Enterprise financial data analysis management system and method based on cloud computing

Abstract

The embodiment of the invention provides an enterprise financial data analysis management system and method based on cloud computing. The system comprises a financial data entry module and a financial data analysis module. The financial data entry module is used for extracting financial data, encrypting the financial data and uploading the financial data to a storage server of cloud computing. And the financial data analysis module is used for acquiring the corresponding basic financial data from the storage server of the cloud computing according to the financial analysis type selected by the user, and inputting the basic financial data into a financial analysis model corresponding to the financial analysis type to obtain a financial data analysis result. According to the method and the device, the structure of the financial document can be accurately identified when the financial data is extracted by matching the financial document with the pre-stored local financial document template, so that the financial data in the financial document can be accurately extracted, and the accuracy during financial data entry and the accuracy of financial data analysis are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of financial data analysis management, and in particular to a cloud computing-based enterprise financial data analysis management system and method. Background Art

[0002] With the rapid development of science and technology, intelligence has become an important direction for the transformation and upgrading of corporate finance. Through intelligent technology, enterprises have realized data processing automation, accurate analysis and prediction, resource allocation optimization, risk control and decision-making efficiency improvement, and innovated financial management models. Financial data analysis and management technology is developing in a more intelligent, mobile and real-time direction. These changes have brought higher management efficiency and more accurate financial decision-making support to enterprises.

[0003] However, accurate collection of financial data has a crucial impact on the results of financial data analysis. In the prior art, when conducting financial data analysis, the credibility of the results of financial data analysis is often affected by inaccurate financial data collection. How to improve the accuracy of financial data collection has become an urgent problem to be solved. Summary of the invention

[0004] The embodiment of the present invention provides an enterprise financial data analysis management system and method based on cloud computing. By matching the financial document with a pre-stored local financial document template, the structure of the financial document can be accurately identified when extracting financial data, and the financial data in the financial document can be accurately extracted, thereby improving the accuracy of financial data entry and the accuracy of financial data analysis.

[0005] An embodiment of the present invention provides an enterprise financial data analysis and management system based on cloud computing, including a financial data entry module and a financial data analysis module;

[0006] The financial data entry module includes a financial data extraction submodule and a financial data upload submodule;

[0007] The financial data extraction submodule is used to extract first financial data of the first financial document by matching the first financial document with a pre-stored local financial document template, where the first financial data includes basic financial data, financial data type, and a second financial document; the first financial document is a paper document, and the second financial document is an electronic document corresponding to the paper document;

[0008] The financial data uploading submodule is used to divide the first financial data into a plurality of first financial data blocks, encrypt the plurality of first financial data blocks, and upload them to a storage server of cloud computing;

[0009] The financial data analysis module is used to obtain the corresponding basic financial data from the cloud computing storage server according to the financial analysis type selected by the user, input the basic financial data into the financial analysis model corresponding to the financial analysis type, and obtain the financial data analysis results.

[0010] Furthermore, extracting first financial data of the first financial document by matching the first financial document with a pre-stored local financial document template specifically includes:

[0011] Capturing a first image of the first financial document, performing paper edge detection on the first image, and regularizing the edge of the first image to obtain a second image;

[0012] Performing line detection and cell segmentation on the second image to obtain a table structure of the second image;

[0013] According to the financial document type of the second image, obtaining a locally pre-stored financial document template corresponding to the financial document type of the second image;

[0014] Comparing the financial document template with the table structure, and if the matching degree between the two exceeds a preset threshold, determining that the financial document template and the first financial document are the same document template;

[0015] Acquire the table structure of the financial document template as the table structure of the electronic document of the second image;

[0016] Positioning the target selection box of the second image according to the table structure of the financial document template, extracting the target data in the target selection box and filling it into the corresponding position of the electronic document; each of the second images corresponds to a group of target selection boxes, each of the target selection boxes corresponds to a selection box number, and each of the selection box numbers is associated with a financial data name;

[0017] All target data of the electronic document and the corresponding financial data names are collected as the basic financial data, and the basic financial data, the financial data type and the electronic document are packaged as the first financial data.

[0018] Furthermore, after obtaining the second image and before performing line detection and cell segmentation on the second image, the following steps are included:

[0019] Determine a maximum frame for the area to be read in the second image, and correct the second image according to the coordinates of four vertices of the maximum frame;

[0020] Performing cell recognition on the second image, reading temporary data in each of the cells, detecting an identification type keyword contained in the temporary data, and determining the financial document type of the second image according to the detection result of the identification type keyword;

[0021] Acquire a corresponding bill template image according to the financial document type, and perform binarization processing on the second image; the bill template image is a standard bill image used to detect the clarity of the second image;

[0022] Calculate the absolute value of the grayscale difference between the second image and the ticket template image in all positioning cells. If the absolute value of the grayscale difference of any positioning cell is less than or equal to the detection threshold, the second image is judged to be an unqualified image and is invalidated; if the absolute values ​​of the grayscale differences of all the positioning cells are greater than the detection threshold, the second image is judged to be a qualified image and proceed to the next step.

[0023] Furthermore, the financial data entry module calls the financial data extraction submodule and the financial data upload submodule at preset intervals.

[0024] Furthermore, the basic financial data is stored in the form of data groups, and one group of the data groups includes a financial data name and a plurality of target data, and the financial data name and the target data are associated with each other.

[0025] Furthermore, the first financial data is segmented into a plurality of first financial data blocks, and the plurality of first financial data blocks are encrypted and uploaded to a storage server of cloud computing, specifically including:

[0026] matching a corresponding encryption algorithm for the first financial data according to the financial data type and transmission scenario of the first financial data;

[0027] Determine a segmentation granularity of the first financial data according to the encryption algorithm, and segment the first financial data according to the segmentation granularity to obtain a plurality of first data blocks;

[0028] The encryption algorithm is used to encrypt the plurality of first data blocks and then the encrypted data blocks are uploaded to a storage server of cloud computing.

[0029] Furthermore, determining the segmentation granularity of the first financial data according to the encryption algorithm specifically includes:

[0030] The segmentation granularity of the first financial data is determined according to a restriction standard of the encryption algorithm and a data size of the first financial data.

[0031] Furthermore, segmenting the first financial data according to the segmentation granularity to obtain a plurality of first data blocks includes the following steps:

[0032] When the security index of the current network is monitored to be lower than a preset threshold and affects data integrity, the segmentation granularity is increased, the first financial data is segmented according to the increased segmentation granularity to obtain a plurality of first data blocks, and redundant codes are added to each of the first data blocks obtained after segmentation.

[0033] When the security index of the current network is monitored to be higher than a preset threshold and the data transmission rate is too low, the segmentation granularity is increased, and the first financial data is segmented according to the increased segmentation granularity to obtain a plurality of first data blocks.

[0034] Furthermore, the uploading process of the first data block is monitored, and when the first data block is intercepted, the segmentation of the first financial data and the uploading of the first data block are stopped.

[0035] Based on the above system embodiment, the present invention provides a method embodiment;

[0036] An embodiment of the present invention provides a method for analyzing and managing enterprise financial data based on cloud computing, comprising the following steps:

[0037] By matching a first financial document with a pre-stored local financial document template, first financial data of the first financial document is obtained, where the first financial data includes basic financial data, financial data type, and a second financial document; the first financial document is a paper document, and the second financial document is an electronic document corresponding to the paper document;

[0038] Segmenting the first financial data into a plurality of first financial data blocks, encrypting the plurality of first financial data blocks, and uploading them to a storage server of cloud computing;

[0039] The corresponding basic financial data is obtained from the storage server of the cloud computing according to the financial analysis type selected by the user, and the basic financial data is input into the financial analysis model corresponding to the financial analysis type to obtain the financial data analysis result.

[0040] The following beneficial effects are achieved by implementing the embodiments of the present invention:

[0041] The embodiment of the present invention provides a cloud computing-based enterprise financial data analysis management system and method. When extracting financial data from a financial document, by matching the financial document with a pre-stored local financial document template, the structure of the financial document can be accurately identified when extracting financial data, and the financial data in the financial document can be accurately extracted, thereby improving the accuracy of financial data entry and the accuracy of financial data analysis. Since the financial document template is a pre-collected standard document, it can help accurately identify and extract the table structure, financial data at each location, and financial data name of the financial document, thereby greatly improving the accuracy of financial data entry and the accuracy of financial data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a structural diagram of a cloud computing-based enterprise financial data analysis and management system provided by an embodiment of the present invention.

[0043] Figure 2 It is a schematic diagram of a terminal interface in which an input data detection submodule of a cloud computing-based enterprise financial data analysis and management system provides a schematic diagram of a terminal interface in which the electronic document and the second image are displayed simultaneously.

[0044] Figure 3 It is a flowchart of a cloud computing-based enterprise financial data analysis and management method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] See also Figure 1 , is a structural diagram of a cloud computing-based enterprise financial data analysis and management system provided by an embodiment of the present invention, including a financial data entry module 11 and a financial data analysis module 12.

[0047] The financial data entry module 11 includes a financial data extraction submodule and a financial data upload submodule.

[0048] The financial data identification submodule is used to extract the first financial data of the first financial document by matching the first financial document with a pre-stored local financial document template, wherein the first financial data includes basic financial data, financial data type, and second financial document; the first financial document is a paper document, and the second financial document is an electronic document corresponding to the paper document. The paper document includes company accounting, account books, customer reconciliation, reimbursement, invoicing, bank reconciliation, and various financial bills. All financial document templates used for corporate financial data are collected in advance and saved locally. In this embodiment, when extracting financial data from a financial document, by matching the pre-stored local financial document template for the financial document, the structure of the financial document can be accurately identified when extracting financial data, and then the financial data in the financial document can be accurately extracted, thereby improving the accuracy of financial data entry, and then improving the accuracy of financial data analysis. Since the financial document template is a pre-collected standard document, it can help accurately identify and extract the table structure, financial data at each location, and financial data name of the financial document, thereby greatly improving the accuracy of financial data entry and the accuracy of financial data analysis.

[0049] The financial data uploading submodule is used to divide the first financial data into a plurality of first financial data blocks, encrypt the plurality of first financial data blocks, and upload them to a cloud computing storage server.

[0050] The financial analysis module 12 is used to obtain the corresponding basic financial data from the cloud computing storage server according to the financial analysis type selected by the user, and input the basic financial data into the financial analysis model corresponding to the financial analysis type to obtain financial data analysis results.

[0051] In a preferred embodiment, the financial data entry module 11 calls the financial data extraction submodule and the financial data upload submodule at preset intervals to perform financial data extraction and upload operations, that is, to perform data extraction and upload processes in the financial data extraction submodule and the financial data upload submodule. By periodically calling the financial data extraction submodule and the financial data upload submodule, the first financial data can be regularly extracted and uploaded to the cloud computing storage server.

[0052] In a preferred embodiment, extracting the first financial data of the first financial document by matching the first financial document with a pre-stored local financial document template specifically includes:

[0053] Step S1: Capture a first image of the first financial document, perform paper edge detection on the first image, and perform regularization processing on the edge of the first image to obtain a second image. The regularization processing includes performing edge thinning processing, edge smoothing processing, edge enhancement processing, and image post-processing on the first image.

[0054] Step S2: Perform line detection and cell segmentation on the second image to obtain a table structure of the second image.

[0055] Step S3: According to the financial document type of the second image, a locally pre-stored financial document template corresponding to the financial document type of the second image is obtained.

[0056] Step S4: comparing the financial document template and the table structure, and if the matching degree between the two exceeds a first preset threshold, determining that the financial document template and the first financial document are the same document template.

[0057] Step S5: Acquire the table structure of the financial document template as the table structure of the electronic document of the second image. At this time, the electronic document of the second image only has the table structure and lacks the financial data in the table.

[0058] Step S6: Position the target selection box of the second image according to the table structure of the financial document template, extract the target data in the target selection box and fill it into the corresponding position of the electronic document. Each of the second images corresponds to a group of target selection boxes, each of the target selection boxes corresponds to a selection box number, and each of the selection box numbers is associated with a financial data name. When extracting the target data in the target selection box, the coordinate information of the target selection box is recorded. Each empty space in the table of the electronic document has a position number. When the empty space is filled with the target data, the position number and the selection box number are associated one-to-one and saved to the local database. The financial data name of the target data in each of the target selection boxes is determined according to the selection box number, and a third association relationship between the financial data name and the target data is established.

[0059] Step S7: Summarize all the target data of the electronic document and the corresponding financial data names as the basic financial data, and package the basic financial data, the financial data type and the electronic document into first financial data. The electronic document is a PDF file or an image file. The basic financial data is stored in the form of a data group, and a group of the data groups includes a financial data name and a plurality of target data, and the financial data name and the target data have a fourth association relationship.

[0060] In a preferred embodiment, the step further includes: establishing a one-to-one fifth association relationship between the electronic document and the second image, and storing the electronic document and the second image in a local database.

[0061] In a preferred embodiment, after obtaining the second image and before performing line detection and cell segmentation on the second image, the following steps are included:

[0062] Step S10: Determine a maximum frame for the area to be read in the second image, and correct the second image according to the four vertex coordinates of the maximum frame. Specifically, determine whether the second image is correct according to the four vertex coordinates. If not, calculate the rotation angle according to the four vertex coordinates, and correct the second image according to the rotation angle. Correcting the second image refers to adjusting the second image to a preset standard direction or angle. When collecting the second image, that is, when the paper document is digitized, due to reasons such as improper operation, the second image will appear obviously tilted, which will affect the subsequent data extraction process of the second image. By correcting the second image, the second image and the pre-stored local financial document template can be perfectly matched to improve the accuracy of financial data extraction. This step improves the readability of the second image, thereby improving the accuracy of subsequent financial data extraction.

[0063] Step S11: Perform cell recognition on the second image, read temporary data in each of the cells, detect the identification type keyword contained in the temporary data, and determine the financial document type of the second image according to the detection result of the identification type keyword. The identification type keyword is a pre-set keyword for identifying the type of financial data. A first association relationship between the identification type keyword and the financial data type is pre-established. Since a financial document contains a certain type or several types of financial data, a second association relationship between the financial data type and the financial document type can also be pre-established. This step determines the financial data type of the second image according to the detected identification type keyword, and then determines the financial document type of the second image according to the financial data type. The financial data types include financial statement data, financial ratio data, budget data, investment data, financing data, cash flow data, cost data, price data, tax data, stock price data and market share data. The financial statement data includes balance sheet data, income statement data, and cash flow statement data. The financial ratio data includes liquidity ratio, leverage ratio, profitability ratio and operating efficiency ratio. The cost data includes direct cost, indirect cost, fixed cost and variable cost.

[0064] Step S12: Acquire a corresponding bill template image according to the financial document type, and perform binarization processing on the second image. The bill template image is a standard bill image used to detect the clarity of the second image.

[0065] Step S13: Calculate the absolute value of the grayscale difference between the second image and the bill template image within the range of all positioning cells. If the absolute value of the grayscale difference of any positioning cell is greater than the detection threshold, the second image is judged to be an unqualified image and the second image is invalidated; if the absolute values ​​of the grayscale differences of all the positioning cells are less than or equal to the detection threshold, the second image is judged to be a qualified image and enter the next step, which is the step S2. In this embodiment, the financial document type of the second image is identified by setting an identification type keyword, and then the locally pre-stored bill template image is searched according to the financial document type, and the clarity of the second image is detected according to the bill template image. That is, in this embodiment, the bill template image (i.e., the standard bill image) is pre-set as a detection reference for the second image, and the second image with qualified clarity is accurately screened out according to the determined clarity standard, thereby improving the accuracy of financial data extraction.

[0066] In a preferred embodiment, the financial data entry module 11 further includes an entry data detection submodule, and the entry data detection submodule is used to help manual inspection and correction of the financial data in the electronic document by simultaneously displaying the electronic document and the second image.

[0067] See also Figure 2 , is a schematic diagram of a terminal interface in which the input data detection submodule of a cloud computing-based enterprise financial data analysis and management system provides a terminal interface in which the electronic document and the second image are displayed simultaneously. The input data detection submodule includes the following steps:

[0068] Step S91: Acquire the electronic document and the second image from the local database, and display the second image and the electronic document on the upper and lower sides or the left and right sides of the terminal display, respectively.

[0069] Step S92: setting the number filling position of the electronic document as button 102, when the user clicks the number filling position, searching for the selection box number of the second image according to the position number of the number filling position.

[0070] Step S93: displaying a red frame 101 on the second image according to the coordinate information of the selection frame number, wherein the red frame 101 is used to identify the data source of the filling position.

[0071] Step S94: the user compares the data in the number-filling position with the data in the red frame 101 to see if they are consistent. If not, the user double-clicks the button 102 of the number-filling position, and the number-filling position becomes editable.

[0072] Step S95: The user corrects the data in the filling position according to the data in the red box 101, and saves the corrected electronic document to the local database by pressing the Enter key, and simultaneously corrects the electronic document in the first financial data.

[0073] In a preferred embodiment, the user corrects the data in the number filling position according to the data in the red frame 101, specifically:

[0074] The user selects the red box 101 and clicks the button 102 at the number filling position to trigger a one-key data acquisition operation, wherein the one-key data acquisition operation refers to extracting the data in the red box 101 and automatically filling it into the number filling position.

[0075] In a preferred embodiment, the step of segmenting the first financial data to obtain a plurality of first financial data blocks, encrypting the plurality of first financial data blocks, and uploading them to a cloud computing storage server specifically includes:

[0076] Step S96: matching a corresponding encryption algorithm for the first financial data according to the financial data type and transmission scenario of the first financial data.

[0077] Step S97: determining the segmentation granularity of the first financial data according to the restriction standard of the encryption algorithm and the data size of the first financial data, and segmenting the first financial data according to the segmentation granularity to obtain a plurality of first data blocks.

[0078] Step S98: encrypt the plurality of first data blocks using the encryption algorithm and upload the encrypted data blocks to a cloud computing storage server.

[0079] In a preferred embodiment, the first financial data is matched with a corresponding encryption algorithm according to the following steps:

[0080] Step S961: determining a transmission path according to a delivery location and a sending location when the first financial data is transmitted, and determining a transmission scenario according to the transmission path;

[0081] Step S962: Calculating a sensitivity coefficient of the first financial data according to the type of financial data and the transmission scenario;

[0082] Step S963: setting variable nodes of the Bayesian network model, taking the financial data type, transmission scenario and sensitivity coefficient as variable parameters of the Bayesian network model, matching the variable parameters to the variable nodes, and connecting the variable nodes with direct dependency relationships;

[0083] Step S964: setting the initialization probability of each variable node, and calculating the posterior probability of the optimal decryption time according to the Bayesian network model;

[0084] Step S965: Calculate the mean of the posterior probabilities that meet the preset conditions, and use the mean as the optimal decryption time after the first financial data is encrypted; the preset condition is that the posterior probability meets the normal distribution.

[0085] Step S966: Match the corresponding encryption algorithm for the first financial data according to the optimal decryption time. Since different encryption algorithms require different times for decryption, this embodiment first determines the optimal decryption time during data transmission, and determines the encryption algorithm according to the optimal decryption time. This ensures that the first financial data can not only be encrypted and uploaded, ensuring the security of data transmission, but also that the decryption time requirement can be met when using the first financial data, thereby improving the analysis efficiency during financial data analysis.

[0086] In a preferred embodiment, segmenting the first financial data according to the segmentation granularity to obtain a plurality of first data blocks includes the following steps:

[0087] Step S967: When the security index of the current network is monitored to be lower than a preset threshold and affects data integrity, the segmentation granularity is increased, and the first financial data is segmented according to the increased segmentation granularity to obtain a plurality of first data blocks, and redundant codes are added to each of the first data blocks obtained after segmentation.

[0088] Step S968: When the security index of the current network is monitored to be higher than a preset threshold and the data transmission rate is too low, the segmentation granularity is increased, and the first financial data is segmented into a plurality of first data blocks according to the increased segmentation granularity. The plurality of first data blocks are numbered, and the first data blocks are combined according to the data block numbers. In this embodiment, by increasing the segmentation granularity and adding redundant codes, the risk of data being intercepted and lost is reduced, the integrity of the data is improved, and the security of data transmission is improved.

[0089] In a preferred embodiment, the upload process of the first data block is monitored, and when the first data block is intercepted, the segmentation of the first financial data and the upload of the first data block are stopped. The upload process of the first data block is monitored, and when it is found that the first data block is intercepted, the segmentation of the first financial data and the upload of the first data block are stopped, thereby ensuring the security of data transmission and preventing the leakage of corporate financial data.

[0090] In a preferred embodiment, the financial data analysis module is used to obtain corresponding financial data from the storage server according to the financial analysis type selected by the user, input the financial data into the financial analysis model, and output the financial data analysis results; the financial analysis types include financial statement analysis, financial ratio analysis, financial trend analysis, financial budget analysis, financial forecast analysis, financial risk analysis, and peer competition analysis. The financial analysis model is established according to the financial analysis type, the financial analysis target corresponding to the financial analysis type, and the machine learning algorithm, and the financial analysis model includes a financial statement analysis model, a financial ratio analysis model, a financial trend analysis model, a financial budget analysis model, a financial forecast analysis model, a financial risk analysis model, and a peer competition analysis model.

[0091] Pre-establishing and training the financial analysis model includes the following steps:

[0092] Step S111: Select a corresponding machine learning algorithm according to the financial analysis type and financial analysis goal.

[0093] Step S112: Determine initial feature data according to the financial analysis type, financial analysis target and machine learning algorithm.

[0094] Step S113: constructing the financial analysis model according to the machine learning algorithm and the initial feature data.

[0095] Step S114: The initial feature data is screened and optimized through feature engineering to obtain first feature data; specifically, the PCA feature dimensionality reduction method is used to reduce the dimension of the initial feature data, and feature data that has an important impact on the analysis result is retained.

[0096] Step S115: constructing a training set and a test set of the financial analysis model according to the first feature data, and optimizing the structure and parameters of the financial analysis model by cross-validation until the financial analysis model reaches a preset performance standard.

[0097] When extracting financial data from a financial document, the embodiment of the present invention matches the financial document with a pre-stored local financial document template, so that the structure of the financial document can be accurately identified when extracting financial data, and the financial data in the financial document can be accurately extracted, thereby improving the accuracy of financial data entry and the accuracy of financial data analysis. Since the financial document template is a pre-collected standard document, it can help accurately identify and extract the table structure, financial data at each location, and financial data name of the financial document, thereby greatly improving the accuracy of financial data entry and the accuracy of financial data analysis.

[0098] Based on the above system embodiment, a corresponding method embodiment is provided;

[0099] See also Figure 3 Another embodiment of the present invention provides a method for analyzing and managing enterprise financial data based on cloud computing, comprising the following steps:

[0100] Step S100: acquiring first financial data of the first financial document by matching a pre-stored local financial document template for the first financial document, wherein the first financial data includes basic financial data, financial data type, and a second financial document; the first financial document is a paper document, and the second financial document is an electronic document corresponding to the paper document;

[0101] Step S200: Segmenting the first financial data into a plurality of first financial data blocks, encrypting the plurality of first financial data blocks, and uploading them to a storage server of cloud computing;

[0102] Step S300: acquiring the corresponding basic financial data from the storage server of the cloud computing according to the financial analysis type selected by the user, inputting the basic financial data into the financial analysis model corresponding to the financial analysis type, and obtaining the financial data analysis result.

[0103] It can be understood that the above method item embodiments correspond to the system item embodiments of the present invention, which can implement the cloud computing-based enterprise financial data analysis and management system provided by any of the above system item embodiments of the present invention.

[0104] It should be noted that the device embodiments described above are merely schematic, wherein the units / modules described as separate components may or may not be physically separated, and the components displayed as units / modules may or may not be physical units / modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without creative work. The schematic diagram is only an example of a control device for a display, and does not constitute a limitation on the control device for the display, and may include more or fewer components than shown in the figure, or a combination of certain components, or different components.

[0105] The above are preferred embodiments of the present invention. It should be noted that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A cloud computing-based enterprise financial data analysis and management system, characterized in that: Including financial data entry module and financial data analysis module; The financial data entry module includes a financial data extraction submodule and a financial data upload submodule; The financial data extraction submodule is used to extract first financial data of the first financial document by matching the first financial document with a pre-stored local financial document template, where the first financial data includes basic financial data, financial data type, and a second financial document; the first financial document is a paper document, and the second financial document is an electronic document corresponding to the paper document; The financial data uploading submodule is used to divide the first financial data into a plurality of first financial data blocks, encrypt the plurality of first financial data blocks, and upload them to a storage server of cloud computing; The financial data analysis module is used to obtain the corresponding basic financial data from the cloud computing storage server according to the financial analysis type selected by the user, input the basic financial data into the financial analysis model corresponding to the financial analysis type, and obtain the financial data analysis results.

2. The enterprise financial data analysis and management system based on cloud computing according to claim 1, characterized in that: By matching the first financial document with a locally stored financial document template, extracting first financial data of the first financial document specifically includes: Capturing a first image of the first financial document, performing paper edge detection on the first image, and regularizing the edge of the first image to obtain a second image; Performing line detection and cell segmentation on the second image to obtain a table structure of the second image; According to the financial document type of the second image, obtaining a locally pre-stored financial document template corresponding to the financial document type of the second image; Comparing the financial document template with the table structure, and if the matching degree between the two exceeds a preset threshold, determining that the financial document template and the first financial document are the same document template; Acquire the table structure of the financial document template as the table structure of the electronic document of the second image; Positioning the target selection box of the second image according to the table structure of the financial document template, extracting the target data in the target selection box and filling it into the corresponding position of the electronic document; each of the second images corresponds to a group of target selection boxes, each of the target selection boxes corresponds to a selection box number, and each of the selection box numbers is associated with a financial data name; All target data of the electronic document and the corresponding financial data names are collected as the basic financial data, and the basic financial data, the financial data type and the electronic document are packaged as the first financial data.

3. The enterprise financial data analysis and management system based on cloud computing as claimed in claim 2, characterized in that: After obtaining the second image and before performing line detection and cell segmentation on the second image, the following steps are included: Determine a maximum frame for the area to be read in the second image, and correct the second image according to the coordinates of four vertices of the maximum frame; Performing cell recognition on the second image, reading temporary data in each of the cells, detecting an identification type keyword contained in the temporary data, and determining the financial document type of the second image according to the detection result of the identification type keyword; Acquire a corresponding bill template image according to the financial document type, and perform binarization processing on the second image; the bill template image is a standard bill image used to detect the clarity of the second image; Calculate the absolute value of the grayscale difference between the second image and the ticket template image in all positioning cells. If the absolute value of the grayscale difference of any positioning cell is less than or equal to the detection threshold, the second image is judged to be an unqualified image and is invalidated; if the absolute values ​​of the grayscale differences of all the positioning cells are greater than the detection threshold, the second image is judged to be a qualified image and proceed to the next step.

4. The cloud computing-based enterprise financial data analysis and management system according to claim 3, characterized in that: The financial data entry module calls the financial data extraction submodule and the financial data upload submodule at preset intervals.

5. The enterprise financial data analysis and management system based on cloud computing according to claim 4, characterized in that: The basic financial data is stored in the form of data groups. One group of the data groups includes a financial data name and a plurality of target data. The financial data name and the target data are associated with each other.

6. The enterprise financial data analysis and management system based on cloud computing according to claim 5, characterized in that: The first financial data is segmented to obtain a plurality of first financial data blocks, and the plurality of first financial data blocks are encrypted and uploaded to a storage server of cloud computing, specifically including: matching a corresponding encryption algorithm for the first financial data according to the financial data type and transmission scenario of the first financial data; Determine a segmentation granularity of the first financial data according to the encryption algorithm, and segment the first financial data according to the segmentation granularity to obtain a plurality of first data blocks; The encryption algorithm is used to encrypt the plurality of first data blocks and then the encrypted data blocks are uploaded to a storage server of cloud computing.

7. The enterprise financial data analysis and management system based on cloud computing according to claim 6, characterized in that: Determining the segmentation granularity of the first financial data according to the encryption algorithm specifically includes: The segmentation granularity of the first financial data is determined according to a restriction standard of the encryption algorithm and a data size of the first financial data.

8. The enterprise financial data analysis and management system based on cloud computing according to claim 7, characterized in that: The step of segmenting the first financial data according to the segmentation granularity to obtain a plurality of first data blocks comprises the following steps: When the security index of the current network is monitored to be lower than a preset threshold and affects data integrity, the segmentation granularity is increased, the first financial data is segmented according to the increased segmentation granularity to obtain a plurality of first data blocks, and a redundant code is added to each of the first data blocks obtained after segmentation; When the security index of the current network is monitored to be higher than a preset threshold and the data transmission rate is too low, the segmentation granularity is increased, and the first financial data is segmented according to the increased segmentation granularity to obtain a plurality of first data blocks.

9. The enterprise financial data analysis and management system based on cloud computing according to any one of claims 1 to 8, characterized in that: The uploading process of the first data block is monitored, and when the first data block is intercepted, the segmentation of the first financial data and the uploading of the first data block are stopped.

10. A method for analyzing and managing enterprise financial data based on cloud computing, characterized in that: The following steps are involved: By matching a first financial document with a pre-stored local financial document template, first financial data of the first financial document is obtained, where the first financial data includes basic financial data, financial data type, and a second financial document; the first financial document is a paper document, and the second financial document is an electronic document corresponding to the paper document; Segmenting the first financial data into a plurality of first financial data blocks, encrypting the plurality of first financial data blocks, and uploading them to a storage server of cloud computing; The corresponding basic financial data is obtained from the storage server of the cloud computing according to the financial analysis type selected by the user, and the basic financial data is input into the financial analysis model corresponding to the financial analysis type to obtain the financial data analysis result.

Images