Advertisement video intelligent information analysis system

Through the modular processing and natural language analysis of the intelligent advertising video information analysis system, the problem that the existing technology cannot effectively analyze video text is solved, and the accurate evaluation and intelligent analysis of advertising videos are realized, ensuring the accuracy of the evaluation results and Comprehensive.

CN119048168AActive Publication Date: 2024-11-29GUANGZHOU TAIDONG TECH CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411533892.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-31
Publication Date
2024-11-29
Estimated Expiration
2044-10-31

AI Technical Summary

Technical Problem

The prior art cannot effectively analyze text information in videos at the visual level, making it difficult for users to fully and comprehensively understand audience responses and interests, and cannot achieve accurate advertising evaluation.

Method used

Provide an intelligent information analysis system for advertising videos, including video segmentation module, text recognition module, screening module, evaluation module and comprehensive analysis module. Through video segmentation, text recognition, cleaning and high-frequency vocabulary extraction, double cycle based on natural language processing Feedback adjusts video cutting to ensure the accuracy of the evaluation results.

Benefits of technology

Accurate evaluation of advertising videos is achieved, and through modular analysis and natural language analysis, the intelligent and accurate evaluation results are ensured, and the audience's reactions and interests can be deeply understood.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119048168A_ABST
    Figure CN119048168A_ABST
Patent Text Reader

Abstract

The invention provides an advertisement video intelligent information analysis system, which also adopts a modular analysis processing mode, is based on natural language analysis, and is sequentially advanced to a character recognition module, a screening module and an evaluation module to extract high-frequency vocabulary and advertisement evaluation data on the basis of segmenting a video. On this basis, two-stage feedback of coarse adjustment feedback and fine adjustment feedback is carried out on the initial video cutting module, and more accurate video cutting is realized under the two-stage feedback, so that more accurate evaluation results are ensured, and more intelligent advertisement evaluation is ensured in a technical mode.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an intelligent information analysis system for advertising videos. Background Art

[0002] Currently, in the advertising industry, the quality of materials directly affects the advertising effect. Especially for video materials, the text information therein (such as brand names, advertising slogans, etc.) is of great value for brand recognition, content understanding, audience interaction, and keyword analysis.

[0003] However, current technologies mainly rely on extracting subtitle files or generating text from audio, which cannot effectively analyze the text appearing in the video at the visual level. In addition, traditional OCR technologies are mainly applied to the text recognition of static images and cannot deeply extract and screen effective information. This makes it difficult for users to comprehensively, quantitatively, and comprehensively understand the reactions and interests of the audience.

[0004] Therefore, there is an urgent need in this field for an accurate and quantitative technology that can achieve precise advertising evaluation. Summary of the Invention

[0005] The present invention provides an intelligent information analysis system for advertising videos, effectively solving the above problems in the prior art.

[0006] Specifically, the present invention provides an intelligent information analysis system for advertising videos, which includes a video segmentation module, a text recognition module, a screening module, an evaluation module, and a comprehensive analysis module. The video segmentation module extracts one frame of advertising image from an advertising video with a duration of T every period t, thus extracting a total of n advertising images. T > t, n = [T / t]. The text recognition module uses an OCR tool to recognize the advertising text in the n advertising images. The screening module cleans the advertising text and then extracts k high-frequency words from the advertising text. Set the rated number of high-frequency words as K. If k < K, send a rough adjustment feedback signal to the video segmentation module, and adjust the period t to t' = t × k 2 / K 2 , the advertising images are adjusted to n' = [T / t']. The evaluation module captures the advertising evaluations of a total number A based on the n' advertising images. Set the advertising evaluation quantity threshold as A*. If A ≤ A*, send a fine adjustment feedback signal to the video segmentation module. The video segmentation module updates the period t' to t'' = t' × A / A*. The video cutting module cuts out n'' advertising images, n'' = [T / t'']. The screening module extracts k'' high-frequency words for the n'' advertising images. The comprehensive analysis module comprehensively and quantitatively determines the comprehensive vocabulary evaluation index obtained by each high-frequency word based on the k'' high-frequency words and the advertising evaluations with a total number of A.

[0007] In particular, the advertising text contains the brand name of the advertisement.

[0008] Preferably, the screening module cleans the advertisement text by removing special characters and meaningless symbols in the advertisement text, unifying the case of letters in the advertisement text, and eliminating stop words.

[0009] Preferably, in the screening module, if k≥K, the time period t remains unchanged, t'=t, and accordingly, the number of advertisement screens also remains unchanged, then n'=n.

[0010] Preferably, the screening module sets a threshold number of times, and the words in the advertisement text that exceed the threshold number of times are determined by the screening module as high-frequency words.

[0011] Preferably, in the evaluation module, all advertisement evaluations of all n' frames corresponding to the n' advertisement pictures are extracted, and the number of advertisement evaluations of each frame in the n' advertisement pictures is a. i , where the integer i ranges from 1 to n', so the total number of advertising evaluations for the n' advertising screens is .

[0012] Preferably, in the evaluation module, if A>A*, the time period t' remains unchanged, t''=t', and accordingly, the number of advertisement screens n''=n'.

[0013] Preferably, in the comprehensive analysis module, the total number A of advertising evaluations is quantified into A evaluation indexes one by one through natural language processing. The evaluation indexes are divided into positive evaluation indexes and negative evaluation indexes. The positive evaluation index is quantified as a positive number, and the negative evaluation index is quantified as a negative number.

[0014] Preferably, in the comprehensive quantitative determination, the number of any high-frequency words extracted from any j-th advertising screen in the n'' advertising screens is counted as b j , where the integer j ranges from 1 to n'', and the quantitative evaluation index of the advertisement evaluation of the frame where the j-th advertisement picture is located is B j , then the vocabulary evaluation index of any high-frequency word in the jth advertising screen is m j =B j / b j If any of the high-frequency words does not appear in the j-th advertising screen, the vocabulary evaluation index m of any of the high-frequency words in the advertising screen is j As the integer j changes from 1 to n'', the comprehensive vocabulary evaluation index of any high-frequency word is calculated. .

[0015] In summary, the present invention provides an advertising video intelligent information analysis system, which also adopts a modular analysis and processing method. Based on natural language analysis, the system sequentially advances to the text recognition module, the screening module, and the evaluation module on the basis of video segmentation to extract high-frequency vocabulary and advertising evaluation data, and on this basis, performs two levels of feedback, namely, coarse adjustment feedback and fine adjustment feedback, to the initial video cutting module. Under the two-level feedback, more accurate video cutting is achieved, thereby ensuring more accurate evaluation results, thereby ensuring more intelligent advertising evaluation in a technical way. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be discussed below. Obviously, the technical solutions described in conjunction with the drawings are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments and their drawings can be obtained based on the embodiments shown in these drawings without paying creative work.

[0017] Figure 1 The operating flow chart of the advertising video intelligent information analysis system according to the present invention is shown. DETAILED DESCRIPTION

[0018] The technical solutions of various embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments described in 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.

[0019] In summary, the present invention provides an intelligent information analysis system for advertising videos. First, the video segmentation module segments the video and extracts the images of the corresponding time frames. The text recognition module uses OCR technology to identify the text information of each frame of the image, including the brand name and other advertising-related information. The screening module uses natural language technology to perform semantic analysis on the recognized text information and screen out high-frequency words. At the same time, based on the number of high-frequency words, it potentially provides feedback to the video segmentation module. The evaluation module deeply analyzes the relationship between high-frequency words and audience evaluations, and gives relevant evaluation results. At the same time, it potentially sends feedback to the video segmentation module to ensure that the evaluation results are more accurate. It can be seen that in the system of the present invention, double-loop feedback is carried out, and large and small feedback ensure that the overall analysis is extremely accurate.

[0020] The following will combine Figure 1 The specific contents of the advertising video intelligent information analysis system provided by the present invention are introduced in detail. Figure 1The operating flow chart of the advertising video intelligent information analysis system according to the present invention is shown.

[0021] like Figure 1 As shown, the system first starts the segmentation module to segment a piece of advertising video. Specifically, for an advertising video with a duration of T, one frame of advertising screen is extracted from the video every time period t, thereby extracting n=[T / t] advertising screens, where T>t. The symbol "[]" here means rounding. For example, for an advertising video with a duration of T=53 minutes, one frame of advertising screen is extracted every 5 minutes, and a total of 10 (=[53 / 5]) frames of advertising screen are extracted.

[0022] It will be mentioned below that for the original advertising video, the duration T is a fixed value, but the time period t may be adjusted in two ways according to the subsequent feedback. The reason why the present invention needs to consider adjusting the time period t is that if the time period t is too long, the number of extracted pictures is small, it will be difficult to feedback the complete content of the advertisement, and the analysis result will be too rough. On the contrary, if the time period t is too short, the number of extracted pictures is too large, which will greatly increase the amount of analysis, resulting in an overly complicated analysis process and easily leading to too much junk information in the analysis.

[0023] Then, a text recognition module is started to extract high-frequency words from the n advertisement images.

[0024] Specifically, the text recognition module uses an OCR tool (such as Tesseract, EasyOCR, etc.) to recognize and extract the advertisement texts in the n advertisement pictures.

[0025] These advertisement texts contain all text information in the screen, including brand names and other advertisement-related information, such as advertising language, performance and price of the corresponding products, etc. The relevant codes can be described as follows: import pytesseract from PlL import Image # Load image file image path = 'frame0.jpg' image = image.open(image_path) # Perform OCR recognition text = pytesseract.image_to_string(image) print(text) At this time, the screening module is started to perform semantic extraction on the advertisement text. First, the screening module cleans the advertisement text. For example, it cleans special characters and meaningless symbols in the advertisement text, unifies the case of letters in the advertisement text, and eliminates stop words (such as words like "of", "is", "in", etc.). For example, the relevant code is as follows: import re from nltk.corpus import stopwords # Assume the data is saved in a list texts = "How to use OCR to identify high-frequency words in a video? "What is the application of OCR in image processing?" Can video analysis extract high-frequency words? "What impact do high-frequency words have on text analysis?] # Download NLTK stop words import nltk nltk.download('stopwords") stop words= set(stopwords.words('chinese”)) # Use in cleaned texts = for text in texts: # Clean the text text = re.sub(r"[^\W\s]”,"”,text) # Remove punctuation text = text.lower() # Convert to lowercase cleaned texts.append(text) # Also remove stop words tokenized texts =[] for text in cleaned texts: words=text.split() # Tokenize filtered words = [word for word in words if word not tokenizedtexts.extend(filtered words) # Add to the total words Then the screening module extracts high-frequency words from the cleaned advertisement text. Specifically, words that appear more than a threshold number of times in the advertisement text can be set as high-frequency words, and vice versa.

[0026] The screening module then counts the extracted high-frequency words, with k high-frequency words counted. Set the rated quantity K of high-frequency words. If k < K, it indicates that the extraction quantity of high-frequency words in the advertisement text is insufficient. At this time, the coarse adjustment feedback is activated, and the screening module sends a coarse adjustment feedback signal to the video segmentation module to perform a multiple adjustment on the time period t.

[0027] For example, through the multiple adjustment, the time period t is adjusted to the time period t’, and the specific calculation formula is: t’ = t × k 2 / K 2 , correspondingly, the video cutting module cuts out n’ advertisement frames, where n’ = [T / t’]. Through this multiple adjustment, the number of advertisement frames that the video segmentation module can segment can be greatly increased. The updated n’ advertisement frames are input into the character recognition module for the recognition of the advertisement text, thereby increasing the number of recognition samples and making the recognition of the advertisement text more accurate.

[0028] If k ≥ K, then the time period t remains unchanged, t’ = t, and correspondingly, the number of advertisement frames also remains unchanged, then n’ = n.

[0029] It should be noted that as mentioned above, the system of the present invention enables a dual feedback mechanism. Therefore, the coarse adjustment feedback here is one of the feedback mechanisms, and another feedback mechanism, that is, the fine adjustment feedback, will be mentioned below.

[0030] Input the current n’ advertisement frames and all high-frequency words into the evaluation module. It should be noted that when k ≥ K, the integer n’ takes the value of the original number of advertisement frames n, that is, n’ = n. When k < K, the integer n’ takes the value of n’ = [T / t’] = [T / (t × k 2 / K 2 )]. In other words, the number n’ of advertisement frames in the evaluation module may have two different values according to the above different situations.

[0031] The ultimate goal of the evaluation module is to give a final quantitative evaluation result for the advertisement video. But before giving the final quantitative evaluation result, the evaluation module also makes a fine adjustment feedback, which is exactly another feedback mechanism in the dual feedback mechanism mentioned above in addition to the coarse adjustment feedback.

[0032] As is well known, the audience may evaluate the advertisement video at any moment during the playback of the advertisement video. In the evaluation module, all advertisement evaluations of the n’ frames corresponding to the n’ advertisement frames will be extracted (for example, by a crawler program), and the number of advertisement evaluations for each frame in the n’ advertisement frames is a i , where the integer i takes values from 1 to n’. Therefore, the total number of advertisement evaluations of the n’ advertisement frames is , set the threshold of the number of advertisement evaluations to A*. If A≤A*, it indicates that the number of advertisement evaluations is insufficient. In other words, the number of n' advertisement screens extracted is insufficient. Then the evaluation module will send a fine-tuning feedback signal to the video segmentation module. The video segmentation module adjusts the time period t' by multiples, thereby updating the time period t' to t''=t'ⅹA / A*. Accordingly, the video cutting module cuts out n'' advertisement screens, where n''=[T / t''].

[0033] If A>A*, the time period t' remains unchanged, t''=t', and accordingly, the number of advertising screens also remains unchanged, then n''=n'. By adjusting this multiple, the number of advertising screens that can be segmented by the video segmentation module can be slightly increased, and the updated n'' advertising screens are input into the text recognition module for advertising text recognition, thereby further increasing the number of recognized samples through fine adjustment on the basis of rough adjustment, and also making the recognition of advertising text more accurate. Accordingly, the screening module extracts k'' high-frequency words for n'' advertising screens.

[0034] Finally, the system starts a comprehensive analysis module to comprehensively and quantitatively determine the evaluation index obtained by each high-frequency word based on the k'' high-frequency words and the A advertisement evaluations.

[0035] For example, the A advertising evaluations can be quantified into A evaluation indexes through natural language processing. It should be noted that the audience's evaluations are often divided into positive evaluations and negative evaluations, and positive evaluations are quantified as positive numbers, while negative evaluations are quantified as negative numbers. For example, if an advertisement is evaluated as "I like this advertising video very much", the word "like" is extracted from the advertisement evaluation through natural language processing and quantified as "+2". Another advertisement is "This advertising video is so wonderful", and the word "wonderful" is extracted from the advertisement evaluation through natural language processing and quantified as "+3"; otherwise, it is a negative evaluation. For example, if an advertisement is evaluated as "I hate this advertising video", the word "hate" is extracted through natural language processing and quantified as "-2".

[0036] In the above comprehensive quantitative determination, the number of any high-frequency words extracted from any j-th advertising screen in the n'' advertising screens is counted as b j The quantitative evaluation index of the advertisement evaluation of the frame where the j-th advertisement picture is located is B j , then the vocabulary evaluation index of any high-frequency word in the jth advertising screen is m j =B j / b j If the jth advertisement screen does not contain any of the high-frequency words, the vocabulary evaluation index m jAs the integer j changes from 1 to n'', the comprehensive vocabulary evaluation index of any high-frequency word can be calculated. .

[0037] The invention has been basically introduced here. In summary, the invention provides an advertising video intelligent information analysis system, which also adopts a modular analysis and processing method. Based on natural language analysis, the system sequentially advances to the text recognition module, the screening module, and the evaluation module on the basis of video segmentation processing to extract high-frequency words and advertising evaluation data, and on this basis, provides two levels of feedback, namely, rough adjustment feedback and fine adjustment feedback, to the initial video cutting module. Under the two-level feedback, the video cutting is made more accurate, thereby ensuring that the evaluation result is more accurate, thereby ensuring that the advertising evaluation is more intelligent in a technical way.

[0038] The above description is only an exemplary embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An advertising video intelligent information analysis system, characterized in that: The system includes a video segmentation module, a text recognition module, a screening module, an evaluation module, and a comprehensive analysis module. The video segmentation module extracts one frame of advertising screen every time period t from an advertising video of duration T, thereby extracting a total of n advertising screens, T>t, n=[T / t], The text recognition module uses an OCR tool to recognize the advertisement texts in the n advertisement images. The screening module cleans the advertisement text and then extracts k high-frequency words from the advertisement text. Set the rated number K of high-frequency words. If k < K, send a coarse adjustment feedback signal to the video segmentation module, and adjust the time period t to the time period t' = t × k 2 / K 2 , and adjust the advertisement screen to n' = [T / t'], The evaluation module captures the total number A of advertising evaluations based on the n' advertising screens, sets the advertising evaluation number threshold to A*, and if A≤A*, sends a fine-tuning feedback signal to the video segmentation module, the video segmentation module updates the time period t' to t''=t'ⅹA / A*, the video cutting module cuts out n'' advertising screens, n''=[T / t''], and returns to the screening module to extract k'' high-frequency words for the n'' advertising screens. The comprehensive analysis module comprehensively and quantitatively determines the comprehensive vocabulary evaluation index obtained for each high-frequency vocabulary based on the k'' high-frequency vocabulary and the total number of advertisement evaluations A.

2. The system according to claim 1, characterized in that The advertisement text includes the brand name of the advertisement.

3. The system according to claim 1, characterized in that The screening module cleans the advertisement text by removing special characters and meaningless symbols in the advertisement text, unifying the case of letters in the advertisement text, and eliminating stop words.

4. The system according to claim 1, characterized in that In the screening module, if k ≥ K, the time period t remains unchanged, t'=t, and accordingly, the number of advertising screens also remains unchanged, then n'=n.

5. The system according to claim 1, characterized in that The screening module sets a threshold number of times, and the words in the advertisement text that exceed the threshold number of times are determined by the screening module as high-frequency words.

6. The system according to claim 1, characterized in that In the evaluation module, all advertisement evaluations of all n' frames corresponding to the n' advertisement pictures are extracted, and the number of advertisement evaluations of each frame in the n' advertisement pictures is a. i , where the integer i ranges from 1 to n', so the total number of advertising evaluations for the n' advertising screens is .

7. The system according to claim 1, characterized in that In the evaluation module, if A>A*, the time period t' remains unchanged, t''=t', and accordingly, the number of advertising screens n''=n'.

8. The system according to claim 1, characterized in that In the comprehensive analysis module, the total number A of advertising evaluations is quantified into A evaluation indexes one by one through natural language processing. The evaluation indexes are divided into positive evaluation indexes and negative evaluation indexes. The positive evaluation index is quantified as a positive number, and the negative evaluation index is quantified as a negative number.

9. The system according to claim 8, characterized in that In the comprehensive quantitative determination, the number of any high-frequency words extracted from any j-th advertising screen in the n'' advertising screens is counted as b j , where the integer j ranges from 1 to n'', and the quantitative evaluation index of the advertisement evaluation of the frame where the j-th advertisement picture is located is B j , then the vocabulary evaluation index of any high-frequency word in the jth advertising screen is m j =B j / b j If any of the high-frequency words does not appear in the j-th advertising screen, the vocabulary evaluation index m of any of the high-frequency words in the advertising screen is j As the integer j changes from 1 to n'', the comprehensive vocabulary evaluation index of any high-frequency word is calculated. .

Citation Information

Patent Citations

  • Advertisement short video quality evaluation method and system based on big data analysis, and storage medium

    CN114639051A

  • Advertisement putting control method and device for video file, equipment and storage medium

    CN114900721A

  • Enterprise operation intelligent management system based on big data

    CN118469649A

  • 3D convolutional neural networks for television advertisement detection

    US10706286B1

  • Video frame rate control method and apparatus, and electronic device and storage medium

    WO2023179161A1