TWI574556B - Method and device for monitoring video advertisement displaying - Google Patents

Description

Video advertisement monitoring method and device

The present invention relates to the field of video advertisements, and in particular, to a video advertisement monitoring method and apparatus.

With the advancement of science and technology, the continuous development of the media industry, the influence of the media on people's daily life is increasing, which in turn promotes the rapid development of the media advertising industry. With the continuous development of the advertising industry, a variety of playback formats are produced, such as TV commercials, outdoor advertising, and Internet advertising; various forms of advertising, especially those for electronic screen broadcasting, are usually mobile, so it is difficult for advertisers to judge whether or not relevant advertisements are played.

In the prior art, the general method is to monitor and broadcast by manual inspection. Specifically, the monitoring personnel need to go to the monitoring site every day to monitor the playback of the video advertisement, and then count it as a monitoring report to report to the advertiser.

However, in the process of implementing the technical solution, the inventors found that the prior art has the following defects: the monitoring personnel cannot track the advertisement playing 24 hours, so it is difficult to judge whether the related advertisement is played, and the method of manually collecting the monitoring is time-consuming. Force and judgment is not accurate.

The invention provides a video advertisement monitoring and broadcasting method and device, which are used for solving the problem that the monitoring personnel who use the personnel monitoring mode in the prior art cannot track the advertisement broadcast 24 hours, so it is difficult to judge whether the related advertisement is played, and the manual collection and monitoring manner is adopted. Time-consuming and labor-intensive and inaccurate.

An aspect of the present invention is to provide a video advertisement monitoring method, including: instantly collecting video advertisement data played by an electronic screen; performing frame cutting processing on the video advertisement data according to a preset interval time, and obtaining a plurality of cut frame images, and The plurality of cut frame images are sorted according to a time point before the frame cutting process; the sorted cut frame images are grouped according to the preset time period, and image recognition is performed on all the cut frame images in each group, and the grouped images are obtained. Cutting a key time period corresponding to the frame picture; comparing the cut frame picture in the key time period with a preset standard picture to determine whether the target advertisement is played.

In the video advertisement monitoring method as described above, the image recognition is performed on all the cut frames in each group, and the key time segments corresponding to the grouped cut frame images are obtained, which specifically includes: if all the cuts in the group after grouping The frame picture feature point change degree is less than or equal to the first preset threshold value, and the time period corresponding to the group is used as the key time period.

In the video advertisement monitoring method as described above, the comparing the image of the cut frame in the critical time period with the preset standard image to determine whether the target advertisement is played, specifically includes: The feature points of all the cut frame pictures are compared with the feature points of the standard picture; if the number of matching points of the feature points of the cut frame picture and the feature points of the standard picture is greater than or equal to the second preset threshold, determining the The target advertisement has been played; or, if the number of matching points of the feature points of the cut frame picture and the feature points of the standard picture is less than the second preset threshold, it is determined that the target advertisement is not played.

The video advertisement monitoring method as described above, before the comparing the cut frame picture in the key time period with the preset standard picture, the method further includes: cutting the standard video data according to the preset interval time Processing to obtain the standard cut frame picture.

Another aspect of the present invention is to provide a video advertisement monitoring device, including: a video image acquisition module, which is used for instantly collecting video advertisement data played by an electronic screen; and a video image frame sorting module for The video advertisement data is subjected to frame cutting processing to obtain a plurality of frame cut images, and the plurality of cut frame images are sorted according to time points before the frame cutting process; the key time segment confirmation module is used to The time segment is used to group the sorted cut frame images, and image recognition is performed on all the cut frame images in each group to obtain a key time segment corresponding to the grouped cut frame image; the image comparison module is used for The cut frame image in the critical time period is compared with the preset standard image to determine whether the target advertisement is played.

In the video advertisement monitoring apparatus as described above, the key time period confirmation module is specifically configured to: if the degree of change of feature points of all the cut frames in the group after the group is less than or equal to the first preset threshold, the group is The corresponding time period is used as the key time period.

In the video advertisement monitoring apparatus as described above, the picture comparison module is further configured to: compare feature points of all the cut frame pictures in the key time period with feature points of the standard picture; if the frame image is cut If the number of matching points of the feature point with the feature point of the standard picture is greater than or equal to the second preset threshold, it is determined that the target advertisement has been played; or if the feature point of the cut frame picture is related to the feature point of the standard picture When the number of matches is less than the second preset threshold, it is determined that the target advertisement is for playing.

The video advertisement monitoring device, as described above, before the image comparison between the cut frame picture and the preset standard picture in the critical time period, further includes: a video image frame cutting processing module, configured to preset according to the preset The standard video information is subjected to frame cutting processing at intervals to obtain the standard cut frame picture.

The video advertisement monitoring and broadcasting method and device provided by the invention collects the video advertisement played by the electronic screen in real time, and then performs frame cutting processing on the video advertisement, and judges the critical time of the degree of change of all the cut frame image information in the preset time period. Segment, and then compare and contrast the cut frame image in the key time period with the standard picture or video information, effectively guarantee the monitoring of the video advertisement for 24 hours, and effectively save the analysis and processing through the confirmation of the key time period. The required time, at the same time, can accurately determine whether the target advertisement is broadcasted, effectively improving the efficiency of advertisement monitoring, reducing the labor cost and time cost, and facilitating the promotion and application of the market.

The specific embodiments of the present invention are further described in detail below with reference to the drawings and embodiments. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.

Example 1

1 is a schematic flowchart of a video advertisement monitoring and broadcasting method according to Embodiment 1. Referring to FIG. 1, the present invention provides a video advertisement monitoring method, which specifically includes: s1: Instantly collecting video advertisement data played by an electronic screen. ; s2: performing frame cutting processing on the video advertisement data according to the preset interval time, obtaining a plurality of cut frame images, and sorting the plurality of cut frame images according to time points before the frame cutting process; s3: according to the preset time The segment groups the sorted cut frame pictures, performs image recognition on all the cut frame pictures in each group, and obtains a key time period corresponding to the grouped cut frame picture; s4: cuts the frame in the key time period The image is compared with a preset standard image to determine if the target ad is playing.

The specific implementation manner of the video advertisement data for real-time acquisition of the electronic screen is not limited, and those skilled in the art can select a method for collecting video advertisement data according to specific conditions, such as a video capture device (such as a camera, a video recorder, etc.). The video data of the video is automatically collected in a 24-hour manner, and then the framed video processing data is subjected to frame cutting processing. In addition, the specific method for the frame cutting process is not limited, and those skilled in the art may use the prior art frame cutting method to target the video. The data is cut into frames, as long as the video advertisement data can be framed and processed, and a plurality of cut frame images are obtained, which will not be described here; after cutting the frames, several cut frame images are obtained, and the cut frame images are taken. The time point information corresponding to the picture before cutting the frame is sorted according to the time point information, and the selected pictures are sorted according to the time point information, and then the sorted pictures are grouped according to the time period according to the preset time period.

In order to more clearly understand the technical solution of the present invention, the inventor exemplifies the following information: if the video information obtained in 24 hours is subjected to frame cutting processing to obtain 240,000 cut frame pictures, assuming that the preset time period is 9 s, each will The 25 frames of cut frames in 9s are divided into 1 group, and then the 240,000 cut frame images obtained in 24 hours are divided into several groups of pictures with 9s time and 25 frames, and then several pictures in each group. The frame image is analyzed and processed, that is, the 25 frame-cut pictures in 9s are analyzed and processed, and then the time period corresponding to the picture group of 9s time is the key time period, wherein the specific preset time period is set by The advertiser and the video monitor perform the setting, and the person skilled in the art can also set according to the needs of the advertiser, and will not go into details here; after judging the information of the key time period, the video data collected for 24 hours is collected. In other words, there will be a number of key time periods, and then the image of the cut frames in the identified key time periods will be compared with the preset standard pictures. The target advertisement (the advertisement that the advertiser asks to play) is broadcasted according to the result of the image comparison. The specific method and implementation manner of the image comparison are not limited, and those skilled in the art can set it as desired, as long as the judgment can be achieved. Whether or not the judgment result of the target advertisement is broadcasted can be omitted, and will not be described here.

The video advertisement monitoring and broadcasting method provided by the invention obtains a video advertisement played by an electronic screen in real time, and then performs frame cutting processing on the video advertisement, and determines a critical time period for the degree of change of all the cut frame image information in the preset time period, Then, by comparing and comparing the cut frame picture in the key time period with the standard picture or video information, the monitoring of the video advertisement is effectively guaranteed for 24 hours, and the confirmation of the key time period effectively saves the analysis processing needs. At the same time, it is also able to accurately determine whether the target advertisement is broadcasted, effectively improving the efficiency of advertisement monitoring, reducing labor costs and time costs, and facilitating the promotion and application of the market.

Example 2

This embodiment is a supplementary description of the foregoing embodiment. Specifically, step 3: grouping the sorted framing frames according to a preset time period, and performing image recognition on all framing pictures in each group to obtain grouping. The critical time period corresponding to the cut frame picture; specifically: if the degree of change of the feature points of all the cut frames in the group after the group is less than or equal to the first preset threshold, the time period corresponding to the group is used as the key time The implementations and processes of the steps 1, 2, and 4 are the same as the implementations and processes of the steps in the first embodiment. For details, refer to the foregoing description, and details are not described herein again. In the judging process, the present invention is mainly determined by determining the degree of change of the feature point of all the framing pictures in the group after the grouping, and continues to take the 9s time and 25 frames of the frame in the embodiment 1 as an example. , comparing the 25 framing frames in the group one by one, if the degree of change of the feature points of the 25 framing pictures is less than or equal to the first preset threshold, then this indicates that the time is The time spent staying in a certain screen is relatively long, which means that the advertisement screen here is more important. Therefore, it is judged that the time period corresponding to the 9s time is a key time period, and if all the cut frame picture feature points in the group are changed after grouping If the average degree is greater than the first preset threshold, it indicates that the advertisement screen has a large degree of change in the period of 9s, and the time period corresponding to the group is not used as the key time period; wherein, the specific range of the first preset threshold is not To be limited, those skilled in the art can set according to the needs of the advertiser, as long as the key time period information can be determined, and details are not described herein.

By judging the degree of change of feature points of all the cut frames in the group after grouping, the key time period information is judged, which effectively increases the efficiency of the key time period judgment, thereby improving the cut frame picture and the preset standard picture through the key time period. Contrast, and then determine whether the accuracy of the target advertisement is played, and the accuracy and reliability of the method is improved.

Example 3

FIG. 2 is a schematic flowchart of the video advertisement monitoring and broadcasting method. Referring to FIG. 2, the embodiment is a supplementary description of the foregoing embodiment. Specifically, the step 4: the key time period is The image of the cut frame is compared with the preset standard image to determine whether the target advertisement is played, and the specifics include: s41: comparing the feature points of all the cut frames in the key time period with the feature points of the standard image. S42: determining whether the number of matching between the feature point of the cut frame picture and the feature point of the standard picture is greater than or equal to the second preset threshold; s43: if the feature point of the cut frame picture matches the feature point of the standard picture If the quantity is greater than or equal to the second preset threshold, determining that the target advertisement has been played; or, s44: if the number of matching points of the feature point of the cut frame picture and the feature point of the standard picture is less than the second preset threshold, It is determined that the target advertisement is not played.

The specific implementation manners and processes of the steps 1, 2, and 3 are the same as the implementations and processes of the steps in the first embodiment and the second embodiment. Those skilled in the art may refer to the foregoing embodiments 1 and 2. The implementation of the steps and the description of the process are not described here. In addition, for step 4, the image of the cut frame in the critical time period is compared with the preset standard image. In the implementation manner, all the cut frame images in the key time period are compared with the standard pictures, wherein the contrast is mainly the feature points in the picture, and in the judgment process, if the feature points of the cut frame picture and the feature points of the standard picture appear If the number of matched matches is greater than or equal to the second preset threshold, then the cut frame picture matches the standard picture, wherein the cut frame picture is a certain cut frame picture in the key time period. That is, when the standard picture matches a certain framed picture in the key time period, the conclusion that the target advertisement is played can be obtained, if the feature point of the cut frame picture appears If the number of matching points in the feature points of the quasi-picture is less than the second preset threshold, it means that all the framing pictures in the critical period of the period do not match the standard picture, wherein the framing picture here is All the cut frame pictures in the critical time period, that is, when the standard picture is compared with all the cut frame pictures in the key time period, if no match is matched, the conclusion that the target advertisement is not played can be obtained; in addition, for the second preset threshold The specific scope is not limited, and those skilled in the art can arbitrarily set according to the needs of the advertiser, and details are not described herein again.

By comparing the feature points of all the cut frame pictures in the key time period with the feature points of the standard picture one by one, and then comparing the number of matching points of the compared feature points with the second preset threshold value, judging whether to play or not The method of target advertising effectively guarantees the accuracy of judging whether or not to play the target advertisement. In addition, the scope of the comparison is improved. In this application, all the cut frame images in several key time periods are compared with the standard pictures. The method effectively prevents the occurrence of the phenomenon of missed advertisements by mistakenly checking the advertisements, and can effectively prevent the occurrence of the phenomenon that the advertisements are not found, and improves the accuracy and reliability of monitoring the advertisements by using the method. Sexuality is conducive to the promotion and application of the market.

Example 4

FIG. 3 is a schematic flowchart of a video advertisement monitoring and broadcasting method according to still another embodiment. Referring to FIG. 3, the embodiment is a supplementary description of the foregoing embodiment, specifically, in step 4: the key is Before the image of the cut frame in the time period is compared with the preset standard image, the method further includes: performing frame cutting processing on the standard video data according to the preset interval time to obtain the standard cut frame image.

The implementations and processes of the steps 1, 2, 3, and 4 are the same as the implementations of the steps in the first embodiment, the second embodiment, and the third embodiment. For details, refer to the foregoing description. This will not be described again; in addition, it should be noted that the preset standard pictures mentioned in Embodiment 1, Embodiment 2 and Embodiment 3 are provided by the advertiser, that is, the advertiser provides the target advertisement to be broadcasted. Standard comparison information, wherein the standard comparison information includes standard picture information and standard video information, and when the advertiser provides standard picture information, it is the case involved in Embodiment 1, Embodiment 2 and Embodiment 3, The schemes in the first embodiment, the second embodiment, and the third embodiment are used to determine whether or not the target advertisement is played.

When the video information provided by the advertiser is the case for the embodiment, the standard video information provided by the advertiser needs to be frame-cut, and the standard frame-cut image information corresponding to the standard video information is obtained. Then, the standard cut frame picture information is compared with the cut frame picture information in the key time period. For the specific comparison process, refer to the description in the foregoing embodiment 3, and details are not described again.

Specifically, those skilled in the art may pass, but are not limited to, the following manners, in a case where the target video to be played by the advertiser is highly targeted, that is, the video playback time of the provided standard video advertisement information is relatively short. In particular, the specific short-term judgment criteria are different, and those skilled in the art can set according to the needs of the advertiser, for example, the short-term time can be set within 5 minutes or within 10 minutes, and then directly All the cut frame images obtained by the standard video frame cutting process are used as the standard cut frame picture, and then all the standard cut frame pictures are compared with the cut frame pictures in the key time period. For the specific comparison process, refer to the above statement. , no longer repeat them.

In another case, when the target advertisement to be played by the advertiser is not highly targeted, that is, when the video playback time of the provided standard video advertisement information is long, the time information corresponding to the above is corresponding, for example, If the video playback time exceeds 5 minutes or 10 minutes, the video playback time is longer. In this case, after the video advertisement is cut according to the preset interval time, several framing pictures can be obtained. The time points corresponding to the pre-cut frame processing are sorted, and then the sorted cut frame pictures are grouped according to a preset time period, and image recognition is performed on all the cut frame pictures in each group to obtain the grouped cut frame picture. Corresponding key time period, then compare all the standard cut frame pictures in the key segment with the cut frame picture to get the conclusion of whether to play the target advertisement, wherein the confirmation of the above key time period and the standard cut frame picture and cut The comparison process of the frame pictures is similar to the confirmation process and the comparison process of the steps 3 and 4 in the embodiment 2 and the embodiment 3, and the specific process can be referred to. The above description is not described here.

When the standard advertisement comparison information provided by the advertiser is video advertisement information, the video advertisement information may be framed to obtain standard cut frame image information, and then the standard cut frame image information is compared with the cut frame image information to obtain whether to broadcast. The conclusion of the target advertisement, through the above method, effectively expands the scope of use of the method, improves the applicability of the method, and is beneficial to the promotion and application of the market.

Example 5

4 is a schematic structural diagram of a video advertisement monitoring apparatus according to the present invention. Referring to FIG. 4, according to another aspect of the present invention, a video advertisement monitoring apparatus is provided, including: a video image acquisition module 1, Instantly collecting video advertisement data played by the electronic screen; the video image cut frame sorting module 2 is configured to perform frame cutting processing on the video advertisement data according to a preset interval time, obtain a plurality of cut frame images, and obtain the plurality of cut frames The cut frame picture is sorted according to the time point before the frame processing; the key time period confirmation module 3 is configured to group the sorted cut frame pictures according to the preset time period, and perform mapping on all the cut frame pictures in each group. The image recognition module 4 is configured to compare the cut frame image in the key time period with a preset standard image to determine a target advertisement. Whether to play.

Among them, the video image acquisition module 1 does not specifically limit the implementation of the video advertisement data, and the video advertisement data can be collected by the camera or the video recorder, and the video advertisement can be collected by other methods. The technician can set according to the design requirements, as long as the function of real-time collecting video advertisement data can be realized, and details are not described herein; in addition, the video advertisement data is cut framed by the video image frame sorting module 2 The specific method is not limited, and a person skilled in the art may use a frame cutting method in the prior art to cut a frame of a video advertisement data, as long as the video advertisement data can be frame-cut, and a plurality of frame-cut images are obtained; After the frame processing, the video image frame sorting module 2 also sorts the plurality of cut frame images after the frame cut processing, wherein a plurality of cut frame images are accompanied by the time corresponding to the cut frame processing. Point information, therefore, the video image cut frame sorting module 2 can sort the time point information by Implementing an action of sorting a plurality of cut frame images; then, by using the key time period confirmation module 3, grouping the sorted cut frame images according to a preset time period, and changing the degree of all cut frame images in the group The analysis process is performed to determine the key time period information. For the specific judgment process, refer to the specific description in the embodiment 1, and details are not described herein. After the key time period confirmation module 3 confirms, there are several keys. The time period information is then compared with the preset standard picture by using the image comparison module 4 to compare the cut frame pictures in the determined key time periods, and judge whether the target is broadcast according to the image comparison result. The advertisement (the advertisement that the advertiser requests to play), wherein the specific method and implementation manner of the image comparison are not limited, and those skilled in the art can arbitrarily set, as long as the judgment result of determining whether the target advertisement is broadcasted can be reached. , will not repeat them here.

The video advertisement monitoring device provided by the invention acquires the video advertisement played by the electronic screen in real time through the video image capturing module 1, and then performs the frame sorting processing on the video advertisement through the video image cut frame sorting module 2, and passes the key The time period confirmation module 3 determines the key time period for the degree of change of all the cut frame picture information in the preset time period, and finally compares the cut frame picture and the standard picture or video information in the key time period by the picture comparison module 4 Analyze and contrast, effectively guarantee 24-hour monitoring of video ads, and through the confirmation of key time periods, effectively save the time required for analysis and processing, and also accurately determine whether the target advertisements are broadcast, effective The efficiency of the advertisement monitoring and broadcasting is improved, the labor cost and the time cost are reduced, thereby ensuring the stable reliability of the device, and improving the market competitiveness of the device, which is beneficial to the promotion and application of the market.

Example 6

This embodiment is a further supplement to the embodiment 5, specifically, wherein the key time period confirmation module 3 is specifically configured to: if all the framing frame picture feature points in the group after grouping are less than or equal to the first The preset threshold is used as the key time period for the corresponding time period of the group.

The key time period confirmation module 3 determines the key time period by comparing the feature point change degree of all the cut frame pictures in each group after the grouping and the first preset threshold value, thereby effectively improving the accuracy of the key time period judgment. And the efficiency of the judgment, wherein the specific range of the first preset threshold is not limited, and those skilled in the art can set according to the requirements or experience of the advertiser, and will not be described here; The criterion for determining the degree of change of the feature point of the picture is not specifically limited, and those skilled in the art may set it according to the specific actual situation; in addition, the process of how to perform the analysis and judgment is the same as the process in the second embodiment, and may be specifically implemented. The description in Example 2 will not be repeated here.

The critical time period confirmation module 3 judges the degree of change of the feature points of all the cut frames in the group after the group to judge the key time period information, effectively increases the efficiency of the key time period judgment, and further improves the frame cutting through the key time period. The picture is compared with the preset standard picture to determine whether the accuracy of the target advertisement is played, and the accuracy and reliability of the device are improved, thereby ensuring the accuracy and reliability of the analysis and judgment result.

Example 7

This embodiment is a further supplement to the foregoing embodiment. Specifically, the picture comparison module 4 is further configured to: perform feature points of all the framing pictures in the key time period and feature points of the standard picture. The comparison data is: whether the number of matching between the feature point of the cut frame picture and the feature point of the standard picture is greater than or equal to the second preset threshold; if the feature point of the cut frame picture is related to the feature point of the standard picture When the number of matches is greater than or equal to the second preset threshold, it is determined that the target advertisement has been played; or, if the number of matching points of the feature points of the cut frame picture and the feature points of the standard picture is less than the second preset threshold, Then it is determined that the target advertisement is for playing.

The above process illustrates the implementation of comparing the cut frame picture in the key time period with the preset standard picture by the picture comparison module 4, and comparing all the cut frame pictures in the critical time period with the standard picture, wherein The comparison is mainly the feature points in the picture, and the specific analysis and determination process by the feature points is the same as the specific analysis process in the third embodiment of the present application. For reference, the description in the foregoing embodiment 3 is omitted. In addition, the specific range of the second preset threshold is not limited, and those skilled in the art may arbitrarily set according to the needs of the advertiser, and details are not described herein again.

The image comparison module 4 compares the feature points of all the cut frame pictures in the key time period with the feature points of the standard picture, and then compares the matched number of the feature points with the second preset threshold. To determine whether the target advertisement is played, which effectively guarantees the accuracy of judging whether or not to play the target advertisement. In addition, the scope of the comparison is improved. In this application, all the cut frames in several key time periods are Compared with the standard picture, it effectively prevents the phenomenon of missing the advertisement and misjudges the phenomenon of not broadcasting the advertisement, and can effectively prevent the occurrence of the phenomenon that the advertisement is not found and the accuracy of the device is improved. Reliability is conducive to the promotion and application of the market.

Example 8

FIG. 5 is a schematic structural diagram of a video advertisement monitoring apparatus according to another embodiment. Referring to FIG. 5, the embodiment is a specific supplementary description of the foregoing embodiment, specifically, in the critical time period. Before the image is compared with the preset standard image, the video image cutting frame processing module 5 is configured to perform frame cutting processing on the standard video information according to the preset interval time to obtain the standard cut image. Frame picture.

It should be noted that the preset standard pictures mentioned in Embodiment 5, Embodiment 6, and Embodiment 7 are provided by the advertiser, that is, the advertiser provides standard comparison information of the target advertisement to be broadcast, wherein the standard The comparison information includes the standard picture information and the standard video information. When the advertiser provides the standard picture information, the cases involved in the embodiment 5, the embodiment 6 and the embodiment 7 can be passed through the embodiment 5 and the embodiment. 6 and the scheme in Embodiment 7 to determine whether the target advertisement is played.

When the video information provided by the advertiser is the case for the embodiment, the standard video information provided by the advertiser needs to be frame-cut, and the standard frame-cut image information corresponding to the standard video information is obtained. Then, the standard cut frame picture information is compared with the cut frame picture information in the key time period. For the specific comparison process, reference may be made to the description in the foregoing embodiment 7, and details are not described again.

In addition, when the duration of the standard video advertisement provided by the advertiser is different, the video image cut frame processing module 5 can also have different refinement functions according to specific actual situations.

Specifically, the inventor exemplifies the following two situations. One case is: when the target advertisement to be played by the advertiser is highly targeted, that is, the video playback time of the provided standard video advertisement information is relatively short. The specific short-term judgment criteria are different. Those skilled in the art can set according to the needs of the advertiser. For example, the short-term time can be set within 5 minutes or within 10 minutes, and then the All the cut frame images obtained by the video image cut frame processing module 5 after the standard video frame cutting process are used as the standard cut frame picture, and then all the standard cut frame pictures and the cut frame pictures in the key time period are performed one by one. For comparison, the specific comparison process can refer to the above statement, and will not be described again.

In another case, when the target advertisement to be played by the advertiser is not highly targeted, that is, when the video playback time of the provided standard video advertisement information is long, the time information corresponding to the above is corresponding, for example, If the video playing time exceeds 5 minutes or 10 minutes, the video playing time is longer. In this case, the video image cutting frame processing module 5 first performs the frame processing according to the preset interval time. After that, the obtained framing frame images may be sorted at a time point corresponding to the frame processing before the frame processing, and then the sorting sub-module included in the video image dicing frame processing module 5 may be according to a preset time period. The sorted cut frame pictures are grouped, and the key time period confirmation sub-module included in the video image cut frame processing module 5 performs image recognition on all the cut frame pictures in each group, and the grouped cut frame picture is obtained. Corresponding key time period, then compare all the standard cut frame images in the key segment with the cut frame image to get a conclusion on whether to play the target advertisement, wherein Key to confirm the above time period and a standard image frame with the tangential cut frame image matching process of Example 2 and Comparative confirmation process and the process in Example 3 is similar to the above-mentioned specific process may be described with reference to the content, which is not repeated herein.

When the standard advertisement comparison information provided by the advertiser is video advertisement information, the video image cutting frame processing module 5 may perform frame cutting processing on the video advertisement information to obtain standard cut frame image information, and then the standard cut frame image information and The frame image information is compared to obtain the conclusion whether the target advertisement is broadcasted. The video image cut frame processing module 5 effectively expands the use range of the device, improves the practicability of the device, and is beneficial to the promotion of the market. application.

It should be noted that the above embodiments are only used to explain the technical solutions of the present invention, and are not limited thereto; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand The technical solutions described in the foregoing embodiments may be modified, or some or all of the technical features may be equivalently replaced; and the modifications or substitutions do not deviate from the technical solutions of the embodiments of the present invention. The scope of the program.

1‧‧‧Video Image Acquisition Module

2‧‧‧Video Image Cut Frame Sorting Module

3‧‧‧Key time period confirmation module

4‧‧‧ Picture comparison module

5‧‧‧Video image cutting frame processing module

S1~s5, s41~s44‧‧‧ steps

FIG. 1 is a schematic flowchart diagram of a video advertisement monitoring and broadcasting method according to an embodiment of the present invention. FIG. 2 is a schematic flowchart diagram of a video advertisement monitoring and broadcasting method according to another embodiment of the present invention. FIG. 3 is a schematic flowchart diagram of a video advertisement monitoring and broadcasting method according to still another embodiment of the present invention. FIG. 4 is a schematic structural diagram of a video advertisement monitoring device according to an embodiment of the present invention. FIG. 5 is a schematic structural diagram of a video advertisement monitoring device according to another embodiment of the present invention.

S1~s4‧‧‧step

Claims (6)

A video advertisement monitoring and broadcasting method comprises: collecting video advertisement data played by an electronic screen in real time; performing frame cutting processing on the video advertisement data according to a preset interval time, obtaining a plurality of cut frame images, and according to the plurality of cut frame images according to the method Sorting the time points before the frame processing; grouping the sorted cut frame pictures according to the preset time period, performing image recognition on all the cut frame pictures in each group, and obtaining the key time corresponding to the grouped cut frame pictures Segmenting; and comparing the cut frame image in the key time period with a preset standard image to determine whether the target advertisement is played; wherein the image recognition is performed on all the cut frame images in each group to obtain a group The key time period corresponding to the subsequent cut frame picture includes: if the degree of change of the feature points of all the cut frames in the group after the group is less than or equal to the first preset threshold, the time period corresponding to the group is used as the key period. The video advertisement monitoring method according to claim 1, wherein the comparing the cut frame image in the critical time period with the preset standard image to determine whether the target advertisement is played, specifically including : comparing the feature points of all the cut frame pictures in the key time period with the feature points of the standard picture; if the feature points of the cut frame picture match the feature points of the standard picture are greater than or equal to the second Determining the threshold, determining that the target advertisement has been played; or determining that the target advertisement is not played if the number of matching points of the feature points of the cut frame picture and the feature points of the standard picture is less than the second preset threshold. The video advertisement monitoring method according to the first or the second aspect of the invention, wherein the comparing the image of the cut frame in the key time period with the preset standard image, the method further includes: The standard video data is subjected to frame cutting processing according to a preset interval time to obtain the standard picture. A video advertisement monitoring device comprises: a video image acquisition module, which is used for instantly collecting video advertisement data played by an electronic screen; and a video image cut frame sorting module, which is used for cutting video advertisement data according to a preset interval time. Frame processing, obtaining a plurality of cut frame pictures, and sorting the plurality of cut frame pictures according to time points before the frame cutting process; the key time period confirming module is configured to sort the sorted frames according to the preset time period The pictures are grouped, image recognition is performed on all the cut frame pictures in each group, and the key time period corresponding to the grouped cut frame picture is obtained; and the picture comparison module is configured to cut the frames in the key time period The image is compared with the preset standard image to determine whether the target advertisement is played, wherein the key time period confirmation module is specifically configured to: if all the cut frame picture feature points in the group are less than or equal to each other The first preset threshold is used as the key time period for the corresponding time period of the group. The video advertisement monitoring device of claim 4, wherein the picture comparison module is further configured to: perform feature points of all the cut frames in the key time period and feature points of the standard picture; One-to-one comparison; if the feature points of the cut frame image match the feature points of the standard picture When the second preset threshold is equal to or equal to the second preset threshold, it is determined that the target advertisement has been played; or, if the number of matching points of the feature point of the cut frame picture and the feature point of the standard picture is less than the second preset threshold, then it is determined The target ad is not played. The video advertisement monitoring apparatus according to the fourth or fifth aspect of the invention, wherein before the image comparison between the cut frame picture and the preset standard picture in the critical time period, the video image is further included: The frame cutting processing module is configured to perform frame cutting processing on the standard video information according to the preset interval time to obtain the standard picture.