An information processing method, device, computer device, and storage medium

Abstract

Embodiments of the present application disclose an information processing method and device, computer equipment and a storage medium, belonging to the technical field of computers. The embodiments of the present application can acquire a to-be-processed image, perform text detection on the to-be-processed image to obtain at least one text content in the to-be-processed image, perform associated processing on the to-be-processed image and the text content to obtain associated relationship information between the text content and the to-be-processed image, encrypt the text content in the to-be-processed image using the text content in a preset image to obtain at least one encrypted text content, send the encrypted text content to a data processing device, acquire a processing result of processing the encrypted text content from the data processing device, and determine a target processing result corresponding to the text content from the processing result based on the associated relationship information, thereby improving the security of information.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to an information processing method and device, computer equipment and storage medium. BACKGROUND

[0002] With the continuous development of information technology and Internet technology, informatization has become a major trend of today's era. In the information age, information has become an important resource, and different information has different values and social influences. In many information processing application scenarios, there are often some information with high sensitivity, for example, the terms of a certain business contract, etc. In the prior art, there is often no desensitization or encryption processing for sensitive information, which threatens information security and easily leads to information leakage, reducing the security of information. SUMMARY

[0003] The embodiments of the present application provide an information processing method, device, computer equipment and storage medium, which can improve the security of information.

[0004] The embodiments of the present application provide an information processing method, comprising:

[0005] Obtaining a to-be-processed image;

[0006] Performing text detection on the to-be-processed image to obtain at least one text content in the to-be-processed image;

[0007] Performing association processing on the to-be-processed image and the text content to obtain association relationship information between the text content and the to-be-processed image;

[0008] Determining a first initial storage address of the preset image text content in a preset storage space, and determining a second initial storage address of the to-be-processed image text content in the preset storage space;

[0009] Calculating the first initial storage address by using a preset rearrangement logic to obtain a first rearranged storage address of the to-be-processed image text content, and calculating the second initial storage address by using the preset rearrangement logic to obtain a second rearranged storage address of the preset image text content;

[0010] Updating the storage position of the to-be-processed image text content in the preset storage space according to the first rearranged storage address, and updating the storage position of the preset image text content in the preset storage space according to the second rearranged storage position to obtain the encrypted text content;

[0011] send the encrypted text content to the data processing device, and obtain a processing result of processing the encrypted text content from the data processing device;

[0012] determine a target processing result corresponding to the text content from the processing result based on the association relationship information.

[0013] Correspondingly, the embodiment of the application further provides an information processing apparatus, comprising:

[0014] an obtaining unit configured to obtain an image to be processed;

[0015] a text detecting unit configured to perform text detection on the image to be processed to obtain at least one text content in the image to be processed;

[0016] an associating unit configured to perform association processing on the image to be processed and the text content to obtain association relationship information between the text content and the image to be processed;

[0017] an encrypting unit configured to determine a first initial storage address of a preset image text content in a preset storage space and a second initial storage address of the text content of the image to be processed in the preset storage space, perform calculation on the first initial storage address by using a preset rearrangement logic to obtain a first rearranged storage address of the text content of the image to be processed, perform calculation on the second initial storage address by using the preset rearrangement logic to obtain a second rearranged storage address of the preset image text content, update a storage position of the text content of the image to be processed in the preset storage space according to the first rearranged storage address, and update a storage position of the preset image text content in the preset storage space according to the second rearranged storage position to obtain the encrypted text content;

[0018] a result obtaining unit configured to send the encrypted text content to the data processing device, and obtain a processing result of processing the encrypted text content from the data processing device;

[0019] a result determining unit configured to determine a target processing result corresponding to the text content from the processing result based on the association relationship information.

[0020] In an embodiment, the text detecting unit can comprise:

[0021] a denoising subunit configured to perform denoising processing on the image to be processed to obtain a denoised image;

[0022] a region detecting subunit configured to detect at least one text region in the denoised image;

[0023] The text recognition subunit is configured to recognize the at least one text region to obtain the at least one text content.

[0024] In an embodiment, the denoising subunit can include:

[0025] The positioning module is configured to perform target positioning on the to-be-processed image to locate target position information from the to-be-processed image.

[0026] The cropping module is configured to crop an effective information region image from the to-be-processed image based on the target position information.

[0027] The correction module is configured to perform correction processing on the effective information region image to obtain the denoised image.

[0028] In an embodiment, the positioning module can include:

[0029] The feature extraction sub-module is configured to perform feature extraction on the to-be-processed image at multiple different scales to obtain feature information of the to-be-processed image at the multiple different scales.

[0030] The full connection sub-module is configured to perform full connection processing on the feature information at the multiple different scales to obtain an effective information region in the to-be-processed image.

[0031] The position recognition sub-module is configured to perform position recognition on the effective information region to obtain the target position information.

[0032] In an embodiment, the text recognition subunit can include:

[0033] The feature extraction module is configured to perform feature extraction on the text region to obtain feature information of the text region.

[0034] The reshaping module is configured to perform reshaping processing on the feature information to obtain reshaped feature information.

[0035] The mapping module is configured to map the reshaped feature information to a preset text probability space to obtain a text mapping probability corresponding to the reshaped feature information.

[0036] The text content determination module is configured to determine the text content based on the text mapping probability.

[0037] In an embodiment, the result determination unit can include:

[0038] The reading subunit is configured to read the association relationship information to obtain identification information corresponding to the text content.

[0039] The matching unit is configured to match the processing identifier of the processing result with the identification information to obtain a matching result.

[0040] The result determination unit is configured to determine the target processing result corresponding to the text content from the processing result based on the matching result.

[0041] In an embodiment, the result determination unit further includes:

[0042] The first content type identification unit is configured to identify the text content to obtain a content type corresponding to the text content.

[0043] The arrangement and recombination unit is configured to arrange and recombine the text content of the to-be-processed image and the target processing result corresponding to the text content according to a preset arrangement logic based on the content type to obtain target structured data.

[0044] The statistics unit is configured to perform statistical processing on the target structured data to obtain a processing indicator corresponding to the target processing result.

[0045] In an embodiment, the association unit can include:

[0046] The second content type identification unit is configured to identify the text content to obtain a content type corresponding to the text content.

[0047] The generation unit is configured to generate identification information of the text content based on the content type and the to-be-processed image.

[0048] The integration unit is configured to perform integration processing on the identification information of the text content of the to-be-processed image to obtain the association relationship information.

[0049] The embodiments of the present application further provide a computer program product or a computer program, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to enable the computer device to perform the method provided in any of the various optional manners of the above-mentioned aspect.

[0050] Correspondingly, the embodiments of the present application further provide a storage medium, which stores instructions. The instructions are executed by a processor to implement the information processing method provided in any of the embodiments of the present application.

[0051] The embodiment of the application can acquire a to-be-processed image; perform text detection on the to-be-processed image to obtain at least one text content in the to-be-processed image; perform association processing on the to-be-processed image and the text content to obtain association relationship information between the text content and the to-be-processed image; encrypt the text content in the to-be-processed image by using the text content in the preset image to obtain at least one encrypted text content; send the encrypted text content to a data processing device and acquire a processing result of processing the encrypted text content from the data processing device; and determine a target processing result corresponding to the text content from the processing result based on the association relationship information, thereby improving the security of information. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can also be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0053] Figure 1 is a scene diagram of an information processing method provided by the embodiment of the application;

[0054] Figure 2 is a flow diagram of an information processing method provided by the embodiment of the application;

[0055] Figure 3 is a scene diagram of a to-be-processed image provided by the embodiment of the application;

[0056] Figure 4 is a scene diagram of a to-be-processed image provided by the embodiment of the application;

[0057] Figure 5 is a scene diagram of establishing a coordinate axis on a to-be-processed image provided by the embodiment of the application;

[0058] Figure 6 is a scene diagram of a denoised image provided by the embodiment of the application;

[0059] Figure 7 is a scene diagram of text content provided by the embodiment of the application;

[0060] Figure 8 is a scene diagram of a relationship between text content and a storage unit provided by the embodiment of the application;

[0061] Figure 9 is a scene diagram of a relationship between text content and a storage unit provided by the embodiment of the application;

[0062] Figure 10is a scenario schematic diagram of target structured data provided by an embodiment of the present application.

[0063] Figure 11 is a scenario schematic diagram of an information processing system provided by an embodiment of the present application.

[0064] Figure 12 is another flow schematic diagram of an information processing method provided by an embodiment of the present application.

[0065] Figure 13 is a structure schematic diagram of an information processing apparatus provided by an embodiment of the present application.

[0066] Figure 14 is a structure schematic diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. However, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making any creative work fall within the scope of protection of the present application.

[0068] Artificial intelligence (AI) is to use digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology of computer science, which tries to understand the essence of intelligence and produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines, so that machines have the functions of perception, reasoning and decision-making.

[0069] Artificial intelligence technology is a comprehensive discipline, involving a wide range of fields, both hardware and software technologies. Artificial intelligence basic technologies generally include technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning, etc.

[0070] Machine Learning (ML) is a multi-disciplinary subject involving probability theory, statistics, approximation theory, convex analysis, algorithmic complexity theory, etc. It is a specialized study of how computers simulate or implement human learning behavior to acquire new knowledge or skills, reorganize existing knowledge structure, and continuously improve their own performance. Machine learning is the core of artificial intelligence and the fundamental approach to making computers intelligent. Its applications are widespread in various fields of artificial intelligence. Machine learning and deep learning generally include artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and teaching learning. Among them, reinforcement learning is a field of machine learning that emphasizes how to act based on the environment to achieve maximum expected benefits. Deep reinforcement learning combines deep learning and reinforcement learning and uses deep learning techniques to solve reinforcement learning problems.

[0071] With the rapid development of artificial intelligence technology, more and more application scenarios use artificial intelligence technology to include information security, avoid information leakage, and thus improve information security.

[0072] To this end, an information processing method is provided in an embodiment of the present application. The information processing method can be executed by an information processing device, which can be integrated in a computer device. The computer device can include at least one of a terminal and a server, etc. That is, the information processing method provided in the embodiment of the present application can be executed by a terminal, a server, or a terminal and a server capable of mutual communication.

[0073] The terminal can be a smartphone, a tablet computer, a notebook computer, a personal computer (PC), a smart home, a wearable electronic device, a VR / AR device, a vehicle-mounted computer, etc. The server can be an interworking server or a background server between multiple heterogeneous systems, or an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, and big data and artificial intelligence platforms, etc. Basic cloud computing services, etc.

[0074] In an embodiment, as shown in Figure 1 the embodiment of the present application shows a schematic diagram of the implementation environment of the information processing method provided in the embodiment of the present application. The implementation environment can include a terminal 11 and a server 12.

[0075] The server 12 may integrate the information processing device proposed in this application embodiment to implement the information processing method proposed in this application embodiment. The terminal 11 may be a data processing device, which is a device capable of processing data. For example, the information processing device may analyze and identify data, etc.

[0076] In one embodiment, server 12 can acquire an image to be processed; perform text detection on the image to be processed to obtain at least one text content in the image to be processed; perform association processing on the image to be processed and the text content to obtain association information between the text content and the image to be processed; use the text content in a preset image to encrypt the text content in the image to be processed to obtain at least one encrypted text content; send the encrypted text content to terminal 11 and obtain the processing result of the encrypted text content from terminal 11; and determine the target processing result corresponding to the text content from the processing result based on the association information.

[0077] The following will provide a detailed description of each example. It should be noted that the order of description of the following embodiments is not intended to limit the preferred order of the embodiments.

[0078] This application will describe the embodiments from the perspective of an information processing device, which can be integrated into a computer device, such as a server or a terminal.

[0079] like Figure 2 The present invention provides an information processing method, the specific process of which includes:

[0080] 201. Obtain the image to be processed.

[0081] The image to be processed may include images that may contain sensitive information. For example, the image to be processed may include images containing information from ID cards, passports, contracts, and / or invoices, etc. For example, the image to be processed may be a photocopy of an ID card. Another example is a scanned copy of a passport, etc.

[0082] In one embodiment, it is understood that sensitive data such as ID cards, passports, and contracts are involved in the specific implementation of this application. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0083] In an embodiment, the user can upload information in the form of an image. For example, the user can upload an ID card, a passport, a contract, an invoice, and the like in the form of an image. For example, the user can take an image of an ID card by using a camera or a mobile phone, and upload the image with the information of the ID card. For another example, the user can scan a contract by using a scanning device, and upload the scanned copy of the contract.

[0084] In an embodiment, there can be various methods to obtain the image to be processed. For example, the image to be processed can be obtained directly or indirectly.

[0085] For example, the user can upload the image to be processed directly to the information processing device, and the information processing device can obtain the image to be processed directly. For example, when the information processing device is integrated in a server, the user can upload the image to be processed directly to the server with the image processing device, and the image processing device can obtain the image to be processed directly.

[0086] For another example, the user can upload the image to be processed to a client, and the client can send the image to be processed to the image processing device.

[0087] The client can include a program that provides local services for the user. For example, the client can include an application, a webpage, and a mini-program, and the like.

[0088] For example, when the information processing device is integrated in a server, the user can first upload the image to be processed to a webpage, and the webpage can send the image to be processed to the server with the image processing device. For another example, the user can upload the image to be processed to an application, and the application can send the image to be processed to the server with the image processing device.

[0089] 202. Perform text detection on the image to be processed to obtain at least one text content in the image to be processed.

[0090] In an embodiment, after obtaining the image to be processed, the image processing device can perform text detection on the image to be processed to obtain the text content recorded in the image to be processed. By detecting the text content in the image to be processed, the image processing device can determine whether the image to be processed includes sensitive information. When the image to be processed includes sensitive information, the sensitive information in the image to be processed can be encrypted to improve the security of the information. For example, when the image processing device determines that the text content in the image to be processed is sensitive information, the image processing device can encrypt the text content in the image to be processed to improve the security of the information.

[0091] The sensitive information can include all information that, if misused or accessed or modified by unauthorized persons, can be detrimental to the implementation of a national interest or a federal government plan or detrimental to the personal privacy rights of individuals. For example, the sensitive information can include the address, name, ID number, and date of birth in an ID card, etc. For another example, the sensitive information can include the contract terms on a contract, etc.

[0092] In an embodiment, the text detection can be performed on the to-be-processed image in various manners to obtain at least one text content in the to-be-processed image.

[0093] For example, the text detection can be performed on the to-be-processed image by using an artificial intelligence method to obtain at least one text content in the to-be-processed image. For example, the text detection can be performed on the to-be-processed image by using any one of a convolutional neural network (CNN), a de-convolutional network (DN), a deep neural network (DNN), a deep convolutional inverse graphics network (DCIGN), a region-based convolutional network (RCNN), a faster region-based convolutional network (Faster RCNN), and a bidirectional encoder representations from transformers (BERT) model, etc. to obtain at least one text content in the to-be-processed image.

[0094] In an embodiment, due to the problem of image shooting by the user or the problem of scanning, the to-be-processed image can have a white space or an image skew, etc. that reduces the quality of the to-be-processed image. For example, as shown in FIG. 1, a to-be-processed image 001 has a white space 002. Figure 3 Figure 3 In an embodiment, due to the problem of image shooting by the user or the problem of scanning, the to-be-processed image can have a white space or an image skew, etc. that reduces the quality of the to-be-processed image. For example, as shown in FIG. 1, a to-be-processed image 001 has a white space 002. Figure 3 In an embodiment, due to the problem of image shooting by the user or the problem of scanning, the to-be-processed image can have a white space or an image skew, etc. that reduces the quality of the to-be-processed image. For example, as shown in FIG. 1, a to-be-processed image 001 has a white space 002. Figure 3 ​It can be seen that the identity card information 002 has a large amount of blank in the to-be-processed image 001, and the identity card information 002 is skewed. In addition, if the information processing apparatus directly detects the text content of the identity information 002 from the to-be-processed image 001, due to the existence of the blank and the skewed image, the accuracy of the information processing apparatus in detecting the text content will be reduced.

[0095] Therefore, when the information processing apparatus performs text detection on the to-be-processed image, the information processing apparatus can perform denoising processing on the to-be-processed image to obtain a denoised image. Then, the denoised image is detected to obtain the text content.

[0096] Specifically, the step of “performing text detection on the to-be-processed image to obtain at least one text content in the to-be-processed image” can include:

[0097] performing denoising processing on the to-be-processed image to obtain a denoised image;

[0098] detecting at least one text region in the denoised image;

[0099] recognizing the at least one text region to obtain at least one text content.

[0100] The denoising processing can include processing for improving the quality of the to-be-processed image. For example, when the to-be-processed image includes a large amount of blank, the denoising processing can include an operation of eliminating the blank in the to-be-processed image. For another example, when the to-be-processed image has a skewed image, the denoising processing can include processing for correcting the skewed image.

[0101] In an embodiment, there are various ways to perform denoising processing on the to-be-processed image.

[0102] For example, when the to-be-processed image includes a large amount of blank, the blank in the to-be-processed image can be eliminated by using image processing software. For example, the blank in the to-be-processed image can be eliminated by using image processing software such as Adobe Photoshop. For another example, the blank in the to-be-processed image can also be eliminated by using artificial intelligence technology.

[0103] For another example, when the to-be-processed image has a skewed image, the skewed image can also be corrected by using image processing software.

[0104] In an embodiment, when the to-be-processed image includes a large amount of blank and has a skewed image, the step of “performing denoising processing on the to-be-processed image to obtain a denoised image” can include:

[0105] performing target positioning on the to-be-processed image to locate target position information from the to-be-processed image;

[0106] Crop an effective information region image from the to-be-processed image based on the target position information.

[0107] Perform correction processing on the effective information region image to obtain a denoised image.

[0108] The target position information can include position information of a region with effective information in the to-be-processed image. For example, as shown in FIG. 1, the target position information can refer to position information of the effective information region 002 in the to-be-processed image 001. For another example, as shown in FIG. 2, the target position information can refer to position information of the effective information region 002 in the to-be-processed image 003. Figure 3 Figure 4

[0109] In an embodiment, the target position information can be represented in the form of a coordinate axis. For example, as shown in FIG. 3, a coordinate axis can be established with the center of the to-be-processed image 005 as the origin of the coordinate axis, and then the target position information of the effective information region 006 is represented by the coordinate axis. Figure 5

[0110] In an embodiment, there are various ways to perform target positioning on the to-be-processed image to locate the target position information from the to-be-processed image.

[0111] For example, an artificial intelligence algorithm can be used to perform target positioning on the to-be-processed image to locate the target position information from the to-be-processed image. For example, a trained target positioning model can be used to perform target positioning on the to-be-processed image to obtain the target position information. The target positioning model can be an image segmentation model or a key point detection model, etc. The image segmentation model or the key point detection model can be a model evolved based on deep learning. For example, the image segmentation model or the key point detection model can be a model evolved based on a CNN, a DNN, an RCNN, a conv3_3, a conv4_3, or a conv5_3, etc.

[0112] In an embodiment, the step of “performing target positioning on the to-be-processed image to locate the target position information from the to-be-processed image” can include:

[0113] performing feature extraction on the to-be-processed image at multiple different scales to obtain feature information of the to-be-processed image at the multiple different scales;

[0114] performing full connection processing on the feature information at the multiple different scales to obtain an effective information region in the to-be-processed image;

[0115] performing position recognition on the effective information region to obtain the target position information.

[0116] The effective information region can include a region with effective information in the to-be-processed image. For example, as shown in FIG. 4, the effective information region 402 in the to-be-processed image 401 can include a region with effective information.​​​Figure 3 As shown in FIG. 2B, the 002 in FIG. 2A can be an effective information region. Figure 3 As shown in FIG. 2B, the 002 in FIG. 2A can be an effective information region. Figure 4 As shown in FIG. 2B, the 002 in FIG. 2A can be an effective information region. Figure 4 As shown in FIG. 2B, the 004 in FIG. 2A can be an effective information region.

[0117] In one embodiment, feature extraction can be performed on the to-be-processed image at multiple different scales to obtain feature information of the to-be-processed image at the multiple different scales.

[0118] For example, the to-be-processed image can be subjected to feature extraction by using multiple different scale convolution kernels to obtain feature information of the to-be-processed image at multiple different scales.

[0119] In one embodiment, the feature information at multiple different scales can be subjected to full connection processing to obtain an effective information region in the to-be-processed image.

[0120] For example, multiple initial effective information regions can be divided in the to-be-processed image based on the feature information at each different scale. Then, the feature information at all scales can be combined to determine a final effective information region from the multiple initial effective information regions.

[0121] In one embodiment, position recognition can be performed on the effective information region to obtain target position information. For example, a coordinate axis can be generated in the to-be-processed image, and the target position information of the effective information region can be located according to the coordinate axis.

[0122] In one embodiment, after the target position information is located, the effective information region image can be cropped from the to-be-processed image based on the target position information.

[0123] The effective information region image can include an image having only the effective information region.

[0124] There are various ways to crop the effective information region image. For example, the effective information region image can be cropped by using an image cropping function. For another example, the effective information region image can be cropped by using image processing software.

[0125] In one embodiment, after the effective information region image is cropped, correction processing can be performed on the effective information region image to obtain a denoised image. For example, when the effective information region image is skewed, the effective information region can be straightened to obtain the denoised image.

[0126] The effective information region image can be corrected in various manners to obtain a denoised image. For example, the effective information region image can be corrected by calling an image processing software to obtain the denoised image. For another example, the effective information region image can be corrected by using a perspective transformation method to obtain the denoised image.

[0127] In an embodiment, after obtaining the denoised image, at least one text region can be detected in the denoised image.

[0128] The text region can refer to a region with text content in the denoised image. For example, as shown in FIG. 1 8, 018 in FIG. 1 8 can be the denoised image, and 019 and 020 in FIG. 1 8 can be text regions. Figure 6 Figure 6 The text region can refer to a region with text content in the denoised image. For example, as shown in FIG. 1 8, 018 in FIG. 1 8 can be the denoised image, and 019 and 020 in FIG. 1 8 can be text regions. Figure 6

[0129] In an embodiment, at least one text region can be detected in the denoised image in various manners.

[0130] For example, at least one text region can be detected in the denoised image by using an Efficient and Accuracy Scene Text (EAST) algorithm, a Connectionist Text Proposal Network, a textboxes algorithm, or a seglink algorithm, or a Real-time Scene Text Detection with Differentiable Binarization (DBNet) algorithm, and the like.

[0131] In an embodiment, after detecting the text region, the at least one text region can be recognized to obtain text content corresponding to each text region.

[0132] The text content can include a content type and / or valuable content in the text. For example, as shown in FIG. 1 9, a name, a gender, an ethnicity, a birth, an address, and an ID number can be the content type. For another example, as shown in FIG. 2 0, “Name: Xiao A” and “Gender: Female” can be the valuable content in the text. Figure 6 Figure 6 The text content can include a content type and / or valuable content in the text. For example, as shown in FIG. 1 9, a name, a gender, an ethnicity, a birth, an address, and an ID number can be the content type. For another example, as shown in FIG. 2 0, “Name: Xiao A” and “Gender: Female” can be the valuable content in the text.

[0133] In an embodiment, the at least one text region can be recognized in various manners to obtain the text content corresponding to each text region.

[0134] ​​​For example, the EAST algorithm, the Connectionist Text Proposal Network, the textboxes algorithm, or the seglink algorithm, or the DBNet algorithm, etc. can be used to identify the at least one text region to obtain the text content corresponding to each text region.

[0135] In an embodiment, the step of "identifying the at least one text region to obtain the at least one text content" can include:

[0136] extracting features of the text region to obtain feature information of the text region;

[0137] reshaping the feature information to obtain reshaped feature information;

[0138] mapping the reshaped feature information to a preset text probability space to obtain a text mapping probability corresponding to the reshaped feature information;

[0139] determining the text content based on the text mapping probability.

[0140] In this way, the feature information of the text region can be extracted by using a convolution kernel or a sampling method.

[0141] In an embodiment, the feature information of the text region is often in the form of a vector or a matrix. Reshaping the feature information can refer to changing the dimension of the vector or the matrix. For example, if the dimension of the feature information is b*c*d*e, the dimension of the reshaped feature information can be (bd)*e*c. Wherein, b, c, d and e can be any integer.

[0142] Reshaping the feature information can improve the amount of information of the feature information.

[0143] Then, the reshaped feature information can be mapped to a preset text probability space to obtain a text mapping probability corresponding to the reshaped feature information, and the text content can be determined based on the text mapping probability.

[0144] The preset text probability space can be a trained space that can reflect the fitting between the feature information and the text.

[0145] The remolded feature information can be mapped into a preset text probability space by using probability theory and the like to obtain a text mapping probability corresponding to the remolded feature information. Then, a text with the maximum text probability can be determined as the text content.

[0146] In an embodiment, the artificial intelligence model can also be pre-trained, and then used to detect the text in the image to be processed to obtain at least one text content in the image to be processed.

[0147] For example, a preset denoising model can be pre-trained, and the image to be processed can be denoised by using the preset denoising model to obtain a denoised image.

[0148] For another example, a preset text processing model can be pre-trained. Then, the preset text processing model can be used to detect at least one text region in the denoised image, and identify the at least one text region to obtain at least one text content.

[0149] The preset denoising model can be any one of image segmentation models or key point detection models and the like in a deep learning model library.

[0150] The preset text processing model can be any one of east, ctpn, textboxes, seglink, dbnet and the like.

[0151] 203. The image to be processed and the text content are associated to obtain association relationship information between the text content and the image to be processed.

[0152] In an embodiment, after the information processing apparatus detects the text content, the image to be processed and the text content can be associated to make the multiple text contents in the image to be processed have an association relationship, and make the image to be processed and the multiple text contents have an association relationship. By establishing the association relationship between the text content and the image to be processed, the text content can still be determined to belong to which object to be processed after being encrypted according to the association relationship information.

[0153] In an embodiment, there are multiple ways to associate the image to be processed and the text content to obtain the association relationship information between the text content and the image to be processed.

[0154] For example, an association table between the image to be processed and the text content can be established, and the association table can be used as the association relationship information between the text content and the image to be processed.

[0155] For example, identification information can be added to the text content, and the text content, the identification information corresponding to the text content, and the to-be-processed image can be integrated together to form the association relationship information between the text content and the to-be-processed image. Specifically, the step of "associating the to-be-processed image and the text content to obtain the association relationship information between the text content and the to-be-processed image" can include:

[0156] identifying the text content to obtain a content type corresponding to the text content;

[0157] generating identification information of the text content based on the content type and the to-be-processed image;

[0158] integrating the identification information of the text content in the to-be-processed image to obtain the association relationship information.

[0159] In an embodiment, when adding identification information to the text content, the identification information of the text content can be generated based on the content type corresponding to the text content and the to-be-processed image corresponding to the text content.

[0160] For example, when the to-be-processed image includes identity card information, the content type corresponding to the text content can be name, address, gender, date of birth, and identity card number, etc. For example, when the to-be-processed image includes contract information, the content type corresponding to the text content can be contract terms, contract signing parties, contract effective date, etc.

[0161] In an embodiment, the identification information of the text content can be generated based on the content type and the to-be-processed image. For example, a first sub-identification can be generated according to the content type corresponding to the text content. In addition, a second sub-identification can be generated according to the to-be-processed image corresponding to the text content, and a third sub-identification can be generated according to the text content itself. Next, the first sub-identification, the second sub-identification, and the third sub-identification can be combined together to obtain the identification information of the text content.

[0162] For example, the sub-identification can be generated by using an MD5 Message-Digest Algorithm (MD5), a Secure Hash Algorithm (SHA), or the like.

[0163] For example, a first sub-identifier of the text content can be generated according to the content identifier corresponding to the text content by using MD5; a second sub-identifier can be generated according to the to-be-processed image corresponding to the text content by using MD5; and a third sub-identifier can be generated according to the text content itself. Next, the first sub-identifier, the second sub-identifier, and the third sub-identifier can be combined together to obtain the identifier information of the text content.

[0164] There are various ways to combine the sub-identifiers together. For example, the sub-identifiers can be directly spliced together to obtain the identifier information of the text content. For another example, the sub-identifiers can be subjected to logical operation processing to obtain the identifier information of the text content.

[0165] In an embodiment, the identifier information of the text content is generated based on the content type corresponding to the text content, the to-be-processed image corresponding to the text content, and the text content itself, so that it is known through the identifier information of the text content what the content type of the text content is and which to-be-processed image it belongs to, without the need to identify the text content again, thereby improving the efficiency of information processing.

[0166] In an embodiment, the identifier information of the text content can also be directly generated. For example, the identifier information of the text content can be randomly generated.

[0167] In an embodiment, after the identifier information of each text content in the to-be-processed image is generated, the identifier information of all the text contents in the to-be-processed image can be integrated to obtain the association relationship information.

[0168] For example, there are three text contents in the to-be-processed image, which are text content w1, text content w2, and text content w3. The identifier information of the text content w1 is v1, the identifier information of the text content w2 is v2, and the identifier information of the text content w3 is v3. Therefore, v1, v2, and v3 can be integrated together to obtain the association relationship information.

[0169] For example, w1, v1, w2, v2, w3, and v3 can be stored in the same document correspondingly, and the document can be taken as the association relationship information. For another example, w1, v1, w2, v2, w3, and v3 can be stored in the same area in a preset document correspondingly, and the information in the area can be taken as the association relationship information of the text contents in the to-be-processed image. For example, as shown in FIG. 1, the to-be-processed image 1 includes text content 1 or text content 2, and so on. Then, the identifier information of the text content 1 and the text content 2, etc. in the to-be-processed image 1 can be stored in the format corresponding to the association relationship information document. Figure 7

[0170] ​In an embodiment, by integrating the identification information of the text content in the image to be processed, the association relationship information is obtained. When the target processing result corresponding to the text content is determined from the processing result based on the association relationship information, the target processing result corresponding to the text content can be quickly determined by querying the association relationship information, thereby improving the information processing efficiency.

[0171] 204. encrypting the text content in the image to be processed by using the text content in the preset image, to obtain at least one encrypted text content.

[0172] In an embodiment, when the information processing device detects that the text content in the image to be processed is sensitive information, in order to avoid causing information leakage and improve the security of information, the text content in the image to be processed can be encrypted by using the text content in the preset image, to obtain at least one encrypted text content.

[0173] The preset image can include an image that has been processed by the information processing device. That is, the preset image can be an image in which the information processing device has detected text content, and the text content can have sensitive information.

[0174] In an embodiment, when the information processing device detects the text content in the image to be processed, the text content belonging to the same image to be processed is often stored together. If the text content is directly sent to the data processing device, the data processing device will know the association relationship between the text content.

[0175] For example, the image to be processed includes identity card information, and the information processing device detects the text content of the name, gender, birth date, address, and identity card number of a citizen in the image to be processed. Generally, the information processing device will store the text content of the name, gender, birth date, address, and identity card number of the citizen together. If the information processing device does not encrypt the text content, but directly sends the detected text content to the data processing device, the data processing device will directly obtain the sensitive information of the name, gender, birth date, address, and identity card number of the citizen, thereby reducing the security of information.

[0176] In an embodiment, since the information processing device will send the text content to the data processing device for processing, the process of encrypting the text content generally will not change the original information of the text content, but will introduce the text content in the preset image, and change the association relationship between the text content in the image to be processed by using the text content in the preset image, to realize the desensitization processing of the text content, so that the data processing device cannot obtain the sensitive information of the user.

[0177] In an embodiment, there are various ways to encrypt the text content in the to-be-processed image using the text content in the preset image, to obtain at least one encrypted text content.

[0178] For example, the text content of the preset image can be inserted into the text content of the to-be-processed image, so as to destroy the original association relationship of the text content of the to-be-processed image. For another example, the text content of the preset image and the text content of the to-be-processed image can be randomly shuffled to obtain disordered text content, so as to destroy the original association relationship of the text content of the to-be-processed image.

[0179] In an embodiment, the step of "encrypting the text content in the to-be-processed image using the text content in the preset image, to obtain at least one encrypted text content" can include:

[0180] determining a first initial storage address of the text content in the preset image in a preset storage space, and determining a second initial storage address of the text content in the to-be-processed image in the preset storage space;

[0181] based on the first initial storage address and the second initial storage address, performing mixed rearrangement processing on the text content in the preset image and the text content in the to-be-processed image, to obtain the encrypted text content.

[0182] The preset storage space can be a region in which the information processing apparatus stores the text content of the preset image. For example, the preset storage space can be a read-only memory or a random access memory in the information processing apparatus, and the like.

[0183] The first initial storage address can be a storage address of the text content in the preset image in the preset storage space. For example, when the preset storage space is a memory, the first initial storage address can be an address of the text content in the preset image in the memory. For example, when the preset storage space is a memory, the first initial storage address can be a number of a storage unit in the memory.

[0184] The second initial storage address can be a storage address of the text content in the to-be-processed image in the preset storage space. For example, when the preset storage space is a memory, the second initial storage address can be an address of the text content in the to-be-processed image in the memory. For example, when the preset storage space is a memory, the second initial storage address can be a number of a storage unit in the memory.

[0185] In an embodiment, for the text content with an association relationship, the initial storage address thereof is also often associated. For example, the initial storage addresses corresponding to each text content in the same ID card can be connected together.

[0186] For example, asFigure 8 As shown, the preset storage space 017 includes 10 storage units. The preset storage space 017 stores the text content of the to-be-processed image 1 and the preset image 2. Since the text content 1 and the text content 2 are detected in the to-be-processed image, the text content 1 and the text content 2 are stored in the continuous storage units in the preset storage space 017. For example, the content type "name" of the text content 1 of the to-be-processed image 1 is stored in the storage unit 007, the content "ABC" is stored in the storage unit 008, and the content type "date of birth" of the text content 2 of the to-be-processed image 1 is stored in the storage unit 009, and the content "19990101" is stored in the storage unit 010. Similarly, since the text content 21 and the text content 22 are detected in the preset image 2, the text content 21 and the text content 22 are stored in the continuous storage units in the preset storage space 017.

[0187] Therefore, the association between the text contents can be destroyed by adjusting the initial storage positions of the text contents, so that the desensitization encryption of the text contents is realized.

[0188] In an embodiment, the text content in the preset image and the text content in the to-be-processed image can be mixed and rearranged based on the first initial storage address and the second initial storage address, so as to destroy the association between the text contents, thereby realizing the desensitization encryption of the text contents. Specifically, the step of "mixing and rearranging the text content in the preset image and the text content in the to-be-processed image based on the first initial storage address and the second initial storage address to obtain the encrypted text content" can include:

[0189] The first initial storage address is calculated by using the preset rearrangement logic to obtain the first rearranged storage address of the text content in the to-be-processed image, and the second initial storage address is calculated by using the preset rearrangement logic to obtain the second rearranged storage address of the text content in the preset image;

[0190] According to the first rearranged storage address, the storage position of the text content in the to-be-processed image in the preset storage space is updated, and according to the second rearranged storage position, the storage position of the text content in the preset image in the preset storage space is updated, to obtain the encrypted text content.

[0191] The preset rearrangement logic can be a logic that is set in advance and is used to change the initial storage address of the text content. For example, the preset rearrangement logic can be a logic with a certain rule. For another example, the preset rearrangement logic can be a logic with random properties.

[0192] In one embodiment, a preset rearrangement logic can be used to calculate the first rearranged storage address corresponding to the first initial storage address and the second rearranged storage address corresponding to the second initial storage address.

[0193] For example, when the preset rearrangement logic is a logic with random properties, the storage location of the rearranged text content can be randomly generated.

[0194] For example, when the preset rearrangement logic is a logic with a certain pattern, the rearranged storage address of the text content can be generated based on the initial storage address of the text content.

[0195] For example, such as Figure 8 As shown, the initial storage address of the content type "Name" in text content 1 is storage unit 007. Using the preset rearrangement logic, the rearranged storage address of the content type "Name" in text content 1 can be calculated to be storage unit 016. For example, as... Figure 8 As shown, the initial storage address of the content type "name" in the text content 21 of the preset image 2 is storage unit 012. Using the preset rearrangement logic, the rearranged storage address of the content type "name" in the text content 21 can be calculated to be storage unit 010.

[0196] In one embodiment, after calculating the first rearranged storage address of the text content in the preset image and the second rearranged storage address of the text content in the image to be processed, the storage location of the text content in the image to be processed in the preset storage space can be updated according to the first rearranged storage address, and the storage location of the text content in the preset image in the preset storage space can be updated according to the second rearranged storage address, and the text content with the updated storage location can be used as the encrypted text content.

[0197] For example, such as Figure 9 As shown, by rearranging the storage locations of the text content, the encrypted text content 018 is obtained. It can be seen from the encrypted text content 018 that the original relationships between the text content have been destroyed; the encrypted text content no longer has any relationships. Therefore, data processing devices cannot obtain sensitive information from the encrypted text content, thus improving information security.

[0198] In one embodiment, this application only provides some embodiments for encrypting the text content in the image to be processed using the text content in the preset image. However, there are various ways to encrypt the text content in the image to be processed using the text content in the preset image to obtain encrypted text content.

[0199] For example, the text content in the image to be processed can be changed using the text content in the preset image, thereby achieving text content desensitization, and so on.

[0200] 205. sending the encrypted text content to the data processing device, and obtaining a processing result of processing the encrypted text content from the data processing device.

[0201] In an embodiment, after the encrypted text content is generated, the encrypted text content can be sent to the data processing device, so that the data processing device processes the encrypted text content.

[0202] In the embodiments of the present application, the manner in which the data processing device processes the encrypted text content is not limited. That is, the data processing device can process the encrypted text content in various ways.

[0203] In an embodiment, after the data processing device finishes processing the encrypted text content, the information processing apparatus can obtain a processing result of processing the encrypted text content from the data processing device.

[0204] 206. determining a target processing result corresponding to the text content from the processing result based on the association information.

[0205] In an embodiment, after the information processing device obtains the processing result of the encrypted text content, the information processing device can determine a target processing result corresponding to the text content in the image to be processed from the processing result based on the association information. That is, the information processing apparatus can map the processing result of the encrypted text content back to the corresponding text content. For example, the encrypted text content C' is obtained by encrypting the text content C. After obtaining the processing result of the encrypted text content C', the information processing apparatus can map the processing result of the encrypted text content C' back to the text content C.

[0206] In an embodiment, since the association information includes the identification information of the text content, the target processing result corresponding to the text content can be determined based on the identification information in the association information. Specifically, the step of "determining a target processing result corresponding to the text content from the processing result based on the association information" can include:

[0207] reading the association information to obtain identification information corresponding to the text content;

[0208] matching the processing identification of the processing result with the identification information to obtain a matching result;

[0209] determining a target processing result corresponding to the text content from the processing result based on the matching result.

[0210] In an embodiment, the data processing device can generate a processing identifier for a processing result when processing the encrypted text content. The processing identifier can include a mark indicating which encrypted text content the processing result corresponds to. Through the processing identifier, the information processing apparatus can know which encrypted text content the processing result corresponds to, and thus know which text content the processing result corresponds to.

[0211] In an embodiment, to improve the efficiency of determining the target processing result corresponding to the text content, the processing identifier of the processing result can be matched with the identification information. When the processing identifier of the processing result is matched with the identification information, the processing result corresponding to the processing identifier can be determined as the target processing result corresponding to the text content.

[0212] In an embodiment, by determining the target processing result corresponding to the text content, the target processing result of the text content can be counted, and thus the processing indicator corresponding to the target processing result can be obtained. Then, the processing indicator can be fed back to the data processing device, so as to assist in improving the capability of the data processing device.

[0213] Therefore, in an embodiment, the information processing method proposed in the embodiments of the present application further includes:

[0214] identifying the text content to obtain a content type corresponding to the text content;

[0215] rearranging the text content and the target processing result corresponding to the text content according to a preset arrangement logic based on the content type, to obtain target structured data;

[0216] counting the target structured data to obtain a processing indicator corresponding to the target processing result.

[0217] The preset arrangement logic can be a rule that needs to be followed when rearranging the text content and the target processing result corresponding to the text content, and the rule can be set in advance.

[0218] The processing indicator can include information indicating the good or bad of the target processing result. For example, when the data processing device identifies the encrypted text content, the processing indicator can be the identification effect of the data processing device on the encrypted text content. For example, the processing indicator can be an identification accuracy rate or an identification error rate, etc. For another example, when the data processing device detects the encrypted text content, the processing indicator can be the detection effect of the data processing device on the encrypted text content. For example, the processing indicator can be a detection accuracy rate or a detection error rate, etc.

[0219] In an embodiment, for the neatness and observability of data, the image text content to be processed and the target processing result corresponding to the text content can be rearranged based on the content type, the case preset arrangement logic, to obtain the target structured data.

[0220] For example, as shown in Figure 10 , the target structured data 020 can be obtained by rearranging the target processing result 019 corresponding to the text content in the image to be processed and the text content in the image to be processed.

[0221] Then, the target structured data can be statistically processed to obtain the processing indicators of the target processing result. For example, the target processing result in the target structured data and the original text content can be compared to statistically obtain the processing indicators of the target processing result.

[0222] Next, the processing indicators can be fed back to the data processing device to assist in improving the ability of the data processing device.

[0223] In an embodiment, the present application also provides an information processing system. For example, as shown in Figure 11 , the information processing system can include a model pre-training module, an information region extraction module, an information encryption and desensitization module, and an information decryption module.

[0224] The model pre-training module can be used to train the model to obtain a trained model. For example, a preset training model can train the model to obtain a trained preset denoising model. For another example, a preset training model can train the model to obtain a trained preset text processing model.

[0225] The information region extraction module is used to obtain an image to be processed; text detection is performed on the image to be processed to obtain at least one text content in the image to be processed. In addition, the information region extraction module can also use the trained model in the model pre-training module to perform text detection on the image to be processed to obtain at least one text content in the image to be processed.

[0226] The region encryption and desensitization module can associate the image to be processed and the text content to obtain the association relationship information between the text content and the image to be processed; encrypt the text content in the image to be processed using the text content in the preset image to obtain at least one encrypted text content; and send the encrypted text content to the data processing device.

[0227] The information decryption module can be used to obtain the processing result of the encrypted text content from the data processing device; and determine the target processing result corresponding to the text content from the processing result based on the association relationship information.

[0228] An information processing method is provided in the embodiments of the present application. The information processing method includes: obtaining a to-be-processed image; performing text detection on the to-be-processed image to obtain at least one text content in the to-be-processed image; performing association processing on the to-be-processed image and the text content to obtain association relationship information between the text content and the to-be-processed image; encrypting the text content in the to-be-processed image by using text content in a preset image to obtain at least one encrypted text content; sending the encrypted text content to a data processing device and obtaining a processing result of processing the encrypted text content from the data processing device; and determining a target processing result corresponding to the text content from the processing result based on the association relationship information. The text content in the to-be-processed image is encrypted, so that desensitization processing of the text content is implemented. Therefore, after the data processing device obtains the encrypted text content, the data processing device cannot obtain sensitive information through the encrypted text content, so that the security of information is improved.

[0229] In addition, the embodiments of the present application can also obtain a processing result of processing the encrypted text content from the data processing device, and determine a target processing result corresponding to the text content from the processing result based on the association relationship information. By determining the target processing result corresponding to the text content, the target processing result of the text content can be counted, so that a processing index corresponding to the target processing result is obtained. Then, the processing index can be fed back to the data processing device, so as to assist in improving the ability of the data processing device.

[0230] According to the method described in the above embodiments, the following will be further described in detail by way of example.

[0231] The embodiments of the present application will be described below by taking an information processing method integrated on a server as an example.

[0232] In an embodiment, as shown in Figure 12 An information processing method is provided. The specific process is as follows:

[0233] 401. The server obtains a to-be-processed image.

[0234] The to-be-processed image can include an image that can have sensitive information. For example, the to-be-processed image can include an image that has information of an identity card, a passport, a contract, and / or an invoice, etc. For example, the to-be-processed image can be a copy of an identity card. For example, the to-be-processed image can be a scanned copy of a passport, etc.

[0235] 402. The server performs text detection on the to-be-processed image to obtain at least one text content in the to-be-processed image.

[0236] In an embodiment, after obtaining the image to be processed, the server can locate the region with valid information in the image to be processed, and save the region with valid information after correction and cropping to obtain the denoised image.

[0237] The server can pre-train a preset denoising model, and then locate the region with valid information in the image to be processed by using the preset denoising model.

[0238] The preset denoising model can be a deep learning model. For example, the preset denoising model can be a deep segmentation model or a deep key point model, etc.

[0239] The preset denoising model can locate the coordinates and content type of the region with valid information. For example, the preset denoising model can output whether the region with valid information is an ID card, a contract, or a passport, etc. In addition, the preset denoising model can also output the coordinates of the region with valid information, so that the server can crop the denoised image from the image to be processed according to the coordinates.

[0240] In an embodiment, after obtaining the denoised image, the region with text content in the denoised image can be detected.

[0241] For example, the server can pre-train a text processing model, and then use the text processing model to detect the region with text content in the denoised image.

[0242] The text processing model can be a deep learning model. For example, the deep learning model can be east, ctpn, textboxes, seglink, dbnet, etc.

[0243] The text processing model can output the coordinates and type of the text region. For example, the text processing model can output whether the text region is a name, a gender, or an ID number, etc.

[0244] The coordinates of the text region can be represented in the form of [x, y, w, h]. Wherein, x represents the horizontal coordinate of the upper left corner of the text region, y represents the vertical coordinate of the upper left corner of the text region, w represents the width of the text region, and h represents the height of the text region. Wherein, x, y, w and h can be represented in pixels. That is, the size of x, y, w and h can be represented by the number of pixels.

[0245] In an embodiment, the type of the detected text region can be used as the text content. For example, the text content of the text region can be “name”. For another example, the text content of the text region can be “gender”, etc.

[0246] 403、The server performs association processing on the to-be-processed image and the text content, to obtain association relationship information between the text content and the to-be-processed image.

[0247] In an embodiment, the server can perform association processing on the to-be-processed image and the text content, to obtain association relationship information between the text content and the to-be-processed image.

[0248] For example, the server can generate MD5 codes for all the text content in the same to-be-processed image. Then, the server can save the MD5 code corresponding to each text content, the text content itself, and the to-be-processed image to which the text content belongs, into an association area of a preset storage area, to obtain the association relationship information.

[0249] 404、The server encrypts the text content in the to-be-processed image by using the text content in the preset image, to obtain at least one encrypted text content.

[0250] In an embodiment, the server can encrypt the text content in the to-be-processed image by using the text content in the preset image, to obtain at least one encrypted text content.

[0251] For example, the server also stores text content detected by other images (equivalent to the preset image). Then, the server can randomly shuffle the text content in the preset image and the text content in the to-be-processed image, to generate an unordered text content package. The unordered text content package includes unordered text content, which can be encrypted text content.

[0252] 405、The server sends the encrypted text content to the data processing device, and obtains a processing result of processing the encrypted text content from the data processing device.

[0253] 406、The server determines a target processing result corresponding to the text content from the processing result based on the association relationship information.

[0254] For example, the server obtains the processing result of the data processing device, and obtains the association relationship information generated in step 403. According to the association relationship information, the server obtains the MD5 of all the text content, and searches for a target processing result corresponding to the text content corresponding to the MD5 code from the processing result generated by the data processing device, to finally obtain the target processing result of all the text content in the same to-be-processed image. Then, the server can organize the target structured data according to the content type in the association relationship information.

[0255] Then, the server can count the target processing result of the text content based on the target structured data, to obtain a processing index, such as the recognition rate of the name field, the field with the lowest recognition rate, and the like, so as to further feed back the processing index to the data processing device, to assist in improving the capability of the data processing device.

[0256] In the embodiments of the present application, the server obtains a to-be-processed image; the server performs text detection on the to-be-processed image to obtain at least one text content in the to-be-processed image; the server performs association processing on the to-be-processed image and the text content to obtain association relationship information between the text content and the to-be-processed image; the server encrypts the text content in the to-be-processed image by using the text content in the preset image to obtain at least one encrypted text content; the server sends the encrypted text content to a data processing device and obtains a processing result of processing the encrypted text content from the data processing device; and the server determines a target processing result corresponding to the text content from the processing result based on the association relationship information. The embodiments of the present application encrypt the text content in the to-be-processed image, thereby realizing desensitization processing of the text content. Therefore, after the data processing device obtains the encrypted text content, the data processing device cannot obtain sensitive information through the encrypted text content, thereby improving the security of information.

[0257] In order to better implement the information processing method provided by the embodiments of the present application, in an embodiment, an information processing device is also provided, which can be integrated in a computer device. The meanings of the terms are the same as those in the above information processing method, and specific implementation details can be referred to the description in the method embodiment.

[0258] In an embodiment, an information processing device is provided, which can be specifically integrated in a computer device, as shown in Figure 13 The information processing device includes an obtaining unit 601, a text detection unit 602, an association unit 603, an encryption unit 604, a result obtaining unit 605 and a result determining unit 606, and the details are as follows.

[0259] The obtaining unit 601 is configured to obtain a to-be-processed image.

[0260] The text detection unit 602 is configured to perform text detection on the to-be-processed image to obtain at least one text content in the to-be-processed image.

[0261] The association unit 603 is configured to perform association processing on the to-be-processed image and the text content to obtain association relationship information between the text content and the to-be-processed image.

[0262] The encryption unit 604 is configured to encrypt the text content in the to-be-processed image by using the text content in the preset image to obtain at least one encrypted text content.

[0263] The result obtaining unit 605 is configured to send the encrypted text content to a data processing device and obtain a processing result of processing the encrypted text content from the data processing device.

[0264] The result determination unit 606 is configured to determine a target processing result corresponding to the text content from the processing results based on the association relationship information.

[0265] In an embodiment, the encryption unit 604 can include:

[0266] The address determination subunit is configured to determine a first initial storage address of the text content in the preset image in a preset storage space, and determine a second initial storage address of the text content in the image to be processed in the preset storage space.

[0267] The mixed rearrangement subunit is configured to perform mixed rearrangement processing on the text content in the preset image and the text content in the image to be processed based on the first initial storage address and the second initial storage address, to obtain the encrypted text content.

[0268] In an embodiment, the mixed rearrangement subunit can include:

[0269] The calculation module is configured to calculate the first initial storage address using a preset rearrangement logic to obtain a first rearranged storage address of the text content in the image to be processed, and calculate the second initial storage address using the preset rearrangement logic to obtain a second rearranged storage address of the text content in the preset image.

[0270] The update module is configured to update a storage location of the text content in the image to be processed in the preset storage space according to the first rearranged storage address, and update a storage location of the text content in the preset image in the preset storage space according to the second rearranged storage location, to obtain the encrypted text content.

[0271] In an embodiment, the text detection unit 602 can include:

[0272] The denoising subunit is configured to perform denoising processing on the image to be processed to obtain a denoised image.

[0273] The region detection subunit is configured to detect at least one text region in the denoised image.

[0274] The text recognition subunit is configured to recognize the at least one text region to obtain the at least one text content.

[0275] In an embodiment, the denoising subunit can include:

[0276] The positioning module is configured to perform target positioning on the image to be processed to locate target position information from the image to be processed.

[0277] a cropping module configured to crop an effective information region image from the to-be-processed image based on the target position information;

[0278] a correction module configured to perform correction processing on the effective information region image to obtain the denoised image.

[0279] In an embodiment, the positioning module can include:

[0280] a feature extraction sub-module configured to perform feature extraction on the to-be-processed image at multiple different scales to obtain feature information of the to-be-processed image at the multiple different scales;

[0281] a full connection sub-module configured to perform full connection processing on the feature information at the multiple different scales to obtain an effective information region in the to-be-processed image;

[0282] a position recognition sub-module configured to perform position recognition on the effective information region to obtain the target position information.

[0283] In an embodiment, the text recognition sub-unit can include:

[0284] a feature extraction module configured to perform feature extraction on the text region to obtain feature information of the text region;

[0285] a reshaping module configured to perform reshaping processing on the feature information to obtain reshaped feature information;

[0286] a mapping module configured to map the reshaped feature information into a preset text probability space to obtain a text mapping probability corresponding to the reshaped feature information;

[0287] a text content determination module configured to determine the text content based on the text mapping probability.

[0288] In an embodiment, the result determination unit 606 can include:

[0289] a reading sub-unit configured to read the association relationship information to obtain identification information corresponding to the text content;

[0290] a matching sub-unit configured to match a processing identifier of the processing result and the identification information to obtain a matching result;

[0291] a result determination sub-unit configured to determine a target processing result corresponding to the text content from the processing result based on the matching result.

[0292] In an embodiment, the result determination unit 606 further includes:

[0293] The first content type identification subunit is used to identify the text content and obtain the content type corresponding to the text content;

[0294] The arrangement and recombination subunit is used to arrange and recombine the text content in the image to be processed and the target processing result corresponding to the text content according to the content type and a preset arrangement logic, so as to obtain the target structured data.

[0295] The statistical subunit is used to perform statistical processing on the target structured data to obtain the processing index corresponding to the target processing result.

[0296] In one embodiment, the association unit 603 may include:

[0297] The second content type identification subunit is used to identify the text content and obtain the content type corresponding to the text content;

[0298] A generation subunit is used to generate identification information for the text content based on the content type and the image to be processed;

[0299] The integration subunit is used to integrate the identification information of the text content in the image to be processed to obtain the association information.

[0300] In practice, each of the above units can be implemented as an independent entity or can be arbitrarily combined to be implemented as the same or several entities. For the specific implementation of each of the above units, please refer to the previous method embodiments, which will not be repeated here.

[0301] The information processing device described above can improve information security.

[0302] This application also provides a computer device, which may include a terminal or a server. For example, the computer device can be an information processing terminal, such as a mobile phone, tablet computer, etc.; or it can be a server, such as an information processing server. Figure 14 As shown, it illustrates the structural diagram of the terminal involved in the embodiments of this application, specifically:

[0303] The computer device may include components such as a processor 801 with one or more processing cores, a memory 802 with one or more computer-readable storage media, a power supply 803, and an input unit 804. Those skilled in the art will understand that... Figure 14 The computer device structure shown does not constitute a limitation on the computer device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein:

[0304] The processor 801 is the control center of the computer device, connects the various parts of the computer device through various interfaces and lines, and performs various functions and processes data of the computer device by running or executing software programs and / or modules stored in the memory 802 and calling data stored in the memory 802, thereby overall detecting the computer device. Optionally, the processor 801 can include one or more processing cores; preferably, the processor 801 can integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user pages and application programs, etc., and the modem processor mainly processes wireless communication. It can be understood that the above-mentioned modem processor can also not be integrated into the processor 801.

[0305] The memory 802 can be used to store software programs and modules, and the processor 801 executes various functions and data processing by running the software programs and modules stored in the memory 802. The memory 802 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required by a function (such as a sound playing function, an image playing function, etc.), etc.; the data storage area can store data created according to the use of the computer device, etc. In addition, the memory 802 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state memory device. Accordingly, the memory 802 can also include a memory controller to provide access for the processor 801 to the memory 802.

[0306] The computer device further includes a power supply 803 for powering various components, and preferably the power supply 803 can be logically connected to the processor 801 through a power management system, so as to realize functions such as management of charging, discharging and power consumption management through the power management system. The power supply 803 can also include one or more than one direct current or alternating current power supply, a recharging system, a power failure detection circuit, a power converter or inverter, a power state indicator, etc. Any component.

[0307] The computer device can also include an input unit 804, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0308] Although not shown, the computer device can also include a display unit, etc., which will not be described here. Specifically, in the present embodiment, the processor 801 in the computer device will load the executable file corresponding to the process of one or more than one application program into the memory 802 according to the following instructions, and run the application program stored in the memory 802 by the processor 801, thereby realizing various functions, as follows:

[0309] obtaining an image to be processed;

[0310] performing text detection on the image to be processed to obtain at least one text content in the image to be processed;

[0311] performing association processing on the image to be processed and the text content to obtain association relationship information between the text content and the image to be processed;

[0312] encrypting the text content in the image to be processed using text content in a preset image to obtain at least one encrypted text content;

[0313] sending the encrypted text content to a data processing device and obtaining a processing result of processing the encrypted text content from the data processing device;

[0314] determining a target processing result corresponding to the text content from the processing result based on the association relationship information.

[0315] The specific implementation of each operation can be referred to the foregoing embodiments, which will not be described here.

[0316] According to an aspect of the present application, a computer program product or computer program is provided, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to enable the computer device to perform the method provided in various optional implementation manners in the foregoing embodiments.

[0317] Those skilled in the art can understand that all or part of the steps in the various methods of the foregoing embodiments can be completed by a computer program, or by relevant hardware controlled by a computer program, which can be stored in a computer readable storage medium and loaded and executed by a processor.

[0318] To this end, the embodiments of the present application further provide a storage medium having a computer program stored therein, which can be loaded by a processor to execute the steps in any information processing method provided by the embodiments of the present application. For example, the computer program can execute the following steps:

[0319] obtaining an image to be processed;

[0320] performing text detection on the image to be processed to obtain at least one text content in the image to be processed;

[0321] performing association processing on the image to be processed and the text content to obtain association relationship information between the text content and the image to be processed;

[0322] encrypting the text content in the to-be-processed image based on the text content in the preset image, to obtain at least one encrypted text content;

[0323] sending the encrypted text content to a data processing device, and obtaining a processing result of processing the encrypted text content from the data processing device;

[0324] determining a target processing result corresponding to the text content from the processing result based on the association relationship information.

[0325] The specific implementation of each operation can be referred to the foregoing embodiments, which will not be described here.

[0326] Since the computer program stored in the storage medium can execute the steps in any of the information processing methods provided by the embodiments of the present application, the beneficial effects that can be achieved by any of the information processing methods provided by the embodiments of the present application can be achieved, which will be described in detail in the foregoing embodiments, and will not be described here.

[0327] The information processing method, device, computer device and storage medium provided by the embodiments of the present application are described in detail above, and the principle and implementation manner of the present application are described by applying specific examples in this paper. The above embodiment is only used to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation manner and application range will be changed, and the above description should not be understood as the limitation of the present application.

Claims

1. An information processing method characterized by comprising: The method comprises: obtaining a to-be-processed image; performing text detection on the to-be-processed image to obtain at least one text content in the to-be-processed image; performing recognition on the text content to obtain a content type corresponding to the text content; generating identification information of the text content based on the content type and the to-be-processed image, the identification information being obtained by combining a first sub-identifier corresponding to the content type, a second sub-identifier corresponding to the to-be-processed image, and a third sub-identifier corresponding to the text content; integrating the identification information of the text content in the to-be-processed image to obtain association relationship information; determining a first initial storage address of preset image text content in a preset storage space and a second initial storage address of the to-be-processed image text content in the preset storage space; the preset image text content is stored in continuous storage units in the preset storage space; and the to-be-processed image text content is stored in continuous storage units in the preset storage space; calculating the first initial storage address by using a preset rearrangement logic to obtain a first rearranged storage address of the to-be-processed image text content, and calculating the second initial storage address by using the preset rearrangement logic to obtain a second rearranged storage address of the preset image text content; the preset rearrangement logic is a logic with random properties; updating the storage location of the to-be-processed image text content in the preset storage space according to the first rearranged storage address, and updating the storage location of the preset image text content in the preset storage space according to the second rearranged storage address, to mix and rearrange the text content in the preset image and the text content in the to-be-processed image, and obtain encrypted text content; sending the encrypted text content to a data processing device and obtaining a processing result of processing the encrypted text content from the data processing device; determining a target processing result corresponding to the text content from the processing result based on the association relationship information.

2. The method of claim 1, wherein, The method comprises: performing denoising processing on the to-be-processed image to obtain a denoised image; detecting at least one text region in the denoised image; performing recognition on the at least one text region to obtain the at least one text content.

3. The method of claim 2, wherein, The method comprises: performing target positioning on the to-be-processed image to locate target position information from the to-be-processed image; cropping an effective information region image from the to-be-processed image based on the target position information; performing correction processing on the effective information region image to obtain the denoised image.

4. The method of claim 3, wherein, The method comprises: performing feature extraction on the to-be-processed image at multiple different scales to obtain feature information of the to-be-processed image at the multiple different scales; The feature information of the plurality of different scales is fully connected to obtain an effective information region in the image to be processed; The target position information is obtained by performing position recognition on the effective information region.

5. The method of claim 2, wherein, The identification of the at least one text region to obtain the at least one text content comprises: feature extraction is performed on the text region to obtain feature information of the text region; The feature information is reshaped to obtain reshaped feature information; The reshaped feature information is mapped into a preset text probability space to obtain a text mapping probability corresponding to the reshaped feature information; The text content is determined based on the text mapping probability.

6. The method of claim 2, wherein, The de-noising of the image to be processed to obtain a de-noised image comprises: The de-noising model is used to de-noise the image to be processed to obtain a de-noised image; The de-noising of the image to be processed to obtain a de-noised image comprises: The text processing model is used to detect at least one text region in the de-noised image; The identification of the at least one text region to obtain the at least one text content comprises: The text processing model is used to identify the at least one text region to obtain the at least one text content.

7. The method of claim 1, wherein, The determination of the target processing result corresponding to the text content from the processing result based on the association information comprises: The association information is read to obtain identification information corresponding to the text content; The processing identification of the processing result and the identification information are matched to obtain a matching result; The target processing result corresponding to the text content is determined from the processing result based on the matching result.

8. The method of claim 7, wherein, The method further comprises: The text content is identified to obtain a content type corresponding to the text content; Based on the content type, the text content in the image to be processed and the target processing result corresponding to the text content are arranged and reorganized according to a preset arrangement logic to obtain target structured data; The target structured data is statistically processed to obtain a processing indicator corresponding to the target processing result.

9. An information processing apparatus, characterized by comprising: Comprise: An acquisition unit is configured to acquire an image to be processed; A text detection unit is configured to perform text detection on the image to be processed to obtain at least one text content in the image to be processed; An association unit is configured to identify the text content to obtain a content type corresponding to the text content; Based on the content type and the image to be processed, identification information of the text content is generated, the identification information being obtained by combining a first sub-identification corresponding to the content type, a second sub-identification corresponding to the image to be processed, and a third sub-identification corresponding to the text content; and the identification information of the text content in the image to be processed is integrated to obtain association information; The encryption unit is configured to determine a first initial storage address of preset image text content in a preset storage space and determine a second initial storage address of the to-be-processed image text content in the preset storage space; the preset image text content is stored in continuous storage units in the preset storage space; and the to-be-processed image text content is stored in continuous storage units in the preset storage space. The first initial storage address is calculated by using preset rearrangement logic to obtain a first rearranged storage address of the to-be-processed image text content, and the second initial storage address is calculated by using preset rearrangement logic to obtain a second rearranged storage address of the preset image text content; the preset rearrangement logic is logic with random properties; the storage position of the to-be-processed image text content in the preset storage space is updated according to the first rearranged storage address, and the storage position of the preset image text content in the preset storage space is updated according to the second rearranged storage address, so as to mix and rearrange the text content in the preset image and the text content in the to-be-processed image to obtain encrypted text content; The result obtaining unit is configured to send the encrypted text content to a data processing device and obtain a processing result of processing the encrypted text content from the data processing device. The result determining unit is configured to determine a target processing result corresponding to the text content from the processing result based on the association relationship information.

10. A computer program product comprising computer programs or instructions, characterized in that, The computer program or instructions are executed by the processor to implement the steps in the information processing method of any one of claims 1 to 8.

11. A computer device, characterized by The memory stores an application program, and the processor is configured to run the application program in the memory to perform the operations in the information processing method of any one of claims 1 to 8.

12. A storage medium, characterized by The storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by the processor to perform the steps in the information processing method of any one of claims 1 to 8.

Images