Counterfeit detection model training method, counterfeit detection method and related equipment

By sample enhancement of the initial icon sample set and training the counterfeit detection model, the problem of low accuracy of counterfeit detection caused by insufficient icon samples in the prior art is solved, and the accuracy and generalization ability of the counterfeit detection model are improved.

CN119006948APending Publication Date: 2024-11-22BEIJING HONGTENG INTELLIGENT TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202310582870.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-22
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The existing counterfeit detection model is difficult to effectively identify counterfeit installation packages due to insufficient icon samples and low accuracy.

Method used

By sample enhancement of the initial icon sample set, the target icon sample set is generated, and the counterfeit detection model is trained using these samples, including reconstruction of the pre-trained model and parameter adjustment until the training end condition is reached.

Benefits of technology

It improves the accuracy of counterfeit detection models, solves the problem of insufficient icon samples, and enhances the generalization and recognition capabilities of the model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119006948A_ABST
    Figure CN119006948A_ABST
Patent Text Reader

Abstract

The invention discloses a counterfeit detection model training method, a counterfeit detection method and related equipment. According to the method, sample enhancement is performed on initial icon samples in an initial icon sample set to obtain a plurality of enhanced icon samples, so that the number of samples of each category in the icon sample set is enriched, and a counterfeit detection model can have enough icon samples for training; the problem that in the prior art, few icon samples are not enough to train the counterfeit detection model is solved, and the accuracy of the counterfeit detection model is improved by training the counterfeit detection model through the target icon sample.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computers, and in particular to a counterfeit detection model training method, a counterfeit detection method and related equipment. Background Art

[0002] As the Internet becomes more and more powerful, there are more and more cyber crimes using mobile phones as carriers, such as fraud and information extraction through counterfeiting installation packages of various applications.

[0003] Currently, due to the limitations of application scenarios, very few icon samples can be collected for the identification of counterfeit installation packages, which are insufficient for training the corresponding counterfeit detection model, resulting in low accuracy of the existing counterfeit detection model. Summary of the invention

[0004] The embodiments of the present application provide a counterfeit detection model training method, a counterfeit detection method and related devices, which can improve the accuracy of the counterfeit detection model.

[0005] In a first aspect, an embodiment of the present application provides a method for training a counterfeit detection model, which obtains an initial icon sample set, wherein each initial icon sample in the initial icon sample set is marked with a sample label, and the sample label includes a counterfeit object; the initial icon sample set is subjected to sample enhancement to obtain a target icon sample set, wherein the target icon sample set includes target icon samples, wherein the target icon samples include the initial icon samples and enhanced icon samples obtained by performing data enhancement processing on the initial icon samples, and the sample label of each enhanced icon sample is consistent with the sample label of the corresponding initial icon sample; the target icon samples are input into a counterfeit detection model one by one to obtain a detection result, and the detection result includes a result of the counterfeit object; according to the sample label and the detection result, the parameters of the counterfeit detection model are adjusted until the training end condition is met to obtain a trained counterfeit detection model.

[0006] In a possible implementation, before inputting the target icon sample into a counterfeit detection model to obtain a detection result including a result of a counterfeit object, the counterfeit detection model training method further includes: training a pre-trained model according to an image sample set to obtain a trained pre-trained model; and reconstructing the trained pre-trained model to obtain the counterfeit detection model.

[0007] In a possible implementation, the training of the pre-trained model according to the image sample set to obtain the trained pre-trained model specifically includes: obtaining the image sample set, each image sample in the image sample set is marked with image content; inputting the image samples one by one into the pre-trained model to obtain the image content; adjusting the parameters of the pre-trained model according to the marked image content and the image content output by the pre-trained model until the conditions for the end of training are met to obtain the trained pre-trained model.

[0008] In a possible implementation, adjusting the parameters of the counterfeit detection model according to the sample labels and the detection results until a training end condition is met to obtain a trained counterfeit detection model specifically includes: determining a loss function according to the sample labels and the detection results; adjusting the parameters of the counterfeit detection model according to the loss function; and stopping training when the loss function reaches the training end condition to obtain the trained counterfeit detection model.

[0009] In a possible implementation, the parameter adjustment includes gradient descent, back propagation and learning rate adjustment, and the parameter adjustment of the counterfeit detection model according to the loss function specifically includes: performing gradient descent and back propagation on the counterfeit detection model according to the loss function; and adjusting the learning rate of the counterfeit detection model according to the number of iterations.

[0010] In a second aspect, an embodiment of the present application provides a counterfeit detection method, the counterfeit detection method comprising:

[0011] Acquire installation package information from the installation package to be detected, the installation package information includes the icon to be detected; input the icon to be detected into a counterfeit detection model to obtain a detection result, the detection result includes a result of the counterfeit object, and the counterfeit detection model is the counterfeit detection model described above; determine whether the installation package to be detected is a counterfeit program installation package based on the detection result.

[0012] In a third aspect, an embodiment of the present application provides a counterfeit detection model training device, comprising: a sample acquisition module, used to acquire an initial icon sample set, each initial icon sample in the initial icon sample set is marked with a sample label, and the sample label includes a counterfeit object; a sample enhancement module, used to perform sample enhancement on the initial icon sample set to obtain a target icon sample set, the target icon sample set includes target icon samples, the target icon samples include the initial icon samples and enhanced icon samples obtained by performing data enhancement processing on the initial icon samples, and the sample label of each enhanced icon sample is consistent with the sample label of the corresponding initial icon sample; a sample input module, used to input the target icon samples one by one into a counterfeit detection model to obtain a detection result, and the detection result includes a result of the counterfeit object; a parameter adjustment module, used to adjust the parameters of the counterfeit detection model according to the sample label and the detection result until the training end condition is met to obtain a trained counterfeit detection model.

[0013] In a possible implementation, the counterfeit detection model training device also includes: a pre-training module, used to train the pre-trained model according to the image sample set to obtain a trained pre-trained model; and a model reconstruction module, used to reconstruct the trained pre-trained model to obtain the counterfeit detection model.

[0014] In a possible implementation, the pre-training module specifically includes: an acquisition submodule, used to acquire the image sample set, each image sample in the image sample set is marked with image content; an input submodule, used to input the image samples one by one into the pre-training model to obtain the image content; an adjustment submodule, used to adjust the parameters of the pre-training model according to the marked image content and the image content output by the pre-training model until the conditions for the end of training are met, thereby obtaining the trained pre-training model.

[0015] In a possible implementation, the parameter adjustment module specifically includes: a loss determination submodule, used to determine the loss function according to the sample label and the detection result; a parameter adjustment submodule, used to adjust the parameters of the counterfeit detection model according to the loss function; and a stop training submodule, used to stop training when the loss function reaches the training end condition to obtain the trained counterfeit detection model.

[0016] In a possible implementation, the parameter adjustment includes gradient descent, back propagation and learning rate adjustment, and the parameter adjustment submodule specifically includes: a first adjustment unit, used to perform gradient descent and back propagation on the counterfeit detection model according to the loss function; a second adjustment unit, used to adjust the learning rate of the counterfeit detection model according to the number of iterations.

[0017] In a fourth aspect, an embodiment of the present application provides a counterfeit detection device, which includes: an icon acquisition module, used to obtain installation package information from the installation package to be detected, and the installation package information includes the icon to be detected; an icon detection module, used to input the icon to be detected into a counterfeit detection model to obtain a detection result, the detection result includes the result of the counterfeit object, and the counterfeit detection model is the counterfeit detection model as described above; a result determination module, used to determine whether the installation package to be detected is a counterfeit program installation package based on the detection result.

[0018] In a fifth aspect, an embodiment of the present application provides an electronic device, comprising: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded and executed by the processor: obtaining an initial icon sample set, each initial icon sample in the initial icon sample set is marked with a sample label, and the sample label includes a counterfeit object; performing sample enhancement on the initial icon sample set to obtain a target icon sample set, the target icon sample set includes target icon samples, the target icon samples include the initial icon samples and enhanced icon samples obtained by performing data enhancement processing on the initial icon samples, and the sample label of each enhanced icon sample is consistent with the sample label of the corresponding initial icon sample; inputting the target icon samples into a counterfeit detection model one by one to obtain a detection result, and the detection result includes a result of the counterfeit object; adjusting the parameters of the counterfeit detection model according to the sample label and the detection result until the training end condition is met to obtain a trained counterfeit detection model.

[0019] In a possible implementation, the processor is further used to execute, before inputting the target icon sample into the counterfeit detection model to obtain the detection result: training the pre-trained model according to the image sample set to obtain the trained pre-trained model; and reconstructing the trained pre-trained model to obtain the counterfeit detection model.

[0020] In a possible implementation, the processor trains a pre-trained model according to an image sample set, and when obtaining a trained pre-trained model, specifically performs the following steps: obtaining the image sample set, wherein each image sample in the image sample set is marked with image content; inputting the image samples one by one into the pre-trained model to obtain image content; and adjusting parameters of the pre-trained model according to the marked image content and the image content output by the pre-trained model until the conditions for the end of training are met, thereby obtaining the trained pre-trained model.

[0021] In a possible implementation, the processor adjusts the parameters of the counterfeit detection model according to the sample label and the detection result until the training end condition is met to obtain the trained counterfeit detection model, and specifically performs: determining a loss function according to the sample label and the detection result; adjusting the parameters of the counterfeit detection model according to the loss function; when the loss function reaches the training end condition, stopping the training to obtain the trained counterfeit detection model.

[0022] In a possible implementation, the parameter adjustment includes gradient descent, back propagation, and learning rate adjustment. When the processor adjusts the parameters of the counterfeit detection model according to the loss function, it specifically performs: performing gradient descent and back propagation on the counterfeit detection model according to the loss function; and adjusting the learning rate of the counterfeit detection model according to the number of iterations.

[0023] In a sixth aspect, an embodiment of the present application provides an electronic device, comprising: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded and executed by the processor: obtaining installation package information from the installation package to be detected, the installation package information including an icon to be detected; inputting the icon to be detected into a counterfeit detection model to obtain a detection result, the detection result including a result of the counterfeit object, and the counterfeit detection model is the counterfeit detection model as described above; based on the detection result, determining whether the installation package to be detected is a counterfeit program installation package.

[0024] In the seventh aspect, an embodiment of the present application provides a computer storage medium, which stores multiple instructions, and the instructions are suitable for being loaded by a processor and executing the method steps provided by the first aspect of the embodiment of the present application or any one of the implementations of the first aspect.

[0025] In an eighth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed on a computer or a processor, enables the above-mentioned computer or processor to execute the encoding method provided by the first aspect of the embodiment of the present application or any possible implementation of the first aspect.

[0026] In the technical solutions provided in some embodiments of the present application, by performing sample enhancement on the initial icon samples in the initial icon sample set, a plurality of enhanced icon samples are obtained, thereby enriching the number of samples of each category in the icon sample set, so that the counterfeit detection model can have sufficient icon samples for training, thereby solving the problem in the prior art that the icon samples are insufficient for the training of the counterfeit detection model. The accuracy of the counterfeit detection model is improved by training the counterfeit detection model with the target icon samples.

[0027] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1 A schematic diagram of an exemplary system architecture to which the technical solution of the embodiments of the present application can be applied is shown.

[0030] Figure 2 A schematic flow chart of a counterfeit detection model training method provided in an embodiment of the present application is shown.

[0031] Figure 3 A flow chart of another counterfeit detection model training method provided in an embodiment of the present application is shown.

[0032] Figure 4 Shown according to Figure 2 A specific implementation flow chart of step S700 in the counterfeit detection model training method shown in the corresponding embodiment.

[0033] Figure 5 A schematic diagram of a flow chart of a counterfeit detection method provided in an embodiment of the present application is shown.

[0034] Figure 6 A schematic diagram of the structure of a counterfeit detection model training device provided in an embodiment of the present application.

[0035] Figure 7 A schematic diagram of the structure of a counterfeit detection device provided in an embodiment of the present application.

[0036] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0037] Fig. 9 A schematic diagram of the structure of another electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0038] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.

[0039] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0040] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0041] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0042] Figure 1 A schematic diagram of an exemplary system architecture to which the technical solution of the embodiments of the present application can be applied is shown.

[0043] like Figure 1 As shown, the system architecture may include terminal devices (such as Figure 1 The embodiment of the present invention is a schematic diagram of a mobile phone 101, a tablet computer 102, a portable computer 103, a desktop computer, etc., a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the terminal device and the server 105. The network 104 may include various connection types, such as a wired communication link, a wireless communication link, etc.

[0044] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. According to the implementation requirements, there may be any number of terminal devices, networks and servers. For example, the server 105 may be a server cluster composed of multiple servers.

[0045] The user can use the terminal device to interact with the server 105 through the network 104 to receive or send messages, etc. The server 105 can be a server that provides various services. For example, the user uses the terminal device 103 (or the terminal device 101 or 102) to upload the initial icon sample set to the server 105, and the server 105 can perform sample enhancement on the initial icon sample set to obtain a target icon sample set, wherein the target icon sample set includes target icon samples, and the target icon samples include the initial icon samples and the enhanced icon samples obtained by performing data enhancement processing on the initial icon samples, and the sample labels of each enhanced icon sample are consistent with the sample labels of the corresponding initial icon samples; the target icon samples are input into the counterfeit detection model one by one to obtain the detection results, and the detection results include the results of the counterfeit objects; according to the sample labels and the detection results, the parameters of the counterfeit detection model are adjusted until the training end conditions are met to obtain a trained counterfeit detection model.

[0046] It should be noted that the counterfeit detection model training method provided in the embodiment of the present application is generally executed by the server 105, and accordingly, the counterfeit detection model training device is generally set in the server 105. However, in other embodiments of the present application, the terminal device may also have similar functions as the server, so as to execute the counterfeit detection model training solution provided in the embodiment of the present application.

[0047] The implementation details of the technical solution of the embodiment of the present application are described in detail below:

[0048] Figure 2 A flowchart of a counterfeit detection model training method according to an embodiment of the present application is shown. The counterfeit detection model training method can be executed by a server, which can be Figure 1 Refer to the server shown in . Figure 2 As shown, the counterfeit detection model training method at least includes:

[0049] Step S100: obtaining an initial icon sample set, wherein each initial icon sample in the initial icon sample set is marked with a sample label, and the sample label has a counterfeit object.

[0050] Step S300, performing sample enhancement on the initial icon sample set to obtain a target icon sample set, wherein the target icon sample set includes target icon samples, and the target icon samples include the initial icon samples and enhanced icon samples obtained by performing data enhancement processing on the initial icon samples, and the sample label of each enhanced icon sample is consistent with the sample label of the corresponding initial icon sample.

[0051] Step S500: Input the target icon samples into a counterfeit detection model one by one to obtain a detection result, wherein the detection result includes a result of the counterfeit object.

[0052] Step S700, adjusting the parameters of the counterfeit detection model according to the sample labels and the detection results until the training end condition is met to obtain a trained counterfeit detection model.

[0053] In an embodiment of the present application, by performing sample enhancement on the initial icon samples in the initial icon sample set, a plurality of enhanced icon samples are obtained, thereby enriching the number of samples of each category in the icon sample set, so that the counterfeit detection model can have sufficient icon samples for training, thereby solving the problem in the prior art that the icon samples are insufficient for the training of the counterfeit detection model. By training the counterfeit detection model with the target icon samples, the accuracy of the counterfeit detection model is improved.

[0054] In step S100, samples are first obtained from a sample library to form an initial icon sample set. Each initial icon sample in the initial icon sample set is marked with a sample label, and the sample label records the counterfeit object.

[0055] In step S300, the initial icon sample set is sample enhanced, and each initial icon sample in the initial icon sample set is subjected to sample enhancement operations such as changing color, changing brightness, changing contrast, filling white borders around, center cropping, and random cropping to obtain a corresponding enhanced icon sample.

[0056] For example, multiple counterfeit icons of Bank A can be obtained by randomly changing the brightness of the icon of Bank A, and multiple counterfeit icons of Bank A can be obtained by randomly cutting the icon.

[0057] Through the above sample enhancement operation, each initial sample icon will get enough counterfeit icon samples. These counterfeit icon samples and initial icon samples form a target icon sample set with enough target icon samples. After having enough target icon samples, the problem of model overfitting can be effectively alleviated, and rich samples can also enhance the generalization ability of the model.

[0058] In some embodiments, Figure 3 As shown, before step S500, the counterfeit detection model training method further includes:

[0059] Step S200: training the pre-trained model according to the image sample set to obtain a trained pre-trained model.

[0060] Step S400, reconstructing the trained pre-trained model to obtain the counterfeit detection model.

[0061] In the embodiments of the present application, due to the small number of samples, in order to ensure the effectiveness of the counterfeit detection model, before training the counterfeit detection model, the pre-trained model is first pre-trained using other image samples so that the model acquires certain image capabilities. Then, the pre-trained model is reconstructed to form a counterfeit detection model, and then the counterfeit detection model is trained, thereby effectively solving the problem of model overfitting.

[0062] In step S200, the image sample set may come from various databases, such as the Image Net database or other collected image libraries containing various images. By training the image samples in the image library, the model can acquire certain image capabilities, thereby solving its overfitting problem.

[0063] Specifically, in some embodiments, the specific implementation of step S200 can refer to the following embodiments. Figure 3 The detailed description of step S200 in the counterfeit detection model training method shown in the corresponding embodiment, in the counterfeit detection model training method, step S200 may include the following steps:

[0064] The image sample set is obtained, wherein each image sample in the image sample set is marked with image content.

[0065] The image samples are input into the pre-trained model one by one to obtain image content.

[0066] According to the marked image content and the image content output by the pre-trained model, the parameters of the pre-trained model are adjusted until the conditions for the end of training are met, thereby obtaining the trained pre-trained model.

[0067] In an embodiment of the present application, by pre-training the pre-trained model for image recognition, that is, training the model to recognize the content of an image, the trained pre-trained model can recognize the content of the image well. For example, it can recognize that there are cats and dogs in the image. At this time, it already has sufficient image recognition capabilities.

[0068] The condition for the end of the pre-trained model can be that when a predetermined proportion of image samples in the image sample set can be recognized by the pre-trained model, that is, when more than a predetermined proportion of image samples in the image sample set are input into the pre-trained model, the obtained image content is consistent with the marked image content, then the training ends and the trained pre-trained model is obtained.

[0069] The condition for the end of the pre-trained model may be that the loss function converges or the loss function is less than a predetermined loss threshold, that is, first calculate the loss function based on the marked image content and the image content output by the pre-trained model, and then adjust the parameters of the pre-trained model according to the loss function until the loss function converges or the loss function is less than the predetermined loss threshold, the training ends, and the trained pre-trained model is obtained.

[0070] In step S400, the output of the preprocessing model is the image content, while the output of the counterfeit detection model required by the present application is the detection result, so the preprocessing model needs to be reconstructed to transform the preprocessing model into a counterfeit detection model that outputs the detection result.

[0071] Specifically, in some embodiments, the specific implementation of step S400 can refer to the following embodiments. Figure 3 The detailed description of step S400 in the counterfeit detection model training method shown in the corresponding embodiment, in the counterfeit detection model training method, step S400 may include the following steps:

[0072] The output layer of the trained pre-trained model is reconstructed to obtain the counterfeit detection model.

[0073] In the embodiments of the present application, since the output of the pre-trained model is the image content rather than the detection result, but the principle of the image recognition network is the same, it is only necessary to change the output layer (top-layer, the top layer can also be called the classification layer) located at the upper layer so that its output vector to the recognition network is processed from the recognition content to the object of detecting counterfeit, that is, the model reconstruction is completed and the counterfeit detection model is obtained.

[0074] Specifically, the output layer of the pre-trained model can be removed first, and then a new output layer that outputs the detection result can be re-added to obtain the counterfeit detection model.

[0075] It should be emphasized that step S100 and step S200 can be executed synchronously or asynchronously, and this application does not make any limitation here. Step S300 is executed after step S100, and step S400 is executed after step S200. This application does not make any other limitations here.

[0076] When the counterfeit detection model is reconstructed in step S400 and the sample data is enhanced to be sufficient in step S300, step S500 can be executed for training. In the embodiment of the present application, the counterfeit detection model is modeled by residual network (ResNet), that is, the idea of ​​residual learning is added to the traditional convolutional neural network to solve the problem of gradient diffusion and precision reduction (training set) in the deep network, so that the network can become deeper and deeper, which not only ensures the precision but also controls the speed.

[0077] In step S700, the parameter adjustment of the counterfeit detection model not only includes the adjustment of each layer of the counterfeit detection model network, such as the adjustment of the convolution layer, activation layer, pooling layer, fully connected layer and output layer network, but also involves some hyperparameters of the model, such as the learning rate.

[0078] Specifically, in some embodiments, the specific implementation of step S700 can be found in Figure 4 . Figure 4 is based on Figure 2 The detailed description of step S700 in the counterfeit detection model training method shown in the corresponding embodiment, in the counterfeit detection model training method, step S700 may include the following steps:

[0079] Step S710: determining a loss function according to the sample label and the detection result.

[0080] Step S720: adjusting parameters of the counterfeit detection model according to the loss function.

[0081] Step S730, when the loss function reaches the training end condition, stop the training to obtain the trained counterfeit detection model.

[0082] In an embodiment of the present application, a loss function is first determined based on sample labels and detection results, and then parameters of the counterfeit detection model are adjusted based on the loss function until the loss function reaches the training end condition, and the training is stopped to obtain a trained counterfeit detection model.

[0083] In step S710, the loss function may be determined by calculating the cross entropy between the sample label and the detection result and using the cross entropy as the loss function, or by directly using the softmax function to calculate and obtain the loss function, which is not limited in this application.

[0084] In step S720, the parameter adjustment of the counterfeit detection model not only includes the adjustment of each layer of the counterfeit detection model network, such as the adjustment of the convolution layer, activation layer, pooling layer, fully connected layer and output layer network, but also involves some hyperparameters of the model, such as the learning rate.

[0085] Specifically, in some embodiments, the specific implementation of step S700 can be found in Figure 4 . Figure 4 is based on Figure 2 The detailed description of step S700 in the counterfeit detection model training method shown in the corresponding embodiment, in the counterfeit detection model training method, the parameter adjustment includes gradient descent, back propagation and learning rate adjustment, and step S700 may include the following steps:

[0086] Gradient descent and back propagation are performed on the counterfeit detection model according to the loss function.

[0087] The learning rate of the counterfeit detection model is adjusted according to the number of iterations.

[0088] In an embodiment of the present application, each time a target icon sample is input, a loss function can be obtained, and the counterfeit detection model can be gradient-descended and back-propagated according to the loss function to update each network layer of the counterfeit model.

[0089] At the same time, a target icon sample is input, and the counterfeit detection model undergoes one iteration (epoch), and its learning rate will be adjusted accordingly.

[0090] Specifically, for the adjustment of the learning rate, a warm up mechanism can be used, that is, the learning rate is gradually adjusted from a lower value (i.e., the baseline learning rate) to a higher value (i.e., the predetermined learning rate). After each iteration (epoch) of the counterfeit detection model, the learning rate is increased by a predetermined amount until the predetermined learning rate is reached, so as to solve the problem that the learning rate is too large, causing the loss function to directly cross the global optimal point, and also solve the problem that if the learning rate is too small, the loss function changes slowly and increases the convergence complexity of the network. Through the warm up mechanism, a better learning rate can be avoided so that the counterfeit detection model can quickly and accurately reach a better convergence.

[0091] In some embodiments of the present application, in addition to adopting the warm up mechanism, the learning rate is also periodically restarted. When the learning rate of the target icon sample reaches the predetermined learning rate, the learning rate is gradually decreased for a predetermined number of iterations, and then the learning rate is gradually adjusted upward to the predetermined learning rate until the loss function meets the training end condition, so as to avoid the learning rate being maintained at a larger predetermined learning rate, causing the loss function to directly cross the global optimal point.

[0092] In step S730, the conditions for the loss function to reach the end of training include: the loss function does not decrease after a predetermined number of iterations (epochs) (that is, the loss function reaches convergence) and the training stops when the accuracy of the counterfeit detection model in detecting the target icon samples in the target icon sample set reaches a set threshold, thereby reducing the possibility of model overfitting and enhancing the generalization ability of the model.

[0093] Figure 5 A flowchart of a counterfeit detection method according to an embodiment of the present application is shown. The counterfeit detection method can be executed by a server, which can be Figure 1 Refer to the server shown in . Figure 5 As shown, the counterfeit detection method at least includes:

[0094] Step S010: obtaining installation package information from the installation package to be detected, wherein the installation package information includes an icon to be detected.

[0095] Step S020, inputting the icon to be detected into a counterfeit detection model to obtain a detection result, wherein the detection result includes a result of the counterfeit object, and the counterfeit detection model is the counterfeit detection model described above.

[0096] Step S030: determining whether the installation package to be detected is a counterfeit program installation package according to the detection result.

[0097] In the embodiment of the present application, when the counterfeit detection model is used for counterfeit detection, the installation package to be detected is first parsed to obtain the application icon of the corresponding application, i.e., the icon to be detected, and then the icon to be detected is input into the above-mentioned counterfeit detection model to obtain the detection result.

[0098] In the embodiments of the present application,

[0099] In step S010, the installation package to be detected is parsed, and the ICON file in the installation package is extracted as the icon to be detected.

[0100] In step S020, the icon to be detected is input into the counterfeit detection model to obtain the result of the counterfeit object. However, since the counterfeit detection model can only detect the counterfeit object of the icon, the icon of the counterfeit object itself is input into the counterfeit detection model and will also be determined to be a counterfeit of the counterfeit object. Therefore, step S030 needs to be executed for further judgment.

[0101] In step S030, if the detection result obtains the result of the counterfeit object, then it is necessary to determine whether the installation package is the counterfeit object itself based on the installation package information. If it is the counterfeit object itself, it is necessary to further analyze it based on the installation package information. That is, in the embodiment of the present application, the installation package information can also include application ID, application signature, developer certificate, etc. At this time, if it is concluded that it is the counterfeit object itself based on the application ID, application signature, developer certificate, etc., then the installation package is not a counterfeit program installation package; if it is concluded that it is not the counterfeit object itself based on the application ID, application signature, developer certificate, etc., then the installation package is a counterfeit program installation package.

[0102] Analyzing whether the installation package is the counterfeit object itself based on the application ID, application signature, and developer certificate can be achieved through a whitelist system, that is, by querying the whitelist, if the developer certificate of the installation package is in the whitelist, then the installation package is the counterfeit object itself, otherwise it is not the counterfeit object itself.

[0103] If the detection result does not contain any counterfeit object, it means that the installation package has no counterfeit object and can be considered that the installation package is safe and not a counterfeit program installation package.

[0104] Meanwhile, in some embodiments, in step S020, the counterfeit detection model not only outputs the result of the counterfeit object but also outputs the confidence level of the result, which can be represented by a confidence value or a confidence score.

[0105] In step S030, when determining whether the installation package to be detected is a counterfeit program installation package, it is also determined based on the confidence level of the counterfeit detection model for the result whether the result can be directly adopted, needs manual review, or can be directly discarded.

[0106] For results with a confidence level in the first confidence interval, the result can be directly adopted; for results with a confidence level in the second confidence interval, the result needs to be manually reviewed; for results in the third confidence interval, the result can be directly discarded.

[0107] In one embodiment, the first confidence interval is (0.95, 1], the second confidence interval is (0.8, 0.95], and the third confidence interval is [0, 0.8].

[0108] For the confidence level in the second confidence interval, if the result of the counterfeit object output by the counterfeit detection model is wrong, it is corrected, and according to the corrected result, the icon to be detected is marked with the corrected result and put into the initial icon sample set as the initial icon sample.

[0109] Figure 6The following is a schematic diagram showing the structure of a counterfeit detection model training device provided by an embodiment of the present application. Figure 6 As shown, the counterfeit detection model training device 600 may at least include: a sample acquisition module 610, a sample enhancement module 620, a sample input module 630, and a parameter adjustment module 640. Among them: the sample acquisition module 610 is used to acquire an initial icon sample set, each initial icon sample in the initial icon sample set is marked with a sample label, and the sample label includes a counterfeit object; the sample enhancement module 620 is used to perform sample enhancement on the initial icon sample set to obtain a target icon sample set, the target icon sample set includes target icon samples, the target icon samples include the initial icon samples and the enhanced icon samples obtained by performing data enhancement processing on the initial icon samples, and the sample label of each enhanced icon sample is consistent with the sample label of the corresponding initial icon sample; the sample input module 630 is used to input the target icon samples one by one into the counterfeit detection model to obtain a detection result, and the detection result includes the result of the counterfeit object; the parameter adjustment module 640 is used to adjust the parameters of the counterfeit detection model according to the sample label and the detection result until the training end condition is met to obtain a trained counterfeit detection model.

[0110] In a possible implementation, the counterfeit detection model training device also includes: a pre-training module, used to train the pre-trained model according to the image sample set to obtain a trained pre-trained model; and a model reconstruction module, used to reconstruct the trained pre-trained model to obtain the counterfeit detection model.

[0111] In a possible implementation, the pre-training module specifically includes: an acquisition submodule, used to acquire the image sample set, each image sample in the image sample set is marked with image content; an input submodule, used to input the image samples one by one into the pre-training model to obtain the image content; an adjustment submodule, used to adjust the parameters of the pre-training model according to the marked image content and the image content output by the pre-training model until the conditions for the end of training are met, thereby obtaining the trained pre-training model.

[0112] In a possible implementation, the parameter adjustment module 640 specifically includes: a loss determination submodule, used to determine the loss function according to the sample label and the detection result; a parameter adjustment submodule, used to adjust the parameters of the counterfeit detection model according to the loss function; and a stop training submodule, used to stop training when the loss function reaches the training end condition to obtain the trained counterfeit detection model.

[0113] In a possible implementation, the parameter adjustment includes gradient descent, back propagation and learning rate adjustment, and the parameter adjustment submodule specifically includes: a first adjustment unit, used to perform gradient descent and back propagation on the counterfeit detection model according to the loss function; a second adjustment unit, used to adjust the learning rate of the counterfeit detection model according to the number of iterations.

[0114] In the technical solutions provided in some embodiments of the present application, by performing sample enhancement on the initial icon samples in the initial icon sample set, a plurality of enhanced icon samples are obtained, thereby enriching the number of samples of each category in the icon sample set, so that the counterfeit detection model can have sufficient icon samples for training, thereby solving the problem in the prior art that the icon samples are insufficient for the training of the counterfeit detection model. The accuracy of the counterfeit detection model is improved by training the counterfeit detection model with the target icon samples.

[0115] It should be noted that the counterfeit detection model training device provided in the above embodiment only uses the division of the above functional modules as an example when executing the counterfeit detection model training method. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the wireless screen projection connection device and the wireless screen projection connection method embodiment provided in the above embodiment belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.

[0116] Figure 7 The following is a schematic diagram showing the structure of a counterfeit detection device provided in an embodiment of the present application. Figure 7 As shown, the counterfeit detection model 700 may at least include: an icon acquisition module 710, an icon detection module 720, and a result determination module 730. Among them: the icon acquisition module 710 is used to obtain installation package information from the installation package to be detected, and the installation package information includes the icon to be detected. The icon detection module 720 is used to input the icon to be detected into the counterfeit detection model to obtain a detection result, and the detection result includes the result of the counterfeit object. The counterfeit detection model is the counterfeit detection model described above. The result determination module 730 is used to determine whether the installation package to be detected is a counterfeit program installation package based on the detection result.

[0117] In the technical solutions provided in some embodiments of the present application, by using the above-mentioned counterfeit detection model to perform counterfeit detection and analyzing the detection results to determine whether the installation package is a counterfeit program installation package, not only the efficiency of counterfeit detection is improved, but also the accuracy of counterfeit detection is improved.

[0118] It should be noted that the counterfeit detection device provided in the above embodiment only uses the division of the above functional modules as an example when executing the counterfeit detection method. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the wireless screen projection connection device and the wireless screen projection connection method embodiment provided in the above embodiment belong to the same concept. The implementation process is detailed in the method embodiment, which will not be repeated here.

[0119] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0120] See also Figure 8 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 8 As shown, the electronic device 800 may include: at least one processor 801 , at least one network interface 804 , a user interface 803 , a memory 805 , and at least one communication bus 802 .

[0121] The communication bus 802 is used to realize the connection and communication between these components.

[0122] The user interface 803 may include a display screen (Display) and a camera (Camera), and the optional user interface 803 may also include a standard wired interface and a wireless interface.

[0123] The network interface 804 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0124] Among them, the processor 801 may include one or more processing cores. The processor 801 uses various interfaces and lines to connect various parts of the entire electronic device 800, and executes various functions and processes data of the electronic device 80 by running or executing instructions, programs, code sets or instruction sets stored in the memory 805, and calling data stored in the memory 805. Optionally, the processor 801 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 801 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 801, and it can be implemented separately through a chip.

[0125] Among them, the memory 805 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 805 includes a non-transitory computer-readable storage medium. The memory 805 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 805 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 805 may optionally be at least one storage device located away from the aforementioned processor 801. As Figure 8 As shown, the memory 805 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a counterfeit detection model training application.

[0126] exist Figure 8In the electronic device 800 shown, the user interface 803 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 801 can be used to call the counterfeit detection model training application stored in the memory 805, and specifically perform the following operations:

[0127] An initial icon sample set is obtained, wherein each initial icon sample in the initial icon sample set is marked with a sample label, and the sample label includes a counterfeit object; sample enhancement is performed on the initial icon sample set to obtain a target icon sample set, wherein the target icon sample set includes target icon samples, and the target icon samples include the initial icon samples and enhanced icon samples obtained by performing data enhancement processing on the initial icon samples, and the sample label of each enhanced icon sample is consistent with the sample label of the corresponding initial icon sample; the target icon samples are input into a counterfeit detection model one by one to obtain a detection result, and the detection result includes a result of the counterfeit object; according to the sample label and the detection result, the parameters of the counterfeit detection model are adjusted until the training end condition is met, and a trained counterfeit detection model is obtained.

[0128] In a possible implementation, the processor is further used to execute, before inputting the target icon sample into the counterfeit detection model to obtain the detection result: training the pre-trained model according to the image sample set to obtain the trained pre-trained model; and reconstructing the trained pre-trained model to obtain the counterfeit detection model.

[0129] In a possible implementation, the processor trains a pre-trained model according to an image sample set, and when obtaining a trained pre-trained model, specifically performs the following steps: obtaining the image sample set, wherein each image sample in the image sample set is marked with image content; inputting the image samples one by one into the pre-trained model to obtain image content; and adjusting parameters of the pre-trained model according to the marked image content and the image content output by the pre-trained model until the conditions for the end of training are met, thereby obtaining the trained pre-trained model.

[0130] In a possible implementation, the processor adjusts the parameters of the counterfeit detection model according to the sample label and the detection result until the training end condition is met to obtain the trained counterfeit detection model, and specifically performs: determining a loss function according to the sample label and the detection result; adjusting the parameters of the counterfeit detection model according to the loss function; when the loss function reaches the training end condition, stopping the training to obtain the trained counterfeit detection model.

[0131] In a possible implementation, the parameter adjustment includes gradient descent, back propagation, and learning rate adjustment. When the processor adjusts the parameters of the counterfeit detection model according to the loss function, it specifically performs: performing gradient descent and back propagation on the counterfeit detection model according to the loss function; and adjusting the learning rate of the counterfeit detection model according to the number of iterations.

[0132] In an embodiment of the present application, by performing sample enhancement on the initial icon samples in the initial icon sample set, a plurality of enhanced icon samples are obtained, thereby enriching the number of samples of each category in the icon sample set, so that the counterfeit detection model can have sufficient icon samples for training, thereby solving the problem in the prior art that the icon samples are insufficient for the training of the counterfeit detection model. By training the counterfeit detection model with the target icon samples, the accuracy of the counterfeit detection model is improved.

[0133] See also Fig. 9 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Fig. 9 As shown, the electronic device 900 may include: at least one processor 901 , at least one network interface 904 , a user interface 903 , a memory 905 , and at least one communication bus 902 .

[0134] The communication bus 902 is used to realize the connection and communication between these components.

[0135] The user interface 903 may include a display screen (Display) and a camera (Camera), and the optional user interface 903 may also include a standard wired interface and a wireless interface.

[0136] The network interface 904 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0137] Among them, the processor 901 may include one or more processing cores. The processor 901 uses various interfaces and lines to connect various parts of the entire electronic device 900, and executes various functions and processes data of the electronic device 60 by running or executing instructions, programs, code sets or instruction sets stored in the memory 905, and calling data stored in the memory 905. Optionally, the processor 901 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 901 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 901, and it can be implemented separately through a chip.

[0138] Among them, the memory 905 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 905 includes a non-transitory computer-readable storage medium. The memory 905 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 905 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 905 may also be optionally at least one storage device located away from the aforementioned processor 901. As Fig. 9 As shown, the memory 905 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program.

[0139] exist Fig. 9In the electronic device 900 shown, the user interface 903 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 901 can be used to call the application stored in the memory 905, and specifically perform the following operations: obtain installation package information from the installation package to be detected, and the installation package information includes the icon to be detected; input the icon to be detected into the counterfeit detection model to obtain a detection result, and the detection result includes the result of the counterfeit object, and the counterfeit detection model is the counterfeit detection model as described above; according to the detection result, determine whether the installation package to be detected is a counterfeit program installation package.

[0140] In the technical solutions provided in some embodiments of the present application, by using the above-mentioned counterfeit detection model to perform counterfeit detection and analyzing the detection results to determine whether the installation package is a counterfeit program installation package, not only the efficiency of counterfeit detection is improved, but also the accuracy of counterfeit detection is improved.

[0141] The embodiment of the present application also provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a computer or a processor, the computer or the processor executes the above-mentioned Figure 2-Figure 5 One or more steps in the illustrated embodiment. If the components of the electronic device described above are implemented in the form of software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0142] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted by the computer-readable storage medium. The computer instructions can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (Digital Subscriber Line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server, a data center, etc. that contains one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0143] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The aforementioned storage medium includes: various media that can store program codes, such as system memory (Read Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk. In the absence of conflict, the technical features in this embodiment and the implementation scheme can be combined arbitrarily.

[0144] The embodiments described above are merely preferred embodiments of the present application and are not intended to limit the scope of the present application. Without departing from the design spirit of the present application, various modifications and improvements made to the technical solutions of the present application by ordinary technicians in this field should fall within the protection scope determined by the claims of the present application.

Claims

1. A method for training a counterfeit detection model, characterized in that: The counterfeit detection model training method comprises: Acquire an initial icon sample set, wherein each initial icon sample in the initial icon sample set is marked with a sample label, and the sample label has a counterfeit object; Performing sample enhancement on the initial icon sample set to obtain a target icon sample set, wherein the target icon sample set includes target icon samples, and the target icon samples include the initial icon samples and enhanced icon samples obtained by performing data enhancement processing on the initial icon samples, and the sample label of each enhanced icon sample is consistent with the sample label of the corresponding initial icon sample; Input the target icon samples one by one into the counterfeit detection model to obtain a detection result, wherein the detection result includes a result of the counterfeit object; According to the sample labels and the detection results, the parameters of the counterfeit detection model are adjusted until the training end condition is met to obtain a trained counterfeit detection model.

2. The method according to claim 1, characterized in that Before inputting the target icon sample into the counterfeit detection model to obtain the detection result, the counterfeit detection model training method further includes: According to the image sample set, the pre-trained model is trained to obtain a trained pre-trained model; The trained pre-trained model is reconstructed to obtain the counterfeit detection model.

3. The method according to claim 2, characterized in that The method of training the pre-trained model according to the image sample set to obtain a trained pre-trained model specifically includes: Acquire the image sample set, each image sample in the image sample set being marked with image content; Input the image samples into the pre-trained model one by one to obtain image content; According to the marked image content and the image content output by the pre-trained model, the parameters of the pre-trained model are adjusted until the conditions for the end of training are met, thereby obtaining the trained pre-trained model.

4. The method according to claim 1, characterized in that The step of adjusting parameters of the counterfeit detection model according to the sample labels and the detection results until a training end condition is met to obtain a trained counterfeit detection model specifically includes: Determine a loss function according to the sample label and the detection result; Adjusting parameters of the counterfeit detection model according to the loss function; When the loss function reaches the training end condition, the training is stopped to obtain the trained counterfeit detection model.

5. The method according to claim 4, characterized in that The parameter adjustment includes gradient descent, back propagation and learning rate adjustment, and the parameter adjustment of the counterfeit detection model according to the loss function specifically includes: Performing gradient descent and back propagation on the counterfeit detection model according to the loss function; The learning rate of the counterfeit detection model is adjusted according to the number of iterations.

6. A counterfeit detection method, characterized in that: The counterfeit detection method comprises: Acquire installation package information from the installation package to be detected, wherein the installation package information includes an icon to be detected; Inputting the icon to be detected into a counterfeit detection model to obtain a detection result, wherein the detection result includes a result of the counterfeit object, and the counterfeit detection model is a counterfeit detection model according to any one of claims 1 to 5; According to the detection result, it is determined whether the installation package to be detected is a counterfeit program installation package.

7. A counterfeit detection model training device, characterized in that: The counterfeit detection model training device comprises: A sample acquisition module, used to acquire an initial icon sample set, wherein each initial icon sample in the initial icon sample set is marked with a sample label, and the sample label has a counterfeit object; A sample enhancement module, configured to perform sample enhancement on the initial icon sample set to obtain a target icon sample set, wherein the target icon sample set includes target icon samples, and the target icon samples include the initial icon samples and enhanced icon samples obtained by performing data enhancement processing on the initial icon samples, and the sample label of each enhanced icon sample is consistent with the sample label of the corresponding initial icon sample; A sample input module, used to input the target icon samples into the counterfeit detection model one by one to obtain a detection result, wherein the detection result includes a result of the counterfeit object; The parameter adjustment module is used to adjust the parameters of the counterfeit detection model according to the sample label and the detection result until the training end condition is met to obtain a trained counterfeit detection model.

8. A counterfeit detection device, characterized in that: The counterfeit detection device comprises: An icon acquisition module, used to acquire installation package information from the installation package to be detected, wherein the installation package information includes the icon to be detected; An icon detection module, used for inputting the icon to be detected into a counterfeit detection model to obtain a detection result, wherein the detection result includes a result of the counterfeit object, and the counterfeit detection model is a counterfeit detection model according to any one of claims 1 to 5; The result determination module is used to determine whether the installation package to be detected is a counterfeit program installation package according to the detection result.

9. An electronic device, characterized in that: include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method steps as claimed in any one of claims 1 to 6.

10. A computer storage medium, characterized in that: The computer storage medium stores a plurality of instructions, which are suitable for being loaded by a processor and executing the method steps according to any one of claims 1 to 6.

Citation Information

Patent Citations

  • Mark detection model training and mark detection method based on multi-stage transfer learning

    CN111062885A

  • Counterfeit APP discovery method and device, medium and equipment

    CN116127460A