the finger vein features of the finger vein image in the maximum matching area and the corresponding positions of the finger vein feature templates as matching scores; the finger boundary positioning module comprises: the rough positioning unit is used for performing scale reduction on the finger vein image to obtain a zoom image, calculating gradient information of the zoom image, and finding a rough finger positioning boundary by utilizing a greedy strategy according to the gradient information of the zoom image; The rotation correction unit is used for carrying out rotation correction on the finger vein image according to the finger rough positioning boundary; And the accurate positioning unit is used for finding the finger boundary by using a greedy strategy according to the gradient information of the finger vein image after the rotation correction.
8. A computer readable storage medium for finger vein recognition, comprising a memory for storing processor executable instructions which when executed by the processor implement steps comprising the finger vein recognition method of any of claims 1-6.
9. An apparatus for finger vein recognition comprising at least one processor and a memory storing computer executable instructions which when executed by the processor implement the steps of the finger vein recognition method as claimed in any one of claims 1 to 6.

Description

Finger vein recognition method, device, computer readable storage medium and apparatus Technical Field The present invention relates to the field of biometric identification, and in particular, to a finger vein identification method, apparatus, computer readable storage medium and device. Background Biometric technology is an operation of identity verification based on physiological and behavioral characteristics of the human body. The finger vein is a physiological characteristic in the human body, is distributed under the skin of the finger, and the finger vein information collected under the near infrared light source can meet the requirement of biological characteristic identification. The finger vein recognition has good stability and safety, because the finger vein is positioned below the skin surface layer, the problems of aging/abrasion and the like are avoided, in addition, the finger vein acquisition needs to irradiate the skin by utilizing near infrared light, and the vein image is captured by utilizing the characteristic that hemoglobin absorbs near infrared light, so that the finger vein recognition has natural anti-counterfeiting property and is not easy to forge. Because of these advantages, finger vein recognition technology is receiving more and more attention and has wide application value. The general flow of finger vein recognition is as shown in fig. 1, and after the finger vein image acquisition is performed under the near infrared illumination condition, finger vein image pretreatment, finger vein feature extraction and feature comparison are sequentially performed. The finger vein image preprocessing is to extract an effective area (i.e. a finger vein area) containing finger vein information, and normalize the scale and enhance the texture of the area. This is because the collected finger vein image is often irregular and there is much disturbance information in the finger vein image in addition to the main body part of the finger vein, which is disturbed by various factors such as the collection device, the collection environment, the difference of human fingers, and the like. The interference can be effectively reduced through finger vein image preprocessing, and a foundation is laid for extracting effective characteristics. Finger vein feature extraction, namely after a stable finger vein region is obtained through finger vein image pretreatment, the finger vein feature is required to be identified through a method for effectively expressing the finger vein feature. An effective finger vein feature extraction method is an important means for improving finger vein recognition effect. The finger vein feature extraction methods commonly found in the prior art are various methods such as shape-based/texture-based/minutiae-based. Feature matching, namely after extracting the finger vein features, the finger veins are required to be subjected to feature matching. In the finger vein recognition algorithm, the feature matching method is generally classified into two types, namely a distance-based method, such as directly calculating Euclidean distance, cosine distance and the like of two features, and a classification-based method, wherein the two features are judged whether to be from the same type by using machine learning methods such as SVM, neural network, fuzzy logic and the like. The finger vein recognition method in the prior art has the following problems: 1. In finger vein image preprocessing, the finger boundaries need to be positioned when the effective area containing finger vein information is extracted, and most of the current finger vein algorithms search discrete finger boundary points based on gray level or gradient characteristics, so that the found finger boundaries are easily interfered by noise, and the effect of finger boundary detection is reduced. 2. Due to the difference of the collecting devices/different finger placement positions and the like, the collected finger vein areas have deviation, and the traditional method cannot exclude the influence of the deviation of the finger vein areas when extracting the effective areas of the finger veins, so that the stability and the effectiveness of the intercepted finger vein areas are poor. 3. In the process of finger vein collection, the finger is placed with a high degree of freedom, so that the finger rotates to a certain extent, the finger gestures collected twice are inconsistent, and the recognition performance of veins can be reduced by utilizing a traditional finger vein feature extraction and feature comparison algorithm. Disclosure of Invention In order to solve the technical problem that the finger vein recognition method in the prior art is easy to be interfered by noise, finger placement positions and the like, the invention provides the finger vein recognition method, the device, the computer-readable storage medium and the equipment. The technical scheme provided by the invention is as follows: in a firs