CN111832296A - Multimedia advertisement intelligent delivery management system based on big data - Google Patents
Multimedia advertisement intelligent delivery management system based on big data Download PDFInfo
- Publication number
- CN111832296A CN111832296A CN202010691474.0A CN202010691474A CN111832296A CN 111832296 A CN111832296 A CN 111832296A CN 202010691474 A CN202010691474 A CN 202010691474A CN 111832296 A CN111832296 A CN 111832296A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- delivery
- field
- playing
- dynamic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/355—Class or cluster creation or modification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- General Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a multimedia advertisement intelligent delivery management system based on big data, which comprises an advertisement type classification module, a delivery field statistic classification module, an advertisement duration division module, a playing mode setting analysis module, a central server and a playing display terminal, the multimedia advertisement intelligent delivery management system based on big data analyzes the delivery fields of advertisements through the advertisement type classification module and the delivery field statistic classification module to obtain an advertisement set of each delivery field, acquires the playing mode of each delivery field by combining the playing mode setting analysis module and plays the playing mode at the delivery field playing display terminal to aim at target people, realizes accurate delivery of advertisements, locks the activity range of the target audience, enhances advertisement interactivity, improves the popularity of advertisement brands, and enables the advertisements to reach the brain of the target audience directly, the advertising effect is enlarged, and the advertising cost is saved.
Description
Technical Field
The invention relates to the technical field of advertisement putting management, in particular to a multimedia advertisement intelligent putting management system based on big data.
Background
The enterprise marketing strategy is greatly changed along with the development of the network advertising market at any time, and the traditional home marketing is excessive to the current advertising marketing, so that the advertising marketing has incomparable advantages, such as comprehensive coverage, close to citizens, wide audience and improvement of brand awareness, and the traditional marketing is undoubtedly superior; the advertising can be released all the day, repeated reading is carried out, the brand awareness of consumption is enhanced, so that a plurality of nameplate enterprises select advertisements to be continuously released, forgetfulness is the natural attribute of consumers, according to the forgetting curve of people, even if the impression is repeatedly strengthened, if the impression is not strengthened for a period of time, the advertisements can slowly disappear in the heart of the consumers, and the advertisements are the reason for continuously releasing the advertisements, and the advertisements play a key role in shaping the brand image, help the product to build the image, and cultivate the trust and loyalty of the consumers to the product, thereby indirectly promoting the sale.
In order to realize accurate delivery of delivered advertisements, advertisement delivery management needs to be carried out on the delivered advertisements, and efficient utilization of advertisement resources is ensured.
Disclosure of Invention
The invention aims to provide a multimedia advertisement intelligent delivery management system based on big data, which analyzes the delivery field of each advertisement through an advertisement type classification module and a delivery field statistical classification module so as to obtain an advertisement set of each delivery field, and combines a playing mode setting analysis module to obtain the playing mode of each delivery field and play the playing mode on a display terminal of the delivery field, thereby solving the problems mentioned in the background technology.
The purpose of the invention can be realized by the following technical scheme:
a multimedia advertisement intelligent delivery management system based on big data comprises an advertisement type classification module, a delivery field statistics classification module, an advertisement duration division module, a playing mode setting analysis module, a central server and a playing display terminal;
the advertisement type classification module is used for classifying a plurality of advertisements stored in an advertisement database into a static advertisement set and a dynamic advertisement set according to whether image signals of the advertisements are in a static or continuous classification mode, wherein the static advertisements are picture advertisements, the display forms of the advertisements are pictures and characters, the dynamic advertisements are animation video advertisements, the display forms of the advertisements are animation or video and audio and characters, and the advertisement type classification module sends the classified static advertisement set and dynamic advertisement set to the delivery field statistics classification module and sends the dynamic advertisement set to the advertisement duration division module;
the delivery field statistic classification module is connected with the advertisement type classification module and receives the static advertisement set and the dynamic advertisement set sent by the advertisement type classification module, identifying the characters on the advertisement picture in each static advertisement in the received static advertisement set, extracting the domain keywords of the recognized characters so as to analyze the advertisement delivery domain, thereby obtaining the delivery domain of each static advertisement in the static advertisement set, identifying the text on the video or animation or audio in each dynamic advertisement in the received dynamic advertisement set, extracting domain keywords of the recognized characters so as to analyze the advertisement delivery domain, thereby obtaining the delivery domains of the dynamic advertisements in the dynamic advertisement set, and sending the counted delivery domains of the static advertisements and the dynamic advertisements to a central server by a delivery domain counting and classifying module;
the central server is connected with the delivery field statistical classification module, receives the delivery fields of the static advertisements and the delivery fields of the dynamic advertisements sent by the delivery field statistical classification module, uniformly stores the static advertisements or the dynamic advertisements in the same delivery field to form a delivery field advertisement set, and sends the delivery field advertisement set to the playing mode setting analysis module;
the advertisement duration dividing module is connected with the advertisement type classifying module, receives the dynamic advertisement sets sent by the advertisement type classifying module, counts the advertisement duration of each dynamic advertisement for the received dynamic advertisement sets to form a dynamic advertisement duration set T (T1, T2,. once, tj,. once, tm) (tj is less than or equal to 60s), tj represents the duration of the jth dynamic advertisement, and sends the dynamic advertisement duration set to the playing mode setting and analyzing module;
the playing mode setting and analyzing module is respectively connected with the advertisement time length dividing module and the central server, receives the dynamic advertisement time length set sent by the advertisement time length dividing module, receives the advertisement set of a release field sent by the central server, and sets the advertisement of a certain release field
askThe kth static advertisement, a, represented as the serving areadgThe method comprises the steps of representing the g-th dynamic advertisement in the launching field, representing sl as the number of the static advertisements in the launching field, representing ds as the number of the dynamic advertisements in the launching field, screening the duration of each dynamic advertisement according to a received dynamic advertisement duration set for each dynamic advertisement in an advertisement set in a certain launching field, dividing each dynamic advertisement into a 5s duration grade, a 10s duration grade, a 15s duration grade, a 30s duration grade and a 60s duration grade according to different lengths of the dynamic advertisement durations, analyzing the advertisement playing mode in a preset single-round playing duration, inserting the dynamic advertisement and the static advertisement combination of each duration grade into a preset single-round playing duration for playing, and sending the analyzed advertisement playing mode to a playing display terminal by a playing mode setting analysis module.
Preferably, the system further comprises an advertisement database, which is respectively connected with the advertisement type classification module, the delivery field statistics classification module, the playing mode setting analysis module and the playing display terminal, and is used for storing a plurality of advertisements, storing field keywords in each delivery field, storing preset static advertisement delivery frequency, and storing preset single-round advertisement playing sequence, wherein the delivery frequency is the number of static advertisements played in 1s, and the delivery fields include markets, hospitals, subways and elevators, but are not limited thereto.
Furthermore, the analysis of the static advertisement delivery field adopts OCR character recognition technology, and the analysis method comprises the following steps:
s1: carrying out binarization processing on the advertisement picture of the static advertisement to obtain a text image and a background image;
s2: carrying out noise removal processing on the obtained text image to obtain a preprocessed text image;
s3: segmenting words on the preprocessed text image to obtain each phrase, further extracting a domain keyword from each phrase, and matching the domain keyword with the domain keyword in each delivery domain stored in a preset advertisement database;
s4: and counting the matching degree of the extracted domain keywords and the domain keywords in each delivery domain stored in the advertisement database, screening the domain keywords with the highest matching degree, and outputting the delivery domain corresponding to the domain keywords with the highest matching degree as the delivery domain of the advertisement when the screened highest matching degree is greater than a set matching degree threshold value.
Further, in the dynamic advertisement, a video text recognition technology is adopted according to the video or animation analysis delivery field, and the analysis method specifically comprises the following steps:
w1: dividing the video or animation in the dynamic advertisement into a plurality of video frames or a plurality of animation frames;
w2: dividing a plurality of video frames or a plurality of animation frames into text video/animation frames and non-text video/animation frames according to whether text information exists in the plurality of divided video frames or the plurality of animation frames or not;
w3: carrying out full-width positioning on the obtained text video/animation frame, and dividing the text video/animation frame into a text candidate region and a non-text region;
w4: the bilinear interpolation method is utilized for the text candidate area image, the pixel density of the character area image is improved, the image resolution is increased, and then the binaryzation processing is carried out on the processed character area image to obtain a character image and a background image;
w5: extracting the keywords of the delivery fields in the character image by using an OCR character recognition technology, and matching the keywords with the keywords in each delivery field stored in a preset advertisement database;
w6: and counting the matching degree of the extracted delivery field keywords and the delivery field keywords in each delivery field stored in the advertisement database, screening the delivery field keywords with the highest matching degree, and outputting the delivery field corresponding to the delivery field keywords with the highest matching degree as the delivery field of the advertisement when the screened highest matching degree is greater than a set matching degree threshold value.
Further, in the dynamic advertisement, the analysis method comprises the following steps:
h1: separating the audio in the dynamic advertisement by an audio separation technology for the video in the dynamic advertisement, and performing endpoint detection and voice enhancement processing on the separated audio to obtain primary audio;
h2: capturing the feature vectors in the primary audio, simultaneously extracting a voice template library prestored in an advertisement database, sequentially matching the captured audio feature vectors with each template in the voice template library, counting the matching similarity between the captured audio feature vectors and each template in the voice template library, and screening the voice template with the maximum similarity;
h3: according to the definition of the screened language module, obtaining the text recognition result of the audio by looking up a table;
h4: segmenting the obtained text recognition result to obtain each phrase, extracting the keywords of the launching field of each phrase, and matching the keywords with the keywords of each launching field stored in a preset advertisement database;
h5: and counting the matching degree of the extracted delivery field keywords and the delivery field keywords in each delivery field stored in the advertisement database, screening the delivery field keywords with the highest matching degree, and outputting the delivery field corresponding to the delivery field keywords with the highest matching degree when the screened highest matching degree is greater than a set matching degree threshold value.
Further, the advertisement playing mode analysis function is T ═ 5a +10b +15c +30d +60e + TsWherein T is the preset single-round playing time length, a, b, c, d and e are the advertisement numbers of 5s, 10s, 15s, 30s and 60s long ranges respectively, a, b, c, d and e are positive integers, TsExpressed as the length of time the static advertisement is played;
randomly extracting the advertisement number corresponding to each time long file from the dynamic advertisement subset in the advertisement set of the delivery field according to the advertisement numbers of the 5s time long file, the 10s time long file, the 15s time long file, the 30s time long file and the 60s time long file analyzed in the advertisement playing mode analysis function, recording the extracted dynamic advertisement numbers,
and according to the time length for playing the static advertisements analyzed in the advertisement playing mode analysis function and the preset static advertisement putting frequency, acquiring the playing number of the static advertisements in the time length, randomly extracting the corresponding number of static advertisements from the static advertisement subset in the advertisement putting field set, and recording the number of each extracted static advertisement.
Furthermore, the playing mode setting analysis module further comprises advertisement playing round statistic analysis, when the next round of advertisement playing is performed, the number of the advertisement played in the previous round is removed from the advertisement set of the playing field, the number corresponding to the dynamic advertisement and the static advertisement of each time long file is extracted from the removed advertisement set of the playing field again, the number of each advertisement is recorded, similarly, the advertisement number played in the previous round is removed from the advertisement set of the playing field, if the removed advertisement set of the playing field has no advertisement, the advertisement of the playing field is played, and if the advertisement set of the playing field still has the advertisement, the advertisement extraction and removal operation is continued until all the advertisements in the advertisement set of the playing field are played.
Furthermore, the playing display terminal is connected with the playing mode setting and analyzing module, receives the advertisement number of each turn of advertisement playing sent by the playing mode setting and analyzing module, extracts each advertisement from the advertisement set corresponding to the putting field, extracts the single turn playing sequence preset in the advertisement database, and plays and displays the advertisement according to the playing sequence.
Has the advantages that:
(1) the invention classifies the types of the advertisements in the advertisement database through the advertisement type classification module, analyzes the putting fields by adopting a targeted putting field analysis mode aiming at the types of the advertisements through the putting field statistic classification module so as to obtain the advertisement sets of the putting fields, sets the playing mode of the analysis module for analyzing the playing mode of each putting field and plays the playing mode at the playing display terminal of the putting field, aims at target people through the playing of the putting field to which the advertisements belong at the playing display terminal of the corresponding putting field, realizes the accurate putting of the advertisements, locks the activity range of the target audiences, enhances the interactivity of the advertisements, improves the popularity of advertisement brands, directly leads the advertisements to the brains of the target audiences, amplifies the advertising effect and saves the advertising cost.
(2) The invention sets the analysis module through the playing mode to play the dynamic advertisement and the static advertisement combination of each time and long grade in the preset single-round playing time length, avoids the viewing fatigue caused by repeating the same advertisement, increases the advertisement viewing experience feeling, displays various advertisement information to the audience in a shorter time, brings strong visual impact to the audience, simultaneously, the advertisement information is repeated at high frequency in the hiding way, so that the audience can feel the advertisement transmission effect more deeply, stimulates the consumption demand, improves the time benefit of the commodity flow, and promotes the economic and prosperous commodity to a certain extent.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic block diagram of the present invention;
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a multimedia advertisement intelligent delivery management system based on big data includes an advertisement database, an advertisement type classification module, a delivery field statistics classification module, an advertisement duration division module, a playing mode setting analysis module, a central server and a playing display terminal.
An advertisement database for storing a plurality of advertisements, domain keywords in each delivery domain, a preset static advertisement delivery frequency, and stores the preset sequence of playing the advertisements in a single round, wherein the frequency of putting is the number of static advertisements played in 1s, the putting fields comprise shopping malls, hospitals, subways, elevators, buses and the like, and the field keywords corresponding to each putting field, for example, the domain keywords corresponding to the mall advertisement include shopping, shop and business recruitment, models, various commodities, etc., the domain keywords corresponding to the elevator advertisement include home, electronic products, building boards, etc., the sequence of the single-turn advertisement playing may be, but is not limited to, a 5 s-hour long file, a static advertisement, a 10 s-hour long file, a static advertisement, a 15 s-hour long file, a static advertisement, a 30 s-hour long file, a static advertisement, and a 60 s-hour long file.
The advertisement type classification module is connected with the advertisement database and used for classifying a plurality of advertisements stored in the advertisement database into a static advertisement set and a dynamic advertisement set according to whether image signals of the advertisements belong to a static or continuous classification mode, wherein the static advertisements are picture advertisements, the display forms of the advertisements are pictures and characters, the dynamic advertisements are animation video advertisements, the display forms of the advertisements are animation or video and audio and characters, the classified static advertisement set and dynamic advertisement set are sent to the delivery field statistics classification module by the advertisement type classification module, and the dynamic advertisement set is sent to the advertisement duration division module.
The advertising field statistical classification module is connected with the advertising type classification module, receives a static advertising set and a dynamic advertising set sent by the advertising type classification module, identifies characters on an advertising picture in each static advertisement in the received static advertising set, extracts field keywords of the identified characters, and analyzes the advertising field of the advertisement by adopting an OCR character identification technology, wherein the OCR identification speed is far faster than manual input, the user experience degree can be improved, and the manual input time is reduced, and the identification method specifically comprises the following steps:
s1: carrying out binarization processing on the advertisement picture of the static advertisement to obtain a text image and a background image;
s2: carrying out noise removal processing on the obtained text image to obtain a preprocessed text image;
s3: segmenting words on the preprocessed text image to obtain each phrase, further extracting a domain keyword from each phrase, and matching the domain keyword with the domain keyword in each delivery domain stored in a preset advertisement database;
s4: and counting the matching degree of the extracted domain keywords and the domain keywords in each delivery domain stored in the advertisement database, screening the domain keywords with the highest matching degree, and outputting the delivery domain corresponding to the domain keywords with the highest matching degree as the delivery domain of the advertisement when the screened highest matching degree is greater than a set matching degree threshold value.
The method comprises the following steps of obtaining the launching field of each static advertisement in a static advertisement set, identifying characters on video or animation or audio in each dynamic advertisement in a received dynamic advertisement set, extracting domain keywords of the identified characters, and analyzing the launching field of the advertisement, wherein a video text identification technology is adopted according to an analysis method of the video or animation analysis launching field, and the method specifically comprises the following steps:
w1: dividing the video or animation in the dynamic advertisement into a plurality of video frames or a plurality of animation frames;
w2: dividing a plurality of video frames or a plurality of animation frames into text video/animation frames and non-text video/animation frames according to whether text information exists in the plurality of divided video frames or the plurality of animation frames or not;
w3: the obtained text video/animation frame is positioned in a whole frame, and the text area has abundant edge, texture, corner and color characteristics and meets a certain rule, the text positioning is carried out by utilizing the characteristics of the text area, and the text video/animation frame is divided into a text candidate area and a non-text area;
w4: the bilinear interpolation method is utilized for the text candidate area image, the pixel density of the character area image is improved, the image resolution is increased, and then the binaryzation processing is carried out on the processed character area image to obtain a character image and a background image;
w5: extracting the keywords of the delivery fields in the character image by using an OCR character recognition technology, and matching the keywords with the keywords in each delivery field stored in a preset advertisement database;
w6: and counting the matching degree of the extracted delivery field keywords and the delivery field keywords in each delivery field stored in the advertisement database, screening the delivery field keywords with the highest matching degree, and outputting the delivery field corresponding to the delivery field keywords with the highest matching degree as the delivery field of the advertisement when the screened highest matching degree is greater than a set matching degree threshold value.
In the embodiment, a video text recognition technology is adopted for detecting and recognizing the video or animation text, the video text is very difficult to detect due to the change of a complex background and a text font, meanwhile, due to the television system, the characters in the video or animation generally have lower resolution, the adopted video text recognition technology has a powerful character positioning function, the positions of the characters in the video can be accurately positioned, the positioned characters are recognized, the recognition result can completely meet the requirements of users, and the difficulty in extracting the video or animation text is solved to a certain extent.
According to the field of audio analysis release, the analysis method specifically comprises the following steps:
h1: separating the audio in the dynamic advertisement by an audio separation technology for the video in the dynamic advertisement, and performing endpoint detection and voice enhancement processing on the separated audio to obtain primary audio;
h2: capturing the feature vectors in the primary audio, simultaneously extracting a voice template library prestored in an advertisement database, sequentially matching the captured audio feature vectors with each template in the voice template library, counting the matching similarity between the captured audio feature vectors and each template in the voice template library, and screening the voice template with the maximum similarity;
h3: according to the definition of the screened language module, obtaining the text recognition result of the audio by looking up a table;
h4: segmenting the obtained text recognition result to obtain each phrase, extracting the keywords of the launching field of each phrase, and matching the keywords with the keywords of each launching field stored in a preset advertisement database;
h5: and counting the matching degree of the extracted delivery field keywords and the delivery field keywords in each delivery field stored in the advertisement database, screening the delivery field keywords with the highest matching degree, and outputting the delivery field corresponding to the delivery field keywords with the highest matching degree when the screened highest matching degree is greater than a set matching degree threshold value.
And the delivery field statistical classification module sends the statistical delivery fields of the static advertisements and the dynamic advertisements to the central server.
The central server is connected with the delivery field statistical classification module, receives the delivery fields of the static advertisements and the delivery fields of the dynamic advertisements sent by the delivery field statistical classification module, uniformly stores the static advertisements or the dynamic advertisements in the same delivery field to form a delivery field advertisement set, and sends the delivery field advertisement set to the playing mode setting analysis module;
the advertisement duration dividing module is connected with the advertisement type classifying module, receives the dynamic advertisement sets sent by the advertisement type classifying module, counts the advertisement duration of each dynamic advertisement for the received dynamic advertisement sets to form a dynamic advertisement duration set T (T1, T2,. once, tj,. once, tm) (tj is less than or equal to 60s), tj represents the duration of the jth dynamic advertisement, and sends the dynamic advertisement duration set to the playing mode setting and analyzing module;
the playing mode setting and analyzing module is respectively connected with the advertisement time length dividing module and the central server, receives the dynamic advertisement time length set sent by the advertisement time length dividing module, receives the advertisement set of a release field sent by the central server, and sets the advertisement of a certain release field
askThe kth static advertisement, a, represented as the serving areadgThe method comprises the steps of representing the g-th dynamic advertisement of the launching field, representing sl as the number of the static advertisements of the launching field, representing ds as the number of the dynamic advertisements of the launching field, screening the duration of each dynamic advertisement according to the received dynamic advertisement duration set for each dynamic advertisement in an advertisement set of a launching field, dividing each dynamic advertisement into a 5s time length grade, a 10s time length grade, a 15s time length grade, a 30s time length grade and a 60s time length grade according to different lengths of the dynamic advertisement durations, analyzing the advertisement playing mode in a preset single-round playing duration, inserting the dynamic advertisement and the static advertisement of each time length grade into a preset single-round playing duration for playing in a combined mode, wherein the advertisement playing mode analysis function is that T is 5a +10b +15c +30d +60e + T ═ TsWherein T is the preset single-round playing time length, a, b, c, d and e are the advertisement numbers of 5s, 10s, 15s, 30s and 60s long ranges respectively, a, b, c, d and e are positive integers, TsExpressed as the length of time the static advertisement is played;
in the advertisement playing mode analysis function provided in this embodiment, all the dynamic advertisements and static advertisements in each time period are displayed in the preset single-round advertisement playing time period, and values of unknown numbers thereof are diversified, and assuming that T is 5min, that is, 300s, values of unknown numbers may be a is 6, b is 5, c is 4, d is 3, e is 1, T is TsThe value of each unknown may be 10, a is 8, b is 4, c is 4, d is 3, e is 1, tsIn this embodiment, the number of the dynamic advertisements in each time period extracted from the dynamic advertisement subset in the advertisement set in the delivery field may be extracted by selecting one group from each group of values, which embodies the intelligence of the system.
And randomly extracting the advertisement number corresponding to each time long file from the dynamic advertisement subset in the advertisement set of the delivery field according to the advertisement numbers of the 5s time long file, the 10s time long file, the 15s time long file, the 30s time long file and the 60s time long file analyzed in the advertisement playing mode analysis function, and recording the extracted dynamic advertisement numbers.
According to the advertisement broadcastThe playing mode analysis function analyzes the time length of playing the static advertisement, obtains the playing number of the static advertisement in the time length according to the preset frequency of putting the static advertisement, and records the number as h (taking a positive integer),
tsexpressing the time length for playing the static advertisement, f expressing the preset static advertisement putting frequency, randomly extracting the corresponding number of static advertisements from the static advertisement subset in the advertisement set of the putting field, recording the extracted static advertisement numbers, removing the advertisement number played in the last round from the advertisement set of the putting field when the next round of advertisement is played, re-extracting the corresponding number of the dynamic advertisements and the static advertisements in each time long file from the removed advertisement set of the putting field, recording the number of each advertisement, similarly removing the advertisement number played in the last round from the advertisement set of the putting field, if the removed advertisement set of the putting field has no advertisement, the advertisement of the putting field is played, if the removed advertisement set of the putting field still has advertisements, continuing the advertisement extraction and removal operation, and all the advertisements in the advertisement set in the release field are completely played, so that the condition that the advertisements which are not played are omitted and loss is brought to advertisers is avoided. The playing mode setting and analyzing module sends the analyzed advertisement playing mode to the playing display terminal, the set playing mode avoids watching fatigue caused by repeatedly playing the same advertisement, the advertisement watching experience feeling is increased, various advertisement information is displayed to audiences in a short time, and strong visual impact is brought to the audiences; meanwhile, the advertisement is repeatedly played in high frequency in turn, so that audiences can feel the advertisement transmission effect more deeply, the advertisement of the product can enter the centers of the audiences in a stealthy manner, the exposure times of brands in target consumer groups are improved, and the requirements of target consumers are stimulated and pulled.
The playing display terminal is connected with the playing mode setting analysis module, a hardware carrier of the playing display terminal comprises a liquid crystal, a plasma display screen, a CRT display, a projector, an LED screen, a DLP splicing wall and the like, the hardware carrier is distributed at each corner of each field, meanwhile, the mobility can be changed at any time according to the following and changing of customer propaganda nodes, the advertisement coverage surface is enlarged, the playing display terminal receives the advertisement number of each turn of advertisement playing sent by the playing mode setting analysis module, each advertisement is extracted from an advertisement set corresponding to the putting field, a preset single turn playing sequence in an advertisement database is extracted, the advertisements are played and displayed according to the playing sequence, the advertisements directly reach the brain of a target audience, the accurate putting effect is achieved, the putting cost is saved, and the cost performance is improved.
The foregoing is merely exemplary and illustrative of the principles of the present invention and various modifications, additions and substitutions of the specific embodiments described herein may be made by those skilled in the art without departing from the principles of the present invention or exceeding the scope of the claims set forth herein.
Claims (8)
1. The utility model provides a multimedia advertisement intelligence puts in management system based on big data which characterized in that: the system comprises an advertisement type classification module, a delivery field statistics classification module, an advertisement duration division module, a playing mode setting analysis module, a central server and a playing display terminal;
the advertisement type classification module is used for classifying a plurality of advertisements stored in an advertisement database into a static advertisement set and a dynamic advertisement set according to whether image signals of the advertisements are static or continuous classification modes, wherein the static advertisements are picture advertisements, the display forms of the advertisements are pictures and characters, the dynamic advertisements are animation video advertisements, the display forms of the advertisements are animation or video and audio and characters, and the advertisement type classification module sends the classified static advertisement set and dynamic advertisement set to the delivery field statistics classification module and sends the dynamic advertisement set to the advertisement duration division module;
the delivery field statistic classification module is connected with the advertisement type classification module and receives the static advertisement set and the dynamic advertisement set sent by the advertisement type classification module, identifying the characters on the advertisement picture in each static advertisement in the received static advertisement set, extracting the domain keywords of the recognized characters so as to analyze the advertisement delivery domain, thereby obtaining the delivery domain of each static advertisement in the static advertisement set, identifying the text on the video or animation or audio in each dynamic advertisement in the received dynamic advertisement set, extracting domain keywords of the recognized characters so as to analyze the advertisement delivery domain, thereby obtaining the delivery domains of the dynamic advertisements in the dynamic advertisement set, and sending the counted delivery domains of the static advertisements and the dynamic advertisements to a central server by a delivery domain counting and classifying module;
the central server is connected with the delivery field statistical classification module, receives the delivery fields of the static advertisements and the delivery fields of the dynamic advertisements sent by the delivery field statistical classification module, uniformly stores the static advertisements or the dynamic advertisements in the same delivery field to form a delivery field advertisement set, and sends the delivery field advertisement set to the playing mode setting analysis module;
the advertisement duration dividing module is connected with the advertisement type classifying module, receives the dynamic advertisement sets sent by the advertisement type classifying module, counts the advertisement duration of each dynamic advertisement for the received dynamic advertisement sets to form a dynamic advertisement duration set T (T1, T2,. once, tj.. once, tm) (tj is less than or equal to 60s), tj represents the duration of the jth dynamic advertisement, and sends the dynamic advertisement duration set to the playing mode setting and analyzing module;
the playing mode setting and analyzing module is respectively connected with the advertisement time length dividing module and the central server, receives the dynamic advertisement time length set sent by the advertisement time length dividing module, receives the advertisement set of a release field sent by the central server, and sets the advertisement of a certain release field
askThe kth static advertisement, a, represented as the serving areadgRepresenting as the g-th dynamic advertisement of the putting field, sl representing as the number of the static advertisements of the putting field, ds representing as the number of the dynamic advertisements of the putting field, and an advertisement set of a certain putting fieldAnd screening the time length of each dynamic advertisement according to the received time length set of the dynamic advertisements, dividing each dynamic advertisement into a 5s time length grade, a 10s time length grade, a 15s time length grade, a 30s time length grade and a 60s time length grade according to different time lengths of the dynamic advertisements, analyzing the advertisement playing mode in the preset single-round playing time length, inserting the combination of the dynamic advertisements and the static advertisements of each time length grade into the preset single-round playing time length for playing, and sending the analyzed advertisement playing mode to the playing display terminal by the playing mode setting and analyzing module.
2. The system for intelligent multimedia advertisement putting management based on big data according to claim 1, characterized in that: still include the advertisement database, be connected with advertisement type classification module, put in the field statistics classification module, broadcast mode setting analysis module and broadcast display terminal respectively for a plurality of advertisements of storage, the field keyword under each field of putting in of storage, the static advertisement frequency of putting in storage and the order of the single round of advertisement broadcast of storage presetting, the number of the static advertisement of broadcast in 1s is put in to the frequency of putting in, the field of putting in includes market, hospital, subway, elevator, but is not limited to this.
3. The system for intelligent multimedia advertisement putting management based on big data according to claim 1, characterized in that: the analysis of the static advertisement delivery field adopts OCR character recognition technology, and the analysis method comprises the following steps:
s1: carrying out binarization processing on the advertisement picture of the static advertisement to obtain a text image and a background image;
s2: carrying out noise removal processing on the obtained text image to obtain a preprocessed text image;
s3: segmenting words on the preprocessed text image to obtain each phrase, further extracting a domain keyword from each phrase, and matching the domain keyword with the domain keyword in each delivery domain stored in a preset advertisement database;
s4: and counting the matching degree of the extracted domain keywords and the domain keywords in each delivery domain stored in the advertisement database, screening the domain keywords with the highest matching degree, and outputting the delivery domain corresponding to the domain keywords with the highest matching degree as the delivery domain of the advertisement when the screened highest matching degree is greater than a set matching degree threshold value.
4. The system for intelligent multimedia advertisement putting management based on big data according to claim 1, characterized in that: in the dynamic advertisement, a video text recognition technology is adopted according to the video or animation analysis delivery field, and the analysis method specifically comprises the following steps:
w1: dividing the video or animation in the dynamic advertisement into a plurality of video frames or a plurality of animation frames;
w2: dividing a plurality of video frames or a plurality of animation frames into text video/animation frames and non-text video/animation frames according to whether text information exists in the plurality of divided video frames or the plurality of animation frames or not;
w3: carrying out full-width positioning on the obtained text video/animation frame, and dividing the text video/animation frame into a text candidate region and a non-text region;
w4: the bilinear interpolation method is utilized for the text candidate area image, the pixel density of the character area image is improved, the image resolution is increased, and then the binaryzation processing is carried out on the processed character area image to obtain a character image and a background image;
w5: extracting the keywords of the delivery fields in the character image by using an OCR character recognition technology, and matching the keywords with the keywords in each delivery field stored in a preset advertisement database;
w6: and counting the matching degree of the extracted delivery field keywords and the delivery field keywords in each delivery field stored in the advertisement database, screening the delivery field keywords with the highest matching degree, and outputting the delivery field corresponding to the delivery field keywords with the highest matching degree as the delivery field of the advertisement when the screened highest matching degree is greater than a set matching degree threshold value.
5. The system for intelligent multimedia advertisement putting management based on big data according to claim 1, characterized in that: in the dynamic advertisement, the analysis method comprises the following steps:
h1: separating the audio in the dynamic advertisement by an audio separation technology for the video in the dynamic advertisement, and performing endpoint detection and voice enhancement processing on the separated audio to obtain primary audio;
h2: capturing the feature vectors in the primary audio, simultaneously extracting a voice template library prestored in an advertisement database, sequentially matching the captured audio feature vectors with each template in the voice template library, counting the matching similarity between the captured audio feature vectors and each template in the voice template library, and screening the voice template with the maximum similarity;
h3: according to the definition of the screened language module, obtaining the text recognition result of the audio by looking up a table;
h4: segmenting the obtained text recognition result to obtain each phrase, extracting the keywords of the launching field of each phrase, and matching the keywords with the keywords of each launching field stored in a preset advertisement database;
h5: and counting the matching degree of the extracted delivery field keywords and the delivery field keywords in each delivery field stored in the advertisement database, screening the delivery field keywords with the highest matching degree, and outputting the delivery field corresponding to the delivery field keywords with the highest matching degree when the screened highest matching degree is greater than a set matching degree threshold value.
6. The system for intelligent multimedia advertisement putting management based on big data according to claim 1, characterized in that: the advertisement playing mode analysis function is T ═ 5a +10b +15c +30d +60e + TsWherein T is the preset single-round playing time length, a, b, c, d and e are the advertisement numbers of 5s, 10s, 15s, 30s and 60s long ranges respectively, a, b, c, d and e are positive integers, TsExpressed as duration of playing static advertisement；
Randomly extracting the advertisement number corresponding to each time long file from the dynamic advertisement subset in the advertisement set of the delivery field according to the advertisement numbers of the 5s time long file, the 10s time long file, the 15s time long file, the 30s time long file and the 60s time long file analyzed in the advertisement playing mode analysis function, recording the extracted dynamic advertisement numbers,
and according to the time length for playing the static advertisements analyzed in the advertisement playing mode analysis function and the preset static advertisement putting frequency, acquiring the playing number of the static advertisements in the time length, randomly extracting the corresponding number of static advertisements from the static advertisement subset in the advertisement putting field set, and recording the number of each extracted static advertisement.
7. The system for intelligent multimedia advertisement putting management based on big data according to claim 1, characterized in that: the playing mode setting and analyzing module further comprises advertisement playing turn statistical analysis, when the next turn of advertisement playing is carried out, the number of the advertisement played in the previous turn is removed from the advertisement set of the playing field, the number corresponding to the dynamic advertisement and the static advertisement of each time long file is extracted from the removed advertisement set of the playing field again, the number of each advertisement is recorded, similarly, the number of the advertisement played in the previous turn is removed from the advertisement set of the playing field, if the removed advertisement set of the playing field has no advertisement, the advertisement in the playing field is played, if the removed advertisement set of the playing field has the advertisement, the operation of extracting and removing the advertisement is continued until all the advertisements in the advertisement set of the playing field are played.
8. The system for intelligent multimedia advertisement putting management based on big data according to claim 1, characterized in that: the playing display terminal is connected with the playing mode setting and analyzing module, receives the advertisement number of each turn of advertisement playing sent by the playing mode setting and analyzing module, extracts each advertisement from the advertisement set corresponding to the putting field, extracts the single turn playing sequence preset in the advertisement database, and plays and displays the advertisement according to the playing sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010691474.0A CN111832296A (en) | 2020-07-17 | 2020-07-17 | Multimedia advertisement intelligent delivery management system based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010691474.0A CN111832296A (en) | 2020-07-17 | 2020-07-17 | Multimedia advertisement intelligent delivery management system based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111832296A true CN111832296A (en) | 2020-10-27 |
Family
ID=72923515
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010691474.0A Withdrawn CN111832296A (en) | 2020-07-17 | 2020-07-17 | Multimedia advertisement intelligent delivery management system based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111832296A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112423078A (en) * | 2020-10-28 | 2021-02-26 | 卡莱特（深圳）云科技有限公司 | Advertisement playing method and device applied to LED display screen |
| CN112561604A (en) * | 2020-12-28 | 2021-03-26 | 惠州华阳通用电子有限公司 | Advertisement pushing method and system |
| CN113205365A (en) * | 2021-05-07 | 2021-08-03 | 武汉连岳传媒有限公司 | Mobile internet advertisement intelligent delivery management method based on big data analysis and cloud service platform |
| CN114501081A (en) * | 2022-02-14 | 2022-05-13 | 浙江新再灵科技股份有限公司 | Screen advertisement playing method and system |
| CN114630187A (en) * | 2022-03-04 | 2022-06-14 | 李亚军 | Accurate customer acquisition system applying signal cloud computing processing |
| CN115222461A (en) * | 2022-09-19 | 2022-10-21 | 杭州数立信息技术有限公司 | Intelligent marketing accurate recommendation method |
| CN116863864A (en) * | 2023-07-11 | 2023-10-10 | 浙江雨林电子科技有限公司 | LED lamp control method and system based on Internet of things |
-
2020
- 2020-07-17 CN CN202010691474.0A patent/CN111832296A/en not_active Withdrawn
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112423078A (en) * | 2020-10-28 | 2021-02-26 | 卡莱特（深圳）云科技有限公司 | Advertisement playing method and device applied to LED display screen |
| CN112561604A (en) * | 2020-12-28 | 2021-03-26 | 惠州华阳通用电子有限公司 | Advertisement pushing method and system |
| CN113205365A (en) * | 2021-05-07 | 2021-08-03 | 武汉连岳传媒有限公司 | Mobile internet advertisement intelligent delivery management method based on big data analysis and cloud service platform |
| CN114501081A (en) * | 2022-02-14 | 2022-05-13 | 浙江新再灵科技股份有限公司 | Screen advertisement playing method and system |
| CN114501081B (en) * | 2022-02-14 | 2023-10-17 | 浙江新再灵科技股份有限公司 | Screen advertisement playing method and system |
| CN114630187A (en) * | 2022-03-04 | 2022-06-14 | 李亚军 | Accurate customer acquisition system applying signal cloud computing processing |
| CN115222461A (en) * | 2022-09-19 | 2022-10-21 | 杭州数立信息技术有限公司 | Intelligent marketing accurate recommendation method |
| CN116863864A (en) * | 2023-07-11 | 2023-10-10 | 浙江雨林电子科技有限公司 | LED lamp control method and system based on Internet of things |
| CN116863864B (en) * | 2023-07-11 | 2024-03-22 | 浙江雨林电子科技有限公司 | LED lamp control method and system based on Internet of things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111832296A (en) | Multimedia advertisement intelligent delivery management system based on big data | |
| CN108322788B (en) | Advertisement display method and device in live video | |
| US9271035B2 (en) | Detecting key roles and their relationships from video | |
| WO2020259510A1 (en) | Method and apparatus for detecting information embedding region, electronic device, and storage medium | |
| US9043860B2 (en) | Method and apparatus for extracting advertisement keywords in association with situations of video scenes | |
| CN101489139B (en) | Video advertisement correlation method and system based on visual saliency | |
| US8929656B2 (en) | Method and apparatus for detecting important information from moving picture | |
| CN101459806A (en) | System and method for video playing | |
| US11330329B2 (en) | System and method for detecting and classifying direct response advertisements using fingerprints | |
| CN101692269B (en) | Method and device for processing video programs | |
| CN103605991A (en) | Automatic video advertisement detection method | |
| CN108540833A (en) | A kind of television advertising recognition methods based on camera lens | |
| CN112567416A (en) | Apparatus and method for processing digital video | |
| CN108509879B (en) | Method and device for realizing information processing | |
| CN103020094A (en) | Method for counting video playing times | |
| CN110414471B (en) | Video identification method and system based on double models | |
| Chattopadhyay et al. | Mash up of breaking news and contextual web information: a novel service for connected television | |
| CN113191811B (en) | Intelligent advertisement pushing method and device and computer readable storage medium | |
| US20170180501A1 (en) | Message pushing method and message pushing device | |
| Goëau et al. | Interactive visualization tool with graphic table of video contents | |
| KR100768074B1 (en) | System for offering advertisement moving picture and service method thereof | |
| CN113569668A (en) | Method, medium, apparatus and computing device for determining highlight segments in video | |
| Wang et al. | Community discovery from movie and its application to poster generation | |
| TW201405462A (en) | Digital signage push system based on crowd correlation determination | |
| Chen et al. | An effective method for video genre classification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20201027 |
|
| WW01 | Invention patent application withdrawn after publication |