A document image encryption method and system

Abstract

The present invention discloses a document image encryption method and system, including: detecting a document image acquisition instruction, judging whether the lens module and the chassis module have not been replaced; if the lens module and the chassis module have not been replaced, judging whether the user has acquisition authority; if the user has acquisition authority, reading the document information of the document image to be acquired, and storing the read document information; if the user does not have acquisition authority, identifying the target document area in the document image to be detected, encrypting the identified target document area, obtaining an encrypted image, and protecting the document information from being acquired. The present invention can realize fast and efficient image reading under the premise of chip encryption verification and authorized login, and directly shielding the document information area when the chip encryption verification passes but the login is not authorized, so as to realize targeted and effective protection of the document information, and greatly improve the user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a document image encryption method and system. Background Art

[0002] With the development of technology, in order to save documents, they are generally stored by scanning or photographing them through image devices. When users need to view documents, they access image devices to obtain the documents.

[0003] In the prior art, imaging devices usually directly capture and store documents. When unauthorized users access files, the files cannot be encrypted and isolated, and the document contents cannot be protected, which can easily cause the loss of important confidential file contents.

[0004] Therefore the prior art needs further development. Summary of the invention

[0005] In response to the above technical problems, an embodiment of the present invention provides a document image encryption method and system, which can solve the technical problem that in the prior art, image devices usually directly capture and store documents, and when unauthorized users access the files, the files cannot be encrypted and isolated, and the document contents cannot be protected, which easily causes the loss of important confidential file contents.

[0006] A first aspect of an embodiment of the present invention provides a document image encryption method, which is applied to an image device, wherein the image device is provided with a lens module and a chassis module, and the method comprises:

[0007] Detecting a document image acquisition instruction, determining whether the lens module and the chassis module have not been replaced;

[0008] If the lens module and the chassis module are not replaced, determining whether the user has the collection authority;

[0009] If the user has the collection authority, the document information of the document image to be collected is read and the read document information is stored;

[0010] If the user does not have the collection authority, the target document area in the document image to be detected is identified, and the identified target document area is encrypted to obtain an encrypted image to protect the document information from being obtained.

[0011] Optionally, the detecting of the document image acquisition instruction and determining whether the lens module and the chassis module are not replaced include:

[0012] Upon detecting a document image acquisition instruction, the reserved keys of the lens module and the encryption chip of the chassis are verified respectively;

[0013] Determining whether the lens module and the chassis module have not been replaced according to the verification result of the reserved key;

[0014] If the reserved key verification is successful, it is determined that the lens module and the chassis module have not been replaced;

[0015] If the reserved key verification fails, the lens module and / or the chassis module has been replaced.

[0016] Optionally, if the lens module and the chassis module are not replaced, determining whether the user has collection authority includes:

[0017] If the lens module and the chassis module are not replaced, obtaining user information;

[0018] Parsing the user information to obtain the encryption key and account password corresponding to the user;

[0019] The encryption key and account password are verified, and whether the user has the collection authority is determined based on the results.

[0020] Optionally, if the user does not have the acquisition authority, the target document area in the document image to be detected is identified, the identified target document area is encrypted to obtain an encrypted image, and the document information is protected from being obtained, including:

[0021] If the user does not have the collection authority, the target document area in the document image to be detected is identified based on the deep learning network model to obtain the identified target document area;

[0022] A mask is added to the target document area to obtain an encrypted image, thereby protecting the document information from being obtained.

[0023] Optionally, the identifying the target document region in the document image to be detected based on the deep learning network model to obtain the identified target document region includes:

[0024] Pre-constructing a deep learning network model, wherein the deep learning network model includes a convolution layer and a feature layer, a classification sub-network and a regression sub-network;

[0025] Based on the convolution layer, the feature map size of the document image is reduced to obtain multi-scale features of the object to be detected;

[0026] The multi-scale features of the object to be detected are fused to obtain a feature layer;

[0027] Input the feature layer into the classification subnetwork to obtain the category prediction result for each spatial position;

[0028] The feature layer is input into the regression sub-network, and the target document region is obtained according to the output result of the regression sub-network.

[0029] A second aspect of an embodiment of the present invention provides a document image encryption system, which is applied to an image device, wherein the image device includes a lens module and a chassis module. The system includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the following steps are implemented:

[0030] Detecting a document image acquisition instruction, determining whether the lens module and the chassis module have not been replaced;

[0031] If the lens module and the chassis module are not replaced, determining whether the user has the collection authority;

[0032] If the user has the collection authority, the document information of the document image to be collected is read and the read document information is stored;

[0033] If the user does not have the collection authority, the target document area in the document image to be detected is identified, and the identified target document area is encrypted to obtain an encrypted image to protect the document information from being obtained.

[0034] Optionally, when the computer program is executed by the processor, the following steps are implemented:

[0035] Upon detecting a document image acquisition instruction, the reserved keys of the lens module and the encryption chip of the chassis are verified respectively;

[0036] Determining whether the lens module and the chassis module have not been replaced according to the verification result of the reserved key;

[0037] If the reserved key verification is successful, it is determined that the lens module and the chassis module have not been replaced;

[0038] If the reserved key verification fails, the lens module and / or the chassis module has been replaced.

[0039] Optionally, when the computer program is executed by the processor, the following steps are further implemented:

[0040] If the lens module and the chassis module are not replaced, obtaining user information;

[0041] Parsing the user information to obtain the encryption key and account password corresponding to the user;

[0042] The encryption key and account password are verified, and whether the user has the collection authority is determined based on the results.

[0043] Optionally, when the computer program is executed by the processor, the following steps are further implemented:

[0044] If the user does not have the collection authority, the target document area in the document image to be detected is identified based on the deep learning network model to obtain the identified target document area;

[0045] A mask is added to the target document area to obtain an encrypted image, thereby protecting the document information from being obtained.

[0046] A third aspect of an embodiment of the present invention provides a non-volatile computer-readable storage medium, characterized in that the non-volatile computer-readable storage medium stores computer-executable instructions, which, when executed by one or more processors, enable the one or more processors to execute the above-mentioned document image encryption method.

[0047] In the technical solution provided by the embodiment of the present invention, a document image acquisition instruction is detected, and it is determined whether the lens module and the chassis module have not been replaced; if the lens module and the chassis module have not been replaced, it is determined whether the user has the acquisition authority; if the user has the acquisition authority, the document information of the document image to be acquired is read, and the read document information is stored; if the user does not have the acquisition authority, the target document area in the document image to be detected is identified, and the identified target document area is encrypted to obtain an encrypted image to protect the document information from being acquired. The present invention can realize fast and efficient image reading under the premise of passing chip encryption verification and authorized login, and directly shield the document information area when the chip encryption verification passes but the login is not authorized, thereby eliminating the cumbersome and time-consuming process of encryption and decryption of the traditional image encryption algorithm, realizing targeted and effective protection of document information, and greatly improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flowchart of a preferred embodiment of a document image encryption method in an embodiment of the present invention;

[0049] Figure 2 A schematic diagram of a network backbone of a deep learning network of a preferred embodiment of a document image encryption method in an embodiment of the present invention;

[0050] Figure 3 A schematic diagram of a classification subnetwork and a regression subnetwork of a deep learning network in a preferred embodiment of a document image encryption method in an embodiment of the present invention;

[0051] Figure 4 The figure is a schematic diagram of the hardware structure of another embodiment of a document image encryption system in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0053] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0054] See also Figure 1 , Figure 1 FIG. 2 is a flow chart of an embodiment of a document image encryption method according to an embodiment of the present invention. Figure 1 As shown, including:

[0055] Step S100, detecting a document image acquisition instruction, determining whether the lens module and the chassis module have not been replaced, if so, executing step S200, if not, executing step S500;

[0056] Step S200, determine whether the user has the collection authority, if yes, execute step S300, if no, execute step S400;

[0057] Step S300, reading document information of the document image to be collected, and storing the read document information;

[0058] Step S400, identifying a target document area in the document image to be detected, encrypting the identified target document area to obtain an encrypted image, and protecting the document information from being obtained;

[0059] Step S500: reject the user's document image acquisition request.

[0060] In specific implementation, the system of the embodiment of the present invention mainly consists of a lens module with a built-in detection model and an encryption chip, a chassis module with a document reading function (here, a hardware module that can read chip information such as identity cards, social security cards, bank cards, etc.) and an encryption chip, and a software with user permission login (an installation software package that can be installed in a Windows system or an Android system). The three together constitute a complete document image encryption system. First, when the system starts, it first determines whether the lens and chassis are modules bound to the chip. Both have built-in encryption chips, using the AES-GCM encryption mode, and using encryption and authentication (AEAD) functions at the same time. It uses the encryption algorithm AES and the Galois Counter Mode (GCM) counter mode to achieve efficient encryption and authentication, while providing confidentiality, integrity and authentication. The encryption and authentication of AES-GCM are performed simultaneously. It uses a multiplication operation on the Galois field called "GHASH" to calculate the lens and chassis authentication tags (MAC) respectively, and uses a counter mode called "CTR" for encryption.

[0061] The following is the detailed process of AES-GCM:

[0062] Separate key and initialization vector: AES-GCM uses the 128-bit AES encryption algorithm, so the key length is 128, 192, or 256 bits. In addition, it requires a unique initialization vector (IV) to ensure that each encryption is different. The length of the IV can be 96 bits or more, and basically 96 bits is enough.

[0063] Initialize counter: A counter is a value used to iteratively generate the keystream during encryption and authentication to encrypt and decrypt messages. In AES-GCM, the counter is initialized using CTR mode. CTR mode is a stream encryption mode that takes the value of the counter as input and uses the AES algorithm to generate the keystream. In AES-GCM, the counter is initialized to the value of the IV.

[0064] Generate keystream: Generate keystream using AES algorithm. For each 64-bit keystream block, the counter value will be incremented to generate a different keystream block.

[0065] Encrypting plaintext: The plaintext and the generated keystream blocks are mixed together using an XOR operation to generate the ciphertext. The encryption process is performed byte by byte.

[0066] Calculate the authentication tag (MAC): An algorithm called GHASH is used to calculate the MAC. GHASH is a Galois field-based algorithm used to calculate the authentication tag of a message. It divides the message into 64-bit blocks and mixes them together using multiplication and XOR operations to ultimately generate a 128-bit MAC.

[0067] Generate complete ciphertext: Combine the ciphertext and MAC together to generate the complete ciphertext.

[0068] Decryption: Using the same key, IV, and MAC keystream, decrypt the ciphertext using an XOR operation.

[0069] Verify MAC: Use the same key, IV and plaintext to recalculate MAC, and compare the calculated MAC with the original MAC. If the MAC matches, it means that the ciphertext has not been tampered with, preventing the risk of document information leakage caused by the module being replaced; then guide the user to log in to the corresponding permission account (the corresponding functional software used with the hardware), and open the corresponding collection and storage functions according to the corresponding permissions. For example, when an unauthorized user logs in and the camera lens module detects paper or ID documents, it will automatically add a mask to the paper and ID area, and the chassis ID reading function will also refuse to read the relevant ID information, while the authorized user can directly obtain the document information of the paper or ID when logging in. Compared with the original image encryption and decryption technology, the present invention is faster, because the present invention only needs to encrypt the image document area for unauthorized users, and authorized users can use it normally without decryption. The combination of software and hardware encryption also makes security more reliable.

[0070] Further, the document image acquisition instruction is detected, and it is determined whether the lens module and the chassis module are not replaced, including:

[0071] Upon detecting a document image acquisition instruction, the reserved keys of the lens module and the encryption chip of the chassis are verified respectively;

[0072] Determining whether the lens module and the chassis module have not been replaced according to the verification result of the reserved key;

[0073] If the reserved key verification is successful, it is determined that the lens module and the chassis module have not been replaced;

[0074] If the reserved key verification fails, the lens module and / or the chassis module has been replaced.

[0075] In specific implementation, after the device is connected and powered on, the first step is self-start detection, with the focus on software verification of the reserved key of the encryption chip on the camera lens and chassis. If the reserved key verification succeeds, it is determined that the lens module and the chassis module have not been replaced; if the reserved key verification fails, the lens module and / or the chassis module have been replaced. Ensure that the lens and chassis of the device have not been replaced, and ensure the security of document information from the hardware.

[0076] Furthermore, if the lens module and the chassis module are not replaced, determining whether the user has collection authority includes:

[0077] If the lens module and the chassis module are not replaced, obtaining user information;

[0078] Parsing the user information to obtain the encryption key and account password corresponding to the user;

[0079] The encryption key and account password are verified, and whether the user has the collection authority is determined based on the results.

[0080] In specific implementation, after the hardware is read correctly, the software further verifies the login user's authority. Here, the built-in license is combined (using an asymmetric encryption algorithm: RSA algorithm to generate a public-private key pair. Before encrypting the configuration information, the MD5 value is calculated using the customer information (business ID) and hardware information (i.e., the identity authentication mark (MAC) of the camera and chassis mentioned above). Then the value is encrypted with our private key and attached to the back of the configuration information as a signature. Then the entire information is encrypted with a symmetric key to generate the final ciphertext, which can be used as our final License. Even if others decompile the code and obtain the symmetric key and RSA public key, they can only see our configuration information and cannot modify it, because the signature attached to the configuration information can only be decrypted with the public key, and it cannot be tampered with and encrypted by yourself without the private key) and the account password (special document processing software used with the camera and chassis module, with account and password login) are used to ensure the security of the software authorized user login.

[0081] In some other embodiments, a Pid-vid combined with firmware information may be added for user verification. The full name of VID is Vender ID, which means manufacturer number in Chinese. The full name of PID is Product ID, which means the internal product number of the manufacturer in Chinese. Usually, the VID and PID of a device are composed of a string of characters such as: VID_05A9&PID_2800, and the "-" represents the specific VID and PID values. The VID and PID values ​​are actually the IDs used as unique identifiers when the USB device is connected to the Windows system. Adding the PID-VID to the environment of software and hardware verification can further ensure the security of the system and prevent leaks caused by hardware replacement.

[0082] Furthermore, if the user does not have the collection authority, the target document area in the document image to be detected is identified, the identified target document area is encrypted to obtain an encrypted image, and the document information is protected from being obtained, including:

[0083] If the user does not have the collection authority, the target document area in the document image to be detected is identified based on the deep learning network model to obtain the identified target document area;

[0084] A mask is added to the target document area to obtain an encrypted image, thereby protecting the document information from being obtained.

[0085] In specific implementation, when an unauthorized user logs in, the built-in target detection model in the lens module starts running. Its main function is to detect and locate areas with document information such as papers and certificates in the image, and directly cover the detected document information areas with the corresponding mask (pure white area) to shield and block the document information.

[0086] Traditional image encryption methods usually require the design of complex encryption and decryption algorithms to encrypt and decrypt images respectively. The reliability of the method is usually positively correlated with the complexity of the algorithm, which often leads to longer time consumption. The embodiment of the present invention directly determines whether to add a mask through authorized login. If you have the authority, you can directly obtain the normal image without decoding. If you don't have the authority, you can directly get the image with the mask added. The time consumption is greatly shortened without the need for complex encryption and decryption operations. At the same time, only adding a mask to the detected document area also reduces the encryption time consumption, and the encryption of the document image is completed without the user noticing.

[0087] Image mask encryption is used in conjunction with lens locking and hardware offline. The former encrypts the image document area according to the login user's permissions (only images of users without permissions will be masked), greatly improving the user's document image encryption experience; the latter prevents system hardware devices from being maliciously replaced through lens locking and hardware offline, eliminating the risk of document image leakage.

[0088] Furthermore, the target document region in the document image to be detected is identified based on the deep learning network model to obtain the identified target document region, including:

[0089] Pre-constructing a deep learning network model, wherein the deep learning network model includes a convolution layer and a feature layer, a classification sub-network and a regression sub-network;

[0090] Based on the convolution layer, the feature map size of the document image is reduced to obtain multi-scale features of the object to be detected;

[0091] The multi-scale features of the object to be detected are fused to obtain a feature layer;

[0092] Input the feature layer into the classification subnetwork to obtain the category prediction result for each spatial position;

[0093] The feature layer is input into the regression sub-network, and the target document region is obtained according to the output result of the regression sub-network.

[0094] When implementing it, Figure 2 and Figure 3 As shown, the convolutional neural network can extract excellent feature maps of objects to be detected in the image. The fusion of these features will improve the richness of the feature information of the objects to be detected, which will facilitate the subsequent use of loss functions to discriminate and locate and lay a solid foundation. An effective multi-scale feature pyramid is constructed by using bottom-up paths and lateral connections. The low-level feature receptive field is small but the scale is large, so small targets can be detected; the high-level feature receptive field is large and small, so it can be used to detect large targets. The object detection network backbone of the present invention is composed of a convolutional layer and a feature layer. The main function is to extract the corresponding features of the objects to be detected while gradually reducing the size of the feature map. Finally, multi-scale fusion is performed to obtain the final multi-scale fusion feature layer 3, multi-scale fusion feature layer 4, multi-scale fusion feature layer 5, multi-scale fusion feature layer 6 and multi-scale fusion feature layer 7, and the output results of each feature layer are respectively input into the subsequent classification subnetwork and regression subnetwork.

[0095] like Figure 2As shown in the figure, in the classification subnet, K refers to the number of categories, A is the number of anchors (that is, the number of predicted boxes), W is the pixel width of the feature map, H is the pixel height of the feature map, and the pyramid fusion feature map with 256 channels obtained in the previous step. The subnet first obtains 4 post-convolution feature layers through 3×3 convolution. Each convolution layer is followed by a ReLU layer activation, and then a convolution feature map with a KA layer is obtained through a 3x3 convolution layer. KA represents the probability of A boxes being K categories respectively. Finally, a sigmoid activation layer is attached to output the category prediction for each spatial position:

[0096]

[0097] The corresponding derivative is:

[0098] The output range of the sigmoid function is 0 to 1. Since the output value is between 0 and 1, it is equivalent to normalizing the output of each neuron.

[0099] The maximum value obtained after sigmoid activation is the category of the object to be detected in the feature map;

[0100] The regression subnet (Box subnet) is parallel to the classification subnet, and the network structure is basically similar. The difference is that the last output convolution layer is width x height x 4A, where 4A represents the offset and scaling change of the output anchor relative to the real anchor. The four values ​​are (Δx, Δy, Δw, Δh), Δx, Δy represent the offset relative to the real anchor center, Δw, Δh represent the size coefficient of scale change, and the updated anchor center position (x', y') and length and width h', w' can be calculated by the following formula:

[0101]

[0102] The regression subnetwork finally determines the position of the circumscribed rectangle of the detected object, and directly masking the rectangular frame completes the encryption of the document information, preventing the leakage of the document information in the event of unauthorized login.

[0103] It can be seen from the above method embodiments that the embodiments of the present invention quickly and efficiently read images under the premise of chip encryption verification and authorized login, and directly shield the document information area when the chip encryption verification is passed but the login is not authorized, thereby eliminating the cumbersome and time-consuming process of encryption and decryption of the traditional image encryption algorithm, effectively protecting the document information in a targeted manner, and greatly improving the user experience.

[0104] It should be noted that there is not necessarily a certain order between the above-mentioned steps. A person skilled in the art can understand, based on the description of the embodiments of the present invention, that in different embodiments, the above-mentioned steps may have different execution orders, that is, they may be executed in parallel, may be executed interchangeably, and so on.

[0105] The document image encryption method in the embodiment of the present invention is described above. The document image encryption system in the embodiment of the present invention is described below. Figure 4 , Figure 4 1 is a schematic diagram of the hardware structure of another embodiment of a document image encryption system according to an embodiment of the present invention. The document image encryption system is applied to an image device, and a lens module and a chassis module are provided on the image device, such as Figure 4 As shown, the system 10 includes: a processor 101, a memory 102, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor 101, the following steps are implemented:

[0106] Detecting a document image acquisition instruction, determining whether the lens module and the chassis module have not been replaced;

[0107] If the lens module and the chassis module are not replaced, determining whether the user has the collection authority;

[0108] If the user has the collection authority, the document information of the document image to be collected is read and the read document information is stored;

[0109] If the user does not have the collection authority, the target document area in the document image to be detected is identified, and the identified target document area is encrypted to obtain an encrypted image to protect the document information from being obtained.

[0110] The specific implementation steps are the same as those in the method embodiment and will not be repeated here.

[0111] Optionally, when the computer program is executed by the processor 101, the following steps are further implemented:

[0112] Upon detecting a document image acquisition instruction, the reserved keys of the lens module and the encryption chip of the chassis are verified respectively;

[0113] Determining whether the lens module and the chassis module have not been replaced according to the verification result of the reserved key;

[0114] If the reserved key verification is successful, it is determined that the lens module and the chassis module have not been replaced;

[0115] If the reserved key verification fails, the lens module and / or the chassis module has been replaced.

[0116] The specific implementation steps are the same as those in the method embodiment and will not be repeated here.

[0117] Optionally, when the computer program is executed by the processor 101, the following steps are further implemented:

[0118] If the lens module and the chassis module are not replaced, obtaining user information;

[0119] Parsing the user information to obtain the encryption key and account password corresponding to the user;

[0120] The encryption key and account password are verified, and whether the user has the collection authority is determined based on the results.

[0121] The specific implementation steps are the same as those in the method embodiment and will not be repeated here.

[0122] Optionally, when the computer program is executed by the processor 101, the following steps are further implemented:

[0123] If the user does not have the collection authority, the target document area in the document image to be detected is identified based on the deep learning network model to obtain the identified target document area;

[0124] A mask is added to the target document area to obtain an encrypted image, thereby protecting the document information from being obtained.

[0125] The specific implementation steps are the same as those in the method embodiment and will not be repeated here.

[0126] Optionally, when the computer program is executed by the processor 101, the following steps are further implemented:

[0127] Pre-constructing a deep learning network model, wherein the deep learning network model includes a convolution layer and a feature layer, a classification sub-network and a regression sub-network;

[0128] Based on the convolution layer, the feature map size of the document image is reduced to obtain multi-scale features of the object to be detected;

[0129] The multi-scale features of the object to be detected are fused to obtain a feature layer;

[0130] Input the feature layer into the classification subnetwork to obtain the category prediction result for each spatial position;

[0131] The feature layer is input into the regression sub-network, and the target document region is obtained according to the output result of the regression sub-network.

[0132] The specific implementation steps are the same as those in the method embodiment and will not be repeated here.

[0133] An embodiment of the present invention provides a non-volatile computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by one or more processors, for example, to execute the above-described Figure 1 The method comprises steps S100 to S500.

[0134] As an example, non-volatile storage media can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) as external cache memory. By way of illustration and not limitation, RAM can be obtained in many forms such as synchronous RAM (SRAM), dynamic RAM, DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM) and direct Rambus RAM (DRRAM). The disclosed memory components or memories of the operating environment described in the embodiments of the present invention are intended to include one or more of these and / or any other suitable types of memories.

[0135] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A document image encryption method, characterized in that: Applied to an image device, the image device is provided with a lens module and a chassis module, the chassis module is a module with a hardware module and an encryption chip that can read chip information of an ID card, a social security card or a bank card, and the method includes: Detecting a document image acquisition instruction, determining whether the lens module and the chassis module have not been replaced; If the lens module and the chassis module are not replaced, determining whether the user has the collection authority; If the user has the collection authority, the document information of the document image to be collected is read and the read document information is stored; If the user does not have the collection authority, the target document area in the document image to be detected will be identified, and the identified target document area will be encrypted to obtain an encrypted image to protect the document information from being obtained. The detecting of the document image acquisition instruction and determining whether the lens module and the chassis module are not replaced include: Upon detecting a document image acquisition instruction, the reserved keys of the encryption chips of the lens module and the chassis module are verified respectively; Determining whether the lens module and the chassis module have not been replaced according to the verification result of the reserved key; If the reserved key verification is successful, it is determined that the lens module and the chassis module have not been replaced; If the reserved key verification fails, the lens module and / or the chassis module has been replaced.

2. The document image encryption method according to claim 1, characterized in that: If the lens module and the chassis module are not replaced, determining whether the user has the collection authority includes: If the lens module and the chassis module are not replaced, obtaining user information; Parsing the user information to obtain the encryption key and account password corresponding to the user; The encryption key and account password are verified, and whether the user has the collection authority is determined based on the results.

3. The document image encryption method according to claim 2, characterized in that: If the user does not have the acquisition authority, the target document area in the document image to be detected is identified, the identified target document area is encrypted to obtain an encrypted image, and the document information is protected from being obtained, including: If the user does not have the collection authority, the target document area in the document image to be detected is identified based on the deep learning network model to obtain the identified target document area; A mask is added to the target document area to obtain an encrypted image, thereby protecting the document information from being obtained.

4. A document image encryption system, characterized in that: Applied to an image device, the image device includes a lens module and a chassis module, the chassis module is a module with a hardware module and an encryption chip that can read chip information of an ID card, a social security card, or a bank card, the system includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program implements the following steps when executed by the processor: Detecting a document image acquisition instruction, determining whether the lens module and the chassis module have not been replaced; If the lens module and the chassis module are not replaced, determining whether the user has the collection authority; If the user has the collection authority, the document information of the document image to be collected is read and the read document information is stored; If the user does not have the collection authority, the target document area in the document image to be detected is identified, and the identified target document area is encrypted to obtain an encrypted image to protect the document information from being obtained; The detecting of the document image acquisition instruction and determining whether the lens module and the chassis module are not replaced include: Upon detecting a document image acquisition instruction, the reserved keys of the encryption chips of the lens module and the chassis module are verified respectively; Determining whether the lens module and the chassis module have not been replaced according to the verification result of the reserved key; If the reserved key verification is successful, it is determined that the lens module and the chassis module have not been replaced; If the reserved key verification fails, the lens module and / or the chassis module has been replaced.

5. The document image encryption system according to claim 4, characterized in that: When the computer program is executed by the processor, the following steps are implemented: Upon detecting a document image acquisition instruction, the reserved keys of the encryption chips of the lens module and the chassis module are verified respectively; Determining whether the lens module and the chassis module have not been replaced according to the verification result of the reserved key; If the reserved key verification is successful, it is determined that the lens module and the chassis module have not been replaced; If the reserved key verification fails, the lens module and / or the chassis module has been replaced.

6. The document image encryption system according to claim 5, characterized in that: When the computer program is executed by the processor, the following steps are also implemented: If the lens module and the chassis module are not replaced, obtaining user information; Parsing the user information to obtain the encryption key and account password corresponding to the user; The encryption key and account password are verified, and whether the user has the collection authority is determined based on the results.

7. The document image encryption system according to claim 6, characterized in that: When the computer program is executed by the processor, the following steps are also implemented: If the user does not have the collection authority, the target document area in the document image to be detected is identified based on the deep learning network model to obtain the identified target document area; A mask is added to the target document area to obtain an encrypted image, thereby protecting the document information from being obtained.

8. A non-volatile computer-readable storage medium, characterized in that: The non-volatile computer-readable storage medium stores computer-executable instructions, which, when executed by one or more processors, enable the one or more processors to execute the document image encryption method according to any one of claims 1 to 3.

Images