A digital media advertisement page monitoring and auditing method and device and a storage medium

Abstract

The application relates to a digital media advertisement page monitoring and auditing method and device and a storage medium, the method of which comprises the following steps: embedding a monitoring plug-in in an advertisement order page, collecting a rendered page panoramic screenshot in a user local browser environment at a key moment of user ordering, and returning the screenshot through an encrypted channel. The screenshot is compared with compliance materials reported by a digital media touch point business in advance by using image similarity comparison and text recognition technology, and an instruction allowing ordering is generated according to a comparison result, so that real-time prevention and control are realized at a business level, and meanwhile, an unalterable data basis is provided for post-auditing. The application realizes real-time and multi-dimensional effective monitoring of the real presentation of advertisement page content at a user end, and binds the monitoring result with an ordering business flow depth, so that pre-prevention and control are realized.

Description

Technical Field

[0001] This invention relates to the field of digital advertising technology, specifically to a method, apparatus, and storage medium for monitoring and auditing digital media advertising pages. Background Technology

[0002] In the digital marketing field, feed ads are a mainstream collaboration model. Advertisers or agencies purchase platform traffic and then target ads precisely based on user tags. However, the final landing page of the ad (i.e., the webpage where the user completes the order) is usually designed and hosted by the agency itself, making it difficult for advertisers to effectively control the page content in real time.

[0003] Currently, the industry's methods for monitoring advertising page content mainly include manual spot checks, URL comparison and content crawling, and simple JavaScript monitoring. For example, some existing technologies require agencies to report page URLs and key elements, and then use timed crawlers to capture the page source code for comparison.

[0004] However, when using URL comparison and content crawling to monitor pages, the "URL remains the same, but page content is modified" behavior cannot be effectively detected when the agent returns different content to the real user and the crawler through server-side technology, thus causing the URL comparison and content crawling monitoring mechanism to fail.

[0005] When using dynamically rendered pages with JavaScript (e.g., pages built with modern front-end frameworks like React and Vue), traditional HTML source code crawlers may not be able to directly obtain the page content that the user ultimately sees. This is because the JavaScript code in the browser executes after the static HTML returned by the server, dynamically modifying the DOM (Document Object Model) and thus changing the content displayed on the page. Consequently, content crawler mechanisms are less effective at monitoring dynamically rendered pages with JavaScript and resisting dynamic rendering and content spoofing.

[0006] URL comparison and content crawling auditing mechanisms all target the webpage code or webpage elements for content comparison auditing. When the webpage code or webpage elements have not been changed, but the agent may use images that are visually very similar to the reported materials but contain misleading words (such as "free" or "lottery"), text comparison cannot detect this, but it will substantially mislead users. It can be seen that the auditing method based on text and code comparison lacks the ability to perceive visual content, resulting in poor auditing ability for webpage images.

[0007] Existing auditing methods typically detect violations only after an order has been placed, making it impossible to identify violations before the order is placed and to block non-compliant orders. This results in a disconnect between monitoring and the order placement process, making it impossible to block violations in real time.

[0008] In existing auditing methods, the collected data is easily tampered with during transmission and storage, leading to unreliable audit results. Summary of the Invention

[0009] To address the shortcomings of existing technologies, this invention provides a method, apparatus, and storage medium for monitoring and auditing digital media advertising pages. The aim is to achieve real-time, multi-dimensional, and effective monitoring of the actual presentation of advertising page content on the user's end, and to deeply integrate the monitoring results with the order placement process to achieve proactive prevention.

[0010] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: A method for monitoring and auditing digital media advertising pages includes the following steps: When a user triggers the intention to place an order, a client-side monitoring script is used to capture a panoramic screenshot of the rendered advertising and marketing page in the user's browser. The similarity between the panoramic screenshot of the advertising marketing page and the preset visual panoramic image is compared based on the image feature matching method to obtain the comparison results; The text information in the panoramic screenshot of the advertising and marketing page is identified and subjected to compliance analysis to obtain the analysis results; Based on the comparison results and the analysis results, a business order placement permission instruction is generated to control the user's order placement process. The validity of the order placement permission instruction is verified to determine whether the order placement operation is allowed to be completed.

[0011] The beneficial effects of this invention are as follows: By embedding a monitoring plugin into the advertising order page, a rendered panoramic screenshot of the page is captured in the user's local browser environment at the critical moment of order placement and transmitted back via an encrypted channel. Using image similarity comparison and text recognition technology, this screenshot is compared with compliant materials pre-reported by the digital media touchpoint partner. Based on the comparison results, a decision is made on whether to generate an order placement instruction, thereby achieving real-time control at the business level and providing tamper-proof data for post-event auditing. Ultimately, this achieves real-time, multi-dimensional, and effective monitoring of the actual presentation of advertising page content on the user's end, and deeply integrates the monitoring results with the order placement business flow, realizing pre-emptive control.

[0012] Based on the above technical solution, the present invention can be further improved as follows.

[0013] Furthermore, it also includes batch collection and storage of panoramic screenshots of advertising and marketing pages corresponding to historical orders according to a preset cycle; batch comparison of the stored panoramic screenshots of advertising and marketing pages with the corresponding preset visual panoramic images of pages; and generation of audit reports or alarm information based on the comparison results.

[0014] Furthermore, the client monitoring script is deployed as an embedded plugin on the agent's advertising order page.

[0015] Furthermore, before capturing and rendering a panoramic screenshot of the advertising and marketing page, a forced waiting mechanism is initiated, and order placement is prohibited until the forced waiting mechanism ends.

[0016] Furthermore, based on image feature matching methods, the panoramic screenshot of the advertising marketing page is compared with the preset visual panoramic image of the page to obtain the comparison results, including the following steps: Extract the image features of the panoramic screenshot of the advertising marketing page and the image features of the preset page visual panoramic image; Based on an image matching algorithm, the matching degree between the image features of the panoramic screenshot of the advertising marketing page and the image features of the preset page visual panoramic image is calculated to obtain a similarity value; wherein, the comparison result includes the similarity value.

[0017] Furthermore, the text information in the panoramic screenshot of the advertising marketing page is identified and subjected to compliance analysis to obtain the analysis results, including the following steps: Identify text information in the panoramic screenshot of the advertising and marketing page; The identified text information is matched with prohibited words in a preset inducement word configuration table to determine whether there are prohibited words in the identified text information. If so, the page violation information is obtained; if not, the page compliance information is obtained. The analysis result includes the page violation information or the page compliance information.

[0018] Furthermore, based on the comparison results and the analysis results, a business order placement permission instruction for controlling the user's order placement process is generated, including the following steps: If the similarity value in the comparison result is greater than a preset threshold, and the analysis result includes the page compliance information, then the page is determined to be compliant, and a key is generated. If the similarity value in the comparison result is less than or equal to the preset threshold, and / or the analysis result includes the page violation information, then the page is determined to be in violation, and a verification failure message is generated; The business order authorization instruction includes the key or the verification failure information.

[0019] Furthermore, the determination of whether to allow the order placement operation to be completed is made by verifying the validity of the business order placement permission instruction, including the following steps: If the business order placement permission instruction includes the key, then the validity of the key is verified to determine whether to allow the order placement operation to be completed. If the order placement permission instruction includes the verification failure information, the order placement operation will be blocked.

[0020] To address the aforementioned technical problems, the present invention also provides a digital media advertising page monitoring and auditing device, the specific technical content of which is as follows: A digital media advertising page monitoring and auditing device, comprising: The embedded monitoring module is used to capture a panoramic screenshot of the rendered advertising and marketing page in the user's browser using a client-side monitoring script when the user triggers the intention to place an order. The real-time audit and decision-making module is used to compare the similarity between the panoramic screenshot of the advertising and marketing page and the preset visual panoramic image of the page based on the image feature matching method, and obtain the comparison result; to identify and perform compliance analysis on the text information in the panoramic screenshot of the advertising and marketing page, and obtain the analysis result; and to generate a business order placement permission instruction for controlling the user's order placement process based on the comparison result and the analysis result. The backward audit scheduling module is used to determine whether to allow the order placement operation to be completed by verifying the validity of the business order placement permission instruction.

[0021] To address the aforementioned technical problems, the present invention also provides a storage medium, the specific technical content of which is as follows: A storage medium storing a computer program or computer instructions, wherein when the computer program or computer instructions are executed by a computer processor, the steps of the above-described digital media advertising page monitoring and auditing method are implemented. Attached Figure Description

[0022] Figure 1 This is a flowchart of a digital media advertising page monitoring and auditing method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a digital media advertising page monitoring and auditing device according to an embodiment of the present invention; Figure 3 This is a flowchart illustrating the configuration of a digital media advertising page monitoring and auditing device in this embodiment. Detailed Implementation

[0023] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0024] Example 1 like Figure 1 As shown, this embodiment provides a method for monitoring and auditing digital media advertising pages, including the following steps: S1. When a user triggers an order placement intention, a client-side monitoring script collects a panoramic screenshot of the rendered advertising and marketing page in the user's browser. This client-side monitoring script is deployed as an embedded plugin within the agent's advertising order placement page. Before collecting the panoramic screenshot of the rendered advertising and marketing page, a forced waiting mechanism is initiated, and order placement is prohibited until the forced waiting mechanism ends. The forced waiting mechanism can be implemented by the client-side monitoring script starting a forced waiting timer, forcing a pre-set waiting time (e.g., 8 seconds), and displaying the waiting status to the user through CSS animation, prohibiting order placement during the waiting period. This ensures the page loads completely and allows the user to fully browse the content, while also providing a stable environment for the plugin to collect data.

[0025] Its client-side monitoring scripts preferentially utilize JS plugins, which are JavaScript plugins; these can effectively improve development efficiency, enrich webpage functionality, and enhance user experience. Commonly used JavaScript plugins mainly include jQuery, Lodash, Moment.js, Chart.js, etc. Among them, jQuery has become the most popular JavaScript library due to its lightweight nature, powerful selectors, excellent browser compatibility, and rich plugin ecosystem.

[0026] When capturing panoramic screenshots of the rendered advertising and marketing page, browser environment data from the user's end is also collected. This browser environment data includes, but is not limited to, user number (serial_number), product code (product_id), browser user agent (browse_ua), complete page URL (page_url), page serial number (page_sn), and client IP (user_ip).

[0027] S2. Based on the image feature matching method, the panoramic screenshot of the advertising marketing page is compared with the preset visual panoramic image of the page to obtain the comparison result; The collected panoramic screenshots of advertising and marketing pages and browser environment data are compressed, encrypted, and digitally signed to obtain encrypted data; After decrypting the encrypted data and verifying the signature, the original panoramic screenshot of the advertising and marketing page and browser environment data are obtained. Based on whether the complete page URL in the browser environment data matches the preset URL, check the validity of the user number (serial_number), product code (product_id), browser user agent (browse_ua), page serial number (page_sn), and client IP (user_ip). If the complete page URL matches the preset URL, and the user number (serial_number), product code (product_id), browser user agent (browse_ua), page serial number (page_sn), and client IP (user_ip) are all valid, then the preliminary review is deemed successful, and the process proceeds to the next step. If the complete page URL does not match the preset URL, and / or any of the user number (serial_number), product code (product_id), browser user agent (browse_ua), page serial number (page_sn), or client IP (user_ip) is invalid, then the preliminary review is deemed unsuccessful, and a preliminary review failure message is returned.

[0028] Image feature extraction (such as SIFT and SURF) and matching algorithms within the OpenCV framework are used to extract image features from the decrypted and verified signed panoramic screenshot of the advertising and marketing page and the image features from the preset page visual panoramic image. Based on the image matching algorithm, the matching degree between the image features of the panoramic screenshot of the advertising and marketing page and the image features of the preset page visual panoramic image is calculated to obtain a similarity value. The comparison result includes the similarity value.

[0029] S3. Identify and perform compliance analysis on the text information in the panoramic screenshot of the advertising marketing page, and obtain the analysis results; this includes the following steps: The system utilizes engines such as Tesseract OCR to recognize text information in the panoramic screenshot of the advertising and marketing page; Tesseract OCR is an open-source optical character recognition (OCR) engine.

[0030] The identified text information is matched with prohibited words in a preset inducement word configuration table to determine whether there are prohibited words in the identified text information. If so, the page violation information is obtained; if not, the page compliance information is obtained. The analysis result includes the page violation information or the page compliance information.

[0031] S4. Generate a business order placement permission instruction for controlling the user's order placement process based on the comparison results and the analysis results; this includes the following steps: If the similarity value in the comparison result is greater than a preset threshold, and the analysis result includes the page compliance information, then the page is deemed compliant, and a key is generated; the preset threshold can be selected as 80%, 85%, 90%, or 95%, etc. If the similarity value in the comparison result is greater than 80%, 85%, 90%, or 95%, and the analysis result includes the page compliance information, then the page is deemed compliant. If the similarity value in the comparison result is less than or equal to the preset threshold, and / or the analysis result includes the page violation information, then the page is determined to be in violation, and a verification failure message is generated; The business order authorization instruction includes the key or the verification failure information.

[0032] S5. Determine whether to allow the order placement operation to be completed by verifying the validity of the business order placement permission instruction; this includes the following steps: If the order placement permission instruction includes the key, the system verifies the validity of the key to determine whether to allow the order placement operation to be completed; the system generates a unique UUID key that is bound to the current user, product, and session, and returns it to the front end; If the order placement permission instruction includes the verification failure information, the order placement operation will be blocked. The system returns the verification failure information to the web frontend, and the web frontend, upon receiving the verification failure information, will prevent the user from proceeding to the next step of the order placement process.

[0033] It also includes the following steps: According to the preset cycle, panoramic screenshots of advertising and marketing pages corresponding to historical orders are collected and stored in batches; the stored panoramic screenshots of advertising and marketing pages are compared with the corresponding preset visual panoramic images of pages in batches, and audit reports or alarm information are generated based on the comparison results.

[0034] A scheduled task is used to trigger at a specified time T (e.g., early morning every day) to batch process the access logs of all orders generated on day T-1.

[0035] Call the same image similarity comparison service as in step S2 above to compare the screenshots in the batch logs with the reported materials.

[0036] Based on the comparison results and preset alarm rules, suspected violation records are filtered out, and alarm notifications are generated. For example, if the similarity value in the comparison results is less than or equal to the preset threshold, an alarm notification is generated.

[0037] Alarm notifications will be generated and sent to operations personnel via SMS or other means. An audit view front-end page will be provided for them to query and export violation details for manual review and follow-up processing.

[0038] By embedding a monitoring plugin into the ad order page, a rendered panoramic screenshot of the page is captured in the user's local browser environment at the critical moment of order placement and transmitted back via an encrypted channel. Using image similarity comparison and text recognition technology, this screenshot is compared with compliant materials pre-submitted by digital media touchpoint partners. Based on the comparison results, a decision is made on whether to generate an order placement permission instruction, thus achieving real-time control at the business level and providing tamper-proof data for post-event auditing. Ultimately, this achieves real-time, multi-dimensional, and effective monitoring of the actual presentation of ad page content on the user's end, and deeply integrates the monitoring results with the order placement business flow, enabling proactive prevention.

[0039] Example 2 Based on Example 1, this example provides the following: Figure 2 The digital media advertising page monitoring and auditing device shown includes an embedded JS plugin and a backend auditing device. The backend auditing device includes a business interface connection module, a backward auditing and alarm scheduling module, an agent reporting material management module, and a real-time auditing and key generation module.

[0040] The embedded JS plugin is deployed on the agent's advertising order page and is responsible for forced waiting, real-time rendering of content collection, data encryption and signing.

[0041] The Agent Reporting Material Management Module is used to receive, store, and manage compliant materials reported by agents. These compliant materials include compliant or preset product codes (product_id), complete page URLs (page_url), page serial numbers (page_sn), etc.

[0042] The real-time audit and key generation module is the core processing module. It is used to receive and decrypt front-end data, perform image similarity comparison and text comparison, and generate a key when the comparison result meets the requirement that the similarity value is higher than a preset threshold (such as 80%) and no leading words (no illegal words or terms) are found in the text of page_image.

[0043] The business interface linkage module provides a verification code verification interface and a modified order placement interface to achieve linkage control with the order placement process.

[0044] The backward audit and alarm scheduling module is used to trigger batch audit tasks on a regular basis and manage alarm configuration and sending.

[0045] Multiple data repositories are used for persistent storage of reported materials, collected data, comparison results, and alarm configurations. Tables for storing reported materials and snapshots, front-end collection logs and comparison results, and alarm configurations are constructed within the data repositories. The table storing reported materials and snapshots stores screenshots of the reported materials (preset page panoramic view), the complete page URL (page_url), and the page serial number (page_sn). The table storing front-end collection logs and comparison results stores front-end collection logs, comparison results, and the aforementioned analysis results. The table storing alarm configurations stores preset alarm rules and generated alarm notifications.

[0046] Example 3 like Figure 3 As shown, this embodiment provides a method for operating or configuring the above-mentioned digital media advertising page monitoring and auditing device, including the following steps: S101. Pre-configuration. Agents submit their marketing materials through the aforementioned material management module. These materials include, but are not limited to, material identifiers, touchpoint codes, product codes, compliant page URLs, and, most importantly, compliant page screenshots (product page snapshots). Compliant page screenshots include the aforementioned preset visual panoramic view of the page. This information is stored in the agent's material management information table. Simultaneously, the system maintains a configuration table for misleading marketing keywords for subsequent OCR text recognition and comparison.

[0047] S102. Page Embedding. The agency integrates a digital media touchpoint management JS plugin (client-side monitoring script) into all of its advertising order pages.

[0048] S103. User Access and Plugin Execution. When a user accesses the page and intends to place an order, the JS plugin is activated and executes the following sub-steps in sequence: S1031. Enabling Forced Waiting. The plugin starts a forced waiting timer (e.g., 8 seconds, or dynamically determined based on DOM events indicating page completion) and displays the waiting status to the user via CSS animation, prohibiting order placement during this period. This ensures the page loads completely and allows the user to fully browse the content, while also providing a stable environment for the plugin to collect data.

[0049] S1032. Collect page screenshots and environmental data. The plugin is deployed in the user's browser environment and performs the following data collection operations: (1) Use html2canvas or a similar library to render the entire element or a specified area as an image (page_image). The element is the core tag in an HTML document used to define the main content of the page. All content visible to the user in the browser (such as text, images, links, tables, forms, etc.) must be placed within the tag.

[0050] (2) Collect browser environment data, including: user number (serial_number), product code (product_id), browser UA (browse_ua), complete page URL (page_url), page serial number (page_sn), client IP (user_ip), etc.

[0051] S1033. Data Security Processing. The plugin compresses, encrypts (e.g., uses RSA algorithm to encrypt key data) and digitally signs the collected panoramic screenshots (page_image) of advertising and marketing pages and browser environment data to obtain encrypted data packets, ensuring that the data is not tampered with during transmission.

[0052] S104. Backend Real-Time Audit and Decision-Making. When a user triggers a verification code request, the plugin sends the encrypted data packet to the All-Customer Platform. The backend real-time audit module processes the data in the following sub-steps: S1041. Data Reception and Decryption. The backend service receives the encrypted data packet, decrypts and verifies the signature to obtain the decrypted data packet, which is the browser environment data mentioned above.

[0053] S1042. Preliminary Consistency Check. The system checks whether the page_url matches the reported URL and verifies the validity of parameters such as page_sn.

[0054] S1043 vs. S1044, multimodal content depth comparison. The system performs the following two comparisons in parallel: S1043. Image Similarity Comparison. The OpenCV framework's image feature extraction algorithms (such as SIFT and SURF) and matching algorithms are called to compare the page_image returned from the front end with the reported material screenshots (preset page panoramic images) and calculate a similarity value (same_ratio).

[0055] S1044, OCR text recognition and lead word filtering. Using engines such as Tesseract OCR, the text in the page_image is recognized and matched against entries in the lead word configuration table to determine if any prohibited words or phrases exist.

[0056] S1045, Decision. The system combines the two comparison results to make a judgment: if the similarity value is higher than the preset threshold (e.g., 80%) and no misleading words are found in the text of page_image (no illegal words or phrases), the page is deemed compliant and proceeds to S1046; if the similarity value is lower than or equal to the preset threshold (e.g., 80%) and / or misleading words are found in the text of page_image (illegal words or phrases), then proceeds to S1047.

[0057] S1046. Generate Key. The system generates a unique UUID key that is bound to the current user, product, and session, and returns it to the front end. Besides a UUID, this unique identifier key can also be other time-sensitive and unique tokens, such as JWT (JSON Web Token).

[0058] S1047, Verification Failure. The system returns a verification failure message, preventing the user from proceeding to the next step of the order placement process.

[0059] S105. Order Placement Integration. When the agent's front-end calls the official order placement interface, it must include this UUID key.

[0060] S1051. Call the order placement interface. The front end calls the order placement interface carrying the UUID key.

[0061] S1052. Key Verification. The order placement interface service in the business interface linkage module will verify the validity of the key. The validity verification includes verifying whether the key exists, whether the key is unused, and whether the key matches the current user and product. When verifying the validity of the key, if the key exists, is unused, and matches the current user and product, the key verification is considered successful. If any of the verifications of whether the key exists, whether the key is unused, or whether the key matches the current user and product is negative, the key verification is considered unsuccessful.

[0062] S1053 and S1054: Business Execution. If the key verification is successful, execute S1053: Allow order placement, order successfully accepted. Otherwise, execute S1054: Reject order placement, process terminated.

[0063] S106. Backward Audit and Alarm. To supplement real-time audit and provide a management view, the system also has a backward process, which is executed asynchronously with the real-time process.

[0064] S1061. Scheduled task triggering. The scheduled task scheduler in the back audit module is triggered at a specified time T (e.g., early morning every day) to batch process the access logs of all orders generated on day T-1.

[0065] S1062. Batch comparison. Call the same image similarity comparison service as in S1043 to compare the screenshots in the batch log with the reported materials.

[0066] S1063. Generate alarms. Based on the comparison results and preset alarm rules, filter out suspected violation records and generate alarm notifications.

[0067] S1064. Notification and Inquiry. Notify operations personnel via SMS or other means, and provide an audit view front-end page for them to query and export violation details for manual review and follow-up processing.

[0068] Compared with traditional methods for monitoring and auditing advertising touchpoints and marketing pages, this invention has the following advantages: 1. By capturing panoramic screenshots of advertising and marketing pages in real time and using them as the core basis for auditing, the problem of server-side crawlers being unable to obtain the real content on the user's end was solved. 2. The image similarity comparison result between real-time screenshots and pre-stored reporting materials in the backend is used to generate the business order placement license key (UUID); this solves the problem of the disconnect between monitoring and order placement business processes, which makes it impossible to block violations in real time; 3. Based on image similarity comparison results, the problem of visual content tampering on web pages was solved; based on OCR text recognition in images, the problem of misleading text on web pages was solved. 4. By introducing a forced waiting mechanism and data encryption strategy in the advertising order process, we ensure the integrity of data collection and prevent skipping of monitoring. 5. By constructing an information table for managing materials distributed by agents, it is easier to compare image similarity and OCR text content, thereby improving the accuracy and efficiency of auditing.

[0069] Example 4 Based on Embodiment 1, this embodiment provides a storage medium storing a computer program or computer instructions. When the computer program or computer instructions are executed by the processor of a computer, the steps of the above-described digital media advertising page monitoring and auditing method are implemented.

[0070] The storage medium can be an internal storage unit of any data processing device described in any of the foregoing embodiments, such as a hard disk or memory. The storage medium can also be an external storage device of any data processing device, such as a plug-in hard disk, smart memory card, SD card, flash memory card, etc., mounted on the device. Furthermore, the storage medium can include both internal storage units and external storage devices of any data processing device. The computer-readable storage medium is used to store the computer program and other programs and data required by the data processing device, and can also be used to temporarily store data that has been output or will be output.

[0071] Example 5 Based on Embodiment 1, this embodiment provides a computer, including a memory and one or more processors. The memory stores executable code. When the one or more processors execute the executable code, they implement the steps of the digital media advertising page monitoring and auditing method in Embodiment 1.

[0072] The memory can be an internal storage unit of any data processing device described in any of the foregoing embodiments, such as a hard disk or RAM. The memory can also be an external storage device of any data processing device, such as a plug-in hard disk, smart memory card, SD card, flash memory card, etc., mounted on the device. Furthermore, the memory can include both internal storage units and external storage devices of any data processing device. The memory is used to store the computer program and other programs and data required by the data processing device, and can also be used to temporarily store data that has been output or will be output.

[0073] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the concept and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for monitoring and auditing digital media advertising pages, characterized in that, Includes the following steps: When a user triggers the intention to place an order, a client-side monitoring script is used to capture a panoramic screenshot of the rendered advertising and marketing page in the user's browser. The similarity between the panoramic screenshot of the advertising marketing page and the preset visual panoramic image is compared based on the image feature matching method to obtain the comparison results; The text information in the panoramic screenshot of the advertising and marketing page is identified and subjected to compliance analysis to obtain the analysis results; Based on the comparison results and the analysis results, a business order placement permission instruction is generated to control the user's order placement process. The validity of the order placement permission instruction is verified to determine whether the order placement operation is allowed to be completed.

2. The digital media advertising page monitoring and auditing method according to claim 1, characterized in that, It also includes batch collection and storage of panoramic screenshots of advertising and marketing pages corresponding to historical orders according to a preset cycle; batch comparison of the stored panoramic screenshots of advertising and marketing pages with the corresponding preset visual panoramic images of pages; and generation of audit reports or alarm information based on the comparison results.

3. The method for monitoring and auditing digital media advertising pages according to claim 1, characterized in that, The client-side monitoring script is deployed as an embedded plugin on the agent's advertising order page.

4. The method for monitoring and auditing digital media advertising pages according to claim 1, characterized in that, Before capturing and rendering a panoramic screenshot of the advertising and marketing page, a forced waiting mechanism is initiated, and order placement is prohibited until the forced waiting mechanism ends.

5. The method for monitoring and auditing digital media advertising pages according to claim 1, characterized in that, The similarity between the panoramic screenshot of the advertising marketing page and the preset visual panoramic image is compared based on the image feature matching method, and the comparison results are obtained, including the following steps: Extract the image features of the panoramic screenshot of the advertising marketing page and the image features of the preset page visual panoramic image; Based on an image matching algorithm, the matching degree between the image features of the panoramic screenshot of the advertising marketing page and the image features of the preset page visual panoramic image is calculated to obtain a similarity value; wherein, the comparison result includes the similarity value.

6. The method for monitoring and auditing digital media advertising pages according to claim 5, characterized in that, The text information in the panoramic screenshot of the advertising marketing page is identified and subjected to compliance analysis to obtain the analysis results, including the following steps: Identify text information in the panoramic screenshot of the advertising and marketing page; The identified text information is matched with prohibited words in a preset inducement word configuration table to determine whether there are prohibited words in the identified text information. If so, the page violation information is obtained; if not, the page compliance information is obtained. The analysis result includes the page violation information or the page compliance information.

7. The method for monitoring and auditing digital media advertising pages according to claim 6, characterized in that, Based on the comparison results and the analysis results, a business order placement permission instruction is generated to control the user's order placement process, including the following steps: If the similarity value in the comparison result is greater than a preset threshold, and the analysis result includes the page compliance information, then the page is determined to be compliant, and a key is generated. If the similarity value in the comparison result is less than or equal to the preset threshold, and / or the analysis result includes the page violation information, then the page is determined to be in violation, and a verification failure message is generated; The business order authorization instruction includes the key or the verification failure information.

8. The method for monitoring and auditing digital media advertising pages according to claim 7, characterized in that, The determination of whether to allow the order placement operation to be completed by verifying the validity of the business order placement permission instruction includes the following steps: If the business order placement permission instruction includes the key, then the validity of the key is verified to determine whether to allow the order placement operation to be completed. If the order placement permission instruction includes the verification failure information, the order placement operation will be blocked.

9. A digital media advertising page monitoring and auditing device employing the method described in any one of claims 1 to 8, characterized in that, include: The embedded monitoring module is used to capture a panoramic screenshot of the rendered advertising and marketing page in the user's browser using a client-side monitoring script when the user triggers the intention to place an order. The real-time audit and decision-making module is used to compare the similarity between the panoramic screenshot of the advertising and marketing page and the preset visual panoramic image of the page based on the image feature matching method, and obtain the comparison result; and to identify and perform compliance analysis on the text information in the panoramic screenshot of the advertising and marketing page, and obtain the analysis result. Based on the comparison results and the analysis results, a business order placement permission instruction is generated to control the user's order placement process. The backward audit scheduling module is used to determine whether to allow the order placement operation to be completed by verifying the validity of the business order placement permission instruction.

10. A storage medium, characterized in that, The storage medium stores a computer program or computer instructions, which, when executed by the computer's processor, implement the steps of the digital media advertising page monitoring and auditing method as described in any one of claims 1 to 8.

Images