CN113240043B - Pseudo-identification method, device, equipment and storage medium based on multi-picture difference - Google Patents
Pseudo-identification method, device, equipment and storage medium based on multi-picture difference Download PDFInfo
- Publication number
- CN113240043B CN113240043B CN202110609695.3A CN202110609695A CN113240043B CN 113240043 B CN113240043 B CN 113240043B CN 202110609695 A CN202110609695 A CN 202110609695A CN 113240043 B CN113240043 B CN 113240043B
- Authority
- CN
- China
- Prior art keywords
- pseudo
- fake
- identifying
- point
- picture
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000012545 processing Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims description 111
- 230000003416 augmentation Effects 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 10
- 230000003190 augmentative effect Effects 0.000 claims description 9
- 230000004913 activation Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000002372 labelling Methods 0.000 abstract description 9
- 230000003321 amplification Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000003199 nucleic acid amplification method Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 3
- 241000246150 Cercis Species 0.000 description 2
- 235000006228 Cercis occidentalis Nutrition 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000003213 activating effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013526 transfer learning Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention relates to the field of image algorithms, and discloses a fake identifying method, a fake identifying device, fake identifying equipment and a storage medium based on multi-picture difference, wherein the fake identifying method comprises the following steps: acquiring a plurality of pseudo-image combinations of the pseudo-object, and acquiring a plurality of pseudo-image combinations according to the plurality of pseudo-image combinations; processing the pseudo-point image combination through the pseudo-point model to obtain a pseudo-point color state information combination; calculating color state variances of the pseudo-discrimination points according to the color state information combination of the pseudo-discrimination points, and calculating variance average values of all the color state variances; if the variance average value is larger than a preset false identification threshold value, the false identification object is judged to be true. According to the invention, the fake identification object pictures of the fake identification objects are obtained from different angles, so that the image samples adopted in the image identification and fake identification have differences, and the accuracy of the image identification and fake identification can be improved. And the error of manual labeling state information is avoided, and the accuracy of image identification and false discrimination is further improved.
Description
Technical Field
The present invention relates to the field of image algorithms, and in particular, to a method, apparatus, device, and storage medium for authenticating multiple pictures based on the difference between the pictures.
Background
Image recognition is an important part of the field of artificial intelligence. Currently, most vision applications require image recognition algorithms. The method comprises the steps of firstly identifying images acquired by a camera, and then entering other processing flows of a vision application system. It can be said that image recognition is the basis of the field of machine vision.
In the prior art, a sample collected based on artificial subjective labeling is generally trained by using a neural network to obtain a corresponding image recognition model. Because of the great difficulty of manual labeling, the sample labeling cost is high. And the difference of the labeling results of different people is large, so that the accuracy of an image recognition algorithm is low, and the authenticity of the image cannot be accurately identified.
Disclosure of Invention
Based on the above, it is necessary to provide a multiple-picture-difference-based fake identification method, device, computer equipment and storage medium to solve the problems that the manual labeling difficulty is high, the difference of different person labeling results is high, the accuracy of an image recognition algorithm is low, and the authenticity of an image cannot be accurately identified.
A false identification method based on multi-picture difference comprises the following steps:
acquiring a plurality of fake identification picture combinations of the fake identification objects; the number of the fake identifying picture combinations is more than three; the fake identifying picture combination comprises a plurality of fake identifying picture images separated by fake identifying object pictures, the difference value of shooting angles of any two fake identifying object pictures is larger than a preset angle threshold value, and one fake identifying picture combination corresponds to one fake identifying object picture;
obtaining a plurality of pseudo-point image combinations according to the plurality of pseudo-point image combinations, wherein the pseudo-point image combinations comprise a specified number of pseudo-point images, and the specified number is equal to the number of the pseudo-point image combinations; one of the pseudo-point image combinations corresponds to one pseudo-point, and the number of the pseudo-point images in the pseudo-point image combination is equal to the number of the pseudo-points;
processing the pseudo point image combination through a pseudo point identification model to obtain a pseudo point color state information combination, wherein the pseudo point color state information combination comprises a plurality of pseudo point color state information, and one pseudo point color state information corresponds to one pseudo point image;
calculating color state variances of the fake identifying points according to the fake identifying point color state information combination, and calculating variance average values of all the color state variances;
and if the variance average value is larger than a preset false identification threshold value, judging that the false identification object is true.
A multiple picture difference based authentication device comprising:
the fake identifying picture combining module is used for obtaining a plurality of fake identifying picture combinations of the fake identifying object; the number of the fake identifying picture combinations is more than three; the fake identifying picture combination comprises a plurality of fake identifying picture images separated by fake identifying object pictures, the difference value of shooting angles of any two fake identifying object pictures is larger than a preset angle threshold value, and one fake identifying picture combination corresponds to one fake identifying object picture;
the fake identifying image combination module is used for obtaining a plurality of fake identifying image combinations according to the plurality of fake identifying image combinations, wherein the fake identifying image combinations comprise a designated number of fake identifying point images, and the designated number is equal to the number of the fake identifying image combinations; one of the pseudo-point image combinations corresponds to one pseudo-point, and the number of the pseudo-point images in the pseudo-point image combination is equal to the number of the pseudo-points;
the fake identifying point color state information module is used for processing the fake identifying point image combination through a fake identifying model to obtain a fake identifying point color state information combination, wherein the fake identifying point color state information combination comprises a plurality of fake identifying point color state information, and one fake identifying point color state information corresponds to one fake identifying point image;
the variance mean module is used for calculating the color state variances of the fake identifying points according to the fake identifying point color state information combination and calculating the variance mean of all the color state variances;
and the judging module is used for judging that the fake identifying object is true if the variance average value is larger than a preset fake identifying threshold value.
A computer device comprising a memory, a processor and computer readable instructions stored in the memory and executable on the processor, the processor implementing the above-described multi-picture difference based authentication method when executing the computer readable instructions.
One or more readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform a multi-picture difference based authentication method as described above.
According to the fake identifying method, the fake identifying device, the computer equipment and the storage medium based on the multi-picture difference, the fake identifying image combinations are obtained according to the fake identifying image combinations by obtaining the fake identifying image combinations of the fake identifying object, and then the fake identifying image combinations are processed through the fake identifying model to obtain the fake identifying point color state information combinations; calculating color state variances of the fake identifying points according to the fake identifying point color state information combination, and calculating variance average values of all the color state variances; and if the variance average value is larger than a preset false identification threshold value, judging that the false identification object is true. According to the invention, the fake identification object pictures of the fake identification objects are obtained from different angles, so that the image samples adopted in the image identification and fake identification have differences, the image samples are more comprehensive, and the accuracy of the image identification and fake identification can be improved. And dividing the fake identifying object picture according to the fake identifying points to obtain a fake identifying picture combination, so as to obtain a fake identifying point image combination, and improve the fake identifying accuracy. The fake identifying image combination is processed through the fake identifying model, so that the fake identifying point color state information combination is obtained, errors of manual labeling state information are avoided, and the fake identifying accuracy of image identification is further improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of an application environment of a multi-picture difference based authentication method according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for authenticating based on multi-picture variability according to an embodiment of the present invention;
FIG. 3 is a diagram showing an exemplary method for authenticating based on multi-picture variability in accordance with one embodiment of the present invention;
FIG. 4 is a schematic diagram of a pseudo-authentication device based on multi-picture variability according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a computer device in accordance with an embodiment of the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The authentication method based on the multi-picture difference provided by the embodiment can be applied to an application environment as shown in fig. 1, wherein a client communicates with a server. Clients include, but are not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server may be implemented by a stand-alone server or a server cluster formed by a plurality of servers.
In an embodiment, as shown in fig. 2, a multi-picture difference-based authentication method is provided, and the method is applied to the server in fig. 1 for illustration, and includes the following steps:
s10, acquiring a plurality of fake identification picture combinations of a fake identification object; the number of the fake identifying picture combinations is more than three; the fake identifying picture combination comprises a plurality of fake identifying picture images separated by fake identifying object pictures, the difference value of shooting angles of any two fake identifying object pictures is larger than a preset angle threshold value, and one fake identifying picture combination corresponds to one fake identifying object picture.
An authentication object is understood to mean an object to be authenticated. For example, when the identity card is brushed to enter and exit the high-speed rail station, the authenticity of the identity card needs to be identified, and at the moment, the identity card is the fake identification object. The fake identifying object picture is a picture obtained by shooting the fake identifying object through the camera, and the difference value of shooting angles of any two fake identifying object pictures is larger than a preset angle threshold value, so that the shooting angles of all the fake identifying object pictures are ensured to be different. The preset angle threshold can be set according to requirements. For example, when the preset angle threshold is 25 degrees and the authenticity of the identity card is identified, the identity card is horizontally placed, and the camera shoots the identity card from angles of 36 degrees, 72 degrees, 108 degrees and 144 degrees with the horizontal plane respectively, so that four fake identification object pictures with different angles are obtained.
Specifically, the fake identifying points contained in the fake identifying object picture are identified, the fake identifying object picture is segmented according to the identified fake identifying points, and the fake identifying object picture can be segmented into one or more fake identifying point images. Each of the authentication point images contains an authentication point. The fake identifying picture combination includes several fake identifying point images corresponding to the fake identifying target picture. More than three fake identifying picture combinations of the fake identifying object are obtained, for example, the fake identifying object is an identity card, the camera shoots the identity card from four different angles to obtain four fake identifying object pictures, each fake identifying object picture corresponds to one fake identifying picture combination, and four fake identifying picture combinations can be obtained.
S20, obtaining a plurality of pseudo-point image combinations according to the pseudo-point image combinations, wherein the pseudo-point image combinations comprise a specified number of pseudo-point images, and the specified number is equal to the number of the pseudo-point image combinations; and one of the fake identifying point image combinations corresponds to one fake identifying point, and the number of the fake identifying point images in the fake identifying picture combination is equal to the number of the fake identifying points.
It can be understood that one fake identifying picture combination includes several fake identifying point images corresponding to one fake identifying object picture. Combining the pseudo-point images of the same pseudo-point in the pseudo-picture combinations to obtain a plurality of pseudo-point image combinations. The pseudo-point image combination includes a specified number of pseudo-point images, the specified number being equal to the number of pseudo-picture combinations.
Specifically, in an example, there are four pseudo-image combinations, namely, a combination, B combination, C combination and D combination, respectively, one pseudo-image combination includes e, f and g three pseudo-point images, and pseudo-point images of the same pseudo-point in the four pseudo-image combinations are combined to obtain three pseudo-point image combinations, namely (a) e ，B e ，C e ，D e )、(A f ，B f ，C f ，D f ) And (A) g ，B g ，C g ，D g ). Wherein one pseudo-point image combination contains four pseudo-point images, and the four pseudo-point images correspond to one pseudo-point. The three pseudo-point image combinations correspond to three pseudo-points, and the three pseudo-point images in the pseudo-picture combinations correspond to three pseudo-points.
S30, processing the pseudo-point image combination through a pseudo-point identification model to obtain a pseudo-point color state information combination, wherein the pseudo-point color state information combination comprises a plurality of pseudo-point color state information, and one pseudo-point color state information corresponds to one pseudo-point image.
It can be understood that the fake identifying model is obtained by obtaining a plurality of first fake identifying training samples with fake identifying objects as true and a plurality of second fake identifying training samples with fake identifying objects as false and training the fake identifying training samples. The authentication model is used for authenticating an object to be authenticated, for example, authenticating a certificate, a living body, or the like with an anti-counterfeit image. The pseudo dot discrimination color state information refers to a color state value of the pseudo dot discrimination image including a color of the pseudo dot discrimination image and a color depth (ink reflectance).
Specifically, the obtained pseudo-point image combination is input into a pseudo-model, the pseudo-model processes a plurality of pseudo-point images contained in the pseudo-point image combination, so that each pseudo-point image obtains color state information of a pseudo-point, and then color state information of a plurality of pseudo-points, namely the pseudo-point color state information combination is obtained.
S40, calculating color state variances of the fake identifying points according to the fake identifying point color state information combination, and calculating variance average values of all the color state variances.
It can be appreciated that the pseudo-point color state information combination includes a plurality of pseudo-point color state information, one pseudo-point corresponding to each pseudo-point color state information combination. The color state information of the plurality of discrimination points refers to color state values of all discrimination point images in the combination of the discrimination point images, wherein the color state values are colors and color depths (ink reflection) of the discrimination point images.
Specifically, the variance of the color state information of a plurality of pseudo-discrimination points in the color state information combination of the pseudo-discrimination points is calculated according to a variance formula, and the variance is the color state variance of the pseudo-discrimination points corresponding to the color state information combination of the pseudo-discrimination points. Each false identification point of the false identification object corresponds to a false identification point color state information combination, and all color state variances of the false identification objects are obtained. And calculating the variance mean value among all color state variances of the false identification object according to the mean value formula.
And S50, if the variance mean value is larger than a preset false identification threshold value, judging that the false identification object is true.
It can be appreciated that the preset false discrimination threshold can be set according to the image color change, shooting angle and other conditions of the false discrimination object.
Specifically, after the variance mean value among all color state variances of the fake identification object is obtained, the size of the variance mean value is judged. And when the variance mean value is larger than a preset false identification threshold value, judging that the false identification object is true. And when the variance mean value is smaller than or equal to a preset false identification threshold value, judging that the false identification object is false. For example, if the preset false discrimination threshold is 0.2, the false discrimination object with the variance average value greater than 0.2 is judged to be true, and the false discrimination object with the variance average value less than or equal to 0.2 is judged to be false.
In the steps S10-S50, a plurality of pseudo-point image combinations are obtained according to the pseudo-image combinations by obtaining the pseudo-image combinations of the pseudo-object, and then the pseudo-point image combinations are processed through a pseudo-model to obtain pseudo-point color state information combinations; calculating color state variances of the fake identifying points according to the fake identifying point color state information combination, and calculating variance average values of all the color state variances; and if the variance average value is larger than a preset false identification threshold value, judging that the false identification object is true. According to the invention, the fake identification object pictures of the fake identification objects are obtained from different angles, so that the image samples adopted in the image identification and fake identification have differences, the image samples are more comprehensive, and the accuracy of the image identification and fake identification can be improved. And dividing the fake identifying object picture according to the fake identifying points to obtain a fake identifying picture combination, so as to obtain a fake identifying point image combination, and improve the fake identifying accuracy. The fake identifying point image combination is processed through the fake identifying model, so that the fake identifying point color state information combination is obtained, errors of manual labeling state information are avoided, and the accuracy of image identification and fake identification is further improved.
Optionally, before step S30, that is, before the processing the pseudo-point image combination by the pseudo-point model, the method further includes:
s301, obtaining a plurality of pre-training samples with true false identification objects, wherein the pre-training samples comprise a plurality of false identification picture combinations;
s302, inputting the pre-training sample into a first preset network structure for training to obtain a pre-training model;
s303, obtaining a pre-training result output by the pre-training model;
s304, acquiring a plurality of first fake identification training samples with fake identification objects being true and a plurality of second fake identification training samples with fake identification objects being false;
s305, inputting the pre-training result, the first false identification training sample and the second false identification training sample into a second preset network structure for training, and obtaining a false identification model.
It is understood that the plurality of authentication objects may be a plurality of different identity cards. The fake identifying object is true, i.e. the fake identifying object is a true identity card, and the fake identifying object is false, i.e. the fake identifying object is a false identity card. Each fake identifying object corresponds to a plurality of fake identifying picture combinations. And acquiring pseudo-object images shot at different angles of a plurality of pseudo-objects, dividing each pseudo-object image of each pseudo-object into a plurality of pseudo-image images, and combining the plurality of pseudo-image images divided by each pseudo-object image as one pseudo-image. The first preset network structure may be a network structure of a Deep Neural Network (DNN) and a multi-layer neural network (MLP) or the like for class learning, or a combination thereof. The second preset network structure may be a deep residual network structure for transfer learning.
Specifically, a plurality of pseudo-discrimination image combinations of pseudo-discrimination objects with true pseudo-discrimination objects are used as pre-training samples to be input into a first preset network structure, so that color state characteristics of pseudo-discrimination point images in the pseudo-discrimination image combinations can be obtained, and the preset network structure carries out training and learning on the color state characteristics, so that a pre-training model can be obtained. The pre-training model learns the color state characteristics that identify the false object as true. The pre-training result output by the pre-training model can be obtained through the pre-training model. And combining the plurality of fake identifying pictures of the fake identifying objects as the fake identifying objects to serve as a first fake identifying training sample, and combining the plurality of fake identifying pictures of the fake identifying objects as the fake identifying objects to serve as a second fake identifying training sample. And inputting the first false identification training sample, the second false identification training sample and the pre-training result into a second preset network structure, so that the second preset network structure carries out training and learning on the false identification training sample and the pre-training result, and a false identification model is generated. The fake identifying model not only learns the information that the fake identifying object is true and the information that the fake identifying object is false, but also carries out migration learning on the pre-training result of the pre-training model, wherein the pre-training result is the result of the pre-training model learning the information that the fake identifying object is true. The false identification point color state information of the false identification object can be predicted through the false identification model, and the predicted false identification point color state information of the false identification object is output. The accuracy of image identification and false identification can be improved, the state information does not need to be marked manually, and the labor cost is saved.
Optionally, before step S10, that is, before the obtaining the several pseudo-image combinations of the pseudo-object, the method includes:
s101, acquiring a plurality of fake identification object pictures of the fake identification object;
s102, identifying the fake identifying points of the fake identifying object picture;
and S103, dividing the fake identifying object picture according to the fake identifying points to obtain a plurality of fake identifying point images.
It can be understood that the image of the fake identifying object refers to an image obtained by shooting the fake identifying object at multiple angles through a camera. A fake identifying object can obtain a plurality of fake identifying object pictures through multi-angle shooting. An authentication object contains one or more authentication points. For example, as shown in fig. 3, the counterfeit identifying object is a hong kong identity card, the complete counterfeit identifying points on the hong kong identity card are three cercis patterns on the right of the figure, wherein each cercis pattern is a counterfeit identifying point.
Specifically, shooting the fake identifying object through the camera at different angles to obtain a plurality of fake identifying object pictures, identifying fake identifying points contained in the fake identifying object pictures, and dividing the fake identifying object pictures according to the identified fake identifying points, so that the fake identifying object pictures can be divided into one or more fake identifying point images. Each of the authentication point images contains an authentication point.
Optionally, in step S30, that is, the processing the pseudo-point image combination by the pseudo-point image processing module, obtaining a pseudo-point color state information combination includes:
s306, inputting a plurality of pseudo-point images contained in the pseudo-point image combination into the pseudo-point model;
s307, acquiring a multi-dimensional feature map of the pseudo point image;
s308, converting the multi-dimensional feature map into a one-dimensional feature value through a full connection layer;
s309, processing the one-dimensional characteristic value through an activation function to obtain the color state information of the pseudo-point; the pseudo-point identifying color state information combination comprises a plurality of pseudo-point identifying color state information.
It will be appreciated that one combination of pseudo-point images comprises several pseudo-point images. And inputting a plurality of pseudo-point images contained in the pseudo-point image combination into a pseudo-model, and obtaining a multi-dimensional characteristic diagram of the pseudo-point image through the pseudo-model. For example, the number of color channels of the input pseudo-point image is 3 (RGB), and a multi-dimensional feature map with 256 color channels of the pseudo-point image can be obtained through the pseudo-point model. Further, the multi-dimensional feature map is converted into one-dimensional feature values through a full connection layer of the fake identification model. And inputs the one-dimensional feature value into an activation function. Wherein the activation function is used for hidden layer neuron output, the value range is (0, 1), and the activation function can map a real number to the interval of (0, 1). And processing the one-dimensional characteristic value by activating the function, and calculating to obtain a false identification point color state value corresponding to the one-dimensional characteristic value, namely false identification point color state information.
Optionally, after step S301, that is, after the obtaining of the pre-training samples in which the plurality of pseudo-objects are true, the method includes:
s3011, carrying out amplification treatment on the plurality of pseudo-discrimination picture combinations contained in the pre-training sample to obtain an amplified sample;
s3012, inputting the pre-training sample and the augmentation sample into a first preset network structure for training to obtain a pre-training model.
It is understood that the augmentation process refers to zooming, cropping, rotating, etc. the picture.
Specifically, a plurality of pseudo-image combinations are randomly obtained from a pre-training sample, a plurality of pseudo-image images contained in the pseudo-image combinations are randomly cut, partial pseudo-image images containing a certain part of pseudo-image points are cut, and the obtained partial pseudo-image images are used as an augmentation sample. And inputting the augmentation sample and the pre-training sample into a first preset network structure for training and learning to obtain a pre-training model. The pre-training model not only learns the fake identifying point image, but also learns the detailed information of the fake identifying point image. Through the study of the augmentation sample, the learning ability of the pre-training model is improved. Even if the pseudo-discrimination point image is scaled, cut, rotated and the like, the pseudo-discrimination model can identify and discriminate the pseudo-discrimination point image, so that the accuracy of image identification and pseudo-discrimination is improved.
Optionally, in step S3011, the performing augmentation processing on the plurality of pseudo-discrimination image combinations included in the pre-training sample to obtain an augmented sample includes:
s3013, obtaining a plurality of pseudo-discrimination picture combinations from the pre-training sample;
s3014, performing the amplification processing of cutting, rotating and zooming on the pseudo-identification picture combination to obtain a plurality of amplified pictures of the pseudo-identification picture combination;
s3015, taking a plurality of amplified pictures combined by the pseudo-identification pictures as amplified samples.
Specifically, a plurality of pseudo-image combinations are randomly obtained from a pre-training sample, and the pseudo-point images contained in the pseudo-image combinations are randomly cut, rotated and scaled to obtain partial pseudo-point images containing a certain part of pseudo-points, pseudo-point scaling images and pseudo-point rotation images. The partial pseudo point image, the pseudo point scaling image and the pseudo point rotation image obtained through the augmentation treatment are multiple augmented pictures corresponding to the pseudo picture combination. And acquiring a plurality of amplified pictures combined by the plurality of pseudo-pictures, and taking the plurality of amplified pictures combined by the plurality of pseudo-pictures as an amplified sample.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.
In an embodiment, a multi-picture-difference-based authentication device is provided, where the multi-picture-difference-based authentication device corresponds to the multi-picture-difference-based authentication method in the above embodiment one by one. As shown in fig. 4, the multi-picture difference based false authentication device includes a false authentication picture combining module 10, a false authentication point image combining module 20, a false authentication point color status information module 30, a variance average module 40 and a judging module 50. The functional modules are described in detail as follows:
the fake identifying picture combining module 10 is used for acquiring a plurality of fake identifying picture combinations of the fake identifying object; the number of the fake identifying picture combinations is more than three; the fake identifying picture combination comprises a plurality of fake identifying picture images separated by fake identifying object pictures, the difference value of shooting angles of any two fake identifying object pictures is larger than a preset angle threshold value, and one fake identifying picture combination corresponds to one fake identifying object picture;
the pseudo-point image combination module 20 is configured to obtain a plurality of pseudo-point image combinations according to the plurality of pseudo-point image combinations, where the pseudo-point image combinations include a specified number of pseudo-point images, and the specified number is equal to the number of pseudo-point image combinations; one of the pseudo-point image combinations corresponds to one pseudo-point, and the number of the pseudo-point images in the pseudo-point image combination is equal to the number of the pseudo-points;
the pseudo-point identifying color state information module 30 is configured to process the pseudo-point identifying image combination through a pseudo-point identifying model to obtain a pseudo-point identifying color state information combination, where the pseudo-point identifying color state information combination includes a plurality of pseudo-point identifying color state information, and one pseudo-point identifying color state information corresponds to one pseudo-point identifying image;
the variance average module 40 is configured to calculate color state variances of the pseudo-discrimination points according to the color state information combination of the pseudo-discrimination points, and calculate variance averages of all the color state variances;
and the judging module 50 is configured to judge that the fake identifying object is true if the variance average value is greater than a preset fake identifying threshold value.
Optionally, before the pseudo-point image combining module 20, it includes:
the pre-training sample module is used for obtaining a plurality of pre-training samples with true false identification objects, wherein the pre-training samples comprise a plurality of false identification picture combinations;
the pre-training model module is used for inputting the pre-training sample into a first preset network structure for training to obtain a pre-training model;
the pre-training result module is used for acquiring a pre-training result output by the pre-training model;
the fake identifying training sample module is used for acquiring a plurality of first fake identifying training samples with fake identifying objects being true and a plurality of second fake identifying training samples with fake identifying objects being false;
and the fake identifying model module is used for inputting the pre-training result, the first fake identifying training sample and the second fake identifying training sample into a second preset network structure for training to obtain a fake identifying model.
Optionally, before the fake identifying picture combining module 10, the method includes:
the fake identification object picture unit is used for acquiring a plurality of fake identification object pictures of the fake identification object;
the complete fake identifying unit is used for identifying the fake identifying point of the fake identifying object picture;
and the fake identifying point image unit is used for dividing the fake identifying object picture according to the fake identifying points to obtain a plurality of fake identifying point images.
Optionally, the fake identifying picture combining module 10 includes:
an input unit for inputting a plurality of pseudo-point images included in the pseudo-point image combination into the pseudo-point model;
the multi-dimensional feature map unit is used for acquiring the multi-dimensional feature map of the pseudo point identification image;
the one-dimensional characteristic value unit is used for converting the multi-dimensional characteristic map into a one-dimensional characteristic value through a full connection layer;
the activation function unit is used for processing the one-dimensional characteristic value through an activation function to obtain the color state information of the fake identifying point; the pseudo-point identifying color state information combination comprises a plurality of pseudo-point identifying color state information.
Optionally, after the pre-training sample module, the method includes:
the amplification sample module is used for carrying out amplification treatment on the plurality of pseudo-discrimination picture combinations contained in the pre-training sample to obtain an amplification sample;
and the augmented sample training module is used for inputting the pre-training sample and the augmented sample into a first preset network structure for training to obtain a pre-training model.
Optionally, the augmentation sample module includes:
the fake identifying picture combining unit is used for acquiring a plurality of fake identifying picture combinations from the pre-training sample;
an augmented picture unit, configured to perform an augmentation process of cutting, rotating, and zooming on the pseudo-identification picture combination, to obtain a plurality of augmented pictures of the pseudo-identification picture combination;
and the augmentation sample unit is used for taking a plurality of augmentation pictures combined by the fake identification pictures as augmentation samples.
For specific limitations of the authentication device based on the multi-picture variability, reference may be made to the above limitation of the authentication method based on the multi-picture variability, and the description thereof will not be repeated here. The modules in the above-mentioned authentication device based on multiple picture differences may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a terminal, and the internal structure of which may be as shown in fig. 5. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a readable storage medium, an internal memory. The non-volatile storage medium stores an operating system and computer readable instructions. The internal memory provides an environment for the execution of an operating system and computer-readable instructions in a readable storage medium. The network interface of the computer device is for communicating with an external server via a network connection. The computer readable instructions, when executed by a processor, implement a multi-picture difference based authentication method. The readable storage medium provided by the present embodiment includes a nonvolatile readable storage medium and a volatile readable storage medium.
In one embodiment, a computer device is provided that includes a memory, a processor, and computer readable instructions stored on the memory and executable on the processor, when executing the computer readable instructions, performing the steps of:
acquiring a plurality of fake identification picture combinations of the fake identification objects; the number of the fake identifying picture combinations is more than three; the fake identifying picture combination comprises a plurality of fake identifying picture images separated by fake identifying object pictures, the difference value of shooting angles of any two fake identifying object pictures is larger than a preset angle threshold value, and one fake identifying picture combination corresponds to one fake identifying object picture;
obtaining a plurality of pseudo-point image combinations according to the plurality of pseudo-point image combinations, wherein the pseudo-point image combinations comprise a specified number of pseudo-point images, and the specified number is equal to the number of the pseudo-point image combinations; one of the pseudo-point image combinations corresponds to one pseudo-point, and the number of the pseudo-point images in the pseudo-point image combination is equal to the number of the pseudo-points;
processing the pseudo point image combination through a pseudo point identification model to obtain a pseudo point color state information combination, wherein the pseudo point color state information combination comprises a plurality of pseudo point color state information, and one pseudo point color state information corresponds to one pseudo point image;
calculating color state variances of the fake identifying points according to the fake identifying point color state information combination, and calculating variance average values of all the color state variances;
and if the variance average value is larger than a preset false identification threshold value, judging that the false identification object is true.
In one embodiment, one or more computer-readable storage media are provided having computer-readable instructions stored thereon, the readable storage media provided by the present embodiment including non-volatile readable storage media and volatile readable storage media. The readable storage medium has stored thereon computer readable instructions which when executed by one or more processors perform the steps of:
acquiring a plurality of fake identification picture combinations of the fake identification objects; the number of the fake identifying picture combinations is more than three; the fake identifying picture combination comprises a plurality of fake identifying picture images separated by fake identifying object pictures, the difference value of shooting angles of any two fake identifying object pictures is larger than a preset angle threshold value, and one fake identifying picture combination corresponds to one fake identifying object picture;
obtaining a plurality of pseudo-point image combinations according to the plurality of pseudo-point image combinations, wherein the pseudo-point image combinations comprise a specified number of pseudo-point images, and the specified number is equal to the number of the pseudo-point image combinations; one of the pseudo-point image combinations corresponds to one pseudo-point, and the number of the pseudo-point images in the pseudo-point image combination is equal to the number of the pseudo-points;
processing the pseudo point image combination through a pseudo point identification model to obtain a pseudo point color state information combination, wherein the pseudo point color state information combination comprises a plurality of pseudo point color state information, and one pseudo point color state information corresponds to one pseudo point image;
calculating color state variances of the fake identifying points according to the fake identifying point color state information combination, and calculating variance average values of all the color state variances;
and if the variance average value is larger than a preset false identification threshold value, judging that the false identification object is true.
Those skilled in the art will appreciate that implementing all or part of the above described embodiment methods may be accomplished by instructing the associated hardware by computer readable instructions stored on a non-volatile readable storage medium or a volatile readable storage medium, which when executed may comprise the above described embodiment methods. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.
Claims (9)
1. The fake identifying method based on the multi-picture difference is characterized by comprising the following steps:
acquiring a plurality of fake identification picture combinations of the fake identification objects; the number of the fake identifying picture combinations is more than three; the fake identifying picture combination comprises a plurality of fake identifying picture images separated by fake identifying object pictures, the difference value of shooting angles of any two fake identifying object pictures is larger than a preset angle threshold value, and one fake identifying picture combination corresponds to one fake identifying object picture;
obtaining a plurality of pseudo-point image combinations according to the plurality of pseudo-point image combinations, wherein the pseudo-point image combinations comprise a specified number of pseudo-point images, and the specified number is equal to the number of the pseudo-point image combinations; one of the pseudo-point image combinations corresponds to one pseudo-point, and the number of the pseudo-point images in the pseudo-point image combination is equal to the number of the pseudo-points;
processing the pseudo point image combination through a pseudo point identification model to obtain a pseudo point color state information combination, wherein the pseudo point color state information combination comprises a plurality of pseudo point color state information, and one pseudo point color state information corresponds to one pseudo point image;
calculating color state variances of the fake identifying points according to the fake identifying point color state information combination, and calculating variance average values of all the color state variances;
if the variance mean value is larger than a preset false identification threshold value, judging that the false identification object is true;
before the obtaining of the combination of the plurality of pseudo-discrimination pictures of the pseudo-discrimination object, the method comprises the following steps:
acquiring a plurality of fake identification object pictures of the fake identification object;
identifying the fake identifying point of the fake identifying object picture;
and dividing the fake identifying object picture according to the fake identifying points to obtain a plurality of fake identifying point images.
2. The multi-picture variability based authentication method according to claim 1, wherein said processing said authentication point image combinations by an authentication model further comprises:
obtaining a plurality of pre-training samples with true false identification objects, wherein the pre-training samples comprise a plurality of false identification picture combinations;
inputting the pre-training sample into a first preset network structure for training to obtain a pre-training model;
obtaining a pre-training result output by the pre-training model;
acquiring a plurality of first false identification training samples with false identification objects as true and a plurality of second false identification training samples with false identification objects as false;
and inputting the pre-training result, the first false identification training sample and the second false identification training sample into a second preset network structure for training to obtain a false identification model.
3. The multi-picture difference based authentication method according to claim 1, wherein the processing the authentication point image combination by an authentication model to obtain an authentication point color state information combination comprises:
inputting a plurality of pseudo-point images contained in the pseudo-point image combination into the pseudo-point model;
acquiring a multidimensional feature map of the pseudo point image;
converting the multi-dimensional feature map into a one-dimensional feature value through a full connection layer;
processing the one-dimensional characteristic value through an activation function to obtain the color state information of the fake identifying point; the pseudo-point identifying color state information combination comprises a plurality of pseudo-point identifying color state information.
4. The multi-picture difference based authentication method as claimed in claim 2, wherein after the obtaining the pre-training samples in which the plurality of authentication objects are true, the method comprises:
performing augmentation treatment on the plurality of pseudo-discrimination picture combinations contained in the pre-training sample to obtain an augmented sample;
and inputting the pre-training sample and the augmented sample into a first preset network structure for training to obtain a pre-training model.
5. The multi-picture difference based authentication method as claimed in claim 4, wherein said performing an augmentation process on the plurality of authentication picture combinations included in the pre-training samples to obtain augmented samples comprises:
obtaining a plurality of pseudo-discrimination picture combinations from the pre-training sample;
cutting, rotating and zooming the fake-identifying picture combination to obtain a plurality of amplified pictures of the fake-identifying picture combination;
and taking a plurality of amplified pictures combined by the pseudo-identification pictures as an amplified sample.
6. A multiple picture difference based authentication device, comprising:
the fake identifying picture combining module is used for obtaining a plurality of fake identifying picture combinations of the fake identifying object; the number of the fake identifying picture combinations is more than three; the fake identifying picture combination comprises a plurality of fake identifying picture images separated by fake identifying object pictures, the difference value of shooting angles of any two fake identifying object pictures is larger than a preset angle threshold value, and one fake identifying picture combination corresponds to one fake identifying object picture;
the fake identifying image combination module is used for obtaining a plurality of fake identifying image combinations according to the plurality of fake identifying image combinations, wherein the fake identifying image combinations comprise a designated number of fake identifying point images, and the designated number is equal to the number of the fake identifying image combinations; one of the pseudo-point image combinations corresponds to one pseudo-point, and the number of the pseudo-point images in the pseudo-point image combination is equal to the number of the pseudo-points;
the fake identifying point color state information module is used for processing the fake identifying point image combination through a fake identifying model to obtain a fake identifying point color state information combination, wherein the fake identifying point color state information combination comprises a plurality of fake identifying point color state information, and one fake identifying point color state information corresponds to one fake identifying point image;
the variance mean module is used for calculating the color state variances of the fake identifying points according to the fake identifying point color state information combination and calculating the variance mean of all the color state variances;
and the judging module is used for judging that the fake identifying object is true if the variance average value is larger than a preset fake identifying threshold value.
7. The multi-picture variability based authentication apparatus of claim 6, wherein prior to said processing said authentication point image combinations by an authentication model, comprising:
the pre-training sample module is used for obtaining a plurality of pre-training samples with true false identification objects, wherein the pre-training samples comprise a plurality of false identification picture combinations;
the pre-training model module is used for inputting the pre-training sample into a first preset network structure for training to obtain a pre-training model;
the pre-training result module is used for acquiring a pre-training result output by the pre-training model;
the fake identifying training sample module is used for acquiring a plurality of first fake identifying training samples with fake identifying objects being true and a plurality of second fake identifying training samples with fake identifying objects being false;
the fake identifying model module is used for inputting the pre-training result, the first fake identifying training sample and the second fake identifying training sample into a second preset network structure for training to obtain a fake identifying model;
the apparatus further comprises:
the fake identification object picture unit is used for acquiring a plurality of fake identification object pictures of the fake identification object;
the complete fake identifying unit is used for identifying the fake identifying point of the fake identifying object picture;
and the fake identifying point image unit is used for dividing the fake identifying object picture according to the fake identifying points to obtain a plurality of fake identifying point images.
8. A computer device comprising a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, wherein the processor, when executing the computer readable instructions, implements a multi-picture difference based authentication method as claimed in any one of claims 1 to 5.
9. One or more readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the multi-picture difference based authentication method of any one of claims 1 to 5.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110609695.3A CN113240043B (en) | 2021-06-01 | 2021-06-01 | Pseudo-identification method, device, equipment and storage medium based on multi-picture difference |
| PCT/CN2022/072288 WO2022252641A1 (en) | 2021-06-01 | 2022-01-17 | Multi-picture difference-based forgery identification method and apparatus, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110609695.3A CN113240043B (en) | 2021-06-01 | 2021-06-01 | Pseudo-identification method, device, equipment and storage medium based on multi-picture difference |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113240043A CN113240043A (en) | 2021-08-10 |
| CN113240043B true CN113240043B (en) | 2024-04-09 |
Family
ID=77136306
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110609695.3A Active CN113240043B (en) | 2021-06-01 | 2021-06-01 | Pseudo-identification method, device, equipment and storage medium based on multi-picture difference |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113240043B (en) |
| WO (1) | WO2022252641A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113240043B (en) * | 2021-06-01 | 2024-04-09 | 平安科技（深圳）有限公司 | Pseudo-identification method, device, equipment and storage medium based on multi-picture difference |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110516739A (en) * | 2019-08-27 | 2019-11-29 | 阿里巴巴集团控股有限公司 | A kind of certificate recognition methods, device and equipment |
| CN110570209A (en) * | 2019-07-30 | 2019-12-13 | 平安科技（深圳）有限公司 | Certificate authenticity verification method and device, computer equipment and storage medium |
| CN111898538A (en) * | 2020-07-30 | 2020-11-06 | 平安科技（深圳）有限公司 | Certificate authentication method and device, electronic equipment and storage medium |
| CN112017352A (en) * | 2020-09-03 | 2020-12-01 | 平安科技（深圳）有限公司 | Certificate authentication method, device, equipment and readable storage medium |
| CN112200136A (en) * | 2020-10-29 | 2021-01-08 | 腾讯科技（深圳）有限公司 | Certificate authenticity identification method and device, computer readable medium and electronic equipment |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9232112B2 (en) * | 2013-07-18 | 2016-01-05 | Electronics For Imaging, Inc. | Method and apparatus for effecting color separation, color variation, color characterization, and gamut mapping in color printing |
| CN113240043B (en) * | 2021-06-01 | 2024-04-09 | 平安科技（深圳）有限公司 | Pseudo-identification method, device, equipment and storage medium based on multi-picture difference |
-
2021
- 2021-06-01 CN CN202110609695.3A patent/CN113240043B/en active Active
-
2022
- 2022-01-17 WO PCT/CN2022/072288 patent/WO2022252641A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110570209A (en) * | 2019-07-30 | 2019-12-13 | 平安科技（深圳）有限公司 | Certificate authenticity verification method and device, computer equipment and storage medium |
| WO2021017610A1 (en) * | 2019-07-30 | 2021-02-04 | 平安科技（深圳）有限公司 | Certificate authenticity verification method and apparatus, computer device and storage medium |
| CN110516739A (en) * | 2019-08-27 | 2019-11-29 | 阿里巴巴集团控股有限公司 | A kind of certificate recognition methods, device and equipment |
| CN111898538A (en) * | 2020-07-30 | 2020-11-06 | 平安科技（深圳）有限公司 | Certificate authentication method and device, electronic equipment and storage medium |
| CN112017352A (en) * | 2020-09-03 | 2020-12-01 | 平安科技（深圳）有限公司 | Certificate authentication method, device, equipment and readable storage medium |
| CN112200136A (en) * | 2020-10-29 | 2021-01-08 | 腾讯科技（深圳）有限公司 | Certificate authenticity identification method and device, computer readable medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113240043A (en) | 2021-08-10 |
| WO2022252641A1 (en) | 2022-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10262190B2 (en) | Method, system, and computer program product for recognizing face | |
| US20230021661A1 (en) | Forgery detection of face image | |
| CN111967033B (en) | Picture encryption method and device based on face recognition, terminal and storage medium | |
| CN111091075B (en) | Face recognition method and device, electronic equipment and storage medium | |
| WO2022033220A1 (en) | Face liveness detection method, system and apparatus, computer device, and storage medium | |
| CN109063628B (en) | Face recognition method, device, computer equipment and storage medium | |
| CN111783629B (en) | Human face in-vivo detection method and device for resisting sample attack | |
| EP3642756B1 (en) | Detecting artificial facial images using facial landmarks | |
| CN110427972B (en) | Certificate video feature extraction method and device, computer equipment and storage medium | |
| CN111222380B (en) | Living body detection method and device and recognition model training method thereof | |
| CN112200136A (en) | Certificate authenticity identification method and device, computer readable medium and electronic equipment | |
| WO2020211396A1 (en) | Silent living body image recognition method and apparatus, computer device and storage medium | |
| CN113111880B (en) | Certificate image correction method, device, electronic equipment and storage medium | |
| CN113240043B (en) | Pseudo-identification method, device, equipment and storage medium based on multi-picture difference | |
| CN113128341A (en) | Dog face identification method based on convolutional neural network | |
| CN116229528A (en) | Living body palm vein detection method, device, equipment and storage medium | |
| CN112232336A (en) | Certificate identification method, device, equipment and storage medium | |
| CN113283388B (en) | Training method, device, equipment and storage medium of living body face detection model | |
| Bharadi et al. | Multi-instance iris recognition | |
| Liu et al. | Face liveness detection based on enhanced local binary patterns | |
| CN108460811B (en) | Face image processing method and device and computer equipment | |
| CN112686847B (en) | Identification card image shooting quality evaluation method and device, computer equipment and medium | |
| CN108875467B (en) | Living body detection method, living body detection device and computer storage medium | |
| CN114612989A (en) | Method and device for generating face recognition data set, electronic equipment and storage medium | |
| WO2021073150A1 (en) | Data detection method and apparatus, and computer device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40053199 Country of ref document: HK |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |