Electronic file verification method and system for digital commerce

Abstract

The invention discloses an electronic file verification method and system for digital commerce. The method comprises the following steps: S1, identifying a plurality of abnormal areas existing in a corresponding scanning image according to local pixel features of each scanning image in an electronic edition file; s2, selecting a to-be-verified region, and matching a plurality of associated regions of which the region similarity is greater than a similarity threshold for the to-be-verified region; s3, obtaining a pixel correction value of each abnormal pixel point, and carrying out simulation correction on pixels of the to-be-verified area to generate a corresponding simulation verification image; and S4, obtaining a plurality of image feature difference values of the simulation verification image, and analyzing to obtain an abnormal type of the corresponding scanning image set so as to identify the authenticity of the corresponding electronic edition file. According to the method, the image error caused by the scanning error of equipment hardware and the image error caused by manually modifying the scanned image can be accurately distinguished, and the authenticity of the electronic file of the corresponding paper file can be identified.

Description

Technical Field

[0001] This invention relates to the field of electronic document verification, and in particular to a method and system for verifying electronic documents for digital commerce. Background Technology

[0002] In today's digital age, digital commerce has become an important development direction. By utilizing advanced technologies such as big data, artificial intelligence, and cloud computing, digital commerce has brought about comprehensive intelligent improvements in business processes, business development, and customer service.

[0003] The digital business industry chain encompasses multiple layers. First, the foundational layer forms the cornerstone of digital business technologies, including technological infrastructure such as artificial intelligence, big data, and cloud computing, as well as data sources and customer base. This infrastructure provides the necessary support and resources for digital business. At the technology layer, cloud computing plays a crucial role. It provides digital business with elastic and scalable computing and storage resources, helping to improve operational efficiency, reduce costs, and respond quickly to market changes.

[0004] In existing application scenarios, when paper documents are converted to electronic versions for sending and storage, manual review of the electronic versions is required for accuracy, especially for the verification of seals. Manual review is labor-intensive and costly, with low efficiency and accuracy heavily reliant on the reviewer's work performance, making errors easy to occur. Furthermore, when image errors arise in the electronic version of the corresponding paper document due to scanning errors in the device hardware, timely correction and response are also impossible. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide an electronic document verification method and system for digital commerce, which can accurately distinguish between image errors caused by scanning errors of device hardware and image errors caused by human modification of scanned images, thereby enabling the authentication of electronic versions of corresponding paper documents.

[0006] The objective of this invention is achieved through the following technical solution: a method for verifying electronic documents in digital commerce, comprising the following steps: S1. Identify several abnormal regions in the corresponding scanned images based on the local pixel features of each scanned image in the electronic file; S2. Select the abnormal region with the largest abnormal value from all abnormal regions corresponding to all scanned images as the region to be verified in the corresponding scanned image set, and match several associated regions with a similarity greater than a similarity threshold from all scanned images according to the first region feature of the region to be verified. S3. Map the second region features of each associated region to the region to be verified to obtain the pixel correction value of each abnormal pixel, and perform simulated correction on the pixels of the region to be verified according to the pixel correction value of each abnormal pixel to generate a corresponding simulated verification image; S4. The simulated verification image is compared with the standard verification image to obtain several image feature differences of the simulated verification image, and the anomaly type of the corresponding scanned image set is analyzed based on the image feature differences and a preset threshold to identify the authenticity of the corresponding electronic file.

[0007] The beneficial effects of this invention are as follows: This invention identifies several abnormal regions with abnormal pixel values ​​in the scanned image of the corresponding paper document uploaded by the user terminal, and corrects the pixels in the corresponding abnormal regions according to the pixel correction values ​​of the abnormal pixels to generate a corresponding simulated verification image. Then, the anomaly type of the corresponding scanned image is verified by the feature difference between the simulated verification image and the standard verification image of the corresponding paper document. This automatically and accurately identifies the cause of pixel anomalies in the electronic version of the corresponding paper document, and can accurately distinguish between image errors caused by scanning errors of the device hardware and image errors caused by human modification of the scanned image, thus realizing the authenticity authentication of the electronic version of the corresponding paper document. Therefore, it can well meet the needs of electronic document authentication in digital business scenarios, reduce the consumption of manpower and material resources, and improve the efficiency of automated processing. Attached Figure Description

[0008] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the system principle of the present invention. Detailed Implementation

[0009] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0010] like Figure 1 As shown, in one embodiment, the electronic document verification method for digital commerce of the present invention includes: S1. The data acquisition module of the digital business cloud platform (for example, commonly used in the field of smart finance, which can be called a smart financial cloud platform) receives the electronic file uploaded by the user terminal after being scanned by a scanner, and identifies several abnormal regions in the corresponding scanned image based on the local pixel features of each scanned image in the electronic file. The electronic file includes a set of scanned images, the number of image pages, the image format, and the image size.

[0011] Optionally, the scanned image set includes several scanned images, wherein each scanned image corresponds to a paper document; the number of pages in the image is the same as the number of pages in the corresponding paper document; and the image format can be bmp, jpg, or png.

[0012] Specifically, the data acquisition module identifies several abnormal regions in the corresponding scanned images based on the local pixel features of each scanned image in the electronic file, including: The data acquisition module identifies the frequency of occurrence of all extreme points contained in each local region of the corresponding scanned image based on the local pixel features of each scanned image. The local pixel features are obtained by analyzing the pixel values ​​and position information of each pixel in the corresponding local region of the image. The data acquisition module constructs a corresponding frequency curve for the corresponding local region of the image based on the pixel values ​​and occurrence frequency of all extreme points, and compares the frequency curve of the corresponding local region of the image with the frequency curve of the adjacent local regions of the image to identify all abnormal extreme points in the local region of the image. The abnormal extreme points are extreme points in the corresponding frequency curve whose occurrence frequency is greater than the frequency threshold. The data acquisition module marks all abnormal extreme points in the corresponding local area of ​​the image to obtain a number of abnormal pixels in the local area of ​​the image, and takes different local areas of the image with a number of abnormal pixels as a number of abnormal areas of the corresponding scanned image.

[0013] Optionally, the occurrence frequency is used to characterize the number of times the corresponding extreme point occurs and the travel frequency; the frequency threshold is a value preset by the system to determine whether the occurrence frequency of the extreme point is abnormal.

[0014] Optionally, the extreme points include the maximum and minimum points of pixels in the corresponding local region of the image, wherein the maximum point is the pixel with the largest pixel value compared with the pixel values ​​of each pixel in the neighboring region; The minimum point is the pixel with the smallest pixel value compared to the pixel values ​​of all pixels in the neighboring region; the neighboring region is a region range preset by the system.

[0015] S2. The region matching module selects the abnormal region with the largest abnormal value from all abnormal regions corresponding to all scanned images as the region to be verified in the corresponding scanned image set, and matches several associated regions with a similarity greater than a similarity threshold from all scanned images for the region to be verified based on the first region feature of the region to be verified.

[0016] Specifically, the region matching module selects the abnormal region with the largest outlier from all abnormal regions corresponding to all scanned images as the region to be verified in the corresponding scanned image set, including: The region matching module counts all abnormal pixels in each abnormal region to obtain the number of abnormal pixels in each abnormal region, and identifies several edge abnormal pixels in the corresponding abnormal region based on the location information of each abnormal pixel. The number of abnormal pixels is used to characterize the ratio between abnormal pixels and normal pixels in the corresponding abnormal region. The region matching module constructs corresponding abnormal structure features for the corresponding abnormal region based on the relative distance and relative angle between each edge abnormal pixel, and analyzes the area of ​​the abnormal region according to the region shape indicated by the abnormal structure features. The abnormal structure features are used to characterize the region shape features formed by all abnormal pixels of the corresponding abnormal region. The region matching module obtains the first abnormal coefficient corresponding to the abnormal region area and the second abnormal coefficient corresponding to the abnormal pixel quantity for each abnormal region. It then performs weighted fusion on the abnormal region area and abnormal pixel quantity of each abnormal region to obtain the abnormal value of the corresponding abnormal region. Finally, it selects the abnormal region with the largest abnormal value from all abnormal regions as the region to be verified in the corresponding scanned image set.

[0017] Optionally, the first anomaly coefficient is used to characterize the weight value of the area of ​​the corresponding abnormal region in the outlier analysis process; the second anomaly coefficient is used to characterize the weight value of the number of abnormal pixels in the corresponding abnormal region in the outlier analysis process, and both the first anomaly coefficient and the second anomaly coefficient are preset by the system.

[0018] Specifically, the region matching module matches several associated regions with a similarity greater than a similarity threshold for the region to be verified from all scanned images based on the first region feature of the region to be verified, including: The region matching module uses the image centroid of the corresponding scanned image as the origin of the two-dimensional plane coordinate system to obtain the region position information of the region to be verified relative to the image centroid of the corresponding scanned image. The image centroid is obtained by analyzing the image shape and image area of ​​the corresponding scanned image. The region matching module performs region localization on all remaining scanned images based on the region location information of the region to be verified to obtain local image regions in each remaining scanned image that are in the same position and have the same area as the region to be verified. The region structure features and region texture features of each local image region are compared with the first region features to obtain the region similarity between each local image region and the region to be verified. The remaining scanned images do not include the scanned images corresponding to the region to be verified. The region matching module uses local image regions with a similarity greater than a similarity threshold as associated regions of the region to be verified.

[0019] Optionally, the similarity threshold is a value pre-set by the system to determine whether the corresponding region to be verified and the corresponding local region of the image are similar regions; the region structure features and region texture features are obtained by analyzing the pixel values ​​and position information of each pixel in the local region of the corresponding image.

[0020] Optionally, the region location information includes the maximum and minimum coordinate values ​​of the corresponding region to be verified; the minimum coordinate values ​​are used to characterize the minimum horizontal and vertical coordinate values ​​of the pixels in the corresponding region to be verified. The maximum coordinate values ​​are used to characterize the maximum horizontal and vertical coordinate values ​​of the corresponding pixels in the region to be verified.

[0021] In another embodiment, the region matching module further includes matching several associated regions with a similarity greater than a similarity threshold for the region to be verified from all scanned images based on the first region feature of the region to be verified: The region matching module obtains the pixel position and pixel value of each normal pixel in the region to be verified, and identifies different local units where each normal pixel in the region to be verified gathers based on the pixel position of each normal pixel. The local unit is used to characterize the local region in the corresponding region to be verified that does not contain abnormal pixels. The region matching module takes all normal pixels belonging to the same local unit as the associated normal pixels of the corresponding local unit, and analyzes the distance features and angle features between each associated normal pixel based on the position information of each associated normal pixel in the corresponding local unit. The associated normal pixels are all normal pixels in the region to be verified that are in the same local unit. The distance features are used to characterize the relative distance between the corresponding associated normal pixels, and the angle features are used to characterize the relative angle between the corresponding associated normal pixels. The region matching module analyzes the unit structure of the corresponding local unit based on the distance and angle features between each associated normal pixel to obtain the unit structure features and unit texture features of the local unit, and aggregates the unit structure features and unit texture features of all local units in the region to be verified to obtain the first region feature of the region to be verified.

[0022] Optionally, the first region feature is obtained by aggregating the unit structure features and unit texture features of all local units corresponding to the region to be verified.

[0023] S3. The image simulation module maps the second region features of each associated region to the region to be verified to obtain the pixel correction value of each abnormal pixel, and performs simulated correction on the pixels of the region to be verified based on the pixel correction value of each abnormal pixel to generate a corresponding simulated verification image.

[0024] Specifically, the image simulation module maps the second region features of each associated region to the region to be verified to obtain the pixel correction value of each abnormal pixel, including: The image simulation module obtains the first feature matching degree between the feature vectors of each normal pixel in the region to be verified and the feature vectors of each pixel in the corresponding associated region, and establishes a corresponding pixel mapping relationship for each normal pixel in the region to be verified based on the first feature matching degree of each feature vector. The pixel mapping relationship is used to characterize the correspondence between the pixel in the associated region and the normal pixel in the region to be verified. The number of regional features contained in the second region feature is greater than the number of regional features contained in the first region feature. The feature vector includes the position feature vector and the pixel feature vector of the normal pixel. The image simulation module maps a portion of the feature vectors in the second region features of each associated region to each abnormal pixel in the region to be verified according to the pixel mapping relationship to obtain the mapping feature value of each abnormal pixel, and uses the mapping feature value as the weight value of the corresponding abnormal pixel. The portion of the feature vectors are the pixels in the corresponding associated region that have not established a pixel mapping relationship with the pixels in the verification region. The image simulation module obtains the pixel values ​​of each pixel in each associated region, and performs weighted fusion of the pixel values ​​of each matching pixel in the associated region according to the weight value of each abnormal pixel in the region to be verified to obtain the pixel correction value of each abnormal pixel. The matching pixel is the pixel in the corresponding associated region with the largest second feature matching degree with the corresponding abnormal pixel.

[0025] Optionally, the first feature matching degree is used to characterize the similarity between each normal pixel in the region to be verified and the corresponding pixel in the associated region; the second feature matching degree is used to characterize the similarity between each abnormal pixel in the region to be verified and the corresponding pixel in the associated region.

[0026] Optionally, the second regional feature is obtained by analyzing the regional structure features and regional texture features of the corresponding associated region.

[0027] S4. The image verification module compares the simulated verification image with the standard verification image to obtain several image feature differences of the simulated verification image, and analyzes the anomaly type of the corresponding scanned image set based on the image feature differences and a preset threshold to identify the authenticity of the corresponding electronic file. The anomaly type includes image anomalies caused by scanning errors of the device hardware and image anomalies caused by human modification.

[0028] Optionally, the standard verification image is an archive of the original electronic version of the corresponding paper document that has not been modified; the preset threshold is a value pre-set by the system to determine whether the content of the file displayed in the simulated verification image after pixel correction is consistent with the content of the original paper document.

[0029] Optionally, when the anomaly type of the scanned image set is identified as an image anomaly caused by human modification, the corresponding electronic file is determined to be a forged file; when the anomaly type of the scanned image set is identified as an image error caused by scanning error of the device hardware, the corresponding electronic file is determined to be a genuine file.

[0030] The technical solution provided by this invention obtains a set of equipment operating values ​​when the production equipment is operating under normal conditions, analyzes the correlation between the changing patterns of various equipment operating values, and detects the operating status of the production equipment within a target monitoring period. This involves comparing the changes in the production equipment's characteristic signals under normal conditions with the changes in the production equipment's characteristic signals within the target monitoring period, thereby accurately and efficiently detecting abnormal states of the equipment. This eliminates the need for highly skilled technicians with extensive field experience, reducing labor costs. Furthermore, it allows for rapid detection in the early stages of abnormal operating conditions of the production equipment, thus lowering equipment maintenance costs.

[0031] See Figure 2 In one embodiment, the electronic document verification system for performing the method of the present invention includes a digital business cloud platform (e.g., a smart financial cloud platform) and a user terminal. The digital business cloud platform and the user terminal have a communication connection. The user terminal is a device used by the document uploader that has computing, storage, and communication functions, including smartphones, desktop computers, and laptops.

[0032] The digital business cloud platform includes a data acquisition module, a region matching module, an image simulation module, and an image verification module.

[0033] The data acquisition module is used to receive electronic files uploaded by user terminals and obtained by scanning, and to identify several abnormal regions in the corresponding scanned images based on the local pixel features of each scanned image in the electronic file. The electronic file includes a set of scanned images, the number of image pages, the image format, and the image size.

[0034] The region matching module is used to select the abnormal region with the largest outlier from all abnormal regions corresponding to all scanned images as the region to be verified in the corresponding scanned image set, and to match several associated regions with a similarity greater than a similarity threshold from all scanned images for the region to be verified based on the first region feature of the region to be verified.

[0035] The image simulation module is used to map the second region features of each associated region to the region to be verified to obtain the pixel correction value of each abnormal pixel, and to perform simulated correction on the pixels of the region to be verified based on the pixel correction value of each abnormal pixel to generate a corresponding simulated verification image.

[0036] The image verification module is used to compare the features of the simulated verification image with the standard verification image to obtain several image feature differences of the simulated verification image, and to analyze the anomaly type of the corresponding scanned image set based on the image feature differences and a preset threshold, so as to identify the authenticity of the corresponding electronic file. The anomaly types include image anomalies caused by scanning errors of the device hardware and image anomalies caused by human modification.

[0037] Furthermore, while specific functions have been discussed above with reference to specific modules, it should be noted that the functions of the modules discussed herein can be divided into multiple modules, and / or at least some functions of multiple modules can be combined into a single module. Additionally, the actions performed by specific modules discussed herein include the specific module itself performing the action, or alternatively, the specific module calling or otherwise accessing another component or module that performs the action (or performing the action in conjunction with the specific module). Therefore, a specific module performing an action can include the specific module performing the action itself and / or another module that the specific module performing the action calls or otherwise accesses.

Claims

1. A method for verifying electronic documents in digital commerce, characterized in that: Includes the following steps: S1. Identify several abnormal regions in the corresponding scanned images based on the local pixel features of each scanned image in the electronic file; S2. Select the abnormal region with the largest abnormal value from all abnormal regions corresponding to all scanned images as the region to be verified in the corresponding scanned image set, and match several associated regions with a similarity greater than a similarity threshold from all scanned images according to the first region feature of the region to be verified. S3. Map the second region features of each associated region to the region to be verified to obtain the pixel correction value of each abnormal pixel, and perform simulated correction on the pixels of the region to be verified according to the pixel correction value of each abnormal pixel to generate a corresponding simulated verification image; S4. The simulated verification image is compared with the standard verification image to obtain several image feature differences of the simulated verification image, and the anomaly type of the corresponding scanned image set is analyzed based on the image feature differences and a preset threshold to identify the authenticity of the corresponding electronic file.

2. The method for verifying electronic documents in digital commerce according to claim 1, characterized in that: The electronic file includes a scanned image set, the number of image pages, the image format, and the image size; The scanned image set includes several scanned images, wherein each scanned image corresponds to one paper document; the number of pages in the image is the same as the number of pages in the corresponding paper document; and the image format is bmp, jpg, or png.

3. The method for verifying electronic documents for digital commerce according to claim 2, characterized in that: Step S1 includes: S101. Receive electronic files uploaded by the user terminal after being scanned by a scanner; S102. Identify several abnormal regions in the corresponding scanned images based on the local pixel features of each scanned image in the electronic file: The frequency of occurrence of all extreme points contained in each local region of the corresponding scanned image is identified based on the local pixel features of each scanned image. The local pixel features are obtained by analyzing the pixel values ​​and position information of each pixel in the corresponding local region of the image. Based on the pixel values ​​and frequency of occurrence of all extreme points, a corresponding frequency curve is constructed for the corresponding local region of the image. The frequency curve of the corresponding local region of the image is compared with the frequency curve of the adjacent local regions of the image to identify all abnormal extreme points in the local region of the image. The abnormal extreme points are the extreme points in the corresponding frequency curves whose frequency of occurrence is greater than the frequency threshold. All abnormal extreme points in the corresponding local area of ​​the image are marked as abnormal to obtain a number of abnormal pixels in the local area of ​​the image, and different local areas of the image containing a number of abnormal pixels are taken as a number of abnormal areas of the corresponding scanned image.

4. The method for verifying electronic documents for digital commerce according to claim 1, characterized in that: In step S2, selecting the abnormal region with the largest outlier from all abnormal regions corresponding to all scanned images as the region to be verified in the corresponding scanned image set includes: The number of abnormal pixels in each abnormal region is counted to obtain the number of abnormal pixels in each abnormal region. Based on the location information of each abnormal pixel, several edge abnormal pixels located at the edge of the corresponding abnormal region are identified. The number of abnormal pixels is used to characterize the ratio between abnormal pixels and normal pixels in the corresponding abnormal region. Based on the relative distance and relative angle between each edge abnormal pixel, a corresponding abnormal structure feature is constructed for the corresponding abnormal region, and the area of ​​the abnormal region is obtained by analyzing the region shape indicated by the abnormal structure feature. The abnormal structure feature is used to characterize the region shape feature formed by all abnormal pixels of the corresponding abnormal region. The first abnormal coefficient corresponding to the area of ​​the abnormal region and the second abnormal coefficient corresponding to the number of abnormal pixels are obtained for each abnormal region. The abnormal region area and the number of abnormal pixels of each abnormal region are weighted and fused to obtain the abnormal value of the corresponding abnormal region. The abnormal region with the largest abnormal value is selected from all abnormal regions as the region to be verified in the corresponding scanned image set.

5. The method for verifying electronic documents in digital commerce according to claim 4, characterized in that: In step S2, based on the first region feature of the region to be verified, several associated regions with a similarity greater than a similarity threshold are matched from all scanned images for the region to be verified, including: The centroid of the corresponding scanned image is used as the origin of the two-dimensional plane coordinate system to obtain the regional position information of the region to be verified relative to the centroid of the corresponding scanned image. The centroid is obtained by analyzing the image shape and image area of ​​the corresponding scanned image. Based on the regional location information of the region to be verified, all remaining scanned images are used to locate the region to obtain local image regions in each remaining scanned image that are in the same position and have the same area as the region to be verified. The regional structure features and regional texture features of each local image region are compared with the first regional features to obtain the regional similarity between each local image region and the region to be verified. The remaining scanned images do not include the scanned images corresponding to the region to be verified. The local regions of the image with a similarity greater than the similarity threshold are used as the associated regions of the region to be verified.

6. The method for verifying electronic documents for digital commerce according to claim 1, characterized in that: In step S3, mapping the second region features of each associated region to the region to be verified to obtain the pixel correction value of each abnormal pixel includes: The first feature matching degree between the feature vectors of each normal pixel in the region to be verified and the feature vectors of each pixel in the corresponding associated region is obtained. Based on the first feature matching degree of each feature vector, a corresponding pixel mapping relationship is established for each normal pixel in the region to be verified. The pixel mapping relationship is used to characterize the correspondence between the pixel in the associated region and the normal pixel in the region to be verified. The number of regional features contained in the second region feature is greater than the number of regional features contained in the first region feature. The feature vector includes the position feature vector and the pixel feature vector of the normal pixel. According to the pixel mapping relationship, a portion of the feature vector in the second region feature of each associated region is mapped to each abnormal pixel in the region to be verified to obtain the mapping feature value of each abnormal pixel, and the mapping feature value is used as the weight value of the corresponding abnormal pixel. The portion of the feature vector is the pixel in the corresponding associated region that has not established a pixel mapping relationship with the pixel in the verification region. The pixel values ​​of each pixel in each associated region are obtained, and the pixel values ​​of each matching pixel in the associated region are weighted and fused according to the weight value of each abnormal pixel in the region to be verified to obtain the pixel correction value of each abnormal pixel. The matching pixel is the pixel in the corresponding associated region with the largest second feature matching degree with the corresponding abnormal pixel.

7. The method for verifying electronic documents in digital commerce according to claim 1, characterized in that: In step S4, the anomaly types include image anomalies caused by scanning errors of the device hardware and image anomalies caused by human modification.

8. The method for verifying electronic documents for digital commerce according to claim 1, characterized in that: The standard verification image is an archive of the original electronic version of the corresponding paper document that has not been modified; the preset threshold is a value pre-set by the system to determine whether the content of the file displayed in the simulated verification image after pixel correction is consistent with the content of the original paper document. When the anomaly type of the scanned image set is identified as an image anomaly caused by human modification, the corresponding electronic file is determined to be a forged file; when the anomaly type of the scanned image set is identified as an image error caused by scanning error of the device hardware, the corresponding electronic file is determined to be a genuine file.

9. An electronic document verification system for digital commerce, comprising verifying electronic documents using the method described in any one of claims 1 to 8, characterized in that: It includes a digital business cloud platform and user terminals, with a communication connection between the digital business cloud platform and the user terminals; The digital business cloud platform includes: The data acquisition module is used to receive electronic files uploaded by the user terminal and obtained by scanning, and to identify several abnormal regions in the corresponding scanned images based on the local pixel features of each scanned image in the electronic file. The electronic file includes a set of scanned images, the number of image pages, the image format, and the image size. The region matching module is used to select the abnormal region with the largest abnormal value from all abnormal regions corresponding to all scanned images as the region to be verified in the corresponding scanned image set, and to match several associated regions with a similarity greater than a similarity threshold from all scanned images for the region to be verified based on the first region feature of the region to be verified. The image simulation module is used to map the second region features of each associated region to the region to be verified to obtain the pixel correction value of each abnormal pixel, and to perform simulated correction on the pixels of the region to be verified based on the pixel correction value of each abnormal pixel to generate a corresponding simulated verification image. The image verification module is used to compare the features of the simulated verification image with the standard verification image to obtain several image feature differences of the simulated verification image, and to analyze the anomaly type of the corresponding scanned image set based on the image feature differences and a preset threshold, so as to identify the authenticity of the corresponding electronic file. The anomaly types include image anomalies caused by scanning errors of the device hardware and image anomalies caused by human modification.

Images