A method, device, equipment and storage medium for selecting high-quality advertisements

Abstract

The embodiment of the present application discloses a method, device, equipment and storage medium for selecting high-quality advertisements, belonging to the field of advertisement application technology, the method comprising: grabbing advertisement videos through advertisement links; performing carousel tests on advertisement videos; performing image acquisition on viewing areas; performing face statistics on the acquired pictures; obtaining the number of people who watch the advertisement videos in the acquired pictures; respectively obtaining the proportional relationship between the number of people who watch the advertisement videos in each acquired picture and the face statistics result; judging the advertisement video corresponding to each acquired picture; respectively obtaining the popularity of different advertisement videos when they are played, and selecting the advertisement video corresponding to the highest popularity as the high-quality advertisement. The present application selects advertisements based on the popularity among users, effectively avoiding the blind delivery of advertisements and saving advertising costs to a certain extent.

Description

Technical Field

[0001] The present application relates to the field of advertising application technology, and in particular to a method, device, equipment and storage medium for selecting high-quality advertisements. Background Art

[0002] Advertising is an important traffic-generating function in most applications and products. This function will attract more users to choose products. At present, there are two main ways to place advertisements: offline and online. Offline advertising mainly refers to graffiti and advertising papers on houses and streets, while online advertising mainly refers to attracting traffic for products through advertising videos on mobile phones or web pages.

[0003] At present, online advertising is mainly delivered by pushing advertisements on the user's mobile phone or web page. However, this delivery method requires more advertising costs. Moreover, the quality of advertisements varies. Advertisements are often delivered blindly, resulting in a large number of "junk" advertisements that occupy most of the advertising area. Therefore, the existing technology has the problem of blind delivery and high costs when delivering advertising videos. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method, device, equipment and storage medium for selecting high-quality advertisements to solve the problem of blind delivery and high cost when delivering advertising videos in the prior art.

[0005] In order to solve the above technical problems, the present application embodiment provides a method for selecting high-quality advertisements, which adopts the following technical solution:

[0006] A method for selecting high-quality advertisements, comprising:

[0007] Obtain N advertisement links of the same product to be launched, capture advertisement videos through the advertisement links, and number the advertisement videos from 1 to N, where N is a positive integer greater than 1;

[0008] Push the advertising videos to a preset play area in the order of numbers, perform a carousel test, and obtain the current time as the node time when each advertisement starts playing;

[0009] Based on a preset image acquisition model, images are acquired in the viewing area corresponding to the preset playback area, and when each image is acquired, the current time is obtained respectively, and the current time is added to the acquired image in the format of a watermark;

[0010] Performing face recognition and face statistics on the collected pictures based on a preset face recognition model;

[0011] Based on a preset eye tracking technology, obtaining the number of people who watch the advertising video in the collected picture;

[0012] Based on a preset probability algorithm model, respectively obtain the ratio between the number of people who watched the advertisement video and the face statistics result in each collected picture;

[0013] Obtain watermark information of all collected pictures in the collected image information, and determine the advertising video corresponding to each collected picture by taking two adjacent time nodes as the starting point and the ending point respectively;

[0014] Based on the preset viewing statistics and comparison model, the popularity of different advertising videos when they are played is obtained respectively, and the advertising video corresponding to the highest popularity is selected as the high-quality advertisement of the product.

[0015] Furthermore, the method for selecting high-quality advertisements, wherein the step of capturing advertisement videos through the advertisement links comprises:

[0016] Based on a preset video capture tool, the advertisement link is used as an address for capture.

[0017] Furthermore, in the method for selecting high-quality advertisements, the preset image acquisition model includes:

[0018] According to a preset collection time interval, image collection is performed on the viewing area corresponding to the preset playback area to obtain a plurality of collected pictures.

[0019] Furthermore, the method for selecting high-quality advertisements, the face recognition based on a preset face recognition model, and the face statistics include:

[0020] Based on the mat l ab image extraction technology, the preset first square frame is used to extract the faces in the collected picture, and the first cut picture is generated, and the faces are numbered respectively to count the number of faces.

[0021] Furthermore, in the method for selecting high-quality advertisements, obtaining the number of people who watched the advertisement video in the collected picture based on the preset eye tracking technology includes:

[0022] The first cut picture is enlarged in proportion, and the MATLAB image extraction technology is used again to extract the eye part of the first cut picture using a preset second grid frame, and a second cut picture is generated;

[0023] Based on the preset two-dimensional coordinate system, the left corner of the left eye in the second cut image is marked as L l , the right corner of the left eye is marked as L r , the left corner of the right eye is marked as Rl , the right corner of the right eye is marked as R r , respectively obtain L l , L r , R l , R r The coordinate value in the two-dimensional coordinate system is the center of the pupil in the left eye, marked as L o , get the pupil center of the right eye marked as R o , respectively obtain L o and R o The coordinate value of

[0024] Judge L o , R o The coordinate value of the y-axis corresponds to the value of L l , L r , R l , R r whether the y-axis values ​​corresponding to the coordinate values ​​are the same, if they are the same, the face corresponding to the second cut image is in a level view state;

[0025] If it is in the level view state, judge L o Is the x-axis value of the coordinate value within the L l , L r The middle position of the x-axis value of the coordinate value is used to determine R o Is the x-axis value of the coordinate value within the R l , R r The coordinate values ​​of the x-axis are in the middle position. If they are all in the middle position, the face corresponding to the second cut picture is in a straight-view state and is watching the advertising video;

[0026] Otherwise, if L o The x-axis value of the coordinate value is not in the L l , L r The x-axis value of the coordinate value is the middle position or R o The x-axis value of the coordinate value is not in the R l , R r The x-axis value of the coordinate value is the middle position, then the L o or R o The offset of the x-coordinate of the second cut image relative to the x-coordinate of the center point, if the offset is greater than the threshold of the allowable offset, the face corresponding to the second cut image is not in a viewing state;

[0027] The number of faces corresponding to the second cut picture that are in a direct-viewing state is obtained as the number of people watching the advertising video in the collected picture.

[0028] Furthermore, in the method for selecting high-quality advertisements, the step of obtaining watermark information of all collected images in the collected image information includes:

[0029] Based on the preset text recognition technology, the time information in the watermark information is obtained.

[0030] Furthermore, in the method for selecting high-quality advertisements, the step of obtaining the popularity of different advertisement videos when they are played includes:

[0031] When the same advertisement video is obtained, the ratio of the number of people who watched the advertisement video and the face statistics result corresponding to all the collected pictures is obtained, and a ratio set is generated;

[0032] The elements in the ratio set are standardized to obtain a standard value, and the standard value is used as a popularity index of the advertising video.

[0033] In order to solve the above technical problems, the embodiment of the present application further provides a device for selecting high-quality advertisements, which adopts the following technical solution:

[0034] A device for selecting high-quality advertisements, comprising:

[0035] An advertising video acquisition module is used to acquire N advertising links of the same product to be launched, capture advertising videos through the advertising links, and number the advertising videos from 1 to N, where N is a positive integer greater than 1;

[0036] A carousel test module is used to push the advertisement videos to a preset play area in the order of numbers, perform a carousel test, and obtain the current time as the node time when each advertisement starts playing;

[0037] An image acquisition module, for acquiring images of the viewing area corresponding to the preset playback area based on a preset image acquisition model, obtaining the current time when acquiring each image, and adding the current time to the acquired image in the format of a watermark;

[0038] A face statistics module, used for performing face recognition and face statistics on the collected pictures based on a preset face recognition model;

[0039] A viewing number determination module, used to obtain the number of people who watch the advertising video in the collected picture based on a preset eye tracking technology;

[0040] A viewing ratio acquisition module is used to obtain the ratio between the number of people who watched the advertisement video and the face statistics result in each collected picture based on a preset probability algorithm model;

[0041] The picture and advertisement corresponding module is used to obtain the watermark information of all collected pictures in the collected image information, and determine the advertisement video corresponding to each collected picture with two adjacent time nodes as the starting point and the end point respectively;

[0042] The high-quality advertisement selection module is used to obtain the popularity of different advertisement videos when they are played based on preset viewing statistics and comparison models, and select the advertisement video corresponding to the highest popularity as the high-quality advertisement of the product.

[0043] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0044] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a method for selecting high-quality advertisements proposed in an embodiment of the present application are implemented.

[0045] In order to solve the above technical problems, the embodiment of the present application further provides a non-volatile computer-readable storage medium, which adopts the following technical solution:

[0046] A non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of a method for selecting high-quality advertisements proposed in an embodiment of the present application are implemented.

[0047] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0048] The embodiment of the present application discloses a method, device, equipment and storage medium for selecting high-quality advertisements, which captures advertisement videos through advertisement links; performs carousel testing on advertisement videos; performs image acquisition on viewing areas; performs face statistics on the collected pictures; obtains the number of people who watch the advertisement videos in the collected pictures; obtains the proportional relationship between the number of people who watch the advertisement videos in each collected picture and the face statistics result; determines the advertisement video corresponding to each collected picture; obtains the popularity of different advertisement videos when they are played, and selects the advertisement video corresponding to the highest popularity as the high-quality advertisement. The present application selects advertisements based on their popularity among users, effectively avoiding the blind delivery of advertisements and saving advertising costs to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 is an exemplary system architecture diagram to which the embodiments of the present application can be applied;

[0051] Figure 2 A flowchart of an embodiment of the method for selecting high-quality advertisements described in the embodiments of the present application;

[0052] Figure 3 A schematic diagram of a two-dimensional coordinate system of the second cutting picture in an embodiment of the present application;

[0053] Figure 4 This is a schematic diagram of the structure of an embodiment of a device for selecting high-quality advertisements described in an embodiment of the present application;

[0054] Figure 5 This is a schematic diagram of the structure of a viewing number determination module in an embodiment of the present application;

[0055] Figure 6 It is a structural diagram of an embodiment of a computer device in the embodiments of the present application. DETAILED DESCRIPTION

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0057] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0058] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0059] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0060] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0061] The terminal devices 101, 102, and 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, and the like.

[0062] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0063] It should be noted that the method for selecting high-quality advertisements provided in the embodiments of the present application is generally executed by a server / terminal device, and accordingly, the device for selecting high-quality advertisements is generally arranged in the server / terminal device.

[0064] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0065] Continue to refer Figure 2 , the figure shows a flow chart of an embodiment of a method for selecting high-quality advertisements of the present application, and the method for selecting high-quality advertisements comprises the following steps:

[0066] Step 201, obtain N advertisement links of the same product to be launched, capture advertisement videos through the advertisement links, and number the advertisement videos from 1 to N, where N is a positive integer greater than 1.

[0067] In this embodiment, the capturing of the advertisement video through the advertisement link includes: based on a preset video capture tool, taking the advertisement link as an address to capture, for example, using a WebView Downloader video capture tool to capture the video.

[0068] Step 202: Push the advertisement videos to a preset play area in a serial number order, perform a carousel test, and obtain the current time as the node time when each advertisement starts playing.

[0069] Step 203, based on a preset image acquisition model, perform image acquisition on the viewing area corresponding to the preset playback area, obtain the current time when acquiring each picture, and add the current time to the acquired picture in the format of a watermark.

[0070] In this embodiment, the preset image acquisition model includes: performing image acquisition on the viewing area corresponding to the preset playback area according to a preset acquisition time interval to obtain a plurality of acquired pictures.

[0071] Step 204, performing face recognition and face statistics on the collected pictures based on a preset face recognition model.

[0072] In this embodiment, face recognition is performed based on a preset facial recognition model, and face statistics are performed. A possible implementation method is to extract faces from the collected pictures using a preset first square frame based on MATLAB image extraction technology, generate a first cut picture, and number them separately to count the number of faces.

[0073] Step 205: Based on the preset eye tracking technology, the number of people who watch the advertisement video in the collected picture is obtained.

[0074] In this embodiment, the method of obtaining the number of people watching the advertisement video in the collected image based on the preset eye tracking technology includes the following steps: proportionally enlarging the first cut image, and again using the matlab image extraction technology to extract the eye part of the first cut image using the preset second grid frame, and generating a second cut image; based on the preset two-dimensional coordinate system, marking the left corner of the left eye in the second cut image as L l , the right corner of the left eye is marked as L r , the left corner of the right eye is marked as R l, the right corner of the right eye is marked as R r , respectively obtain L l , L r , R l , R r The coordinate value in the two-dimensional coordinate system is the center of the pupil in the left eye, marked as L o , get the pupil center of the right eye marked as R o , respectively obtain L o and R o Coordinate value of L o , R o The coordinate value of the y-axis corresponds to the value of L l , L r , R l , R r The y-axis values ​​corresponding to the coordinate values ​​of L are the same. If they are the same, the face corresponding to the second cut image is in a level view state; if it is in a level view state, determine L o Is the x-axis value of the coordinate value within the L l , L r The middle position of the x-axis value of the coordinate value is used to determine R o Is the x-axis value of the coordinate value within the R l , R r The x-axis value of the coordinate value of L is in the middle position. If they are all in the middle position, the face corresponding to the second cut picture is in a straight-looking state and is watching the advertising video; otherwise, if L o The x-axis value of the coordinate value is not in the L l , L r The x-axis value of the coordinate value is the middle position or R o The x-axis value of the coordinate value is not in the R l , R r The x-axis value of the coordinate value is the middle position, then the L o or R o The offset of the x-coordinate relative to the x-coordinate of the center point, if the offset is greater than the threshold value of the allowable offset, the face corresponding to the second cut image is not in a viewing state; the number of faces corresponding to the second cut image that are in a direct viewing state is obtained as the number of people watching the advertising video in the collected image.

[0075] Specific reference Figure 3 , Figure 3 Schematic diagram of the two-dimensional coordinate system of the second cutting picture in the embodiment of the present application, showing the x-axis and the y-axis, 301 in the figure shows the second cutting picture, and 302 in the figure shows L l , 303 in the figure shows L r , FIG. 304 shows R l, FIG. 305 shows R r , FIG. 306 shows L o , Figure 307 shows R o .

[0076] In the embodiment of the present application, the method of obtaining the number of people watching the advertising video in the collected picture based on the preset eye tracking technology also includes: proportionally enlarging the first cut picture, and again using the matlab image extraction technology to extract the eyes and eyebrows of the first cut picture using a preset third grid frame, and generating a third cut picture; based on the preset two-dimensional coordinate system, marking the left corner of the left eye in the third cut picture as L l1 , the right corner of the right eye is marked as R r1 , respectively obtain L l1 , R r1 The coordinate value in the two-dimensional coordinate system. At the same time, based on the preset two-dimensional coordinate system, the highest point of the upper edge of the left eyebrow in the third cut image is marked as B. l , the highest point of the right eyebrow is marked as B r , respectively obtain B l , B r The coordinate value in the two-dimensional coordinate system; obtain the straight-line distance between the left eye corner of the left eye and the highest point of the upper edge of the left eyebrow in the third cut image, recorded as L l1 B l , obtain the straight-line distance between the left eye corner of the left eye and the highest point of the upper edge of the left eyebrow in the third cut image, recorded as L l1 B r , obtain the straight-line distance between the right eye corner of the right eye and the highest point of the upper edge of the left eyebrow in the third cut image, recorded as R r1 B l , obtain the straight-line distance between the right eye corner of the right eye and the highest point of the upper edge of the right eyebrow in the third cut image, recorded as R r1 B r .

[0077] Step 206, based on a preset probability algorithm model, respectively obtain the proportional relationship between the number of people who watch the advertisement video in each collected picture and the face statistics result.

[0078] In the embodiment of the present application, the method of obtaining the proportional relationship between the number of people who watch the advertising video in each collected picture and the face statistics result includes: obtaining the number of people who watch the advertising video in the collected picture through the second cut picture, counting as Pi2, obtaining the number of the first cut pictures, counting as Pi1, based on the preset probability algorithm formula: The ratio between the number of people watching the advertisement video in the collected pictures and the face statistics result is obtained, which is expressed as P.

[0079] In the embodiment of the present application, the obtaining of the proportional relationship between the number of people watching the advertisement video in each collected picture and the face counting result further includes: obtaining the L corresponding to each of the third cut pictures based on a preset collection time interval. l1 B l , L l1 B r , R r1 B l , R r1 B r value, and respectively correspond to the L corresponding to the first third cut picture l1 B l , L l1 B r , R r1 B l , R r1 B r The L l1 B l , L l1 B r , R r1 B l , R r1 B r The change value, based on the L l1 B l , L l1 B r , R r1 B l , R r1 B r The change value of the eyebrow-eye distance of the person watching the advertising video in each collected picture is determined, and the interest degree of the person watching the advertising video in the advertising video is determined based on the change of the eyebrow-eye distance.

[0080] In the embodiment of the present application, the L l1 B l , L l1 B r , R r1 B l , R r1 B r , determine the change in the distance between the eyebrows and eyes of the person watching the advertising video in each collected picture, and determine the degree of interest of the person watching the advertising video in the advertising video based on the change in the distance between the eyebrows and eyes. The specific implementation method is: set the L l1 B l , L l1 B r , R r1 B l , R r1 Br The change values ​​are B1, B2, B3, and B4 respectively; based on the preset first distance algorithm formula: Obtain the change in the distance between the left eyebrow and the eyebrow; based on a preset second distance algorithm formula: Obtain the change in the distance between the right eyebrow and the right eyebrow; based on a preset threshold Q, determine the Q l And the Q r Whether it exceeds the threshold value Q, if it exceeds, the people watching the advertising video are more interested in the advertising video, otherwise, the people watching the advertising video are not interested in the advertising video enough; count the number of people Pi3 who are more interested in the advertising video.

[0081] Step 207, obtaining watermark information of all collected pictures in the collected image information, taking two adjacent time nodes as the starting point and the ending point respectively, and determining the advertising video corresponding to each collected picture.

[0082] In this embodiment, the acquisition of watermark information of all collected images in the collected image information includes: based on a preset text recognition technology, acquiring time information in the watermark information, for example, using OCR recognition technology to acquire time information on the watermark in the image.

[0083] Step 208, based on the preset viewing statistics and comparison model, respectively obtain the popularity of different advertising videos when they are played, and select the advertising video corresponding to the highest popularity as the high-quality advertisement of the product.

[0084] In this embodiment, one way to obtain the popularity of different advertising videos when they are played is as follows: when belonging to the same advertising video, obtain the ratio of the number of people who watch the advertising video corresponding to all the collected pictures and the face statistics result, and generate a ratio set; standardize the elements in the ratio set to obtain a standard value, and use the standard value as the popularity index of the advertising video.

[0085] In this embodiment, another way to obtain the popularity of different advertising videos when they are played is to obtain the proportional relationship between the number of people who watch the advertising video and the face statistics corresponding to all the collected pictures belonging to the same advertising video, and at the same time, obtain the number of people Pi3 who have a strong interest in the advertising video, based on the preset interest degree algorithm formula: An interest level value is obtained, and the interest level value ω is used as a popularity index of the advertisement video.

[0086] The method for selecting high-quality advertisements described in the embodiments of the present application can capture advertisement videos through advertisement links; perform carousel tests on advertisement videos; perform image acquisition on viewing areas; perform face statistics on the acquired pictures; obtain the number of people who watch the advertisement videos in the acquired pictures; respectively obtain the proportional relationship between the number of people who watch the advertisement videos in each acquired picture and the face statistics result; determine the advertisement video corresponding to each acquired picture; respectively obtain the popularity of different advertisement videos when they are played, and select the advertisement video corresponding to the highest popularity as the high-quality advertisement. The present application selects advertisements based on their popularity among users, effectively avoiding the blind delivery of advertisements and saving advertising costs to a certain extent.

[0087] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0088] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0089] Further references Figure 4 , as a response to the above Figure 2 The present application provides an embodiment of a device for selecting high-quality advertisements. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0090] like Figure 4As shown, the high-quality advertisement selection device 7 described in this embodiment includes: an advertisement video acquisition module 401, a carousel test module 402, an image acquisition module 403, a face counting module 404, a viewing number determination module 405, a viewing ratio acquisition module 406, a picture and advertisement correspondence module 407 and a high-quality advertisement selection module 408. Among them:

[0091] The advertisement video acquisition module 401 is used to acquire N advertisement links of the same product to be launched, capture advertisement videos through the advertisement links, and number the advertisement videos from 1 to N, where N is a positive integer greater than 1;

[0092] The carousel test module 402 is used to push the advertisement videos to the preset play area in the order of numbers, perform a carousel test, and obtain the current time as the node time when each advertisement starts playing;

[0093] The image acquisition module 403 is used to acquire images of the viewing area corresponding to the preset playback area based on a preset image acquisition model, obtain the current time when acquiring each picture, and add the current time to the acquired picture in the format of a watermark;

[0094] The face counting module 404 is used to perform face recognition and face counting on the collected pictures based on a preset face recognition model;

[0095] The viewing number determination module 405 is used to obtain the number of people who watch the advertisement video in the collected picture based on a preset eye tracking technology;

[0096] The viewing ratio acquisition module 406 is used to respectively acquire the ratio relationship between the number of people who watched the advertisement video and the face statistics result in each collected picture based on a preset probability algorithm model;

[0097] The picture and advertisement corresponding module 407 is used to obtain the watermark information of all collected pictures in the collected image information, and determine the advertisement video corresponding to each collected picture by taking two adjacent time nodes as the starting point and the ending point respectively;

[0098] The high-quality advertisement selection module 408 is used to obtain the popularity of different advertisement videos when they are played based on preset viewing statistics and comparison models, and select the advertisement video corresponding to the highest popularity as the high-quality advertisement of the product.

[0099] In some embodiments of the present application, the advertisement video acquisition module 401 is used to capture the advertisement link as an address based on a preset video capture tool, for example, using a WebVi deo Downloader video capture tool to capture the video.

[0100] In some embodiments of the present application, the image acquisition module 403 is used to acquire images of the viewing area corresponding to the preset playback area at a preset acquisition time interval to obtain a plurality of acquired pictures.

[0101] In some embodiments of the present application, the face counting module 404 is used to extract faces from the captured image based on the mat la b image extraction technology using a preset first square frame, generate a first cut image, and number them respectively to count the number of faces.

[0102] In some embodiments of the present application, Figure 5 , Figure 5 4 is a schematic diagram of the structure of the viewing number determination module in the embodiment of the present application. The viewing number determination module 405 includes a second cutting unit 405a, a level viewing state judgment unit 405b, a direct viewing state judgment unit 405c, a viewing state judgment unit 405d, and a viewing number counting unit 405e.

[0103] In some embodiments of the present application, the second cutting unit 405a is used to perform proportional enlargement processing on the first cutting picture, and use the matlab image extraction technology again to extract the eye part of the first cutting picture using a preset second grid frame, and generate a second cutting picture;

[0104] In some embodiments of the present application, the horizontal state judgment unit 405b is used to mark the left eye corner of the left eye in the second cut picture as L based on a preset two-dimensional coordinate system. l , the right corner of the left eye is marked as L r , the left corner of the right eye is marked as R l , the right corner of the right eye is marked as R r , respectively obtain L l , L r , R l , R r The coordinate value in the two-dimensional coordinate system is the center of the pupil in the left eye, marked as L o , get the pupil center of the right eye marked as R o , respectively obtain L o and R o Coordinate value of L o , R o The coordinate value of the y-axis corresponds to the value of L l, L r , R l , R r Whether the y-axis values ​​corresponding to the coordinate values ​​are the same, if they are the same, the face corresponding to the second cut image is in a level view state.

[0105] In some embodiments of the present application, the direct-view state judgment unit 405c is used to judge whether L is in the level-view state. o Is the x-axis value of the coordinate value within the L l , L r The middle position of the x-axis value of the coordinate value is used to determine R o Is the x-axis value of the coordinate value within the R l , R r The x-axis value of the coordinate value is in the middle position. If they are all in the middle position, the face corresponding to the second cut picture is in a direct viewing state and is watching the advertising video.

[0106] In some embodiments of the present application, the viewing status determination unit 405d is used if L o The x-axis value of the coordinate value is not in the L l , L r The x-axis value of the coordinate value is the middle position or R o The x-axis value of the coordinate value is not in the R l , R r The middle position of the x-axis value of the coordinate value is obtained, then the Lo or R o The offset of the x-coordinate of the second cut image relative to the x-coordinate of the center point, if the offset is greater than the threshold of the allowable offset, the face corresponding to the second cut image is not in a viewing state;

[0107] In some embodiments of the present application, the viewing number counting unit 405e is used to obtain the number of faces corresponding to the second cut picture that are in a direct viewing state as the number of people watching the advertising video in the collected picture.

[0108] In some embodiments of the present application, the viewing ratio acquisition module 406 is used to obtain the number of people who watch the advertising video in the collected picture through the second cut picture, counted as Pi2, obtain the number of first cut pictures, counted as Pi1, based on the preset probability algorithm formula: The ratio between the number of people watching the advertisement video in the collected pictures and the face statistics result is obtained, which is expressed as P.

[0109] In some embodiments of the present application, the picture and advertisement correspondence module 407 is used to obtain time information in the watermark information based on a preset text recognition technology, for example, using OCR recognition technology to obtain time information on the watermark in the picture.

[0110] In some embodiments of the present application, the high-quality advertisement selection module 408 is used to obtain the proportional relationship between the number of people who watch the advertisement video and the face statistics results corresponding to all the collected pictures belonging to the same advertisement video, and generate a ratio set; standardize the elements in the ratio set to obtain a standard value, use the standard value as the popularity index of the advertisement video, and select the advertisement video corresponding to the highest popularity as the high-quality advertisement for the product.

[0111] In addition, in the embodiment of the present application, the viewing number determination module may further include a third cutting unit, which is used to perform proportional enlargement processing on the first cutting picture, and use the mat lab image extraction technology again to extract the eyes and eyebrows of the first cutting picture using a preset third grid frame, and generate a third cutting picture; based on the preset two-dimensional coordinate system, the left corner of the left eye in the third cutting picture is marked as L l1 , the right corner of the right eye is marked as R r1 , respectively obtain L l1 , R r1 The coordinate value in the two-dimensional coordinate system. At the same time, based on the preset two-dimensional coordinate system, the highest point of the upper edge of the left eyebrow in the third cut image is marked as B. l , the highest point of the right eyebrow is marked as B r , respectively obtain B l , B r The coordinate value in the two-dimensional coordinate system; obtain the straight-line distance between the left eye corner of the left eye and the highest point of the upper edge of the left eyebrow in the third cut image, recorded as L l1 B l , obtain the straight-line distance between the left eye corner of the left eye and the highest point of the upper edge of the left eyebrow in the third cut image, recorded as L l1 B r , obtain the straight-line distance between the right eye corner of the right eye and the highest point of the upper edge of the left eyebrow in the third cut image, recorded as R r1 B l , obtain the straight-line distance between the right eye corner of the right eye and the highest point of the upper edge of the right eyebrow in the third cut image, recorded as R r1 B r .

[0112] In addition, in some embodiments of the present application, the viewing ratio acquisition module may further include: based on a preset acquisition time interval, acquiring the L corresponding to each of the third cut pictures l1 B l , L l1 B r , R r1 B l , R r1 B rvalue, and respectively correspond to the L corresponding to the first third cut picture l1 B l , L l1 B r , R r1 B l , R r1 B r The L l1 B l , L l1 B r , R r1 B l , R r1 B r The change value, based on the L l1 B l , L l1 B r , R r1 B l , R r1 B r , determining the change in the distance between the eyebrows and eyes of the person watching the advertising video in each collected picture, and determining the interest level of the person watching the advertising video in the advertising video based on the change in the distance between the eyebrows and eyes, wherein the l1 B l , L l1 B r , R r1 B l , R r1 B r , determine the change in the distance between the eyebrows and eyes of the person watching the advertising video in each collected picture, and determine the degree of interest of the person watching the advertising video in the advertising video based on the change in the distance between the eyebrows and eyes. The specific implementation method is: set the L l1 B l , L l1 B r , R r1 B l , R r1 B r The change values ​​are B1, B2, B3, and B4 respectively; based on the preset first distance algorithm formula: Obtain the change in the distance between the left eyebrow and the eyebrow; based on a preset second distance algorithm formula: Obtain the change in the distance between the right eyebrow and the right eyebrow; based on a preset threshold Q, determine the Q l And the Q rWhether it exceeds the threshold value Q, if it exceeds, the people watching the advertising video are more interested in the advertising video, otherwise, the people watching the advertising video are not interested in the advertising video enough; count the number of people Pi3 who are more interested in the advertising video, obtain the proportional relationship between the number of people watching the advertising video and the face statistics result corresponding to all the collected pictures belonging to the same advertising video, and at the same time, obtain the number of people Pi3 who are more interested in the advertising video, based on the preset interest degree algorithm formula: An interest level value is obtained, and the interest level value ω is used as a popularity index of the advertisement video.

[0113] The high-quality advertisement selection device described in the embodiment of the present application captures advertisement videos through advertisement links; performs carousel testing on advertisement videos; performs image acquisition on viewing areas; performs face statistics on the acquired pictures; obtains the number of people who watch the advertisement videos in the acquired pictures; obtains the proportional relationship between the number of people who watch the advertisement videos in each acquired picture and the face statistics result; determines the advertisement video corresponding to each acquired picture; obtains the popularity of different advertisement videos when they are played, and selects the advertisement video corresponding to the highest popularity as the high-quality advertisement. The present application selects advertisements based on their popularity among users, effectively avoiding the blind delivery of advertisements and saving advertising costs to a certain extent.

[0114] To solve the above technical problems, the present application also provides a computer device. Figure 6 , Figure 6 This is a basic structural block diagram of the computer device in this embodiment.

[0115] The computer device 6 includes a memory 6a, a processor 6b, and a network interface 6c that are interconnected through a system bus. It should be noted that the figure only shows a computer device 6 with components 6a-6c, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to a microprocessor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital processor (DSP), an embedded device, etc.

[0116] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0117] The memory 6a includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 6a can be an internal storage unit of the computer device 6, such as a hard disk or memory of the computer device 6. In other embodiments, the memory 6a can also be an external storage device of the computer device 6, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 6. Of course, the memory 6a can also include both the internal storage unit of the computer device 6 and its external storage device. In this embodiment, the memory 6a is usually used to store the operating system and various application software installed on the computer device 6, such as the program code of the method for selecting high-quality advertisements, etc. In addition, the memory 6a can also be used to temporarily store various types of data that have been output or are to be output.

[0118] The processor 6b may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 6b is generally used to control the overall operation of the computer device 6. In this embodiment, the processor 6b is used to run the program code stored in the memory 6a or process data, such as running the program code of the method for selecting high-quality advertisements.

[0119] The network interface 6c may include a wireless network interface or a wired network interface. The network interface 6c is generally used to establish a communication connection between the computer device 6 and other electronic devices.

[0120] The present application also provides another embodiment, namely, providing a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a high-quality advertisement selection program, and the high-quality advertisement selection program can be executed by at least one processor so that the at least one processor performs the steps of the high-quality advertisement selection method as described above.

[0121] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0122] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A method for selecting high-quality advertisements, characterized in that: The steps include: Obtain N advertisement links of the same product to be launched, capture advertisement videos through the advertisement links, and number the advertisement videos from 1 to N, where N is a positive integer greater than 1; Push the advertising videos to a preset play area in the order of numbers, perform a carousel test, and obtain the current time as the node time when each advertisement starts playing; Based on a preset image acquisition model, images are acquired in the viewing area corresponding to the preset playback area, and when each image is acquired, the current time is obtained respectively, and the current time is added to the acquired image in the format of a watermark; The preset image acquisition model includes: performing image acquisition on the viewing area corresponding to the preset playback area according to a preset acquisition time interval to obtain a plurality of acquired pictures; For the collected pictures, face recognition is performed based on a preset face recognition model, and face statistics are performed; based on a preset eye tracking technology, the number of people who watch the advertising video in the collected pictures is obtained; the face recognition and face statistics based on the preset face recognition model include: based on the matlab image extraction technology, using a preset first square frame to extract the faces in the collected pictures, and generating a first cut picture, and numbering them respectively, and counting the number of faces; based on the preset eye tracking technology, obtaining the number of people who watch the advertising video in the collected pictures includes: proportionally enlarging the first cut picture, and using the matlab image extraction technology again, using a preset second square frame to extract the eye part of the first cut picture, and generating a second cut picture; based on a preset two-dimensional coordinate system, marking the left corner of the left eye in the second cut picture as L l , the right corner of the left eye is marked as L r , the left corner of the right eye is marked as R l , the right corner of the right eye is marked as R r , respectively obtain L l , L r , R l , R r The coordinate value in the two-dimensional coordinate system is the center of the pupil in the left eye, marked as L o , get the pupil center of the right eye marked as R o , respectively obtain L o and R o Coordinate value of L o , R o The coordinate value of the y-axis corresponds to the value of L l , L r , R l , R r The y-axis values ​​corresponding to the coordinate values ​​of L are the same. If they are the same, the face corresponding to the second cut image is in a level view state; if it is in a level view state, determine L o Is the x-axis value of the coordinate value within the L l , L r The middle position of the x-axis value of the coordinate value is used to determine R o Is the x-axis value of the coordinate value within the R l , R r The x-axis value of the coordinate value of L is in the middle position. If they are all in the middle position, the face corresponding to the second cut picture is in a direct view state and is watching the advertising video; otherwise, if L o The x-axis value of the coordinate value is not in the L l , L r The x-axis value of the coordinate value is the middle position or R o The x-axis value of the coordinate value is not in the R l , R r The x-axis value of the coordinate value is the middle position, then the L o or R o The offset of the x-coordinate of the second cut picture relative to the x-coordinate of the center point, if the offset is greater than the threshold value of the allowed offset, the face corresponding to the second cut picture is not in the viewing state; the number of faces corresponding to the second cut picture that are in the direct viewing state is obtained as the number of people watching the advertising video in the collected picture; Based on a preset probability algorithm model, respectively obtain the ratio between the number of people who watched the advertisement video and the face statistics result in each collected picture; Obtain watermark information of all collected pictures in the collected image information, and determine the advertising video corresponding to each collected picture by taking two adjacent time nodes as the starting point and the ending point respectively; Based on the preset viewing statistics and comparison model, the popularity of different advertising videos when they are played is obtained respectively, and the advertising video corresponding to the highest popularity is selected as the high-quality advertisement of the product.

2. The method for selecting high-quality advertisements according to claim 1, characterized in that: The method of capturing the advertisement video through the advertisement link comprises: Based on a preset video capture tool, the advertisement link is used as an address for capture.

3. The method for selecting high-quality advertisements according to claim 1, characterized in that: The watermark information of all collected pictures in the collected image information is obtained, including: Based on the preset text recognition technology, the time information in the watermark information is obtained.

4. The method for selecting high-quality advertisements according to claim 1, characterized in that: The obtaining of the popularity of different advertising videos when they are played includes: When the same advertisement video is obtained, the ratio of the number of people who watched the advertisement video and the face statistics result corresponding to all the collected pictures is obtained, and a ratio set is generated; The elements in the ratio set are standardized to obtain a standard value, and the standard value is used as a popularity index of the advertising video.

5. A device for selecting high-quality advertisements, characterized in that: include: An advertising video acquisition module is used to acquire N advertising links of the same product to be launched, capture advertising videos through the advertising links, and number the advertising videos from 1 to N, where N is a positive integer greater than 1; A carousel test module is used to push the advertisement videos to a preset play area in the order of numbers, perform a carousel test, and obtain the current time as the node time when each advertisement starts playing; An image acquisition module, for acquiring images of the viewing area corresponding to the preset playback area based on a preset image acquisition model, obtaining the current time when acquiring each image, and adding the current time to the acquired image in the format of a watermark; The preset image acquisition model includes: performing image acquisition on the viewing area corresponding to the preset playback area according to a preset acquisition time interval to obtain a plurality of acquired pictures; A face statistics module, used for performing face recognition and face statistics on the collected pictures based on a preset face recognition model; The module for determining the number of viewers is used to perform face recognition and face counting on the collected pictures based on a preset face recognition model; based on a preset eye tracking technology, the number of people watching the advertising video in the collected pictures is obtained; the face recognition and face counting based on the preset face recognition model include: based on the matlab image extraction technology, using a preset first square frame to extract the faces in the collected pictures, and generating a first cut picture, and numbering them respectively to count the number of faces; based on the preset eye tracking technology, the number of people watching the advertising video in the collected pictures includes: performing proportional enlargement processing on the first cut picture, and using the matlab image extraction technology again, using a preset second square frame to extract the eye part of the first cut picture, and generating a second cut picture; based on a preset two-dimensional coordinate system, marking the left corner of the left eye in the second cut picture as L l , the right corner of the left eye is marked as L r , the left corner of the right eye is marked as R l , the right corner of the right eye is marked as R r , respectively obtain L l , L r , R l , R r The coordinate value in the two-dimensional coordinate system is the center of the pupil in the left eye, marked as L o , get the pupil center of the right eye marked as R o , respectively obtain L o and R o Coordinate value of L o , R o The coordinate value of the y-axis corresponds to the value of L l , L r , R l , R r The y-axis values ​​corresponding to the coordinate values ​​of L are the same. If they are the same, the face corresponding to the second cut image is in a level view state; if it is in a level view state, determine L o Is the x-axis value of the coordinate value within the L l , L r The middle position of the x-axis value of the coordinate value is used to determine R o Is the x-axis value of the coordinate value within the R l , R r The x-axis value of the coordinate value of L is in the middle position. If they are all in the middle position, the face corresponding to the second cut picture is in a direct view state and is watching the advertising video; otherwise, if L o The x-axis value of the coordinate value is not in the L l , L r The x-axis value of the coordinate value is the middle position or R o The x-axis value of the coordinate value is not in the R l , R r The x-axis value of the coordinate value is the middle position, then the L o or R o The offset of the x-coordinate of the second cut picture relative to the x-coordinate of the center point, if the offset is greater than the threshold value of the allowed offset, the face corresponding to the second cut picture is not in the viewing state; the number of faces corresponding to the second cut picture that are in the direct viewing state is obtained as the number of people watching the advertising video in the collected picture; The viewing ratio acquisition module is used to obtain the ratio between the number of people who watch the advertising video and the face statistics results in each collected picture based on a preset probability algorithm model; the picture and advertisement correspondence module is used to obtain the watermark information of all collected pictures in the collected image information, and determine the advertising video corresponding to each collected picture with two adjacent time nodes as the starting point and the end point respectively; The high-quality advertisement selection module is used to obtain the popularity of different advertisement videos when they are played based on preset viewing statistics and comparison models, and select the advertisement video corresponding to the highest popularity as the high-quality advertisement of the product.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method for selecting high-quality advertisements according to any one of claims 1 to 4 when executing the computer program.

7. A non-volatile computer-readable storage medium, characterized in that: The non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for selecting high-quality advertisements according to any one of claims 1 to 4 are implemented.

Images