Data desensitization method, terminal, device and storage medium

Abstract

The present invention discloses a data desensitization method, comprising the following steps: performing image segmentation on a text image based on a preset image segmentation algorithm to obtain a set of image fragments corresponding to the text image, hiding the attribute information of each image fragment in the image fragment set, deforming each image fragment to obtain a deformed image fragment set, encrypting the file name of each image fragment in the deformed image fragment set to obtain an encrypted image fragment set, grouping the image fragments in the encrypted image fragment set, and sending each group to different preset terminals. The present invention also discloses a terminal, a device, and a storage medium. By performing image segmentation on a text image to obtain image fragments, then deforming, encrypting the file name, and grouping the image fragments, and sending the groups to different preset terminals, desensitization processing of the text image is achieved, and data security when internal and external terminals interact with text images is effectively protected.

Description

Technical Field

[0001] The present invention relates to the field of machine learning technology, and in particular to a data desensitization method, terminal, device and readable storage medium. Background Art

[0002] In banks and other financial institutions, a lot of user data is very sensitive and needs to be kept confidential, such as the user's ID number, the user's vehicle model, and many of the user's transaction data. This data is often displayed on pictures, such as the user's ID card photo, vehicle registration certificate photo, invoice photo, etc. Therefore, these pictures containing sensitive text information are also sensitive data.

[0003] In the field of machine learning, data annotation is the starting point for machines to perceive the real world. To some extent, unlabeled data is useless data. When we train some models based on text images, such as character recognition (OCR), we often need to annotate a large number of single-line text images. This requires us to find a large number of annotators to manually annotate these images. Compared with the massive amount of text image data, internal data annotation is very inefficient, so in many cases, these data need to be handed over to external specialized annotation companies for annotation. However, due to the sensitivity of text images, directly transmitting text images to external annotation companies will cause the leakage of customer security data or some commercial sensitive data.

[0004] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of the present invention is to provide a data desensitization method, terminal, device and readable storage medium, aiming to solve the technical problem of privacy data leakage caused by the lack of desensitization of pictures when text and pictures are interacted between existing internal and external terminals.

[0006] To achieve the above object, the present invention provides a data desensitization method, which is applied to a data desensitization terminal. The data desensitization method comprises the following steps:

[0007] Segment the text image based on a preset image segmentation algorithm to obtain a set of image segments corresponding to the text image;

[0008] Hiding the attribute information of each picture segment in the picture segment set, and performing deformation processing on each picture segment to obtain a deformed picture segment set;

[0009] Encrypting the file names of the various image segments in the deformed image segment set to obtain an encrypted image segment set;

[0010] The encrypted picture segments are concentrated into groups, and each group is sent to different preset terminals respectively.

[0011] Furthermore, the step of performing image segmentation on the text image based on a preset image segmentation algorithm to obtain a set of image segments corresponding to the text image includes:

[0012] Obtaining preset filtering parameters, and performing filtering processing on the text image based on the filtering parameters to obtain a filtered text image;

[0013] Compare the text image with the filtered text image to obtain a pure text image;

[0014] Scanning the plain text image pixel by pixel based on a threshold value and determining an image segmentation line;

[0015] The text image is segmented based on the image segmentation line to obtain a set of image segments corresponding to the text image.

[0016] Furthermore, the step of hiding the attribute information of each picture segment in the picture segment set includes:

[0017] The attribute information of each picture segment is obtained, and each attribute information is randomly rewritten to hide the association relationship between the attribute information of each picture segment.

[0018] Furthermore, the step of deforming each image segment to obtain a deformed image segment set includes:

[0019] Get the preset deformation parameter set;

[0020] A deformation process is performed on each image segment based on the preset deformation parameter set, wherein the deformation process is to randomly obtain a deformation parameter value from the deformation parameter set, and scale the image segment based on the deformation parameter value.

[0021] Furthermore, the step of encrypting the file name of each picture segment in the deformed picture segment set to obtain the encrypted picture segment set includes:

[0022] A preset encryption algorithm is obtained, and based on the encryption algorithm, a multi-level encryption is performed on the file name of each picture segment in the deformed picture segment set to obtain an encrypted picture segment set.

[0023] Furthermore, the step of grouping the encrypted picture fragments in the picture fragment set includes:

[0024] Disrupting the order of the picture segments in the encrypted picture segment set to obtain a new picture segment set;

[0025] Acquire a preset grouping parameter, and group the picture segments in the new picture segment set based on the grouping parameter;

[0026] Each group is sent to a different preset terminal.

[0027] Furthermore, after the step of grouping the encrypted picture segment sets and sending each grouped segment set to different preset terminals, the method further includes:

[0028] Upon receiving the file name and annotation information comparison table sent by each preset terminal, decrypting each file name in each file name and annotation information comparison table respectively to obtain the decrypted file name of each file name, wherein after receiving the group segment set, the preset terminal performs data annotation processing on the group segment set to generate the file name and annotation information comparison table;

[0029] Obtain a sorted list corresponding to the text image, and based on the sorted list and the decrypted file name, sort all the annotation information in the file name and annotation information comparison table to obtain the target annotation information corresponding to the text image.

[0030] Furthermore, the data desensitization device includes:

[0031] A segmentation module, used to segment the text image based on a preset image segmentation algorithm to obtain a set of image segments corresponding to the text image;

[0032] A deformation module, used for hiding the attribute information of each picture segment in the picture segment set, and performing deformation processing on each picture segment to obtain a deformed picture segment set;

[0033] An encryption module, used for encrypting the file name of each picture segment in the deformed picture segment set to obtain an encrypted picture segment set;

[0034] The grouping sending module is used to group the encrypted picture fragments into groups and send each group to different preset terminals respectively.

[0035] In addition, to achieve the above-mentioned purpose, the present invention also provides a data desensitizing terminal, which includes a memory, a processor, and a data desensitizing program stored in the memory and executable on the processor, and the data desensitizing program, when executed by the processor, implements the steps of the data desensitizing method described above.

[0036] In addition, to achieve the above-mentioned purpose, the present invention also provides a readable storage medium, on which a data desensitization program is stored, and when the data desensitization program is executed by a processor, the steps of the data desensitization method described above are implemented.

[0037] The present invention performs image segmentation on a text image based on a preset image segmentation algorithm to obtain a set of image fragments corresponding to the text image, and then hides the attribute information of each image fragment in the image fragment set, and deforms each image fragment to obtain a deformed image fragment set, then encrypts the file name of each image fragment in the deformed image fragment set to obtain an encrypted image fragment set, and finally groups the image fragments in the encrypted image fragment set, and sends each group to different preset terminals. By performing image segmentation on a text image to obtain image fragments, and then deforming, encrypting the file name, and grouping the image fragments, and sending different groups to different preset terminals, desensitization processing of the text image is achieved, and data security when internal and external terminals interact with text images is effectively protected. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the structure of a terminal system in a hardware operating environment involved in an embodiment of the present invention;

[0039] Figure 2 This is a flow chart of the first embodiment of the data desensitization method of the present invention;

[0040] Figure 3 This is a flow chart of a second embodiment of the data desensitization method of the present invention;

[0041] Figure 4 This is a schematic diagram of the functional modules of an embodiment of the data desensitization device of the present invention.

[0042] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0043] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0044] like Figure 1 As shown, Figure 1 It is a structural diagram of the hardware operating environment involved in the embodiment of the present invention.

[0045] It should be noted that Figure 1 The data desensitization device of the embodiment of the present invention can be a PC, or a terminal device with display function such as a smart phone, smart TV, tablet computer, portable computer, etc.

[0046] like Figure 1As shown, the data desensitization terminal may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or it may be a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.

[0047] Those skilled in the art will understand that Figure 1 The system structure shown in the figure does not constitute a limitation on the terminal system, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0048] like Figure 1 As shown, the memory 1005 as a readable storage medium may include an operating system, a network communication module, a user interface module, and a data desensitization program.

[0049] exist Figure 1 In the system shown, the network interface 1004 is mainly used to connect to the background server and communicate data with the background server; the user interface 1003 is mainly used to connect to the client (client) and communicate data with the client; and the processor 1001 can be used to call the data desensitization program stored in the memory 1005.

[0050] In this embodiment, the terminal system includes: a memory 1005, a processor 1001, and a data desensitization program stored on the memory 1005 and executable on the processor 1001, wherein when the processor 1001 calls the data desensitization program stored in the memory 1005, the steps of the data desensitization method provided in each embodiment of the present application are executed.

[0051] The present invention also provides a data desensitization method, referring to Figure 2 , Figure 2 This is a flow chart of the first embodiment of the data desensitization method of the present invention.

[0052] The embodiments of the present invention provide embodiments of the data desensitization method. It should be noted that although the logical order is shown in the flow chart, in some cases, the steps shown or described may be performed in an order different from that shown here.

[0053] In this embodiment, the data desensitization method includes:

[0054] Step S10, performing image segmentation on the text image based on a preset image segmentation algorithm to obtain a set of image segments corresponding to the text image;

[0055] In this embodiment, data desensitization refers to the deformation of certain sensitive information through desensitization rules to achieve reliable protection of sensitive privacy data, which is one of the data security technologies. In the case of customer security data or some commercial sensitive data, real data is transformed and provided for testing without violating system rules. For example, personal information such as ID card number, mobile phone number, card number, customer number, etc. need to be desensitized.

[0056] As mentioned above, training some text-image-based models, such as text recognition (OCR), often requires labeling a large number of single-line text images, which requires us to find a large number of labelers to manually label these images. Compared with the massive amount of text image data, internal data labeling efficiency is very low, so in many cases, these data need to be handed over to external specialized labeling companies for labeling. However, due to the sensitivity of text images, directly transmitting text images to external labeling companies will cause the leakage of customer security data or some commercial sensitive data.

[0057] Taking the above defects into consideration, in this embodiment, a data desensitization method is proposed, which performs image segmentation on the text image to obtain image fragments, then deforms the image fragments, encrypts the file names, groups them, and sends the groups to different preset terminals, thereby achieving desensitization of the text image and effectively protecting the data security when the internal and external terminals interact with the text image. It should be noted that if the text image involves multiple lines of text, the text between different lines needs to be aligned up and down to avoid cutting the text in different lines into two halves when the image is cut.

[0058] Specifically, step S10 includes:

[0059] Step S11, obtaining preset filtering parameters, and filtering the text image based on the filtering parameters to obtain a filtered text image;

[0060] Step S12, comparing the text image with the filtered text image to obtain a pure text image;

[0061] Step S13, performing pixel scanning on the plain text image based on a threshold value and determining an image segmentation line;

[0062] Step S14, segmenting the text image based on the image segmentation line to obtain a set of image segments corresponding to the text image.

[0063] In this embodiment, the text image is first filtered, and the filtering parameters are set according to the type of the text image. The text image is filtered according to the filtering parameters to obtain the filtered text image. The specific filtering process is: obtain the grayscale image of the text image, and then perform filtering operations on the grayscale image, mainly corrosion and expansion, wherein the corrosion operation is described as: scan each pixel of the image, and use the structure element to perform an "AND" operation with the binary image covered by it: if both are 1, the pixel of the result image is 1, otherwise it is 0; the expansion operation is described as: scan each pixel of the image, and use the structure element to perform an "AND" operation with the binary image covered by it: if both are 0, the pixel of the result image is 0, otherwise it is 1. The corrosion operation is to select the minimum value of the difference between the image value and the structure element value in the neighborhood block determined by the structure element, and the expansion operation is to select the maximum value of the sum of the image value and the structure element value in the neighborhood block determined by the structure element. Since corrosion and expansion are both existing technologies, the specific processing process is not described in detail.

[0064] Furthermore, each pixel of the filtered text image is divided by each pixel corresponding to the original text image to change the grayscale of the text image, thereby identifying the text area in the image and obtaining a pure text image, that is, the area other than the text is basically a white area.

[0065] Then, the plain text image is pixel-scanned, the pixel values ​​corresponding to the pixels in the image are read, and the pixel values ​​greater than the threshold are determined as text, and the pixel values ​​less than the threshold are determined as text intervals. The number of characters that each image segment can contain is determined, and then the image segmentation line is determined at the corresponding text interval. Finally, the text image is segmented along the image segmentation line to obtain multiple image segments.

[0066] Step S20, hiding the attribute information of each picture segment in the picture segment set, and performing deformation processing on each picture segment to obtain a deformed picture segment set;

[0067] In this embodiment, the attribute information of the image segments after the image segmentation may contain the same or related information, such as the same creation time. The image segments can be easily restored to the original text image based on the same or similar attributes, so the attribute information of the image segments needs to be hidden. Similarly, the image segments are deformed to prevent the original text image from being restored based on the size or size of the image segments.

[0068] Specifically, step S20 includes:

[0069] Step S21, obtaining attribute information of each picture segment, and randomly rewriting each attribute information to hide the association relationship between the attribute information of each picture segment.

[0070] In this embodiment, the attribute information of the image fragment can be randomly rewritten, and the attribute information does not affect the content of the image fragment. The rewriting of the attribute information is a link in the desensitization process of this application, which is equivalent to hiding the attribute information, and can prevent the image fragment from being restored to the original text image based on the same or similar attribute information, thereby improving the desensitization quality of the text image. The specific content of the random rewriting is determined according to the actual situation and is not limited in this embodiment.

[0071] Step S22, obtaining a preset deformation parameter set;

[0072] Step S23, performing deformation processing on each image segment based on the preset deformation parameter set, wherein the deformation processing is to randomly obtain a deformation parameter value from the deformation parameter set, and perform scaling processing on the image segment based on the deformation parameter value.

[0073] In this embodiment, each image fragment is deformed to different degrees, so that the height, distortion, etc. of each image fragment after the same text image is cut are not necessarily the same. This can ensure that the person who gets the image fragment cannot regularly find the image fragments corresponding to the same original text image through these image attributes, and thus cannot restore the original text image.

[0074] Get a preset deformation parameter set, which includes many deformation parameter values. Traverse all image fragments and perform deformation processing on each image fragment. The specific process of deformation processing is to randomly obtain a deformation parameter value in the deformation parameter set. This is to achieve that the deformation parameters of each image fragment are not exactly the same. Then, the image fragment is scaled according to the deformation parameter value, so that the height and distortion of each image fragment are different.

[0075] Step S30, encrypting the file name of each picture segment in the deformed picture segment set to obtain an encrypted picture segment set;

[0076] In this embodiment, after the attribute hiding and deformation processing is performed on each picture segment in the picture segment set, the file name of each picture segment in the picture segment set is further encrypted.

[0077] Specifically, step S30 includes: obtaining a preset encryption algorithm, and performing multi-level encryption on the file name of each picture segment in the deformed picture segment set based on the encryption algorithm to obtain an encrypted picture segment set.

[0078] In this embodiment, for each segmented image segment, a preset encryption algorithm is used to encrypt the file name of the image segment, and multiple encryption algorithms are used in combination to achieve multi-level encryption of the file name of the image segment. For example, the file name of the image segment is encrypted by using a combination of high-strength one-way and two-way encryption algorithms recognized by the industry, such as the AES algorithm, so that the encrypted file name has sufficient randomness, so that the person who gets the image segment cannot obtain the association between the image segments through the name of the image segment, and thus cannot restore the original text image.

[0079] Step S40: group the encrypted picture segments into groups, and send each group to different preset terminals respectively.

[0080] In this embodiment, after the attributes of each picture segment in the picture segment set are hidden and deformed, the file names of the picture segments are encrypted, and then the picture segments are further grouped, and each group is sent to a different preset terminal.

[0081] Specifically, step S40 includes:

[0082] Step S41, disrupting the order of the picture segments in the encrypted picture segment set to obtain a new picture segment set;

[0083] Step S42, obtaining a preset grouping parameter, and grouping the picture segments in the new picture segment set based on the grouping parameter;

[0084] Step S43, sending each group to different preset terminals respectively.

[0085] In this embodiment, the order of the picture segments in the encrypted picture segment set is firstly disrupted to obtain a new picture segment set; then a grouping parameter is obtained, wherein the grouping parameter may be the number of groups or the number of picture segments to be included in each group, and the picture segments in the new picture segment set are grouped according to the grouping parameter. For example, if there are 8 picture segments in the new picture segment set and the obtained grouping parameter is 4 groups, the 8 picture segments may be evenly divided into 4 groups, each including 2 picture segments; or if the obtained grouping parameter is that each group includes 4 picture segments, the 8 picture segments may be evenly divided into 2 groups, each including 4 picture segments. Finally, each group is sent to different preset terminals respectively.

[0086] The data desensitization method proposed in this embodiment performs image segmentation on the text image based on a preset image segmentation algorithm to obtain a set of image fragments corresponding to the text image, and then hides the attribute information of each image fragment in the image fragment set, and deforms each image fragment to obtain a deformed image fragment set, then encrypts the file name of each image fragment in the deformed image fragment set to obtain an encrypted image fragment set, and finally groups the image fragments in the encrypted image fragment set, and sends each group to different preset terminals. By performing image segmentation on the text image to obtain image fragments, and then deforming, encrypting the file name, and grouping the image fragments, and sending different groups to different preset terminals, desensitization processing of the text image is achieved, and data security when internal and external terminals interact with text images is effectively protected.

[0087] Based on the first embodiment, refer to Figure 4 , a second embodiment of the data processing method of the present invention is proposed. In this embodiment, after step S40, the method further includes:

[0088] Step S50, when receiving the file name and annotation information comparison table sent by each preset terminal, decrypt each file name in each file name and annotation information comparison table to obtain the decrypted file name of each file name, wherein after receiving the group segment set, the preset terminal performs data annotation processing on the group segment set to generate the file name and annotation information comparison table;

[0089] Step S60, obtaining a sorted list corresponding to the text image, and sorting all the annotation information in the file name and annotation information comparison table based on the sorted list and the decrypted file name to obtain the target annotation information corresponding to the text image.

[0090] In this embodiment, the text image needs to be sent to the preset terminal for data annotation processing, and the text image is segmented to obtain a set of image fragments. Then, the attributes of each image fragment are hidden and deformed, and the file name of the image fragment is encrypted. Finally, the image fragments are grouped, and the text image is desensitized. The image is then sent to the preset terminal for data annotation, which effectively protects the data security when internal and external terminals interact with text images.

[0091] Furthermore, after receiving the grouped fragment set, the preset terminal performs data annotation processing on the grouped fragment set to generate a file name and annotation information comparison table, and the preset terminal sends the file name and annotation information comparison table to the data desensitizing terminal; when the data desensitizing terminal receives the file name and annotation information comparison table sent by each preset terminal, it decrypts each file name in each file name and annotation information comparison table respectively to obtain the decrypted file name of each file name, which file name is the original file name of the picture fragment.

[0092] Next, obtain the sorted list corresponding to the text image. The sorted table stores the correspondence between the original file name of the image fragment and the position of the image fragment. According to the sorted list and the decrypted file name, sort all the annotation information in the file name and annotation information comparison table to obtain the target annotation information corresponding to the text image.

[0093] The data desensitization method proposed in this embodiment, after receiving the file name and annotation information comparison table sent by each preset terminal, decrypts each file name in each file name and annotation information comparison table to obtain the decrypted file name of each file name, and obtains the sorted list corresponding to the text image, and sorts all the annotation information in the file name and annotation information comparison table based on the sorted list and the decrypted file name, so as to obtain the target annotation information corresponding to the text image. In the text image standard process, the text image is desensitized, which effectively protects the data security when the internal and external terminals interact with the text image.

[0094] The present invention further provides a data processing device, referring to Figure 4 , Figure 4 This is a schematic diagram of the functional modules of an embodiment of the data desensitization device of the present invention.

[0095] A segmentation module 10 is used to segment the text image based on a preset image segmentation algorithm to obtain a set of image segments corresponding to the text image;

[0096] A deformation module 20, used for hiding the attribute information of each picture segment in the picture segment set, and performing deformation processing on each picture segment to obtain a deformed picture segment set;

[0097] An encryption module 30 is used to encrypt the file name of each picture segment in the deformed picture segment set to obtain an encrypted picture segment set;

[0098] The grouping sending module 40 is used to group the encrypted picture segment set into picture segments, and send each group to different preset terminals respectively.

[0099] Furthermore, the segmentation module 10 is also used for:

[0100] Obtaining preset filtering parameters, and performing filtering processing on the text image based on the filtering parameters to obtain a filtered text image;

[0101] Compare the text image with the filtered text image to obtain a pure text image;

[0102] Scanning the plain text image pixel by pixel based on a threshold value and determining an image segmentation line;

[0103] The text image is segmented based on the image segmentation line to obtain a set of image segments corresponding to the text image.

[0104] Furthermore, the deformation module 20 is also used for:

[0105] The attribute information of each picture segment is obtained, and each attribute information is randomly rewritten to hide the association relationship between the attribute information of each picture segment.

[0106] Furthermore, the deformation module 20 is also used for:

[0107] Get the preset deformation parameter set;

[0108] A deformation process is performed on each image segment based on the preset deformation parameter set, wherein the deformation process is to randomly obtain a deformation parameter value from the deformation parameter set, and scale the image segment based on the deformation parameter value.

[0109] Furthermore, the encryption module 30 is also used for:

[0110] A preset encryption algorithm is obtained, and based on the encryption algorithm, a multi-level encryption is performed on the file name of each picture segment in the deformed picture segment set to obtain an encrypted picture segment set.

[0111] Furthermore, the packet sending module 40 is also used for:

[0112] Disrupting the order of the picture segments in the encrypted picture segment set to obtain a new picture segment set;

[0113] Acquire a preset grouping parameter, and group the picture segments in the new picture segment set based on the grouping parameter;

[0114] Each group is sent to a different preset terminal.

[0115] Furthermore, the data desensitization device also includes:

[0116] a decryption module, configured to, upon receiving the file name and annotation information comparison table sent by each preset terminal, decrypt each file name in each file name and annotation information comparison table to obtain the decrypted file name of each file name, wherein after receiving the group segment set, the preset terminal performs data annotation processing on the group segment set to generate the file name and annotation information comparison table;

[0117] The restoration module is used to obtain the sorted list corresponding to the text image, sort all the annotation information in the file name and annotation information comparison table based on the sorted list and the decrypted file name, and obtain the target annotation information corresponding to the text image.

[0118] In addition, an embodiment of the present invention further proposes a readable storage medium, on which a data desensitization program is stored. When the data desensitization program is executed by a processor, the steps of the data desensitization method in the above-mentioned embodiments are implemented.

[0119] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0120] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0121] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for enabling a system device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0122] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A data desensitization method, characterized in that: Applied to the data desensitization terminal, the data desensitization method includes: Segment the text image based on a preset image segmentation algorithm to obtain a set of image segments corresponding to the text image; Acquire attribute information of each image segment, randomly rewrite each attribute information to hide the association relationship between the attribute information of each image segment, and acquire a preset deformation parameter set, wherein the preset deformation parameter set includes a plurality of deformation parameter values; Performing deformation processing on each image segment based on the preset deformation parameter set, wherein the deformation processing is to randomly obtain a deformation parameter value from the deformation parameter set, and perform scaling processing on the image segment based on the deformation parameter value; Encrypting the file names of the various image segments in the deformed image segment set to obtain an encrypted image segment set; The encrypted image fragments are grouped into groups, and each group is sent to different preset terminals respectively; Upon receiving the file name and annotation information comparison table sent by each preset terminal, decrypting each file name in each file name and annotation information comparison table respectively to obtain the decrypted file name of each file name, wherein after receiving the group segment set, the preset terminal performs data annotation processing on the group segment set to generate the file name and annotation information comparison table; Obtain a sorted list corresponding to the text image, and based on the sorted list and the decrypted file name, sort all the annotation information in the file name and annotation information comparison table to obtain the target annotation information corresponding to the text image.

2. The data desensitization method according to claim 1, characterized in that: The step of segmenting the text image based on a preset image segmentation algorithm to obtain a set of image segments corresponding to the text image comprises: Obtaining preset filtering parameters, and performing filtering processing on the text image based on the filtering parameters to obtain a filtered text image; Compare the text image with the filtered text image to obtain a pure text image; Scanning the plain text image pixel by pixel based on a threshold value and determining an image segmentation line; The text image is segmented based on the image segmentation line to obtain a set of image segments corresponding to the text image.

3. The data desensitization method according to claim 1, characterized in that: The step of encrypting the file names of the various picture segments in the deformed picture segment set to obtain the encrypted picture segment set comprises: A preset encryption algorithm is obtained, and based on the encryption algorithm, a multi-level encryption is performed on the file name of each picture segment in the deformed picture segment set to obtain an encrypted picture segment set.

4. The data desensitization method according to claim 1, characterized in that: The step of grouping the encrypted picture fragments into picture fragments comprises: Disrupting the order of the picture segments in the encrypted picture segment set to obtain a new picture segment set; Acquire a preset grouping parameter, and group the picture segments in the new picture segment set based on the grouping parameter; Each group is sent to a different preset terminal.

5. A data desensitization device, characterized in that: The data desensitization device comprises: A segmentation module, used to segment the text image based on a preset image segmentation algorithm to obtain a set of image segments corresponding to the text image; A deformation module, used to obtain attribute information of each image segment, randomly rewrite each attribute information to hide the association relationship between the attribute information of each image segment, and obtain a preset deformation parameter set, wherein the preset deformation parameter set includes a plurality of deformation parameter values; perform deformation processing on each image segment based on the preset deformation parameter set, wherein the deformation processing is to randomly obtain a deformation parameter value in the deformation parameter set, and scale the image segment based on the deformation parameter value; An encryption module, used for encrypting the file name of each picture segment in the deformed picture segment set to obtain an encrypted picture segment set; A grouping sending module, used for grouping the encrypted picture segments into groups, and sending each group to different preset terminals respectively; The data desensitization device also includes: Upon receiving the file name and annotation information comparison table sent by each preset terminal, decrypting each file name in each file name and annotation information comparison table respectively to obtain the decrypted file name of each file name, wherein after receiving the group segment set, the preset terminal performs data annotation processing on the group segment set to generate the file name and annotation information comparison table; Obtain a sorted list corresponding to the text image, and based on the sorted list and the decrypted file name, sort all the annotation information in the file name and annotation information comparison table to obtain the target annotation information corresponding to the text image.

6. A data desensitization terminal, characterized in that: The data desensitizing terminal includes a memory, a processor, and a data desensitizing program stored in the memory and executable on the processor. When the data desensitizing program is executed by the processor, the steps of the data desensitizing method according to any one of claims 1 to 4 are implemented.

7. A readable storage medium, characterized in that: The readable storage medium stores a data desensitization program, and when the data desensitization program is executed by the processor, the steps of the data desensitization method according to any one of claims 1 to 4 are implemented.

Images