A gate control system and method based on palm vein technology

Abstract

The application relates to a gate control system based on palm vein technology, which can reduce various noises contained in the image, prevent the existence of a larger protrusion or many burrs on the vein lines in the obtained image, and is beneficial to obtaining smooth and less-distorted palm vein line features of the palm and back of the hand; and the established palm vein feature model contains the macroscopic features of the palm veins, the fifth metacarpophalangeal joint is taken as a coordinate origin to segment and obtain the vein features in the rectangular area, the vein features in the area are limited features and do not change with the palm movement, feature matching is carried out by combining the macroscopic features of the veins and the microscopic features of the veins, the accuracy can be greatly improved, and the adaptive space for the acquisition error caused by the human body state change is provided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of palm vein control, in particular to a gate control system and method based on palm vein technology. BACKGROUND

[0002] The palm vein technology is based on a near-infrared sensor to obtain a palm vein distribution feature map, establish a palm vein database unique to a user, and match the palm vein distribution feature map collected again to log in the system to achieve a biological recognition function management. The palm vein technology is usually connected with an access controller to limit non-users from entering, and can be configured with a voice prompt and display function, and has advantages of being difficult to forge, high recognition rate, fast recognition speed, non-contact characteristics, etc.

[0003] At present, the palm vein recognition technology is used to control the opening and closing of a gate, and only the palm palm vein distribution information is obtained. However, in the palm vein recognition process, an infrared wavelength needs to pass through a palm fat layer to obtain vein information. If a person's weight changes significantly within a certain period, the palm vein information logged in later may be different from the palm vein information stored in the database initially, such as when a person's weight is relatively heavy, the palm fat layer of the palm will increase significantly, which may result in that the palm vein information logged in initially is not comprehensive. When the weight decreases, the palm vein information is fully visible, and the palm vein information logged in later may be different from the palm vein information logged in initially, which may result in that the system information does not match. Therefore, it is necessary to further improve and optimize, and try to extract more palm vein information to make the system have a better vein recognition rate. SUMMARY

[0004] To solve the above problems, the present application provides a gate control system and method based on palm vein technology.

[0005] The technical scheme adopted by the present application is: a gate control system based on palm vein technology, comprising a gate control system and a palm vein recognition system, the palm vein recognition system comprising the following modules:

[0006] An image acquisition module is provided with two groups, which are respectively used for acquiring palm palm vein images and back of hand vein images;

[0007] A preprocessing module is used for image gray scale normalization and image enhancement processing of the palm palm vein images and the back of hand vein images, so that the vein features are visualized. The images after the vein features are visualized are subjected to image binarization processing, and the palm palm vein line image and the back of hand vein line image are obtained by using Rosenfeld thinning.

[0008] The feature extraction module fuses the palm vein pattern and the back of the hand vein pattern by weighting, establishes a palm vein feature model, establishes a target region coordinate system on the front and back of the palm vein feature model, and takes the fifth metacarpophalangeal joint as the coordinate origin, draws a horizontal line x-axis and a vertical line y-axis, intercepts a fixed size of 6cm*4cm rectangular region at 0.5-2cm away from the x-axis and at 0-1cm away from the y-axis, divides the region into 6 square regions for image segmentation, and extracts the vein micro features in each square region;

[0009] The storage module stores the palm vein feature model and the vein micro features in a palm vein dataset of a database;

[0010] The feature matching module is configured to match the newly collected palm vein feature model and vein micro features with the palm vein feature model and vein micro features of the palm vein dataset respectively, and outputs a matching result to the gate control system. When the coincidence degree of the palm vein feature model is less than 0.75, the output result is mismatched. When the coincidence degree of the palm vein feature model is greater than or equal to 0.75, and the coincidence degree of the vein micro features is greater than or equal to 0.9, the output result is matched. When the coincidence degree of the palm vein feature model is less than 0.75, and the coincidence degree of the vein micro features is greater than or equal to 0.9, the output result is re-matched.

[0011] As a preferred embodiment, the image acquisition module comprises an acquisition probe and an image compensation model. The acquisition probe acquires at least 3 groups of images at a time. The image compensation model performs image cutting on the acquired images by using a Graph-Based Segmentation algorithm after weighting and fusing the images, performs normalization processing on the weighted and fused images, divides the images into 6 layers according to brightness, merges and extracts details from the region blocks cut out from the images according to the brightness layers, calculates the proportion of the pixels of each layer region in the entire image pixels, obtains the histogram of each layer region, calculates the maximum displacement of the overlapping region of the histograms between adjacent two layers, performs brightness mapping on the pixels of each region, obtains the region layer combination brightness value, obtains the S-shaped Gamma curve of the image, and constrains the horizontal pixels and vertical pixels of the image by using the curve to automatically compensate the brightness of the image.

[0012] As a preferred embodiment, the pre-processing module performs the image gray normalization and image enhancement processing steps based on low-pass filtering, which comprises:

[0013] (1) The image size is scaled and transformed to 100*100 pixels by matrix operation, which is expressed as:

[0014]

[0015] where (x, y) is the coordinate of the original image, (x', y') is the coordinate of the transformed image, R is the matrix, and the size of the original image is m x n pixels, then The pixel value of the output image is obtained by inverse mapping R - {(x,y)} The coordinates of the output image are calculated, and then the nearest neighbor interpolation method is used to interpolate them into the nearest input pixel value to determine the value of the output pixel.

[0016] (2) The gray scale of the image is unified to the same gray scale range [I min , I max ], and the image is transformed to the range of [0, 225] by the gray scale normalization algorithm to obtain a gray scale normalized image, wherein the gray scale normalization algorithm is:

[0017]

[0018] I' is the original image, I is the gray scale normalized image, I max is the minimum gray scale of the original image, and I min is the maximum gray scale of the original image.

[0019] (3) The gray scale normalized image is divided into low frequency region and high frequency region, and corresponding enhancement coefficient K(x, y) is selected in the low frequency region and the high frequency region, K(x, y) ≥ 1, the low frequency region is divided according to the local variance according to the 3x3 window, and the local variance formula is: The high frequency region is calculated based on the following algorithm: G(x, y) = (1 + λ)f(x, y) - f(x, y)H(x, y), wherein

[0020] According to the above solution, the image enhancement algorithm is: g(x, y) = f(x, y) + k(x, y) x G(x, y), wherein f(x, y) is the input image, G(x, y) is the detail image or high frequency image of the original image, and g(x, y) is the transformed image.

[0021] As preferred, the image is divided according to the 3x3 window, the mean and variance of the pixel points in the region are calculated, and the binaryzation formula is used for processing:

[0022] P(x, y) = Q(x, y) + κ x S(x, y), P(x, y) is the threshold value of the point, Q(x, y) is the mean value of the pixel points in the 3x3 region, S(x, y) is the variance of the pixel points in the region, and κ is the correction coefficient, wherein f(i, j) is the gray scale value of point (i, j).

[0023] As preferred, the feature extraction module acquires 6 vein micro-features on the front and back of the palm vein feature model respectively, arranges the vein micro-features according to s1, s2...s12, and then calculates 7 Hu invariant moments η1, η2...η7 of each image si (i=1, 2...12) in the vein micro-features s1, s2...s12, and obtains η1, η2...η7 by the following steps:

[0024]

[0025] HMIF=(M1, M2,...M12), wherein m+n=2, 3,...; i=1, 2..., 12; j=1, 2,...7.

[0026] As preferred, the palm vein feature model establishing method comprises:

[0027] (1) registering the palm vein pattern image and the back vein pattern image, and the registration accuracy is within one pixel;

[0028] (2) performing n-layer wavelet decomposition on the palm vein pattern image and the back vein pattern image, in the form of:

[0029]

[0030] wherein K 1(N) and K 2(N) represent the low-frequency components of the palm vein pattern image and the back vein pattern image at the n-th decomposition layer respectively; w i (K1) and w i (K2), i=1, 2,..., n represent the wavelet coefficients of the palm vein pattern image and the back vein pattern image at the i-th layer respectively. n is 2-4 decomposition layers;

[0031] (3) performing fusion processing on the wavelet coefficients by using a weighted fusion rule, to obtain:

[0032] wherein f1 and f2 are corresponding fusion operators;

[0033] (4) performing inverse wavelet transform on w i (F) and G n to obtain the fusion result image.

[0034] As preferred, the coincidence degree matching method comprises:

[0035] (1) mapping the newly collected palm vein feature model and the palm vein feature model of the palm vein data set to an A vector space model, and then calculating the cosine value size between the two vectors;

[0036] (2) The newly collected vein micro-features are mapped to the B vector space model with the vein micro-features of the palm vein dataset, and then the cosine value size between the two vectors is calculated.

[0037] As a preferred, a gate control method based on palm vein technology is characterized in that:

[0038] Step 1): Collecting palm palm vein images and back vein images;

[0039] Step 2): Preprocessing the palm vein images and back vein images to obtain palm vein line images and back vein line images, and obtaining palm vein line features and back vein line features;

[0040] Step 3): Weighted fusion of palm vein line images and back vein line images to establish a palm vein feature model, and obtaining palm vein micro-features of palm and back through image cutting;

[0041] Step 4): Storing the palm vein feature model, palm vein micro-features of palm and back in the storage module, and storing the palm vein feature model, palm vein micro-features of palm and back from the same palm as a palm vein dataset;

[0042] Step 5): Collecting new palm vein feature model and vein micro-features, matching them with the palm vein dataset stored in the storage module, and outputting the matching result to the gate control system for display, when the output result is matching, the gate control system controls the gate to open.

[0043] As a preferred, when the coincidence degree of the palm vein feature model is less than 0.75, the gate control system outputs a result of not matching, and the gate is not opened; when the coincidence degree of the palm vein feature model is greater than or equal to 0.75, and the coincidence degree of the vein micro-features is greater than or equal to 0.9, the gate control system outputs a result of matching, and the gate is opened; when the coincidence degree of the palm vein feature model is less than 0.75, and the coincidence degree of the vein micro-features is greater than or equal to 0.9, the gate control system outputs a result of re-matching.

[0044] The beneficial effects of the present application are:

[0045] The gate control system and method based on palm vein technology of the present application can simultaneously collect palm vein images and back vein images, expand the palm vein information amount through back vein features, improve the accuracy of feature matching, allow a certain coincidence degree deviation, and solve the problem of inaccurate palm vein information amount caused by traditional acquisition.

[0046] The palm vein technology-based gate control system and method can reduce various noises contained in the image, prevent the vein lines in the obtained image from having a large protrusion or many burrs on the edge, and is beneficial to obtaining smooth and less distorted palm vein line features and back of hand vein line features; and the established palm vein feature model contains the palm vein macro feature, the vein feature in the rectangular region is segmented and obtained by taking the fifth metacarpophalangeal joint as the coordinate origin, the vein feature in the region is a limited feature and does not change with the palm movement, the feature matching is performed by combining the vein macro feature and the vein micro feature, the accuracy can be greatly improved, and the adaptation space for the acquisition error caused by the human body state change is provided. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 It is a palm vein technology-based gate control system image acquisition module schematic diagram of the application;

[0048] Figure 2 It is a palm vein technology-based gate control system principle diagram of the application;

[0049] Figure 3 It is a palm vein feature model front target area coordinate system schematic diagram;

[0050] Figure 4 It is a palm vein feature model back target area coordinate system schematic diagram;

[0051] Figure 5 It is a palm vein technology-based gate control method flow chart of the application. DETAILED DESCRIPTION

[0052] The features and exemplary embodiments of various aspects of the application will be described in detail below, in order to make the purposes, technical solutions and advantages of the application more clear and apparent, the application will be further described in detail below in combination with the drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the application, but not to limit the application. The application can be implemented without some of these specific details by those skilled in the art. The following description of the embodiments is only to provide a better understanding of the application by showing examples of the application.

[0053] It should be understood that when used in the specification and the appended claims of the application, the term "comprising" indicates the presence of the described features, whole, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, whole, steps, operations, elements, components and / or sets thereof.

[0054] It should also be understood that the term "and / or" as used herein is used to associate together one or more of the associated listed items and all possible combinations of the items, and includes these combinations.

[0055] As used in the specification and the appended claims, the term "if' can be interpreted as meaning "when" or "once" or "in response to a determination" or "in response to a detection" depending on the context. Similarly, the phrase "if determined" or "if detected [the described condition or event]" can be interpreted as meaning "once determined" or "in response to a determination" or "once detected [the described condition or event]" or "in response to a detection [the described condition or event]" depending on the context.

[0056] Embodiment One:

[0057] This embodiment refers to Figures 1-4 A gate control system based on palm vein technology, comprising a gate control system and a palm vein identification system, the palm vein identification system is connected to the gate control system, the palm vein identification system comprises an image acquisition module, a preprocessing module, a feature extraction module, a storage module and a feature matching module, the gate control system and the image acquisition module are arranged on the gate, the image acquisition module is provided with two groups, which are symmetrically arranged above and below, and are respectively used for acquiring palm palm vein image and back of hand vein image.

[0058] In one embodiment, the image acquisition module includes an acquisition probe and an image compensation model, the acquisition probe acquires at least 3 groups of images at a time, the image compensation model performs weighted fusion on the acquired images, and at least 3 groups of images are fused in an overlapping manner to form an image with complete feature quantity, then the Graph-Based Segmentation algorithm is used for image cutting, the weighted fused image is normalized, the image is divided into 6 layers according to brightness, then the region blocks cut out from the image according to brightness layer are merged and details are extracted, the proportion of pixels of each layer region to the whole image is calculated, the histogram of each layer region is obtained, the maximum displacement of the overlapping region between adjacent two layers is calculated, and the brightness mapping of each region pixel is performed, and the brightness mapping formula is: The formula is: p=1 / (1+exp(-t), according to the formula algorithm, all possible area layer brightness combinations are listed, the Y value of each combination is calculated, the combination with the minimum Y value is taken, the area layer brightness value is obtained, the S-shaped Gamma curve of the image is obtained, the horizontal pixels and the vertical pixels of the image are constrained by using the curve to automatically compensate the image brightness, the value of Gamma greater than 1 can expand the light value area and compress the dark value area, the greater the value of Gamma, the greater the expansion of the light value area, and the greater the compression of the dark value area, and vice versa when the value of Gamma is less than 1.

[0059] The collection probe is composed of a CCD camera equipped with an infrared filter and an infrared LED array, the infrared LED array is arranged around the CCD camera, the infrared LED array emits infrared light with a wavelength of 750-820nm, can penetrate into human tissue by 3-5mm, and obtain the venous blood vessel distribution image of the palm.

[0060] In another embodiment, the collection probe can also select at least one of a palm vein sensor, an infrared camera, a palm vein identification instrument, a near-infrared sensor, an infrared image converter, an imaging sensor silicon palladium camera tube.

[0061] The image acquisition module acquires the palm vein image and the back of the hand vein image, and transmits them to the preprocessing module for preprocessing. First, the preprocessing module respectively performs image gray normalization and image enhancement processing on the palm vein image and the back of the hand vein image based on low-pass filtering, and the processing steps include:

[0062] (1) The image size is scaled and transformed to 100x100 pixels through matrix operation, and the matrix operation is expressed as:

[0063]

[0064] Where (x, y) is the coordinate of the original image, (x', y') is the coordinate after transformation, R is the matrix, and the size of the original image is m*n pixels, then The pixel value of the output image is calculated by inverse mapping R - {(x,y)} calculates the coordinates of the output image corresponding to the input image, and then uses the nearest neighbor interpolation method to interpolate the nearest input pixel value to determine the value of the output pixel;

[0065] (2) The gray scale of the image is unified to the same gray scale range [I min , I max ], the image is transformed to the range of [0, 225] through the gray normalization algorithm, and the gray normalized image is obtained, wherein the gray normalization algorithm is:

[0066]

[0067] I' is the original image, I is the image after gray scale normalization, I max Imin is the minimum gray scale of the original image, I min Imax is the maximum gray scale of the original image;

[0068] (3) The gray scale normalized image is divided into a low frequency region and a high frequency region, and corresponding enhancement coefficients K(x, y) are selected in the low frequency region and the high frequency region, K(x, y) >= 1, the low frequency region is divided according to a local variance in a 3*3 window, and the local variance formula is: The high frequency region is calculated based on the following algorithm: G(x, y) = (1 + lambda) f(x, y) - f(x, y) H(x, y), wherein

[0069] According to the above solution, the image enhancement algorithm is: g(x, y) = f(x, y) + k(x, y) * G(x, y), wherein, f(x, y) is the input image, G(x, y) is the detail image or high frequency image of the original image, and g(x, y) is the transformed image.

[0070] Since the size, fat thickness and other factors of the palm and back of each person are different, in order to improve the accuracy of module recognition, the image gray scale normalization is performed on the palm vein image and the back vein image through the preprocessing module, the gray scale is unified to the same gray scale range, and then the image enhancement processing is performed, the high frequency region is appropriately improved to improve the visual effect, and the vein features are better and more obvious and clear.

[0071] In addition, the image binarization processing is performed on the image in which the vein features are visualized by the preprocessing module, the image is divided according to a 3*3 window, the mean and variance of the pixel points in the region are calculated, and the binarization formula is used for processing:

[0072] P(x, y) = Q(x, y) + kappa * S(x, y), P(x, y) is a threshold value of the point, Q(x, y) is the mean of the pixel points in the 3*3 region, S(x, y) is the variance of the pixel points in the region, and kappa is a correction coefficient, wherein, f(i, j) is the gray scale value of the point (i, j), the image is converted into a binary image by using the binarization algorithm, the palm vein line graph and the back vein line graph are obtained by using the Rosenfeld thinning, the accuracy and efficiency of the preprocessing can be greatly improved, and the problems of holes, burrs and unsmooth edges of the vein features in the image are reduced, so that the edges of the vein features in the processed binary image are smooth.

[0073] The palm vein line graph and the back vein line graph after preprocessing are fused by the feature extraction module, a palm vein feature model is established, and the method for establishing the palm vein feature model comprises:

[0074] (1) Register the palm vein pattern and the back of the hand vein pattern, and the registration accuracy is within one pixel;

[0075] (2) Perform n-layer wavelet decomposition on the palm vein pattern and the back of the hand vein pattern, in the form of:

[0076]

[0077] wherein K 1(N) and K 2(N) represent the low-frequency components of the palm vein pattern and the back of the hand vein pattern at the n-th decomposition layer, respectively; w i (K1) and w i (K2), i = 1, 2, …, n represent the wavelet coefficients of the palm vein pattern and the back of the hand vein pattern at the i-th layer, respectively. n is 2-4 decomposition layers;

[0078] (3) Perform fusion processing on the wavelet coefficients using a weighted fusion rule, to obtain:

[0079] wherein f1 and f2 are corresponding fusion operators;

[0080] (4) Perform inverse wavelet transform using w i (F) and G n to obtain the fusion result image.

[0081] The palm vein feature model has palm vein features and back of the hand vein features. Target region coordinate systems are established on the front and back of the palm vein feature model, and the target region coordinate systems take the fifth metacarpophalangeal joint as the coordinate origin, draw a horizontal line x-axis and a vertical line y-axis, and intercept a fixed size of 6cm x 4cm rectangular region 0.5-2cm away from the x-axis and 0-1cm away from the y-axis. The vein features in the rectangular region are relatively stable and generally do not change with palm opening and closing activities. The rectangular region is divided into 6 square regions for image segmentation, and vein micro features in each square region are extracted. The feature extraction module acquires 6 vein micro features on the front and back of the palm vein feature model, respectively, arranges the vein micro features in s1, s2...s12, and then calculates 7 Hu invariants η1, η2...η7 of each image si (i = 1, 2...12) in the vein micro features s1, s2...s12. η1, η2...η7 are obtained by the following steps:

[0082]

[0083] HMIF=(M1,M2,...M12), wherein, m+n=2, 3,...; i=1, 2,..., 12; j=1, 2,..., 7.

[0084] The obtained vein micro features contain the distribution of veins in each square region, which is conducive to subsequent micro feature matching and improves the accuracy of palm vein recognition.

[0085] The storage module stores the palm vein feature model and the vein micro features in the palm vein dataset of the database as an entry template, and the palm vein feature model, the palm vein micro features of the palm and the back of the palm from the same palm are coded as one palm vein dataset, and in the later stage, only the palm vein information of the newly collected palm vein needs to be compared with the palm vein information in the corresponding palm vein dataset in the storage module.

[0086] The feature matching module is used for matching the newly collected palm vein feature model and the vein micro features with the palm vein feature model and the vein micro features of the palm vein dataset respectively, and outputting the matching result to the gate control system.

[0087] (1) mapping the newly collected palm vein feature model and the palm vein feature model of the palm vein dataset to an A vector space model, and then calculating the cosine value size of the included angle between the two vectors;

[0088] (2) mapping the newly collected vein micro features and the vein micro features of the palm vein dataset to a B vector space model, and then calculating the cosine value size of the included angle between the two vectors.

[0089] The greater the cosine value of the included angle is, the more similar the two vectors are, and the cosine value calculation formula is: The numerator is the inner product of the two vectors, the cosine value of the included angle between the two vectors, the greater the included angle between the vectors is, the smaller the cosine value is, and the smaller the final similarity is.

[0090] When the coincidence degree of the palm vein feature model is <0.75, the output result is not matched; when the coincidence degree of the palm vein feature model is ≥0.75 and the coincidence degree of the vein micro features is ≥0.9, the output result is matched; when the coincidence degree of the palm vein feature model is <0.75 and the coincidence degree of the vein micro features is ≥0.9, the output result is re-matching.

[0091] Example two:

[0092] Reference Figure 5 The embodiment provides a gate control method based on palm vein technology, which controls the opening and closing of the gate based on the gate control system based on palm vein technology, and the specific steps are as follows:

[0093] Step 1): Collecting palm palm vein image and dorsal vein image;

[0094] Step 2): Preprocessing palm palm vein image and dorsal vein image, obtaining palm palm vein pattern and dorsal vein pattern, obtaining palm vein feature and dorsal vein feature;

[0095] Step 3): Palm vein feature model is established by weighted fusion of palm vein pattern and dorsal vein pattern, and the microstructure of palm vein and dorsal vein is obtained by image cutting;

[0096] Step 4): Store palm vein feature model, palm vein microstructure and dorsal vein microstructure in storage module, palm vein feature model, palm vein microstructure and dorsal vein microstructure from the same palm are coded as a palm vein data set;

[0097] Step 5): Collect new palm vein feature model and vein microstructure, match it with palm vein data set stored in storage module, output matching result to gate control system for display, when the output result is matching, gate control system controls gate to open.

[0098] The specific judgment process is that when the coincidence degree of palm vein feature model is <0.75, the output result of gate control system is not matching, and the gate is not opened; when the coincidence degree of palm vein feature model is ≥0.75, and the coincidence degree of vein microstructure is ≥0.9, the output result of gate control system is matching, and the gate is opened; when the coincidence degree of palm vein feature model is <0.75, and the coincidence degree of vein microstructure is ≥0.9, the output result of gate control system is re matching.

[0099] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is taken as an example, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional module in the embodiment can be integrated in one processing unit, or each unit can be physically independent, or two or more units can be integrated in one unit. The above integrated unit can be realized in the form of hardware or software. In addition, the specific name of each functional module is only for easy distinction, and does not limit the protection scope of the application. The specific working process of the modules in the above system can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0100] In the above embodiments, the description of each embodiment is focused on, and the parts not described or recorded in a certain embodiment can be referred to the relevant description of other embodiments.

[0101] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0102] In the embodiments provided in the present application, it should be understood that the disclosed apparatus / equipment and methods can be implemented in other ways. For example, the apparatus / equipment embodiments described above are merely illustrative, for example, the division of modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed units can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0103] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment.

[0104] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0105] The integrated module, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. The computer readable storage medium can be nonvolatile or volatile. Based on such understanding, all or part of the processes in the above-mentioned embodiments can be implemented by a computer program, and the computer program can be stored in a computer readable storage medium. The computer program, when executed by a processor, can implement the steps of the above-mentioned various method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer readable storage medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer readable storage medium does not include electrical carrier signals and telecommunication signals.

[0106] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

[0107] The above content is a further detailed description of the present application in combination with specific preferred embodiments, and the specific implementation of the present application cannot be limited to these descriptions. For ordinary skilled persons in the technical field to which the present application belongs, some simple deductions or replacements can be made without departing from the concept of the present application, and all of them should be regarded as falling within the protection scope of the present application.

Claims

1. A gate control system based on palm vein technology, comprising a gate control system and a palm vein identification system, characterized in that, The palm vein identification system comprises the following modules: The image acquisition module is provided with two groups, which are respectively used for collecting palm palm vein images and back vein images; The preprocessing module is used for image gray scale normalization and image enhancement processing of the palm vein images and the back vein images, so that the vein features are highlighted, the images after the vein feature highlighting are subjected to image binarization processing, and the palm palm vein line graph and the back vein line graph are obtained by using the Rosenfeld thinning; The feature extraction module is used for weighted fusion of the palm vein line graph and the back vein line graph, establishment of a palm vein feature model, establishment of a target region coordinate system on the front and back of the palm vein feature model, and taking the fifth metacarpophalangeal joint as a coordinate origin, a horizontal line x axis and a vertical line y axis, a fixed size of 6cm*4cm rectangular region is intercepted at 0.5-2cm away from the x axis and 0-1cm away from the y axis, the region is divided into 6 square regions for image segmentation, and vein micro features in each square region are extracted respectively; The storage module is used for storing the palm vein feature model and the vein micro features in a palm vein data set of a database; The feature matching module is used for matching the newly collected palm vein feature model and the vein micro features with the palm vein feature model and the vein micro features of the palm vein data set respectively, and outputting a matching result to a gate control system, when the coincidence degree of the palm vein feature model is less than 0.75, the output result is mismatching, when the coincidence degree of the palm vein feature model is greater than or equal to 0.75 and the coincidence degree of the vein micro features is greater than or equal to 0.9, the output result is matching, and when the coincidence degree of the palm vein feature model is less than 0.75 and the coincidence degree of the vein micro features is greater than or equal to 0.9, the output result is re-matching. The method for establishing the palm vein feature model comprises: (1) registration of the palm vein line graph and the back vein line graph, and the registration accuracy is within one pixel; (2) n-layer wavelet decomposition of the palm vein line graph and the back vein line graph, in the form of: where K 1(N) and K 2(N) represent the low frequency components of the palm and dorsal hand vein pattern at the nth resolution level; w i (K1) and w i (K2), i = 1, 2,..., n represent the wavelet coefficients of the palm and dorsal hand vein pattern at the ith level, n takes 2-4 resolution levels; (3) fusion processing of the wavelet coefficients by using a weighted fusion rule, and calculation of: where fi, f2are the respective fusion operators; (4) inverse wavelet transformation by using wi(F) and Gn to obtain a fusion result image.

2. The gate control system based on palm vein technology according to claim 1, characterized in that: The image acquisition module comprises an acquisition probe and an image compensation model, the acquisition probe acquires at least 3 groups of images at a time, the image compensation model performs image cutting on the acquired images by using a Graph-Based Segmentation algorithm after weighted fusion, performs normalization processing on the weighted fusion images, divides the images into 6 layers according to brightness, merges and extracts details of the region blocks cut out from the images according to the brightness layers, calculates the proportion of the pixels of each layer region in the entire image pixels, obtains a histogram of each layer region, calculates the maximum displacement of the overlapping region of the histograms between two adjacent layers, performs brightness mapping on the pixels of each region, obtains a region layer combination brightness value, obtains an S-shaped Gamma curve of the image, and constrains the horizontal pixels and the vertical pixels of the image by using the curve to automatically compensate the brightness of the image.

3. The palm vein technology based gate control system according to claim 1, characterized in that: The preprocessing module performs image gray normalization and image enhancement processing steps based on low-pass filtering, which includes: (1) The image size is scaled and transformed to 100x100 pixels by matrix operation, which is expressed as: where (x, y) is the coordinate of the original image, (x', y') is the coordinate of the transformed image, R is the matrix, and the size of the original image is m x n pixels, then The pixel value of the output image is mapped by the inverse mapping R - {(x,y)} The coordinates corresponding to the input image are calculated, and then the nearest neighbor interpolation method is used to interpolate them into the nearest input pixel value to determine the value of the output pixel. (2) the gray scale of the image is unified into the same gray scale range [I min , I max ], the image is transformed into the range of [0, 225] through a gray scale normalization algorithm to obtain a gray scale normalized image, wherein the gray scale normalization algorithm is: I' is the original image, I is the image after gray scale normalization, I max is the minimum value of the gray scale of the original image, I min is the maximum value of the gray scale of the original image; (3) The gray normalized image is divided into low frequency region and high frequency region, and the corresponding enhancement coefficient K(x, y) is selected in the low frequency region and the high frequency region, K(x, y)≥1, the image is divided into 3x3 windows according to the local variance in the low frequency region The local variance formula is: The high frequency area is calculated based on the following algorithm: G(x, y) = (1 + λ)f(x, y) - f(x, y)H(x, y), where From the above solution, the image enhancement algorithm is: g(x, y) = f(x, y) + k(x, y) x G(x, y), where, f(x,y) is the input image, G(x,y) is the detail image or high frequency image of the original image, and g(x,y) is the transformed image.

4. The gate control system based on palm vein technology according to claim 1, characterized in that: The image is divided into 3x3 windows, the mean and variance of the pixels in the region are calculated, and the binary formula is used for processing: P(x, y) = Q(x, y) + κ × S(x, y), P(x,y) is the threshold value of the point, Q(x,y) is the mean of the pixels in the 3x3 region, S(x,y) is the variance of the pixels in the region, κ is the correction coefficient, wherein, f(i,j) is the gray value of point (i,j).

5. The gate control system based on palm vein technology according to claim 1, characterized in that: The feature extraction module acquires 6 micro-vein features of the palm vein feature model from the front and back respectively, arranges the micro-vein features according to s1, s2...s12, and then calculates 7 Hu invariants η1, η2...η7 of each image si(i=1, 2...12) in the micro-vein features s1, s2...s12, and η1, η2...η7 are obtained by the following steps: HMIF=(M1,M2,...M12), where m+n=2, 3,...; i=1, 2,..., 12; j=1, 2,...

7.

6. The gate control system based on palm vein technology according to claim 1, characterized in that: Matching method of coincidence degree: (1) Map the newly collected palm vein feature model and the palm vein feature model of the palm vein data set to the A vector space model, and then calculate the cosine value size between the two vectors; (2) Map the newly collected micro-vein features and the micro-vein features of the palm vein data set to the B vector space model, and then calculate the cosine value size between the two vectors.

7. A gate control method based on palm vein technology, using the system of any one of claims 1-6, characterized in that: Step 1): Collect the palm vein image of the palm and the palm vein image of the back of the hand; Step 2): Preprocess the palm vein image and the back of the hand vein image to obtain the palm vein image and the back of the hand vein image, and obtain the palm vein feature and the back of the hand vein feature; Step 3): Weighted fusion of palm vein image and back of the hand vein image, palm vein feature model is established, and micro-vein features of palm and back of the hand are obtained by image cutting; Step 4): Store the palm vein feature model, the micro-vein features of the palm and the back of the hand in the storage module, and the palm vein feature model, the micro-vein features of the palm and the back of the hand from the same palm are coded as a palm vein data set; Step 5): Collect new palm vein feature model and micro-vein features, and match them with the palm vein data set stored in the storage module, and output the matching result to the gate control system for display, when the output result is matched, the gate control system controls the gate to open.

8. The gate control method based on palm vein technology according to claim 7, characterized in that: When the coincidence degree of the palm vein feature model is less than 0.75, the gate control system outputs a result of mismatch, and the gate is not opened; when the coincidence degree of the palm vein feature model is greater than or equal to 0.75, and the coincidence degree of the vein micro-feature is greater than or equal to 0.9, the gate control system outputs a result of match, and the gate is opened; when the coincidence degree of the palm vein feature model is less than 0.75, and the coincidence degree of the vein micro-feature is greater than or equal to 0.9, the gate control system outputs a result of re-match.

Images