Bank customer identification method and device, electronic equipment and product
By performing grayscale conversion and key point recognition on multi-frame head area images of bank customers, the problem of traditional bank self-service terminals relying on ID cards has been solved, achieving fast and accurate user identification and improving the efficiency of self-service and customer experience.
Patent Information
- Application Number
- CN202510917194.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-03
- Publication Date
- 2025-11-14
AI Technical Summary
Traditional bank self-service terminals rely on ID cards, which prevents customers from using self-service, resulting in low business flow efficiency. This can easily lead to queuing and congestion, especially during peak hours, affecting customer experience and branch operational efficiency.
By acquiring multiple frames of head region images, performing grayscale conversion, multi-threshold binarization processing, and connected component analysis, the left eye, right eye, and mouth regions are identified. The facial images are corrected based on the iris center point, and the key point distance ratio is calculated to determine the user's identity.
Accurately identifying key points in facial regions improves the ease of use of self-service, avoids queuing congestion, and enhances customer experience and operational efficiency.
Smart Images
Figure CN120954073A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of facial recognition technology, specifically relating to a bank customer identification method, device, electronic device, and product. Background Technology
[0002] In traditional banking, self-service terminals are usually set up in the bank lobby to facilitate banking transactions for bank customers. Users can use the self-service terminals to handle related business or queue up to get a number by swiping their ID cards. The self-service terminals can read the customer's identity information through Optical Character Recognition (OCR) to complete the identity verification and assign a business queue.
[0003] While this approach can make it more convenient for bank customers to conduct related business to some extent, it has significant limitations. It relies heavily on physical identification documents, and in practice, customers may not be able to use self-service if they do not have their ID cards and need to go to the manual window, which reduces the efficiency of business diversion. Especially during peak hours, it can easily cause queuing congestion, affecting customer experience and branch operation efficiency.
[0004] Therefore, how to provide an effective solution to facilitate users' use of bank self-service has become a pressing problem to be solved in existing technologies. Summary of the Invention
[0005] The purpose of this invention is to provide a bank customer identification method, device, electronic device, and product to solve the aforementioned problems existing in the prior art.
[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for identifying bank customers, comprising: Acquire multi-frame head region images of the target user for whom banking business is pending; The multi-frame head region images are converted into grayscale images to obtain multi-frame head region grayscale images that correspond one-to-one with the multi-frame head region images. After performing multi-threshold binarization on the grayscale image of the head region of each frame, the binary images are merged to obtain the binary merged image of the head region corresponding to the grayscale image of the head region of each frame. Perform connected component analysis on each frame of the binary merged head region image, and determine the connected component region with the largest area that is located in the upper part of the binary merged head region image as the hair region. Based on the position of the hair region in the binary merged head region image of each frame, and the positional relationship between the hair region and the other connected regions in the binary merged head region image of each frame, the left eye region, right eye region and mouth region in the grayscale image of the head region corresponding to the binary merged head region image of each frame are determined. Based on the pixel grayscale distribution of the left and right eye regions in each frame of the head region grayscale image, the coordinates of the center point of the left iris and the center point of the right iris in each frame of the head region grayscale image are determined. Based on the coordinates of the center point of the left and right irises in each frame of the grayscale image of the head region, planar rotation correction is performed on each frame of the grayscale image of the head region to obtain the corrected grayscale image of the head region corresponding to each frame of the grayscale image of the head region. Based on the coordinates of the center point of the left and right irises in each frame of the corrected grayscale image of the head region, the horizontal coordinate range of the center point of the mouth region in each frame of the corrected grayscale image of the head region is determined. The window region located in the mouth region and within the horizontal coordinate range in each frame of the corrected head region grayscale image is integrally projected in the horizontal direction to obtain the grayscale integral curve of the mouth region in the vertical direction in each frame of the corrected head region grayscale image. Based on the coordinates of the endpoints of the mouth region on both sides in the grayscale image of the head region after correction in each frame, and the grayscale integral curve of the mouth region in the vertical direction, the coordinates of the lip point in the grayscale image of the head region after correction in each frame are determined. The lip point, mouth region boundary point, left eye region boundary point, and right eye region boundary point in each frame of the corrected grayscale image of the head region are used as facial recognition key points. The pixel distance between each pair of facial recognition key points is calculated to obtain multiple sets of key point distances corresponding to each frame of the corrected grayscale image of the head region. Calculate the ratio between the distances of multiple key points corresponding to each frame of the corrected head region grayscale image and the corresponding binocular distance to obtain the ratio sequence corresponding to each frame of the corrected head region grayscale image. The binocular distance corresponding to the corrected head region grayscale image is the pixel distance between the coordinates of the center point of the left iris and the center point of the right iris in the corrected head region grayscale image. Calculate the Euclidean distance between the ratio sequence corresponding to the grayscale image of the corrected head region of each frame and the ratio sequence corresponding to the grayscale image of the corrected head region of the known bank customer, and determine the bank customer identification result of the target user based on the Euclidean distance between the ratio sequence corresponding to the grayscale image of the corrected head region of each frame and the ratio sequence corresponding to the grayscale image of the corrected head region of the known bank customer.
[0007] In one possible design, the grayscale image of the head region in each frame is subjected to multi-threshold binarization processing, followed by binary image merging to obtain a binary merged image of the head region corresponding to each frame's grayscale image, including: By using multiple pre-set binary thresholds, the grayscale image of the head region of each frame is binarized multiple times to obtain a multi-frame binarized image corresponding to the grayscale image of the head region of each frame. Perform logical AND, OR, and OR operations on the multiple binarized images corresponding to each frame of the head region grayscale image to obtain the merged binary image of the head region corresponding to each frame of the head region grayscale image; or The multi-frame binarized images corresponding to the grayscale image of the head region of each frame are weighted and then normalized to obtain the merged binary image of the head region corresponding to the grayscale image of the head region of each frame.
[0008] In one possible design, based on the position of the hair region in each frame of the binary merged head region image, and the positional relationship between the hair region and other connected regions in each frame of the binary merged head region image, the left eye region, right eye region, and mouth region in the corresponding grayscale image of the head region for each frame of the binary merged head region image are determined, including: Based on the position of the hair region in each frame of the binary merged head region image, the center position of the other connected regions in each frame of the binary merged head region image, and the area of the other connected regions in each frame of the binary merged head region image, two connected regions in each frame of the binary merged head region image that are located below the hair region, whose corresponding center ordinate difference is lower than a first preset threshold, whose corresponding abscissa difference is within the range of the first preset threshold, and whose corresponding area is greater than the first preset area are respectively selected as the initial left eye region and the initial right eye region. In each frame of the binary merged head region image, the following region is selected as the initial mouth region: located below the initial left eye region and the initial right eye region, with its vertical coordinate between the center points of the initial left eye region and the initial right eye region, its corresponding area is greater than the second preset area, and the ratio of the distance between the corresponding center point and the center point of the initial left eye region to the center distance of the initial left and right eye regions is within the second preset threshold range, and the ratio of the distance between the corresponding center point and the center point of the initial right eye region to the center distance of the initial left and right eye regions is within the second preset threshold range. The center distance of the initial left and right eye regions is the pixel distance between the center points of the initial left eye region and the initial right eye region. The region in the grayscale image of the head region corresponding to the initially selected left eye region in the corresponding binary merged image of the head region is taken as the left eye region; the region in the grayscale image of the head region corresponding to the initially selected right eye region in the corresponding binary merged image of the head region is taken as the right eye region; and the region in the grayscale image of the head region corresponding to the initially selected mouth region in the corresponding binary merged image of the head region is taken as the mouth region.
[0009] In one possible design, based on the pixel grayscale distribution of the left and right eye regions in each frame of the head region grayscale image, the coordinates of the center point of the left and right iris in each frame of the head region grayscale image are determined, including: In each frame of the grayscale image of the head region, all neighboring pixels in the left eye region are located in the left eye target pixel of the left eye region, and in each frame of the grayscale image of the head region, all neighboring pixels in the right eye region are located in the right eye target pixel of the right eye region. The support function corresponding to each left-eye target pixel and the mean gray value of the neighborhood corresponding to each right-eye target pixel are calculated according to the following formulas.
[0010] Where A represents the neighborhood range of the target pixel in the left eye or the target pixel in the right eye. This represents the grayscale value of the pixel with coordinates (x, y) within the neighborhood of the left-eye target pixel or the right-eye target pixel, and N represents the total number of pixels within the neighborhood of the left-eye target pixel or the right-eye target pixel. The coordinates of the left eye target pixel with the lowest gray-scale mean in its corresponding neighborhood are taken as the coordinates of the center point of the left eye iris in the gray-scale image of the head region, and the coordinates of the right eye target pixel with the lowest gray-scale mean in its corresponding neighborhood are taken as the coordinates of the center point of the right eye iris in the gray-scale image of the head region.
[0011] In one possible design, based on the coordinates of the center point of the left and right irises in each frame of the head region grayscale image, a planar rotation correction is performed on each frame of the head region grayscale image to obtain a corrected head region grayscale image corresponding to each frame of the head region grayscale image, including: Based on the coordinates of the center point of the left iris and the center point of the right iris in each frame of the grayscale image of the head region, the tilt angle corresponding to each frame of the grayscale image of the head region is determined. Based on the tilt angle corresponding to the grayscale image of the head region in each frame, the rotation angle corresponding to the grayscale image of the head region in each frame is determined, and the corresponding grayscale image of the head region is corrected by planar rotation based on the rotation angle corresponding to the grayscale image of the head region in each frame, so as to obtain the corrected grayscale image of the head region corresponding to each frame.
[0012] In one possible design, the bank customer identification result of the target user is determined based on the Euclidean distance between the ratio sequence corresponding to the grayscale image of the corrected head region of each frame and the ratio sequence corresponding to the grayscale image of the corrected head region of a known bank customer, including: If the Euclidean distance between the ratio sequence corresponding to at least one frame of the corrected grayscale image of the head region and the corrected grayscale image of the head region corresponding to a known bank customer is lower than a preset distance, then the target user is determined to be the known bank customer.
[0013] In one possible design, multiple frames of head region images of the target user to be processed for banking services are acquired, including: Acquire multi-frame sample images containing the head region of the target user; The head region in the multi-frame sample images is identified by facial contour detection. The head region in the multi-frame sample images is expanded to obtain multi-frame head region images of the target user to be processed for banking business.
[0014] In a second aspect, the present invention provides a bank customer identification device, comprising: The acquisition unit is used to acquire multiple frames of head region images of the target user for whom banking business is to be processed. The grayscale conversion unit is used to convert the multi-frame head region images into grayscale images to obtain multi-frame head region grayscale images that correspond one-to-one with the multi-frame head region images. The binarization unit is used to perform multi-threshold binarization processing on the grayscale image of the head region of each frame and then merge the binary images to obtain the head region binary merged image corresponding to the grayscale image of the head region of each frame. The connected component analysis unit is used to perform connected component analysis on each frame of the binary merged head region image, and to determine the connected component region with the largest area that is located in the upper part of the binary merged head region image as the hair region. The first determining unit is used to determine the left eye region, right eye region, and mouth region in the grayscale image of the head region corresponding to each frame of the head region based on the position of the hair region in the binary merged head region image of each frame, and the positional relationship between the hair region and the other connected region regions in the binary merged head region image of each frame. The second determining unit is used to determine the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the head region grayscale image based on the pixel grayscale distribution of the left eye region and the right eye region in each frame of the head region grayscale image. The correction unit is used to perform planar rotation correction on each frame of the grayscale image of the head region based on the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the grayscale image of the head region, so as to obtain the corrected grayscale image of the head region corresponding to each frame of the grayscale image of the head region. The third determining unit is used to determine the horizontal coordinate range of the center point of the mouth region in each frame of the corrected head region grayscale image based on the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the corrected head region grayscale image. The projection unit is used to perform horizontal integral projection on the window area located in the mouth region and within the horizontal coordinate range in each frame of the corrected head region grayscale image to obtain the grayscale integral curve of the mouth region in the vertical direction in each frame of the corrected head region grayscale image. The fourth determining unit is used to determine the lip point coordinates in each frame of the corrected head region grayscale image based on the coordinates of the endpoints of the mouth region on both sides in the grayscale image of the head region in each frame and the grayscale integral curve of the mouth region in the vertical direction. The first calculation unit is used to take the lip point, mouth region boundary point, left eye region boundary point and right eye region boundary point in each frame of the corrected head region grayscale image as facial recognition key points, and calculate the pixel distance between each pair of facial recognition key points to obtain multiple sets of key point distances corresponding to each frame of the corrected head region grayscale image. The second calculation unit is used to calculate the ratio between the distance of multiple key points corresponding to each frame of the corrected head region grayscale image and the corresponding binocular distance, so as to obtain the ratio sequence corresponding to each frame of the corrected head region grayscale image. The binocular distance corresponding to the corrected head region grayscale image is the pixel distance between the coordinates of the center point of the left eye iris and the center point of the right eye iris in the corrected head region grayscale image. The third calculation unit is used to calculate the Euclidean distance between the ratio sequence corresponding to the grayscale image of the head region of each frame after correction and the ratio sequence corresponding to the grayscale image of the head region of the known bank customer, and to determine the bank customer identification result of the target user based on the Euclidean distance between the ratio sequence corresponding to the grayscale image of the head region of each frame after correction and the ratio sequence corresponding to the grayscale image of the head region of the known bank customer.
[0015] Thirdly, the present invention provides an electronic device comprising a memory, a processor, and a transceiver sequentially and communicatively connected, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the bank customer identification method as described in the first aspect or any possible design of the first aspect.
[0016] Fourthly, the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, perform the bank customer identification method described in the first aspect or any possible design of the first aspect.
[0017] Fifthly, the present invention provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform the bank customer identification method as described in the first aspect or any possible design of the first aspect.
[0018] Beneficial effects: This invention acquires multiple frames of head region images of a target user to be processed for banking transactions; converts these multiple frames of head region images into grayscale images, obtaining multiple grayscale head region images corresponding one-to-one with the original images; performs multi-threshold binarization processing on each grayscale head region image and then merges the binary images to obtain a merged head region image corresponding to each grayscale head region image; performs connected component analysis on each merged head region image and identifies the connected component region with the largest area located in the upper part of the merged head region image as the hair region; based on the position of the hair region in each merged head region image and the head region in each frame... The positional relationship between the remaining connected regions and the hair region in the binary merged head region image is used to determine the left eye region, right eye region, and mouth region in the corresponding grayscale image of the head region for each frame of the binary merged head region image. Based on the pixel grayscale distribution of the left and right eye regions in each frame of the grayscale image of the head region, the coordinates of the center point of the left iris and the center point of the right iris in each frame of the grayscale image of the head region are determined. Based on the coordinates of the center point of the left and right iris in each frame of the grayscale image of the head region, a planar rotation correction is performed on each frame of the grayscale image of the head region to obtain the corrected grayscale image of the head region corresponding to each frame of the grayscale image of the head region. The coordinates of the center points of the left and right irises in the corrected grayscale images of the head region are used to determine the horizontal coordinate range of the center point of the mouth region in each corrected grayscale image of the head region. The window region located within the mouth region and falling within the horizontal coordinate range in each corrected grayscale image of the head region is then subjected to horizontal integral projection to obtain the vertical grayscale integral curve of the mouth region in each corrected grayscale image of the head region. Based on the coordinates of the two endpoints of the mouth region in each corrected grayscale image of the head region and the vertical grayscale integral curve of the mouth region, the coordinates of the lip point in each corrected grayscale image of the head region are determined. The horizontal coordinates of the center point of the mouth region in each corrected grayscale image of the head region are then determined. The lip point, mouth region boundary point, left eye region boundary point, and right eye region boundary point in the grayscale image of the posterior head region are used as facial recognition key points. The pixel distance between each pair of facial recognition key points is calculated to obtain multiple sets of key point distances corresponding to each frame of the corrected grayscale image of the head region. The ratio between the multiple sets of key point distances corresponding to each frame of the corrected grayscale image of the head region and the corresponding binocular distance is calculated to obtain the ratio sequence corresponding to each frame of the corrected grayscale image of the head region. The binocular distance corresponding to the corrected grayscale image of the head region is the pixel distance between the coordinates of the center point of the left iris and the center point of the right iris in the corrected grayscale image of the head region.The Euclidean distance between the ratio sequence corresponding to the grayscale image of the head region in each frame after correction and the ratio sequence corresponding to the grayscale image of the head region of a known bank customer is calculated. Based on this Euclidean distance, the bank customer identification result of the target user is determined. This method accurately identifies key points in the facial region and, based on the ratio of the distance between each set of key points to the binocular distance, as well as the ratio of the distance between each set of key points to the binocular distance of a known user, facilitates quick and easy identification of the target user. This makes it easier for users to use self-service, avoids queuing congestion during peak hours, improves customer experience and branch operational efficiency, and is suitable for practical application and promotion. Attached Figure Description
[0019] Figure 1 A flowchart illustrating the bank customer identification method provided in this application embodiment; Figure 2 A block diagram of a bank customer identification device provided in an embodiment of this application; Figure 3 This is a block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation
[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in conjunction with the accompanying drawings and descriptions of the embodiments or the prior art. Obviously, the following description of the structure of the accompanying drawings is only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention.
[0021] It should be understood that although the terms first, second, etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit, without departing from the scope of the exemplary embodiments of the invention.
[0022] It should be understood that the term "and / or" that may appear in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" that may appear in this document describes another relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " that may appear in this document generally indicates that the related objects before and after it are in an "or" relationship.
[0023] To facilitate users' access to bank self-service, this application provides a bank customer identification method, device, electronic device, and product. This bank customer identification method, device, electronic device, and product can easily and quickly identify target users, making it convenient for users to use self-service.
[0024] The bank customer identification method provided in this application can be applied to bank self-service terminals. It is understood that the execution entity described does not constitute a limitation on the embodiments of this application.
[0025] The bank customer identification method provided in the embodiments of this application will be described in detail below.
[0026] like Figure 1 The diagram shown is a flowchart of a bank customer identification method provided in the first aspect of the present application. The bank customer identification method may include, but is not limited to, the following steps S101-S113.
[0027] Step S101. Obtain multi-frame head region images of the target user for whom banking business is to be processed.
[0028] In this embodiment, the self-service terminal can be equipped with an image acquisition unit. When a target user conducts self-service business through the self-service terminal, the image acquisition unit on the self-service terminal can acquire multiple frame sample images containing the target user's head region. The head region in the multiple frame sample images can be identified through facial contour detection, and then the head region in the multiple frame sample images can be expanded to obtain multiple frame head region images of the target user to be conducting banking business.
[0029] The facial contour detection and recognition can employ existing conventional facial contour detection algorithms, and no specific limitations are imposed on the embodiments of this application.
[0030] Step S102. Convert the multi-frame head region images into grayscale images to obtain multi-frame head region grayscale images that correspond one-to-one with the multi-frame head region images.
[0031] Step S103. Perform multi-threshold binarization processing on the grayscale image of the head region of each frame, and then merge the binary images to obtain the merged binary image of the head region corresponding to the grayscale image of the head region of each frame.
[0032] Specifically, after performing multi-threshold binarization processing on the grayscale image of the head region of each frame, the binary images are merged, which may include, but is not limited to, the following steps S1031-S1032.
[0033] Step S1031. By using multiple preset binary thresholds, perform multiple binarization processes on the grayscale image of the head region of each frame to obtain a multi-frame binarized image corresponding to the grayscale image of the head region of each frame.
[0034] In this embodiment, multiple binary thresholds can be preset, for example, two binary thresholds of 100 and 150 can be set. Then, the grayscale image of the head region of each frame is binarized multiple times using the multiple binary thresholds to obtain multiple binarized images corresponding to each frame of the grayscale image of the head region. For example, if two binary thresholds are set, the frame image needs to be binarized twice using the two binary thresholds, and after binarization, each frame of the grayscale image of the head region corresponds to two frames of binarized images.
[0035] Step S1032. Perform a logical AND operation on the multi-frame binarized images corresponding to the grayscale image of the head region of each frame to obtain the binary merged image of the head region corresponding to the grayscale image of the head region of each frame.
[0036] Specifically, a logical AND operation can be performed on the pixels corresponding to the same pixel coordinate in the multi-frame binarized images corresponding to the grayscale image of the head region of each frame. If the binarized values of the pixels corresponding to the same pixel coordinate in the multi-frame binarized images are all 1, then the binarized value corresponding to the same pixel coordinate is 1; otherwise, it is 0.
[0037] Understandably, in some other embodiments, the multiple binarized images corresponding to each frame of the grayscale image of the head region can be logically ORed to obtain a binary merged image of the head region corresponding to each frame of the grayscale image of the head region. Alternatively, the multiple binarized images corresponding to each frame of the grayscale image of the head region can be weighted and then normalized to obtain a binary merged image of the head region corresponding to each frame of the grayscale image of the head region.
[0038] Step S104. Perform connected component analysis on each frame of the binary merged head region image, and determine the connected component region with the largest area located in the upper region of the binary merged head region image as the hair region.
[0039] In one or more embodiments, after performing connected component analysis on the binary merged image of the header region of each frame, some connected regions with smaller areas can be removed.
[0040] Step S105. Based on the position of the hair region in the binary merged head region image of each frame, and the positional relationship between the hair region and the other connected regions in the binary merged head region image of each frame, determine the left eye region, right eye region, and mouth region in the grayscale image of the head region corresponding to the binary merged head region image of each frame.
[0041] Specifically, based on the position of the hair region in each frame of the binary merged head region image, the center position of the remaining connected regions in each frame of the binary merged head region image, and the area of the remaining connected regions in each frame of the binary merged head region image, two connected regions located below the hair region, whose corresponding center ordinate difference is lower than a first preset threshold, whose corresponding abscissa difference is within the range of the first preset threshold, and whose corresponding area is greater than a first preset area, can be selected as the initial left eye region and the initial right eye region, respectively. The first preset threshold, the first preset threshold range, and the first preset area can be set empirically.
[0042] Then, the following region in the binary merged image of the head region in each frame, located below the initially selected left and right eye regions, with its ordinate between the center points of the initially selected left and right eye regions, an area greater than a second preset area, and a ratio between the distance between its corresponding center point and the center point of the initially selected left eye region and the center distance between the centers of the initially selected left and right eye regions within a second preset threshold range, and a ratio between the distance between its corresponding center point and the center point of the initially selected right eye region and the center distance between the centers of the initially selected left and right eye regions within a second preset threshold range, can be designated as the initial mouth region. Here, the center distance between the initially selected left and right eye regions is the pixel distance between the center points of the initially selected left and right eye regions, and the second preset area and the second preset threshold range can be set empirically.
[0043] Finally, the region in the grayscale image of the head region that corresponds to the initially selected left eye region in the corresponding binary merged image of the head region can be taken as the left eye region, the region in the grayscale image of the head region that corresponds to the initially selected right eye region in the corresponding binary merged image of the head region can be taken as the right eye region, and the region in the grayscale image of the head region that corresponds to the initially selected mouth region in the corresponding binary merged image of the head region can be taken as the mouth region.
[0044] Therefore, the left eye region, right eye region, and mouth region in the grayscale image of the head region can be identified.
[0045] Step S106. Based on the pixel grayscale distribution of the left and right eye regions in each frame of the head region grayscale image, determine the coordinates of the center point of the left iris and the center point of the right iris in each frame of the head region grayscale image.
[0046] In this embodiment, when determining the coordinates of the center point of the left and right irises in each frame of the grayscale image of the head region, it is possible to first determine the left-eye target pixel in the left-eye region of each frame of the grayscale image of the head region, where all its neighboring pixels are located in the left-eye region, and the right-eye target pixel in the right-eye region of each frame of the grayscale image of the head region, where all its neighboring pixels are located in the right-eye region. The neighborhood radius can be set according to actual conditions.
[0047] Then, the support function corresponding to each left-eye target pixel and the mean gray value of the neighborhood corresponding to each right-eye target pixel can be calculated using the following formula:
[0048] Where A represents the neighborhood range of the target pixel in the left eye or the target pixel in the right eye. This represents the grayscale value of the pixel with coordinates (x, y) within the neighborhood of the left-eye or right-eye target pixel, and N represents the total number of pixels within the neighborhood of the left-eye or right-eye target pixel.
[0049] Finally, the coordinates of the left eye target pixel with the lowest gray-scale mean in its corresponding neighborhood can be used as the coordinates of the center point of the left eye iris in the gray-scale image of the head region, and the coordinates of the right eye target pixel with the lowest gray-scale mean in its corresponding neighborhood can be used as the coordinates of the center point of the right eye iris in the gray-scale image of the head region.
[0050] Step S107. Based on the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the grayscale image of the head region, perform planar rotation correction on each frame of the grayscale image of the head region to obtain the corrected grayscale image of the head region corresponding to each frame of the grayscale image of the head region.
[0051] Specifically, based on the coordinates of the center point of the left iris and the center point of the right iris in each frame of the grayscale image of the head region, the tilt angle corresponding to each frame of the grayscale image of the head region can be determined, that is, the tilt of the line connecting the coordinates of the center point of the left iris and the center point of the right iris.
[0052] Then, based on the tilt angle corresponding to the grayscale image of the head region in each frame, the rotation angle corresponding to the grayscale image of the head region in each frame is determined, and the corresponding grayscale image of the head region is corrected by planar rotation based on the rotation angle corresponding to the grayscale image of the head region in each frame, so as to obtain the corrected grayscale image of the head region corresponding to each frame.
[0053] Step S108. Based on the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the corrected grayscale image of the head region, determine the horizontal coordinate range of the center point of the mouth region in each frame of the corrected grayscale image of the head region.
[0054] In this embodiment, the horizontal coordinate range of the center point of the mouth region in the corrected grayscale image of the head region in each frame can be selected between the horizontal coordinates of the center point of the left iris and the horizontal coordinates of the center point of the right iris in the corrected grayscale image of the head region in the corresponding frame.
[0055] Step S109. Perform horizontal integral projection on the window area located in the mouth region and within the horizontal coordinate range in the grayscale image of the head region of each frame after correction, to obtain the grayscale integral curve of the mouth region in the vertical direction in the grayscale image of the head region of each frame after correction.
[0056] Step S110. Based on the coordinates of the endpoints of the mouth region on both sides in the grayscale image of the head region after correction in each frame and the grayscale integral curve of the mouth region in the vertical direction, determine the coordinates of the lip point in the grayscale image of the head region after correction in each frame.
[0057] Specifically, the center coordinates of the horizontal coordinates of the two endpoints of the mouth region in each frame of the corrected grayscale image of the head region can be used as the horizontal coordinates of the lip points in each frame of the corrected grayscale image of the head region. Then, the vertical coordinate of the corresponding minimum grayscale integral value in the grayscale integral curve can be used as the vertical coordinate of the lip points in each frame of the corrected grayscale image of the head region. Thus, the coordinates of the lip points in each frame of the corrected grayscale image of the head region can be obtained.
[0058] Step S111. Take the lip point, mouth region boundary point, left eye region boundary point and right eye region boundary point in each frame of the corrected head region grayscale image as facial recognition key points, and calculate the pixel distance between each pair of facial recognition key points to obtain multiple sets of key point distances corresponding to each frame of the corrected head region grayscale image.
[0059] For example, the pixel distances between the left and right boundaries of the left eye region can be used as a set of keypoint distances; the pixel distances between the left and right boundaries of the right eye region can be used as a set of keypoint distances; the pixel distances between the upper and lower boundaries of the left eye region can be used as a set of keypoint distances; the pixel distances between the upper and lower boundaries of the right eye region can be used as a set of keypoint distances; the pixel distances between the left and right boundaries of the mouth region can be used as a set of keypoint distances; the pixel distances between the upper and lower boundaries of the mouth region can be used as a set of keypoint distances; the pixel distances between the left and right boundaries of the mouth region and the lip point can be used as a set of keypoint distances; the pixel distances between the left and right boundaries of the mouth region and the lip point can be used as a set of keypoint distances, and so on. Thus, for each frame of the corrected grayscale image of the head region, multiple sets of keypoint distances can be obtained.
[0060] Step S112. Calculate the ratio between the distances of multiple key points corresponding to each frame of the corrected head region grayscale image and the corresponding binocular distance, to obtain the ratio sequence corresponding to each frame of the corrected head region grayscale image.
[0061] Specifically, the binocular distance corresponding to the corrected grayscale image of the head region is the pixel distance between the center coordinates of the left and right irises in the corrected grayscale image of the head region. It should be noted that multiple sets of keypoint distances must be sorted in a specified order.
[0062] In one or more embodiments, after obtaining the ratio sequence corresponding to the grayscale image of the head region in each frame after correction, the ratio sequence corresponding to the grayscale image of the head region in each frame after correction can be normalized.
[0063] Step S113. Calculate the Euclidean distance between the ratio sequence corresponding to the grayscale image of the corrected head region of each frame and the ratio sequence corresponding to the grayscale image of the corrected head region of the known bank customer, and determine the bank customer identification result of the target user based on the Euclidean distance between the ratio sequence corresponding to the grayscale image of the corrected head region of each frame and the ratio sequence corresponding to the grayscale image of the corrected head region of the known bank customer.
[0064] In this embodiment, the ratio sequence corresponding to the corrected head region grayscale image of the known bank customer can be obtained in the same manner as steps S101-S112 described above. After obtaining the ratio sequence corresponding to each frame of the corrected head region grayscale image, the Euclidean distance between the ratio sequence corresponding to each frame of the corrected head region grayscale image and the ratio sequence corresponding to the corrected head region grayscale image of the known bank customer can be calculated. If the Euclidean distance between the ratio sequence corresponding to at least one frame of the corrected head region grayscale image and the corrected head region grayscale image corresponding to a certain known bank customer is lower than a preset distance, then the target user is determined to be the known bank customer.
[0065] In summary, the bank customer identification method provided by this invention can accurately identify key points in the facial region. Based on the ratio of the distance between each set of key points to the binocular distance, and the ratio of the distance between each set of key points to the binocular distance for a known user, it can easily and quickly identify the target user. This facilitates the use of self-service by users, avoids queuing congestion during peak hours, improves customer experience and branch operation efficiency, and is easy to apply and promote in practice.
[0066] Please see Figure 2 The second aspect of this application provides a bank customer identification device, comprising: The acquisition unit is used to acquire multiple frames of head region images of the target user for whom banking business is to be processed. The grayscale conversion unit is used to convert the multi-frame head region images into grayscale images to obtain multi-frame head region grayscale images that correspond one-to-one with the multi-frame head region images. The binarization unit is used to perform multi-threshold binarization processing on the grayscale image of the head region of each frame and then merge the binary images to obtain the head region binary merged image corresponding to the grayscale image of the head region of each frame. The connected component analysis unit is used to perform connected component analysis on each frame of the binary merged head region image, and to determine the connected component region with the largest area that is located in the upper part of the binary merged head region image as the hair region. The first determining unit is used to determine the left eye region, right eye region, and mouth region in the grayscale image of the head region corresponding to each frame of the head region based on the position of the hair region in the binary merged head region image of each frame, and the positional relationship between the hair region and the other connected region regions in the binary merged head region image of each frame. The second determining unit is used to determine the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the head region grayscale image based on the pixel grayscale distribution of the left eye region and the right eye region in each frame of the head region grayscale image. The correction unit is used to perform planar rotation correction on each frame of the grayscale image of the head region based on the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the grayscale image of the head region, so as to obtain the corrected grayscale image of the head region corresponding to each frame of the grayscale image of the head region. The third determining unit is used to determine the horizontal coordinate range of the center point of the mouth region in each frame of the corrected head region grayscale image based on the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the corrected head region grayscale image. The projection unit is used to perform horizontal integral projection on the window area located in the mouth region and within the horizontal coordinate range in each frame of the corrected head region grayscale image to obtain the grayscale integral curve of the mouth region in the vertical direction in each frame of the corrected head region grayscale image. The fourth determining unit is used to determine the lip point coordinates in each frame of the corrected head region grayscale image based on the coordinates of the endpoints of the mouth region on both sides in the grayscale image of the head region in each frame and the grayscale integral curve of the mouth region in the vertical direction. The first calculation unit is used to take the lip point, mouth region boundary point, left eye region boundary point and right eye region boundary point in each frame of the corrected head region grayscale image as facial recognition key points, and calculate the pixel distance between each pair of facial recognition key points to obtain multiple sets of key point distances corresponding to each frame of the corrected head region grayscale image. The second calculation unit is used to calculate the ratio between the distance of multiple key points corresponding to each frame of the corrected head region grayscale image and the corresponding binocular distance, so as to obtain the ratio sequence corresponding to each frame of the corrected head region grayscale image. The binocular distance corresponding to the corrected head region grayscale image is the pixel distance between the coordinates of the center point of the left eye iris and the center point of the right eye iris in the corrected head region grayscale image. The third calculation unit is used to calculate the Euclidean distance between the ratio sequence corresponding to the grayscale image of the head region of each frame after correction and the ratio sequence corresponding to the grayscale image of the head region of the known bank customer, and to determine the bank customer identification result of the target user based on the Euclidean distance between the ratio sequence corresponding to the grayscale image of the head region of each frame after correction and the ratio sequence corresponding to the grayscale image of the head region of the known bank customer.
[0067] The working process, working details and technical effects of the bank customer identification device provided in the second aspect of this embodiment can be found in the first aspect of the embodiment, and will not be repeated here.
[0068] like Figure 3 As shown, a third aspect of this application provides an electronic device, including a memory, a processor, and a transceiver that are sequentially and communicatively connected, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the bank customer identification method as described in the first aspect of the embodiment.
[0069] Specifically, the memory may include, but is not limited to, random access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out (FIFO) memory, and / or last-in-first-out (FILO) memory, etc.; the processor may not be limited to microprocessors of the STM32F105 series, ARM (Advanced RISC Machines), x86 architecture processors, or processors with integrated NPU (neural-network processing units); the transceiver may be, but is not limited to, WiFi (Wireless Fidelity) wireless transceivers, Bluetooth wireless transceivers, General Packet Radio Service (GPRS) wireless transceivers, ZigBee (a low-power LAN protocol based on the IEEE 802.15.4 standard), 3G transceivers, 4G transceivers, and / or 5G transceivers, etc.
[0070] This fourth aspect of the embodiment provides a computer-readable storage medium storing instructions containing the bank customer identification method described in the first aspect of the embodiment. Specifically, the computer-readable storage medium stores instructions that, when executed on a computer, perform the bank customer identification method as described in the first aspect. The computer-readable storage medium refers to a data storage medium, which may include, but is not limited to, floppy disks, optical disks, hard disks, flash memory, USB flash drives, and / or Memory Sticks. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
[0071] The fifth aspect of this embodiment provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform the bank customer identification method as described in the first aspect of this embodiment, wherein the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.
[0072] It should be understood that specific details are provided in the following description to facilitate a complete understanding of the exemplary embodiments. However, those skilled in the art will understand that the exemplary embodiments can be implemented without these specific details. For example, the system may be shown in block diagrams to avoid obscuring the example with unnecessary details. In other instances, well-known processes, structures, and techniques may be shown without unnecessary details to avoid obscuring the exemplary embodiments.
[0073] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for identifying bank customers, characterized in that, include: Acquire multi-frame head region images of the target user for whom banking business is pending; The multi-frame head region images are converted into grayscale images to obtain multi-frame head region grayscale images that correspond one-to-one with the multi-frame head region images. After performing multi-threshold binarization on the grayscale image of the head region of each frame, the binary images are merged to obtain the binary merged image of the head region corresponding to the grayscale image of the head region of each frame. Perform connected component analysis on each frame of the binary merged head region image, and determine the connected component region with the largest area that is located in the upper part of the binary merged head region image as the hair region. Based on the position of the hair region in the binary merged head region image of each frame, and the positional relationship between the hair region and the other connected regions in the binary merged head region image of each frame, the left eye region, right eye region and mouth region in the grayscale image of the head region corresponding to the binary merged head region image of each frame are determined. Based on the pixel grayscale distribution of the left and right eye regions in each frame of the head region grayscale image, the coordinates of the center point of the left iris and the center point of the right iris in each frame of the head region grayscale image are determined. Based on the coordinates of the center point of the left and right irises in each frame of the grayscale image of the head region, planar rotation correction is performed on each frame of the grayscale image of the head region to obtain the corrected grayscale image of the head region corresponding to each frame of the grayscale image of the head region. Based on the coordinates of the center point of the left and right irises in each frame of the corrected grayscale image of the head region, the horizontal coordinate range of the center point of the mouth region in each frame of the corrected grayscale image of the head region is determined. The window region located in the mouth region and within the horizontal coordinate range in each frame of the corrected head region grayscale image is integrally projected in the horizontal direction to obtain the grayscale integral curve of the mouth region in the vertical direction in each frame of the corrected head region grayscale image. Based on the coordinates of the endpoints of the mouth region on both sides in the grayscale image of the head region after correction in each frame, and the grayscale integral curve of the mouth region in the vertical direction, the coordinates of the lip point in the grayscale image of the head region after correction in each frame are determined. The lip point, mouth region boundary point, left eye region boundary point, and right eye region boundary point in each frame of the corrected grayscale image of the head region are used as facial recognition key points. The pixel distance between each pair of facial recognition key points is calculated to obtain multiple sets of key point distances corresponding to each frame of the corrected grayscale image of the head region. Calculate the ratio between the distances of multiple key points corresponding to each frame of the corrected head region grayscale image and the corresponding binocular distance to obtain the ratio sequence corresponding to each frame of the corrected head region grayscale image. The binocular distance corresponding to the corrected head region grayscale image is the pixel distance between the coordinates of the center point of the left iris and the center point of the right iris in the corrected head region grayscale image. Calculate the Euclidean distance between the ratio sequence corresponding to the grayscale image of the corrected head region of each frame and the ratio sequence corresponding to the grayscale image of the corrected head region of the known bank customer, and determine the bank customer identification result of the target user based on the Euclidean distance between the ratio sequence corresponding to the grayscale image of the corrected head region of each frame and the ratio sequence corresponding to the grayscale image of the corrected head region of the known bank customer.
2. The bank customer identification method according to claim 1, characterized in that, After performing multi-threshold binarization on the grayscale image of the head region in each frame, the binary images are merged to obtain the merged binary image of the head region corresponding to the grayscale image of the head region in each frame, including: By using multiple pre-set binary thresholds, the grayscale image of the head region of each frame is binarized multiple times to obtain a multi-frame binarized image corresponding to the grayscale image of the head region of each frame. Perform logical AND, OR, and OR operations on the multiple binarized images corresponding to each frame of the head region grayscale image to obtain the merged binary image of the head region corresponding to each frame of the head region grayscale image; or The multi-frame binarized images corresponding to the grayscale image of the head region of each frame are weighted and then normalized to obtain the merged binary image of the head region corresponding to the grayscale image of the head region of each frame.
3. The bank customer identification method according to claim 1, characterized in that, Based on the position of the hair region in the binary merged head region image of each frame, and the positional relationship between the hair region and other connected regions in the binary merged head region image of each frame, the left eye region, right eye region, and mouth region in the corresponding grayscale head region image of each frame are determined, including: Based on the position of the hair region in each frame of the binary merged head region image, the center position of the other connected regions in each frame of the binary merged head region image, and the area of the other connected regions in each frame of the binary merged head region image, two connected regions in each frame of the binary merged head region image that are located below the hair region, whose corresponding center ordinate difference is lower than a first preset threshold, whose corresponding abscissa difference is within the range of the first preset threshold, and whose corresponding area is greater than the first preset area are respectively selected as the initial left eye region and the initial right eye region. In each frame of the binary merged head region image, the following region is selected as the initial mouth region: located below the initial left eye region and the initial right eye region, with its vertical coordinate between the center points of the initial left eye region and the initial right eye region, its corresponding area is greater than the second preset area, and the ratio of the distance between the corresponding center point and the center point of the initial left eye region to the center distance of the initial left and right eye regions is within the second preset threshold range, and the ratio of the distance between the corresponding center point and the center point of the initial right eye region to the center distance of the initial left and right eye regions is within the second preset threshold range. The center distance of the initial left and right eye regions is the pixel distance between the center points of the initial left eye region and the initial right eye region. The region in the grayscale image of the head region corresponding to the initially selected left eye region in the corresponding binary merged image of the head region is taken as the left eye region; the region in the grayscale image of the head region corresponding to the initially selected right eye region in the corresponding binary merged image of the head region is taken as the right eye region; and the region in the grayscale image of the head region corresponding to the initially selected mouth region in the corresponding binary merged image of the head region is taken as the mouth region.
4. The bank customer identification method according to claim 1, characterized in that, Based on the pixel grayscale distribution of the left and right eye regions in each frame of the head region grayscale image, the coordinates of the center point of the left and right iris in each frame of the head region grayscale image are determined, including: In each frame of the grayscale image of the head region, all neighboring pixels in the left eye region are located in the left eye target pixel of the left eye region, and in each frame of the grayscale image of the head region, all neighboring pixels in the right eye region are located in the right eye target pixel of the right eye region. The support function corresponding to each left-eye target pixel and the mean gray value of the neighborhood corresponding to each right-eye target pixel are calculated according to the following formulas. Where A represents the neighborhood range of the target pixel in the left eye or the target pixel in the right eye. This represents the grayscale value of the pixel with coordinates (x, y) within the neighborhood of the left-eye target pixel or the right-eye target pixel, and N represents the total number of pixels within the neighborhood of the left-eye target pixel or the right-eye target pixel. The coordinates of the left eye target pixel with the lowest gray-scale mean in its corresponding neighborhood are taken as the coordinates of the center point of the left eye iris in the gray-scale image of the head region, and the coordinates of the right eye target pixel with the lowest gray-scale mean in its corresponding neighborhood are taken as the coordinates of the center point of the right eye iris in the gray-scale image of the head region.
5. The bank customer identification method according to claim 1, characterized in that, Based on the coordinates of the center point of the left and right irises in each frame of the head region grayscale image, planar rotation correction is performed on each frame of the head region grayscale image to obtain the corrected head region grayscale image corresponding to each frame of the head region grayscale image, including: Based on the coordinates of the center point of the left iris and the center point of the right iris in each frame of the grayscale image of the head region, the tilt angle corresponding to each frame of the grayscale image of the head region is determined. Based on the tilt angle corresponding to the grayscale image of the head region in each frame, the rotation angle corresponding to the grayscale image of the head region in each frame is determined, and the corresponding grayscale image of the head region is corrected by planar rotation based on the rotation angle corresponding to the grayscale image of the head region in each frame, so as to obtain the corrected grayscale image of the head region corresponding to each frame.
6. The bank customer identification method according to claim 1, characterized in that, Based on the Euclidean distance between the ratio sequence corresponding to the grayscale image of the corrected head region of each frame and the ratio sequence corresponding to the grayscale image of the corrected head region of a known bank customer, the bank customer identification result of the target user is determined, including: If the Euclidean distance between the ratio sequence corresponding to at least one frame of the corrected grayscale image of the head region and the corrected grayscale image of the head region corresponding to a known bank customer is lower than a preset distance, then the target user is determined to be the known bank customer.
7. The bank customer identification method according to claim 1, characterized in that, Obtain multi-frame head region images of the target user for whom banking business is to be processed, including: Acquire multiple sample images containing the head region of the target user; The head region in the multi-frame sample images is identified by facial contour detection. The head region in the multi-frame sample images is expanded to obtain multi-frame head region images of the target user to be processed for banking business.
8. A bank customer identification device, characterized in that, include: The acquisition unit is used to acquire multiple frames of head region images of the target user for whom banking business is to be processed. The grayscale conversion unit is used to convert the multi-frame head region images into grayscale images to obtain multi-frame head region grayscale images that correspond one-to-one with the multi-frame head region images. The binarization unit is used to perform multi-threshold binarization processing on the grayscale image of the head region of each frame and then merge the binary images to obtain the head region binary merged image corresponding to the grayscale image of the head region of each frame. The connected component analysis unit is used to perform connected component analysis on each frame of the binary merged head region image, and to determine the connected component region with the largest area that is located in the upper part of the binary merged head region image as the hair region. The first determining unit is used to determine the left eye region, right eye region, and mouth region in the grayscale image of the head region corresponding to each frame of the head region based on the position of the hair region in the binary merged head region image of each frame, and the positional relationship between the hair region and the other connected region regions in the binary merged head region image of each frame. The second determining unit is used to determine the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the head region grayscale image based on the pixel grayscale distribution of the left eye region and the right eye region in each frame of the head region grayscale image. The correction unit is used to perform planar rotation correction on each frame of the grayscale image of the head region based on the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the grayscale image of the head region, so as to obtain the corrected grayscale image of the head region corresponding to each frame of the grayscale image of the head region. The third determining unit is used to determine the horizontal coordinate range of the center point of the mouth region in each frame of the corrected head region grayscale image based on the coordinates of the center point of the left eye iris and the center point of the right eye iris in each frame of the corrected head region grayscale image. The projection unit is used to perform horizontal integral projection on the window area located in the mouth region and within the horizontal coordinate range in each frame of the corrected head region grayscale image to obtain the grayscale integral curve of the mouth region in the vertical direction in each frame of the corrected head region grayscale image. The fourth determining unit is used to determine the lip point coordinates in each frame of the corrected head region grayscale image based on the coordinates of the endpoints of the mouth region on both sides in the grayscale image of the head region in each frame and the grayscale integral curve of the mouth region in the vertical direction. The first calculation unit is used to take the lip point, mouth region boundary point, left eye region boundary point and right eye region boundary point in each frame of the corrected head region grayscale image as facial recognition key points, and calculate the pixel distance between each pair of facial recognition key points to obtain multiple sets of key point distances corresponding to each frame of the corrected head region grayscale image. The second calculation unit is used to calculate the ratio between the distance of multiple key points corresponding to each frame of the corrected head region grayscale image and the corresponding binocular distance, so as to obtain the ratio sequence corresponding to each frame of the corrected head region grayscale image. The binocular distance corresponding to the corrected head region grayscale image is the pixel distance between the coordinates of the center point of the left eye iris and the center point of the right eye iris in the corrected head region grayscale image. The third calculation unit is used to calculate the Euclidean distance between the ratio sequence corresponding to the grayscale image of the head region of each frame after correction and the ratio sequence corresponding to the grayscale image of the head region of the known bank customer, and to determine the bank customer identification result of the target user based on the Euclidean distance between the ratio sequence corresponding to the grayscale image of the head region of each frame after correction and the ratio sequence corresponding to the grayscale image of the head region of the known bank customer.
9. An electronic device, characterized in that, The device includes a memory, a processor, and a transceiver that are sequentially and communicatively connected. The memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the bank customer identification method as described in any one of claims 1 to 7.
10. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or the instructions are executed by the computer, they implement the bank customer identification method as described in any one of claims 1 to 7.