Financial management system for sports economic analysis

Abstract

The invention discloses a financial management system for sports economic analysis, and relates to the technical field of financial management, and the system comprises three modules: a data collection module obtains a user network signal and equipment image data in real time and marks an ID; the preprocessing module preprocesses the marked data; and the central calculation processing module calculates and standardizes to obtain a picture tearing model, analyzes whether the picture is torn or not, predicts severity if the picture is torn, analyzes financial income increase, returns the picture to the acquisition module after adjustment through a scheduling algorithm, and ends analysis if the picture is not torn. According to the method, the prediction type picture tearing model is constructed by fusing network signals and image contour analysis, abnormity is detected in real time, user behaviors are pre-judged, a dynamic advertisement scheduling algorithm is combined, problem users are intelligently shielded, and normal users are switched, so that the advertisement putting efficiency and the financial revenue stability are remarkably improved, and the playing amount and the duration are effectively prevented from sliding down.

Description

Technical Field

[0001] This invention relates to the field of financial management technology, specifically a financial management system for sports economic analysis. Background Technology

[0002] In the sports economy landscape, integrating advertising with event viewership and playback duration is a mature and efficient profit system. Event viewership and playback duration are not only core indicators for measuring attention, but also directly determine the breadth and depth of advertising reach. The higher the event viewership and the longer the playback duration, the greater the exposure effectiveness of the advertisement. Based on this, advertisers pay corresponding fees to lock in these exposure scenarios, while event organizers obtain stable financial revenue, which in turn supports event operations, thus forming a virtuous cycle of mutual empowerment between sports events and commercial advertising.

[0003] For example, the RPA-based digital fiscal management method and system provided in Chinese Patent Publication No. CN116720973B includes: automatically extracting script fields from the unit's budget system database by comparing the digital fiscal system's data entry information; creating an automated execution script for the digital fiscal management RPA robot process; obtaining project budget reporting data from the unit's budget system; creating automated execution actions for the digital fiscal management RPA robot process by mimicking manual computer operation based on the automated execution script; generating an automatic digital fiscal reporting process based on the intelligent digital fiscal reporting actions, project budget reporting data, and digital fiscal system data entry information; and recording the reporting status of all project budget reporting data based on the generated project budget reporting data reporting execution log, thus performing digital fiscal management through automated operation and maintenance of the RPA robot process.

[0004] However, the above plan failed to consider that in a profit model that relies on sports event viewership and playback time for inserting advertisements, network signal problems would trigger a series of chain reactions. When users watch live sports events, if the picture becomes torn due to poor signal, not only will the inserted advertisement content be damaged and unable to effectively convey information, but it will also seriously wear down the user's viewing patience. As the viewing experience continues to decline, the user's willingness to continue watching will drop significantly, and they will choose to exit the live broadcast. This reduction in user viewership and playback time directly affects the exposure effect and placement value of advertisements, causing advertising revenue to shrink and ultimately having a negative impact on overall fiscal revenue. Summary of the Invention

[0005] Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a financial management system for sports economic analysis. It solves the problem that in a profit model that relies on the number and duration of sports event broadcasts for advertising, poor network signal can cause screen tearing between live streams and advertisements, which reduces user viewing willingness, reduces broadcast volume and duration, thereby weakening advertising value, reducing advertising revenue, and ultimately affecting overall fiscal revenue.

[0006] Technical solution To achieve the above objectives, the present invention provides the following technical solution: a financial management system for sports economic analysis, comprising the following specific modules: a data acquisition module: which calls up the user's network signal data and device image data in real time, and marks the user with an ID; the network signal data includes actual bandwidth, actual latency, and actual packet loss rate; and the device image data includes pixel coordinates and the number of pixels; a data preprocessing module: which preprocesses the ID-marked network signal data and device image data; and a central computing and processing module: which performs comprehensive calculations based on the preprocessed network signal data and device image data, and then performs standardization processing to obtain a screen tearing model; based on the screen tearing model, it analyzes whether the screen is torn; if the analysis shows no screen tearing, the analysis ends directly; if screen tearing is detected, it predicts the severity of screen tearing, analyzes the increase in fiscal revenue, and adjusts it through a dynamic scheduling algorithm for full-domain reach advertising, before returning to the data acquisition module.

[0007] Furthermore, the specific method for obtaining the screen tearing model is as follows: network signal data is comprehensively calculated to obtain network anomaly parameters; device image data is comprehensively calculated to obtain network anomaly parameters; and the screen tearing model is obtained by comprehensively calculating the network anomaly parameters and the screen anomaly parameters. ;in, This indicates a screen tearing model. Indicates abnormal parameters in the image. Indicates abnormal network parameters.

[0008] Furthermore, the specific method for obtaining the network anomaly parameters is as follows: Set a standard bandwidth, standard latency, and standard packet loss rate; calculate the difference between the standard bandwidth and the actual bandwidth to obtain the bandwidth deviation value; calculate the difference between the actual latency and the standard latency to obtain the latency deviation value; calculate the difference between the actual packet loss rate and the standard packet loss rate to obtain the packet loss rate deviation value; and sum the bandwidth deviation value, latency deviation value, and packet loss rate deviation value to obtain the network anomaly parameters.

[0009] Furthermore, the specific method for obtaining the image anomaly parameters is as follows: The device image data is processed by an image contour segmentation algorithm and an image contour tracking algorithm to obtain an image contour. The difference between the x-axis coordinates of the pixel coordinates of the previous frame image contour and the x-axis coordinates of the corresponding pixel coordinates of the next frame image contour is calculated, or the difference between the y-axis coordinates of the pixel coordinates of the previous frame image contour and the y-axis coordinates of the corresponding pixel coordinates of the next frame image contour is calculated, and the absolute value is taken to obtain a continuous value. The continuous value is summed according to the number of pixels in the image contour to obtain the image anomaly parameters.

[0010] Furthermore, the specific steps for analyzing whether the screen is torn based on the screen tearing model are as follows: set a screen tearing threshold, compare the screen tearing model with the screen tearing threshold, if the screen tearing model is greater than the screen tearing threshold, it indicates that the screen is torn, if the screen tearing model is less than or equal to the screen tearing threshold, it indicates that the screen is not torn.

[0011] Furthermore, the specific method for obtaining the screen tearing threshold is as follows: When the image is normal, if both the outline of the previous frame and the outline of the next frame are stationary, the difference between the x-axis or y-axis coordinates of the pixel coordinates of the previous frame and the x-axis or y-axis coordinates of the corresponding pixel coordinates of the next frame is zero. If the outline of the next frame moves following the outline of the previous frame, the absolute value of the difference between the x-axis or y-axis coordinates of the pixel coordinates of the next frame and the x-axis or y-axis coordinates of the corresponding pixel coordinates of the previous frame is one. Therefore, by summing the continuous values ​​based on the number of pixels in the image outline, the range of the screen tearing threshold is 0 to... , The product of the number of pixels representing the image outline and one.

[0012] Further, the specific steps for predicting the severity of screen tearing are as follows: S11: Record the values ​​of the screen tearing model as tear values ​​and count them to obtain a tear value set; S12: Record the time interval between the previous tear value and the next tear value according to the tear value set to obtain the tear interval time, and count them to obtain a tear interval time set; S13: Establish a two-dimensional coordinate system, with the horizontal axis representing the tear interval time and the vertical axis representing the tear value, to obtain the tear point. Since both the tear interval time and the tear value are greater than zero, the tear point is in the first quadrant; S14: Let the... -1 tear point, the first The tear point and the first tear point +1 tear point, for the first The tear point and the first The slope is calculated at the -1 tear point to obtain the Nth tear slope. The tear point and the first S15: Calculate the slope of each tear point to obtain the Mth tear slope; S16: Compare the Mth tear slope with the Nth tear slope. If the Mth tear slope is greater than the Nth tear slope, the image tearing is predicted to be severe. If the Mth tear slope is less than or equal to the Nth tear slope, the image tearing is predicted to be mild.

[0013] Furthermore, the specific method for obtaining the Nth tear slope or the Mth tear slope is as follows: or ;in, Indicates the slope of the Nth tear or This represents the slope of the Mth tear. Indicates the first The tear value at each tear point Indicates the first The tear value at each tear point Indicates the first The tear value at each tear point Indicates the first The tear interval at each tear point Indicates the first The tear interval at each tear point Indicates the first The tear interval time at each tear point.

[0014] Furthermore, the specific steps for analyzing the growth rate of fiscal revenue are as follows: S21: During the time period when the Mth tear slope is greater than the Nth tear slope, the fiscal revenue at this moment is recorded as risky fiscal revenue. Before the moment when the Mth tear slope is detected to be greater than the Nth tear slope, the fiscal revenue is recorded as normal fiscal revenue; S22: The risky fiscal revenue is compared with the normal fiscal revenue. If the risky fiscal revenue is less than the normal fiscal revenue, it indicates that the growth rate of fiscal revenue has slowed down, and then it is adjusted through the dynamic scheduling algorithm of the full-domain reach advertising. If the risky fiscal revenue is greater than or equal to the normal fiscal revenue, it indicates that the growth rate of fiscal revenue is normal, and then the comparison continues.

[0015] Furthermore, the specific steps of the dynamic scheduling algorithm for full-domain reach advertising are as follows: S31: When revenue slows down, temporarily block the user IDs that are detected to have screen tearing; S32: Continue to deliver advertisements to new users. If screen tearing is detected for this user, return to S31, and so on. If screen tearing is detected for this user, continue to deliver advertisements.

[0016] Beneficial effects Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: 1. By combining network signal data with image contour analysis technology, a predictive screen tearing model is constructed. The system can not only detect screen anomalies in real time, but also predict user behavior trends, thereby achieving intelligent scheduling of advertising resources and significantly improving advertising efficiency and fiscal revenue stability.

[0017] 2. Through the dynamic scheduling algorithm of full-domain reach advertising, real-time optimization of ad delivery is achieved. When severe screen tearing is detected and causes a slowdown in the growth of fiscal revenue, the problematic user ID will be temporarily blocked and the ads will be delivered to new users with normal screens. This dynamic adjustment mechanism effectively avoids the problem of a sharp drop in playback volume and duration caused by user exit.

[0018] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0019] Figure 1 This invention relates to a flowchart of a financial management method for sports economic analysis.

[0020] Figure 2 This invention relates to a structural diagram of a financial management system for sports economic analysis. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0023] Example 1: like Figures 1-2 As shown, this embodiment of the invention provides a financial management system for sports economic analysis, comprising the following specific modules: Data acquisition module: With the user's consent, the module calls the user's network signal data and device image data in real time through the application programming interface, and marks the user with an ID to achieve unique user identification. The network signal data includes the actual bandwidth, actual latency and actual packet loss rate. The device image data includes pixel coordinates and the number of pixels, where the pixel coordinates include x-axis coordinates and y-axis coordinates.

[0024] Data preprocessing module: Cleans the network signal data marked with ID, removes redundant data, and performs noise reduction processing on the device image data marked with ID to improve image quality.

[0025] The central computing and processing module performs comprehensive calculations based on the cleaned network signal data and the noise-reduced device image data, while simultaneously standardizing the data to eliminate dimensional differences between the two. It transforms values ​​of different orders of magnitude into a unified numerical range, thus obtaining a screen tearing model. This model is used to analyze whether screen tearing exists, as screen tearing affects advertising content and sports event footage. It not only damages the advertising content and fails to effectively convey information but also severely diminishes viewers' patience, leading to a significant decrease in viewership and ultimately, exiting the live stream. Reduced viewership and viewing time directly impact advertising revenue, thus affecting financial income. If no screen tearing is detected, the analysis ends directly. If screen tearing is detected, the severity of the tearing is predicted, and the severity of subsequent screen tearing in the user's image determines whether to continue advertising to that user. If the user's subsequent image tearing is not severe, the probability of this user abandoning the sports event is low. To avoid a decrease in playback time due to frequent user switching, content will continue to be delivered to them. However, if the user's subsequent image tearing is severe, the probability of this user abandoning the sports event is high. Therefore, it is necessary to analyze the increase in revenue. That is, even if the user's subsequent image tearing is severe, if the user insists on continuing to watch, then it will not affect the slowdown in revenue growth. Unless the user abandons the sports event, it will affect revenue. The dynamic scheduling algorithm of the whole-domain reach advertising is used to select new users who are watching the sports event. After obtaining the new user's consent, the data collection module is returned to collect the network signal data and device image data of the new user. This ensures that the network signal data and device image data of users watching the sports event are normal. Only in this way can the number of plays and playback time be increased, thereby increasing the revenue brought by advertising.

[0026] Example 2 differs from Example 1 in that: The specific method for obtaining the screen tear model is as follows: Network anomaly parameters are obtained by comprehensively calculating network signal data and device image data. Based on the comprehensive calculation of network anomaly parameters and image anomaly parameters, an image tearing model is obtained. ; in, This indicates a screen tearing model, reflecting whether or not the screen is torn. These parameters indicate whether the image is abnormal. This represents network anomaly parameters, reflecting whether the network is abnormal. If the network anomaly parameter is zero, it will not affect the screen tearing model. The network anomaly parameter and the screen anomaly parameter are non-linearly fused to more accurately predict the risk of user exit and achieve dynamic optimization of ad placement.

[0027] The specific methods for obtaining network anomaly parameters are as follows: Set standard bandwidth, standard latency, and standard packet loss rate, which represent the bandwidth, latency, and packet loss rate under normal network signal conditions. Calculate the difference between the standard bandwidth and the actual bandwidth to obtain the bandwidth deviation value. The larger the bandwidth deviation value, the smaller the amount of data the network can transmit per unit time. Calculate the difference between the actual latency and the standard latency to obtain the latency deviation value. The larger the latency deviation value, the longer the time required for data to travel from the sender to the receiver. Calculate the difference between the actual packet loss rate and the standard packet loss rate to obtain the packet loss rate deviation value. The larger the packet loss rate deviation value, the larger the proportion of lost data packets among the total sent data packets during data transmission. Sum the bandwidth deviation value, latency deviation value, and packet loss rate deviation value to obtain the network anomaly parameters. ; in, Indicates abnormal network parameters. Indicates standard bandwidth. Indicates the actual bandwidth. This represents the bandwidth deviation value; a positive bandwidth deviation value indicates... Indicates the actual delay. Indicates standard delay, This represents the delay deviation value; a positive delay deviation value indicates... This represents the actual packet loss rate. Indicates the standard packet loss rate. This represents the packet loss rate deviation value, which is positive.

[0028] The specific methods for obtaining abnormal image parameters are as follows: Image contour segmentation and image contour tracking algorithms are used to process device image data. For example, the Sobel algorithm first converts the device image data into a grayscale image to simplify calculations. Then, it convolves the image data with 3×3 kernels in the horizontal and vertical directions to obtain the horizontal and vertical gradient matrices. Next, it calculates the gradient magnitude and direction, with the former determining the edge and the latter determining the edge extension direction. Finally, it performs threshold binarization, marking pixels above the threshold as contours to obtain the final image contour. Image contour tracking algorithms, such as the Lucas-Kanade optical flow method, establish optical flow constraint equations based on the assumption of constant brightness between adjacent frames. It calculates spatial and temporal gradients using the Sobel algorithm and frame difference. It assumes consistent neighborhood motion, constructs an overdetermined system of equations, and solves for the optical flow velocity using the least squares method. Finally, it performs tracking and position updates to track the image contour. The difference between the x-axis coordinates of the pixel coordinates of the outline of the previous frame and the x-axis coordinates of the corresponding pixel coordinates of the outline of the next frame is calculated, or the difference between the y-axis coordinates of the pixel coordinates of the outline of the previous frame and the y-axis coordinates of the corresponding pixel coordinates of the outline of the next frame is calculated, and the absolute value is taken to obtain a continuous value. The continuous values ​​are then summed according to the number of pixels in the image outline to obtain the image anomaly parameters. ; in, Indicates abnormal parameters in the image. The number of pixels representing the image outline. The x-axis coordinate represents the pixel coordinates corresponding to the outline of the next frame image. The x-axis coordinates representing the pixel coordinates of the outline of the previous frame image. The y-axis coordinate represents the pixel coordinates corresponding to the outline of the next frame image. The y-axis coordinates represent the pixel coordinates of the outline of the previous frame image. or Indicates continuous values.

[0029] The specific steps for analyzing whether there is screen tearing based on the screen tearing model are as follows: Set a screen tearing threshold, compare the screen tearing model with the screen tearing threshold. If the screen tearing model is greater than the screen tearing threshold, it means that the screen is torn. If the screen tearing model is less than or equal to the screen tearing threshold, it means that the screen is not torn.

[0030] The specific method for obtaining the screen tearing threshold is as follows: When the image is normal, if both the outline of the previous frame and the outline of the next frame are stationary, the difference between the x-axis or y-axis coordinates of the pixels in the previous frame and the corresponding pixels in the next frame is zero. If the outline of the next frame moves following the outline of the previous frame, the absolute value of the difference between the x-axis or y-axis coordinates of the pixels in the next frame and the corresponding pixels in the previous frame is one. Therefore, by summing consecutive values ​​based on the number of pixels in the image outline, the range of the screen tearing threshold is 0 to... , This represents the product of the number of pixels in the image contour and one. If the screen tearing model is greater than the screen tearing threshold, it not only indicates that the abnormal network parameters are greater than zero, making the screen tearing model larger (i.e., the abnormal network parameters affect the screen tearing model), but also that the screen tearing model is greater than the screen tearing threshold. In other words, the network difference increases the probability of screen tearing. Moreover, when screen tearing occurs, the absolute value of the difference between the x-axis or y-axis coordinate of a pixel in the image contour of the previous frame and the x-axis or y-axis coordinate of the corresponding pixel in the image contour of the next frame is greater than one.

[0031] The specific steps for predicting the severity of screen tearing are as follows: S11: Record the values ​​of the screen tearing model as tearing values ​​and count them. That is, the number of times the screen goes from normal to tearing and back to normal is represented by the number of times the screen goes through the tearing process. This gives us the set of tearing values. S12: Record the time interval between the previous tear value and the next tear value according to the tear value set to obtain the tear interval time, and perform a count to obtain the tear interval time set. S13: Establish a two-dimensional coordinate system, with the horizontal axis representing the tearing interval time and the vertical axis representing the tearing value, to obtain the tearing point. Since both the tearing interval time and the tearing value are greater than zero, the tearing point is in the first quadrant. S14: Let the first -1 tear point, the first The tear point and the first tear point +1 tear point, for the first The tear point and the first The slope is calculated at the -1 tear point to obtain the Nth tear slope. The tear point and the first The slope of the Mth tear is obtained by calculating the slope of each tear point. S15: Compare the Mth tear slope with the Nth tear slope. If the Mth tear slope is greater than the Nth tear slope, it indicates that when the tear value remains unchanged (i.e., the severity of screen tearing remains unchanged), the tear interval time decreases (i.e., the frequency of tear value occurrence increases), reflecting the first type of screen tearing severity. When the tear value increases (i.e., the severity of screen tearing increases), the tear interval time remains unchanged (i.e., the frequency of tear value occurrence remains unchanged), reflecting the second type of screen tearing severity. When the tear value increases (i.e., the severity of screen tearing increases), the tear interval time decreases (i.e., the frequency of tear value occurrence increases), reflecting the third type of screen tearing severity. In other words, comparing the tear slopes reflects the three types of screen tearing severity, which not only reduces the computational complexity but also improves the accuracy of the analysis, thus predicting severe screen tearing. If the Mth tear slope is less than or equal to the Nth tear slope, then the screen tearing is predicted to be not severe.

[0032] The specific methods for obtaining the Nth tear slope or the Mth tear slope are as follows: or ; in, Indicates the slope of the Nth tear or This represents the slope of the Mth tear. Indicates the first The tear value at each tear point Indicates the first The tear value at each tear point Indicates the first The tear value at each tear point Indicates the first The tear interval at each tear point Indicates the first The tear interval at each tear point Indicates the first The tear interval time at each tear point.

[0033] The specific steps for analyzing the growth rate of fiscal revenue are as follows: S21: During the time period when the Mth tear slope is greater than the Nth tear slope, the fiscal revenue at this moment is recorded as risky fiscal revenue; before the moment when the Mth tear slope is detected to be greater than the Nth tear slope, the fiscal revenue is recorded as normal fiscal revenue. S22: Compare risky fiscal revenue with normal fiscal revenue. If risky fiscal revenue is less than normal fiscal revenue, it indicates that the growth rate of fiscal revenue has slowed down. Then, the dynamic scheduling algorithm of full-domain reach advertising is used to adjust the situation. If risky fiscal revenue is greater than or equal to normal fiscal revenue, it indicates that the growth rate of fiscal revenue is normal. Then, the comparison continues.

[0034] The specific steps for adjusting through the dynamic scheduling algorithm of full-domain reach ads are as follows: S31: When fiscal revenue slows down, which means a decrease in play count and play duration, user IDs that are detected to have screen tearing will be temporarily blocked. S32: Since users who are blocked cannot be advertised temporarily, ads will be advertised to new users. If the screen of this user is also torn, return to S31 and so on. If the screen of this user is normal, ads will continue to be advertised to maintain the continuous growth of play count and play duration and keep the revenue normal.

[0035] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A financial management system for sports economic analysis, characterized in that: Includes the following specific modules: Data acquisition module: Real-time access to user's network signal data and device image data, and tag the user with an ID. Network signal data includes actual bandwidth, actual latency, and actual packet loss rate. Device image data includes pixel coordinates and number of pixels. Data preprocessing module: preprocesses the network signal data marked with ID and the device image data; Central computing and processing module: Based on the preprocessed network signal data and device image data, it performs comprehensive calculations and standardization to obtain a screen tearing model. Based on the screen tearing model, it analyzes whether the screen is torn. If the analysis shows no screen tearing, the analysis ends directly. If screen tearing is detected, it predicts the severity of screen tearing and analyzes the increase in fiscal revenue. It then adjusts the data through a dynamic scheduling algorithm for full-domain reach advertising before returning to the data acquisition module.

2. The financial management system for sports economic analysis according to claim 1, characterized in that: The specific method for obtaining the screen tearing model is as follows: Network anomaly parameters are obtained by comprehensively calculating network signal data and device image data. Based on the comprehensive calculation of network anomaly parameters and image anomaly parameters, an image tearing model is obtained. ； in, This indicates a screen tearing model. Indicates abnormal parameters in the image. Indicates abnormal network parameters.

3. A financial management system for sports economic analysis according to claim 2, characterized in that: The specific method for obtaining the network anomaly parameters is as follows: Set standard bandwidth, standard latency, and standard packet loss rate. Calculate the difference between the standard bandwidth and the actual bandwidth to obtain the bandwidth deviation value. Calculate the difference between the actual latency and the standard latency to obtain the latency deviation value. Calculate the difference between the actual packet loss rate and the standard packet loss rate to obtain the packet loss rate deviation value. Sum the bandwidth deviation value, latency deviation value, and packet loss rate deviation value to obtain the network anomaly parameters.

4. A financial management system for sports economic analysis according to claim 2, characterized in that: The specific method for obtaining the abnormal image parameters is as follows: The device image data is processed by image contour segmentation and image contour tracking algorithms to obtain image contours. The difference between the x-axis coordinates of the pixels of the previous frame image contour and the x-axis coordinates of the corresponding pixels of the next frame image contour is calculated, or the difference between the y-axis coordinates of the pixels of the previous frame image contour and the y-axis coordinates of the corresponding pixels of the next frame image contour is calculated, and the absolute value is taken to obtain continuous values. The continuous values ​​are summed according to the number of pixels of the image contour to obtain the image anomaly parameters.

5. A financial management system for sports economic analysis according to claim 1, characterized in that: The specific steps for analyzing whether the image is torn based on the image tearing model are as follows: Set a screen tearing threshold, compare the screen tearing model with the screen tearing threshold. If the screen tearing model is greater than the screen tearing threshold, it means that the screen is torn. If the screen tearing model is less than or equal to the screen tearing threshold, it means that the screen is not torn.

6. A financial management system for sports economic analysis according to claim 5, characterized in that: The specific method for obtaining the screen tearing threshold is as follows: When the image is normal, if both the outline of the previous frame and the outline of the next frame are stationary, the difference between the x-axis or y-axis coordinates of the pixels in the previous frame and the corresponding pixels in the next frame is zero. If the outline of the next frame moves following the outline of the previous frame, the absolute value of the difference between the x-axis or y-axis coordinates of the pixels in the next frame and the corresponding pixels in the previous frame is one. Therefore, by summing consecutive values ​​based on the number of pixels in the image outline, the range of the screen tearing threshold is 0 to... , The product of the number of pixels representing the image outline and one.

7. A financial management system for sports economic analysis according to claim 1, characterized in that: The specific steps for predicting the severity of screen tearing are as follows: S11: Record the values ​​of the screen tearing model as tearing values ​​and count them to obtain a set of tearing values; S12: Record the time interval between the previous tear value and the next tear value according to the tear value set to obtain the tear interval time, and perform a count to obtain the tear interval time set. S13: Establish a two-dimensional coordinate system, with the horizontal axis representing the tearing interval time and the vertical axis representing the tearing value, to obtain the tearing point. Since both the tearing interval time and the tearing value are greater than zero, the tearing point is in the first quadrant. S14: Let the first -1 tear point, the first The tear point and the first tear point +1 tear point, for the first The tear point and the first The slope is calculated at the -1 tear point to obtain the Nth tear slope. The tear point and the first The slope of the Mth tear is obtained by calculating the slope of each tear point. S15: Compare the Mth tear slope with the Nth tear slope. If the Mth tear slope is greater than the Nth tear slope, the image tearing is predicted to be severe. If the Mth tear slope is less than or equal to the Nth tear slope, the image tearing is predicted to be mild.

8. A financial management system for sports economic analysis according to claim 7, characterized in that: The specific method for obtaining the Nth tear slope or the Mth tear slope is as follows: or ; in, Indicates the slope of the Nth tear or This represents the slope of the Mth tear. Indicates the first The tear value at each tear point Indicates the first The tear value at each tear point Indicates the first The tear value at each tear point Indicates the first The tear interval at each tear point Indicates the first The tear interval at each tear point Indicates the first The tear interval time at each tear point.

9. A financial management system for sports economic analysis according to claim 1, characterized in that: The specific steps for analyzing the growth rate of fiscal revenue are as follows: S21: During the time period when the Mth tear slope is greater than the Nth tear slope, the fiscal revenue at this moment is recorded as risky fiscal revenue; before the moment when the Mth tear slope is detected to be greater than the Nth tear slope, the fiscal revenue is recorded as normal fiscal revenue. S22: Compare risky fiscal revenue with normal fiscal revenue. If risky fiscal revenue is less than normal fiscal revenue, it indicates that the growth rate of fiscal revenue has slowed down. Then, the dynamic scheduling algorithm of full-domain reach advertising is used to adjust the situation. If risky fiscal revenue is greater than or equal to normal fiscal revenue, it indicates that the growth rate of fiscal revenue is normal. Then, the comparison continues.

10. A financial management system for sports economic analysis according to claim 9, characterized in that: The specific steps for adjusting the dynamic scheduling algorithm for full-domain reach advertising are as follows: S31: When fiscal revenue slows down, temporarily block user IDs that are detected to have screen tearing; S32: Continue to serve ads to new users. If the screen of this user is also torn, return to S31 and so on. If the screen of this user is normal, continue to serve ads.

Images