Visual data processing method, security detection method, computing device and storage medium

Abstract

The present application relates to the technical field of image processing, and in particular to a visual data processing method, a security detection method, a computing device and a storage medium. The visual data processing method comprises: determining an object to be processed which is included in initial visual data, and on the basis of said object, performing segmentation processing on the initial visual data, so as to obtain first visual data that includes said object and second visual data that does not include said object (202); performing enhancement processing on the second visual data, so as to obtain a plurality of pieces of third visual data (204); and on the basis of the first visual data and the plurality of pieces of third visual data, constructing a plurality of visual data verification codes, so as to perform security detection on the basis of the visual data verification codes (206).

Description

Visual data processing methods, security detection methods, computing devices and storage media Technical Field

[0001] This disclosure relates to the field of image processing technology, and in particular to visual data processing methods, security detection methods, computing devices, and storage media. Background Technology

[0002] A CAPTCHA can be understood as a technology used to distinguish between a human and a computer program. Typically, when a user wants to log in to an account, website, or application, a CAPTCHA is used to verify their identity. To counter evolving automated attacks and improve user experience, the form of CAPTCHAs has undergone several changes. CAPTCHAs can now be text-based, visual data-based (e.g., image recognition CAPTCHAs), icon-based click CAPTCHAs, and spatial semantic CAPTCHAs, among others.

[0003] In current visual data CAPTCHAs, visual data is usually presented in the form of images. However, image CAPTCHAs generated based on image editing methods are usually of poor quality and costly. They can usually only generate image CAPTCHAs with specific content, making them easy to crack when attacked, further increasing the risk to users. Therefore, there is an urgent need for an effective technical solution to solve the above problems. Summary of the Invention

[0004] In view of the above, embodiments of this disclosure provide a visual data processing method. One or more embodiments of this disclosure also relate to a visual data processing apparatus, a security detection method, a security detection device, a computing device, a computer-readable storage medium, and a computer program product, to address the technical deficiencies existing in the prior art.

[0005] According to a first aspect of the present disclosure, a visual data processing method is provided, comprising:

[0006] The objects to be processed contained in the initial visual data are determined, and the initial visual data is segmented according to the objects to be processed to obtain first visual data containing the objects to be processed and second visual data not containing the objects to be processed.

[0007] The second visual data is enhanced to obtain multiple third visual data.

[0008] Based on the first visual data and the plurality of third visual data, multiple visual data verification codes are constructed to perform security detection based on the visual data verification codes.

[0009] According to a second aspect of the present disclosure, a visual data processing apparatus is provided, comprising:

[0010] The segmentation module is configured to determine the objects to be processed contained in the initial visual data, and to segment the initial visual data according to the objects to be processed to obtain first visual data containing the objects to be processed and second visual data not containing the objects to be processed.

[0011] The enhancement module is configured to enhance the second visual data to obtain multiple third visual data.

[0012] The construction module is configured to construct multiple visual data verification codes based on the first visual data and the multiple third visual data, so as to perform security detection based on the visual data verification codes.

[0013] According to a third aspect of the present disclosure, a security detection method is provided, comprising:

[0014] In response to a data processing request, a visual data verification code is displayed containing movable visual data and fixed visual data, wherein the visual data verification code is determined according to the visual data processing method provided in the first aspect of the present disclosure, the movable visual data is a first visual data, and the fixed visual data is a third visual data;

[0015] In response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data, a security check is performed on the data processing request.

[0016] According to a fourth aspect of the present disclosure, a security detection device is provided, comprising:

[0017] The display module is configured to respond to a data processing request by displaying the visual data to be moved and the fixed visual data in the visual data verification code, wherein the visual data verification code is determined according to the visual data processing method provided in the first aspect of the present disclosure, the visual data to be moved is the first visual data, and the fixed visual data is the third visual data.

[0018] The detection module is configured to perform security detection on the data processing request in response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data.

[0019] According to a fifth aspect of the present disclosure, a computing device is provided, comprising:

[0020] Memory and processor;

[0021] The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions, which, when executed by the processor, implement the steps of the above method.

[0022] According to a sixth aspect of the present disclosure, a computer-readable storage medium is provided that stores a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0023] According to a seventh aspect of the present disclosure, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0024] One embodiment of this disclosure provides a visual data processing method, comprising: determining an object to be processed contained in initial visual data, and segmenting the initial visual data according to the object to be processed to obtain first visual data containing the object to be processed and second visual data not containing the object to be processed; enhancing the second visual data to obtain multiple third visual data; and constructing multiple visual data verification codes according to the first visual data and the multiple third visual data for security detection based on the visual data verification codes.

[0025] The above method identifies the object to be processed within the initial visual data and segments the initial visual data based on the object to be processed, obtaining first visual data containing the object to be processed and second visual data not containing the object to be processed. This enables image matting of the initial visual data. Furthermore, the second visual data is enhanced to obtain multiple third visual data, increasing the number of visual data and improving the image quality of the second visual data through enhancement. Subsequently, a visual data CAPTCHA is constructed based on the first visual data and multiple third visual data, further improving both the quantity and quality of the CAPTCHA. Security detection is then performed based on this visual data CAPTCHA to prevent it from being cracked during attacks, thus further ensuring the security of user access. Attached Figure Description

[0026] Figure 1 is a schematic diagram of an application scenario of a visual data processing method provided in an embodiment of this disclosure;

[0027] Figure 2 is a flowchart of a visual data processing method provided in an embodiment of this disclosure;

[0028] Figure 3 is a schematic diagram of a visual data verification code in a visual data processing method provided in an embodiment of this disclosure;

[0029] Figure 4 is a schematic diagram of the editing area in a visual data processing method provided in an embodiment of this disclosure;

[0030] Figure 5 is a flowchart of a visual data processing method provided in an embodiment of this disclosure;

[0031] Figure 6 is a schematic diagram of the structure of a visual data processing device provided in an embodiment of the present disclosure;

[0032] Figure 7 is a flowchart of a security detection method provided in an embodiment of this disclosure;

[0033] Figure 8 is a schematic diagram of a security detection device provided in an embodiment of this disclosure;

[0034] Figure 9 is a structural block diagram of a computing device provided in an embodiment of this disclosure. Detailed Implementation

[0035] Numerous specific details are set forth in the following description to provide a full understanding of this disclosure. However, this disclosure can be implemented in many other ways than those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this disclosure. Therefore, this disclosure is not limited to the specific implementations disclosed below.

[0036] The terminology used in one or more embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of this disclosure. The singular forms “a,” “the,” and “the” as used in one or more embodiments of this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in one or more embodiments of this disclosure refers to and includes any or all possible combinations of one or more associated listed items.

[0037] It should be understood that although the terms first, second, etc., may be used to describe various information in one or more embodiments of this disclosure, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second without departing from the scope of one or more embodiments of this disclosure, and similarly, second may also be referred to as first. Depending on the context, the word “if” as used herein may be interpreted as “when”, “in response to a determination”, or “when…”.

[0038] Furthermore, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this disclosure are all information and data authorized by the user or fully authorized by all parties. Moreover, the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0039] In one or more embodiments of this disclosure, a large model refers to a deep learning model with a large number of model parameters, typically containing hundreds of millions, tens of billions, hundreds of billions, trillions, or even tens of trillions of model parameters. A large model can also be called a foundation model. It is pre-trained using large-scale unlabeled corpora to produce a pre-trained model with hundreds of millions of parameters. Such models can adapt to a wide range of downstream tasks and have good generalization ability. Examples include Large Language Models (LLMs) and multi-modal pre-training models.

[0040] In practical applications, large models only require a small number of samples to fine-tune the pre-trained model before they can be applied to different tasks. Large models can be widely used in fields such as Natural Language Processing (NLP) and Computer Vision. Specifically, they can be applied to computer vision tasks such as Visual Question Answering (VQA), Image Captioning (IC), and Image Generation, as well as NLP tasks such as text-based sentiment classification, text summarization, and machine translation. The main application scenarios for large models include digital assistants, intelligent robots, search, online education, office software, e-commerce, and intelligent design.

[0041] This disclosure provides a visual data processing method, and also relates to a visual data processing apparatus, a security detection method, a security detection apparatus, a computing device, a computer-readable storage medium, and a computer program product, which will be described in detail in the following embodiments.

[0042] Referring to Figure 1, Figure 1 shows a schematic diagram of an application scenario of a visual data processing method according to an embodiment of the present disclosure. The specific steps of the visual data processing method are as follows.

[0043] The objects to be processed contained in the initial visual data are determined, and the initial visual data is segmented according to the objects to be processed to obtain first visual data containing the objects to be processed and second visual data not containing the objects to be processed.

[0044] The second visual data is enhanced to obtain multiple third visual data.

[0045] Based on the first visual data and the plurality of third visual data, multiple visual data verification codes are constructed to perform security detection based on the visual data verification codes.

[0046] Specifically, Figure 1 may include an edge device 102 and a cloud device 104, wherein the cloud device 104 may be deployed with a variety of models for visual data processing.

[0047] In practice, the user can send an image processing request to the cloud device 104 through the terminal device 102. The cloud device 104 can determine the initial visual data and the object to be processed contained in the initial visual data based on the image processing request. It can then call the visual data processing model to segment the initial visual data according to the object to be processed, obtain the first visual data containing the object to be processed and the second visual data not containing the object to be processed, and perform enhancement processing on the second visual data to obtain multiple third visual data. Based on the first visual data and the multiple third visual data, multiple visual data verification codes are constructed, realizing the low-cost and high-quality construction of visual data verification codes, and further enabling subsequent security detection based on visual data verification codes.

[0048] The edge device 102 may include a browser, an app (application), or a web application such as an H5 (Hypertext Markup Language 5) application, a lightweight application (also known as a mini-program), or a cloud application. The edge device can be developed based on a software development kit (SDK) provided by the server, such as a real-time communication (RTC) SDK. The edge device can be deployed in an electronic device and depends on the device's operation or certain apps within the device to run. The electronic device may have a display screen and support information browsing, such as a personal mobile terminal like a mobile phone, tablet, or personal computer. Various other types of applications can also be configured in the electronic device, such as human-computer interaction applications, model training applications, text processing applications, web browser applications, shopping applications, search applications, instant messaging tools, email clients, and social media platform software.

[0049] Cloud-side device 104 can be understood as a server providing various services, including physical servers and cloud servers. Examples include servers providing communication services to multiple clients, servers supporting backend training of models used on clients, and servers processing data sent by clients. It should be noted that cloud-side device 104 can be implemented as a distributed server cluster composed of multiple servers, or as a single server. Cloud-side device 104 can also be a server for a distributed system, or a server integrated with blockchain. Cloud-side device 104 can also be a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms, or an intelligent cloud computing server or intelligent cloud host with artificial intelligence technology.

[0050] It is worth noting that the visual data processing method provided in this embodiment can be executed by the cloud-side device 104. In other embodiments of this disclosure, the end-side device 102 can also execute the visual data processing method provided in this embodiment. In other embodiments, the visual data processing method provided in this embodiment can also be jointly executed by the end-side device 102 and the cloud-side device 104.

[0051] Referring to Figure 2, Figure 2 shows a flowchart of a visual data processing method provided according to an embodiment of the present disclosure, which specifically includes the following steps.

[0052] Step 202: Determine the objects to be processed contained in the initial visual data, and segment the initial visual data according to the objects to be processed to obtain first visual data containing the objects to be processed and second visual data not containing the objects to be processed.

[0053] The visual data can include image data, video data, and animation data. For ease of understanding, this embodiment uses image data as an example. The initial visual data can be understood as the initial image used to generate a visual data verification code. The object to be processed can be understood as the object in the initial visual data that needs to be detected and segmented. For example, if the initial visual data is an image of an umbrella, then the object to be processed can be the umbrella handle in the initial visual data. The first visual data can be understood as an image containing only the object to be processed. It can be understood that the first visual data can be an image obtained by cutting out the initial visual data. For example, if the object to be processed is the umbrella handle in the initial visual data, the first visual data can be an image of the umbrella handle. The second visual data can be understood as an image after removing the object to be processed from the initial visual data. For example, the second visual data can be an image after removing the umbrella handle from the initial visual data.

[0054] Based on this, the initial visual data used to generate the visual data verification code can be determined, and the object to be processed that needs to be detected and segmented in the initial visual data can be determined. According to the object to be processed, the object to be processed is removed from the initial visual data to obtain the first visual data containing the object to be processed after removal, and the second visual data that does not contain the object to be processed after removal.

[0055] In practical applications, before determining the objects to be processed contained in the initial visual data, the process further includes:

[0056] The initial visual data is generated based on the prompt text for the initial visual data.

[0057] The prompt text for the initial visual data can be understood as the image description text (i.e., text-to-image prompt words) describing the initial visual data. For example, if you want to generate an initial visual data containing an alarm clock, the prompt text for the initial visual data could be "Generate an image of an alarm clock". Or, if you want to generate an initial visual data of a dog running on grass, the prompt text for the initial visual data could be "Generate an image of a dog running on grass".

[0058] Specifically, it can generate the initial visual data corresponding to the image description text input by the user for the initial visual data.

[0059] In practical applications, the initial visual data can be generated based on the text-to-image model. The prompt text can then be used to guide the text-to-image model in generating the corresponding initial visual data. The prompt text can be input into the text-to-image model to obtain the initial visual data output by the model. Optionally, the initial visual data can also be generated using large models, generative adversarial networks, diffusion models, variational autoencoders, or multimodal pre-trained models, etc., but this embodiment does not limit the scope of the invention.

[0060] Understandably, the prompt text can be natural language text input by the user. Based on the input pattern of natural language, it is possible to generate visual data CAPTCHA questions with different content at low cost, without having to train models for different scenarios, thus achieving the generation of visual data CAPTCHAs with a wider range of image semantics.

[0061] In addition, initial visual data can be generated based on computer graphics technology. For example, initial visual data can be generated through 3D modeling and texture synthesis. Alternatively, photos taken by a camera can be used as initial visual data. Alternatively, photos taken by a camera can be input into an image processing model, and the image processing model can be used to process the photos, with the processed photos used as initial visual data. Alternatively, initial visual data can be generated based on random noise using a diffusion model. This disclosure does not limit the scope of the embodiments.

[0062] In summary, by generating corresponding initial visual data based on the prompt text, high-quality generation of the background image in the visual data CAPTCHA can be achieved. Furthermore, the ability to customize the generation of initial visual data based on the prompt text can meet the different needs of various tasks.

[0063] Specifically, determining the objects to be processed contained in the initial visual data includes:

[0064] Target detection is performed on the initial visual data to obtain the object to be processed contained in the initial visual data.

[0065] Specifically, object detection models can be used to detect objects in the initial visual data and identify the objects to be removed from the initial visual data.

[0066] In specific implementation, the step of performing target detection on the initial visual data to obtain the object to be processed contained in the initial visual data includes:

[0067] Based on the text prompts, target detection is performed on the initial visual data to obtain the object to be processed contained in the initial visual data.

[0068] Specifically, text prompts (i.e., object recognition prompts) can be used to perform object detection on the initial visual data using an object detection model. The text prompts can be understood as prompts used to guide the object detection model to detect which object in the initial visual data. For example, if the initial visual data contains a dog and a cat running on the grass, the text prompt could be "detect the dog's left leg in the image". In this case, the object to be processed is the dog's left leg in the initial visual data.

[0069] In practical applications, object detection can also be performed on the initial visual data based on pre-trained machine learning models, deep learning models, convolutional neural networks, and other object detection models. Alternatively, object detection can be achieved based on large models. Feature extractors and classifiers can be used to perform object detection on the initial visual data. Or, object detection algorithms can be used to divide the initial visual data into multiple grid cells and predict a fixed number of bounding boxes and their corresponding categories in each grid cell, thereby achieving object detection on the initial visual data. This disclosure does not limit the scope of the embodiments.

[0070] Furthermore, in one embodiment of this disclosure, the initial visual data can be segmented based on an image segmentation model according to the object to be processed, that is, the object to be processed can be removed from the initial visual data to obtain first visual data containing the object to be processed and second visual data not containing the object to be processed. In practical applications, the image segmentation model can be a pre-trained machine learning model, a deep learning model, a convolutional neural network, etc. In another embodiment of this disclosure, the removal of the object to be processed can also be achieved using user interaction methods, such as using image editing tools such as the magic wand tool or the lasso tool.

[0071] In summary, by using object detection and image segmentation models to identify and segment the objects to be processed in the initial visual data, we can obtain first and second visual data, which can be used to form a visual data CAPTCHA for image restoration, providing a data foundation for the subsequent creation of high-quality visual data CAPTCHAs.

[0072] Step 204: Enhance the second visual data to obtain multiple third visual data.

[0073] In practical applications, before performing enhancement processing on the second visual data to obtain multiple third visual data, the process further includes:

[0074] If the second visual data meets the preset repair conditions, the second visual data is repaired to obtain the repaired second visual data.

[0075] The enhancement processing of the second visual data to obtain multiple third visual data includes:

[0076] The repaired second visual data is enhanced to obtain multiple third visual data.

[0077] The preset repair conditions can be understood as pre-set conditions requiring the repair of the second visual data. In one embodiment of this disclosure, the preset repair conditions may be that the image metrics of the second visual data reach a preset metric threshold. These image metrics may include, but are not limited to, peak signal-to-noise ratio, structural similarity index, feature loss, edge preservation index, color fidelity, texture similarity, and sharpness. When the image metrics reach the preset metric threshold, it indicates that the second visual data may have quality degradation, blurring, or missing data. In another embodiment of this disclosure, the preset repair conditions may also be that the user's evaluation score for the second visual data reaches a preset score threshold. Specifically, the user can manually determine whether the second visual data needs repair and evaluate it. For example, if the user's evaluation score for the second visual data is less than 50 points, it indicates that the second visual data meets the preset repair conditions, meaning the second visual data needs to be repaired. In another embodiment of this disclosure, all second visual data may also be repaired; however, this disclosure does not limit this approach.

[0078] In practical applications, once the second visual data meets the preset restoration conditions, an image restoration model can be used to restore the second visual data, obtaining restored second visual data. Subsequently, this restored second visual data can be enhanced to obtain multiple third visual data sets. The image restoration model can be a filter for removing image noise, a deconvolution algorithm for solving image blurring problems, or a pre-trained convolutional neural network, etc.

[0079] In summary, by performing image restoration on the second visual data, the problems of smearing and cutout marks that exist in the second visual data are avoided, thereby ensuring the high quality of the second visual data, and further ensuring the high quality of the third visual data generated based on the restored second visual data.

[0080] Furthermore, after performing repair processing on the second visual data to obtain the repaired second visual data, the process further includes:

[0081] Based on the first visual data and the repaired second visual data, a visual data verification code is constructed for security detection.

[0082] Specifically, the first visual data can be used as the answer, and the repaired second visual data can be used as the question to construct a visual data CAPTCHA. Security checks can then be performed based on this visual data CAPTCHA.

[0083] In summary, based on the first visual data and the repaired second visual data, a high-quality visual data CAPTCHA was constructed, which facilitates subsequent security testing based on this visual data CAPTCHA.

[0084] Further, the step of constructing a visual data verification code based on the first visual data and the repaired second visual data, and performing security detection based on the visual data verification code, includes:

[0085] Using the first visual data as the visual data to be moved and the repaired second visual data as the fixed visual data, a visual data verification code is constructed.

[0086] Based on the visual data verification code, a security detection is performed in response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data.

[0087] The visual data to be moved can be understood as the image that needs to be moved to a designated position during the security detection process, which can be understood as the answer to the visual data CAPTCHA. The fixed visual data can be understood as the background image that is fixed during the security detection process, which can be understood as the question of the visual data CAPTCHA. The designated position to which the visual data to be moved is located in the fixed visual data. Specifically, the designated position (i.e. the corresponding position) can be understood as the position of the object to be removed in the fixed visual data.

[0088] In practical applications, refer to Figure 3, which shows a schematic diagram of a visual data verification code in a visual data processing method according to an embodiment of the present disclosure. As shown in Figure 3, the first visual data is an image of a cup lid, and the second visual data is an image of a boy holding a cup. The first visual data can be overlaid on any position of the second visual data, or it can be placed outside the second visual data. In subsequent applications, the user needs to move the cup lid to a specified position in the second visual data using the mouse pointer. The image of the cup lid at the specified position can be understood as the second visual data obtained after being removed in step 202 above.

[0089] In summary, visual data CAPTCHAs are constructed to enable subsequent security checks.

[0090] In specific implementation, the enhancement processing of the second visual data to obtain multiple third visual data includes:

[0091] The second visual data is augmented to obtain intermediate visual data;

[0092] Based on the position information of the object to be processed in the initial visual data, the intermediate visual data is randomly cropped to obtain multiple third visual data.

[0093] The process of augmenting the second visual data can be understood as augmenting the pixels of the second visual data.

[0094] Specifically, the pixels of the second visual data can be expanded to obtain intermediate visual data, and the intermediate visual data can be randomly cropped to obtain multiple third visual data.

[0095] In one embodiment of this disclosure, when augmenting the second visual data, an image augmentation model can be used to augment the pixels of the second visual data. For example, second visual data with a pixel size of 300×300 can be augmented into intermediate visual data with a pixel size of 600×600. The image augmentation model may include, but is not limited to, pre-trained machine learning models, deep learning models, etc.

[0096] In another embodiment of this disclosure, when augmenting the second visual data, a portion of the second visual data can be redrawn to obtain multiple intermediate visual data, which are then used as multiple third visual data. For example, if the second visual data shows a cabinet filled with kitchen utensils, with each row containing different utensils (e.g., the first row contains pots, cups, and bowls, and the second row contains plates), a portion of the second visual data can be redrawn. For instance, the pots in the first row can be redrawn as canned goods, and the plates in the second row as cups. This yields multiple intermediate visual data, which can then be directly identified as multiple third visual data. Alternatively, after obtaining multiple intermediate visual data, random cropping or truncating operations can be performed on the intermediate visual data to obtain multiple third visual data.

[0097] In another embodiment of this disclosure, when augmenting the second visual data, color transformation processing can also be performed on the second visual data. For example, image indicators such as color, saturation, and brightness of the second visual data can be modified. Grayscale transformation can also be performed on the second visual data. Specifically, the grayscale values ​​of the second visual data can be linearly mapped, and the amplitude and direction of the color transformation can be controlled by adjusting the mapping parameters during the linear mapping process. Random noise can also be added to the second visual data, such as Gaussian noise, salt-and-pepper noise, etc. The addition of noise makes the model more robust to the influence of image noise.

[0098] Specifically, after expanding the second visual data to obtain intermediate visual data, the other regions in the intermediate visual data that are outside the position information of the object to be processed in the initial visual data can be randomly cropped to obtain multiple third visual data.

[0099] In practical applications, when performing random cropping on intermediate visual data, random cropping can be performed within a region that does not affect the position of the answer (i.e., the position of the object to be processed in the initial visual data).

[0100] In summary, by performing image augmentation and random cropping, a large amount of third visual data can be constructed using a small amount of second visual data, further improving the efficiency of visual data CAPTCHA question expansion. The generation cost of visual data CAPTCHA is low, and it only requires continuous random cropping of the second visual data, requiring only image editing operations and no need to run algorithm models, further reducing costs. By randomly cropping within the area that does not affect the answer position, the number of visual data CAPTCHA questions can be expanded at low cost.

[0101] Furthermore, after enhancing the second visual data to obtain multiple third visual data, the process further includes:

[0102] Based on the position information of the object to be processed in the initial visual data, determine the position to be edited in any one of the plurality of third visual data, excluding the position information;

[0103] The position to be edited is edited to obtain third visual data containing the editing area, wherein the editing area is determined based on the position to be edited;

[0104] The visual data verification code is constructed based on the first visual data and the third visual data containing the editing area.

[0105] In this context, the position information of the object to be processed in the initial visual data can be understood as the answer position, the position to be edited can be understood as other positions in the third visual data besides the answer position, and the editing area can be understood as the editing area obtained after editing the position to be edited.

[0106] Specifically, since some attacks can use modeling techniques to identify which areas of a visual data CAPTCHA have been edited, thus pinpointing the answer, to counter such attacks, additional editable areas can be added outside the actual answer location for obfuscation. Therefore, based on the actual answer location, the editable position in each third visual data set can be determined, and this editable position can be processed to obtain third visual data containing the edited area. Then, a visual data CAPTCHA can be constructed based on the first visual data and the third visual data containing the edited area.

[0107] Understandably, the same editing operations can be performed on the aforementioned repaired second visual data.

[0108] In practical applications, refer to Figure 4, which illustrates a schematic diagram of the editing region in a visual data processing method according to an embodiment of this disclosure. As shown in Figure 4, the third visual data includes the actual answer location (i.e., the position of the object to be processed in the initial visual data), as well as editable location 1 and editable location 2. Specifically, the editable locations can be edited and repaired based on an image restoration model to obtain third visual data containing the actual answer location, editable region 1, and editable region 2. Based on this, by erasing multiple editable regions, subsequent attack methods using model technology will be able to identify multiple editable regions, making it impossible to locate the answer and thus preventing the cracking of the visual data verification code, achieving security detection.

[0109] Step 206: Based on the first visual data and the plurality of third visual data, construct a plurality of visual data verification codes to perform security detection based on the visual data verification codes.

[0110] Specifically, the step of constructing multiple visual data verification codes based on the first visual data and the multiple third visual data includes:

[0111] Construct a visual data pair based on the first visual data and any one of the plurality of third visual data;

[0112] The visual data pairs are identified as the visual data verification codes, and multiple visual data verification codes are obtained.

[0113] Specifically, a visual data pair can be constructed by combining the first visual data with any one of the multiple third visual data. This visual data pair can serve as a visual data CAPTCHA. Therefore, it can be understood that the number of visual data CAPTCHAs can be the same as the number of third visual data. For example, five visual data CAPTCHAs can be constructed based on the first visual data and five third visual data.

[0114] Further, the step of constructing a visual data pair based on the first visual data and any one of the plurality of third visual data includes:

[0115] The first visual data is used as the visual data to be moved, and any one of the third visual data is used as the fixed visual data to construct a visual data pair, and the visual data pair is determined as the visual data verification code.

[0116] The security detection based on the visual data CAPTCHA includes:

[0117] Based on the visual data verification code, a security detection is performed in response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data.

[0118] Specifically, a visual data pair can be constructed by using the first visual data as the visual data to be moved and any third visual data as the fixed visual data. This visual data pair can then be used as a visual data verification code. In this case, when performing security checks based on the visual data verification code, the security check can be performed in response to the movement operation of moving the visual data to be moved to the corresponding position in the fixed visual data.

[0119] In addition, after obtaining multiple third-vision data, the third-vision data can be used as a base map according to the above steps to identify other objects to be processed in the third-vision data, and the above steps of object detection, segmentation processing, and enhancement processing can be continued to construct the visual data CAPTCHA.

[0120] In summary, the above method identifies the object to be processed within the initial visual data, segments the initial visual data based on this object, and obtains first visual data containing the object and second visual data excluding it. This enables image matting of the initial visual data. Furthermore, the second visual data is enhanced to obtain multiple third visual data, increasing the quantity of visual data and improving its image quality. Subsequently, a visual data CAPTCHA is constructed based on the first visual data and multiple third visual data, further enhancing both the quantity and quality of the CAPTCHA. Security checks are then performed based on this CAPTCHA to prevent it from being cracked during attacks, thus further ensuring the security of user access.

[0121] The following description, in conjunction with Figure 5, uses the application of the visual data processing method provided in this disclosure in image verification code generation as an example to further illustrate the visual data processing method. Figure 5 shows a flowchart of the processing procedure of a visual data processing method according to an embodiment of this disclosure, specifically including the following steps.

[0122] Step 502: Using the text-generated image model, generate the initial image (i.e., the base image) based on the prompt text.

[0123] Step 504: Using the recognition and segmentation model, the initial image is segmented according to the object recognition prompts to obtain a first image containing the object to be processed (i.e., cutout) and a second image not containing the object to be processed (i.e., background image).

[0124] Step 506: Use the large image restoration model to restore the second image and obtain the restored second image.

[0125] Step 508: Construct an image verification code based on the repaired second image and the first image.

[0126] Step 510: Using the image augmentation model, augment the second image obtained in step 504 to obtain an intermediate image.

[0127] Step 512: Perform random cropping on the intermediate image to obtain multiple third images.

[0128] Step 514: Construct an image verification code based on the first image and multiple third images.

[0129] The above method identifies the object to be processed within the initial image, segments the initial image based on the object to be processed, and obtains a first image containing the object to be processed and a second image not containing the object to be processed, thus performing image cutout processing on the initial image. Furthermore, the second image is enhanced to obtain multiple third images, increasing the number of images and improving the image quality of the second images through enhancement processing. Subsequently, an image CAPTCHA is constructed based on the first image and multiple third images, improving both the quantity and quality. Security detection is performed based on this image CAPTCHA to prevent it from being cracked during attacks, further ensuring the security of user access.

[0130] Corresponding to the above method embodiments, this disclosure also provides a visual data processing device embodiment. FIG6 shows a schematic diagram of the structure of a visual data processing device provided in one embodiment of this disclosure. As shown in FIG6, the device includes:

[0131] The segmentation module 602 is configured to determine the object to be processed contained in the initial visual data, and to segment the initial visual data according to the object to be processed to obtain first visual data containing the object to be processed and second visual data not containing the object to be processed.

[0132] Enhancement module 604 is configured to enhance the second visual data to obtain multiple third visual data.

[0133] The construction module 606 is configured to construct multiple visual data verification codes based on the first visual data and the multiple third visual data, so as to perform security detection based on the visual data verification codes.

[0134] In an optional embodiment, the enhancement module 604 is further configured to:

[0135] The second visual data is augmented to obtain intermediate visual data;

[0136] Based on the position information of the object to be processed in the initial visual data, the intermediate visual data is randomly cropped to obtain multiple third visual data.

[0137] In an optional embodiment, the building module 606 is further configured to:

[0138] Construct a visual data pair based on the first visual data and any one of the plurality of third visual data;

[0139] The visual data pairs are identified as the visual data verification codes, and multiple visual data verification codes are obtained.

[0140] In an optional embodiment, the building module 606 is further configured to:

[0141] The first visual data is used as the visual data to be moved, and any one of the third visual data is used as the fixed visual data to construct a visual data pair, and the visual data pair is determined as the visual data verification code.

[0142] Based on the visual data verification code, a security detection is performed in response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data.

[0143] In an optional embodiment, the enhancement module 604 is further configured to:

[0144] If the second visual data meets the preset repair conditions, the second visual data is repaired to obtain the repaired second visual data.

[0145] The repaired second visual data is enhanced to obtain multiple third visual data.

[0146] In an optional embodiment, the building module 606 is further configured to:

[0147] Based on the first visual data and the repaired second visual data, a visual data verification code is constructed for security detection.

[0148] In an optional embodiment, the building module 606 is further configured to:

[0149] Using the first visual data as the visual data to be moved and the repaired second visual data as the fixed visual data, a visual data verification code is constructed.

[0150] Based on the visual data verification code, a security detection is performed in response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data.

[0151] In an optional embodiment, the enhancement module 604 is further configured to:

[0152] Based on the position information of the object to be processed in the initial visual data, determine the position to be edited in any one of the plurality of third visual data, excluding the position information;

[0153] The position to be edited is edited to obtain third visual data containing the editing area, wherein the editing area is determined based on the position to be edited;

[0154] The visual data verification code is constructed based on the first visual data and the third visual data containing the editing area.

[0155] In an optional embodiment, the segmentation module 602 is further configured to:

[0156] Target detection is performed on the initial visual data to obtain the object to be processed contained in the initial visual data.

[0157] In an optional embodiment, the segmentation module 602 is further configured to:

[0158] Based on the text prompts, target detection is performed on the initial visual data to obtain the object to be processed contained in the initial visual data.

[0159] In an optional embodiment, the segmentation module 602 is further configured to:

[0160] The initial visual data is generated based on the prompt text for the initial visual data.

[0161] In summary, the aforementioned device determines the object to be processed contained in the initial visual data, segments the initial visual data based on the object to be processed, and obtains first visual data containing the object to be processed and second visual data not containing the object to be processed, thereby performing image matting processing on the initial visual data. Furthermore, it enhances the second visual data to obtain multiple third visual data, increasing the number of visual data and improving the image quality of the second visual data through enhancement processing. Subsequently, a visual data CAPTCHA constructed based on the first visual data and multiple third visual data further improves both the quantity and quality. Security detection is performed based on this visual data CAPTCHA to prevent it from being cracked during attacks, further ensuring the security of user access.

[0162] The above is an illustrative scheme of a visual data processing device according to this embodiment. It should be noted that the technical solution of this visual data processing device and the technical solution of the above-described visual data processing method belong to the same concept. For details not described in detail in the technical solution of the visual data processing device, please refer to the description of the technical solution of the above-described visual data processing method.

[0163] Referring to Figure 7, which shows a flowchart of a security detection method according to an embodiment of the present disclosure, the method includes the following steps.

[0164] Step 702: In response to a data processing request, display the visual data to be moved and the fixed visual data in the visual data verification code, wherein the visual data verification code is determined according to the visual data processing method provided in the first aspect of the present disclosure, the visual data to be moved is the first visual data, and the fixed visual data is the third visual data;

[0165] Step 704: In response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data, perform a security check on the data processing request.

[0166] Data processing requests can include network access requests and login requests, among others.

[0167] Specifically, this security detection method can be applied to the client side. The client side can respond to the data processing request by displaying a visual data verification code. The user can use the client's mouse pointer to move the visual data to be moved in the visual data verification code to the corresponding position in the fixed visual data. The client side can determine whether the user who initiated the data processing request is a human or a computer program based on this movement operation, thereby realizing the security detection of the data processing request.

[0168] It is understood that the visual data verification code in this security detection method is constructed based on the above-described visual data processing method, and this disclosure will not repeat the details.

[0169] The above is an illustrative scheme of a security detection method according to this embodiment. It should be noted that the technical solution of this security detection method and the technical solution of the above-described visual data processing method belong to the same concept. For details not described in detail in the technical solution of the security detection method, please refer to the description of the technical solution of the above-described visual data processing method.

[0170] Corresponding to the above method embodiments, this disclosure also provides a security detection device embodiment. Figure 8 shows a schematic diagram of the structure of a security detection device provided in one embodiment of this disclosure. As shown in Figure 8, the device includes:

[0171] The display module 802 is configured to display, in response to a data processing request, visual data to be moved and fixed visual data in a visual data verification code, wherein the visual data verification code is determined according to the visual data processing method provided in the first aspect of the present disclosure, the visual data to be moved is the first visual data, and the fixed visual data is the third visual data.

[0172] The detection module 804 is configured to perform security detection on the data processing request in response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data.

[0173] The above is a schematic diagram of a security detection device according to this embodiment. It should be noted that the technical solution of this security detection device and the technical solution of the security detection method described above belong to the same concept. For details not described in detail in the technical solution of the security detection device, please refer to the description of the technical solution of the security detection method described above.

[0174] Figure 9 shows a structural block diagram of a computing device 900 according to an embodiment of the present disclosure. The components of the computing device 900 include, but are not limited to, a memory 910 and a processor 920. The processor 920 is connected to the memory 910 via a bus 930, and a database 950 is used to store data.

[0175] The computing device 900 also includes an access device 940, which enables the computing device 900 to communicate via one or more networks 960. Examples of these networks include Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN), Personal Area Network (PAN), or combinations of communication networks such as the Internet. The access device 940 may include one or more of any type of wired or wireless network interface (e.g., a network interface controller (NIC)), such as an IEEE 802.11 Wireless Local Area Network (WLAN) interface, a Wi-MAX (Worldwide Interoperability for Microwave Access) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a Near Field Communication (NFC) interface, and so on.

[0176] In one embodiment of this application, the aforementioned components of the computing device 900, as well as other components not shown in FIG. 9, may be interconnected, for example, via a bus. It should be understood that the computing device structural block diagram shown in FIG. 9 is merely for illustrative purposes and is not intended to limit the scope of this application. Those skilled in the art can add or replace other components as needed.

[0177] The computing device 900 can be any type of stationary or mobile computing device, including mobile computers or mobile computing devices (e.g., tablet computers, personal digital assistants, laptop computers, notebook computers, netbooks, etc.), mobile phones (e.g., smartphones), wearable computing devices (e.g., smartwatches, smart glasses, etc.) or other types of mobile devices, or stationary computing devices such as desktop computers or personal computers (PCs). The computing device 900 can also be a mobile or stationary server.

[0178] The processor 920 is used to execute the following computer program / instructions, which, when executed by the processor, implement the steps of the above method.

[0179] The various embodiments in this disclosure are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the computing device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0180] An embodiment of this disclosure also provides a computer-readable storage medium storing a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0181] The various embodiments in this disclosure are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the computer-readable storage medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0182] An embodiment of this disclosure also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0183] The above is an illustrative scheme of a computer program product according to this embodiment. It should be noted that the technical solution of this computer program product and the technical solution of the above method belong to the same concept, and all details not described in detail in the technical solution of the computer program product can be referred to the description of the technical solution of the above method.

[0184] The foregoing has described specific embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0185] The computer instructions include computer program code, which may be in the form of source code, object code, executable file, or certain intermediate forms. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium may be appropriately added or removed according to the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media may not include electrical carrier signals and telecommunication signals.

[0186] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this disclosure are not limited to the described order of actions, because according to the embodiments of this disclosure, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the embodiments of this disclosure.

[0187] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0188] The preferred embodiments disclosed above are merely illustrative of this disclosure. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the embodiments of this disclosure. These embodiments are selected and specifically described in this disclosure to better explain the principles and practical applications of the embodiments of this disclosure, thereby enabling those skilled in the art to better understand and utilize this disclosure. This disclosure is limited only by the claims and their full scope and equivalents.

Claims

1. A visual data processing method, comprising: The objects to be processed contained in the initial visual data are determined, and the initial visual data is segmented according to the objects to be processed to obtain first visual data containing the objects to be processed and second visual data not containing the objects to be processed. The second visual data is enhanced to obtain multiple third visual data. Based on the first visual data and the plurality of third visual data, multiple visual data verification codes are constructed to perform security detection based on the visual data verification codes.

2. The method according to claim 1, wherein, The enhancement processing of the second visual data to obtain multiple third visual data includes: The second visual data is augmented to obtain intermediate visual data; Based on the position information of the object to be processed in the initial visual data, the intermediate visual data is randomly cropped to obtain multiple third visual data.

3. The method according to claim 1 or 2, wherein, The step of constructing multiple visual data verification codes based on the first visual data and the multiple third visual data includes: Construct a visual data pair based on the first visual data and any one of the plurality of third visual data; The visual data pairs are identified as the visual data verification codes, and multiple visual data verification codes are obtained.

4. The method according to claim 3, wherein, The step of constructing a visual data pair based on the first visual data and any one of the plurality of third visual data includes: The first visual data is used as the visual data to be moved, and any one of the third visual data is used as the fixed visual data to construct a visual data pair, and the visual data pair is determined as the visual data verification code. The security detection based on the visual data CAPTCHA includes: Based on the visual data verification code, a security detection is performed in response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data.

5. The method according to claim 1 or 2, wherein, Before performing enhancement processing on the second visual data to obtain multiple third visual data, the method further includes: If the second visual data meets the preset repair conditions, the second visual data is repaired to obtain the repaired second visual data. The enhancement processing of the second visual data to obtain multiple third visual data includes: The repaired second visual data is enhanced to obtain multiple third visual data.

6. The method according to claim 5, wherein, After performing repair processing on the second visual data to obtain the repaired second visual data, the process further includes: Based on the first visual data and the repaired second visual data, a visual data verification code is constructed for security detection.

7. The method according to claim 6, wherein, The step of constructing a visual data verification code based on the first visual data and the repaired second visual data, and performing security detection based on the visual data verification code, includes: Using the first visual data as the visual data to be moved and the repaired second visual data as the fixed visual data, a visual data verification code is constructed. Based on the visual data verification code, a security detection is performed in response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data.

8. The method according to any one of claims 1 or 2, wherein, After enhancing the second visual data to obtain multiple third visual data, the process further includes: Based on the position information of the object to be processed in the initial visual data, determine the position to be edited in any one of the plurality of third visual data, excluding the position information; The position to be edited is edited to obtain third visual data containing the editing area, wherein the editing area is determined based on the position to be edited; The visual data verification code is constructed based on the first visual data and the third visual data containing the editing area.

9. The method according to any one of claims 1 or 2, wherein, The process of determining the objects to be processed contained in the initial visual data includes: Target detection is performed on the initial visual data to obtain the object to be processed contained in the initial visual data.

10. The method according to claim 9, wherein, The step of performing target detection on the initial visual data to obtain the object to be processed contained in the initial visual data includes: Based on the text prompts, target detection is performed on the initial visual data to obtain the object to be processed contained in the initial visual data.

11. The method according to any one of claims 1 or 2, wherein, Before determining the objects to be processed contained in the initial visual data, the process also includes: The initial visual data is generated based on the prompt text for the initial visual data.

12. A security detection method, comprising: In response to a data processing request, a visual data verification code is displayed containing movable visual data and fixed visual data, wherein the visual data verification code is determined by the method according to any one of claims 1-11, the movable visual data is a first visual data, and the fixed visual data is a third visual data; In response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data, a security check is performed on the data processing request.

13. A computing device, comprising: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer programs / instructions are executed by the processor, the processor performs the following method: determining the object to be processed contained in the initial visual data, and segmenting the initial visual data according to the object to be processed to obtain first visual data containing the object to be processed and second visual data not containing the object to be processed. The second visual data is enhanced to obtain multiple third visual data. Based on the first visual data and the plurality of third visual data, multiple visual data verification codes are constructed to perform security detection based on the visual data verification codes.

14. A computer-readable storage medium storing a computer program / instructions that, when executed by a processor, cause the processor to perform the following method: determining an object to be processed contained in initial visual data, and segmenting the initial visual data according to the object to be processed to obtain first visual data containing the object to be processed and second visual data not containing the object to be processed; The second visual data is enhanced to obtain multiple third visual data. Based on the first visual data and the plurality of third visual data, multiple visual data verification codes are constructed to perform security detection based on the visual data verification codes.

15. The computer-readable storage medium according to claim 14, wherein, When the computer program / instruction is executed by the processor, the processor also performs the following method: augmenting the second visual data to obtain intermediate visual data; and randomly cropping the intermediate visual data according to the position information of the object to be processed in the initial visual data to obtain multiple third visual data.

16. The computer-readable storage medium according to claim 14 or 15, wherein, When the computer program / instruction is executed by the processor, the processor also performs the following method: constructing a visual data pair based on the first visual data and any one of the plurality of third visual data; determining the visual data pair as the visual data verification code, and obtaining a plurality of visual data verification codes.

17. The computer-readable storage medium of claim 16, wherein, When the computer program / instruction is executed by the processor, the processor also performs the following method: constructing a visual data pair by using the first visual data as the visual data to be moved and any one of the third visual data as the fixed visual data, and determining the visual data pair as the visual data verification code; the security detection based on the visual data verification code includes: performing security detection based on the visual data verification code in response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data.

18. The computer-readable storage medium according to claim 14 or 15, wherein, When the computer program / instruction is executed by the processor, the processor is also made to perform the following method: if it is determined that the second visual data meets the preset repair conditions, the second visual data is repaired to obtain the repaired second visual data; The step of enhancing the second visual data to obtain multiple third visual data includes: enhancing the repaired second visual data to obtain multiple third visual data.

19. The computer-readable storage medium according to claim 18, wherein, When the computer program / instruction is executed by the processor, the processor is also made to perform the following method: construct a visual data verification code based on the first visual data and the repaired second visual data, so as to perform security detection based on the visual data verification code.

20. The computer-readable storage medium according to claim 19, wherein, When the computer program / instruction is executed by the processor, the processor also performs the following method: using the first visual data as visual data to be moved and the repaired second visual data as fixed visual data, constructing a visual data verification code; and performing security detection based on the visual data verification code in response to a movement operation that moves the visual data to be moved to a corresponding position in the fixed visual data.

Images