阅读说明：本技术 一种智能设备视频广告处理方法以及智能设备 (Intelligent device video advertisement processing method and intelligent device ) 是由 肖必龙 蔡幸波 余刚 蔡海艇 祝可颖 于 2021-10-11 设计创作，主要内容包括：本发明提供了一种智能设备视频广告处理方法以及智能设备。智能设备视频广告处理方法包括：智能设备在运行视频应用时,周期性地检测视频应用中播放的视频内容；若检测到视频内容为视频广告时,智能设备进入视频广告预处理模式；预处理模式包括：在智能设备第一次播放视频广告,且在视频广告的广告总时长T时,智能设备在第一识别时长T-(1)内识别视频广告的视频信息,并且将视频信息记录在第一视频信息表中；在智能设备第二次播放视频广告的情况下,将视频信息记录在第二视频信息表中,且智能设备在第二识别时长T-(2)内对第二次视频广告进行识别。本发明能够解决在智能设备在运行视频广告过程,过滤效率较低的技术问题。(The invention provides an intelligent device video advertisement processing method and an intelligent device. The intelligent device video advertisement processing method comprises the following steps: the method comprises the steps that when the intelligent equipment runs a video application, video content played in the video application is periodically detected; if the video content is detected to be the video advertisement, the intelligent device enters a video advertisement preprocessing mode; the preprocessing mode comprises the following steps: playing the video advertisement for the first time on the intelligent equipment, and when the total advertisement duration T of the video advertisement is reached, the intelligent equipment is in the first identification duration T 1 Video information of the video advertisement is internally identified, and the video information is recorded in a first video information table; under the condition that the intelligent equipment plays the video advertisement for the second time, the video information is recorded in the second video information table, and the intelligent equipment has the second identification time length T 2 Inner pair second videoThe advertisement is identified. The invention can solve the technical problem of low filtering efficiency in the process of running the video advertisement by the intelligent equipment.)

1. A video advertisement processing method for intelligent equipment is characterized by comprising the following steps:

the method comprises the steps that when the intelligent equipment runs a video application, video content played in the video application is periodically detected;

if the video content is detected to be the video advertisement, the intelligent equipment enters a video advertisement preprocessing mode;

the pre-processing mode comprises: when the intelligent equipment plays the video advertisement for the first time and the total advertisement duration T of the video advertisement, the intelligent equipment has a first identification duration T₁Identifying video information of the video advertisement and recording the video information in a first video information table;

under the condition that the intelligent equipment plays the video advertisement for the second time, the video information is recorded in a second video information table, and the intelligent equipment has a second identification time length T₂Identifying the second video advertisement;

comparing the video information in the first video information table with the video information in the second video information table, and judging whether the intelligent equipment meets video advertisement processing conditions or not according to a comparison result;

when the intelligent equipment meets the video advertisement processing condition, the intelligent equipment enters a video advertisement processing mode;

wherein the second recognition duration T₂< the first recognition duration T₁< the total advertisement duration T.

2. The intelligent device video advertisement processing method according to claim 1, wherein the intelligent device meeting the video advertisement processing condition comprises:

the video information in the first video information table is the same as the video information in the second video information table.

3. The intelligent device video advertisement processing method according to claim 2, wherein the comparing the video information includes:

if the similarity between the video information in the first video information table and the video information in the second video information table is greater than or equal to a first preset value H₁If so, judging that the video information in the first video information table is the same as the video information in the second video information table;

if the similarity between the video information in the first video information table and the video information in the second video information table is less than or equal to a second preset value H₂And if so, judging that the video information in the first video information table is different from the video information in the second video information table.

4. The intelligent device video advertisement processing method according to claim 3, wherein the first preset value H is₁∈[75％，85％]The second preset value H₁∈[35％，45％]。

5. The smart device video advertisement processing method as claimed in claim 1, wherein the second recognition duration T is₂Satisfies the following conditions:

T₂=Asin[（T₁-T₀）π/T₁]+B；

wherein, T₀To correct the difference; a is the first duration correction value and B is the second duration correction value.

6. The smart device video advertisement processing method according to claim 5, wherein the first recognition duration T₁∈[30s，50s]Said corrected difference T₀∈[15s，25s]。

7. The intelligent device video advertisement processing method according to claim 1, wherein comparing the video information in the first video information table and the second video information table further comprises:

if the similarity between the video information in the first video information table and the video information in the second video information table is at a second preset value H₂With a first predetermined value H₁When the intelligent device is in the middle of the time, the intelligent device enters into identificationAnd (5) a duration correction mode.

8. The intelligent device video advertisement processing method of claim 7, wherein the identifying duration correction mode comprises:

the intelligent equipment continues to identify the video advertisements until the video information in the first video information table and the video information in the second video information table are the same or different, or a third identification duration T is met₃The same condition as the total advertisement time length T.

9. The intelligent device video advertisement processing method according to claim 8, wherein the identified duration correction mode satisfies:

T₃=T₂+NC；

wherein, T₃Is a third recognition duration, N is the number of times, C is a correction constant, and the third recognition duration T₃The total advertisement time length T is less than or equal to.

10. An intelligent device, comprising a computer-readable storage medium storing a computer program and a processor, wherein the computer program, when read by the processor, implements the intelligent device video advertisement processing method according to any one of claims 1 to 9.

Technical Field

The invention relates to the technical field of intelligent terminals, in particular to a video advertisement processing method of intelligent equipment and the intelligent equipment.

Background

When the intelligent device runs a video application, advertisements which are often inserted into a video film are usually inserted at the beginning or the end of the video film, the video advertisements are generally set to be not skipped over by non-members, and a user can watch the video film only after the video advertisements are played.

In the prior art, although there are filtering and processing methods for video advertisements in smart devices, a user often uses a conventional method that the user can shoot an advertisement image or record a segment of advertisement audio, and in the process of filtering and identifying the advertisement, the video advertisements can be processed in a unified manner, and the types of the video advertisements cannot be accurately selected, so that the efficiency of filtering the video advertisements is reduced, and the user experience is influenced.

Disclosure of Invention

The invention can solve the technical problem of low filtering efficiency in the process of running the video advertisement by the intelligent equipment.

In order to solve the above problem, an embodiment of the present invention provides a method for processing a video advertisement of an intelligent device, where the method for processing a video advertisement includes: the method comprises the steps that when the intelligent equipment runs a video application, video content played in the video application is periodically detected; if the video content is detected to be the video advertisement, the intelligent device enters a video advertisement preprocessing mode; the preprocessing mode comprises the following steps: playing the video advertisement for the first time on the intelligent equipment, and when the total advertisement duration T of the video advertisement is reached, the intelligent equipment is in the first identification duration T₁Video information of the video advertisement is internally identified, and the video information is recorded in a first video information table; under the condition that the intelligent equipment plays the video advertisement for the second time, the video information is recorded in the second timeIn the video information table, the intelligent equipment identifies the duration T at the second time₂Identifying the second video advertisement; comparing the video information in the first video information table with the video information in the second video information table, and judging whether the intelligent equipment meets the video advertisement processing conditions or not according to the comparison result; when the intelligent equipment meets the video advertisement processing condition, the intelligent equipment enters a video advertisement processing mode; wherein the second identification duration T₂< first recognition duration T₁< Total advertisement duration T.

Compared with the prior art, the technical scheme has the following technical effects: by periodically detecting the condition of the intelligent equipment running the video application, whether the intelligent equipment has the playing condition of the video advertisement can be known in real time, and the aims of accurately selecting and filtering the video advertisement can be further fulfilled; in a preprocessing mode, the video advertisement played for the first time is recorded in the first video information table, and the video advertisement played again is recorded in the second video information table, so that comparison can be formed between the video advertisement played for the first time and the video advertisement played again, the time spent on identifying the video advertisement in the process of playing the advertisement video again by the intelligent equipment is further shortened, and the filtering effect of the video advertisement is improved; the purpose of rapidly identifying the video advertisements can be achieved by comparing the video information in the first video information table with the video information in the second video information table, and the intelligent device can further achieve the purpose of accurately screening the video advertisement types by processing the video advertisements when meeting the video advertisement processing conditions.

Further, in one embodiment of the present invention, the video advertisement processing condition includes: the video information in the first video information table is the same as the video information in the second video information table.

Compared with the prior art, the technical scheme has the following technical effects: in the two-time video advertisements, if the video information screened twice is the same, the type of the two-time played video advertisements is the same or similar, and when the same video advertisement is played again, the intelligent device can rapidly identify and classify in a short time, so that the time spent on identifying the video advertisements is reduced, and the efficiency of screening the video advertisements by the intelligent device is further improved.

Further, in an embodiment of the present invention, comparing the video information in the first video information table and the second video information table includes: if the similarity between the video information in the first video information table and the video information in the second video information table is greater than or equal to a first preset value H₁If so, judging that the video information in the first video information table is the same as that in the second video information table; if the similarity between the video information in the first video information table and the video information in the second video information table is less than or equal to a second preset value H₂And if so, judging that the video information in the first video information table is different from the video information in the second video information table.

Compared with the prior art, the technical scheme has the following technical effects: when the video information in the first video information table and the second video information table is compared, if the similarity of the video information in the first video information table and the second video information table is greater than or equal to a first preset value H₁When the first video advertisement is similar to or close to the second video advertisement, the video contents of the first video advertisement and the second video advertisement can be summarized into the same category, and then the two video advertisements can be processed uniformly; similarly, if the similarity of the video information in the first video information table and the second video information table is less than or equal to the second preset value H₂When the video advertisements are played, the video contents in the first video advertisement and the second video advertisement are different, and the first video advertisement and the second video advertisement can be summarized into two different types, so that the video advertisements played twice can be processed differently; the intelligent equipment can judge whether the contents of the two video advertisements are the same or not according to the information such as characters, images, audio and the like appearing in the video advertisements, and sets a first preset value H as the conditions that the language characters are similar and the like possibly exist in the video information for identifying the first video advertisement and the second video advertisement are set₁And a second preset value H₂The similarity of the information such as characters, images and audios in the first video advertisement and the second video advertisement is judged, and the similarity is compared with the first video advertisementSet value H₁And a second preset value H₂The relationship between the two video advertisements obtains the conclusion that whether the two played video advertisements are the same or different, and finally the video advertisements can be processed by setting a first preset value H₁And a second preset value H₂The identification efficiency of the intelligent device during advertisement playing is further improved.

Further, in one embodiment of the present invention, the first preset value H₁∈[75％，85％]Second preset value H₁∈[35％，45％]。

Compared with the prior art, the technical scheme has the following technical effects: according to the comparison of the similarity of the video information of the video advertisements played twice, the purposes of screening and filtering the video advertisements are improved, and the similarity is further greater than or equal to a first preset value H₁Under the condition of (1), and setting the first preset value H₁The setting is between 75% and 85%, so that the intelligent equipment can accurately judge whether the two video advertisements belong to the same type; likewise, the second preset value H₂The setting is 35% -45%, so that the intelligent equipment can also accurately judge whether the two video advertisements belong to different types; first preset value H₁And a second preset value H₂The range is set, and the purposes of accurately selecting and filtering the video advertisements are achieved.

Further, in one embodiment of the present invention, the second recognition duration T₂Satisfies the following conditions: t is₂=Asin[（T₁-T₀）π/T₁]+ B; wherein, T₀To correct the difference; a is the first duration correction value and B is the second duration correction value.

Compared with the prior art, the technical scheme has the following technical effects: if the video advertisements played twice by the intelligent equipment are the same, when the same video advertisement is played again, correcting the difference value T in the formula₀And a first recognition duration T₁The identification time length spent by the same video advertisement after the video advertisement is played for the second time can be obtained, namely the identification time length is obtained through the formula T₂=Asin[（T₁-T₀）π/T₁]+ B the second recognition duration T₂Intelligent deviceCan be in the second recognition duration T₂Identifying the video advertisements quickly; by setting the correction difference T₀The time error caused when the video advertisement is identified for the second time is reduced; in order to further improve the identification efficiency of the video advertisements, the set correction difference value T₀Satisfies the following conditions: correcting the difference T₀= total advertisement duration T-first recognition duration T₁According to the total advertisement time length T and the first identification time length T₁Is determined to correct the difference value T₀By correcting the difference T₀And a first recognition duration T₁Calculating to obtain a second identification duration T₂A value of (d); by the second recognition duration T in the sine formula₂The time spent on identifying the video advertisement for the second time under the same video advertisement is obtained through the calculation of (1); it will be appreciated that the first recognition duration T₁And a second recognition duration T₂Is related to the total advertisement time length T, i.e. the longer the total advertisement time length T is, the first recognition time length T₁And a second recognition duration T₂The greater the value of (A), i.e. the total advertisement duration T and the first recognition duration T₁In an increasing functional relationship, and further, in the above formula, by pairing sin [ (T)₁-T₀）π/T₁]The limitation of the value range realizes the first identification duration T₁Based on the second recognition duration T₂The purpose of to promoted the smart machine to the filtration efficiency of video advertisement when playing the video for the second time.

Further, in one embodiment of the present invention, the first recognition duration T₁∈[30s，50s]Correcting the difference T₀∈[15s，25s]。

Compared with the prior art, the technical scheme has the following technical effects: by comparing the first identification time length T and correcting the difference value T₀The time spent in identifying the video advertisements is further reduced by limiting the value range of (a).

Further, in an embodiment of the present invention, comparing the video information in the first video information table and the second video information table further includes: if the video information in the first video information table and the second video information tableThe similarity between them is at the second preset value H₂With a first predetermined value H₁And in the meantime, the intelligent equipment enters an identification duration correction mode.

Compared with the prior art, the technical scheme has the following technical effects: in the comparison of the two video advertisements, if the comparison is within the second identification duration T₂In the method, the video information of the two times cannot be judged whether the video information is the same or different, that is, the similarity of the video information in the two video advertisements is at a second preset value H₂With a first predetermined value H₁Meanwhile, the intelligent equipment enters the identification duration correction mode, and after the energy equipment enters the identification duration correction mode, the second identification duration T is set₂The intelligent device can accurately realize the purposes of filtering and screening the video advertisement types.

Further, in an embodiment of the present invention, identifying the duration correction pattern includes: the intelligent equipment continues to identify the video advertisements until the video information in the first video information table and the video information in the second video information table are the same or different, or the third identification duration T is met₃The same condition as the total advertisement time length T.

Compared with the prior art, the technical scheme has the following technical effects: it should be noted that, if in the video advertisement processing mode, if in the second recognition duration T₂In the method, when the similarity of the two times of video information cannot be determined, if the similarity of the two times of video information can reach the same condition or different conditions, the two times of video advertisements are summarized into the same category or different categories, and if the similarity of the two times of video information can reach the same condition or different conditions, the second identification duration T is the second identification duration₂In the internal identification, if the similarity of the two video information cannot be determined, entering an identification duration correction mode, and further identifying the two video advertisements; by identifying the setting of the time length correction mode, the efficiency of identifying and processing the video advertisements is further improved.

Further, in an embodiment of the present invention, the identified duration correction pattern satisfies: t is₃=T₂+ NC; wherein, T₃Is a third recognition duration, N is the number of times, C is a correction constant, and the third recognition duration T₃The total advertisement time length T is less than or equal to.

Compared with the prior art, the technical scheme has the following technical effects: it will be appreciated that in the two comparisons of the video information, at the second recognition duration T₂In the method, the similarity between the first video advertisement and the second video advertisement is uncertain, that is, the similarity of the video information of the two video advertisements is at a second preset value H₂With a first predetermined value H₁In between, the second identification duration T is needed at this time₂Extended and extended to the third recognition duration T₃Further identifying the video advertisements of the two times; since the total advertisement duration of the video advertisement is T, the duration T is identified at the second time₂In the course of extension, the third recognition duration T₃To satisfy T₃T is less than or equal to T; if the time is in the third recognition duration T₃If the similarity of the two video advertisements still cannot be identified, the two video advertisements are summarized into two different types; further by the formula: t is₃=T₂+ NC, enabling a third recognition duration T₃And finally, screening the video advertisements is completed, and the accuracy of video advertisement identification is further improved.

Further, an embodiment of the present invention further provides an intelligent device, where the intelligent device includes a computer-readable storage medium storing a computer program and a processor, and when the computer program is read by the processor, the method for processing the video advertisement of the intelligent device in the foregoing embodiment is implemented.

Compared with the prior art, the technical scheme has the following technical effects: when the intelligent device implements the intelligent device video advertisement processing method in the above embodiment, all the beneficial effects brought by the implementation of the intelligent device video advertisement processing method in the above embodiment are achieved, and details are not repeated here.

In summary, after the technical scheme of the invention is adopted, the following technical effects can be achieved:

i) by periodically detecting the condition of the intelligent equipment running the video application, whether the intelligent equipment has the playing condition of the video advertisement can be known in real time, and the aims of accurately selecting and filtering the video advertisement can be further fulfilled; through a preprocessing mode, a video advertisement played for the first time is recorded in a first video information table, and a video advertisement played again is recorded in a second video information table, so that comparison can be formed between the video advertisement played for the first time and the video advertisement played again, the time spent on identifying the video advertisement in the process of playing the advertisement video again by the intelligent equipment is further shortened, and the filtering effect of the video advertisement is improved;

ii) the purpose of rapidly identifying the video advertisements can be realized by comparing the video information in the first video information table with the video information in the second video information table, and the purpose of accurately screening the video advertisement types can be further achieved by processing the video advertisements when the intelligent equipment meets the video advertisement processing conditions;

iii) when the video advertisements played twice by the intelligent device are the same, the formula T is used₂=Asin[（T₁-T₀）π/T₁]+ B the second recognition duration T₂The intelligent device can identify the duration T at the second time₂Identifying the video advertisements quickly; by setting the correction difference T₀The time error caused when the video advertisement is identified for the second time is reduced; further, through the above formula, the first recognition duration T can be realized₁Based on the second recognition duration T₂The purpose of the method is that the filtering efficiency of the intelligent equipment on the video advertisements is improved when the video is played for the second time;

iv) by the formula: t is₃=T₂+ NC for second recognition duration T₂The identification duration is prolonged, and finally the intelligent equipment can complete the screening of the video advertisements, so that the accuracy of video advertisement identification is further improved.

Description of the drawings:

fig. 1 is a schematic diagram of a method for processing a video advertisement of an intelligent device according to an embodiment of the present invention.

Fig. 2 is a flowchart of a video advertisement processing method for an intelligent device.

Detailed Description

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.

[ first embodiment ] A method for manufacturing a semiconductor device

Referring to fig. 1, an embodiment of the present invention provides a method for processing a video advertisement of an intelligent device, where the method for processing a video advertisement of an intelligent device includes:

s10: the method comprises the steps that when the intelligent equipment runs a video application, video content played in the video application is periodically detected;

s20: if the video content is detected to be the video advertisement, the intelligent device enters a video advertisement preprocessing mode;

s30: under the condition that the intelligent equipment plays the video advertisement for the second time, the video information is recorded in the second video information table, and the intelligent equipment has the second identification time length T₂Identifying the second video advertisement;

s40: comparing the video information in the first video information table with the video information in the second video information table, and judging whether the intelligent equipment meets the video advertisement processing conditions or not according to the comparison result;

s50: and when the intelligent equipment meets the video advertisement processing condition, the intelligent equipment enters a video advertisement processing mode.

It can be understood that, when a user watches videos through some intelligent devices such as a computer and a smart phone, the user often inserts advertisements after the videos are finished and the videos are started, the total duration of some advertisements can reach about one minute to two minutes, which brings more inconvenience to the user, and the advertisements are usually set to be that a non-member user cannot skip, the non-member user can only continue to watch video prints after the advertisements are played, so that the use efficiency of the user using the intelligent devices is influenced.

In the embodiment of the invention, different video advertisements are identified and classified by detecting the video playing condition of the intelligent equipment in the operation process, so that the use efficiency of the intelligent equipment used by a user is improved.

Preferably, in S10, when the user is using the smart device, if the smart device is running the video application, the video content played in the video application is periodically detected, and the advertisement played by the smart device is correspondingly processed according to the detected video information, so as to improve the experience of the user using the smart device; in the process of periodically detecting the video played by the intelligent device, a corresponding detection period can be set according to the video feature watched by the user.

For example, when the user is watching a television show, the detection period can be set to 40-50 min; if the user watches the film, setting the detection period to be 90-120 min; if the user watches some cartoon works, the detection period of the video can be set to be 15-30 min; of course, the user may also set other detection periods, specifically, the detection period may be determined according to the video feature played in the video application of the smart device and the actual requirements of the user, and is not limited to the limitation on the detection period in this embodiment.

Preferably, in S20, in the detection process, if it is detected that the smart device has an advertisement playing, the smart device enters a video advertisement preprocessing mode, specifically, in the preprocessing mode: the video advertisement is played for the first time, and when the total advertisement duration T of the video advertisement is reached, the intelligent equipment is in the first recognition duration T₁Identifying video advertisements and recording the identified video information in a first video information table; it will be appreciated that the first identified duration of an identified video advertisement is T, when the total duration of the advertisement is T₁Here, the first recognition duration T should be satisfied₁The total advertisement duration T is less than the total advertisement duration T, that is, the intelligent device needs to complete the identification of the video advertisement before the video advertisement is completely played.

It should be noted that, when the intelligent device identifies the video information, the video information includes: the intelligent device records the information such as the image, the characters and the voice in the video advertisement in the first video information table, and certainly, when the video information is recorded, the form of extracting keywords and key sentences can be adopted, or the whole sentence words and the whole video can be stored.

Furthermore, after the intelligent device records the identified video information in the first video information table, the intelligent device can classify the video advertisements in the first video information table, for example, the video advertisements can be classified into two types; wherein, if the video advertisement is a positive energy advertisement, the video advertisement is selected to class a, and if the video advertisement is a negative energy, the video advertisement is selected to class B, for example, the class a video advertisement includes: public welfare, legal promotions, etc., and class B video advertisements include malicious promotions, phishing, etc.

Preferably, referring to fig. 2, in S30, in the case that the smart device plays the video advertisement for the second time, the video information is recorded in the second video information table, and the smart device is operated for the second recognition duration T₂The second video advertisement is identified, and the purposes of accurately filtering and selecting the video advertisement can be achieved by comparing the two video advertisements.

Specifically, in S40, the second recognition duration T₂< first recognition duration T₁(ii) a The video advertisements can be rapidly identified by comparing the second video advertisements in the second video information table with the first video advertisements in the first video information table, and in the comparison process of the first video information table and the second video information table, the video advertisements are processed differently according to the similarity of the first video advertisements and the video information in the second video advertisements.

Specifically, in the comparison of the first video advertisement and the second video advertisement, the following requirements are met:

s41: if the similarity of the video information in the first video information table and the second video information table is greater than or equal to a first preset value H₁If so, judging that the video information in the first video information table is the same as that in the second video information table;

s42: if the similarity of the video information in the first video information table and the second video information table is less than or equal to a second preset value H₂And if so, judging that the video information in the first video information table is different from the video information in the second video information table.

Preferably, in S41, if the first view isThe similarity of the video information in the two video advertisements in the video information table and the second video information table is more than or equal to a first preset value H₁And if so, indicating that the video information of the two times is the same, and further classifying the second video advertisement and the first video advertisement into the same category.

It should be noted that, in the comparison of the similarity between the first video advertisement and the second video advertisement, the compared information includes: the information such as images, characters, voice and the like of the advertisement video can be information of keywords and key sentences appearing in the video advertisement, or homophones, homophones and the like; as long as in the comparison process, the similarity of the two video information is greater than the first preset value H₁Then, it can be determined that the two video advertisements are identical.

It can be understood that the advertisement contents of the two video advertisements may be the same or similar, and in the identification of the two video advertisements, the synonym and the synonym are considered to exist in the two video advertisements, so that in the comparison of the two video advertisements, the set similarity is greater than or equal to the first preset value H₁Then, determining that the video information of the two times is the same; it should be noted that the first preset value H set here₁∈[75％，85％]Wherein H is₁May be selected to be 80%, i.e. for a second recognition duration T₂In the method, if the content similarity of the two video advertisements reaches 80%, the content of the two video advertisements is judged to be the same, and the two same video advertisements can be summarized into the same type.

Similarly, in S42, the advertisement contents of the two video advertisements may be different or similar, and in the identification of the two video advertisements, the contents of synonyms and synonyms are considered, so that in the comparison of the two video advertisements, the set similarity is not more than the second preset value H₂Then, it is determined that the two times of video information are different; it should be noted that the second preset value H is set here₁∈[35％，45％]Wherein H is₂May be selected to be 40%, i.e. during the second recognition period T₂In addition, if the content similarity of the two video advertisements is about 40%, then the content similarity is determinedAnd judging that the contents of the two video advertisements are different.

Preferably, in S50, to further implement the screening and identification of the video advertisements, when the video information in the first video information table is the same as the video information in the second video information table, the smart device satisfies the video advertisement processing condition, and further, the user may perform corresponding processing according to the classified video advertisements, that is, when the smart device enters the video advertisement processing mode, after the smart device enters the video advertisement processing mode, different processing may be performed on different types of video advertisements, for example, the type a video advertisements with positive energy, for example, the type a video advertisements include: public welfare activities, legal propaganda and the like, and processing such as downloading, propaganda, sharing and the like is carried out; for video advertisements with negative energy in class B, for example, class B video advertisements include: malicious promotion, phishing and the like, and the processes of shielding, deleting and even reporting to relevant departments are carried out; and in the video advertisement processing mode, the second identification duration T₂Satisfies the following conditions: t is₂=Asin[（T₁-T₀）π/T₁]+B；

T₀To correct for differences, and T₀Satisfy T₀=T-T₁T is the total advertisement duration, T₁A first recognition duration; a is a first duration correction value and is a constant in units of s, a =10s, B is a second duration correction value and is a constant in units of s, B =5 s; first recognition duration T₁∈[30s，50s]Wherein, T₁The time can be selected to be 40 s; correcting the difference T₀∈[15s，25s]Wherein T is₀15s may be selected.

For example, the total advertisement time T is selected to be 60s, T₁At 40s, the second recognition duration T₂=10sin[（40-20）π/40]+5=15s, i.e. can be understood as: the total time length of the video advertisements is 60s, the first identification time length spent by the intelligent equipment in identifying the video information is 40s, when the same video advertisements are played again, the intelligent equipment can identify the video advertisements within 15s, and the screening and the filtering of the video advertisements are further completed; it will be appreciated that the first recognition duration T₁And a second recognition duration T₂Size and advertisement ofThe total duration T being related, i.e. the longer the total duration T of the advertisement, the first recognition duration T₁And a second recognition duration T₂The larger the value of (a), sin [ (T) in the above sine formula₁-T₀）π/T₁]Is in the range of [0, pi/2 ]]I.e. the first recognition duration T₁And the total advertisement time length T satisfies: 1.5T₁≤T≤2T₁I.e. when the first recognition duration T₁At 40s, the total advertisement time length T is 60s, 80s]When the first identification duration T is₁50s, the total advertisement time length T is 75s, 100s]Meanwhile, it should be noted that, in order to improve the identification efficiency of the video advertisement, when the first identification duration T is longer₁In [30s, 50s ]]Meanwhile, the total advertisement time length T of the video advertisements identified by the intelligent device is 45s, 100s]Meanwhile, the total advertisement time length T of the video advertisement is limited within 100s in the embodiment; when the total advertisement time length T is less than 45s, the corresponding adjustable first identification time length T₁As long as 1.5T is satisfied₁≤T≤2T₁Then the method is finished; of course, if the total advertisement time period T is between 10s and 20s, the first recognition time period T may be set in consideration of the fact that the total video advertisement time period T is short₁And a second recognition duration T₂The same as the total advertisement duration T.

It should be noted that in S42, when the video information in the first video information table is different from the video information in the second video information table, that is, the similarity between the first video advertisement and the second video advertisement is low, at this time, the intelligent device does not satisfy the video advertisement processing condition, and the intelligent device reenters the video advertisement preprocessing mode, so as to classify the second video advertisement.

Preferably, the comparing, by the intelligent device, the video information in the first video information table and the video information in the second video information table further includes:

s43: if the similarity of the video information in the first video information table and the second video information table is at a second preset value H₂With a first predetermined value H₁And in the meantime, the intelligent equipment enters an identification duration correction mode.

It will be appreciated that in the comparison of two video advertisements, if at the second identified duration T₂Two-time video informationCan not judge whether the video information is the same or different, namely the video information similarity in the two video advertisements is at a second preset value H₂With a first predetermined value H₁In between, smart machine entering discernment duration correction mode this moment, and discernment duration correction mode includes:

s431: the intelligent equipment continues to identify the video advertisements until the video information in the first video information table and the video information in the second video information table are the same or different, or the third identification duration T is met₃The same condition as the total advertisement time length T.

It is understood that, in the entering of the smart device into the recognition duration correction mode, the third recognition duration T is set₃The duration correction is performed on the video information of the two video advertisements, and the purpose of screening the video information in the second video advertisement is achieved sequentially, wherein the first preset value H is described here₁∈[75％，85％]Wherein H is₁80% can be selected, and the second preset value H₁∈[35％，45％]Wherein H is₂40% may be selected.

Preferably, in the two comparisons of the video information, there is a second recognition duration T₂In other words, the similarity of the video information of the two video advertisements is within a second preset value H₂With a first predetermined value H₁In between, need to discern the second video advertisement again, promptly this moment discerns duration correction mode and should satisfy: t is₃=T₂+ NC; wherein, T₃Is the third identification time length, N is the correction times, C is the correction value of the third time length and is a constant, and C belongs to [4s, 6s ]]Third recognition duration T₃Satisfies the following conditions: third recognition duration T₃The total advertisement time length T is less than or equal to.

For example, when the total advertisement duration T of a video advertisement is 60s, T₂When the selection time is 15s, the C selection time is 5s, if the similarity of the two video advertisements is difficult to determine by the intelligent device within 15s, the identification time of the two video advertisements needs to be prolonged to identify the second video advertisement again, namely, the third identification time T₃=15s +1 × 5s =20s, which is understood to mean the recognition of the second video advertisementThe difference time is from the original second identification time T₂(15 s) extension to a third recognition duration T₃(20 s) continuing the identification; of course, the similarity of the video information of the two video advertisements is still at the second preset value H within 20s₂With a first predetermined value H₁In between, the second identification duration T is continued₂Is extended, i.e. T is₃=15s +2 × 5s =25s, and so on until the condition that the video information in the first video information table and the video information in the second video information table are the same or different is satisfied, or the third recognition duration T is satisfied₃The condition that the total time length T of the advertisement is the same; it will be appreciated that the third recognition duration T is due to₃Cannot be greater than the video duration T of the video advertisement, so the duration T is identified for the second time₂In the process of prolonging, the third identification duration T can be satisfied₃The same condition as the total advertisement time length T when the third identification time length T₃If the total time length T of the advertisements is equal, if the similarity of the video information of the two video advertisements is still at a second preset value H₂With a first predetermined value H₁In between, it can be determined that the video information of the two times is different, and the video advertisements of the two times are summarized into two different types.

[ second embodiment ]

A second embodiment of the present invention provides an intelligent device, where the intelligent device includes a computer-readable storage medium storing a computer program and a processor, and when the computer program is read by the processor, the method for processing a video advertisement of the intelligent device in the foregoing embodiments can be implemented.

When the intelligent device implements the intelligent device video advertisement processing method in the above embodiment, all the beneficial effects brought by the implementation of the intelligent device video advertisement processing method in the above embodiment are achieved, and details are not repeated here.

Although the present invention is disclosed above, the present invention is not limited thereto. Various changes and modifications may be effected therein by one skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.