CN111143813B

CN111143813B - Verification problem generation method, verification method and device

Info

Publication number: CN111143813B
Application number: CN201911380167.4A
Authority: CN
Inventors: 朱康峰; 冯潞潞; 范长杰; 胡志鹏
Original assignee: Netease Hangzhou Network Co Ltd
Current assignee: Netease Hangzhou Network Co Ltd
Priority date: 2019-12-27
Filing date: 2019-12-27
Publication date: 2022-02-22
Anticipated expiration: 2039-12-27
Also published as: CN111143813A

Abstract

The application relates to the technical field of verification codes, in particular to a method, a method and a device for generating a verification problem. Based on the mode, a large number of verification problems can be generated, and the difficulty of cracking the generated verification problems by a computer can be improved because the verification problems are generated through the image characteristic information.

Description

Verification problem generation method, verification method and device

Technical Field

The present application relates to the field of verification code technologies, and in particular, to a method for generating a verification problem, a method for verifying a verification problem, and an apparatus for verifying a verification problem.

Background

With the rapid development of the internet technology, the information security problem is increasingly prominent, and in order to prevent phenomena such as malicious password cracking, ticket swiping, forum water filling and the like, the prior art mainly adopts a verification code to distinguish whether a user is a computer or a human.

The existing verification mode is mainly a scheme of 'inputting characters', 'clicking characters', 'splicing pictures' and 'recognizing pictures', but the scheme of 'inputting characters' and 'clicking characters' is easy to crack characters or letters with noise due to the development of image processing related technologies, and the scheme of 'splicing pictures' and 'recognizing pictures' needs a large amount of picture resources and is difficult to generate a sufficient number of verification problems in a short time.

Therefore, how to enrich the verification problem and improve the difficulty of the verification problem generated by computer cracking is a problem to be solved urgently at present.

Disclosure of Invention

In view of this, an object of the embodiments of the present application is to provide a method, a device and a computer program product for generating a verification problem, which can not only generate a large number of verification problems, but also improve the difficulty of breaking the generated verification problems by a computer.

Mainly comprises the following aspects:

in a first aspect, an embodiment of the present application provides a method for generating a verification problem, where the method includes:

acquiring a target image, and extracting image characteristic information from the target image;

and generating at least one verification question of the target image according to the image characteristic information.

In a possible implementation, the acquiring a target image and extracting image feature information from the target image includes:

acquiring the target image and determining at least one verification subject matched with the target image;

and inputting the target image into the determined image recognition model corresponding to each verification subject, and outputting the image characteristic information of the target image under each verification subject.

In one possible embodiment, the verification topic includes at least one of the following topics:

gender of the person, hair color of the person, skin color of the person, location of skeletal points of the person, type of animal, location of the object in the image.

In a possible implementation manner, if the verification subject is the position of a human skeleton point, the image identification model corresponding to the verification subject is an identification model of the human skeleton point; the step of inputting the target image into the determined image recognition model corresponding to each verification subject and outputting the image characteristic information of the target image under each verification subject comprises the following steps:

inputting the target image into the character skeleton point recognition model, and outputting the position information of each skeleton point of the game role in the target image; the target image is an image from a game character in a game database.

In a possible implementation, the target image is plural, and after the generating at least one verification question of the target image according to the image feature information, the generating method further includes:

obtaining a first type of questions marked with verification answers and a second type of questions not marked with verification answers from a plurality of verification questions corresponding to the target images;

sending the second type of questions without marked verification answers and the corresponding target images to terminal equipment of a plurality of sample users, and determining the non-marked verification answers of the second type of questions according to the feedback results;

storing the first type of questions, the corresponding labeled verification answers and the corresponding target images in a verification question bank in an associated manner, and storing the second type of questions, the corresponding non-labeled verification answers and the corresponding target images in the verification question bank in an associated manner; the verification question bank is used for providing questions for verifying the user.

In a possible implementation manner, the sending the second type of questions without labeled verification answers and the corresponding target images to the terminal devices of a plurality of sample users, and determining the non-labeled verification answers of the second type of questions according to the feedback result includes:

counting the number of times each second-class problem in the plurality of second-class problems is sent;

obtaining a second type of questions meeting a first preset condition from the plurality of second type of questions, and obtaining a plurality of candidate verification answers corresponding to the second type of questions meeting the first preset condition according to the feedback result; the first preset condition is that the sending times of the second type of problems are more than or equal to preset times;

for each second type question meeting the first preset condition, calculating the proportion of the number of times that each candidate verification answer in the plurality of corresponding candidate verification answers is answered to the total number of times that the plurality of candidate verification answers are answered;

if the candidate verification answer with the proportion larger than or equal to the preset threshold exists, determining the candidate verification answer as a non-labeled verification answer of the second type of questions;

and if the candidate verification answers with the proportion larger than or equal to the preset threshold value do not exist, abandoning the corresponding second type of problems.

In one possible embodiment, after storing the first type of question, the corresponding annotated verification answer, and the corresponding target image association in a verification question bank, and storing the second type of question, the corresponding non-annotated verification answer, and the corresponding target image association in the verification question bank, the generation method further includes:

aiming at the obtained multiple first-class problems, determining a predicted verification answer of each first-class problem according to each first-class problem and image characteristic information corresponding to the first-class problem;

acquiring a first-class problem meeting a second preset condition from the plurality of first-class problems; the second preset condition is that the predicted verification answer corresponding to the first type of question is different from the marked verification answer;

and adjusting network parameters in the image recognition model for generating the image characteristic information through the first kind of questions meeting the second preset condition and the marking verification answers corresponding to the first kind of questions.

In one possible implementation, after storing the first type of question, the corresponding annotated verification answer, and the corresponding target image association in a verification question bank, and storing the second type of question, the corresponding non-annotated verification answer, and the corresponding target image association in the verification question bank, the generation method further includes:

aiming at the plurality of acquired second-class problems, determining a predictive verification answer of each second-class problem according to each second-class problem and image characteristic information corresponding to the second-class problem;

acquiring a second type of problem meeting a third preset condition from the plurality of second type of problems; the third preset condition is that the predicted verification answers corresponding to the second type of questions are different from the non-labeled verification answers;

and adjusting network parameters in the image recognition model for generating the image characteristic information through the second type of questions meeting the third preset condition and the non-labeling verification answers corresponding to the second type of questions.

In a second aspect, an embodiment of the present application further provides a verification method for verifying a user operating a game, where the verification method includes:

acquiring a group of verification information in response to a verification event triggered by a target user; the verification information comprises an image of a game character in the game, a verification question asking a position of a target skeleton point, and position information of the target skeleton point as a verification answer;

sending the image of the game role and a verification question for inquiring the position of a target skeleton point to the terminal equipment of the target user;

and determining a verification result for verifying the target user according to the feedback result of the terminal equipment and the verification answer.

In one possible embodiment, the verification event comprises one of the following events:

a game account login event, a game payment event, and a game cheating event.

In one possible embodiment, the obtaining a set of authentication information in response to an authentication event triggered by a target user includes:

and acquiring the image of any game character in the game, a verification question related to the image of the game character and inquiring the position of the target skeleton point, and a verification answer corresponding to the verification question from a verification information base.

acquiring an image of any game role in the game from a game database;

inputting the obtained image of the game role into a character skeleton point identification model, and outputting the position information of each skeleton point in the image of the game role;

generating a plurality of verification questions inquiring the positions of all the skeleton points according to the position information of all the skeleton points;

a verification question is selected from a plurality of verification questions asking for a location of a respective skeletal site.

In a possible implementation manner, the feedback result of the terminal device includes: and the target user clicks the position information of any position of the image presented by the terminal equipment.

In a third aspect, an embodiment of the present application further provides a device for generating a verification problem, where the device includes:

the extraction module is used for acquiring a target image and extracting image characteristic information from the target image;

and the generating module is used for generating at least one verification problem of the target image according to the image characteristic information.

In a fourth aspect, an embodiment of the present application further provides an authentication apparatus for authenticating a user operating a game, where the authentication apparatus includes:

the acquisition module is used for responding to a verification event triggered by a target user and acquiring a group of verification information; the verification information comprises an image of a game character in the game, a verification question asking a position of a target skeleton point, and position information of the target skeleton point as a verification answer;

the sending module is used for sending the image of the game role and a verification question for inquiring the position of a target skeleton point to the terminal equipment of the target user;

and the determining module is used for determining a verification result for verifying the target user according to the feedback result of the terminal equipment and the verification answer.

In a fifth aspect, an embodiment of the present application further provides an electronic device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating via the bus when the electronic device is running, the machine-readable instructions being executable by the processor to perform the steps of the method for generating a verification problem as described in any one of the above-mentioned first aspect or first possible implementation manners, and/or to perform the steps of the method for verifying as described in any one of the above-mentioned second aspect or second possible implementation manners.

In a sixth aspect, this embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to perform the steps of the method for generating the verification problem described in the first aspect or any one of the possible embodiments of the first aspect, and/or to perform the steps of the method for verifying described in the second aspect or any one of the possible embodiments of the second aspect.

In the embodiment of the application, at least one verification problem of the target image is generated according to the image characteristic information by acquiring the target image and extracting the image characteristic information from the target image. Based on the mode, a large number of verification problems can be generated, and the difficulty of cracking the generated verification problems by a computer can be improved because the verification problems are generated through the image characteristic information.

Further, a group of verification information is obtained through a verification event triggered by the target user, the image of the game role and the verification question inquiring the position of the target skeleton point are sent to the terminal equipment of the target user, and further, the verification result for verifying the target user is determined according to the feedback result and the verification answer of the terminal equipment. Based on the mode, the verification information for providing the verification service is complex, so that the verification problem provided for the target user to carry out verification is difficult to crack by a computer.

In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

FIG. 1 is a flow chart illustrating a method for generating a verification problem according to an embodiment of the present application;

FIG. 2 is a flow chart illustrating a verification method provided by an embodiment of the present application;

FIG. 3 is a functional block diagram of an apparatus for generating a verification problem according to an embodiment of the present application;

FIG. 4 shows a functional block diagram of the extraction module of FIG. 3;

FIG. 5 is a second functional block diagram of an apparatus for generating a verification problem according to an embodiment of the present application;

FIG. 6 is a functional block diagram of an authentication apparatus provided in an embodiment of the present application;

fig. 7 shows a schematic structural diagram of an electronic device provided in an embodiment of the present application.

Detailed Description

To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are for illustrative and descriptive purposes only and are not used to limit the scope of protection of the present application. Additionally, it should be understood that the schematic drawings are not necessarily drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be performed out of order, and that steps without logical context may be performed in reverse order or concurrently. One skilled in the art, under the guidance of this application, may add one or more other operations to, or remove one or more operations from, the flowchart.

In addition, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.

To enable those skilled in the art to utilize the present disclosure, in connection with certain application scenarios "generation and application of verification questions," the following embodiments are presented to enable those skilled in the art to apply the general principles defined herein to other embodiments and application scenarios without departing from the spirit and scope of the present disclosure.

The method, apparatus, electronic device or computer-readable storage medium described in the embodiments of the present application may be applied to any scenario in which generation and application of a verification problem are required, and the embodiments of the present application do not limit a specific application scenario, and any scheme that uses the method, apparatus and apparatus for generating a verification problem provided in the embodiments of the present application is within the scope of protection of the present application.

It should be noted that, before the present application is proposed, the existing verification methods are mainly schemes of "inputting characters", "braille", "jigsaw", and "learning to see pictures", however, due to the development of image processing related technologies, the schemes of "inputting characters" and "braille" have been easy to be cracked by characters or letters with noise, while the schemes of "jigsaw" and "learning to see pictures" require a large amount of picture resources, and it is difficult to generate a sufficient number of verification codes in a short time.

In view of the above problems, in the embodiment of the present application, at least one verification problem of a target image is generated according to image feature information by acquiring the target image and extracting the image feature information from the target image. Based on the mode, a large number of verification problems can be generated, and the difficulty of cracking the generated verification problems by a computer can be improved because the verification problems are generated through the image characteristic information.

For the convenience of understanding of the present application, the technical solutions provided in the present application will be described in detail below with reference to specific embodiments.

Fig. 1 is a flowchart of a method for generating a verification problem according to an embodiment of the present application. As shown in fig. 1, the method for generating a verification problem provided in the embodiment of the present application includes the following steps:

s101: and acquiring a target image, and extracting image characteristic information from the target image.

In a specific implementation, an image, i.e., a target image, is acquired, and image feature information may be extracted from the target image, where the image feature information may represent features of the target image.

Further, a specific process of extracting image feature information from a target image is explained, that is, the step S101 of acquiring the target image and extracting the image feature information from the target image includes the following steps:

acquiring the target image and determining at least one verification subject matched with the target image; and inputting the target image into the determined image recognition model corresponding to each verification subject, and outputting the image characteristic information of the target image under each verification subject.

In a specific implementation, a plurality of verification subjects may be preset, and a target image may be obtained, where at least one verification subject may be matched for the target image, and the target image may be an image of a person or an image of an animal, and the target image is input into an image recognition model corresponding to the verification subject matched with the target image, and image feature information of the target image under the corresponding verification subject may be output. Here, the image recognition model is a model trained in advance.

Wherein the verification topic comprises at least one of the following topics: gender of the person, hair color of the person, skin color of the person, location of skeletal points of the person, type of animal, location of the object in the image.

In particular implementations, the target image may be matched to at least one authentication topic, and the target image may obtain different image characteristic information under different authentication topics.

In one example, if the preset verification subject includes a person gender, a person color, a person skin color and a position of a person skeleton point, if the obtained target image is a person image, the image feature information of the person image under the subject of the person gender is the gender of the person in the image, such as gender "woman"; the image characteristic information of the person image under the theme of the person color development is the color development of the person in the image, such as the color development 'black'; the image characteristic information of the person image under the subject of the person skin color is the skin color of the person in the image, such as the skin color of yellow; the image feature information of the person image under the subject of the position of the person skeleton point is the position of the person skeleton point in the person image, such as the coordinates "(30, 40)" of the person left skeleton point.

Here, if the verification subject is the character gender, the corresponding image recognition model is a character gender recognition model; if the verification subject is the skin color of the object, the corresponding image recognition model is a skin color recognition model of the object; if the verification subject is the color development of the human object, the corresponding image recognition model is the color development recognition model of the human object; if the verification subject is the position of the human skeleton point, the corresponding image recognition model is a human skeleton point recognition model; if the verification subject is the animal type, the corresponding image identification model is the animal type identification model; and if the verification subject is the position of the article in the image, the corresponding image identification model is the article position identification model.

Here, when the user is authenticated, the target image and the authentication question corresponding to the target image may be sent to the terminal device of the user together, where the target image and the corresponding authentication question are both presented on the terminal device, and the user gives an authentication answer according to the authentication question and the target image.

Further, if the verification subject is the position of the human skeleton point, the image identification model corresponding to the verification subject is the identification model of the human skeleton point; the step of inputting the target image into the determined image recognition model corresponding to each verification subject and outputting the image characteristic information of the target image under each verification subject comprises the following steps:

In a specific implementation, the image feature information of each image may be obtained by using images of a plurality of game characters in the game database, and if the image recognition model is a recognition model of a character skeleton point, specifically, the images of the plurality of game characters may be obtained from the game database, the image of each game character may be input into the recognition model of the character skeleton point, and the position information of each skeleton point in the image of the game character may be output.

S102: and generating at least one verification question of the target image according to the image characteristic information.

In specific implementation, a target image determines image feature information under each verification subject, and then a verification problem of the target image under the verification subject is generated according to the image feature information of the target image under one verification subject. Here, the same target image may generate a plurality of verification questions corresponding to the same verification subject, the same target image may generate a larger number of verification questions corresponding to different verification subjects, each verification subject may correspond to a different kind of verification questions, thus, a large number of verification problems of different kinds can be generated rapidly through one target image, the number of generated verification problems is increased, the generated verification problems are diversified, and the verification problems are obtained through the image characteristic information, because the difficulty is improved on the algorithm, the verification problem obtained by adopting the method is difficult to crack by a computer, compared with the verification scheme which is easy to crack by the computer through inputting characters, braille, jigsaw puzzle and learning pictures in the prior art, the difficulty that the generated verification problem is cracked by a computer can be improved, and the safety of information can be further improved.

It should be noted that, for each target image, verification problems with different difficulty levels may be generated according to business needs, for example, a verification problem with low difficulty level is to click on the thigh of a person in the image, and a verification problem with high difficulty level is to click on the bare foot and the left foot of the person in the image in sequence.

In an example, the target image is an image of a game character, the verification subject is positions of skeleton points of a human body, the corresponding image feature information is position information of each skeleton point in the image of the game character, and the generated verification problem may be clicking a left foot of the game character, or sequentially clicking a left eye and a right eye of the game character in the image.

Further, the target image may be a plurality of target images, and after the step S102 of generating at least one verification question of the target image from a plurality of verification questions corresponding to the plurality of images according to the image feature information, the generating method further includes:

step (1): and acquiring a first type of questions marked with verification answers and a second type of questions not marked with verification answers from a plurality of verification questions corresponding to the target images.

In specific implementation, after a large number of verification questions corresponding to a plurality of acquired target images are obtained respectively, labeling of verification answers can be performed on part of the obtained verification questions manually, further, a first class of questions with the labeled verification answers can be obtained, the remaining verification questions without the labeled verification answers are a second class of questions, and the first class of questions, the corresponding target images and the corresponding labeled verification answers can be stored in a verification information base after being associated so as to be used when a user is verified.

Here, the first question and the second question can be simultaneously sent to the same user, the user is verified through the first question, the non-labeled answer of the second question is determined through the answer of the user to the second question, the corresponding image and the corresponding verification answer are further stored in the verification information base after being associated, and the verification information base can be expanded without marking the verification answer on the verification question by a large amount of manual work.

Step (2): and sending the second type of questions without marked verification answers and the corresponding images to terminal equipment of a plurality of sample users, and determining the non-marked verification answers of the second type of questions according to the feedback results.

In specific implementation, each second-class question and corresponding image which are not labeled with verification answers can be sent to terminal equipment of a plurality of sample users, so that the non-labeled verification answer of each second-class question can be determined according to the feedback result of each second-class question given by a large number of sample users, and the second-class question, the corresponding image and the corresponding non-labeled verification answer are stored in a verification information base, so that the verification information base can be expanded.

Here, the sample user may be a user who specially determines the non-labeled verification answers to the second type of questions, or may be a user who performs verification, and the user may obtain the non-labeled verification answers corresponding to the second type of questions.

Further, the step (2) of sending the second type of questions without marked authentication answers and the corresponding images to the terminal devices of the plurality of sample users, and determining the non-marked authentication answers of the second type of questions according to the feedback result includes the following steps:

step a: and counting the number of times each second-class question in the plurality of second-class questions is sent.

In a specific implementation, for each of a plurality of second-class questions, the number of times each second-class question is sent to a respective sample user is counted.

Step b: obtaining a second type of questions meeting a first preset condition from the plurality of second type of questions, and obtaining a plurality of candidate verification answers corresponding to the second type of questions meeting the first preset condition according to the feedback result; the first preset condition is that the sending times of the second type of problems are more than or equal to the preset times.

In specific implementation, according to the number of times of sending each of the plurality of second-class questions, selecting a second-class question with the sending number greater than or equal to a preset number from the plurality of second-class questions, and obtaining a plurality of candidate verification answers corresponding to the second-class question meeting a first preset condition, where the plurality of candidate verification answers are feedback results given by a plurality of sample users for the second-class question.

The preset number may be determined according to an actual service requirement, an empirical value, or accuracy, where the preset number is preferably 50.

It should be noted that, for the second type of questions that do not satisfy the first preset condition, the second type of questions may be continuously sent to the sample user, and when the second type of questions satisfy the first preset condition, non-labeled verification answers are determined for the second type of questions.

Step c: for each second type question meeting the first preset condition, calculating the proportion of the number of times that each candidate verification answer in the plurality of candidate verification answers is answered to the total number of times that the plurality of candidate verification answers are answered.

In a specific implementation, for each second type of question that satisfies the first preset condition, the number of times that each candidate verification answer corresponding to each second type of question is answered is calculated, and the total number of times that a plurality of candidate answers corresponding to each second type of question are answered is calculated, and the proportion of the number of times to the total number of times is calculated, that is, the proportion of each candidate verification answer corresponding to each second type of question to the corresponding plurality of candidate verification answers is calculated.

Step d: and if the candidate verification answers with the proportion larger than or equal to the preset threshold exist, determining the candidate verification answers as the non-labeled verification answers of the second type of questions.

In a specific implementation, after calculating a ratio corresponding to each candidate verification answer corresponding to each second type of question meeting a first preset condition, selecting the candidate verification answer with the ratio greater than or equal to a preset threshold value to determine the candidate verification answer as a non-labeled verification answer of the second type of question.

Step e: and if the candidate verification answers with the proportion larger than or equal to the preset threshold value do not exist, abandoning the corresponding second type of problems.

In a specific implementation, if there is no candidate verification answer with a ratio greater than or equal to a preset threshold, it is described that the distribution of the candidate verification answers corresponding to the second type of question is relatively scattered, and a non-labeled verification answer of the second type of question cannot be determined, the second type of question is discarded, and the second type of question is not stored.

And (3): storing the first type of questions, the corresponding labeled verification answers and the corresponding target images in a verification question bank in an associated manner, and storing the second type of questions, the corresponding non-labeled verification answers and the corresponding target images in the verification question bank in an associated manner; the verification question bank is used for providing questions for verifying the user.

In a specific implementation, each first-class question, the corresponding annotated verification answer and the corresponding image may be associated and then stored in a verification information base as a set of verification information, and each second-class question, the corresponding non-annotated verification answer and the corresponding image may be associated and then stored in the verification information base as a set of verification information, so that the user may be verified through the verification information in the verification information base.

Further, after the step (3), the method for generating the verification problem further comprises the following steps: aiming at the obtained multiple first-class problems, determining a predicted verification answer of each first-class problem according to each first-class problem and image characteristic information corresponding to the first-class problem; acquiring a first-class problem meeting a second preset condition from the plurality of first-class problems; the second preset condition is that the predicted verification answer corresponding to the first type of question is different from the marked verification answer; and adjusting network parameters in the image recognition model for generating the image characteristic information through the first kind of questions meeting the second preset condition and the marking verification answers corresponding to the first kind of questions.

In specific implementation, according to each first-class question and corresponding image feature information, a predicted verification answer corresponding to each first-class question can be determined, for example, the image feature information is position information of each bone point, the first-class question is a position of a clicked left-foot bone point, the predicted verification answer is position information of a left bone point in an image, here, the labeled verification answer is an answer manually labeled on the first-class question, if it is detected that the predicted verification answer corresponding to a certain first-class question is inconsistent with the labeled verification answer, it is indicated that the image feature information obtained according to the image recognition model is inaccurate, and the image recognition model needs to be further trained, so as to further improve the accuracy of the output image feature information of the image recognition model.

It should be noted that the image recognition model may be further trained by the first type of question that satisfies the second preset condition and the labeled verification answer corresponding to the first type of question, and the accuracy of the output image feature information of the image recognition model may be improved by adjusting the network parameters in the image recognition model. Here, training the image recognition model only by the first-class problem satisfying the second preset condition, instead of training the image recognition model by all the first-class problems, can reduce the number of times of training and the amount of calculation.

Further, after the step (3), the method for generating the verification problem further comprises the following steps: aiming at the plurality of acquired second-class problems, determining a predictive verification answer of each second-class problem according to each second-class problem and image characteristic information corresponding to the second-class problem; acquiring a second type of problem meeting a third preset condition from the plurality of second type of problems; the third preset condition is that the predicted verification answers corresponding to the second type of questions are different from the non-labeled verification answers; and adjusting network parameters in the image recognition model for generating the image characteristic information through the second type of questions meeting the third preset condition and the non-labeling verification answers corresponding to the second type of questions.

In specific implementation, according to each second type of question and corresponding image feature information, a predicted verification answer corresponding to each second type of question can be determined, for example, the image feature information is position information of each bone point, the second type of question is a position of clicking a left bone point, the predicted verification answer is position information of a left bone point in an image, here, a non-labeled verification answer is a feedback result of the second type of question given by a sample user, and if it is detected that the predicted verification answer corresponding to a certain second type of question is not consistent with the non-labeled verification answer, it indicates that the image feature information obtained according to the image recognition model is not accurate, the image recognition model needs to be further trained, so as to further improve the accuracy of the output image feature information of the image recognition model.

It should be noted that the image recognition model may be further trained by the second type of questions meeting the third preset condition and the non-labeled verification answers corresponding to the second type of questions, and the accuracy of the output image feature information of the image recognition model may be improved by adjusting network parameters in the image recognition model. Here, training the image recognition model only by the second-class problem satisfying the third preset condition, instead of training the image recognition model by all the second-class problems, can reduce the number of times of training and the amount of calculation.

Fig. 2 is a flowchart of a verification method according to an embodiment of the present disclosure. As shown in fig. 2, the verification method provided in the embodiment of the present application is used for verifying a user operating a game, and includes the following steps:

s201: acquiring a group of verification information in response to a verification event triggered by a target user; the verification information includes an image of a game character in the game, a verification question asking a position of a target skeleton point, and position information of the target skeleton point as a verification answer.

In a specific implementation, after a target user triggers a verification event for verifying the user, a set of verification information is obtained, where the verification information may be generated in real time or obtained from a verification information library, where the set of verification information is composed of a verification question, a corresponding verification answer, and a corresponding image, and specifically, the image may be an image of a game character in a game, the verification question is an inquiry target skeleton point location, and the verification answer is position information of the target skeleton point.

It should be noted that the verification information may be matched with a verification service scenario corresponding to the verification event, and if the current verification service is the game a, the verification information related to the game a may be selected, for example, an image of any game character in the game a is selected as a verification image, a question for the image of the game character is a verification question, and an answer corresponding to the verification question is a verification answer, so that the verification information and the verification service scenario are more integrated.

Further, the verification event comprises one of the following events: a game account login event, a game payment event, and a game cheating event.

In a specific implementation, in a game scene, if a target user triggers a verification event, the target user is verified, and the verification event includes a game account login event, a game payment event, and a game cheating event.

Further, step S201, in response to an authentication event triggered by a target user, acquires a set of authentication information, which includes the following two ways:

the first method is as follows: and acquiring the image of any game character in the game, a verification question related to the image of the game character and inquiring the position of the target skeleton point, and a verification answer corresponding to the verification question from a verification information base.

In a specific implementation, the verification information may be obtained from a verification information base, and specifically, an image of any one game character in the current game, a verification question of any one skeletal point position associated with the obtained image of the game character, and a verification answer corresponding to the verification question may be obtained from the verification information base. Here, the image of each game character stored in the authentication information base may be associated with a plurality of authentication questions about the positions of the respective skeletal points.

The second method comprises the following steps: acquiring an image of any game role in the game from a game database; inputting the obtained image of the game role into a character skeleton point identification model, and outputting the position information of each skeleton point in the image of the game role; generating a plurality of verification questions inquiring the positions of all the skeleton points according to the position information of all the skeleton points; a verification question is selected from a plurality of verification questions asking for a location of a respective skeletal site.

In a specific implementation, the verification information may be generated in real time, specifically, an image of any one game character in the current game may be acquired from a game database of the current game, the image of the game character is input into the identification model of the task bone point, the position information of each bone point in the image of the game character is output, and further, a corresponding verification question may be generated for each bone point, and one verification question that asks the position of a target bone point may be selected from a plurality of verification questions that ask the position of each bone point.

S202: and sending the image of the game role and a verification question for inquiring the position of a target skeleton point to the terminal equipment of the target user.

In a specific implementation, the image of the game character and the verification question inquiring about the position of the target skeleton point are sent to the terminal equipment of the target user so as to verify the target user. Here, authentication questions of different authentication levels for an authentication event may be chosen.

S203: and determining a verification result for verifying the target user according to the feedback result of the terminal equipment and the verification answer.

In specific implementation, according to a feedback result given by the target user through the terminal device and the verification answer in the verification information, a verification result of the target user for verification can be determined.

Further, the feedback result of the terminal device includes: and the target user clicks the position information of any position of the image presented by the terminal equipment.

In a specific implementation, a verification result of the target user for verification may be determined by calculating a distance between a click position given by the target user and a position of a target bone point in the verification answer, specifically, when the distance is less than or equal to a preset distance, it is determined that the verification is passed, and when the distance is greater than the preset distance, it is determined that the verification is failed.

In the embodiment of the application, a group of verification information is acquired by responding to a verification event triggered by a target user, an image of a game role and a verification question inquiring about the position of a target skeleton point are sent to a terminal device of the target user, and further, a verification result for verifying the target user is determined according to a feedback result and a verification answer of the terminal device. Based on the mode, the verification information for providing the verification service is complex, so that the verification problem provided for the target user to carry out verification is difficult to crack by a computer.

Based on the same application concept, the embodiment of the present application further provides a device for generating the verification problem corresponding to the method for generating the verification problem, and since the principle of solving the problem by the device in the embodiment of the present application is similar to the method for generating the verification problem in the embodiment of the present application, the implementation of the device may refer to the implementation of the method, and repeated details are not repeated.

Referring to fig. 3 to 5, fig. 3 shows one of functional block diagrams of a verification problem generation apparatus 300 according to an embodiment of the present application, fig. 4 shows a functional block diagram of an extraction module 310 in fig. 3, and fig. 5 shows a second functional block diagram of the verification problem generation apparatus 300 according to an embodiment of the present application.

As shown in fig. 3, the verification problem generation apparatus 300 includes:

an extraction module 310, configured to obtain a target image and extract image feature information from the target image;

a generating module 320, configured to generate at least one verification question of the target image according to the image feature information.

In one possible implementation, as shown in fig. 4, the extraction module 310 includes:

an obtaining unit 312, configured to obtain the target image and determine at least one verification subject matching the target image;

the output unit 314 is configured to input the target image into the determined image recognition model corresponding to each verification subject, and output image feature information of the target image under each verification subject.

In a possible implementation manner, if the verification subject is the position of a human skeleton point, the image identification model corresponding to the verification subject is an identification model of the human skeleton point; as shown in fig. 4, the output unit 314 is configured to output the image feature information according to the following steps:

In a possible embodiment, the target image is multiple, as shown in fig. 5, the apparatus 300 for generating a verification question further includes:

an obtaining module 330, configured to obtain, from multiple verification questions corresponding to multiple target images, a first type of question labeled with a verification answer and a second type of question not labeled with a verification answer;

the determining module 340 is configured to send the second type of questions that are not labeled with the verification answers and the corresponding target images to the terminal devices of the multiple sample users, and determine the non-labeled verification answers of the second type of questions according to the feedback result;

a storage module 350, configured to store the first type of question, the corresponding annotated verification answer, and the corresponding target image in a verification question bank in an associated manner, and store the second type of question, the corresponding non-annotated verification answer, and the corresponding target image in the verification question bank in an associated manner; the verification question bank is used for providing questions for verifying the user.

In one possible implementation, as shown in fig. 5, the determining module 340 is configured to not determine the non-labeled verification answer for the second type of question according to the following:

In one possible embodiment, as shown in fig. 5, the generation apparatus 300 of the verification problem further includes a first adjusting module 360; the first adjusting module 360 is configured to adjust network parameters in the image recognition model according to the following steps:

In one possible embodiment, as shown in fig. 5, the generation apparatus 300 of the verification problem further includes a second adjustment module 370; the second adjusting module 370 is configured to adjust the network parameters in the image recognition model according to the following steps:

Based on the same application concept, a verification device corresponding to the verification method is further provided in the embodiment of the present application, and as the principle of solving the problem of the device in the embodiment of the present application is similar to the verification method in the embodiment of the present application, the implementation of the device may refer to the implementation of the method, and repeated details are not repeated.

Referring to fig. 6, a functional block diagram of an authentication apparatus 600 according to an embodiment of the present application is shown, where the authentication apparatus 600 includes:

an obtaining module 610, configured to obtain a set of verification information in response to a verification event triggered by a target user; the verification information comprises an image of a game character in the game, a verification question asking a position of a target skeleton point, and position information of the target skeleton point as a verification answer;

a sending module 620, configured to send the image of the game character and a verification question asking a position of a target skeleton point to a terminal device of the target user;

a determining module 630, configured to determine, according to the feedback result of the terminal device and the verification answer, a verification result for verifying the target user.

a game account login event, a game payment event, and a game cheating event.

In one possible implementation, as shown in fig. 6, the obtaining module 610 obtains a set of verification information according to the following steps:

acquiring an image of any game role in the game from a game database;

Based on the same application concept, referring to fig. 7, a schematic structural diagram of an electronic device 700 provided in the embodiment of the present application includes: a processor 710, a memory 720 and a bus 730, wherein the memory 720 stores machine-readable instructions executable by the processor 710, the processor 710 and the memory 720 communicate via the bus 730 when the electronic device 700 is operated, and the machine-readable instructions are executed by the processor 710 to perform the steps of the method for generating a verification problem according to any of the embodiments and/or to perform the steps of the method for verifying according to any of the embodiments.

In particular, the machine readable instructions, when executed by the processor 710, may perform the following:

Based on the same application concept, embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored, and the computer program is executed by a processor to perform the steps of the method for generating the verification problem provided in the foregoing embodiments and/or perform the steps of the verification method provided in the foregoing embodiments.

Specifically, the storage medium can be a general-purpose storage medium, such as a mobile disk, a hard disk, or the like, and when a computer program on the storage medium is run, the generation of the verification problem can be performed, so that not only a large number of verification problems can be generated, but also the difficulty of cracking the generated verification problem by a computer can be improved because the verification problem is generated through image feature information.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for generating a verification question, the method comprising:

generating at least one verification question of the target image according to the image characteristic information;

2. The generation method according to claim 1, wherein the acquiring a target image and extracting image feature information from the target image includes:

3. The generation method according to claim 2, wherein the verification topic comprises at least one of the following topics:

4. The generation method according to claim 2, wherein if the verification subject is the position of the human skeleton point, the image recognition model corresponding to the verification subject is the recognition model of the human skeleton point; the step of inputting the target image into the determined image recognition model corresponding to each verification subject and outputting the image characteristic information of the target image under each verification subject comprises the following steps:

5. The method for generating the verification result according to claim 1, wherein the step of sending the second type of questions without marked verification answers and the corresponding target images to the terminal devices of a plurality of sample users and determining the non-marked verification answers of the second type of questions according to the feedback result comprises:

6. The method of generating as claimed in claim 1, wherein after storing the first type of question, the corresponding annotated verification answer and the corresponding target image association in a verification question bank, and storing the second type of question, the corresponding non-annotated verification answer and the corresponding target image association in the verification question bank, the method further comprises:

7. The method of generating as claimed in claim 1, wherein after storing the first type of question, the corresponding annotated verification answer and the corresponding target image association in a verification question bank, and storing the second type of question, the corresponding non-annotated verification answer and the corresponding target image association in the verification question bank, the method further comprises:

8. An authentication method for authenticating a user operating a game, the authentication method comprising:

acquiring a group of verification information in response to a verification event triggered by a target user; the verification information comprises an image of a game character in the game, a verification question asking a position of a target skeleton point, and position information of the target skeleton point as a verification answer; the verification information is obtained from a verification information base; the verification information base stores a first type of questions marked with verification answers, corresponding marked verification answers and corresponding target images, a second type of questions not marked with verification answers, corresponding non-marked verification answers fed back by a user and corresponding target images;

9. The authentication method of claim 8, wherein the authentication event comprises one of:

a game account login event, a game payment event, and a game cheating event.

10. The authentication method of claim 8, wherein said obtaining a set of authentication information in response to an authentication event triggered by a target user comprises:

11. The authentication method of claim 8, wherein said obtaining a set of authentication information in response to an authentication event triggered by a target user comprises:

acquiring an image of any game role in the game from a game database;

12. The authentication method according to claim 8, wherein the feedback result of the terminal device comprises: and the target user clicks the position information of any position of the image presented by the terminal equipment.

13. A verification question generation apparatus, comprising:

the generating module is used for generating at least one verification problem of the target image according to the image characteristic information;

the first acquisition module is used for acquiring a first type of questions marked with verification answers and a second type of questions not marked with verification answers from a plurality of verification questions corresponding to a plurality of target images;

the determining module is used for sending the second type of questions which are not marked with the verification answers and the corresponding target images to the terminal equipment of a plurality of sample users, and determining the non-marked verification answers of the second type of questions according to the feedback results;

the storage module is used for storing the first type of questions, the corresponding labeling verification answers and the corresponding target images in a verification question bank in a correlated manner, and storing the second type of questions, the corresponding non-labeling verification answers and the corresponding target images in the verification question bank in a correlated manner; the verification question bank is used for providing questions for verifying the user.

14. An authentication apparatus for authenticating a user operating a game, the authentication apparatus comprising:

the second acquisition module is used for responding to a verification event triggered by a target user and acquiring a group of verification information; the verification information comprises an image of a game character in the game, a verification question asking a position of a target skeleton point, and position information of the target skeleton point as a verification answer; the verification information is obtained from a verification information base; the verification information base stores a first type of questions marked with verification answers, corresponding marked verification answers and corresponding target images, a second type of questions not marked with verification answers, corresponding non-marked verification answers fed back by a user and corresponding target images;

15. An electronic device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating over the bus when the electronic device is run, the machine-readable instructions when executed by the processor performing the steps of the method of generating a verification problem according to any one of claims 1 to 7 and/or the steps of the method of verifying according to any one of claims 8 to 12.

16. A computer-readable storage medium, characterized in that a computer program is stored thereon, which computer program, when being executed by a processor, performs the steps of the method for generating an authentication question as claimed in any one of claims 1 to 7, and/or the steps of the method for authenticating as claimed in any one of claims 8 to 12.