CN113743281A - Program advertisement material identification method, system, computer device and storage medium - Google Patents
Program advertisement material identification method, system, computer device and storage medium Download PDFInfo
- Publication number
- CN113743281A CN113743281A CN202111004373.2A CN202111004373A CN113743281A CN 113743281 A CN113743281 A CN 113743281A CN 202111004373 A CN202111004373 A CN 202111004373A CN 113743281 A CN113743281 A CN 113743281A
- Authority
- CN
- China
- Prior art keywords
- program
- advertisement
- data set
- region
- target detection
- 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.)
- Pending
Links
- 239000000463 material Substances 0.000 title claims abstract description 334
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000001514 detection method Methods 0.000 claims abstract description 98
- 238000012549 training Methods 0.000 claims abstract description 31
- 230000009466 transformation Effects 0.000 claims abstract description 27
- 238000012216 screening Methods 0.000 claims abstract description 17
- 238000004590 computer program Methods 0.000 claims description 11
- 238000002513 implantation Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 abstract description 10
- 238000002372 labelling Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 4
- 238000007619 statistical method Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 239000002131 composite material Substances 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- 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
- G06Q30/0276—Advertisement creation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
- H04N21/23418—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/81—Monomedia components thereof
- H04N21/812—Monomedia components thereof involving advertisement data
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a program advertisement material identification method, a system, a computer device and a storage medium, wherein the program advertisement material identification method comprises the following steps: a material screening step, namely acquiring video data, extracting a program picture key frame containing an advertisement material in the video data and screening the program picture key frame based on the advertisement material; a data set acquisition step, namely, keying advertisement materials in the key frames of the program pictures as material template pictures, and carrying out random image transformation on the material template pictures to obtain a data set; and a target identification step, namely training at least one target detection model based on the data set, and detecting the region of the advertisement material in the video data in the program picture and the region area thereof based on the target detection model. Through the method and the device, the material marking labor cost is reduced, and the data processing efficiency is improved.
Description
Technical Field
The present application relates to the field of internet technologies, and in particular, to a method, a system, a computer device, and a computer-readable storage medium for identifying program advertisement material.
Background
In many programs of the integrated art, the scene of the theme event has posters, props and related products of the relevant sponsors, and the titles, transition advertisements, corner marks and spoken prompts are inserted into the recorded programs. The advertiser can evaluate the area and position of the product such as the product and the prop of the related sponsor in the program by detecting the area and position of the product, for example, from the center to the edge, the program picture is divided into a golden area, a middle area and an edge area, as shown in fig. 1, the product and the prop are exposed in the golden area of the program, which is beneficial to attracting the attention of the consumer. Since the consumer's gaze is primarily focused on guests being spoken, it may be advisable to enhance the interaction of guests with props, products, etc. The detection and evaluation of these implanted products, props, etc. therefore, is advantageous in that the consumer is given a more impressive impression in the subsequent display in a novel form and creative manner.
The image recognition technology is rapidly developed in recent years, and the appearance of the deep learning technology enables the image recognition technology to take a great step on the basis of the traditional method. Through the obtained video or picture of the comprehensive art program, props, oral broadcast prompt bars, products and the like in the program can be positioned by utilizing a target detection technology. The positioned object and the pre-intercepted template can be matched and aligned by utilizing the scale invariant feature transform, namely, the SIFT algorithm to obtain the position and the category of the corresponding object, and corresponding evaluation is carried out.
The related art is labeled with advertiser-provided material, as shown in fig. 2, followed by model training. Under the condition, the cost for marking the materials is high, if new materials appear, the materials need to be marked again, and the working efficiency is low. Moreover, the new materials need to be detected by full-time personnel, so that the effectiveness is poor, and massive materials cannot be processed in time.
Disclosure of Invention
The embodiment of the application provides a program advertisement material identification method, a program advertisement material identification system, computer equipment and a computer readable storage medium, so that the material marking labor cost is reduced, and the data processing efficiency is improved.
In a first aspect, an embodiment of the present application provides a method for identifying program advertisement material, including:
a material screening step, namely acquiring video data, extracting a program picture key frame containing an advertisement material in the video data and screening the program picture key frame based on the advertisement material; specifically, the advertisement material may be a product entity, a prop, an oral play cue, or the like.
A data set acquisition step, namely, keying advertisement materials in the key frames of the program pictures as material template pictures, and carrying out random image transformation on the material template pictures to obtain a data set;
and a target identification step, namely training at least one target detection model based on the data set, and detecting the region of the advertisement material in the video data in the program picture and the region area thereof based on the target detection model.
Through the steps, on the premise of less data samples, the embodiment of the application utilizes the limited samples to rapidly construct a rich data set for model training, so that the regions and the region areas of advertising materials such as sponsor products, props, and oral broadcast prompt bars implanted in programs can be rapidly detected, the manual labeling cost is greatly reduced, errors and omissions in manual labeling detection are effectively solved, and the data processing efficiency and accuracy are improved.
In some of these embodiments, the dataset acquisition step further comprises:
a material template drawing obtaining step, namely obtaining the program picture key frame and extracting advertisement materials in the program picture key frame to obtain a material template drawing;
a material class counting step, counting material classes in the material template drawings, and if the material classes are larger than a set threshold value, dividing the material template drawings based on the material classes to obtain a plurality of subclass material template drawings;
and a data set construction step, wherein random image transformation is carried out on the material template drawing and/or the plurality of subclass material template drawings to obtain a material data set and/or a plurality of subclass material data sets. Specifically, the random image transformation includes: and (4) random image transformation of the size, position and rotation angle of the material.
Through the steps, rich data sets are constructed based on image transformation of the materials, meanwhile, in order to improve the class identification accuracy of the target detection model, the material classes are divided, and then a plurality of subclass material data sets are constructed, so that a data base is provided for training a more accurate target detection model.
In some of these embodiments, the target identifying step further comprises:
a target detection model training step, wherein the target detection model is trained on the basis of the material data set, and/or a plurality of target detection models are respectively trained on the basis of a plurality of subclass material data sets;
and detecting the advertisement material, namely detecting the advertisement material in the video data based on the target detection model and acquiring the region of the advertisement material and the region area of the advertisement material.
In some of these embodiments, the method further comprises:
and a detection result acquisition step, namely, pre-configuring a scoring standard based on the region and the region area of the advertisement material, calculating the comprehensive score of the advertisement material in the video data based on the scoring standard, and generating a detection result file based on the comprehensive score to realize the implantation of the advertisement material in the evaluation program. Specifically, the scoring criteria of the embodiment of the present application configures scores according to the region to which the advertisement material belongs and the region area, where the region to which the advertisement material belongs includes a golden region, a middle region, and an edge region, and the region to which the scoring corresponds is configured as a golden region score > a middle region score > an edge region score, and the region area to which the scoring corresponds is configured to be proportional to the area ratio of the region area in the region to which the region area belongs, and the area ratio of the advertisement material in the region to which the advertisement material belongs is obtained through region area calculation.
Through the steps, the comprehensive score of the advertisement material is calculated through statistical analysis and detection data, so that whether the advertisement material is normally implanted in the program is judged for a program sponsor, and data support is provided for adjustment of subsequent material exhibition.
In a second aspect, an embodiment of the present application provides a program advertisement material identification system, including:
the material screening module is used for acquiring video data, extracting program picture key frames containing advertisement materials in the video data and screening the program picture key frames based on the advertisement materials; specifically, the advertisement material may be a product entity, a prop, an oral play cue, or the like.
The data set acquisition module is used for matting advertisement materials in the program picture key frames as material template pictures and carrying out random image transformation on the material template pictures to obtain a data set;
and the target identification module trains at least one target detection model based on the data set and detects the region of the advertisement material in the video data in the program picture and the region area thereof based on the target detection model.
Through the modules, on the premise of less data samples, the embodiment of the application utilizes the limited samples to rapidly construct a rich data set for model training, so that the regions and the region areas of advertising materials such as sponsor products, props, and oral broadcast prompt bars implanted in programs can be rapidly detected, the manual labeling cost is greatly reduced, errors and omissions in manual labeling detection are effectively solved, and the data processing efficiency and accuracy are improved.
In some of these embodiments, the dataset acquisition module further comprises:
the material template picture acquisition module is used for acquiring the program picture key frame and extracting advertisement materials in the program picture key frame to obtain a material template picture;
the material class counting module is used for counting the material classes in the material template pictures, and if the material classes are larger than a set threshold value, the material template pictures are divided based on the material classes to obtain a plurality of subclass material template pictures;
and the data set construction module is used for carrying out random image transformation on the material template graph and/or the plurality of subclass material template graphs to obtain a material data set and/or a plurality of subclass material data sets. Specifically, the random image transformation includes: and (4) random image transformation of the size, position and rotation angle of the material.
Through the modules, rich data sets are constructed based on image transformation of the materials, meanwhile, in order to improve the class identification accuracy of the target detection model, a plurality of subclass material data sets are constructed after material classes are divided, and a data basis is provided for training a more accurate target detection model.
In some of these embodiments, the object recognition module further comprises:
the target detection model training module is used for training the target detection model based on the material data set and/or respectively training a plurality of target detection models based on a plurality of subclass material data sets;
and the advertisement material detection module is used for detecting the advertisement material in the video data based on the target detection model and acquiring the region of the advertisement material and the region area of the advertisement material.
In some of these embodiments, the system further comprises:
and the detection result acquisition module is used for pre-configuring a scoring standard based on the region and the region area of the advertisement material, calculating the comprehensive score of the advertisement material in the video data based on the scoring standard, and generating a detection result file based on the comprehensive score to realize the implantation of the advertisement material in the evaluation program. Specifically, the scoring criteria of the embodiment of the present application configures scores according to the region to which the advertisement material belongs and the region area, where the region to which the advertisement material belongs includes a golden region, a middle region, and an edge region, and the region to which the scoring corresponds is configured as a golden region score > a middle region score > an edge region score, and the region area to which the scoring corresponds is configured to be proportional to the area ratio of the region area in the region to which the region area belongs, and the area ratio of the advertisement material in the region to which the advertisement material belongs is obtained through region area calculation.
Through the modules, the comprehensive score of the advertisement material is calculated through statistical analysis and detection data, so that whether the advertisement material is normally implanted in the program is judged for a program sponsor, and data support is provided for adjustment of subsequent material exhibition.
In a third aspect, an embodiment of the present application provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to implement the program advertisement material identification method according to the first aspect.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements the program advertising material identification method as described in the first aspect above.
Compared with the related art, the program advertisement material identification method, the system, the computer equipment and the computer readable storage medium provided by the embodiment of the application particularly relate to a deep learning technology, specifically, a data set is constructed through material template map extraction and transformation operation, a target detection model is trained by using the data set, the advertisement material implanted in a program is rapidly detected, errors and omissions caused by manual labeling detection are effectively solved, and the data processing efficiency and accuracy are improved.
The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below to provide a more thorough understanding of the application.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a schematic diagram of program picture region division according to the related art;
fig. 2 is a schematic view of a material marking effect according to the related art;
FIG. 3 is a flow chart of a method of identifying program advertising material according to an embodiment of the present application;
FIG. 4 is a flow chart of substeps of a method for identifying program advertising material according to an embodiment of the present application;
FIG. 5 is a flow chart of substeps of a method for identifying program advertising material according to an embodiment of the present application;
fig. 6 is a flow chart of a method of identifying program advertising material in accordance with a preferred embodiment of the present application;
FIG. 7 is a key frame diagram of a program picture according to a preferred embodiment of the present application;
FIG. 8 is a schematic diagram of a material template according to a preferred embodiment of the present application;
FIG. 9(a-d) is a schematic diagram of the image transformation effect of the material template map according to the preferred embodiment of the present application;
fig. 10 is a block diagram of the structure of a program advertising material identification system according to an embodiment of the present application.
Wherein:
1. a material screening module; 2. a dataset acquisition module; 3. a target identification module;
4. a detection result acquisition module; 201. a material template picture acquisition module;
202. a material class counting module; 203. a data set construction module;
301. a target detection model training module; 302. advertisement material detection module.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments provided in the present application without any inventive step are within the scope of protection of the present application.
It is obvious that the drawings in the following description are only examples or embodiments of the present application, and that it is also possible for a person skilled in the art to apply the present application to other similar contexts on the basis of these drawings without inventive effort. Moreover, it should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another.
Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the specification. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of ordinary skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments without conflict.
Unless defined otherwise, technical or scientific terms referred to herein shall have the ordinary meaning as understood by those of ordinary skill in the art to which this application belongs. Reference to "a," "an," "the," and similar words throughout this application are not to be construed as limiting in number, and may refer to the singular or the plural. The present application is directed to the use of the terms "including," "comprising," "having," and any variations thereof, which are intended to cover non-exclusive inclusions; for example, a process, method, system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to the listed steps or elements, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. Reference to "connected," "coupled," and the like in this application is not intended to be limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The term "plurality" as referred to herein means two or more. "and/or" describes an association relationship of associated objects, meaning that three relationships may exist, for example, "A and/or B" may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. Reference herein to the terms "first," "second," "third," and the like, are merely to distinguish similar objects and do not denote a particular ordering for the objects.
In order to solve the problems that the manual labeling cost of the program advertisement material is high and misjudgment and misleakage are prone to occur, the embodiment provides a program advertisement material identification method. Fig. 3-5 are flowcharts of a method of identifying program advertising material according to an embodiment of the present application, the flowcharts including the steps of, as shown in fig. 3-5:
a material screening step S1 of acquiring video data, extracting program picture key frames containing advertisement materials in the video data, and screening the program picture key frames based on the advertisement materials; specifically, the advertisement material may be a product entity, a prop, an oral play cue, or the like.
A data set acquisition step S2, wherein, the advertisement material in the key frame of the program picture is extracted as a material template picture, and the random image transformation is carried out on the material template picture to obtain a data set;
and a target identification step S3, training at least one target detection model based on the data set, and detecting the region of the advertisement material in the video data in the program picture and the region area thereof based on the target detection model.
Through the steps, on the premise of less data samples, the embodiment of the application utilizes the limited samples to rapidly construct a rich data set for model training, so that the regions and the region areas of advertising materials such as sponsor products, props, and oral broadcast prompt bars implanted in programs can be rapidly detected, the manual labeling cost is greatly reduced, errors and omissions in manual labeling detection are effectively solved, and the data processing efficiency and accuracy are improved.
As shown in fig. 4, the data set obtaining step S2 further includes:
a material template drawing obtaining step S201, obtaining a program picture key frame and extracting advertisement materials in the program picture key frame to obtain a material template drawing; optionally, the matting of the advertisement material in this embodiment may be performed based on the outline of the advertisement material, or may be performed based on a rectangular frame.
A material class counting step S202, counting material classes in the material template drawings, and if the material classes are larger than a set threshold, dividing the material template drawings based on the material classes to obtain a plurality of subclass material template drawings;
and a data set construction step S203, performing random image transformation on the material template drawing and/or the plurality of subclass material template drawings to obtain a material data set and/or a plurality of subclass material data sets. Specifically, the random image transformation includes: the random image transformation of the material size, position, and rotation angle is shown in fig. 9.
Through the steps, rich data sets are constructed based on image transformation of the materials, meanwhile, in order to improve the class identification accuracy of the target detection model, the material classes are divided, and then a plurality of subclass material data sets are constructed, so that a data base is provided for training a more accurate target detection model.
As shown in fig. 5, the target recognizing step S3 further includes:
a target detection model training step S301, training a target detection model based on a material data set, and/or respectively training a plurality of target detection models based on a plurality of subclass material data sets; optionally, the target detection model may be implemented based on a fastern target detection algorithm, or may be implemented based on other target detection networks, which is not described herein again.
And an advertisement material detection step S302, which is to detect the advertisement material in the video data based on the target detection model and acquire the region of the advertisement material and the region area thereof. According to the method and the device, the category of the advertisement material is obtained by using the target detection model, and the area of the area to which the advertisement material belongs are obtained by calculating based on the coordinate information and the size of the detection frame.
In addition, in this embodiment, in consideration that in practical application, a program sponsor may determine whether an advertisement material in a program is normally embedded according to a position and an area of the material, and in order to facilitate quantitative embedding of the advertisement material by the program sponsor, the method for identifying a program advertisement material in an embodiment of the present application further includes:
a detection result obtaining step S4, pre-configuring a scoring criterion based on the region and region area to which the advertisement material belongs, calculating a comprehensive score of the advertisement material in the video data based on the scoring criterion, and generating a detection result file based on the comprehensive score to realize the implantation of the advertisement material in the evaluation program. Specifically, the scoring criteria of the embodiment of the present application configures scores according to the area and the area of the advertisement material, and as shown in fig. 1, the area includes a golden area, a middle area, and an edge area, and the area is configured to be the golden area score > the middle area score > the edge area score, considering that the area occupied ratio of the advertisement material is small in the golden area during the playing of the program, the scoring criteria based on the area of the area is added to the scoring criteria. The area corresponding score is configured to be in direct proportion to the area proportion of the area in the region to which the area corresponds, and the area proportion of the advertisement material in the region to which the area corresponds is obtained through the area calculation of the region.
It should be noted that the above scoring criteria and classification of the detection results in this embodiment are not intended to limit the scoring criteria in this embodiment, and modifications and improvements made to the scoring criteria based on the technical solution of the present application are all within the scope of the present application.
Through the steps, the comprehensive score of the advertisement material is calculated through statistical analysis and detection data, so that whether the advertisement material is normally implanted in the program is judged for a program sponsor, and data support is provided for adjustment of subsequent material exhibition.
The embodiments of the present application are described and illustrated below by means of preferred embodiments.
Fig. 6 is a flowchart of a program advertising material identification method according to a preferred embodiment of the present application, which includes the steps of, as shown in fig. 6:
step S601, extracting program picture key frames containing product advertisement materials in the video according to the collected video data, and screening out the key frames based on the products, props, and the oral broadcasting prompt bars and the like required in the advertisement materials;
step S602, accurately extracting products, props and prompt bars in the pictures according to the screened advertisement materials, such as shown in figures 7-8, wherein figure 7 is an original picture, and figure 8 is a template picture of the extracted materials; optionally, the manner of this embodiment to extract the material may have multiple methods: (1) and (2) carrying out rectangular frame type digging according to the outline of the material.
And step S603, counting the variety types of the material template pictures, dividing the material template pictures if the variety types of the material template pictures are more than 8, otherwise, performing no treatment, and entering the next step.
Step S604, constructing a data set by using the extracted material template pictures, and constructing a rich data set by randomly transforming the size, position and rotation angle of the product in each picture, as shown in fig. 9(a-d) below.
In step S605, a target detection model is trained using the constructed data set. It is worth noting that if the material variety number is larger than a certain number and needs to be divided, the divided data is subjected to independent training of constructing a data set and a model, and the detection model of the step only detects the foreground and the background. The model algorithm is a commonly used target detection algorithm, such as the FasterCNN target detection algorithm. The task of the object detection algorithm is not only to obtain the class of the object, but also where the object is located.
And step S606, after the training of the detection models is finished, the detection and the matching of the advertisement materials are realized by utilizing the combination of the models, and the areas of the detected areas and the detected areas of the advertisement materials are calculated.
And step S607, generating a result file according to the result obtained by calculation in the step 606, generating a report, and reminding relevant personnel to check. In the result file, a score is assigned to the advertisement material based on the belonging area and the area of the area, and the belonging area corresponding score is assigned by:
if the advertisement material belongs to the program picture golden area, the configuration score is 10 points;
if the advertisement material belongs to the middle area of the program picture, the configuration score is 8;
if the advertisement material belongs to the program picture edge area, the configuration score is 4;
and calculating the area ratio of the advertisement material in the region to which the advertisement material belongs through the region area, wherein the corresponding score of the area ratio is configured as:
if the area percentage P of the advertisement material in the area is more than or equal to 80 percent, the distribution is 10 points;
if the area proportion P of the advertisement material in the region is more than 80% and more than or equal to 50%, configuring the advertisement material into 8 points;
if the area proportion P of the advertisement material in the region is less than 50%, configuring the area proportion P into 4 points;
based on the scoring standard, four types of detection results of excellence, good, qualified and unqualified are set for the comprehensive Score, specifically:
if the composite Score is 20, the placement of the advertising material is evaluated as excellent;
if the comprehensive Score is more than 20 and more than Score and more than or equal to 16, the implantation evaluation of the advertisement material is good;
if the comprehensive Score is more than 16 and more than or equal to 14, the implantation evaluation of the advertisement material is qualified;
if the composite Score is < 14, then the placement of the advertising material is assessed as disqualified.
It should be noted that the steps illustrated in the above-described flow diagrams or in the flow diagrams of the figures may be performed in a computer system, such as a set of computer-executable instructions, and that, although a logical order is illustrated in the flow diagrams, in some cases, the steps illustrated or described may be performed in an order different than here.
The embodiment further provides a program advertisement material identification system, which includes a material screening module 1, a data set obtaining module 2, a target identifying module 3, a detection result obtaining module 4, and other modules, and is used to implement the foregoing embodiments and preferred embodiments, which have been described and will not be described again. As used hereinafter, the terms "module," "unit," "subunit," and the like may implement a combination of software and/or hardware for a predetermined function. While the system described in the embodiments below is preferably implemented in software, implementations in hardware, or a combination of software and hardware are also possible and contemplated.
Fig. 10 is a block diagram of the structure of a program advertising material identification system according to an embodiment of the present application, as shown in fig. 10, the system including:
the material screening module 1 is used for acquiring video data, extracting program picture key frames containing advertisement materials in the video data and screening the program picture key frames based on the advertisement materials; specifically, the advertisement material may be a product entity, a prop, an oral play cue, or the like.
The data set acquisition module 2 is used for matting advertisement materials in the key frames of the program pictures as material template pictures and carrying out random image transformation on the material template pictures to obtain a data set; specifically, the data set obtaining module 2 further includes: a material template drawing obtaining module 201, which obtains the program picture key frame and extracts the advertisement material in the program picture key frame to obtain a material template drawing; optionally, the matting of the advertisement material in this embodiment may be performed based on the outline of the advertisement material, or may be performed based on a rectangular frame. The material class counting module 202 counts material classes in the material template drawings, and if the material classes are larger than a set threshold, the material template drawings are divided based on the material classes to obtain a plurality of subclass material template drawings; and the data set construction module 203 performs random image transformation on the material template drawing and/or the plurality of subclass material template drawings to obtain a material data set and/or a plurality of subclass material data sets. Specifically, the random image transformation includes: the random image transformation of the material size, position, and rotation angle is shown in fig. 9. Through the modules, rich data sets are constructed based on image transformation of the materials, meanwhile, in order to improve the class identification accuracy of the target detection model, a plurality of subclass material data sets are constructed after material classes are divided, and a data basis is provided for training a more accurate target detection model.
And the target identification module 3 is used for training at least one target detection model based on the data set and detecting the region of the advertisement material in the video data in the program picture and the region area thereof based on the target detection model. Specifically, the object recognition module 3 further includes: a target detection model training module 301 for training a target detection model based on the material data set and/or respectively training a plurality of target detection models based on a plurality of subclass material data sets; optionally, the target detection model may be implemented based on a fastern target detection algorithm, or may be implemented based on other target detection networks, which is not described herein again. The advertisement material detection module 302 detects an advertisement material in the video data based on the target detection model and obtains the region to which the advertisement material belongs and the region area thereof. According to the method and the device, the category of the advertisement material is obtained by using the target detection model, and the area of the area to which the advertisement material belongs are obtained by calculating based on the coordinate information and the size of the detection frame.
Through the modules, on the premise of less data samples, the embodiment of the application utilizes the limited samples to rapidly construct a rich data set for model training, so that the regions and the region areas of advertising materials such as sponsor products, props, and oral broadcast prompt bars implanted in programs can be rapidly detected, the manual labeling cost is greatly reduced, errors and omissions in manual labeling detection are effectively solved, and the data processing efficiency and accuracy are improved.
Considering that in practical application, a program sponsor can judge whether an advertisement material is normally embedded in a program according to the position and the area of the material, and in order to facilitate quantitative embedding of the advertisement material by the program sponsor, the program advertisement material identification system of the embodiment of the present application further includes:
the detection result acquisition module 4 pre-configures a scoring standard based on the region and the region area to which the advertisement material belongs, calculates the comprehensive score of the advertisement material in the video data based on the scoring standard, generates a detection result file based on the comprehensive score, and realizes the implantation of the advertisement material in the evaluation program. Specifically, the scoring criteria of the embodiment of the present application configures scores according to the area to which the advertisement material belongs and the area of the area, and as shown in fig. 1, the area to which the advertisement material belongs includes a golden area, a middle area, and an edge area, and the area to which the advertisement material belongs is configured to be the golden area score > the middle area score > the edge area score. The area corresponding score is configured to be in direct proportion to the area proportion of the area in the region to which the area corresponds, and the area proportion of the advertisement material in the region to which the area corresponds is obtained through the area calculation of the region.
It should be noted that the above scoring criteria and classification of the detection results in this embodiment are not intended to limit the scoring criteria in this embodiment, and modifications and improvements made to the scoring criteria based on the technical solution of the present application are all within the scope of the present application.
Through the modules, the comprehensive score of the advertisement material is calculated through statistical analysis and detection data, so that whether the advertisement material is normally implanted in the program is judged for a program sponsor, and data support is provided for adjustment of subsequent material exhibition.
The above modules may be functional modules or program modules, and may be implemented by software or hardware. For a module implemented by hardware, the modules may be located in the same processor; or the modules can be respectively positioned in different processors in any combination.
In addition, the program advertising material identification methods of the embodiments of the present application described in connection with fig. 3-5 may be implemented by a computer device. The computer device may include a processor and a memory storing computer program instructions. In particular, the processor may include a Central Processing Unit (CPU), or A Specific Integrated Circuit (ASIC), or may be configured to implement one or more Integrated circuits of the embodiments of the present Application.
The memory may include, among other things, mass storage for data or instructions. By way of example, and not limitation, memory may include a Hard Disk Drive (Hard Disk Drive, abbreviated to HDD), a floppy Disk Drive, a Solid State Drive (SSD), flash memory, an optical Disk, a magneto-optical Disk, tape, or a Universal Serial Bus (USB) Drive or a combination of two or more of these. The memory may include removable or non-removable (or fixed) media, where appropriate. The memory may be internal or external to the data processing apparatus, where appropriate. In a particular embodiment, the memory is a Non-Volatile (Non-Volatile) memory. In particular embodiments, the Memory includes Read-Only Memory (ROM) and Random Access Memory (RAM). The ROM may be mask-programmed ROM, Programmable ROM (PROM), Erasable PROM (EPROM), Electrically Erasable PROM (EEPROM), Electrically rewritable ROM (EAROM), or FLASH Memory (FLASH), or a combination of two or more of these, where appropriate. The RAM may be a Static Random-Access Memory (SRAM) or a Dynamic Random-Access Memory (DRAM), where the DRAM may be a Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), an Extended data output Dynamic Random-Access Memory (EDODRAM), a Synchronous Dynamic Random-Access Memory (SDRAM), and the like.
The memory may be used to store or cache various data files for processing and/or communication use, as well as possibly computer program instructions for execution by the processor.
The processor implements any of the program advertising material identification methods in the above embodiments by reading and executing computer program instructions stored in the memory.
In addition, in combination with the program advertisement material identification method in the above embodiments, the embodiments of the present application may be implemented by providing a computer-readable storage medium. The computer readable storage medium having stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement any of the program advertising material identification methods of the above embodiments.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A method for identifying program advertising material, comprising:
a material screening step, namely acquiring video data, extracting a program picture key frame containing an advertisement material in the video data and screening the program picture key frame based on the advertisement material;
a data set acquisition step, namely, keying advertisement materials in the key frames of the program pictures as material template pictures, and carrying out random image transformation on the material template pictures to obtain a data set;
and a target identification step, namely training at least one target detection model based on the data set, and detecting the region of the advertisement material in the video data in the program picture and the region area thereof based on the target detection model.
2. The method of identifying programming advertising material as recited in claim 1, wherein said data set obtaining step further comprises:
a material template drawing obtaining step, namely obtaining the program picture key frame and extracting advertisement materials in the program picture key frame to obtain a material template drawing;
a material class counting step, counting material classes in the material template drawings, and if the material classes are larger than a set threshold value, dividing the material template drawings based on the material classes to obtain a plurality of subclass material template drawings;
and a data set construction step, wherein random image transformation is carried out on the material template drawing and/or the plurality of subclass material template drawings to obtain a material data set and/or a plurality of subclass material data sets.
3. The method of identifying programming advertising material as recited in claim 2, wherein said target identifying step further comprises:
a target detection model training step, wherein the target detection model is trained on the basis of the material data set, and/or a plurality of target detection models are respectively trained on the basis of a plurality of subclass material data sets;
and detecting the advertisement material, namely detecting the advertisement material in the video data based on the target detection model and acquiring the region of the advertisement material and the region area of the advertisement material.
4. A method for identifying program advertising material as recited in any one of claims 1-3, further comprising:
and a detection result acquisition step, namely, pre-configuring a scoring standard based on the region and the region area of the advertisement material, calculating the comprehensive score of the advertisement material in the video data based on the scoring standard, and generating a detection result file based on the comprehensive score to realize the implantation of the advertisement material in the evaluation program.
5. A program advertising material identification system, comprising:
the material screening module is used for acquiring video data, extracting program picture key frames containing advertisement materials in the video data and screening the program picture key frames based on the advertisement materials;
the data set acquisition module is used for matting advertisement materials in the program picture key frames as material template pictures and carrying out random image transformation on the material template pictures to obtain a data set;
and the target identification module trains at least one target detection model based on the data set and detects the region of the advertisement material in the video data in the program picture and the region area thereof based on the target detection model.
6. The program advertising material identification system of claim 5, wherein the data set acquisition module further comprises:
the material template picture acquisition module is used for acquiring the program picture key frame and extracting advertisement materials in the program picture key frame to obtain a material template picture;
the material class counting module is used for counting the material classes in the material template pictures, and if the material classes are larger than a set threshold value, the material template pictures are divided based on the material classes to obtain a plurality of subclass material template pictures;
and the data set construction module is used for carrying out random image transformation on the material template graph and/or the plurality of subclass material template graphs to obtain a material data set and/or a plurality of subclass material data sets.
7. The program advertising material identification system of claim 6, wherein the target identification module further comprises:
the target detection model training module is used for training the target detection model based on the material data set and/or respectively training a plurality of target detection models based on a plurality of subclass material data sets;
and the advertisement material detection module is used for detecting the advertisement material in the video data based on the target detection model and acquiring the region of the advertisement material and the region area of the advertisement material.
8. The program advertising material identification system of any one of claims 5-7, further comprising:
and the detection result acquisition module is used for pre-configuring a scoring standard based on the region and the region area of the advertisement material, calculating the comprehensive score of the advertisement material in the video data based on the scoring standard, and generating a detection result file based on the comprehensive score to realize the implantation of the advertisement material in the evaluation program.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program implements a program advertising material identification method as claimed in any one of claims 1 to 4.
10. A computer-readable storage medium on which a computer program is stored, which program, when executed by a processor, carries out the program advertisement material identification method of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111004373.2A CN113743281A (en) | 2021-08-30 | 2021-08-30 | Program advertisement material identification method, system, computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111004373.2A CN113743281A (en) | 2021-08-30 | 2021-08-30 | Program advertisement material identification method, system, computer device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113743281A true CN113743281A (en) | 2021-12-03 |
Family
ID=78733783
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111004373.2A Pending CN113743281A (en) | 2021-08-30 | 2021-08-30 | Program advertisement material identification method, system, computer device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113743281A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106982380A (en) * | 2017-04-20 | 2017-07-25 | 上海极链网络科技有限公司 | The method for implantation of virtual interactive advertisement in internet video |
| WO2019062841A1 (en) * | 2017-09-30 | 2019-04-04 | 北京国双科技有限公司 | Brand exposure effect analysis method and device |
| CN109598527A (en) * | 2017-09-30 | 2019-04-09 | 北京国双科技有限公司 | Analysis of advertising results method and device |
| CN110148102A (en) * | 2018-02-12 | 2019-08-20 | 腾讯科技（深圳）有限公司 | Image composition method, ad material synthetic method and device |
| CN111144937A (en) * | 2019-12-20 | 2020-05-12 | 北京达佳互联信息技术有限公司 | Advertisement material determination method, device, equipment and storage medium |
| CN111401228A (en) * | 2020-03-13 | 2020-07-10 | 中科创达软件股份有限公司 | Video target labeling method and device and electronic equipment |
| CN112287800A (en) * | 2020-10-23 | 2021-01-29 | 北京中科模识科技有限公司 | Advertisement video identification method and system under no-sample condition |
| CN112598448A (en) * | 2020-12-29 | 2021-04-02 | 恩亿科(北京)数据科技有限公司 | Dynamic material data processing method, system, computer and readable storage medium |
-
2021
- 2021-08-30 CN CN202111004373.2A patent/CN113743281A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106982380A (en) * | 2017-04-20 | 2017-07-25 | 上海极链网络科技有限公司 | The method for implantation of virtual interactive advertisement in internet video |
| WO2019062841A1 (en) * | 2017-09-30 | 2019-04-04 | 北京国双科技有限公司 | Brand exposure effect analysis method and device |
| CN109598527A (en) * | 2017-09-30 | 2019-04-09 | 北京国双科技有限公司 | Analysis of advertising results method and device |
| CN110148102A (en) * | 2018-02-12 | 2019-08-20 | 腾讯科技（深圳）有限公司 | Image composition method, ad material synthetic method and device |
| CN111144937A (en) * | 2019-12-20 | 2020-05-12 | 北京达佳互联信息技术有限公司 | Advertisement material determination method, device, equipment and storage medium |
| CN111401228A (en) * | 2020-03-13 | 2020-07-10 | 中科创达软件股份有限公司 | Video target labeling method and device and electronic equipment |
| CN112287800A (en) * | 2020-10-23 | 2021-01-29 | 北京中科模识科技有限公司 | Advertisement video identification method and system under no-sample condition |
| CN112598448A (en) * | 2020-12-29 | 2021-04-02 | 恩亿科(北京)数据科技有限公司 | Dynamic material data processing method, system, computer and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109858371B (en) | Face recognition method and device | |
| Bruni et al. | Distributional semantics in technicolor | |
| US20180081978A1 (en) | Method and Apparatus for Processing Information | |
| CN111191067A (en) | Picture book identification method, terminal device and computer readable storage medium | |
| CN110427859A (en) | A kind of method for detecting human face, device, electronic equipment and storage medium | |
| CN105159913B (en) | Method and device for determining file to be cleaned | |
| CN111401238B (en) | Method and device for detecting character close-up fragments in video | |
| CN105869008A (en) | Targeted delivery method and device of advertisement | |
| CN108376164B (en) | Display method and device of potential anchor | |
| CN106961559B (en) | A kind of production method and electronic equipment of video | |
| Theisen et al. | Automatic discovery of political meme genres with diverse appearances | |
| CN105303449B (en) | The recognition methods and system of social network user based on camera fingerprint characteristic | |
| Dantone et al. | Augmented faces | |
| Bozkir et al. | Layout-based computation of web page similarity ranks | |
| CN111488847B (en) | Sports game video ball-feeding segment acquisition system, method and terminal | |
| Lumini et al. | Detector of image orientation based on Borda Count | |
| CN107729528A (en) | A kind of building information acquisition methods and device, computer installation and computer-readable recording medium | |
| CN111738042A (en) | Identification method, device and storage medium | |
| US20110044530A1 (en) | Image classification using range information | |
| CN109816004A (en) | Source of houses picture classification method, device, equipment and storage medium | |
| CN113743281A (en) | Program advertisement material identification method, system, computer device and storage medium | |
| WO2024087358A1 (en) | Target detection method and apparatus, and device and non-volatile readable storage medium | |
| CN115474093A (en) | Method and device for calculating importance of video elements, storage medium and electronic equipment | |
| TW201816659A (en) | Method and apparatus for identifying bar code | |
| CN104021220A (en) | Image retrieval method based on significant model |
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 |