CN113778591A

CN113778591A - Method, device, server and storage medium for obtaining display card surface

Info

Publication number: CN113778591A
Application number: CN202110971051.9A
Authority: CN
Inventors: 易晨辉; 章澄; 宋伟男; 江航; 刘志群
Original assignee: China Unionpay Co Ltd
Current assignee: China Unionpay Co Ltd
Priority date: 2021-08-23
Filing date: 2021-08-23
Publication date: 2021-12-10
Anticipated expiration: 2041-08-23
Also published as: CN113778591B

Abstract

The application discloses a method, a device, a server and a storage medium for obtaining a display card surface, and belongs to the field of data processing. The method comprises the following steps: acquiring a target picture, wherein the target picture is a shot picture of the card surface of the entity card; extracting a first key point and a first descriptor vector of a target picture by using a feature extraction algorithm; matching the target picture with the card surface sample picture in the card surface database based on the first descriptor vector and the card surface sample data in the card surface database to obtain the matching similarity of the target picture and the card surface sample picture; selecting the first N card surface sample pictures as alternative card surface pictures according to the sequence of the matching similarity from high to low; and sending the alternative card surface picture to the user terminal, and enabling the user terminal to display the alternative card surface picture for the user to select so as to determine that the entity card displays the card surface picture at the user terminal. According to the embodiment of the application, the maintenance cost is reduced on the basis of providing the display card surface picture required by the user terminal to display the entity card.

Description

Method, device, server and storage medium for obtaining display card surface

Technical Field

The present application belongs to the field of data processing, and in particular, to a method, an apparatus, a server, and a storage medium for obtaining a display card surface.

Background

In order to meet the user's needs for the physical card image, the card issuing organization issues some physical cards with personalized card surfaces. With the development of electronic information technology, a user binds an entity card by using an application installed in a terminal device, and realizes the function of the bound entity card through the operation of the application.

In order to improve the visibility of the entity card in the application program of the terminal device, the personalized card surface of the entity card bound by the user can be displayed under the condition that the application program of the terminal device is operated. At present, a background server of the application program needs to determine a card surface number corresponding to the card number by using the card number of the physical card and using a corresponding relationship between the card number and the card surface number in a server of the card issuing institution through a special query interface, and determine a card surface picture corresponding to the card surface number according to the card surface number, and provide the card surface picture to the user terminal, so that the user terminal can display the card surface picture.

However, the acquisition of the personalized card surface of the entity card can be realized only by depending on the card number and the service of the server of the card issuing institution, and the server of the card issuing institution needs to be maintained on the basis of maintaining the background server, so that the maintenance cost is high.

Disclosure of Invention

The embodiment of the application provides a method, a device, a server and a storage medium for obtaining a display card surface, which can reduce the maintenance cost on the basis of providing a display card surface picture required by a user terminal to display an entity card.

In a first aspect, an embodiment of the present application provides a method for obtaining a display card surface, including: the method comprises the steps that a server obtains a target picture, wherein the target picture is a shot picture of a card surface of an entity card; the server extracts a first key point and a first descriptor vector of the target picture by using a feature extraction algorithm, wherein the first descriptor vector is a descriptor vector of the first key point; the server matches the target picture with a card face sample picture in a card face database based on the first descriptor vector and card face sample data in the card face database established in advance in the server to obtain matching similarity of the target picture and the card face sample picture, wherein the card face sample data is used for representing the card face sample picture; the server selects the first N card face sample pictures as alternative card face pictures according to the sequence of the matching similarity from high to low, wherein N is a positive integer; the server sends the alternative card face picture to the user terminal, and the user terminal is enabled to display the alternative card face picture for the user to select so as to determine that the entity card displays the card face picture at the user terminal.

In a second aspect, an embodiment of the present application provides an apparatus for obtaining a display card surface, including: the acquisition module is used for acquiring a target picture, and the target picture is a shot picture of the card surface of the entity card; the extraction module is used for extracting a first key point and a first descriptor vector of the target picture by using a feature extraction algorithm, wherein the first descriptor vector is a descriptor vector of the first key point; the matching module is used for matching the target picture with the card surface sample picture in the card surface database based on the first descriptor vector and the card surface sample data in the card surface database which is pre-established in the device for displaying the card surface, so as to obtain the matching similarity between the target picture and the card surface sample picture, and the card surface sample data is used for representing the card surface sample picture; the selecting module is used for selecting the first N card surface sample pictures as alternative card surface pictures according to the sequence of the matching similarity from high to low, wherein N is a positive integer; and the sending module is used for sending the alternative card surface picture to the user terminal, so that the user terminal displays the alternative card surface picture for the user to select so as to determine the card surface picture displayed by the entity card at the user terminal.

In a third aspect, an embodiment of the present application provides a server, where the server includes: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements the method of obtaining a display card surface of the first aspect.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method for obtaining a display card surface in the first aspect is implemented.

The embodiment of the application provides a method, a device, a server and a storage medium for obtaining a display card surface, wherein the server extracts key points and descriptor sub-vectors of a picture of the card surface of a shot entity card, matches the picture of the card surface of the shot entity card with card surface sample pictures in a card surface database based on a first descriptor sub-vector and card surface sample data in a preset card surface database, and sends N card surface sample pictures with highest matching similarity with the picture of the card surface of the shot entity card as alternative card surface pictures to a user terminal, so that the user terminal can determine the display card surface picture used by the entity card displayed by the user terminal in the alternative card surface pictures. The server provides the display card surface picture required by the display entity card for the user terminal through the image identification technology, the acquisition of the display card surface picture does not depend on the card number of the entity card and the service of the server of the card issuing mechanism, only the server for image identification needs to be maintained, and the maintenance cost is reduced on the basis that the display card surface picture required by the display entity card of the user terminal can be provided.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly described below, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic diagram of an example of an application interface displayed by a user terminal according to an embodiment of the present application;

fig. 2 is a schematic diagram of an example of an application scenario of a method for obtaining a display card surface according to an embodiment of the present application;

FIG. 3 is a flowchart of an embodiment of a method for obtaining a display card surface provided herein;

FIG. 4 is a flow chart of another embodiment of a method for obtaining a display card face provided herein;

FIG. 5 is a flow chart of yet another embodiment of a method for obtaining a display card surface provided herein;

FIG. 6 is a flowchart of yet another embodiment of a method for obtaining a display card surface provided herein;

FIG. 7 is a diagram illustrating an example of a card face of a card and a common element and entity provided by an embodiment of the present application;

FIG. 8 is a flow chart of yet another embodiment of a method of obtaining a display card face provided herein;

FIG. 9 is a flowchart of yet another embodiment of a method for obtaining a display card surface provided herein;

FIG. 10 is a flow chart of yet another embodiment of a method for obtaining a display card face provided herein;

fig. 11 is a flowchart of an example of a card surface database establishment process in the embodiment of the present application;

fig. 12 is a flowchart of an example of a flow of obtaining a display card surface in the method for obtaining a display card surface according to the embodiment of the present application;

FIG. 13 is a schematic structural diagram illustrating an embodiment of an apparatus for obtaining a display card surface according to the present disclosure;

FIG. 14 is a schematic structural diagram of another embodiment of an apparatus for obtaining a display card surface according to the present application;

FIG. 15 is a schematic structural diagram illustrating a further embodiment of an apparatus for obtaining a display card surface according to the present application;

FIG. 16 is a schematic structural diagram illustrating a further embodiment of an apparatus for obtaining a display card surface according to the present application;

FIG. 17 is a schematic structural diagram of yet another embodiment of an apparatus for obtaining a display card surface provided herein;

fig. 18 is a schematic structural diagram of an embodiment of a server provided in the present application.

Detailed Description

Features and exemplary embodiments of various aspects of the present application will be described in detail below, and in order to make objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are intended to be illustrative only and are not intended to be limiting. It will be apparent to one skilled in the art that the present application may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present application by illustrating examples thereof.

In order to meet the user's needs for the physical card image, the card issuing organization issues some physical cards with personalized card surfaces. With the development of electronic information technology, a user binds an entity card by using an application installed in a terminal device, and realizes the function of the bound entity card through the operation of the application. In order to improve the visibility of the entity card in the application program of the terminal device, the personalized card surface of the entity card bound by the user can be displayed under the condition that the application program of the terminal device is operated. Fig. 1 is a schematic diagram of an example of an application interface displayed by a user terminal according to an embodiment of the present application. As shown in fig. 1, the application program interface displayed by the user terminal includes two personalized card surface pictures of the bank card and two airline service membership cards, so that the user can visually confirm the entity card corresponding to the personalized card surface.

The user terminal can display the personalized card face picture only after acquiring the personalized card face picture corresponding to the entity card. The user terminal needs to initiate a request to the background server of the application. The background server sends the card number of the entity card corresponding to the personalized card surface requested by the user terminal to the server of the card issuing organization through a special query interface between the background server and the server of the card issuing organization. The server of the card issuing mechanism determines the card surface number corresponding to the requested card number by using the corresponding relation between the card number and the card surface number, acquires the corresponding card surface picture according to the card surface number, and sends the card surface picture to the user terminal through the background server so that the user terminal can display the card surface picture. The function of the background server for acquiring the card surface picture required to be displayed by the user terminal is closely coupled with the service provided by the server of the card issuing mechanism, namely, the acquisition of the personalized card surface can be realized only by depending on the card number and the service of the server of the card issuing mechanism, so that the system structure required for acquiring the personalized card surface is more complicated, the server of the card issuing mechanism needs to be maintained on the basis of maintaining the background server, and the maintenance cost is higher.

The application provides a method, a device, a server and a storage medium for obtaining a display card surface, which can identify and obtain a card surface sample picture corresponding to a shot card surface picture of an entity card in a card surface database by using the shot card surface picture of the entity card, and provide the card surface sample picture for a user terminal so that the user terminal can display the card surface picture corresponding to the entity card.

An application scenario of the method for obtaining the display card surface provided by the embodiment of the application can relate to a user terminal and a server. Fig. 2 is a schematic diagram of an example of an application scenario of a method for obtaining a display card surface according to an embodiment of the present application. As shown in fig. 2, the user terminal 11 may interact with the server 12 to transfer a picture of a card face of a photographed physical card, a picture of a display card face displayed at the user terminal 11, and the like.

The user terminal 11 may be installed with an application program, which may be bound with the physical card for implementing the function of the physical card. The type of the physical card is not limited herein, for example, the physical card may be a bank card, a membership card, etc. The terminal 11 may specifically include a device with a display function, such as a mobile phone, a computer, a tablet computer, and the like, but is not limited thereto.

The server 12 may be a background server for applications installed in the user terminal 11. The server 12 stores a card surface database, which may include card surface sample data of a card surface sample picture and a card surface sample picture. The card surface sample data of the card surface sample picture and the card surface sample picture can be used for matching and identifying the displayed card surface picture displayed by the user terminal 11.

The method, the apparatus, the server and the storage medium for obtaining the display card surface provided by the embodiment of the present application are sequentially described below.

The application provides a method for obtaining a display card surface, which can be applied to a server, namely the method for obtaining the display card surface can be executed by the server. Fig. 3 is a flowchart of an embodiment of a method for obtaining a display card surface according to the present disclosure. As shown in fig. 3, the method for obtaining the display card surface may include steps S201 to S205.

In step S201, the server acquires a target picture.

The target picture is a shot picture of the card surface of the entity card. In some examples, the server may obtain the target picture from the user terminal. Specifically, the user terminal may photograph the card surface of the entity card to obtain a photographed picture of the card surface of the entity card, and send the picture, i.e., the target picture, to the server, so that the server obtains the target picture.

In step S202, the server extracts a first key point and a first descriptor vector of the target picture by using a feature extraction algorithm.

The feature extraction algorithm is an algorithm that can extract the key points in the picture and the descriptor vectors of the key points, and is not limited herein. In some examples, the key points and the descriptor vectors of the target picture may be extracted using a Scale-Invariant Feature Transform (SIFT) algorithm, a Speeded Up Robust Features (SURF) algorithm, or other Feature extraction algorithms, but the Feature extraction algorithms are not limited to the SIFT algorithm and the SURF algorithm.

The first key point is a feature point of the target picture. The first descriptor vector is a descriptor vector of the first keypoint and can be used to describe the first keypoint. In step S202, a plurality of first keypoints and a plurality of first descriptor vectors of the target picture may be extracted, and the first descriptor vectors may correspond to the first keypoints one to one.

In step S203, the server matches the target picture with the card surface sample picture in the card surface database based on the first descriptor vector and the card surface sample data in the card surface database pre-established in the server, so as to obtain a matching similarity between the target picture and the card surface sample picture.

The server stores a card face database established in advance. The card face database may include a card face sample picture and card face sample data corresponding to the card face sample picture. The card face sample picture is a picture as a sample of the card face of the physical card. Card face sample data can be obtained according to the card face sample picture, can be used for the card face sample picture of characterization. In some examples, the card face sample data may include key points extracted from the card face sample picture and descriptor vectors of the key points. In other examples, the key points extracted from the card face sample picture and the descriptor vectors of the key points may be processed, and the processed data may be used as the card face sample data.

Under the condition that the card surface database comprises more than two card surface sample pictures, the target picture can be sequentially matched with the card surface sample pictures in the card surface database. In some examples, the matching of the target picture and the card-face sample picture may be specifically realized as matching of a first descriptor vector of the target picture and the card-face sample data corresponding to the card-face sample picture. In other examples, matching between the target picture and the card surface sample picture may be specifically implemented as matching between data obtained after the first descriptor vector of the target picture is processed and the card surface sample data corresponding to the card surface sample picture. The implementation manner of matching the target picture and the card surface sample picture may be determined according to the type of the card surface sample data, and is not limited herein.

The matching similarity between the target picture and the card surface sample picture is a parameter capable of reflecting the similarity between the target picture and the card surface sample picture. The higher the matching similarity between the target picture and the card surface sample picture is, the higher the possibility that the card surface of the entity card in the target picture is consistent with the card surface sample picture is. The lower the matching similarity between the target picture and the card surface sample picture is, the lower the possibility that the card surface of the entity card in the target picture is consistent with the card surface sample picture is.

In step S204, the server selects the first N card face sample pictures as alternative card face pictures according to the sequence of the matching similarity from high to low.

The alternative card face picture is a card face sample picture for providing the user with a selection. The first N card face sample pictures with the similarity from high to low are the N card face sample pictures with the highest similarity with the card face of the entity card in the target picture.

N is a positive integer. Specifically, N may be 1 or an integer of 2 or more, and is not limited herein.

In step S205, the server sends the alternative card face picture to the user terminal, so that the user terminal displays the alternative card face picture for the user to select, so as to determine that the entity card displays the card face picture at the user terminal.

The alternative card face picture can be used for providing a user with a selection to obtain a display card face picture of the entity card at the user terminal. And displaying the card surface picture as a card surface picture consistent with the card surface of the entity card in the target picture. And when the user terminal needs to display the entity card, the user terminal displays the card surface picture.

The card face pictures consistent with the card face of the entity card in the target picture have a very high possibility in the first N card face sample pictures with the similarity from high to low, but the situation that the card face pictures consistent with the card face of the entity card in the target picture are not found in the first N card face sample pictures with the similarity from high to low exists, and in order to avoid displaying wrong card face pictures on the user terminal, the selected alternative card face pictures can be sent to the user terminal. The user terminal can display N optional card surface pictures for the user to select.

In some examples, N ═ 1, the user terminal determines, in response to a confirmation input by the user, the alternative card face picture as the presentation card face picture of the physical card at the user terminal. Namely, the alternative card face picture is the card face picture consistent with the card face of the entity card in the target picture.

In other examples, N is an integer greater than or equal to 2, and the user terminal determines, in response to a selection input by the user, one of the N candidate card surface pictures indicated by the selection input as the display card surface picture of the physical card at the user terminal. Namely, the N candidate card surface pictures comprise card surface pictures consistent with the card surfaces of the entity cards in the target picture.

In still other examples, N is a positive integer, and the user terminal determines that the picture of the display card surface of the physical card at the user terminal is not obtained in response to a negative input from the user. Namely, the card surface pictures which are consistent with the card surface of the entity card in the target picture are not included in the N candidate card surface pictures.

In the embodiment of the application, the server extracts key points and descriptor vectors of a shot card face picture of the entity card, matches the shot card face picture of the entity card with a card face sample picture in a card face database based on a first descriptor vector and card face sample data in a preset card face database, and sends N card face sample pictures with highest matching similarity with the shot card face picture of the entity card to the user terminal as alternative card face pictures so that the user terminal can determine a display card face picture used by the entity card displayed at the user terminal in the alternative card face pictures. The server provides the display card surface picture required by the display entity card for the user terminal through the image identification technology, the acquisition of the display card surface picture does not depend on the card number of the entity card and the service of the server of the card issuing mechanism, only the server for image identification needs to be maintained, and the maintenance cost is reduced on the basis that the display card surface picture required by the display entity card of the user terminal can be provided.

In some embodiments, the card face sample data may include keypoints and descriptor vectors extracted from the card face sample picture, and correspondingly, the server may match the card face sample data with the first descriptor vector. Fig. 4 is a flowchart of another embodiment of a method for obtaining a display card surface provided in the present application. Fig. 4 is different from fig. 3 in that step S203 in fig. 3 can be specifically detailed as step S2031 to step S2034 in fig. 4.

In step S2031, the server calculates euclidean distances between the first descriptor vector and the second descriptor vectors corresponding to the respective card-face sample pictures.

In some examples, the target picture has a plurality of first keypoints, each of which may correspond to one first descriptor vector. Each card face sample picture may have a plurality of second keypoints, each corresponding to one second descriptor vector. Specifically, the euclidean distance between each first descriptor vector and each second descriptor vector corresponding to each card surface sample picture can be calculated. The smaller the Euclidean distance between a first descriptor vector and a second descriptor vector is, the higher the matching degree of a first key point corresponding to the first descriptor vector and a second key point corresponding to the second descriptor vector is.

In step S2032, the server selects the first M second descriptor vectors in the order of decreasing euclidean distance.

M is a positive integer. Specifically, for each first descriptor sub-vector, the first M second descriptor sub-vectors are selected in order of decreasing euclidean distance from the first descriptor sub-vector to increasing euclidean distance from the first descriptor sub-vector. The Euclidean distances between the second descriptor vector and the first descriptor vector are in the order from small to large, that is, the matching degree between the second key point corresponding to the second descriptor vector and the first key point corresponding to the first descriptor vector is in the order from high to low. For example, the number of the first descriptor vectors is P, and M second descriptor vectors are obtained by selecting each first descriptor vector, so that P × M second descriptor vectors can be obtained by selecting P first descriptor vectors.

In step S2033, the server determines whether the first keypoint corresponding to the first descriptor vector matches the second keypoint corresponding to the selected second descriptor vector according to the euclidean distance between the selected second descriptor vector and the first descriptor vector.

For each first descriptor vector, whether a first keypoint corresponding to the first descriptor vector matches a second keypoint corresponding to a selected second descriptor vector may be determined according to the euclidean distance between the first descriptor vector and the selected second descriptor vector.

In some examples, M ═ 1, i.e., one second descriptor vector is chosen for one first descriptor vector. And the server determines that a first key point corresponding to the first descriptor vector is matched with a second key point corresponding to the selected second descriptor vector under the condition that the Euclidean distance between the selected second descriptor vector and the first descriptor vector is smaller than or equal to a matching distance threshold.

The matching distance threshold is a determination threshold for determining whether the first key point and the second key point are matched, and may be set according to a scene, a requirement, experience, and the like, which is not limited herein. The Euclidean distance between the first descriptor vector and the second descriptor vector is smaller than a matching distance threshold value, and a first key point corresponding to the first descriptor vector is matched with a second key point corresponding to the second descriptor vector; the Euclidean distance between the first descriptor vector and the second descriptor vector is larger than the matching distance threshold value, and the fact that a first key point corresponding to the first descriptor vector is not matched with a second key point corresponding to the second descriptor vector is represented.

In other examples, M-2, i.e. two second descriptor vectors are chosen for one first descriptor vector. And the server calculates a Euclidean distance ratio of the first Euclidean distance to the second Euclidean distance, and determines that a first key point corresponding to the first descriptor vector is matched with a second key point corresponding to the obtained first second descriptor vector when the Euclidean distance ratio is smaller than or equal to a matching ratio threshold value. The first Euclidean distance is the Euclidean distance between the first descriptor vector and the second descriptor vector, and the second Euclidean distance is the Euclidean distance between the second descriptor vector and the first descriptor vector. For example, for the first descriptor vector a1, the second descriptor vector b1 and the second descriptor vector b2 are selected in descending order of euclidean distance from the first descriptor vector a1, the euclidean distance between the first descriptor vector a1 and the second descriptor vector b1 is smaller than the euclidean distance between the first descriptor vector a1 and the second descriptor vector b2, the euclidean distance between the first descriptor vector a1 and the second descriptor vector b1 is the first euclidean distance, and the euclidean distance between the first descriptor vector a1 and the second descriptor vector b2 is the second euclidean distance.

The matching ratio threshold is a determination threshold for determining whether the second keypoint has uniqueness so as to match the first keypoint, and may be set according to a scene, a requirement, experience, and the like, which is not limited herein. For example, the matching ratio threshold is set to 0.5. The Euclidean distance ratio is the ratio of the first Euclidean distance to the second Euclidean distance. The Euclidean distance ratio corresponding to a first descriptor vector can reflect the difference between a selected second descriptor vector closest to the first descriptor vector and a second descriptor vector second closest to the first descriptor vector. The Euclidean distance ratio is less than or equal to a matching ratio threshold value, which indicates that the difference between a second descriptor vector closest to the first descriptor vector and a second descriptor vector second closest to the first descriptor vector is large, namely, a second key point corresponding to the second descriptor vector closest to the first descriptor vector has uniqueness, and can be considered to be matched with a first key point corresponding to the first descriptor vector; the euclidean distance ratio is greater than the matching ratio threshold, indicating that the difference between the second descriptor vector closest to the first descriptor vector and the second descriptor vector second closest to the first descriptor vector is small, i.e. the second keypoint corresponding to the second descriptor vector closest to the first descriptor vector has no uniqueness, and cannot be considered as matching the first keypoint corresponding to the first descriptor vector.

In step S2034, the server counts the matching number of second key points in each card surface sample picture matching the first key points, and obtains the matching similarity between the target picture and the card surface sample picture according to the matching number.

The matching quantity of the second key points matched with the first key points in one card surface sample picture can reflect the matching similarity of the card surface sample picture and the target picture. The matching quantity is positively correlated with the matching similarity, namely the larger the matching quantity is, the higher the matching similarity is; the smaller the number of matches, the lower the similarity of the matches. In this embodiment of the application, the matching number may be directly used as the matching similarity, or a parameter obtained by calculating according to the matching number may be used as the matching similarity, which is not limited herein.

In some embodiments, the card surface sample data may include a multidimensional vector of the card surface sample picture, and correspondingly, the server may process the first description sub-vector to obtain a multidimensional vector of the target picture, and perform matching using the multidimensional vector of the target picture and the card surface sample data. Fig. 5 is a flowchart of a method for obtaining a display card surface according to another embodiment of the present disclosure. Fig. 5 is different from fig. 3 in that step S203 in fig. 3 may be specifically detailed as step S2035 to step S2037.

In step S2035, the server obtains a second multidimensional vector of the target picture according to the first descriptor vector by using a vector coding algorithm.

The vector encoding algorithm is an algorithm for encoding a descriptor vector of a picture to obtain a multidimensional vector corresponding to the picture, and is not limited herein. In some examples, the Vector encoding algorithm may include a local Aggregated Vector (VLAD) algorithm, a Fisher Vector (FV) algorithm, and the like, without limitation.

Specifically, a vector coding algorithm is utilized to process a plurality of first descriptor sub-vectors of the target picture, so as to obtain a second multi-dimensional vector. The second multi-dimensional vector is a multi-dimensional vector of the target picture. For a target picture, the target picture may correspond to a plurality of first descriptor sub-vectors and a second multi-dimensional vector. The multiple first description sub-vectors of the target picture are coded into a second multi-dimensional vector through the vector, and the calculation efficiency of matching the subsequent target picture and the card surface sample picture can be improved. Under the condition that card surface sample data in the card surface database comprises a first multi-dimensional limit of a card surface sample picture, vector coding is carried out on a first description sub-vector of a target picture, and the total retrieval efficiency of the target picture in the card surface database can be improved.

In order to facilitate the second multi-dimensional vector to participate in the subsequent Euclidean distance calculation, the second multi-dimensional vector can be subjected to normalization processing, and the second multi-dimensional vector after the normalization processing is used for participating in the subsequent Euclidean distance calculation. The algorithm of the normalization process is not limited herein, and for example, the normalization process may be performed by using the L2 normalization algorithm. The normalization process can normalize the measurement scale of the multi-dimensional vector, such as the vector length, so as to improve the efficiency of subsequent Euclidean distance calculation.

In step S2036, the server calculates a target euclidean distance corresponding to each card surface sample picture.

The card face sample data comprises a first multi-dimensional vector of the card face sample picture. The first multi-dimensional vector is a multi-dimensional vector of the card face sample picture. And obtaining the first multi-dimensional vector according to the second key point and the second descriptor vector of the card surface sample picture. The second key point is the key point of the card face sample picture. The second descriptor vector is a descriptor vector of the second keypoint. The first multi-dimensional vectors correspond to the card surface sample pictures one by one, namely, one card surface sample picture corresponds to one first multi-dimensional vector.

The target Euclidean distance is the Euclidean distance between the second multi-dimensional vector and the first multi-dimensional vector corresponding to the card surface sample picture.

In step S2037, the server obtains the matching similarity between the target picture and the card surface sample picture according to the target euclidean distance.

The target Euclidean distance can reflect the matching similarity of the target picture and the card surface sample picture. The target Euclidean distance and the matching similarity are in negative correlation, namely the larger the target Euclidean distance between the second multi-dimensional vector and the first multi-dimensional vector is, the lower the matching similarity between the target picture corresponding to the second multi-dimensional vector and the card surface sample picture corresponding to the first multi-dimensional vector is; the smaller the target Euclidean distance between the second multi-dimensional vector and the first multi-dimensional vector is, the higher the matching similarity between the target picture corresponding to the second multi-dimensional vector and the card surface sample picture corresponding to the first multi-dimensional vector is. In the embodiment of the present application, the inverse of the target euclidean distance may be used as the matching similarity, and other parameters obtained by calculation according to the target euclidean distance may also be used as the matching similarity, which is not limited herein.

In some embodiments, the target picture may be further processed before the first keypoints and the first descriptor sub-vectors of the target picture are extracted, so that the extracted first keypoints and the first descriptor sub-vectors can be more accurate. And the extracted first key points and the first descriptor vectors can be screened, so that invalid data which easily influence the matching similarity are filtered, and the accuracy of the matching similarity is improved. Fig. 6 is a flowchart of another embodiment of a method for obtaining a display card surface according to the present application. Fig. 6 is different from fig. 3 in that the method for obtaining the display card surface shown in fig. 6 may further include steps S206 to S210.

In step S206, the server scales the target picture to a predetermined size using an image scaling algorithm.

The target pictures may have different sizes due to different sources of the target pictures. To facilitate processing, the target picture may be scaled to a uniform size. The image scaling algorithm may include, but is not limited to, region interpolation and the like. The predetermined size may be set according to a scene, a requirement, experience, and the like, and is not limited herein. For example, the predetermined size may be 170 × 107 pixels. The accuracy of matching similarity is further improved by scaling the target picture to a uniform size.

In step S207, the server identifies the card surface boundary of the entity card in the target picture by using a boundary identification algorithm.

The target picture is a shot picture of the card surface of the entity card. Although the card surface of the physical card can be framed through a view frame during photographing of the user terminal, the target picture may include contents other than the card surface of the physical card. And the card surface boundary of the entity card in the target picture is identified by utilizing a boundary identification algorithm, so that noise information interfering the matching of the target picture and the card surface sample picture can be reduced.

The boundary identification algorithm is an algorithm capable of identifying an edge of a pattern in a picture, and is not limited herein. In some examples, the boundary identification algorithm may include, without limitation, a Canny algorithm, an inter-Hough detection algorithm, and the like.

In step S208, the server determines an area enclosed by the card surface boundary as an effective area of the target picture.

The effective region is used for extracting the first key point and the first descriptor vector, namely extracting the first key point and the first descriptor vector from the effective region, and excluding noise information outside the effective region. For example, the edge part in the target picture can be identified through a Canny algorithm, then an edge straight line belonging to the card surface of the entity card is extracted through a Hough straight line detection algorithm, the edge straight line is intersected, and a rectangular area formed by the minimum value and the maximum value of the horizontal coordinate and the vertical coordinate in the intersection point of the edge straight line is obtained, and the rectangular area is an effective area.

By determining the effective area and eliminating the noise information in the target picture, the accuracy of matching similarity can be further improved, and therefore the accuracy of the alternative card surface picture provided for the user is improved.

In step S209, the server matches the first descriptor vector with a preset third descriptor vector, and obtains a first descriptor vector matching the third descriptor vector.

The third descriptor vector is a descriptor vector of the third key point. The third keypoint is a keypoint of a common element. The third keypoint and the third descriptor vector may be extracted from the common element using a feature extraction algorithm. The common elements are the same pattern in the card face of the physical card and are not limited herein. For example, the physical card includes a bank card, and the card surface of the bank card has the same card organization identification pattern, so that the card organization identification pattern can be used as a public element; the card surface of the bank card is also provided with the same chip pattern, and the chip pattern can also be used as a common element.

For the matching between the first descriptor vector and the third descriptor vector, reference may be made to the related description of the matching between the first descriptor vector and the second descriptor vector in the above embodiments, and details are not repeated here.

The first descriptor vector matched with the third descriptor vector is the first descriptor vector with the similarity exceeding a certain degree with the third descriptor vector, and represents that the first key point corresponding to the first descriptor vector matched with the third descriptor vector is the key point of the common element. The card surface of the entity card has common elements, and the common elements can influence the matching of the target picture and the card surface sample picture, so that the accuracy of matching similarity is reduced.

In step S210, the server filters out the first descriptor vector matching the third descriptor vector from the first descriptor vectors.

In order to avoid the influence of the common element on the matching of the card surface of the entity card and the card surface sample picture, the first descriptor vector influencing the matching of the target picture and the card surface sample picture is filtered, and the first descriptor vector influencing the matching of the target picture and the card surface sample picture is the first descriptor vector matched with the third descriptor vector. And filtering the first descriptor vector matched with the third descriptor vector, namely, the first descriptor vector remaining after the first descriptor vector matched with the third descriptor vector is filtered is involved in the matching of the subsequent target picture and the card surface sample picture, so that the influence of common elements on the matching of the target picture and the card surface sample picture is avoided, and the accuracy of matching similarity is improved.

For example, fig. 7 is a schematic diagram of an example of a card surface of a card and a physical card provided in an embodiment of the present application. As shown in fig. 7, the left side is a common element, specifically, a card organization identification pattern, and the right side is a picture of the card surface of the physical card, which has the card organization identification pattern. The common elements correspond to the card organization identification patterns on the picture of the card surface of the entity card, the circles on the common elements are third key points, the circles on the card organization identification patterns on the picture of the card surface of the entity card are first key points matched with the third key points, and connecting lines among the circles represent the matching relation between the third key points and the first key points.

In the process of executing the method for obtaining the display card surface, one or more groups of the steps S206, S207, S208, S209, and S210 may be selected for execution, and is not limited herein.

In some embodiments, after the matching between the target picture and the card surface sample picture is completed, the target picture may be stored in the card surface database as the card surface sample picture, and the card surface data of the target picture may be stored in the card surface database as the card surface sample data, so as to improve and enrich the card surface database. Fig. 8 is a flowchart of yet another embodiment of a method for obtaining a display card surface provided by the present application. Fig. 8 is different from fig. 3 in that the method for obtaining the display card surface shown in fig. 8 may further include steps S211 to S213, and step S214.

In step S211, the server obtains a slave card surface sample picture based on the target picture.

The card surface sample picture is a desensitized picture of the card surface including the solid card. Desensitization here refers to the removal of user sensitive information in the target picture, for example, the card number, card expiration date, etc. in the picture of the card face of a physical card may be removed.

In some examples, after desensitizing the target picture, the card face of the physical card in the target picture may also be identified, a portion of the card face of the physical card in the target picture may be retained, and the portion may be scaled to a predetermined size, resulting in a sample picture of the card face. That is, the slave card face sample picture may also be a picture of the card face including the physical card after desensitization processing, boundary identification processing, and scaling processing, which is not limited herein.

In step S212, the server obtains card surface sample data corresponding to the slave card surface sample picture according to the first keypoint and the first descriptor vector corresponding to the slave card surface sample picture.

And the first key point corresponding to the card surface sample picture is the first key point of the target picture, and the first description sub-vector corresponding to the card surface sample picture is the first description sub-vector of the target picture.

The type of the card face sample data corresponding to the card face sample picture may be consistent with the type of the card face sample data in the card face database. In some examples, the card face sample data corresponding from the card face sample picture may include the first keypoint and the first descriptor vector corresponding from the card face sample picture. In other examples, the card face sample data corresponding to the card face sample picture may include the second multidimensional vector of the target picture in the above embodiments.

In step S213, the server stores the slave card face sample picture as a card face sample picture in a card face database, and stores card face sample data corresponding to the slave card face sample picture in the card face database.

And the card surface sample picture corresponding to the target picture and serving as the display card surface picture is a main card surface sample picture corresponding to the slave card surface sample picture. Card face sample data of the slave card face sample picture and the slave card face sample picture in the card face database are used for being matched with the target picture, card face sample data of the master card face sample picture and the master card face sample picture are used for being matched with the target picture, and the master card face sample picture can also be used as an alternative card face picture.

In step S214, in a case that the selected first N card surface sample pictures include slave card surface sample pictures, the server replaces the slave card surface sample pictures in the selected first N card surface sample pictures with corresponding master card surface sample pictures.

Under the condition that the card surface database comprises card surface sample data of the slave card surface sample pictures and the slave card surface sample pictures, if the first N card surface sample pictures selected according to the sequence from high matching similarity to low matching similarity comprise the slave card surface sample pictures, the master open surface sample picture corresponding to the slave card surface sample picture can be used for replacing the slave card surface sample picture as an alternative card surface picture.

The card surface sample picture participates in matching, the matching success rate can be improved, namely the probability of identifying the card surface sample picture corresponding to the target picture can be improved, and the data in the card surface database is expanded and enriched. The main card face sample picture is a standard card face sample picture, the main card face sample picture is obtained from a picture of a card face of a shot entity card, the definition and the attractiveness are lower than those of the main card face sample picture, namely the main card face sample picture is more suitable for being displayed on a user terminal than the auxiliary card face sample picture, the main card face sample picture is used for replacing the auxiliary card face sample picture as an alternative card face picture, the definition and the attractiveness of the card face picture displayed on the user terminal can be improved, a user can conveniently perform card selecting operation and the like on the user terminal, and user experience is improved.

In some embodiments, establishing the card face database may be foreseen in the server. Fig. 9 is a flowchart of yet another embodiment of a method for obtaining a display card surface according to the present application. Fig. 9 is different from fig. 3 in that the method for obtaining the display card surface shown in fig. 9 may further include steps S215 to S218.

In step S215, the server enters a card face sample picture.

The card surface sample picture can be manually input, and can also be imported from equipment such as a server of a card issuing mechanism through an interface, and the input mode of the card surface sample picture is not limited herein.

In step S216, the server extracts a second key point and a second descriptor vector of the card surface sample picture by using a feature extraction algorithm.

The second key point is the key point of the card face sample picture. The second descriptor vector is a descriptor vector of the second keypoint. In some examples, a card surface sample picture has a plurality of second keypoints, and correspondingly, one second keypoint corresponds to one second descriptor vector, that is, a card surface sample picture has a plurality of second descriptor vectors.

The feature extraction algorithm is the same as the feature extraction algorithm in step S202 in the above embodiment, and the specific contents of the feature extraction algorithm, the second key point of the card face sample picture, and the second descriptor vector extraction may refer to the related descriptions of the feature extraction algorithm, the first key point, and the first descriptor vector extraction in the above embodiment, and are not described herein again.

In some examples, before performing step S216, the second descriptor vector may be further matched with a preset third descriptor vector to obtain a second descriptor vector matched with the third descriptor vector, and the second descriptor vector matched with the third descriptor vector is filtered out from the second descriptor vector. The third descriptor vector is a descriptor vector of the third key point. The third keypoint is a keypoint of a common element. The common elements are the same patterns in at least part of the card face sample picture. The specific content of the common elements can be referred to the related description in the above embodiments, and will not be described herein.

The content of the matching between the second descriptor vector and the third descriptor vector can refer to the related description of the matching between the first descriptor vector and the third descriptor vector in the above embodiments, and is not repeated herein. And in the second descriptor vector, filtering out the second descriptor vector matched with the third descriptor vector in order to filter out data related to the common element in the card surface sample data, so as to avoid the interference of the common element on the matching of the card surface sample picture of the target picture, and further improve the accuracy of the obtained matching similarity.

In step S217, the server obtains card surface sample data according to the second keypoint and the second descriptor vector.

In some examples, the card face sample data includes a second keypoint and a second descriptor vector if the number of card face sample pictures is less than or equal to the complex number threshold. The complex number threshold is a number threshold used for determining whether the number of card surface sample pictures in the card surface database to be established reaches a complex level, and may be set according to a scene, a requirement, experience, and the like, which is not limited herein. The number of the card surface sample pictures in the card surface database to be established is less than or equal to the complex number threshold, which indicates that the number of the card surface sample pictures in the card surface database to be established is less, the number of the corresponding second key points and the second descriptor vectors is also less, and the card surface database can directly store the second key points and the second descriptor vectors. Specifically, the second keypoints and the second descriptor vectors may be stored in a list manner in the established card face database.

In other examples, when the number of the card-face sample pictures is greater than the complex number threshold, a vector coding algorithm is used to obtain a first multi-dimensional vector of each card-face sample picture according to the second keypoints and the second descriptor sub-vectors of each card-face sample picture, and the first multi-dimensional vector of each card-face sample picture is determined as card-face sample data. The number of the card surface sample pictures is larger than the complex number threshold, which indicates that the number of the card surface sample pictures in the card surface database to be established is large, and the number of the corresponding second key points and the second descriptor vectors is also large.

In the case where the card face sample data comprises the first multi-dimensional vector, the card face sample data may also be stored in the card face database in an indexed structure. The index structure may be constructed using an index construction algorithm and a first multi-dimensional vector. In some examples, the index construction algorithm may include, without limitation, a ball-tree algorithm, a locality sensitive hashing algorithm, a Hierarchical navigatable Small World algorithm, i.e., HNSW algorithm, and the like.

In step S218, the server establishes a card face database based on the card face sample picture and the card face sample data.

The card surface database comprises a card surface sample picture and card surface sample data.

In some embodiments, the card face database may include more than two sub-databases. The sub-databases may be divided according to the card issuing organization to which the card face sample picture belongs. Card surface sample pictures in one sub-database belong to the same card issuing organization. Fig. 10 is a flowchart of yet another embodiment of a method for obtaining a display card surface provided in the present application. Fig. 10 is different from fig. 3 in that the method for obtaining the display card surface shown in fig. 10 may further include step S219, and step S203 in fig. 3 may be specifically detailed as step S2038 and step S2039 in fig. 10.

In step S219, the server acquires the binding information of the entity card in the target picture.

The server is a background server of the application program, and card binding information of the entity card is stored in the server. The binding information includes a first identifier characterizing an issuer of the physical card. The first identifier may be used to indicate the issuer, for example, the first identifier may include an issuer number, and the like, and is not limited herein. The binding information may also include other information, and is not limited herein.

In step S2038, the server determines a target sub-database according to the first identifier.

The target sub-database is a sub-database corresponding to the card issuing organization indicated by the first identification. And the sub-databases in the card surface database are divided according to the card issuing organizations, and the sub-database corresponding to the card issuing organization to which the entity card belongs can be determined according to the first identification.

In step S2039, the server matches the target picture with the card-face sample picture in the target sub-database based on the first descriptor vector and the card-face sample data in the target sub-database, so as to obtain a matching similarity between the target picture and the card-face sample picture.

And the card surface sample in the target sub-database is directly used for matching with the target picture, so that the matching range can be reduced, and the matching speed and the matching accuracy are improved.

For specific contents of the matching similarity between the target picture and the card surface sample picture obtained by matching the target picture and the card surface sample picture in the target sub-database based on the first descriptor vector and the card surface sample data in the target sub-database, reference may be made to the relevant description in the above embodiment, and details are not repeated here.

For easy understanding, the following describes an example of a process of establishing a card surface database by a server in the method for obtaining and displaying a card surface. Fig. 11 is a flowchart of an example of a card surface database establishment process in the embodiment of the present application. As shown in fig. 11, the card surface database establishing process may include steps S301 to S305.

In step S301, the server receives an entered card face sample picture.

In step S302, the server extracts a second keypoint and a second descriptor vector of each card face sample picture.

In step S303, if the number of the card surface sample pictures is less than or equal to the complex number threshold, the server uses the second keypoints and the second descriptor vectors of each card surface sample picture as card surface sample data.

In step S304, if the number of the card surface sample pictures is greater than the complex number threshold, the server encodes the second descriptor vector of each card surface sample picture to obtain a first multidimensional vector of each card surface sample picture as card surface sample data, and constructs an index structure of the first multidimensional vector by using the first multidimensional vector of each card surface sample picture.

In step S305, the server establishes a card surface database based on the card surface sample picture and the card surface sample data.

The details of the steps S301 to S305 can refer to the related descriptions in the above embodiments, and are not repeated herein.

For ease of understanding, the following describes an example of a process for obtaining the presentation card surface between the user terminal and the server. Fig. 12 is a flowchart of an example of a flow of obtaining a display card surface in the method for obtaining a display card surface according to the embodiment of the present application. As shown in fig. 12, the process of obtaining the display card surface may include steps S401 to S410.

In step S401, the user terminal photographs the card surface of the physical card to obtain a target picture.

In step S402, the user terminal uploads the target picture to the server.

In step S403, the server extracts a first keypoint and a first descriptor vector from the target picture.

In step S404, if the number of the card surface sample pictures in the card surface database is less than or equal to the complex number threshold, the server matches the target picture with the card surface sample pictures in the card surface database by using the first descriptor sub-vector and the second descriptor sub-vector in the card surface database, so as to obtain a matching similarity.

In step S405, if the number of the card surface sample pictures in the card surface database is greater than the complex number threshold, the server obtains a second multidimensional vector of each card surface sample picture according to the first descriptor vector of each card surface sample picture, and matches the target picture with the card surface sample pictures in the card surface database by using the second multidimensional vector and the first multidimensional vector in the card surface database, so as to obtain matching similarity.

In step S406, the server selects the first N card face sample pictures as candidate card face pictures according to the sequence of the matching similarity from high to low.

In step S407, the server transmits the alternative card face picture to the user terminal.

In step S408, the user terminal displays the alternative card face picture.

In step S409, the user terminal receives an operation input by the user.

In step S410, the user terminal responds to the operation input, and selects the alternative card face picture indicated by the operation input as the display card face picture of the entity card at the user terminal.

The details of the steps S401 to S410 can be referred to the related descriptions in the above embodiments, and are not repeated herein.

The application provides a device for obtaining a display card surface. Fig. 13 is a schematic structural diagram of an embodiment of an apparatus for obtaining a display card surface according to the present disclosure. As shown in fig. 13, the apparatus 500 for acquiring a display card surface may include a first acquiring module 501, an extracting module 502, a matching module 503, a selecting module 504, and a sending module 505.

The first obtaining module 501 may be used to obtain a target picture.

The target picture is a shot picture of the card surface of the entity card.

The extraction module 502 may be configured to extract a first keypoint and a first descriptor vector of the target picture using a feature extraction algorithm.

The first descriptor vector is a descriptor vector of the first keypoint.

The matching module 503 may be configured to match the target picture with a card surface sample picture in a card surface database based on the first descriptor vector and card surface sample data in the card surface database pre-established in the device for obtaining and displaying the card surface, so as to obtain a matching similarity between the target picture and the card surface sample picture.

The card surface sample data is used for representing the card surface sample picture.

The selecting module 504 may be configured to select the first N card surface sample pictures as alternative card surface pictures in an order from high matching similarity to low matching similarity.

N is a positive integer.

The sending module 505 may be configured to send the alternative card surface picture to the user terminal, so that the user terminal displays the alternative card surface picture for the user to select, so as to determine that the entity card displays the card surface picture at the user terminal.

In the embodiment of the application, the device for obtaining the card surface display extracts key points and description sub-vectors of a shot picture of the card surface of the entity card, matches the shot picture of the card surface of the entity card with a card surface sample picture in a card surface database based on a first description sub-vector and card surface sample data in a preset card surface database, and sends N card surface sample pictures with the highest matching similarity with the shot picture of the card surface of the entity card as alternative card surface pictures to a user terminal, so that the user terminal can determine the card surface display picture used by the entity card in the alternative card surface pictures. The device for acquiring the display card surface provides a display card surface picture required by displaying the entity card for the user terminal through an image identification technology, the acquisition of the display card surface picture does not depend on the card number of the entity card and the service of a server of a card issuing mechanism, only the device for acquiring the display card surface for image identification needs to be maintained, and the maintenance cost is reduced on the basis that the display card surface picture required by displaying the entity card by the user terminal can be provided.

In some examples, the set of card face sample data includes a second keypoint of one card face sample picture and a second descriptor vector, and the second descriptor vector is a descriptor vector of the second keypoint.

The matching module 503 may be configured to: calculating Euclidean distances between the first descriptor vector and second descriptor vectors corresponding to the card surface sample pictures; selecting the first M second descriptor vectors according to the sequence of the Euclidean distance from small to large, wherein M is a positive integer; determining whether a first key point corresponding to the first descriptor vector is matched with a second key point corresponding to the selected second descriptor vector according to the Euclidean distance between the selected second descriptor vector and the first descriptor vector; and counting the matching number of second key points matched with the first key points in each card surface sample picture, and obtaining the matching similarity between the target picture and the card surface sample picture according to the matching number, wherein the matching number and the matching similarity are in positive correlation.

Specifically, the matching module 503 may be configured to: and under the condition that the Euclidean distance between the selected second descriptor vector and the first descriptor vector is smaller than or equal to a matching distance threshold value, determining that a first key point corresponding to the first descriptor vector is matched with a second key point corresponding to the selected second descriptor vector, wherein M is 1.

Alternatively, the matching module 503 may be configured to: calculating a Euclidean distance ratio of a first Euclidean distance and a second Euclidean distance, wherein the first Euclidean distance is the Euclidean distance between a selected first second descriptor vector and the first descriptor vector, the second Euclidean distance is the Euclidean distance between a selected second descriptor vector and the first descriptor vector, and M is 2; and under the condition that the Euclidean distance ratio is smaller than or equal to the matching ratio threshold, determining that a first key point corresponding to the first descriptor vector is matched with a second key point corresponding to the obtained first second descriptor vector.

In some examples, the card face sample data includes a first multidimensional vector of the card face sample picture, the first multidimensional vector is obtained according to a second keypoint of the card face sample picture and a second descriptor vector, and the second descriptor vector is a descriptor vector of the second keypoint.

The matching module 503 may be configured to: obtaining a second multi-dimensional vector of the target picture according to the first descriptor vector by using a vector coding algorithm; calculating target Euclidean distances corresponding to the card surface sample pictures, wherein the target Euclidean distances are Euclidean distances between the second multi-dimensional vectors and the first multi-dimensional vectors corresponding to the card surface sample pictures; and obtaining the matching similarity between the target picture and the card surface sample picture according to the target Euclidean distance, wherein the target Euclidean distance and the matching similarity are in negative correlation.

Fig. 14 is a schematic structural diagram of another embodiment of the apparatus for obtaining a display card surface provided in the present application. Fig. 14 is different from fig. 13 in that the apparatus 500 for obtaining the display card surface shown in fig. 14 may further include an interference filtering module 506, a scaling module 507, and a boundary dividing module 508.

The interference filtering module 506 may be configured to: matching the first descriptor vector with a preset third descriptor vector to obtain a first descriptor vector matched with the third descriptor vector, wherein the third descriptor vector is a descriptor vector of a third key point, the third key point is a key point of a public element, and the public element is the same pattern in the card surface of the entity card; and filtering out the first descriptor vector matched with the third descriptor vector from the first descriptor vectors.

The scaling module 507 may be used to scale the target picture to a predetermined size using an image scaling algorithm.

The boundary partitioning module 508 may be configured to: identifying a card surface boundary of the entity card in the target picture by using a boundary identification algorithm; and determining an area enclosed by the card surface boundary as an effective area of the target picture, wherein the effective area is used for extracting a first key point and a first descriptor vector.

Fig. 15 is a schematic structural diagram of a device for obtaining a display card surface according to still another embodiment of the present disclosure. Fig. 15 is different from fig. 13 in that the apparatus 500 for obtaining a display card surface shown in fig. 15 may further include a first processing module 509, a storage module 510, and a replacement module 511.

The first processing module 509 is operable to: obtaining a secondary card surface sample picture based on the target picture, wherein the secondary card surface sample picture is a desensitized card surface picture comprising the entity card; and obtaining card surface sample data corresponding to the card surface sample picture according to the first key point and the first descriptor vector corresponding to the card surface sample picture.

The storage module 510 may be configured to: and storing the slave card face sample picture as a card face sample picture into a card face database, storing the card face sample data corresponding to the slave card face sample picture into the card face database, wherein the card face sample picture corresponding to the target picture and used as a display card face picture is a main card face sample picture corresponding to the slave card face sample picture.

The replacement module 511 may be configured to: and under the condition that the selected first N card surface sample pictures comprise the slave card surface sample pictures, replacing the slave card surface sample pictures in the selected first N card surface sample pictures with the corresponding master card surface sample pictures.

Fig. 16 is a schematic structural diagram of a device for obtaining a display card surface according to still another embodiment of the present disclosure. FIG. 16 differs from FIG. 13 in that the device 500 for obtaining a display card face shown in FIG. 16 may also include an entry module 512, a second processing module 513, and a database module 514.

The entry module 512 can be used to enter a card face sample picture.

The extracting module 502 may be configured to extract a second keypoint and a second descriptor vector of the card surface sample picture by using a feature extraction algorithm.

The second descriptor vector is a descriptor vector of the second keypoint.

The second processing module 513 may be configured to obtain card face sample data according to the second keypoint and the second descriptor vector.

The database module 514 may be configured to build a card face database based on the card face sample picture and the card face sample data.

In some examples, the card face sample data includes a second keypoint and a second descriptor vector if the number of card face sample pictures is less than or equal to the complex number threshold.

In other examples, the second processing module 513 may be configured to: under the condition that the number of the card surface sample pictures is larger than the threshold value of the complex number, obtaining a first multi-dimensional vector of each card surface sample picture according to a second key point and a second descriptor vector of each card surface sample picture by using a vector coding algorithm; and determining a first multi-dimensional vector of the card surface sample picture as card surface sample data.

Specifically, the card face sample data is stored in the card face database in an index structure manner.

In some embodiments, the interference filtering module 506 may further be configured to: matching the second descriptor vector with a preset third descriptor vector to obtain a second descriptor vector matched with the third descriptor vector, wherein the third descriptor vector is a descriptor vector of a third key point, the third key point is a key point of a common element, and the common element is the same pattern in at least part of card surface sample pictures; and filtering out a second descriptor vector matched with the third descriptor vector from the second descriptor vector.

In some embodiments, the card face database comprises more than two sub-databases, and the card face sample pictures in one sub-database belong to the same card issuing institution. Fig. 17 is a schematic structural diagram of yet another embodiment of an apparatus for obtaining a display card surface provided in the present application. Fig. 17 differs from fig. 13 in that the apparatus 500 for obtaining a display card surface shown in fig. 17 may further include a second obtaining module 515.

The second obtaining module 515 may be configured to obtain binding information of the entity card in the target picture, where the binding information includes a first identifier that represents a card issuing mechanism of the entity card.

The matching module 503 may be configured to: determining a target sub-database according to the first identifier, wherein the target sub-database is a sub-database corresponding to the card issuing organization indicated by the first identifier; and matching the target picture and the card surface sample picture in the target sub-database based on the first descriptor vector and the card surface sample data in the target sub-database to obtain the matching similarity of the target picture and the card surface sample picture.

The application also provides a server. Fig. 18 is a schematic structural diagram of an embodiment of a server provided in the present application. As shown in fig. 18, the server 600 includes a memory 601, a processor 602, and a computer program stored on the memory 601 and executable on the processor 602.

In one example, the processor 602 may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or may be configured to implement one or more Integrated circuits of the embodiments of the present Application.

The Memory 601 may include Read-Only Memory (ROM), Random Access Memory (RAM), magnetic disk storage media devices, optical storage media devices, flash Memory devices, electrical, optical, or other physical/tangible Memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software comprising computer-executable instructions and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method of obtaining a presentation card surface in accordance with embodiments of the present application.

The processor 602 runs a computer program corresponding to the executable program code by reading the executable program code stored in the memory 601, so as to implement the method for acquiring the display card surface in the above embodiment.

In one example, server 600 can also include a communication interface 603 and bus 604. As shown in fig. 18, the memory 601, the processor 602, and the communication interface 603 are connected via a bus 604 to complete communication therebetween.

The communication interface 603 is mainly used for implementing communication between modules, apparatuses, units and/or devices in the embodiments of the present application. Input devices and/or output devices are also accessible through communication interface 603.

The bus 604 comprises hardware, software, or both to couple the components of the server 600 to one another. By way of example, and not limitation, Bus 604 may include an Accelerated Graphics Port (AGP) or other Graphics Bus, an Enhanced Industry Standard Architecture (EISA) Bus, a Front-Side Bus (Front Side Bus, FSB), a Hyper Transport (HT) Interconnect, an Industry Standard Architecture (ISA) Bus, an infiniband Interconnect, a Low Pin Count (LPC) Bus, a memory Bus, a microchannel Architecture (MCA) Bus, a Peripheral Component Interconnect (PCI) Bus, a PCI-Express (PCI-X) Bus, a Serial Advanced Technology Attachment (attached) Bus, a Local Electronics Standard Association (vldo) Bus, a Local Architecture (vldo) Bus, or a combination of two or more of these buses. Bus 604 may include one or more buses, where appropriate. Although specific buses are described and shown in the embodiments of the application, any suitable buses or interconnects are contemplated by the application.

The embodiment of the present application further provides a computer-readable storage medium, where computer program instructions are stored on the computer-readable storage medium, and when the computer program instructions are executed by a processor, the method for obtaining a display card surface in the foregoing embodiment can be implemented, and the same technical effect can be achieved. The computer-readable storage medium may include a non-transitory computer-readable storage medium, such as a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like, which is not limited herein.

It should be clear that the embodiments in this specification are described in a progressive manner, and the same or similar parts in the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. For apparatus embodiments, server embodiments, computer-readable storage medium embodiments, reference may be made in the descriptive section to method embodiments for relevant aspects. The present application is not limited to the particular steps and structures described above and shown in the drawings. Those skilled in the art may make various changes, modifications and additions or change the order between the steps after appreciating the spirit of the present application. Also, a detailed description of known process techniques is omitted herein for the sake of brevity.

Aspects of the present application are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware for performing the specified functions or acts, or combinations of special purpose hardware and computer instructions.

It will be appreciated by persons skilled in the art that the above embodiments are illustrative and not restrictive. Different features which are present in different embodiments may be combined to advantage. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art upon studying the drawings, the specification, and the claims. In the claims, the term "comprising" does not exclude other means or steps; the word "a" or "an" does not exclude a plurality; the terms "first" and "second" are used to denote a name and not to denote any particular order. Any reference signs in the claims shall not be construed as limiting the scope. The functions of the various parts appearing in the claims may be implemented by a single hardware or software module. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

Claims

1. A method for obtaining a display card surface is characterized by comprising the following steps:

the method comprises the steps that a server obtains a target picture, wherein the target picture is a shot picture of a card surface of an entity card;

the server extracts a first key point and a first descriptor vector of the target picture by using a feature extraction algorithm, wherein the first descriptor vector is a descriptor vector of the first key point;

the server matches the target picture with a card face sample picture in a card face database established in advance in the server based on the first descriptor vector and the card face sample data in the card face database to obtain matching similarity of the target picture and the card face sample picture, wherein the card face sample data is used for representing the card face sample picture;

the server selects the first N card surface sample pictures as alternative card surface pictures according to the sequence of matching similarity from high to low, wherein N is a positive integer;

and the server sends the alternative card surface picture to the user terminal, so that the user terminal displays the alternative card surface picture for the user to select so as to determine the card surface picture displayed by the entity card at the user terminal.

2. The method according to claim 1, wherein a set of said card face sample data comprises a second keypoint of said card face sample picture and a second descriptor vector, said second descriptor vector being a descriptor vector of said second keypoint;

the server matches the target picture with a card surface sample picture in a card surface database based on the first descriptor vector and card surface sample data in the card surface database pre-established in the server to obtain matching similarity between the target picture and the card surface sample picture, and the method comprises the following steps:

the server calculates Euclidean distances between the first descriptor vector and the second descriptor vector corresponding to each card surface sample picture;

the server selects the first M second descriptor vectors according to the sequence of the Euclidean distances from small to large, wherein M is a positive integer;

the server determines whether the first key point corresponding to the first descriptor vector is matched with the second key point corresponding to the selected second descriptor vector according to the Euclidean distance between the selected second descriptor vector and the first descriptor vector;

the server counts the matching number of second key points matched with the first key points in each card face sample picture, and obtains the matching similarity of the target picture and the card face sample picture according to the matching number, wherein the matching number is positively correlated with the matching similarity.

3. The method of claim 2, wherein the server determines whether the first keypoint corresponding to the first descriptor vector matches the second keypoint corresponding to the selected second descriptor vector according to the euclidean distance between the selected second descriptor vector and the first descriptor vector, comprising:

the server determines that the first key point corresponding to the first descriptor vector is matched with the second key point corresponding to the selected second descriptor vector under the condition that the Euclidean distance between the selected second descriptor vector and the first descriptor vector is smaller than or equal to a matching distance threshold, wherein M is 1;

or,

the server calculates a Euclidean distance ratio of a first Euclidean distance and a second Euclidean distance, wherein the first Euclidean distance is the Euclidean distance between a first selected second descriptor vector and the first descriptor vector, the second Euclidean distance is the Euclidean distance between a second selected second descriptor vector and the first descriptor vector, and M is 2; and the server determines that the first key point corresponding to the first descriptor vector is matched with the second key point corresponding to the first obtained second descriptor vector under the condition that the Euclidean distance ratio is smaller than or equal to a matching ratio threshold.

4. The method according to claim 1, wherein the card face sample data comprises a first multidimensional vector of the card face sample picture, the first multidimensional vector is obtained from a second keypoint of the card face sample picture and a second descriptor vector, and the second descriptor vector is a descriptor vector of the second keypoint;

the server obtains a second multi-dimensional vector of the target picture according to the first descriptor vector by using a vector coding algorithm;

the server calculates a target Euclidean distance corresponding to each card surface sample picture, wherein the target Euclidean distance is the Euclidean distance between the second multi-dimensional vector and the first multi-dimensional vector corresponding to the card surface sample picture;

and the server obtains the matching similarity between the target picture and the card surface sample picture according to the target Euclidean distance, wherein the target Euclidean distance is in negative correlation with the matching similarity.

5. The method according to claim 1, before the server matches the target picture with a card face sample picture in a card face database established in advance in the server based on the first descriptor vector and the card face sample data in the card face database, further comprising:

the server matches the first descriptor vector with a preset third descriptor vector to obtain the first descriptor vector matched with the third descriptor vector, wherein the third descriptor vector is a descriptor vector of a third key point, the third key point is a key point of a public element, and the public element is the same pattern in a card surface of an entity card;

the server filters out the first descriptor vector matching the third descriptor vector from the first descriptor vector.

6. The method according to claim 1, before the server extracts the first keypoint and the first descriptor vector of the target picture by using a feature extraction algorithm, further comprising:

and the server scales the target picture to a preset size by using an image scaling algorithm.

7. The method according to claim 1, before the server extracts the first keypoint and the first descriptor vector of the target picture by using a feature extraction algorithm, further comprising:

the server identifies the card surface boundary of the entity card in the target picture by using a boundary identification algorithm;

and the server determines an area enclosed by the card surface boundary as an effective area of the target picture, wherein the effective area is used for extracting the first key point and the first descriptor vector.

8. The method of claim 1, further comprising:

the server obtains a secondary card surface sample picture based on the target picture, wherein the secondary card surface sample picture is a desensitized picture comprising the card surface of the entity card;

the server obtains the card surface sample data corresponding to the card surface sample picture according to the first key point and the first descriptor vector corresponding to the card surface sample picture;

and the server stores the slave card surface sample picture as the card surface sample picture to the card surface database, stores the card surface sample data corresponding to the slave card surface sample picture in the card surface database, and uses the card surface sample picture corresponding to the target picture as the display card surface picture as a main card surface sample picture corresponding to the slave card surface sample picture.

9. The method according to claim 8, wherein after the server selects the first N card face sample pictures as the candidate card face pictures in an order from high matching similarity to low matching similarity, the method further comprises:

and the server replaces the slave card surface sample pictures in the first N selected card surface sample pictures with the corresponding main card surface sample pictures under the condition that the first N selected card surface sample pictures comprise the slave card surface sample pictures.

10. The method according to claim 1, wherein before the server obtains the target picture, the target picture being a picture of a photographed physical card, further comprising:

the server inputs a card face sample picture;

the server extracts a second key point and a second descriptor vector of the card surface sample picture by using the feature extraction algorithm, wherein the second descriptor vector is a descriptor vector of the second key point;

the server obtains the card surface sample data according to the second key point and the second descriptor vector;

and the server establishes the card surface database based on the card surface sample picture and the card surface sample data.

11. The method of claim 10,

in a case that a number of the card face sample pictures is less than or equal to a complex number threshold, the card face sample data includes the second keypoint and the second descriptor vector.

12. The method of claim 10, wherein obtaining the card-face sample data according to the second keypoint and the second descriptor vector comprises:

the server obtains a first multi-dimensional vector of each card surface sample picture according to the second key point and the second descriptor sub-vector of each card surface sample picture by using a vector coding algorithm under the condition that the number of the card surface sample pictures is larger than a complex number threshold;

the server determines a first multi-dimensional vector of the card face sample picture as the card face sample data.

13. The method of claim 4 or 12, wherein the card face sample data is stored in the card face database in an indexed structure.

14. The method according to claim 10, before the server extracts the second keypoints and the second descriptor vectors of the card face sample picture by using the feature extraction algorithm, further comprising:

the server matches the second descriptor vector with a preset third descriptor vector to obtain the second descriptor vector matched with the third descriptor vector, wherein the third descriptor vector is a descriptor vector of a third key point, the third key point is a key point of a common element, and the common element is the same pattern in at least part of the card surface sample pictures;

the server filters out the second descriptor vector matched with the third descriptor vector in the second descriptor vector.

15. The method according to claim 1, wherein the card surface database comprises more than two sub-databases, and the card surface sample pictures in one sub-database belong to the same card issuing institution;

the method further comprises the following steps:

the server acquires binding information of the entity card in the target picture, wherein the binding information comprises a first identifier representing a card issuing mechanism of the entity card;

the server determines the target sub-database according to the first identifier, wherein the target sub-database is the sub-database corresponding to the card issuing organization indicated by the first identifier;

and the server matches the target picture with the card surface sample picture in the target sub-database based on the first descriptor vector and the card surface sample data in the target sub-database to obtain the matching similarity between the target picture and the card surface sample picture.

16. An apparatus for obtaining a display card face, comprising:

the acquisition module is used for acquiring a target picture, wherein the target picture is a shot picture of the card surface of the entity card;

the extraction module is used for extracting a first key point and a first descriptor vector of the target picture by using a feature extraction algorithm, wherein the first descriptor vector is the descriptor vector of the first key point;

the matching module is used for matching the target picture with a card surface sample picture in a card surface database based on the first descriptor vector and card surface sample data in the card surface database pre-established in the device for obtaining and displaying the card surface to obtain matching similarity between the target picture and the card surface sample picture, wherein the card surface sample data is used for representing the card surface sample picture;

the selecting module is used for selecting the first N card surface sample pictures as alternative card surface pictures according to the sequence of the matching similarity from high to low, wherein N is a positive integer;

and the sending module is used for sending the alternative card surface picture to a user terminal, so that the user terminal displays the alternative card surface picture for a user to select so as to determine the card surface picture displayed by the entity card at the user terminal.

17. A server, characterized in that the device comprises: a processor and a memory storing computer program instructions;

the processor, when executing the computer program instructions, implements a method of obtaining a display card surface as claimed in any one of claims 1 to 15.

18. A computer-readable storage medium, having stored thereon computer program instructions, which, when executed by a processor, implement a method of obtaining a display card surface according to any one of claims 1 to 15.