Financing material tampering detection method and device, electronic equipment and storage medium

Abstract

The invention discloses a financing material tampering detection method and device, electronic equipment and a storage medium, the method and device are applied to the electronic equipment, and the method specifically comprises the steps that a to-be-detected financing material is acquired, and the financing material is an image file in a digital image format; and processing the financing material based on a pre-constructed PDDNet network structure model to obtain a tampering score, the tampering score being used for reflecting the possible tampering degree of each part of the financing material. In other words, the user can automatically obtain the tampering score reflecting the tampering degree only by inputting the financing material, and tampering does not need to be detected manually, so that the detection efficiency is improved, and misjudgment caused by human reasons can be avoided.

Description

Technical Field

[0001] The present application relates to the field of anti-counterfeiting technology, and more specifically, to a method, device, electronic device and storage medium for detecting tampering of financing materials. Background Art

[0002] In the process of corporate financing, financing materials are usually scanned and backed up in digital image format. In order to ensure the authenticity and fairness of loan decisions, maintain fair market order, and improve loan efficiency and accuracy, it is necessary to conduct authenticity detection on financing materials, or tamper detection on them, and determine the authenticity and fairness of the materials through the test results. At present, tamper detection of financing materials is generally carried out manually, resulting in low detection efficiency and misjudgment due to human factors. Summary of the invention

[0003] In view of this, the present application provides a financing material tampering detection method, device, electronic device and storage medium, which are used to detect whether the financing materials have been tampered with, so as to improve the detection efficiency.

[0004] In order to achieve the above objectives, the proposed solution is as follows:

[0005] A method for detecting tampering of financing materials, applied to electronic equipment, comprises the following steps:

[0006] Acquire financing materials to be tested, wherein the financing materials are image files in a digital image format;

[0007] The financing materials are processed based on a pre-built PDDNet network structure model to obtain a tampering score, which is used to reflect the extent to which various parts of the financing materials may have been tampered with.

[0008] Optionally, the PDDNet network structure model includes an encoding module, a detection module and a decoding module, wherein:

[0009] The encoding module is used to extract the features of different layers of the financing material, and the features of different layers constitute a feature map;

[0010] The decoding module is used to adaptively adjust the feature map using an attention mechanism;

[0011] The detection module is used to compress and remap the high-dimensional features of the feature map into low-dimensional numerical values ​​to generate the tampering score.

[0012] Optionally, when extracting features of different layers, the encoding module performs feature extraction according to the original scale of the financing material.

[0013] Optionally, the encoding module uses a 2x2 convolution kernel to achieve downsampling.

[0014] Optionally, the encoding module uses transposed convolution and linear sampling to achieve upsampling.

[0015] Optionally, a binary cross entropy loss function is used to define a loss function when training the PDDNet network structure model to perform detection and positioning tasks.

[0016] Optionally, the loss function is:

[0017] L'=L bce (s d ,l d )+(∑ 4 m＝1 L bce (M m ,G m ))÷4

[0018] Where L' represents the loss function, L bce represents the binary cross entropy loss function, M m represents the predicted mask, G m represents the mask of the true label, m represents four masks of different scales, and the value range of m is 1 to 4.

[0019] A tamper detection device for financing materials, applied to electronic equipment, comprising:

[0020] A material acquisition module is configured to acquire financing materials to be detected, wherein the financing materials are image files in a digital image format;

[0021] The detection execution module is configured to process the financing materials based on a pre-built PDDNet network structure model to obtain a tampering score, where the tampering score is used to reflect the extent to which various parts of the financing materials may have been tampered with.

[0022] An electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0023] The memory is used to store computer programs or instructions;

[0024] The processor is used to execute the computer program or instruction so that the electronic device implements the tampering detection method as described above.

[0025] A computer-readable storage medium is applied to an electronic device, wherein the storage medium carries one or more computer programs, and the one or more computer programs can be executed by the electronic device, so that the electronic device can implement the tampering detection method as described above.

[0026] It can be seen from the above technical solutions that the present application discloses a method, device, electronic device and storage medium for detecting tampering of financing materials. The method and device are applied to electronic devices, specifically to obtain financing materials to be detected, which are image files in digital image format; the financing materials are processed based on the pre-built PDDNet network structure model to obtain tampering scores, which are used to reflect the extent to which each part of the financing materials may be tampered with. That is, the user only needs to input the financing materials to automatically obtain the tampering score reflecting the degree of tampering, without the need to manually detect tampering, thereby improving the detection efficiency and avoiding the occurrence of misjudgments due to human factors.

[0027] In addition, the PDDNetwork structure model combines the encoding module, decoding module and attention mechanism to make tampering detection more efficient. This combination can effectively detect tampering in financing materials, improve the security and integrity of documents, and thus enhance data protection capabilities.

[0028] The PDDNet network structure model provides an end-to-end detection solution, which simplifies the tampering detection process. This makes detection more intuitive and efficient, reduces the possibility of manual intervention and misjudgment, and improves the credibility of detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0030] Figure 1 A flowchart of a method for detecting tampering of financing materials according to an embodiment of the present application;

[0031] Figure 2 A network structure diagram of the PDDNet network structure model of an embodiment of the present application;

[0032] Figure 3 A schematic diagram of the internal backbone of the encoding module of an embodiment of the present application;

[0033] Figure 4 A schematic diagram of the internal backbone of a decoding module according to an embodiment of the present application;

[0034] Figure 5 A block diagram of a tamper detection device for financing materials according to an embodiment of the present application;

[0035] Figure 6 A block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0037] Figure 1 This is a flow chart of a method for detecting tampering of financing materials according to an embodiment of the present application.

[0038] like Figure 1 As shown, the tampering abbreviation method provided in this embodiment is applied to an electronic device, which is used to tamper with all or part of the financing materials to determine the authenticity and fairness of the financing materials. The electronic device can be understood as a computer, server or cloud platform with data computing capabilities. The tampering detection method specifically includes the following steps:

[0039] S1. Obtain the financing materials to be tested.

[0040] The financing material is an image file in digital image format, which actually contains various documents and certification materials in the enterprise financing process.

[0041] S2. The tampering score is obtained by processing the financing materials based on the PDDNet network structure model.

[0042] That is, the financing materials are input into the PDDNet network structure model obtained by pre-training, so that the model outputs a tampering score, which is used to reflect the extent to which each part of the financing materials may be tampered with. Generally speaking, the larger the value, the greater the possibility of tampering. Of course, the opposite regulation or setting can also be made, that is, the smaller the value, the greater the possibility of tampering.

[0043] like Figure 2 As shown, the PDDNet network structure model of the present application includes three modules, namely, an encoding module, a decoding module and a detection module.

[0044] The encoding module is implemented through a modified residual module, which extracts features of different layers at the original scale. For layers with larger feature maps, more encoding modules need to be stacked to expand the receptive field and gradually capture the key feature information of the image. The decoding module uses the attention mechanism to perform feature adaptive adjustment to ensure that the output detection results are robust and not easily disturbed by factors such as noise. At the same time, the decoding module can decode the encoded features into feature maps of the same size and dimension as the original image to maintain the validity and integrity of the features. The detection module compresses and remaps high-dimensional features into low-dimensional values ​​to generate specific tampering scores. The tampering score reflects the degree to which each part of the image may have been tampered with. The larger the value, the greater the possibility of tampering.

[0045] The encoding module uses a 2×2 convolution kernel to achieve downsampling during the encoding process, and reduces the dimension of the feature map through convolution operations to improve computational efficiency and retain key information. Upsampling includes the use of transposed convolution and linear upsampling. Transposed convolution upsampling helps to restore potential tampering features, while linear upsampling helps to reduce the risk of overfitting and hardware computing power consumption when the feature map size is greater than twice.

[0046] In addition, for decoding modules at different depths, maximum pooling downsampling and linear upsampling are used to obtain cross-layer progressive features. These progressive features can be passed as input to the next module layer by layer, deeply mining feature information, which is conducive to the network to achieve pixel-level tampering detection.

[0047] The internal backbone of the encoding module is as follows Figure 3 As shown in Figure 1, where CA stands for channel attention mechanism. In the design, two important strategies are adopted, which bring the following two benefits: ① Parallel calculation of features of different scales. Features of different scales are calculated in parallel, so effective information exchange can be achieved, which is conducive to handling scale changes. Dense connections enable features between different scales to exchange information directly, thereby enhancing the network's adaptability to scale changes. ② Local and global feature transfer. Local and global feature transfer is performed at each scale to ensure that each feature contains enough information to predict the tampering mask at the corresponding scale. The overall feature transfer conforms to the progressive mechanism, and the prediction of each mask depends on all local and global features, which improves the accuracy of the prediction. In addition to the predicted mask at the last scale, other masks are used as a priori for the prediction of the mask at the next scale, further improving the accuracy and reliability of tampering detection.

[0048] The detection module first performs the reverse operation of the encoding module to extract features from the top-down path to predict the detection score. Subsequently, the operation mask is generated through a fully supervised progressive mechanism. This process achieves a progressive processing of features from coarse to fine, forming a bionic structural design. Inside the detection module, a large number of dimensionality reduction convolution operations are used.

[0049] Assume that the input financing material image is I∈R H×W×3 , so the output features of the first to fourth layers of the encoding module are: F1∈R H×W×C , F2∈R H / 2×W / 2×2C , F3∈R H / 4×W / 4×4C , F4∈R H / 8×W / 8×8C The mask feature Mn (n = 1, 2, 3, 4) is the identity mapping of Fn, and its mapping relationship can be expressed as formula (1):

[0050] M n-1 = f n-1 (τ(M n )·F n-1 ), n = 2,3,4 (1)

[0051] Among them, f n represents the decoder output feature of the nth layer, and τ is the upsampling operation. For the mask of the bottom layer M4, it can be directly expressed as M4=f4(F4).

[0052] The internal backbone of the decoding module is as follows Figure 4 As shown in , Trans represents transposed convolution. In order to reduce the difficulty of prediction, the progressive mechanism we proposed avoids directly generating masks at the finest scale. Since the attention mechanism is usually used to adjust the learned feature weights according to the relative importance of the features, the decoding module uses spatial attention to aggregate pixel-level features based on contextual relevance, and uses channel attention to merge feature maps based on channel relevance. The spatial attention association formula formed after Softmax in the figure can be expressed as formula (2):

[0053] Y' s =A s X' g =softmax(X' θ X' T φ )X' g (2)

[0054] Y' s Represents the spatial attention feature output, and the spatial feature weight is represented by A s ∈R HW / 4×HW / 4 , the feature after 1×1 convolution is represented as X' g , X' T φ Represents X' g The result after transposed convolution. Similarly, the channel attention association formula can be expressed as formula (3):

[0055] Y' c =X' g A c=X' g softmax(X' T θ X' φ ) (3)

[0056] Y' c represents the channel attention feature output, and the spatial feature weight is represented by A c Therefore, the entire decoding module can be expressed as formula (4):

[0057] Z=X+ɑ s ×ω s (Y s )+α c ω c (Y c ) (4)

[0058] ω s and ω c Respectively represent 1×1 convolution, ɑ s and α c represents the learning parameters, X represents the input features, and Z is the feature representation output by the decoding module, which is used to generate the prediction mask.

[0059] In this application, the binary cross entropy loss function (L bce ) to perform detection and localization tasks. The predicted detection score (s d ) is subject to the true label (l d ) supervision. Where d = 0 represents the original image and d = 1 represents the forged image. In addition, by downsampling the mask G1 of the real label to G2, G3 and G4. Through the progressive feature transfer mechanism, predictions are made at different scales. Therefore, the final loss function of this patent can be expressed as:

[0060] L'=L bce (s d ,l d )+(∑ 4 m＝1 L bce (M m ,G m ))÷4 (5)

[0061] Among them, L' represents the final loss function, L bce represents the binary cross entropy loss function, M m and G m They represent the predicted mask and the mask of the true label respectively, and m ranges from 1 to 4, indicating masks of four different scales.

[0062] It can be seen from the above technical solution that this embodiment provides a method for detecting tampering of financing materials, which is applied to electronic devices, specifically obtaining financing materials to be detected, which are image files in digital image format; processing the financing materials based on the pre-built PDDNet network structure model to obtain tampering scores, which are used to reflect the degree to which each part of the financing materials may be tampered. That is, the user only needs to input the financing materials to automatically obtain the tampering score reflecting the degree of tampering, without the need to manually detect tampering, thereby improving the detection efficiency and avoiding the occurrence of misjudgment due to human factors.

[0063] In addition, the PDDNetwork structure model combines the encoding module, decoding module and attention mechanism to make tampering detection more efficient. This combination can effectively detect tampering in financing materials, improve the security and integrity of documents, and thus enhance data protection capabilities.

[0064] The PDDNet network structure model provides an end-to-end detection solution, which simplifies the tampering detection process. This makes detection more intuitive and efficient, reduces the possibility of manual intervention and misjudgment, and improves the credibility of detection.

[0065] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0066] Although operations are depicted in a particular order, this should not be understood as requiring that the operations be performed in the particular order shown or in a sequential order.Multitasking and parallel processing may be advantageous under certain circumstances.

[0067] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0068] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as C or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer.

[0069] Figure 5 This is a block diagram of a device for detecting tampering of financing materials according to an embodiment of the present application.

[0070] like Figure 5 As shown, the tampering abbreviation device provided in this embodiment is applied to an electronic device, and is used to tamper with all or part of the financing materials to determine the authenticity and fairness of the financing materials. The electronic device can be understood as a computer, server or cloud platform with data computing capabilities, and the tampering detection device specifically includes a material acquisition module 10 and a detection execution module 20.

[0071] The material acquisition module is used to acquire the financing materials to be tested.

[0072] The financing material is an image file in digital image format, which actually contains various documents and certification materials in the enterprise financing process.

[0073] The detection execution module is used to obtain tampering scores by processing financing materials based on the PDDNet network structure model.

[0074] That is, the financing materials are input into the PDDNet network structure model obtained by pre-training, so that the model outputs a tampering score, which is used to reflect the extent to which each part of the financing materials may be tampered with. Generally speaking, the larger the value, the greater the possibility of tampering. Of course, the opposite regulation or setting can also be made, that is, the smaller the value, the greater the possibility of tampering.

[0075] It can be seen from the above technical solution that this embodiment provides a tampering detection device for financing materials, which is applied to electronic devices, specifically to obtain financing materials to be detected, which are image files in digital image format; based on the pre-built PDDNet network structure model, the financing materials are processed to obtain tampering scores, which are used to reflect the degree to which each part of the financing materials may be tampered. That is, the user only needs to input the financing materials to automatically obtain the tampering score reflecting the degree of tampering, without the need to manually detect tampering, thereby improving the detection efficiency and avoiding the occurrence of misjudgment due to human factors.

[0076] In addition, the PDDNetwork structure model combines the encoding module, decoding module and attention mechanism to make tampering detection more efficient. This combination can effectively detect tampering in financing materials, improve the security and integrity of documents, and thus enhance data protection capabilities.

[0077] The PDDNet network structure model provides an end-to-end detection solution, which simplifies the tampering detection process. This makes detection more intuitive and efficient, reduces the possibility of manual intervention and misjudgment, and improves the credibility of detection.

[0078] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a unit does not limit the unit itself in some cases. For example, the first acquisition unit may also be described as a "unit for acquiring at least two Internet Protocol addresses".

[0079] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0080] Figure 6 A block diagram of an electronic device according to an embodiment of the present application.

[0081] Reference below Figure 6, which shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure. The terminal device in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. The electronic device is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0082] The electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory ROM 602 or a program loaded from an input device 606 into a random access memory RAM 603. In the RAM, various programs and data required for the operation of the electronic device are also stored. The processing device, ROM, and RAM are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0083] Typically, the following devices may be connected to the I / O interface: input devices including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 608 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 609. The communication device 609 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows an electronic device with various devices, it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have alternatively.

[0084] The present application also provides a computer-readable storage medium embodiment.

[0085] The computer-readable storage medium is applied to an electronic device and carries one or more computer programs. When the one or more computer programs are executed by the electronic device, the electronic device obtains the financing materials to be detected, which are image files in digital image format; the financing materials are processed based on the pre-built PDDNet network structure model to obtain tampering scores, which are used to reflect the degree to which each part of the financing materials may be tampered. That is, the user only needs to input the financing materials to automatically obtain the tampering score reflecting the degree of tampering, without the need to manually detect the tampering, thereby improving the detection efficiency and avoiding the occurrence of misjudgment due to human factors.

[0086] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0087] In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0088] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0089] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0090] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0091] The technical solution provided by the present invention is introduced in detail above. Specific examples are used in this article to illustrate the principle and implementation mode of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation mode and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for detecting tampering of financing materials, applied to electronic equipment, characterized in that: The tampering detection method comprises the steps of: Acquire financing materials to be tested, wherein the financing materials are image files in a digital image format; The financing materials are processed based on a pre-built PDDNet network structure model to obtain a tampering score, which is used to reflect the extent to which various parts of the financing materials may have been tampered with.

2. The tampering detection method according to claim 1, characterized in that: The PDDNet network structure model includes an encoding module, a detection module and a decoding module, wherein: The encoding module is used to extract the features of different layers of the financing material, and the features of different layers constitute a feature map; The decoding module is used to adaptively adjust the feature map using an attention mechanism; The detection module is used to compress and remap the high-dimensional features of the feature map into low-dimensional numerical values ​​to generate the tampering score.

3. The tampering detection method according to claim 2, characterized in that: When extracting features of different layers, the encoding module performs feature extraction according to the original scale of the financing material.

4. The tampering detection method according to claim 2, characterized in that: The encoding module uses a 2x2 convolution kernel to achieve downsampling.

5. The tampering detection method according to claim 2, characterized in that: The encoding module uses transposed convolution and linear sampling to achieve upsampling.

6. The tampering detection method according to claim 2, characterized in that: A binary cross entropy loss function is used to define the loss function when training the PDDNet network structure model to perform detection and positioning tasks.

7. The tampering detection method according to claim 6, characterized in that: The loss function is: L ’ =L bce (s d ,l d )+(∑ 4 m＝1 L bce (M m ,G m ))÷4 Among them, L ’ represents the loss function, L bce represents the binary cross entropy loss function, M m represents the predicted mask, G m represents the mask of the true label, m represents four masks of different scales, and the value range of m is 1 to 4.

8. A tamper detection device for financing materials, applied to electronic equipment, characterized in that: The tamper detection device comprises: A material acquisition module is configured to acquire financing materials to be detected, wherein the financing materials are image files in a digital image format; The detection execution module is configured to process the financing materials based on a pre-built PDDNet network structure model to obtain a tampering score, where the tampering score is used to reflect the extent to which various parts of the financing materials may have been tampered with.

9. An electronic device, characterized in that: The electronic device comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs or instructions; The processor is used to execute the computer program or instruction so that the electronic device implements the tampering detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, applied to an electronic device, characterized in that: The storage medium carries one or more computer programs, and the one or more computer programs can be executed by the electronic device, so that the electronic device can implement the tampering detection method according to any one of claims 1 to 7.

Images