CN110706221A - Verification method, verification device, storage medium and device for customizing pictures - Google Patents
Verification method, verification device, storage medium and device for customizing pictures Download PDFInfo
- Publication number
- CN110706221A CN110706221A CN201910938011.7A CN201910938011A CN110706221A CN 110706221 A CN110706221 A CN 110706221A CN 201910938011 A CN201910938011 A CN 201910938011A CN 110706221 A CN110706221 A CN 110706221A
- Authority
- CN
- China
- Prior art keywords
- picture
- verification
- style
- verified
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Abstract
The invention relates to the technical field of verification, and discloses a verification method, verification equipment, a storage medium and a device for customizing a picture. In the invention, a user interface UI style picture and a client brand picture are obtained; analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics; determining a corresponding picture to be verified based on the image style characteristics; and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity. Obviously, the invention customizes the picture to be verified through the analyzed image style, constructs the verification framework based on the picture to be verified, can effectively intercept the verification attack behavior implemented in the exhaustive form of massive pictures, greatly improves the verification safety, and solves the technical problem of lower safety of the verification mode.
Description
Technical Field
The present invention relates to the field of verification technologies, and in particular, to a verification method, a verification device, a storage medium, and an apparatus for customizing a picture.
Background
With the continuous development of authentication technology, there are various authentication methods for man-machine identification, i.e. whether the authentication behavior is a human-operated authentication behavior or a hacking-controlled authentication behavior. For example, short message verification and character filling verification with a picture as background, etc.
However, with the continuous development of Optical Character Recognition (OCR) technology, if a verification method involving pictures is adopted, hackers can easily implement verification attack behaviors in an exhaustive manner by using a large number of pictures, thereby greatly reducing the security of the verification process.
Therefore, the verification method has the technical problem of low safety.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide a verification method, verification equipment, a storage medium and a device for customizing a picture, and aims to solve the technical problem of low safety of a verification mode.
In order to achieve the above object, the present invention provides a method for verifying picture customization, which comprises the following steps:
acquiring a user interface UI style picture and a client brand picture;
analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics;
determining a corresponding picture to be verified based on the image style characteristics;
and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity.
Preferably, the analyzing the image style of the UI style picture and the customer brand picture by a preset convolutional neural network algorithm to obtain the image style characteristics specifically includes:
carrying out convolution operation on the UI style picture and the customer brand picture in a forward propagation mode through a preset convolution neural network algorithm to obtain a response matrix corresponding to each layer of neural network;
determining a degree of correlation between the response matrices within a feature layer;
and carrying out normalization processing on the correlation degree to obtain a texture matrix, and taking the texture matrix as an image style characteristic.
Preferably, the image style features comprise a gram matrix;
correspondingly, the determining a corresponding picture to be verified based on the image style characteristics specifically includes:
acquiring a first preset drawing library;
performing convolution operation on the image library images in the first preset image library through the preset convolution neural network algorithm to obtain a characteristic matrix;
calculating a sum of difference squares between the gram matrix and the feature matrix;
and performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as pictures to be verified.
Preferably, the style rendering is performed on the gallery picture in the first preset gallery based on the difference sum, and the rendered gallery picture is used as a picture to be verified, specifically including:
performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as initial rendering pictures;
counting style loss, content loss and pixel loss in the rendering process corresponding to the initial rendering picture;
and readjusting the rendering process according to the values of the style loss, the content loss and the pixel loss, and taking the re-rendered gallery picture as a picture to be verified.
Preferably, the determining a corresponding picture to be verified based on the image style characteristics specifically includes:
acquiring a second preset drawing library;
traversing the gallery pictures in the second preset gallery, and matching the traversed gallery pictures with the image style characteristics;
and taking the gallery picture successfully matched with the image style characteristics as a picture to be verified.
Preferably, the constructing a verification frame based on the picture to be verified and operating the verification frame to perform validity verification specifically include:
determining a region to be spliced in the picture to be verified;
removing the area image corresponding to the area to be spliced in the picture to be verified to obtain a display picture;
constructing a verification frame through the display picture and the area image corresponding to the area to be spliced;
and operating the verification frame to carry out validity verification by splicing the region image to the region to be spliced in the picture to be verified.
Preferably, the operating the verification framework to perform validity verification by splicing the region image to the region to be spliced in the picture to be verified specifically includes:
running the verification frame and acquiring a current splicing behavior, wherein the current splicing behavior is a behavior of splicing the region image to a current region in the picture to be verified;
extracting a behavior track from the current splicing behavior, and acquiring the region matching degree between the current region and the region to be spliced;
and carrying out validity verification according to the matching degree of the behavior track and the region.
Furthermore, to achieve the above object, the present invention further proposes an authentication apparatus, which includes a memory, a processor, and a picture customized authentication program stored on the memory and executable on the processor, the picture customized authentication program being configured to implement the steps of the picture customized authentication method as described above.
In addition, to achieve the above object, the present invention further provides a storage medium having a customized photo verification program stored thereon, wherein the customized photo verification program realizes the steps of the customized photo verification method as described above when being executed by a processor.
In order to achieve the above object, the present invention further provides a picture customization verification apparatus, including:
the picture acquisition module is used for acquiring a user interface UI style picture and a client brand picture;
the style analysis module is used for analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm so as to obtain image style characteristics;
the picture generation module is used for determining a corresponding picture to be verified based on the image style characteristics;
and the verification module is used for constructing a verification frame based on the picture to be verified and operating the verification frame to verify the validity.
In the invention, a user interface UI style picture and a client brand picture are obtained; analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics; determining a corresponding picture to be verified based on the image style characteristics; and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity. Obviously, the invention customizes the picture to be verified through the analyzed image style, constructs the verification framework based on the picture to be verified, can effectively intercept the verification attack behavior implemented in the exhaustive form of massive pictures, greatly improves the verification safety, and solves the technical problem of lower safety of the verification mode.
Drawings
FIG. 1 is a schematic diagram of a verification device of a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a verification method for customized pictures according to a first embodiment of the present invention;
FIG. 3 is a flowchart illustrating a second embodiment of a verification method for customized pictures according to the present invention;
FIG. 4 is a flowchart illustrating a third embodiment of a verification method for customized pictures according to the present invention;
fig. 5 is a block diagram of a first embodiment of the apparatus for verifying customized pictures according to the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a verification device of a hardware operating environment according to an embodiment of the present invention.
As shown in fig. 1, the authentication apparatus may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), the optional user interface 1003 may also include a standard wired interface and a wireless interface, and the wired interface of the user interface 1003 may be a Universal Serial Bus (USB) interface in the present invention. The network interface 1004 may optionally include a standard wired interface as well as a wireless interface (e.g., WI-FI interface). The Memory 1005 may be a high speed Random Access Memory (RAM); or a stable Memory, such as a Non-volatile Memory (Non-volatile Memory), and may be a disk Memory. The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the configuration shown in FIG. 1 does not constitute a limitation of the verification device, and may include more or fewer components than those shown, or some components in combination, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a kind of computer storage medium, may include therein an operating system, a network communication module, a user interface module, and a picture-customized authentication program.
In the authentication apparatus shown in fig. 1, the network interface 1004 is mainly used for connecting to a backend server and performing data communication with the backend server; the user interface 1003 is mainly used for connecting peripheral equipment; the authentication apparatus calls, by the processor 1001, an authentication program for picture customization stored in the memory 1005, and performs the following operations:
acquiring a user interface UI style picture and a client brand picture;
analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics;
determining a corresponding picture to be verified based on the image style characteristics;
and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity.
Further, the processor 1001 may call the verification program for picture customization stored in the memory 1005, and further perform the following operations:
carrying out convolution operation on the UI style picture and the customer brand picture in a forward propagation mode through a preset convolution neural network algorithm to obtain a response matrix corresponding to each layer of neural network;
determining a degree of correlation between the response matrices within a feature layer;
and carrying out normalization processing on the correlation degree to obtain a texture matrix, and taking the texture matrix as an image style characteristic.
Further, the processor 1001 may call the verification program for picture customization stored in the memory 1005, and further perform the following operations:
acquiring a first preset drawing library;
performing convolution operation on the image library images in the first preset image library through the preset convolution neural network algorithm to obtain a characteristic matrix;
calculating a sum of difference squares between the gram matrix and the feature matrix;
and performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as pictures to be verified.
Further, the processor 1001 may call the verification program for picture customization stored in the memory 1005, and further perform the following operations:
performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as initial rendering pictures;
counting style loss, content loss and pixel loss in the rendering process corresponding to the initial rendering picture;
and readjusting the rendering process according to the values of the style loss, the content loss and the pixel loss, and taking the re-rendered gallery picture as a picture to be verified.
Further, the processor 1001 may call the verification program for picture customization stored in the memory 1005, and further perform the following operations:
acquiring a second preset drawing library;
traversing the gallery pictures in the second preset gallery, and matching the traversed gallery pictures with the image style characteristics;
and taking the gallery picture successfully matched with the image style characteristics as a picture to be verified.
Further, the processor 1001 may call the verification program for picture customization stored in the memory 1005, and further perform the following operations:
determining a region to be spliced in the picture to be verified;
removing the area image corresponding to the area to be spliced in the picture to be verified to obtain a display picture;
constructing a verification frame through the display picture and the area image corresponding to the area to be spliced;
and operating the verification frame to carry out validity verification by splicing the region image to the region to be spliced in the picture to be verified.
Further, the processor 1001 may call the verification program for picture customization stored in the memory 1005, and further perform the following operations:
running the verification frame and acquiring a current splicing behavior, wherein the current splicing behavior is a behavior of splicing the region image to a current region in the picture to be verified;
extracting a behavior track from the current splicing behavior, and acquiring the region matching degree between the current region and the region to be spliced;
and carrying out validity verification according to the matching degree of the behavior track and the region.
In the embodiment, a user interface UI style picture and a client brand picture are obtained; analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics; determining a corresponding picture to be verified based on the image style characteristics; and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity. Obviously, the embodiment customizes the picture to be verified through the analyzed image style, constructs the verification framework based on the picture to be verified, can effectively intercept the verification attack behavior implemented in the exhaustive form of massive pictures, greatly improves the verification security, and solves the technical problem of low security of the verification mode.
Based on the above hardware structure, an embodiment of the verification method for picture customization of the present invention is provided.
Referring to fig. 2, fig. 2 is a schematic flow chart of a verification method for customizing a picture according to a first embodiment of the present invention.
In a first embodiment, the method for verifying customized pictures comprises the following steps:
step S10: and acquiring a UI style picture and a customer brand picture of the user interface.
It can be understood that the execution subject of this embodiment is an authentication device, and the authentication device may specifically be a server that provides an authentication service for a backend, and the user device performs validity authentication by accessing the authentication device and passing the authentication service. The user equipment may be a personal computer or a smart phone operated by a user.
It is noted that the present invention relates to a service party providing authentication service technology, a client party using the authentication service, and an individual user performing validity authentication through the authentication service, and the individual user can access a data resource or a network resource provided by the client party only after the individual user passes the validity authentication. The client side here may be a portal operator or a video website operator, etc. Note that the client side and individual users referred to herein are distinguished.
In specific implementation, the embodiment can customize the gallery to enrich the verification pictures of the client side and effectively prevent hackers from performing verification attacks in an exhaustive form by using a large number of pictures. Specifically, a User Interface (UI) style picture and a client brand picture may be obtained first, where the UI style picture includes a UI picture externally displayed by a client, for example, a picture material of an operation page provided by the client to a website visitor; the client brand picture comprises a brand picture which is externally displayed by a client, such as a trademark typeface, a trademark component element, a cartoon element typeface element and a main tone element of a common advertisement and the like related to the client.
Step S20: and analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics.
It should be appreciated that after the UI style picture and the customer brand picture are obtained, the image style can be summarized by presetting a convolutional neural network algorithm and characterizing the image style in the form of image style features.
In a specific implementation, the preset convolutional neural network algorithm can be a convolutional neural network algorithm based on a Tensorflow platform, and the Tensorflow platform is an artificial intelligence platform supporting a cyclic neural network.
Step S30: and determining a corresponding picture to be verified based on the image style characteristics.
It will be appreciated that the image style characteristics obtained based on the predetermined convolutional neural network algorithm may be characterized as a type of matrix from which a customized picture may be generated for the customer. The generated picture to be verified is more fit with the image style of the client side, and if the picture to be verified is used for providing verification service, verification attack behaviors which are implemented in a massive picture exhaustion mode can be effectively stopped. After all, the picture to be verified applied at this time has certain specificity, and is not easy to obtain by hackers.
Step S40: and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity.
In specific implementation, there are various implementation modes of the verification frame, and the verification mode related to the picture, such as the presence of a click-type verification frame, is a background, the picture to be verified is used as the background, a plurality of single characters are displayed on the background, and the user clicks the single characters according to a prompt sequence to complete validity verification; the user can drag the small part to the blank area of the picture to be verified, from which the image is originally removed, to restore the blank area to the original state of the picture, thereby completing the validity verification.
In the embodiment, a user interface UI style picture and a client brand picture are obtained; analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics; determining a corresponding picture to be verified based on the image style characteristics; and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity. Obviously, the embodiment customizes the picture to be verified through the analyzed image style, constructs the verification framework based on the picture to be verified, can effectively intercept the verification attack behavior implemented in the exhaustive form of massive pictures, greatly improves the verification security, and solves the technical problem of low security of the verification mode.
Referring to fig. 3, fig. 3 is a flowchart illustrating a second embodiment of the verification method for customizing a picture according to the present invention, and the second embodiment of the verification method for customizing a picture according to the present invention is proposed based on the first embodiment illustrated in fig. 2.
In the second embodiment, the step S20 specifically includes:
step S201: and carrying out convolution operation on the UI style picture and the customer brand picture in a forward propagation mode through a preset convolution neural network algorithm so as to obtain a response matrix corresponding to each layer of neural network.
It will be appreciated that in order to analyse the image style, a convolution operation will be performed first. Specifically, after the UI style picture and the customer brand picture are input, a corresponding response matrix is generated through each layer of neural network.
Step S202: a degree of correlation between the response matrices is determined within a feature layer.
It should be appreciated that after obtaining multiple response matrices, to determine the textural nature of the UI style picture in common with the customer brand picture, a degree of correlation between the different response matrices will be calculated.
Step S203: and carrying out normalization processing on the correlation degree to obtain a texture matrix, and taking the texture matrix as an image style characteristic.
It will be appreciated that by normalizing the correlation, the common textural property is characterized in the form of a matrix, and a texture matrix is obtained. The texture matrix is an image style characteristic and is used for representing the image style of a client side.
Further, the image style feature includes a Gram Matrix (Gram Matrix), in other words, the texture Matrix may exist in the form of a Gram Matrix.
Correspondingly, the step S30 specifically includes:
step S301: and acquiring a first preset gallery.
In a specific implementation, as for a picture generation manner for generating the customized picture on the client side through the matrix, a picture rendering manner may be adopted. Moreover, the client side customized picture is generated by drawing up a specific rendering mode, so that the difficulty of verification and solution can be further improved.
It can be understood that a first preset gallery can be obtained, and the first preset gallery is a preset collected picture material which is difficult to crack.
Step S302: and carrying out convolution operation on the image library images in the first preset image library through the preset convolution neural network algorithm to obtain a characteristic matrix.
It should be appreciated that in order to obtain a picture to be verified that is similar in style to the client-side image, the picture material in the first preset gallery may be stylized. For example, a convolution operation may be performed on a gallery picture to obtain a feature matrix representing the gallery picture.
Step S303: and calculating the sum of the difference squares between the gram matrix and the feature matrix.
It will be appreciated that the sum of the differences between the matrices may characterize the degree of similarity between the picture material within the first preset gallery and the image style characteristics of the client side.
Step S304: and performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as pictures to be verified.
It should be understood that style rendering may be performed on the gallery pictures in the first preset gallery based on the similarity degree, so that the similarity degree between the feature matrix corresponding to the rendered gallery pictures and the image style features of the client side is higher, and the rendered gallery pictures with higher similarity degree may be used as the pictures to be verified. If the generated picture to be verified is used for validity verification, the difficulty of cracking is further improved.
Further, style rendering is performed on the gallery picture in the first preset gallery based on the difference sum, and the rendered gallery picture is used as a picture to be verified, which specifically includes:
performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as initial rendering pictures;
counting style loss, content loss and pixel loss in the rendering process corresponding to the initial rendering picture;
and readjusting the rendering process according to the values of the style loss, the content loss and the pixel loss, and taking the re-rendered gallery picture as a picture to be verified.
It can be understood that, in order to reduce the data loss caused by the style rendering process, the present embodiment also provides a secondary adjustment scheme of the style rendering process.
In a specific implementation, for the secondary adjustment scheme, after performing primary style rendering on a gallery picture in a first preset gallery based on a difference sum, the gallery picture after the primary rendering is recorded as an initial rendering picture, and data loss in the primary style rendering process is counted, where the data loss may relate to style loss, content loss, pixel loss, and the like.
It should be understood that the rendering process may be readjusted based on the data loss, and secondary style rendering may be performed based on the adjusted rendering process to obtain a secondarily rendered gallery picture and record the secondarily rendered gallery picture as a picture to be verified. And, the data loss during the rescaled rendering is less than the data loss during the primary style rendering. In addition, the gallery picture after the secondary style rendering is smoother than the initially rendered picture, because the readjusted rendering process can take the mean square error between adjacent pixels into account and control the mean square error within a certain numerical range.
In the embodiment, the picture to be verified with the similar style to that of the image on the client side can be obtained by stylizing the picture material in the first preset picture base, so that the difficulty of cracking verification is further improved.
Referring to fig. 4, fig. 4 is a flowchart illustrating a third embodiment of the verification method for customizing a picture according to the present invention, and the third embodiment of the verification method for customizing a picture according to the present invention is proposed based on the first embodiment illustrated in fig. 2.
In a third embodiment, the determining a corresponding picture to be verified based on the image style characteristics specifically includes:
acquiring a second preset drawing library;
traversing the gallery pictures in the second preset gallery, and matching the traversed gallery pictures with the image style characteristics;
and taking the gallery picture successfully matched with the image style characteristics as a picture to be verified.
It can be understood that, as for the picture generation manner for generating the customized picture on the client side by the matrix, besides the implementation by the picture rendering manner mentioned in the second embodiment of the verification method for customizing a picture according to the present invention, the picture generation manner can also be directly selected by the style matching manner.
In a specific implementation, the first preset gallery and the second preset gallery may be the same gallery or different galleries, and are only named for distinction. The gallery pictures of the second preset gallery can be traversed, the picture features of the traversed gallery pictures can be matched with the image style features, and the gallery pictures corresponding to the picture features with higher matching degree are determined as the gallery pictures which are successfully matched.
The feature types of the picture features of the gallery picture may include corner features, hue features, combination features, and the like. The combination features are high-order features, and angular point features, tone features, texture features and the like can be integrated. Similarly, the feature types of the image style features may also include corner features, hue features, combination features, and the like.
Further, the step S40 specifically includes:
step S401: and determining a region to be spliced in the picture to be verified.
In a specific implementation, a splicing type verification framework may be adopted, for example, a small circular area may be defined in the center of a stylized square to-be-verified picture, and the circular area may be an area to be spliced.
Step S402: and removing the area image corresponding to the area to be spliced in the picture to be verified so as to obtain a display picture.
It can be understood that, removing the area image corresponding to the circular area from the center of the square picture forms a remaining square picture in which the circular area is blank, and the remaining square picture is a display picture.
Step S403: and constructing a verification frame through the display picture and the area image corresponding to the area to be spliced.
Step S404: and operating the verification frame to carry out validity verification by splicing the region image to the region to be spliced in the picture to be verified.
It should be appreciated that the authentication framework can be operated to authenticate an individual user who, after successful authentication, has normal access to data resources, network resources, video resources, etc. on the client side. When the validity is verified, the personal user can drag the area image to the blank circular area of the rest square picture through the mouse at the personal computer side, so that the area image is matched with the blank circular area. If they match, the verification is deemed to be successful.
In addition, the removed area image can be displayed at any position of the verification frame, and the dragging operation is provided for the individual user.
Further, the operating the verification framework to perform validity verification by splicing the region image to the region to be spliced in the picture to be verified specifically includes:
running the verification frame and acquiring a current splicing behavior, wherein the current splicing behavior is a behavior of splicing the region image to a current region in the picture to be verified;
extracting a behavior track from the current splicing behavior, and acquiring the region matching degree between the current region and the region to be spliced;
and carrying out validity verification according to the matching degree of the behavior track and the region.
In a specific implementation, as for the determination method of the validity verification, in addition to determining whether the area image matches the blank circular area, the behavior of the operation of the individual user may also be determined. For example, in the process of stitching the image of the stitching area to the blank circular area, a mouse connected to the user equipment or a touch screen of the user equipment itself is used, and a behavior track in the stitching behavior exists. The behavior track refers to a dragging track of a mouse dragged by an individual user or a touch track of the touch screen touched by the individual user.
It can be understood that, when the validity verification is determined, it may be determined that the current region and the region to be spliced are completely matched, where the region matching degree is 100% if the current region and the region to be spliced are completely matched, or it may be determined whether the behavior trajectory is a machine movement trajectory, where the machine movement trajectory is greatly different from a human movement trajectory, for example, the human movement trajectory generally has a certain jitter property.
The current area is the final area where the individual user moves the area image, and ideally, the current area is matched with the area to be spliced.
It should be understood that if the behavior track is not the machine movement track and the area matching degree is high, the verification may be considered to be successful; otherwise, the verification is determined to be failed.
Further, after the verification frame is constructed based on the picture to be verified and the verification frame is operated for validity verification, the customized picture verification method further includes:
monitoring for abnormal behavior while running the verification framework;
and if the abnormal behavior is monitored, initiating secondary verification on the verification equipment side corresponding to the abnormal behavior.
It can be understood that if abnormal data is monitored in the verification process, verification is temporarily added again to improve the safety of the verification.
Further, in the process of analyzing the image style, the voice information of the activity of the client can be introduced. The introduction information of the relevant contents such as customer products, strategies and the like is recorded in the customer activity voice information, and the introduction information can be converted into characters. Then, when the image style of the UI style picture and the brand picture of the client is analyzed, the character content can be simultaneously combined, and the image style characteristic is obtained.
The embodiment relates to a splicing type verification mode, and enriches the application modes of the pictures to be verified; meanwhile, the collection and analysis of the behavior track are introduced, and the difficulty of program cracking is further improved.
In addition, an embodiment of the present invention further provides a storage medium, where the storage medium stores a customized verification program for a picture, and when executed by a processor, the customized verification program for a picture implements the following operations:
acquiring a user interface UI style picture and a client brand picture;
analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics;
determining a corresponding picture to be verified based on the image style characteristics;
and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity.
Further, the customized photo verification program when executed by the processor further performs the following operations:
carrying out convolution operation on the UI style picture and the customer brand picture in a forward propagation mode through a preset convolution neural network algorithm to obtain a response matrix corresponding to each layer of neural network;
determining a degree of correlation between the response matrices within a feature layer;
and carrying out normalization processing on the correlation degree to obtain a texture matrix, and taking the texture matrix as an image style characteristic.
Further, the customized photo verification program when executed by the processor further performs the following operations:
acquiring a first preset drawing library;
performing convolution operation on the image library images in the first preset image library through the preset convolution neural network algorithm to obtain a characteristic matrix;
calculating a sum of difference squares between the gram matrix and the feature matrix;
and performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as pictures to be verified.
Further, the customized photo verification program when executed by the processor further performs the following operations:
performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as initial rendering pictures;
counting style loss, content loss and pixel loss in the rendering process corresponding to the initial rendering picture;
and readjusting the rendering process according to the values of the style loss, the content loss and the pixel loss, and taking the re-rendered gallery picture as a picture to be verified.
Further, the customized photo verification program when executed by the processor further performs the following operations:
acquiring a second preset drawing library;
traversing the gallery pictures in the second preset gallery, and matching the traversed gallery pictures with the image style characteristics;
and taking the gallery picture successfully matched with the image style characteristics as a picture to be verified.
Further, the customized photo verification program when executed by the processor further performs the following operations:
determining a region to be spliced in the picture to be verified;
removing the area image corresponding to the area to be spliced in the picture to be verified to obtain a display picture;
constructing a verification frame through the display picture and the area image corresponding to the area to be spliced;
and operating the verification frame to carry out validity verification by splicing the region image to the region to be spliced in the picture to be verified.
Further, the customized photo verification program when executed by the processor further performs the following operations:
running the verification frame and acquiring a current splicing behavior, wherein the current splicing behavior is a behavior of splicing the region image to a current region in the picture to be verified;
extracting a behavior track from the current splicing behavior, and acquiring the region matching degree between the current region and the region to be spliced;
and carrying out validity verification according to the matching degree of the behavior track and the region.
In the embodiment, a user interface UI style picture and a client brand picture are obtained; analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics; determining a corresponding picture to be verified based on the image style characteristics; and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity. Obviously, the embodiment customizes the picture to be verified through the analyzed image style, constructs the verification framework based on the picture to be verified, can effectively intercept the verification attack behavior implemented in the exhaustive form of massive pictures, greatly improves the verification security, and solves the technical problem of low security of the verification mode.
In addition, referring to fig. 5, an embodiment of the present invention further provides a verification apparatus for customizing a picture, where the verification apparatus for customizing a picture includes:
and the picture acquisition module 10 is used for acquiring the UI style picture of the user interface and the brand picture of the client.
It is to be understood that the authentication device may specifically be a server providing an authentication service for the backend, and the user device performs the validity authentication by accessing the authentication device and passing the authentication service. The user equipment may be a personal computer or a smart phone operated by a user.
It is noted that the present invention relates to a service party providing authentication service technology, a client party using the authentication service, and an individual user performing validity authentication through the authentication service, and the individual user can access a data resource or a network resource provided by the client party only after the individual user passes the validity authentication. The client side here may be a portal operator or a video website operator, etc. Note that the client side and individual users referred to herein are distinguished.
In specific implementation, the embodiment can customize the gallery to enrich the verification pictures of the client side and effectively prevent hackers from performing verification attacks in an exhaustive form by using a large number of pictures. Specifically, a User Interface (UI) style picture and a client brand picture may be obtained first, where the UI style picture includes a UI picture externally displayed by a client, for example, a picture material of an operation page provided by the client to a website visitor; the client brand picture comprises a brand picture which is externally displayed by a client, such as a trademark typeface, a trademark component element, a cartoon element typeface element and a main tone element of a common advertisement and the like related to the client.
And the style analysis module 20 is configured to perform image style analysis on the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain an image style characteristic.
It should be appreciated that after the UI style picture and the customer brand picture are obtained, the image style can be summarized by presetting a convolutional neural network algorithm and characterizing the image style in the form of image style features.
In a specific implementation, the preset convolutional neural network algorithm can be a convolutional neural network algorithm based on a Tensorflow platform, and the Tensorflow platform is an artificial intelligence platform supporting a cyclic neural network.
And the picture generation module 30 is configured to determine a corresponding picture to be verified based on the image style characteristics.
It will be appreciated that the image style characteristics obtained based on the predetermined convolutional neural network algorithm may be characterized as a type of matrix from which a customized picture may be generated for the customer. The generated picture to be verified is more fit with the image style of the client side, and if the picture to be verified is used for providing verification service, verification attack behaviors which are implemented in a massive picture exhaustion mode can be effectively stopped. After all, the picture to be verified applied at this time has certain specificity, and is not easy to obtain by hackers.
And the verification module 40 is used for constructing a verification frame based on the picture to be verified and operating the verification frame to verify the validity.
In specific implementation, there are various implementation modes of the verification frame, and the verification mode related to the picture, such as the presence of a click-type verification frame, is a background, the picture to be verified is used as the background, a plurality of single characters are displayed on the background, and the user clicks the single characters according to a prompt sequence to complete validity verification; the user can drag the small part to the blank area of the picture to be verified, from which the image is originally removed, to restore the blank area to the original state of the picture, thereby completing the validity verification.
In the embodiment, a user interface UI style picture and a client brand picture are obtained; analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics; determining a corresponding picture to be verified based on the image style characteristics; and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity. Obviously, the embodiment customizes the picture to be verified through the analyzed image style, constructs the verification framework based on the picture to be verified, can effectively intercept the verification attack behavior implemented in the exhaustive form of massive pictures, greatly improves the verification security, and solves the technical problem of low security of the verification mode.
In an embodiment, the style analysis module 20 is further configured to perform convolution operation on the UI style picture and the customer brand picture in a forward propagation manner through a preset convolution neural network algorithm to obtain a response matrix corresponding to each layer of neural network; determining a degree of correlation between the response matrices within a feature layer; and carrying out normalization processing on the correlation degree to obtain a texture matrix, and taking the texture matrix as an image style characteristic.
In an embodiment, the image generating module 30 is further configured to obtain a first preset gallery; performing convolution operation on the image library images in the first preset image library through the preset convolution neural network algorithm to obtain a characteristic matrix; calculating a sum of difference squares between the gram matrix and the feature matrix; and performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as pictures to be verified.
In an embodiment, the picture generating module 30 is further configured to perform style rendering on the gallery pictures in the first preset gallery based on the difference sum, and use the rendered gallery pictures as initial rendering pictures; counting style loss, content loss and pixel loss in the rendering process corresponding to the initial rendering picture; and readjusting the rendering process according to the values of the style loss, the content loss and the pixel loss, and taking the re-rendered gallery picture as a picture to be verified.
In an embodiment, the image generating module 30 is further configured to obtain a second preset gallery; traversing the gallery pictures in the second preset gallery, and matching the traversed gallery pictures with the image style characteristics; and taking the gallery picture successfully matched with the image style characteristics as a picture to be verified.
In an embodiment, the verification module 40 is further configured to determine an area to be spliced in the picture to be verified; removing the area image corresponding to the area to be spliced in the picture to be verified to obtain a display picture; constructing a verification frame through the display picture and the area image corresponding to the area to be spliced; and operating the verification frame to carry out validity verification by splicing the region image to the region to be spliced in the picture to be verified.
In an embodiment, the verification module 40 is further configured to run the verification framework and obtain a current splicing behavior, where the current splicing behavior is a behavior of splicing the region image to a current region in the to-be-verified image; extracting a behavior track from the current splicing behavior, and acquiring the region matching degree between the current region and the region to be spliced; and carrying out validity verification according to the matching degree of the behavior track and the region.
Other embodiments or specific implementation manners of the image customization verification apparatus according to the present invention may refer to the above method embodiments, and are not described herein again.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third, etc. do not denote any order, but rather the words first, second, third, etc. are to be interpreted as names.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory, a RAM, a magnetic disk, and an optical disk), and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A verification method for picture customization is characterized by comprising the following steps:
acquiring a user interface UI style picture and a client brand picture;
analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm to obtain image style characteristics;
determining a corresponding picture to be verified based on the image style characteristics;
and constructing a verification frame based on the picture to be verified, and operating the verification frame to verify the validity.
2. The method for verifying picture customization according to claim 1, wherein the analyzing the UI style picture and the customer brand picture for the image style by a preset convolutional neural network algorithm to obtain the image style characteristics specifically comprises:
carrying out convolution operation on the UI style picture and the customer brand picture in a forward propagation mode through a preset convolution neural network algorithm to obtain a response matrix corresponding to each layer of neural network;
determining a degree of correlation between the response matrices within a feature layer;
and carrying out normalization processing on the correlation degree to obtain a texture matrix, and taking the texture matrix as an image style characteristic.
3. The method of validating a customization of pictures according to claim 2, wherein the image style features include a gram matrix;
correspondingly, the determining a corresponding picture to be verified based on the image style characteristics specifically includes:
acquiring a first preset drawing library;
performing convolution operation on the image library images in the first preset image library through the preset convolution neural network algorithm to obtain a characteristic matrix;
calculating a sum of difference squares between the gram matrix and the feature matrix;
and performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as pictures to be verified.
4. The method for verifying picture customization according to claim 3, wherein the style rendering of the gallery picture in the first preset gallery based on the difference sum, and taking the rendered gallery picture as a picture to be verified specifically include:
performing style rendering on the gallery pictures in the first preset gallery based on the difference sum, and taking the rendered gallery pictures as initial rendering pictures;
counting style loss, content loss and pixel loss in the rendering process corresponding to the initial rendering picture;
and readjusting the rendering process according to the values of the style loss, the content loss and the pixel loss, and taking the re-rendered gallery picture as a picture to be verified.
5. The method for verifying picture customization according to claim 1, wherein the determining of the corresponding picture to be verified based on the image style characteristics specifically comprises:
acquiring a second preset drawing library;
traversing the gallery pictures in the second preset gallery, and matching the traversed gallery pictures with the image style characteristics;
and taking the gallery picture successfully matched with the image style characteristics as a picture to be verified.
6. The method for verifying customized pictures according to any one of claims 1 to 5, wherein the constructing a verification framework based on the picture to be verified and operating the verification framework for validity verification specifically comprises:
determining a region to be spliced in the picture to be verified;
removing the area image corresponding to the area to be spliced in the picture to be verified to obtain a display picture;
constructing a verification frame through the display picture and the area image corresponding to the area to be spliced;
and operating the verification frame to carry out validity verification by splicing the region image to the region to be spliced in the picture to be verified.
7. The method for verifying customized pictures according to claim 6, wherein the running of the verification framework for performing validity verification by stitching the region image to the region to be stitched in the picture to be verified specifically comprises:
running the verification frame and acquiring a current splicing behavior, wherein the current splicing behavior is a behavior of splicing the region image to a current region in the picture to be verified;
extracting a behavior track from the current splicing behavior, and acquiring the region matching degree between the current region and the region to be spliced;
and carrying out validity verification according to the matching degree of the behavior track and the region.
8. An authentication apparatus, characterized in that the authentication apparatus comprises: memory, a processor and a verification program stored on the memory and executable on the processor for customizing a picture, the verification program when executed by the processor implementing the steps of the method for verifying the customization of the picture according to any one of claims 1 to 7.
9. A storage medium having a picture customization validation program stored thereon, the picture customization validation program when executed by a processor implementing the steps of the picture customization validation method according to any one of claims 1 to 7.
10. A picture customization verification apparatus, comprising:
the picture acquisition module is used for acquiring a user interface UI style picture and a client brand picture;
the style analysis module is used for analyzing the image style of the UI style picture and the customer brand picture through a preset convolutional neural network algorithm so as to obtain image style characteristics;
the picture generation module is used for determining a corresponding picture to be verified based on the image style characteristics;
and the verification module is used for constructing a verification frame based on the picture to be verified and operating the verification frame to verify the validity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910938011.7A CN110706221A (en) | 2019-09-29 | 2019-09-29 | Verification method, verification device, storage medium and device for customizing pictures |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910938011.7A CN110706221A (en) | 2019-09-29 | 2019-09-29 | Verification method, verification device, storage medium and device for customizing pictures |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110706221A true CN110706221A (en) | 2020-01-17 |
Family
ID=69197473
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910938011.7A Pending CN110706221A (en) | 2019-09-29 | 2019-09-29 | Verification method, verification device, storage medium and device for customizing pictures |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110706221A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI770947B (en) * | 2021-04-20 | 2022-07-11 | 國立清華大學 | Verification method and verification apparatus based on attacking image style transfer |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846274A (en) * | 2018-04-09 | 2018-11-20 | 腾讯科技（深圳）有限公司 | A kind of safe verification method, device and terminal |
| CN108959303A (en) * | 2017-05-22 | 2018-12-07 | 阿里巴巴集团控股有限公司 | A kind of exhibiting pictures generate, layout generation method and data processing server |
| CN109800559A (en) * | 2019-01-02 | 2019-05-24 | 平安科技（深圳）有限公司 | Generation method, device, computer equipment and the storage medium of sliding block identifying code |
-
2019
- 2019-09-29 CN CN201910938011.7A patent/CN110706221A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108959303A (en) * | 2017-05-22 | 2018-12-07 | 阿里巴巴集团控股有限公司 | A kind of exhibiting pictures generate, layout generation method and data processing server |
| CN108846274A (en) * | 2018-04-09 | 2018-11-20 | 腾讯科技（深圳）有限公司 | A kind of safe verification method, device and terminal |
| CN109800559A (en) * | 2019-01-02 | 2019-05-24 | 平安科技（深圳）有限公司 | Generation method, device, computer equipment and the storage medium of sliding block identifying code |
Non-Patent Citations (2)
| Title |
|---|
| RANJITHA KUMAR ET.AL: "Bricolage: Example-Based Retargeting for Web Design", 《ACM》 * |
| YUNKE ZHANG ET.AL: "Layout Style Modeling for Automating Banner Design", 《THEMATIC WORKSHOPS’17》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI770947B (en) * | 2021-04-20 | 2022-07-11 | 國立清華大學 | Verification method and verification apparatus based on attacking image style transfer |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106453209B (en) | Identity verification method and device | |
| CN109194689B (en) | Abnormal behavior recognition method, device, server and storage medium | |
| CN109409349B (en) | Credit certificate authentication method, credit certificate authentication device, credit certificate authentication terminal and computer readable storage medium | |
| CN110780789B (en) | Game application starting method and device, storage medium and electronic device | |
| CN106656944B (en) | Method and device for sliding verification of handheld mobile equipment | |
| CN104361285B (en) | The safety detection method and device of mobile device application program | |
| CN109547426B (en) | Service response method and server | |
| CN109391620B (en) | Method, system, server and storage medium for establishing abnormal behavior judgment model | |
| CN109413047B (en) | Behavior simulation judgment method, behavior simulation judgment system, server and storage medium | |
| CN110866239A (en) | Verification code request processing method, device, equipment and computer storage medium | |
| JP2018504681A (en) | Method, apparatus, system, storage medium, program, and computer apparatus for providing authentication information on a web page | |
| CN105740670A (en) | Application encryption method and device, and application startup method and device | |
| CN106789973B (en) | Page security detection method and terminal equipment | |
| CN109299592B (en) | Man-machine behavior characteristic boundary construction method, system, server and storage medium | |
| CN113326045B (en) | Interface code generation method based on design file | |
| CN109284590B (en) | Method, equipment, storage medium and device for access behavior security protection | |
| CN108418797B (en) | Webpage access method and device, computer equipment and storage medium | |
| CN110795706B (en) | Hash-based verification method, equipment, storage medium and device | |
| CN107231358B (en) | Questionnaire data acquisition method, server and mobile terminal | |
| CN110706221A (en) | Verification method, verification device, storage medium and device for customizing pictures | |
| CN111090849A (en) | Memory, verification code implementation method, device and equipment | |
| CN106161388B (en) | Information verification method, server and system | |
| CN112307464A (en) | Fraud identification method and device and electronic equipment | |
| CN108182355B (en) | Login verification method, server and computer readable storage medium | |
| US20220414193A1 (en) | Systems and methods for secure adaptive illustrations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200117 |