WO2022088805A1 - Artificial intelligence-based online credit method and apparatus, computer device, and medium - Google Patents

Abstract

The present application relates to the technical field of artificial intelligence, and provides an artificial intelligence-based online credit method and apparatus, a computer device, and a medium. The method comprises: obtaining a dual-recording video of a preset text read by a credit requester; determining, according to the preset text, the dual-recording video, and a document image of the credit requester, whether the reading is successful; upon determining that the reading is successful, using a willingness recognition model to recognize a willingness type of the credit requester on the basis of the dual-recording video; when the willingness type is a target willingness type, receiving a digital password inputted by the credit requester; generating a first public key according to the digital password, and generating a digital signature according to the first public key and the document image; and generating a credit contract according to the digital signature. The present application can improve the efficiency of credit application.

Description

Artificial intelligence-based online credit method, device, computer equipment and medium

This application claims the priority of the Chinese patent application filed on October 30, 2020 with the application number 202011198141.0 and the invention titled "Artificial Intelligence-based Online Credit Method, Device, Computer Equipment and Medium", all of which are The contents are incorporated herein by reference.

technical field

The present application relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based online credit method, device, computer equipment and medium.

Background technique

The traditional bank credit business is carried out offline and based on the face-to-face signing model of offline outlets, resulting in a very limited business scope of bank credit business, which cannot grant credit to the vast number of Internet users on a large scale and in batches.

Although with the development of network technology gradually penetrated into people's daily consumption life, more and more online business activities choose to conduct online, avoiding the trouble of signing, exchanging and saving traditional paper contracts, but the Internet-based credit business Electronic contracts still require the user's handwritten electronic signature. The inventor realized that, on the one hand, handwritten electronic signatures are prone to imitation and forgery, which leads to the inability to guarantee the security of credit business electronic contracts. Low.

SUMMARY OF THE INVENTION

In view of the above, it is necessary to propose an online credit method, device, computer equipment and medium based on artificial intelligence, which can improve the efficiency of credit application.

A first aspect of the present application provides an artificial intelligence-based online credit method, the method comprising:

Obtain a double-recorded video of a credit requester reading preset text;

Determine whether the reading is passed according to the preset text, the double-recorded video and the credit requester's certificate image;

using a willingness recognition model to identify the type of willingness of the credit requester based on the double-recorded video when a pass reading is determined;

When the willingness type is the target willingness type, receive the digital password input by the credit requester;

generating a first public key according to the digital password, and generating a digital signature according to the first public key and the certificate image;

A credit contract is generated based on the digital signature.

A second aspect of the present application provides an artificial intelligence-based online credit device, the device comprising:

An acquisition module for acquiring a double-recorded video of a credit requester reading a preset text;

a judging module for judging whether the reading is passed according to the preset text, the double-recorded video and the credit requester's certificate image;

an identification module for identifying the credit requester's willingness type based on the double-recorded video using a willingness recognition model when it is determined to pass the reading;

a receiving module, configured to receive the digital password input by the credit requester when the willingness type is the target willingness type;

a signature module, configured to generate a first public key according to the digital password, and generate a digital signature according to the first public key and the certificate image;

A generating module for generating a credit contract according to the digital signature.

A third aspect of the present application provides a computer device, the computer device includes a processor, and the processor is configured to implement the following steps when executing computer-readable instructions stored in a memory:

Obtain a double-recorded video of a credit requester reading preset text;

Determine whether the reading is passed according to the preset text, the double-recorded video and the credit requester's certificate image;

using a willingness recognition model to identify the type of willingness of the credit requester based on the double-recorded video when a pass reading is determined;

When the willingness type is the target willingness type, receive the digital password input by the credit requester;

generating a first public key according to the digital password, and generating a digital signature according to the first public key and the certificate image;

A credit contract is generated based on the digital signature.

A fourth aspect of the present application provides a computer-readable storage medium, where computer-readable instructions are stored thereon, and when the computer-readable instructions are executed by a processor, the following steps are implemented:

Obtain a double-recorded video of a credit requester reading preset text;

Determine whether the reading is passed according to the preset text, the double-recorded video and the credit requester's certificate image;

using a willingness recognition model to identify the type of willingness of the credit requester based on the double-recorded video when a pass reading is determined;

When the willingness type is the target willingness type, receive the digital password input by the credit requester;

generating a first public key according to the digital password, and generating a digital signature according to the first public key and the certificate image;

A credit contract is generated based on the digital signature.

To sum up, the artificial intelligence-based online credit method, device, computer equipment and medium described in this application can be applied in areas such as smart government affairs that require encryption of massive data, thereby promoting the development of smart cities. When the double-recorded video of the credit requester reading the preset text is obtained, it is determined whether the credit requester has passed the reading according to the preset text, the double-recorded video and the image of the credit requester's certificate; Through reading, the willingness recognition model is used to identify the willingness type of the credit requester based on the double-recorded video, and only when the willingness type is the target willingness type, the digital password input by the credit requester is received, and then according to the The digital password generates a first public key, and a digital signature is generated according to the first public key and the certificate image, and finally a credit contract is generated according to the digital signature. This application can generate a credit contract based on the double-recorded video of the credit requester, which improves the efficiency of credit application, and based on the double-recorded video, it can avoid credit application data fraud and ensure the security of the credit contract.

Description of drawings

FIG. 1 is a flowchart of an artificial intelligence-based online credit method provided in Embodiment 1 of the present application.

FIG. 2 is a structural diagram of an artificial intelligence-based online credit device provided in Embodiment 2 of the present application.

FIG. 3 is a schematic structural diagram of a computer device provided in Embodiment 3 of the present application.

Detailed ways

In order to more clearly understand the above objects, features and advantages of the present application, the present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments may be combined with each other in the case of no conflict.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs. The terms used herein in the specification of the application are for the purpose of describing specific embodiments only, and are not intended to limit the application.

The artificial intelligence-based online credit method provided by the embodiments of the present application is executed by computer equipment, and correspondingly, the artificial intelligence-based online credit device runs in the computer equipment.

FIG. 1 is a flowchart of an artificial intelligence-based online credit method provided in Embodiment 1 of the present application. The artificial intelligence-based online credit method specifically includes the following steps. According to different requirements, the sequence of the steps in the flowchart can be changed, and some can be omitted.

S11, obtain a double-recorded video of the credit requester reading the preset text.

The preset text refers to the text that displays business risks, loan instructions, etc. on the client of the credit requester in the form of text in order to achieve the obligation of the customer to know when the online credit is performed.

A credit application dedicated to credit service is pre-installed on the client of the credit requester, and the credit requester records a double-recorded video of reading the preset text through the credit application on the client. Specifically, the credit requester activates the recording function on the credit application, reads against the preset text, and the credit application records a double-recorded video through the recording function. After the double-recording video recording is completed, the double-recording video is uploaded to the credit server through the client.

In this embodiment, by obtaining a double-recorded video of the credit requester reading the preset text, it can be ensured that the credit requester actually reads, thereby realizing the obligation of informing the credit requester and reducing the risk of subsequent credit complaints.

S12: Determine whether the reading is passed according to the preset text, the double-recorded video, and the image of the credit requester's certificate.

The credit requester successfully installs the credit application program on the client, and needs to perform identity information registration when logging in for the first time. For example, the registration of identity information is completed by uploading the certificate image through the certificate image upload function in the credit application. The certificate image may be an identity card image, a passport image, a driver's license image, and the like.

In an optional embodiment, the determining whether to pass the reading according to the preset text, the double-recorded video and the certificate image of the credit requester includes:

Extracting audio in the double-recording video and extracting multiple frame images in the dual-recording video;

Recognizing the audio to obtain voice text, and comparing the voice text and the preset text to obtain a first comparison result;

Comparing the plurality of frame images and the certificate images to obtain a second comparison result;

Whether the credit requester has passed the reading is determined according to the first comparison result and the second comparison result.

The double-recorded video includes audio and frame image sequences. Audio separation technology is used to separate the audio from the double-recorded video, and speech recognition technology is used to recognize the audio to obtain speech text. The audio separation technology and the speech recognition technology may be the prior art. The computer device may extract a plurality of frame images from the frame image sequence according to a preset fixed collection frequency.

Since the credit requester has a dialect difference, the computer device presets a first similarity threshold to judge whether the audio comparison of the credit requester passes. In a specific implementation, after obtaining the voice text, the computer device calculates a first similarity between the voice text and the preset text; compares the first similarity with a preset first similarity threshold; When the first similarity is greater than or equal to the preset first similarity threshold, a first comparison result that passes the audio comparison is obtained; when the first similarity is less than the preset first similarity threshold, obtain The result of the first alignment that failed the audio alignment.

Since the faces in the dual-recording video are different due to light and other reasons when recording the dual-recording video, the computer device presets a second similarity threshold to determine whether the image comparison of the credit requester passes. In a specific implementation, after obtaining a plurality of frame images, the computer device calculates the second similarity between each frame image and the certificate image; compares each second similarity with a preset second similarity threshold; When every second similarity is greater than or equal to the preset second similarity threshold, a second comparison result that passes the image comparison is obtained; when any second similarity is less than the preset second similarity threshold, A second comparison result that fails the audio comparison is obtained.

The computer device finally determines whether the credit requester has passed the reading in combination with the first comparison result and the second comparison result. In specific implementation, when the first comparison result is that the audio comparison is passed and the second comparison result is that the image comparison is passed, it is determined that the credit requester has passed the reading; when the first comparison result is If the audio comparison fails and the second comparison result is that the image comparison fails, it is determined that the credit requester fails the reading.

In an optional embodiment, the extracting multiple frame images in the dual-recording video includes:

Calculate the double-recording time of the double-recording video;

Generate the number of detections according to the double recording time;

Calculate the detection frame rate according to the double recording time and the detection times;

Extracting a plurality of frame images from the dual recording video using the detected frame rate.

In this optional embodiment, the computer device calculates the double-recording time of each double-recording video according to the start time node and the end time node of the process link corresponding to each double-recording sub-video.

In order to avoid data fraud during the credit application process, the computer equipment generates the number of detections according to the double-recording time of the double-recording video, so as to determine how to extract frame images from the dual-recording video for face detection according to the number of detections. According to the double-recording time, different detection times can be generated for different double-recording videos, so that the extracted frame images have greater randomness, so that the face detection on the frame images also has greater randomness, which can effectively Ensure the authenticity and reliability of face detection.

The detection frame rate can be obtained by calculating the quotient between the double recording time and the number of detections by the computer equipment. For example, if the double recording time is 5 minutes and the number of detections is 10, the detection frame rate is 5 minutes/10=30 seconds, that is, every 30 seconds to extract a frame image from the double-recorded video.

In an optional embodiment, the generating the detection times according to the double recording time includes:

Define the upper and lower limits of the first detection times and the upper and lower limits of the second detection times;

Calculate the average time value of the historical double recording time;

comparing the double recording time with the average time value;

When the double-recording time is greater than or equal to the average time value, a first random number is generated within the numerical range corresponding to the upper and lower limits of the first detection times, as the detection times of the double-recording video;

When the double-recording time is less than the average time value, a second random number is generated within a numerical range corresponding to the upper and lower limits of the second detection times as the detection times of the double-recording video.

Wherein, the upper and lower limits of the first detection times are [x1, x2], the upper and lower limits of the second detection times are [x3, x4], and the lower limit of the first detection times is greater than the upper limit of the second detection times, that is, the second detection times The lower limit of the number of times x3 < the upper limit of the second number of detection times x4 < the lower limit of the first number of detection times x1 < the upper limit of the first number of detection times x2.

As an example, suppose there are 3 historical dual-recording videos, the dual-recording time of the first historical dual-recording video is T1, the dual-recording time of the second historical dual-recording video is T2, and the dual-recording time of the third historical dual-recording video is T2. If the time is T3, the average time value of the historical double-recording time is calculated as B1=T1/(T1+T2+T3).

When the double-recording time is greater than or equal to the average time value, it indicates that the dual-recording video belongs to a relatively long dual-recording video, so more video frames can be extracted; when the dual-recording time is less than the The average time value indicates that the dual-recording video is relatively short dual-recording video, so fewer video frames can be extracted.

In this optional embodiment, by generating the first random number within the numerical range corresponding to the upper and lower limits of the first detection times, the generated first random number can be made larger, so that the first random number is used as the When the number of detections of the double-recorded video is performed, more video frames can be extracted, and by generating the second random number within the numerical range corresponding to the upper and lower limits of the second detection number, the generated second random number can be smaller, Therefore, when the second random number is used as the detection times of the double recording video, fewer video frames can be extracted. In this way, the number of extracted video frames can be automatically determined, and the extraction efficiency of video frames is high.

S13, when it is determined to pass the reading, use a willingness identification model to identify the willingness type of the credit requester based on the double-recorded video.

The willingness recognition model is obtained by offline training of computer equipment in advance, and is used to recognize the willingness type of the credit requester based on the double-recorded video. Wherein, the willingness type includes: willingness and unwillingness. If the willingness type is willing, it indicates that the credit requester is truly willing to apply for credit. If the willingness type is unwilling, it indicates that the credit requester is not really willing to apply for credit, for example, is coerced or forced to apply for credit.

In an optional embodiment, the training process of the intention recognition model may include:

Acquire multiple double-recorded videos corresponding to multiple will types, and identify the voice text in each double-recorded video corresponding to each will type;

Carrying out line-by-line encoding processing to the speech text to obtain a plurality of encoding vectors;

Input each coding vector into the forward long short-term memory network layer to obtain the first vector, and input each coding vector into the reverse long short-term memory network layer to obtain the second vector;

The multiple first vectors and multiple second vectors corresponding to the voice text are spliced according to the row order of the voice text to obtain a first input vector;

generating a feature vector according to the willingness type and the first input vector corresponding to the willingness type;

A support vector machine is trained based on a plurality of the feature vectors to obtain a willingness type identification model.

In this optional embodiment, the computer device first obtains a preset paragraph label in the language text, and performs block processing on the language text according to the preset paragraph label to obtain a plurality of paragraph texts; and then obtains For the preset line breaks in the language text, each segment of text is processed into lines according to the preset line breaks to obtain multiple lines of text.

The computer device may divide a plurality of feature vectors by using a cross-validation method to obtain the first set and the second set. During specific implementation, the computer device randomly divides the plurality of feature vectors into at least one data packet according to a preset ratio, and determines any one data packet in the at least one data packet as the second set, and the rest The data packets are determined to be the first set, and the above steps are repeated until all the data packets are sequentially used as the second set. Wherein, the preset ratio can be set by self-definition, which is not limited in this application. A binary classification model is trained based on the first set, and the binary classification model is validated based on the second set. The specific training process and verification process are in the prior art, which will not be described in detail in this application.

After the computer equipment is trained to obtain a willingness recognition model, the voice text corresponding to the double-recorded video is subjected to line-by-line encoding processing to obtain a plurality of encoding vectors, and each encoding vector is input into the forward long short-term memory network layer to obtain the first number of encoding vectors. three vectors, and input each coding vector into the reverse long short-term memory network layer to obtain a fourth vector; according to the row order of the voice text, multiple third vectors and multiple fourth vectors corresponding to the voice text The vectors are spliced to obtain a second input vector; the second input vector is input into the intention identification model, and the intention type is obtained by identifying the second input vector by the intention identification model.

In this optional embodiment, by performing block processing on the language text and then performing branch processing, it can be avoided that the end of the segment in the speech text does not have a preset newline character and cannot be divided into lines, thereby improving the accuracy of line branching; After the accuracy of the branch is improved, the encoding accuracy of the branch code can be improved, so that the forward long short-term memory network layer and the reverse long short-term memory network layer are used to process the encoding vector, so that each encoding vector can be More in line with contextual semantics, the training accuracy of the willingness recognition model is improved; by dividing multiple feature vectors, each data row in the multiple feature vectors participates in training and verification, thereby improving the training of the willingness recognition model. the fit.

S14, when the willingness type is the target willingness type, receive the digital password input by the credit requester.

The target willingness type is a willingness type pre-designated by the computer device. Exemplarily, the target willingness type may be willingness.

When determining that the credit requester's willingness type is the target willingness type, the computer device displays a digital password input interface for the credit requester to input a digital password in the digital password digital interface. The digital password is used for the password required to be input for identity verification during subsequent bank lending.

S15: Generate a first public key according to the digital password, and generate a digital signature according to the first public key and the certificate image.

A key generation algorithm is pre-stored in the computer device, the digital password is used as an input parameter of the key generation algorithm, and the obtained key value is used as the first public key. The certificate image is encrypted using the first public key, and the encrypted encrypted bits are extracted to obtain a digital signature.

In an optional embodiment, the generating the first public key according to the digital password includes:

determining a character string corresponding to the digital password;

compute the hash value of the string;

Obtain the first system parameter and the second system parameter;

Using the first system parameter, the second system parameter and the hash value to calculate the first verification parameter that complies with the El Gamal commitment, and determine the first verification parameter as the first public key.

In this optional embodiment, a correspondence table between numbers and characters is stored in the computer device, and according to the correspondence table, the character corresponding to each password in the digital password is determined, and then the The characters are concatenated in the order of the numeric password to form a string. The computer device may use a hash function to calculate the character string to obtain a hash value. The hash function may be a message digest algorithm (Message-Digest Algorithm, MD5). By calculating the hash value of the character string through MD5, a 128-bit (16-byte) hash value can be generated.

The first system parameter and the second system parameter are both public parameters, the first system parameter may be an elliptic curve group generator, and the second system parameter may be a natural number. Exemplarily, assuming that the first system parameter is g, the second system parameter is n, and the hash value is r, then use the first system parameter g, the second system parameter n, and the hash value r to calculate The first verification parameter F=gr ^modn in accordance with the El Gamal commitment.

S16. Generate a credit contract according to the digital signature.

A credit contract template is pre-stored in the computer device, the basic information of the credit requester is filled in the position corresponding to the key field in the credit contract template, and the digital signature is added at the signature position, thereby generating a credit Requester's credit contract.

The digital signature generates a credit contract, provides solid and effective real data for bank credit, provides data basis for credit decision-making, and simplifies the application process and application cost.

In an optional embodiment, after generating the credit contract according to the digital signature, the method further includes:

using the first public key as the contract code of the credit contract;

After receiving the sign-off instruction from the signer, sending a digital password acquisition instruction to the client of the credit requester;

Receive the digital password sent by the client and input by the credit requester according to the digital password acquisition instruction;

generating a second public key according to the received digital password;

verifying whether the second public key is the same as the contract code;

When it is determined that the second public key is the same as the contract code, a preset lending operation is performed.

In this optional embodiment, if the credit requester really voluntarily applies for credit, when receiving the digital password acquisition instruction sent by the computer device through the client, the same digital password will be input again and sent to the client through the client. computer equipment, then the public key calculated by the computer equipment using the same secret key generation algorithm based on the digital password will be consistent with the contract code on the credit contract.

By re-entering the digital password and verifying the credit requester, the credit requester can be authenticated twice, which ensures the security of the credit application, reduces the time spent on verification, and improves the efficiency of the review. After that, the loan operation is completed online, without the need for paper documents and offline manual processing, making the operation of applying for loan simple and fast.

It should be emphasized that, in order to further ensure the privacy and security of the above-mentioned willingness type identification model, the above-mentioned willingness type identification model can be stored in the nodes of the blockchain.

When the double-recorded video of the credit requester reading the preset text is obtained, it is determined whether the credit requester has passed the reading according to the preset text, the double-recorded video and the image of the credit requester's certificate; Through reading, the willingness recognition model is used to identify the willingness type of the credit requester based on the double-recorded video, and only when the willingness type is the target willingness type, the digital password input by the credit requester is received, and then according to the The digital password generates a first public key, and a digital signature is generated according to the first public key and the certificate image, and finally a credit contract is generated according to the digital signature. This application can be applied in smart government affairs and other fields that need to encrypt massive data, thereby promoting the development of smart cities. Video recording can avoid fraudulent credit application data and ensure the security of credit contracts.

FIG. 2 is a structural diagram of an artificial intelligence-based online credit device provided in Embodiment 2 of the present application.

In some embodiments, the artificial intelligence-based online credit device 20 may include a plurality of functional modules composed of computer-readable instruction segments. Computer readable instructions for each program segment in the artificial intelligence-based online credit device 20 may be stored in the memory of the computer device and executed by at least one processor to execute (detailed in the description of FIG. 1 ) artificial intelligence-based Smart online credit features.

In this embodiment, the online credit device 20 based on artificial intelligence can be divided into a plurality of functional modules according to the functions it performs. The functional modules may include: an acquisition module 201 , a judgment module 202 , an identification module 203 , a training module 204 , a reception module 205 , a signature module 206 , a generation module 207 and an execution module 208 . A module referred to in this application refers to a series of computer-readable instruction segments that can be executed by at least one processor and can perform fixed functions, and are stored in a memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

The obtaining module 201 is configured to obtain a double-recorded video of the credit requester reading the preset text.

The preset text refers to the text that displays business risks, loan instructions, etc. on the client of the credit requester in the form of text in order to achieve the obligation of the customer to know when the online credit is performed.

A credit application dedicated to credit service is pre-installed on the client of the credit requester, and the credit requester records a double-recorded video of reading the preset text through the credit application on the client. Specifically, the credit requester activates the recording function on the credit application, reads against the preset text, and the credit application records a double-recorded video through the recording function. After the double-recording video recording is completed, the double-recording video is uploaded to the credit server through the client.

In this embodiment, by obtaining a double-recorded video of the credit requester reading the preset text, it can be ensured that the credit requester actually reads, thereby realizing the obligation of informing the credit requester and reducing the risk of subsequent credit complaints.

The judging module 202 is configured to judge whether the reading is passed according to the preset text, the double-recorded video and the certificate image of the credit requester.

The credit requester successfully installs the credit application program on the client, and needs to perform identity information registration when logging in for the first time. For example, the registration of identity information is completed by uploading a credential image through the credential image upload function in the credit application. The certificate image may be an identity card image, a passport image, a driver's license image, and the like.

In an optional embodiment, the judging module 202 judges whether to pass the reading according to the preset text, the double-recorded video and the certificate image of the credit requester, including:

Extracting audio in the double-recording video and extracting multiple frame images in the dual-recording video;

Recognizing the audio to obtain voice text, and comparing the voice text and the preset text to obtain a first comparison result;

Comparing the plurality of frame images and the certificate images to obtain a second comparison result;

Whether the credit requester has passed the reading is determined according to the first comparison result and the second comparison result.

The double-recorded video includes audio and frame image sequences. Audio separation technology is used to separate the audio from the double-recorded video, and speech recognition technology is used to recognize the audio to obtain speech text. The audio separation technology and the speech recognition technology may be the prior art. The computer device may extract a plurality of frame images from the frame image sequence according to a preset fixed collection frequency.

Since the credit requester has a dialect difference, the computer device presets a first similarity threshold to judge whether the audio comparison of the credit requester passes. In a specific implementation, after obtaining the voice text, the computer device calculates a first similarity between the voice text and the preset text; compares the first similarity with a preset first similarity threshold; When the first similarity is greater than or equal to the preset first similarity threshold, a first comparison result that passes the audio comparison is obtained; when the first similarity is less than the preset first similarity threshold, obtain The result of the first alignment that failed the audio alignment.

Since the faces in the dual-recording video are different due to light and other reasons when recording the dual-recording video, the computer device presets a second similarity threshold to determine whether the image comparison of the credit requester passes. In a specific implementation, after obtaining a plurality of frame images, the computer device calculates the second similarity between each frame image and the certificate image; compares each second similarity with a preset second similarity threshold; When every second similarity is greater than or equal to the preset second similarity threshold, a second comparison result that passes the image comparison is obtained; when any second similarity is less than the preset second similarity threshold, A second comparison result that fails the audio comparison is obtained.

The computer device finally determines whether the credit requester has passed the reading in combination with the first comparison result and the second comparison result. In specific implementation, when the first comparison result is that the audio comparison is passed and the second comparison result is that the image comparison is passed, it is determined that the credit requester has passed the reading; when the first comparison result is If the audio comparison fails and the second comparison result is that the image comparison fails, it is determined that the credit requester fails the reading.

In an optional embodiment, the extracting multiple frame images in the dual-recording video includes:

Calculate the double-recording time of the double-recording video;

Generate the number of detections according to the double recording time;

Calculate the detection frame rate according to the double recording time and the detection times;

Extracting a plurality of frame images from the dual recording video using the detected frame rate.

In this optional embodiment, the computer device calculates the double-recording time of each double-recording video according to the start time node and the end time node of the process link corresponding to each double-recording sub-video.

In order to avoid data fraud during the credit application process, the computer equipment generates the number of detections according to the double-recording time of the double-recording video, so as to determine how to extract frame images from the dual-recording video for face detection according to the number of detections. According to the double-recording time, different detection times can be generated for different double-recording videos, so that the extracted frame images have greater randomness, so that the face detection on the frame images also has greater randomness, which can effectively Ensure the authenticity and reliability of face detection.

The detection frame rate can be obtained by calculating the quotient between the double recording time and the number of detections by the computer equipment. For example, if the double recording time is 5 minutes and the number of detections is 10, the detection frame rate is 5 minutes/10=30 seconds, that is, every 30 seconds to extract a frame image from the double-recorded video.

In an optional embodiment, the generating the detection times according to the double recording time includes:

Define the upper and lower limits of the first detection times and the upper and lower limits of the second detection times;

Calculate the average time value of the historical double recording time;

comparing the double recording time with the average time value;

When the double-recording time is greater than or equal to the average time value, a first random number is generated within the numerical range corresponding to the upper and lower limits of the first detection times, as the detection times of the double-recording video;

When the double-recording time is less than the average time value, a second random number is generated within a numerical range corresponding to the upper and lower limits of the second detection times as the detection times of the double-recording video.

Wherein, the upper and lower limits of the first detection times are [x1, x2], the upper and lower limits of the second detection times are [x3, x4], and the lower limit of the first detection times is greater than the upper limit of the second detection times, that is, the second detection times The lower limit of the number of times x3 < the upper limit of the second number of detection times x4 < the lower limit of the first number of detection times x1 < the upper limit of the first number of detection times x2.

As an example, suppose there are 3 historical dual-recording videos, the dual-recording time of the first historical dual-recording video is T1, the dual-recording time of the second historical dual-recording video is T2, and the dual-recording time of the third historical dual-recording video is T2. If the time is T3, the average time value of the historical double-recording time is calculated as B1=T1/(T1+T2+T3).

When the double-recording time is greater than or equal to the average time value, it indicates that the dual-recording video belongs to a relatively long dual-recording video, so more video frames can be extracted; when the dual-recording time is less than the The average time value indicates that the dual-recording video is relatively short dual-recording video, so fewer video frames can be extracted.

In this optional embodiment, by generating the first random number within the numerical range corresponding to the upper and lower limits of the first detection times, the generated first random number can be made larger, so that the first random number is used as the When the number of detections of the double-recorded video is performed, more video frames can be extracted, and by generating the second random number within the numerical range corresponding to the upper and lower limits of the second detection number, the generated second random number can be smaller, Therefore, when the second random number is used as the detection times of the double recording video, fewer video frames can be extracted. In this way, the number of extracted video frames can be automatically determined, and the extraction efficiency of video frames is high.

The identifying module 203 is configured to identify the willingness type of the credit requester based on the double-recorded video using a willingness recognition model when it is determined that the reading is passed.

The willingness recognition model is obtained by offline training of computer equipment in advance, and is used to recognize the willingness type of the credit requester based on the double-recorded video. Wherein, the willingness type includes: willingness and unwillingness. If the willingness type is willing, it indicates that the credit requester is truly willing to apply for credit. If the willingness type is unwilling, it indicates that the credit requester is not really willing to apply for credit, for example, is coerced or forced to apply for credit.

The training module 204 is used for training a willingness recognition model.

In an optional embodiment, the training module 204 training the intention recognition model includes:

Acquire multiple double-recorded videos corresponding to multiple will types, and identify the voice text in each double-recorded video corresponding to each will type;

Carrying out line-by-line encoding processing to the speech text to obtain a plurality of encoding vectors;

Input each coding vector into the forward long short-term memory network layer to obtain the first vector, and input each coding vector into the reverse long short-term memory network layer to obtain the second vector;

The multiple first vectors and multiple second vectors corresponding to the voice text are spliced according to the row order of the voice text to obtain a first input vector;

generating a feature vector according to the willingness type and the first input vector corresponding to the willingness type;

A support vector machine is trained based on a plurality of the feature vectors to obtain a willingness type identification model.

In this optional embodiment, the computer device first obtains a preset paragraph label in the language text, and performs block processing on the language text according to the preset paragraph label to obtain a plurality of paragraph texts; and then obtains For the preset line breaks in the language text, each segment of text is processed into lines according to the preset line breaks to obtain multiple lines of text.

The computer device may divide a plurality of feature vectors by using a cross-validation method to obtain the first set and the second set. During specific implementation, the computer device randomly divides the plurality of feature vectors into at least one data packet according to a preset ratio, and determines any one data packet in the at least one data packet as the second set, and the rest The data packets are determined to be the first set, and the above steps are repeated until all the data packets are sequentially used as the second set. Wherein, the preset ratio can be set by self-definition, which is not limited in this application. A binary classification model is trained based on the first set, and the binary classification model is validated based on the second set. The specific training process and verification process are in the prior art, which will not be described in detail in this application.

After the computer equipment is trained to obtain a willingness recognition model, the voice text corresponding to the double-recorded video is subjected to line-by-line encoding processing to obtain a plurality of encoding vectors, and each encoding vector is input into the forward long short-term memory network layer to obtain the first number of encoding vectors. three vectors, and input each coding vector into the reverse long short-term memory network layer to obtain a fourth vector; according to the row order of the voice text, multiple third vectors and multiple fourth vectors corresponding to the voice text The vectors are spliced to obtain a second input vector; the second input vector is input into the intention identification model, and the intention type is obtained by identifying the second input vector by the intention identification model.

In this optional embodiment, by performing block processing on the language text and then performing branch processing, it can be avoided that the end of the segment in the speech text does not have a preset newline character and cannot be divided into lines, thereby improving the accuracy of line branching; After the accuracy of the branch is improved, the encoding accuracy of the branch code can be improved, so that the forward long short-term memory network layer and the reverse long short-term memory network layer are used to process the encoding vector, so that each encoding vector can be More in line with contextual semantics, the training accuracy of the willingness recognition model is improved; by dividing multiple feature vectors, each data row in the multiple feature vectors participates in training and verification, thereby improving the training of the willingness recognition model. of fit.

The receiving module 205 is configured to receive the digital password input by the credit requester when the willingness type is the target willingness type.

The target willingness type is a willingness type pre-designated by the computer device. Exemplarily, the target willingness type may be willingness.

When determining that the credit requester's willingness type is the target willingness type, the computer device displays a digital password input interface for the credit requester to input a digital password in the digital password digital interface. The digital password is used for the password required to be input for identity verification during subsequent bank lending.

The signature module 206 is configured to generate a first public key according to the digital password, and generate a digital signature according to the first public key and the certificate image.

A key generation algorithm is pre-stored in the computer device, the digital password is used as an input parameter of the key generation algorithm, and the obtained key value is used as the first public key. The certificate image is encrypted using the first public key, and the encrypted encrypted bits are extracted to obtain a digital signature.

In an optional embodiment, the signature module 206 generating the first public key according to the digital password includes:

determining a character string corresponding to the digital password;

compute the hash value of the string;

Obtain the first system parameter and the second system parameter;

Using the first system parameter, the second system parameter and the hash value to calculate the first verification parameter that complies with the El Gamal commitment, and determine the first verification parameter as the first public key.

In this optional embodiment, a correspondence table between numbers and characters is stored in the computer device, and according to the correspondence table, the character corresponding to each password in the digital password is determined, and then the The characters are concatenated in the order of the numeric password to form a string. The computer device may use a hash function to calculate the character string to obtain a hash value. The hash function may be a message digest algorithm (Message-Digest Algorithm, MD5). By calculating the hash value of the character string through MD5, a 128-bit (16-byte) hash value can be generated.

The first system parameter and the second system parameter are both public parameters, the first system parameter may be an elliptic curve group generator, and the second system parameter may be a natural number. Exemplarily, assuming that the first system parameter is g, the second system parameter is n, and the hash value is r, then use the first system parameter g, the second system parameter n, and the hash value r to calculate The first verification parameter F=gr ^modn in accordance with the El Gamal commitment.

The generating module 207 is configured to generate a credit contract according to the digital signature.

A credit contract template is pre-stored in the computer device, the basic information of the credit requester is filled in the position corresponding to the key field in the credit contract template, and the digital signature is added at the signature position, thereby generating a credit Requester's credit contract.

The digital signature generates a credit contract, provides solid and effective real data for bank credit, provides data basis for credit decision-making, and simplifies the application process and application cost.

The generating module 207 is further configured to use the first public key as the contract code of the credit contract.

The receiving module 205 is further configured to send a digital password obtaining instruction to the client of the credit requester after receiving the sign-off instruction of the signer; Password The numeric password entered by the command to get the password.

The generating module 207 is further configured to generate a second public key according to the received digital password.

The execution module 208 is configured to verify whether the second public key is the same as the contract code, and when it is determined that the second public key is the same as the contract code, execute a preset lending operation.

In this optional embodiment, if the credit requester really voluntarily applies for credit, when receiving the digital password acquisition instruction sent by the computer device through the client, the same digital password will be input again and sent to the client through the client. computer equipment, then the public key calculated by the computer equipment using the same secret key generation algorithm based on the digital password will be consistent with the contract code on the credit contract.

By re-entering the digital password and verifying the credit requester, the credit requester can be authenticated twice, which ensures the security of the credit application, reduces the time spent on verification, and improves the efficiency of the review. After that, the loan operation is completed online, without the need for paper documents and offline manual processing, making the operation of applying for loan simple and fast.

It should be emphasized that, in order to further ensure the privacy and security of the above-mentioned willingness type identification model, the above-mentioned willingness type identification model can be stored in the nodes of the blockchain.

When the double-recorded video of the credit requester reading the preset text is obtained, it is determined whether the credit requester has passed the reading according to the preset text, the double-recorded video and the image of the credit requester's certificate; Through reading, the willingness recognition model is used to identify the willingness type of the credit requester based on the double-recorded video, and only when the willingness type is the target willingness type, the digital password input by the credit requester is received, and then according to the The digital password generates a first public key, and a digital signature is generated according to the first public key and the certificate image, and finally a credit contract is generated according to the digital signature. This application can be applied in smart government affairs and other fields that need to encrypt massive data, thereby promoting the development of smart cities. Video recording can avoid fraudulent credit application data and ensure the security of credit contracts.

Referring to FIG. 3 , it is a schematic structural diagram of a computer device according to Embodiment 3 of the present application. In a preferred embodiment of the present application, the computer device 3 includes a memory 31 , at least one processor 32 , at least one communication bus 33 and a transceiver 34 .

Those skilled in the art should understand that the structure of the computer device shown in FIG. 3 does not constitute a limitation of the embodiments of the present application, and may be a bus-type structure or a star-shaped structure. more or less other hardware or software, or a different arrangement of components is shown.

In some embodiments, the computer device 3 is a device that can automatically perform numerical calculation and/or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits, Programmable gate arrays, digital processors and embedded devices, etc. The computer equipment 3 may also include client equipment, including but not limited to any electronic product that can interact with the client through a keyboard, a mouse, a remote control, a touchpad or a voice-activated device, etc., for example, Personal computers, tablets, smartphones, digital cameras, etc.

It should be noted that the computer equipment 3 is only an example, and other existing or future electronic products, if applicable to the present application, should also be included within the protection scope of the present application, and incorporated herein by reference .

In some embodiments, the memory 31 stores computer readable instructions that, when executed by the at least one processor 32, implement all of the artificial intelligence-based online credit methods described above. or part of the steps. The memory 31 includes volatile and non-volatile memories, such as random access memory (Random Access Memory, RAM), read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable Read-Only memory) Memory, PROM), Erasable Programmable Read-Only Memory (EPROM), One-time Programmable Read-Only Memory (OTPROM), Electronically Erasable Rewritable Only Memory Read-only memory (Electrically-Erasable Programmable Read-Only Memory, EEPROM), Compact Disc Read-Only Memory (CD-ROM), or other optical disk storage, magnetic disk storage, tape storage, or other capable of carrying or storing data computer-readable storage medium. The computer-readable storage medium may be non-volatile or volatile. .

Further, the computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function, and the like; The data created by the use of the node, etc.

The blockchain referred to in this application is a new application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm. Blockchain, essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information to verify its Validity of information (anti-counterfeiting) and generation of the next block. The blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

In some embodiments, the at least one processor 32 is a control core (Control Unit) of the computer device 3, using various interfaces and lines to connect various components of the entire computer device 3, and by running or executing storage in the computer device 3 The programs or modules in the memory 31 and the data stored in the memory 31 are called to perform various functions of the computer device 3 and process data. For example, when the at least one processor 32 executes the computer-readable instructions stored in the memory, it implements all or part of the steps of the artificial intelligence-based online credit method described in the embodiments of the present application; or implements an artificial intelligence-based online credit method. All or part of the functionality of the online credit facility. The at least one processor 32 may be composed of integrated circuits, for example, may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits packaged with the same function or different functions, including one or more central processing units. (Central Processing unit, CPU), microprocessor, digital processing chip, graphics processor and combination of various control chips, etc.

In some embodiments, the at least one communication bus 33 is configured to enable connection communication between the memory 31 and the at least one processor 32 and the like.

Although not shown, the computer device 3 may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 32 through a power management device, so as to be implemented by the power management device Manage charging, discharging, and power management functions. The power source may also include one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and any other components. The computer device 3 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

The above-mentioned integrated units implemented in the form of software functional modules may be stored in a computer-readable storage medium. The above-mentioned software function modules are stored in a storage medium, and include several instructions to make a computer device (which may be a personal computer, a computer device, or a network device, etc.) or a processor (processor) to execute the methods described in the various embodiments of the present application. part.

In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of the modules is only a logical function division, and there may be other division manners in actual implementation.

The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, and may be located in one place or distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional module in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware, or can be implemented in the form of hardware plus software function modules.

It will be apparent to those skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, but that the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Accordingly, the embodiments are to be regarded in all respects as illustrative and not restrictive, and the scope of the application is to be defined by the appended claims rather than the foregoing description, which is therefore intended to fall within the scope of the claims. All changes within the meaning and scope of the equivalents of , are included in this application. Any reference signs in the claims shall not be construed as limiting the involved claim. Furthermore, it is clear that the word "comprising" does not exclude other units or, and the singular does not exclude the plural. A plurality of units or devices stated in the device embodiments may also be implemented by one unit or device through software or hardware. The terms first, second, etc. are used to denote names and do not denote any particular order.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application rather than limitations. Although the present application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present application can be Modifications or equivalent substitutions can be made without departing from the spirit and scope of the technical solutions of the present application.

Claims (20)

1. An artificial intelligence-based online credit method, wherein the method includes:

   Obtain a double-recorded video of a credit requester reading preset text;

   Determine whether the reading is passed according to the preset text, the double-recorded video and the credit requester's certificate image;

   using a willingness recognition model to identify the type of willingness of the credit requester based on the double-recorded video when a pass reading is determined;

   When the willingness type is the target willingness type, receive the digital password input by the credit requester;

   generating a first public key according to the digital password, and generating a digital signature according to the first public key and the certificate image;

   A credit contract is generated based on the digital signature.
2. The online credit method based on artificial intelligence according to claim 1, wherein the determining whether to pass the reading according to the preset text, the double-recorded video and the certificate image of the credit requester comprises:

   Extracting audio in the double-recording video and extracting multiple frame images in the dual-recording video;

   Recognizing the audio to obtain voice text, and comparing the voice text and the preset text to obtain a first comparison result;

   Comparing the plurality of frame images and the certificate images to obtain a second comparison result;

   Whether the credit requester has passed the reading is determined according to the first comparison result and the second comparison result.
3. The artificial intelligence-based online credit method according to claim 2, wherein the extracting a plurality of frame images in the double-recorded video comprises:

   Calculate the double-recording time of the double-recording video;

   Generate the number of detections according to the double recording time;

   Calculate the detection frame rate according to the double recording time and the detection times;

   Extracting a plurality of frame images from the dual recording video using the detected frame rate.
4. The online credit method based on artificial intelligence as claimed in claim 3, wherein the generating the detection times according to the double recording time comprises:

   Define the upper and lower limits of the first detection times and the upper and lower limits of the second detection times;

   Calculate the average time value of the historical double recording time;

   comparing the double recording time with the average time value;

   When the double-recording time is greater than or equal to the average time value, a first random number is generated within the numerical range corresponding to the upper and lower limits of the first detection times, as the detection times of the double-recording video;

   When the double-recording time is less than the average time value, a second random number is generated within a numerical range corresponding to the upper and lower limits of the second detection times as the detection times of the double-recording video.
5. The online credit method based on artificial intelligence according to any one of claims 1 to 4, wherein the training process of the willingness recognition model comprises:

   Acquire multiple double-recorded videos corresponding to multiple will types, and identify the voice text in each double-recorded video corresponding to each will type;

   Carrying out line-by-line encoding processing to the speech text to obtain a plurality of encoding vectors;

   Input each coding vector into the forward long short-term memory network layer to obtain the first vector, and input each coding vector into the reverse long short-term memory network layer to obtain the second vector;

   The multiple first vectors and multiple second vectors corresponding to the voice text are spliced according to the row order of the voice text to obtain an input vector;

   generating a feature vector according to the willingness type and the input vector corresponding to the willingness type;

   A support vector machine is trained based on a plurality of the feature vectors to obtain a willingness type identification model.
6. The artificial intelligence-based online credit method according to claim 5, wherein the generating the first public key according to the digital password comprises:

   determining a character string corresponding to the digital password;

   compute the hash value of the string;

   Obtain the first system parameter and the second system parameter;

   Using the first system parameter, the second system parameter and the hash value to calculate the first verification parameter that complies with the El Gamal commitment, and determine the first verification parameter as the first public key.
7. The artificial intelligence-based online credit method according to claim 6, wherein, after the credit contract is generated according to the digital signature, the method further comprises:

   using the first public key as the contract code of the credit contract;

   After receiving the sign-off instruction from the signer, sending a digital password acquisition instruction to the client of the credit requester;

   Receive the digital password sent by the client and input by the credit requester according to the digital password acquisition instruction;

   generating a second public key according to the received digital password;

   verifying whether the second public key is the same as the contract code;

   When it is determined that the second public key is the same as the contract code, a preset lending operation is performed.
8. An artificial intelligence-based online credit device, wherein the device includes:

   An acquisition module for acquiring a double-recorded video of a credit requester reading a preset text;

   a judging module for judging whether the reading is passed according to the preset text, the double-recorded video and the credit requester's certificate image;

   an identification module for identifying the credit requester's willingness type based on the double-recorded video using a willingness recognition model when it is determined to pass the reading;

   a receiving module, configured to receive the digital password input by the credit requester when the willingness type is the target willingness type;

   a signature module, configured to generate a first public key according to the digital password, and generate a digital signature according to the first public key and the certificate image;

   A generating module for generating a credit contract according to the digital signature.
9. A computer device, wherein the computer device includes a processor for executing computer-readable instructions stored in a memory to implement the following steps:

   Obtain a double-recorded video of a credit requester reading preset text;

   Determine whether the reading is passed according to the preset text, the double-recorded video and the credit requester's certificate image;

   using a willingness recognition model to identify the type of willingness of the credit requester based on the double-recorded video when a pass reading is determined;

   When the willingness type is the target willingness type, receive the digital password input by the credit requester;

   generating a first public key according to the digital password, and generating a digital signature according to the first public key and the certificate image;

   A credit contract is generated based on the digital signature.
10. The computer device of claim 9, wherein the processor executes the computer-readable instructions to determine whether to read or not based on the preset text, the double-recorded video, and a credential image of the credit requester , including:

    Extracting audio in the double-recording video and extracting multiple frame images in the dual-recording video;

    Recognizing the audio to obtain voice text, and comparing the voice text and the preset text to obtain a first comparison result;

    Comparing the plurality of frame images and the certificate images to obtain a second comparison result;

    Whether the credit requester has passed the reading is determined according to the first comparison result and the second comparison result.
11. The computer device according to claim 10, wherein, when the processor executes the computer-readable instructions to extract a plurality of frame images in the dual-recording video, it specifically includes:

    Calculate the double-recording time of the double-recording video;

    Generate the number of detections according to the double recording time;

    Calculate the detection frame rate according to the double recording time and the detection times;

    Extracting a plurality of frame images from the dual recording video using the detected frame rate.
12. The computer device according to claim 11, wherein, when the processor executes the computer-readable instructions to generate the number of detections according to the double-recording time, it specifically includes:

    Define the upper and lower limits of the first detection times and the upper and lower limits of the second detection times;

    Calculate the average time value of the historical double recording time;

    comparing the double recording time with the average time value;

    When the double-recording time is greater than or equal to the average time value, a first random number is generated within the numerical range corresponding to the upper and lower limits of the first detection times, as the detection times of the double-recording video;

    When the double-recording time is less than the average time value, a second random number is generated within a numerical range corresponding to the upper and lower limits of the second detection times as the detection times of the double-recording video.
13. The computer device according to any one of claims 9 to 12, wherein the training process of the intention recognition model comprises:

    Acquire multiple double-recorded videos corresponding to multiple will types, and identify the voice text in each double-recorded video corresponding to each will type;

    Carrying out line-by-line encoding processing to the speech text to obtain a plurality of encoding vectors;

    Input each coding vector into the forward long short-term memory network layer to obtain the first vector, and input each coding vector into the reverse long short-term memory network layer to obtain the second vector;

    The multiple first vectors and multiple second vectors corresponding to the voice text are spliced according to the row order of the voice text to obtain an input vector;

    generating a feature vector according to the willingness type and the input vector corresponding to the willingness type;

    A support vector machine is trained based on a plurality of the feature vectors to obtain a willingness type identification model.
14. The computer device according to claim 13, wherein, when the processor executes the computer-readable instructions to generate the first public key according to the digital password, it specifically includes:

    determining a character string corresponding to the digital password;

    compute the hash value of the string;

    Obtain the first system parameter and the second system parameter;

    Using the first system parameter, the second system parameter and the hash value to calculate the first verification parameter that complies with the El Gamal commitment, and determine the first verification parameter as the first public key.
15. 15. The computer device of claim 14, wherein, after said generating a credit contract based on said digital signature, said processor executing said computer-readable instructions further implements the following steps:

    using the first public key as the contract code of the credit contract;

    After receiving the sign-off instruction from the signer, sending a digital password acquisition instruction to the client of the credit requester;

    Receive the digital password sent by the client and input by the credit requester according to the digital password acquisition instruction;

    generating a second public key according to the received digital password;

    verifying whether the second public key is the same as the contract code;

    When it is determined that the second public key is the same as the contract code, a preset lending operation is performed.
16. A computer-readable storage medium storing computer-readable instructions on the computer-readable storage medium, wherein the computer-readable instructions realize the following steps when executed by a processor:

    Obtain a double-recorded video of a credit requester reading preset text;

    Determine whether the reading is passed according to the preset text, the double-recorded video and the credit requester's certificate image;

    using a willingness recognition model to identify the type of willingness of the credit requester based on the double-recorded video when a pass reading is determined;

    When the willingness type is the target willingness type, receive the digital password input by the credit requester;

    generating a first public key according to the digital password, and generating a digital signature according to the first public key and the certificate image;

    A credit contract is generated based on the digital signature.
17. 17. The computer-readable storage medium of claim 16, wherein the computer-readable instructions are executed by the processor to implement a credential image based on the preset text, the double-recorded video, and the credit requester When judging whether to pass the reading, it specifically includes:

    Extracting audio in the double-recording video and extracting multiple frame images in the dual-recording video;

    Recognizing the audio to obtain voice text, and comparing the voice text and the preset text to obtain a first comparison result;

    Comparing the plurality of frame images and the certificate images to obtain a second comparison result;

    Whether the credit requester has passed the reading is determined according to the first comparison result and the second comparison result.
18. The computer-readable storage medium according to claim 17, wherein, when the computer-readable instructions are executed by the processor to extract multiple frame images in the dual-recording video, the instructions specifically include:

    Calculate the double-recording time of the double-recording video;

    Generate the number of detections according to the double recording time;

    Calculate the detection frame rate according to the double recording time and the detection times;

    Extracting a plurality of frame images from the dual recording video using the detected frame rate.
19. The computer-readable storage medium according to claim 18, wherein when the computer-readable instructions are executed by the processor to realize the generation of the number of detections according to the double-recording time, it specifically includes:

    Define the upper and lower limits of the first detection times and the upper and lower limits of the second detection times;

    Calculate the average time value of the historical double recording time;

    comparing the double recording time with the average time value;

    When the double-recording time is greater than or equal to the average time value, a first random number is generated within the numerical range corresponding to the upper and lower limits of the first detection times, as the detection times of the double-recording video;

    When the double-recording time is less than the average time value, a second random number is generated within a numerical range corresponding to the upper and lower limits of the second detection times as the detection times of the double-recording video.
20. The computer-readable storage medium according to any one of claims 16 to 19, wherein the training process of the intention recognition model comprises:

    Acquire multiple double-recorded videos corresponding to multiple will types, and identify the voice text in each double-recorded video corresponding to each will type;

    Carrying out line-by-line encoding processing to the speech text to obtain a plurality of encoding vectors;

    Input each coding vector into the forward long short-term memory network layer to obtain the first vector, and input each coding vector into the reverse long short-term memory network layer to obtain the second vector;

    The multiple first vectors and multiple second vectors corresponding to the voice text are spliced according to the row order of the voice text to obtain an input vector;

    generating a feature vector according to the willingness type and the input vector corresponding to the willingness type;

    A support vector machine is trained based on a plurality of the feature vectors to obtain a willingness type identification model.

Images