CN112330371A - AI-based intelligent advertisement pushing method, device, system and storage medium - Google Patents
AI-based intelligent advertisement pushing method, device, system and storage medium Download PDFInfo
- Publication number
- CN112330371A CN112330371A CN202011351491.6A CN202011351491A CN112330371A CN 112330371 A CN112330371 A CN 112330371A CN 202011351491 A CN202011351491 A CN 202011351491A CN 112330371 A CN112330371 A CN 112330371A
- Authority
- CN
- China
- Prior art keywords
- program
- advertisement
- program scene
- scene recognition
- screen
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 230000008569 process Effects 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 5
- 230000003631 expected effect Effects 0.000 abstract description 10
- 238000013473 artificial intelligence Methods 0.000 description 62
- 238000012545 processing Methods 0.000 description 10
- ZYXYTGQFPZEUFX-UHFFFAOYSA-N benzpyrimoxan Chemical compound O1C(OCCC1)C=1C(=NC=NC=1)OCC1=CC=C(C=C1)C(F)(F)F ZYXYTGQFPZEUFX-UHFFFAOYSA-N 0.000 description 8
- 238000002716 delivery method Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000010079 rubber tapping Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0242—Determining effectiveness of advertisements
- G06Q30/0244—Optimization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Abstract
The invention discloses an AI-based intelligent advertisement pushing method, an AI-based intelligent advertisement pushing device, an AI-based intelligent advertisement pushing system and a storage medium. The method comprises the following steps: acquiring screen information after the terminal is started; according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result; acquiring a corresponding advertisement according to the program scene identification result; and playing the corresponding advertisement on the screen. The invention solves the problems that the advertisement is not in accordance with the requirements of consumers and the expected effect is not achieved, and realizes the intelligent pushing of the advertisement.
Description
Technical Field
The invention relates to the field of Artificial Intelligence (AI), in particular to an AI-based intelligent advertisement pushing method, an AI-based intelligent advertisement pushing device, an AI-based intelligent advertisement pushing system and a storage medium.
Background
With the improvement of the consumption ability of people, the advertisement pushing mode is more and more abundant, wherein the mode of pushing the advertisement on the screen end of a consumer is most favored by advertisers, and the mode often appears in the forms of screen-opening advertisement and patch advertisement, which is the most direct mode of touching the advertisement message to the user side. However, the traditional screen advertisement push mostly adopts a blind advertisement mode, and the product or service of the advertisement is easily inconsistent with the demand of the consumer, so the consumer often chooses to finish the advertisement play in advance, further possibly chooses complaints after receiving uninteresting advertisements for many times, the brand value is reduced, and the advertisement is not in the expected effect.
Disclosure of Invention
The invention mainly aims to provide an AI-based intelligent advertisement pushing method, an AI-based intelligent advertisement pushing device, an AI-based intelligent advertisement pushing system and a storage medium, and aims to solve the problems that advertisements are inconsistent with the requirements of consumers and cannot achieve the expected effect and realize the intelligent advertisement pushing.
In order to achieve the above object, the present invention provides an AI-based intelligent advertisement delivery method, which is applied to an AI-based intelligent advertisement delivery system, and comprises the following steps:
acquiring screen information after the terminal is started;
according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result;
acquiring a corresponding advertisement according to the program scene identification result;
and playing the corresponding advertisement on the screen.
Optionally, the step of obtaining the screen information after the terminal is turned on further includes:
establishing a program scene recognition model through an AI technology;
collecting relevant pictures and/or audio of historical programs watched by a user, and obtaining corresponding program scene information;
and inputting the relevant pictures and/or audios of the historical programs and the corresponding program scene information into the program scene recognition model, and training the program scene recognition model to obtain a trained program scene recognition model.
Optionally, the step of obtaining the screen information after the terminal is turned on includes:
when the opening of a terminal screen is detected or in the process of playing programs by the terminal is detected, screen information on the terminal screen is acquired according to a preset frequency.
Optionally, the screen information includes: the step of obtaining a program scene recognition result by combining the trained program scene recognition model according to the screen information comprises the following steps:
and inputting the picture and/or the audio into the trained program scene recognition model in a data stream format, and recognizing the picture and/or the audio by the program scene recognition model according to an object and/or an identifier appearing in the picture or according to a voiceprint in the audio to obtain a program scene recognition result.
Optionally, the program scene recognition result includes: one or more of program scene information and a weight corresponding to the program scene information.
Optionally, the step of obtaining the corresponding advertisement according to the program scene identification result includes:
screening the program scene information and the weights, and obtaining the distribution condition of the program scenes when N continuous program scene information are consistent with the corresponding weights, wherein N is an integer and is more than or equal to 3;
and obtaining corresponding advertisements according to the distribution situation of the program scenes.
Optionally, the step of obtaining the corresponding advertisement according to the distribution situation of the program scene includes:
according to the distribution situation of the program scenes, counting the corresponding watching duration of the distribution situation of the program scenes;
sequencing the watching duration, and synchronizing the sequenced watching duration and the distribution situation of the program scene to a cloud server;
and receiving the advertisement fed back by the cloud server according to the sequenced watching duration and the program scene distribution condition, and pushing the advertisement to a terminal user.
In order to achieve the above object, the present invention further provides an AI-based intelligent advertisement delivery apparatus, including:
the screen capture/recording module is used for acquiring screen information after the terminal is started and sending the screen information to the model operation module;
the model operation module is used for obtaining a program scene recognition result by combining the trained program scene recognition model according to the screen information and sending the program scene recognition result to the advertisement pushing module;
the advertisement pushing module is used for obtaining corresponding advertisements according to the program scene recognition result and sending the corresponding advertisements to the advertisement playing module;
and the advertisement playing module is used for playing the corresponding advertisement on the screen.
In addition, to achieve the above object, the present invention further provides an AI-based intelligent advertisement delivery system, which includes: the system comprises a memory, a processor and an AI-based smart advertisement push program stored on the memory and operable on the processor, wherein the AI-based smart advertisement push program realizes the steps of the AI-based smart advertisement push method when the processor executes the AI-based smart advertisement push program.
In addition, to achieve the above object, the present invention further provides a storage medium having an AI-based smart advertisement push program stored thereon, wherein the AI-based smart advertisement push program, when executed by a processor, implements the steps of the AI-based smart advertisement push method as described above.
The invention provides an AI-based intelligent advertisement pushing method, an AI-based intelligent advertisement pushing device, an AI-based intelligent advertisement pushing system and a storage medium. In the above way, the screen information after the terminal is started is obtained; according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result; acquiring a corresponding advertisement according to the program scene identification result; and playing the corresponding advertisement on the screen. Through the mode, the intelligent advertisement pushing system and the intelligent advertisement pushing method solve the problems that the advertisement is not in accordance with the requirements of consumers and the expected effect is not achieved, and realize intelligent advertisement pushing.
Drawings
FIG. 1 is a system diagram of a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a first embodiment of an AI-based smart advertisement delivery method according to the present invention;
FIG. 3 is a schematic diagram of an AI-based smart advertisement delivery apparatus according to the present invention;
FIG. 4 is a flowchart illustrating a second embodiment of the AI-based smart advertisement delivery method according to the present invention;
fig. 5 is a flowchart illustrating a third embodiment of the AI-based smart advertisement delivery method according to the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The main solution of the embodiment of the invention is as follows: acquiring screen information after the terminal is started; according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result; acquiring a corresponding advertisement according to the program scene identification result; and playing the corresponding advertisement on the screen.
The screen advertisement push of the existing television and other terminals mostly adopts a blind advertisement mode, products or services of advertisements are easy to be inconsistent with the requirements of consumers, the consumers often choose to finish the advertisement play in advance, further, the consumers may choose to complain after receiving uninteresting advertisements for many times, the brand value is reduced, and meanwhile, the advertisements do not achieve the expected effect.
The invention provides a solution, which solves the problems that the advertisement is not in accordance with the requirements of consumers and the expected effect is not achieved, and realizes the intelligent pushing of the advertisement.
As shown in fig. 1, fig. 1 is a system structural diagram of a hardware operating environment according to an embodiment of the present invention.
The system of the embodiment of the invention can be a PC, and can also be a mobile system device with a display function, such as a smart phone, a tablet computer and the like.
As shown in fig. 1, the system may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a DisPlay screen (displayability), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., a magnetic disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.
Preferably, the system may further include a camera, RF (Radio Frequency) circuitry, sensors, audio circuitry, a WiFi module, and the like. Such as light sensors, motion sensors, and other sensors. In particular, the light sensor may include an ambient light sensor that adjusts the brightness of the display screen based on the intensity of ambient light, and a proximity sensor that turns off the display screen and/or backlight when the mobile system is moved to the ear. As one of the motion sensors, the gravity acceleration sensor can detect the magnitude of acceleration in each direction (generally, three axes), detect the magnitude and direction of gravity when the mobile system is stationary, and can be used for applications of recognizing the posture of the mobile system (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer and tapping), and the like; of course, the mobile system may also be configured with other sensors such as a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, etc., which are not described herein again.
Those skilled in the art will appreciate that the system architecture shown in FIG. 1 is not intended to be limiting of the system, and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a kind of computer storage medium, may include therein an operating system, a network communication module, a user interface module, and an AI-based smart advertisement push program.
In the system shown in fig. 1, the network interface 1004 is mainly used for connecting to a backend server and communicating with the backend server; the user interface 1003 is mainly used for connecting a client (user side) and performing data communication with the client; and the processor 1001 may be configured to call the AI-based smart advertisement push program stored in the memory 1005 and perform the following operations:
acquiring screen information after the terminal is started;
according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result;
acquiring a corresponding advertisement according to the program scene identification result;
and playing the corresponding advertisement on the screen.
Further, the processor 1001 may call the AI-based smart advertisement push program stored in the memory 1005, and further perform the following operations:
establishing a program scene recognition model through an AI technology;
collecting relevant pictures and/or audio of historical programs watched by a user, and obtaining corresponding program scene information;
and inputting the relevant pictures and/or audios of the historical programs and the corresponding program scene information into the program scene recognition model, and training the program scene recognition model to obtain a trained program scene recognition model.
Further, the processor 1001 may call the AI-based smart advertisement push program stored in the memory 1005, and further perform the following operations:
when the opening of a terminal screen is detected or in the process of playing programs by the terminal is detected, screen information on the terminal screen is acquired according to a preset frequency.
Further, the processor 1001 may call the AI-based smart advertisement push program stored in the memory 1005, and further perform the following operations:
and inputting the picture and/or the audio into the trained program scene recognition model in a data stream format, and recognizing the picture and/or the audio by the program scene recognition model according to an object and/or an identifier appearing in the picture or according to a voiceprint in the audio to obtain a program scene recognition result.
Further, the processor 1001 may call the AI-based smart advertisement push program stored in the memory 1005, and further perform the following operations:
the program scene recognition result comprises: one or more of program scene information and a weight corresponding to the program scene information.
Further, the processor 1001 may call the AI-based smart advertisement push program stored in the memory 1005, and further perform the following operations:
screening the program scene information and the weights, and obtaining the distribution condition of the program scenes when N continuous program scene information are consistent with the corresponding weights, wherein N is an integer and is more than or equal to 3;
and obtaining corresponding advertisements according to the distribution situation of the program scenes.
Further, the processor 1001 may call the AI-based smart advertisement push program stored in the memory 1005, and further perform the following operations:
according to the distribution situation of the program scenes, counting the corresponding watching duration of the distribution situation of the program scenes;
sequencing the watching duration, and synchronizing the sequenced watching duration and the distribution situation of the program scene to a cloud server;
and receiving the advertisement fed back by the cloud server according to the sequenced watching duration and the program scene distribution condition, and pushing the advertisement to a terminal user.
The invention provides an AI-based intelligent advertisement pushing method, an AI-based intelligent advertisement pushing device, an AI-based intelligent advertisement pushing system and a storage medium. In the above way, the screen information after the terminal is started is obtained; according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result; acquiring a corresponding advertisement according to the program scene identification result; and playing the corresponding advertisement on the screen. Through the mode, the intelligent advertisement pushing system and the intelligent advertisement pushing method solve the problems that the advertisement is not in accordance with the requirements of consumers and the expected effect is not achieved, and realize intelligent advertisement pushing.
Based on the above hardware structure, the embodiment of the method of the present invention is provided.
Referring to fig. 2, fig. 2 is a schematic flowchart of a first embodiment of an AI-based intelligent advertisement delivery method according to the present invention, and with reference to fig. 2, the method includes:
and step S10, acquiring screen information after the terminal is started.
As an implementation manner, in this embodiment, when the terminal screen is detected to be opened or during the terminal is detected to play a program, the screen information on the terminal screen is obtained according to a preset frequency, and in this embodiment, taking the frequency of 300 ms/time as an example, the screen information includes: pictures and audio.
Specifically, as an embodiment, in this embodiment, with reference to fig. 3, fig. 3 is a schematic diagram of an AI-based intelligent advertisement delivery apparatus according to the present invention, and the apparatus includes the following modules: host system, screen capture/recording module, model operation module, advertisement propelling movement module and advertisement broadcast module, wherein advertisement propelling movement module includes: the system comprises an identification result processing unit, a background advertisement management unit, an advertisement control playing unit and a background advertisement server.
Furthermore, the main control module has the functions of coordinately controlling the screen capturing/recording module, the model operation module, the recognition result processing unit and the advertisement control playing unit. When detecting that the user opens electronic terminal playing equipment such as a television, a Pad and the like, or when detecting that the user electronic terminal plays a program, the main control module sends a control instruction and sends the control instruction to the screen capturing/recording module, wherein the control instruction comprises the following steps: a screen capture control instruction and a recording control instruction.
Furthermore, the screen capture/recording module acquires screen information of the user terminal after the user terminal is started according to the screen capture control instruction and the recording control instruction of the control instruction, and performs screen capture and recording on the electronic terminal playing device of the user at a certain frequency. In the present embodiment, the frequency of 300 ms/time is taken as an example to obtain the screen picture and the recorded audio of the program currently watched by the user. In addition, the user screen may be captured in addition to the screen capturing mode, and the screen capturing mode is taken as an example in this embodiment for description.
In this embodiment, the screen capture/recording module captures a screen of a user screen at a frequency of 300 ms/time to obtain a screen capture picture of a program currently watched by the user, records program content at a frequency of 300 ms/time to obtain a recording audio of the program currently watched by the user, and obtains a screen picture and an audio of the user terminal after being turned on.
And step S20, according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result.
As an implementation manner, in this embodiment, the picture and/or the audio are input into the trained program scene recognition model in a data stream format, and the program scene recognition model performs recognition according to an object and/or an identifier appearing in the picture, or performs recognition according to a voiceprint in the audio, so as to obtain a program scene recognition result. The program scene recognition result comprises: one or more of program scene information and a weight corresponding to the program scene information. In this embodiment, the program scene recognition result includes: the program scene information and the weight corresponding to the program scene information are taken as examples.
Specifically, as an embodiment, in this embodiment, the model operation module combines the program scene recognition model, inputs the picture and/or the audio into the program scene recognition model in a data stream format, and performs scene recognition on the acquired program screenshot picture and the acquired program recording audio to obtain a program scene recognition result. And the program scene recognition model recognizes the program scene according to the program screenshot picture and aiming at the factors such as objects, marks, characters and the like appearing in the program screenshot picture. For example, when a sports scene appears on the screen, it is determined that the user may be watching a sports program, and when a word or a logo of "NBA" appears on the screen, it is determined that the user is watching a NBA basketball program, resulting in a program scene recognition result of "NBA basketball program".
Further, the program scene recognition model acquires special voiceprint information in the voiceprint fragment according to the voiceprint fragment of the program recording audio to recognize the program scene. For example, when the voiceprint information shows a music type voiceprint, it is determined that the user may be watching a music type program, and further, when the music voiceprint is a voiceprint such as orchestra, a program scene recognition result of "orchestra" is obtained.
Further, the program scene recognition model gives a weight of the corresponding program scene information. The program scene recognition model recognizes and judges the user program screenshot picture and the program recording audio to obtain the program scene recognition result watched by the user, and calculates the proportion of the program scene recognition result, namely the weight of the corresponding program scene information according to the frequency of the program scene recognition result, such as 'NBA basketball program 0.6', news live broadcast 0.1 'and symphony music 0.3'.
And step S30, obtaining corresponding advertisement according to the program scene identification result.
As an implementation manner, in this embodiment, the program scene information and the weights are filtered, and when N consecutive program scene information are consistent with the corresponding weights, a program scene distribution condition is obtained, where N is an integer and N is greater than or equal to 3; and obtaining corresponding advertisements according to the distribution situation of the program scenes.
Specifically, as an embodiment, in this embodiment, the identification result processing unit performs screening according to the program scene identification result given by the model operation module, including screening the program scene information and the corresponding weight, that is, when the program scene identification result is the same for two or more consecutive times, it is determined that the program scene identification result is valid, the jitter identification error is eliminated, and the program scene distribution condition is obtained, in this embodiment, the program scene identification results for three consecutive times are the same as an example; and obtaining corresponding advertisements from a cloud server according to the distribution condition of the program scene.
Further, according to the program scene recognition results given by the model operation module, such as "NBA basketball program 0.6, live news 0.1 and symphony 0.3", the recognition result processing unit further judges the program scene recognition results "NBA basketball program 0.6, live news 0.1 and symphony 0.3", continuously receives the program scene recognition results three times, if the program scene recognition results three times are the same, namely "NBA basketball program 0.6, live news 0.1 and symphony 0.3", confirms that the program scene recognition model is not misjudged, eliminates the shake recognition error of the program scene recognition model, and obtains the distribution condition of the program scene, namely "NBA basketball program 0.6, live news 0.1 and symphony 0.3".
Further, when the recognition results of the program scenes received three times continuously are not completely consistent, for example, when the user is at a turnplate, the screen capture/recording module obtains the screen capture pictures and the program recording audio of other programs appearing in the middle of the turnplate, the results transmitted by the program scene recognition model may be "NBA basketball program 0.6, cultural relic appreciation 0.1, orchestra 0.3", "NBA basketball program 0.6, news live broadcast 0.1, orchestra 0.3", the recognition result processing unit determines the shake recognition error of the program scene recognition model, waits until the recognition results of the program scenes received three times continuously, such as "NBA basketball program 0.6, news live broadcast 0.1, orchestra 0.3", excludes the shake recognition error of the program scene recognition model, and obtains the program distribution scene situation, namely "NBA basketball 0.6", the news is live 0.1, and the symphony is 0.3 ".
Further, according to the obtained program scene distribution situation, advertisement content corresponding to the program scene distribution situation is obtained from a cloud server, namely a background advertisement server, and the background advertisement server pushes the advertisement content according with the watching habit of the user. For example, advertisement contents of "NBA events" peripheral products such as shoes, uniforms, and souvenirs are pushed according to "NBA basketball program", and local latest symphony show advertisements are pushed according to "symphony".
Furthermore, when the distribution situation of the program scenes changes, the device can update the program scenes to a background advertisement server in real time, and pushes advertisement contents according to the latest program data.
Step S40, playing the corresponding advertisement on the screen.
As an implementation manner, in this embodiment, the advertisement control playing unit plays the corresponding advertisement on the screen of the electronic user terminal according to the advertisement content pushed by the background advertisement server, the playing manner is not limited to the form of a patch advertisement or an open screen advertisement, and since the played advertisements are all contents that conform to the habit of the user, the user is prevented from ending the advertisement playing in advance, and the purpose of improving the advertisement pushing effect is achieved.
According to the AI-based intelligent advertisement pushing method, the screen information of the terminal after being started is obtained in the mode; according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result; acquiring a corresponding advertisement according to the program scene identification result; and playing the corresponding advertisement on the screen. Through the mode, the intelligent advertisement pushing system and the intelligent advertisement pushing method solve the problems that the advertisement is not in accordance with the requirements of consumers and the expected effect is not achieved, and realize intelligent advertisement pushing.
Further, referring to fig. 4, fig. 4 is a flowchart illustrating a second embodiment of the AI-based intelligent advertisement delivery method according to the present invention. Based on the embodiment shown in fig. 2, in step S10: the method also comprises the following steps before the screen information after the terminal is started is acquired:
step S50, establishing a program scene recognition model through AI technology; collecting relevant pictures and/or audio of historical programs watched by a user, and obtaining corresponding program scene information; and inputting the relevant pictures and/or audios of the historical programs and the corresponding program scene information into the program scene recognition model, and training the program scene recognition model to obtain a trained program scene recognition model.
As an implementation manner, in this embodiment, a system developer may develop a convolution algorithm based on an Android nn interface, establish a program scene recognition model through an AI technology, and collect a large amount of materials including program-related pictures and audio. The program related pictures and the audios are classified to obtain a mapping relation of the program related pictures and the audio-program scene information, then the mapping relation of the program related pictures and the audio-program scene information is input into a program scene recognition model, and the program scene recognition model is trained.
After the program scene recognition model learns the mapping relation of the program related pictures and the audio-program scene information, the capability of obtaining the corresponding program scene information according to the program related pictures and the audio is obtained, meanwhile, system developers carry out manual correction, and the recognition accuracy of the program scene recognition model to the materials is improved through continuous training.
Specifically, as an embodiment, in this embodiment, the program scene recognition model also collects pictures and/or audios related to a historical program watched by a user, obtains corresponding program scene information according to a mapping relationship between "program related pictures and audio-program scene information" for the pictures and/or audios related to the historical program, inputs the program scene recognition model, trains the program scene recognition model, obtains the trained program scene recognition model, further improves the material recognition accuracy of the program scene recognition model, and is in accordance with a usage habit close to the user.
Furthermore, the identification process can be operated independently, and does not occupy the Central Processing Unit (CPU) and Graphic Processing Unit (GPU) resources of the host chip Soc (System on chip), so as to perform artificial intelligent identification without affecting the normal use of the electronic device. Meanwhile, by means of further improvement of the computing power of the current embedded neural network processor, the image recognition capability reaches the level of frame by frame, and the recognition speed and accuracy are further improved.
Further, in this embodiment, the steps S10 to S40 refer to the first embodiment of the AI-based intelligent advertisement delivery method of the present invention, and are not repeated herein.
According to the scheme, the program scene recognition model is established through an AI technology; collecting relevant pictures and/or audio of historical programs watched by a user, and acquiring corresponding scene information; inputting the relevant pictures and/or audios of the historical program and the corresponding scene information into the program scene recognition model, and training the program scene recognition model to obtain a trained program scene recognition model; acquiring screen information after the terminal is started; according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result; acquiring a corresponding advertisement according to the program scene identification result; and playing the corresponding advertisement on the screen. Through the mode, the intelligent advertisement pushing system and the intelligent advertisement pushing method solve the problems that the advertisement is not in accordance with the requirements of consumers and the expected effect is not achieved, and realize intelligent advertisement pushing.
Further, referring to fig. 5, fig. 5 is a flowchart illustrating a third embodiment of the AI-based smart advertisement delivery method according to the present invention. Based on the embodiment shown in fig. 2, in step S30: according to the program scene recognition result, obtaining the corresponding advertisement specifically comprises:
step S301, the program scene information and the weights are screened, and when N continuous program scene information are consistent with the corresponding weights, the distribution situation of the program scenes is obtained, wherein N is an integer and is more than or equal to 3.
As an implementation manner, in this embodiment, the identification result processing unit, according to the program scene identification result given by the model operation module, includes screening the program scene information and the corresponding weight, that is, when the program scene identification result is the same for two or more consecutive times, it is determined that the program scene identification result is valid, the jitter identification error is eliminated, and the program scene distribution condition is obtained, in this embodiment, three consecutive times are taken as an example.
Specifically, as an embodiment, in this embodiment, according to the program scene recognition results given by the model operation module, such as "NBA basketball program 0.6, news live broadcast 0.1, and symphony music 0.3", where "NBA basketball program, news live broadcast, and symphony music" are program scene information and "0.6, 0.1, and 0.3" are corresponding weights, the recognition result processing unit further determines the program scene recognition results "NBA basketball program 0.6, news live broadcast 0.1, and symphony music 0.3", continuously receives the program scene recognition results three times, if the program scene recognition results three times are all the same, that is, the program scene recognition results three times are all "NBA basketball program 0.6, news 0.1, and symphony music 0.3", confirms that the program scene recognition model has no erroneous judgment, excludes the shaking recognition error of the program scene recognition model, and obtains the program scene distribution situation, that is "NBA basketball program 0.6, the news is live 0.1, and the symphony is 0.3 ".
Further, when the recognition results of the program scenes received three times continuously are not completely consistent, for example, when the user is in a channel switching, the screen capturing/recording module obtains the captured images of other programs and the recorded audio of the programs which appear in the middle of the channel switching, the results transmitted by the program scene recognition model may be "NBA basketball program 0.6, cultural relic appreciation 0.1, orchestra 0.3", "NBA basketball program 0.6, news live broadcast 0.1, orchestra 0.3", the recognition result processing unit determines that the program scene recognition model has a shake recognition error, waits until the recognition results of the program scenes received three times continuously, such as "NBA basketball program 0.6, news live broadcast 0.1, orchestra 0.3", excludes the shake recognition error of the program scene recognition model, and obtains the program distribution scene condition, namely 'NBA basketball program 0.6, news live broadcast 0.1 and symphony music 0.3'.
Step S302, according to the distribution situation of the program scenes, obtaining corresponding advertisements.
As an implementation manner, in this embodiment, according to the distribution situation of the program scenes, the corresponding watching duration of the distribution situation of the program scenes is counted; sequencing the watching duration, and synchronizing the sequenced watching duration and the distribution situation of the program scene to a cloud server; and receiving the advertisement fed back by the cloud server according to the sequenced watching duration and the program scene distribution condition, and pushing the advertisement to a terminal user.
Specifically, as an embodiment, in this embodiment, the background advertisement management unit counts the usage duration corresponding to the program scene distribution situation according to the program scene distribution situation, that is, "NBA basketball program 0.6, symphony music 0.3, and news live broadcast 0.1", that is, the background advertisement management unit counts the watching duration of each program of "NBA basketball program", "symphony music", and "news live broadcast" of the user, obtains the results such as "NBA basketball program 1.5h, symphony music 0.75h, and news live broadcast 0.5 h", and sends the results such as "NBA basketball program 1.5h, symphony music 0.75h, and news live broadcast 0.5 h" to the advertisement control playing unit.
Further, the advertisement control playing unit sorts the counted watching time length and selects the distribution condition of the program scenes three or more before the watching time length. For example, in this embodiment, the advertisement control playing unit sorts the results of "NBA basketball program 1.5h, symphony 0.75h, and news live broadcast 0.5 h" according to the weights, to obtain "NBA basketball program 1.5h, symphony 0.75h, and news live broadcast 0.5 h", and selects the distribution of the top three or more program scenes and the corresponding duration.
Further, the advertisement control playing unit synchronizes the distribution condition and the corresponding duration of the selected program scenes, namely 'NBA basketball program 1.5h, symphony music 0.75h and news live broadcast 0.5 h', to the background advertisement server at a certain frequency. In this embodiment, the advertisement control playback unit synchronizes the program data to the background advertisement server at a rate of 10 s/time.
Further, the background advertising server pushes the corresponding advertising content according to the NBA basketball program of 1.5h, the symphony of 0.75h and the news broadcast of 0.5h, for example, the advertising content of the products around the NBA match such as sneakers, uniforms and souvenirs is pushed according to the NBA basketball program, and the local latest symphony show advertisement is pushed according to the symphony to the terminal user.
Furthermore, when the program data are changed, the device can be updated to the background advertisement server in real time, and advertisement content terminal users can be pushed according to the latest program data.
According to the AI-based intelligent advertisement pushing method, the screen information of the terminal after being started is obtained in the mode; according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result; screening the program scene recognition results, and obtaining the distribution condition of the program scenes when N consecutive program scene recognition results are consistent, wherein N is an integer and is more than or equal to 3; according to the distribution situation of the program scenes, counting the corresponding watching duration of the distribution situation of the program scenes; sequencing the watching duration, and synchronizing the sequenced watching duration and the distribution situation of the program scene to a cloud server; receiving the advertisements fed back by the cloud server according to the sorted watching duration and the program scene distribution condition, and pushing the advertisements to a terminal user to play the corresponding advertisements on the screen. Through the mode, the intelligent advertisement pushing system and the intelligent advertisement pushing method solve the problems that the advertisement is not in accordance with the requirements of consumers and the expected effect is not achieved, and realize intelligent advertisement pushing.
The invention also provides an AI-based intelligent advertisement pushing system.
The AI-based intelligent advertisement push system comprises: the system comprises a memory, a processor and an AI-based smart advertisement push program stored on the memory and operable on the processor, wherein the AI-based smart advertisement push program realizes the steps of the AI-based smart advertisement push method when the processor executes the AI-based smart advertisement push program.
The method implemented when the AI-based smart advertisement push program running on the processor is executed may refer to various embodiments of the AI-based smart advertisement push method of the present invention, and details thereof are not repeated herein.
The invention also provides a storage medium.
The storage medium of the present invention stores an AI-based smart advertisement push program, and the AI-based smart advertisement push program implements the steps of the AI-based smart advertisement push method as described above when executed by a processor.
The method implemented when the AI-based smart advertisement push program running on the processor is executed may refer to various embodiments of the AI-based smart advertisement push method of the present invention, and details thereof are not repeated herein.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be substantially or partially embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for causing a system device (e.g., a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. An AI-based intelligent advertisement push method is characterized by comprising the following steps:
acquiring screen information after the terminal is started;
according to the screen information, combining the trained program scene recognition model to obtain a program scene recognition result;
acquiring a corresponding advertisement according to the program scene identification result;
and playing the corresponding advertisement on the screen.
2. The AI-based smart advertisement push method according to claim 1, wherein the step of obtaining the screen information after the terminal is powered on further comprises:
establishing a program scene recognition model through an AI technology;
collecting relevant pictures and/or audio of historical programs watched by a user, and obtaining corresponding program scene information;
and inputting the relevant pictures and/or audios of the historical programs and the corresponding program scene information into the program scene recognition model, and training the program scene recognition model to obtain a trained program scene recognition model.
3. The AI-based smart advertisement push method according to claim 2, wherein the step of acquiring the screen information after the terminal is powered on comprises:
when the opening of a terminal screen is detected or in the process of playing programs by the terminal is detected, screen information on the terminal screen is acquired according to a preset frequency.
4. The AI-based smart advertisement push method according to claim 3, wherein the screen information includes: the step of obtaining a program scene recognition result by combining the trained program scene recognition model according to the screen information comprises the following steps:
and inputting the picture and/or the audio into the trained program scene recognition model in a data stream format, and recognizing the picture and/or the audio by the program scene recognition model according to an object and/or an identifier appearing in the picture or according to a voiceprint in the audio to obtain a program scene recognition result.
5. The AI-based smart advertisement push method according to claim 4, wherein the program scene recognition result includes: one or more of program scene information and a weight corresponding to the program scene information.
6. The AI-based smart advertisement push method according to claim 5, wherein the step of obtaining the corresponding advertisement according to the program scene recognition result comprises:
screening the program scene information and the weights, and obtaining the distribution condition of the program scenes when N continuous program scene information are consistent with the corresponding weights, wherein N is an integer and is more than or equal to 3;
and obtaining corresponding advertisements according to the distribution situation of the program scenes.
7. The AI-based smart advertisement push method according to claim 6, wherein the step of obtaining corresponding advertisements according to the distribution of the program scenes comprises:
according to the distribution situation of the program scenes, counting the corresponding watching duration of the distribution situation of the program scenes;
sequencing the watching duration, and synchronizing the sequenced watching duration and the distribution situation of the program scene to a cloud server;
and receiving the advertisement fed back by the cloud server according to the sequenced watching duration and the program scene distribution condition, and pushing the advertisement to a terminal user.
8. The utility model provides an intelligent advertisement pusher based on AI which characterized in that includes:
the screen capture/recording module is used for acquiring screen information after the terminal is started and sending the screen information to the model operation module;
the model operation module is used for obtaining a program scene recognition result by combining the trained program scene recognition model according to the screen information and sending the program scene recognition result to the advertisement pushing module;
the advertisement pushing module is used for obtaining corresponding advertisements according to the program scene recognition result and sending the corresponding advertisements to the advertisement playing module;
and the advertisement playing module is used for playing the corresponding advertisement on the screen.
9. An AI-based smart advertisement push system, the system comprising: a memory, a processor and an AI-based smart advertisement push program stored on the memory and executable on the processor, the AI-based smart advertisement push program implementing the steps of the AI-based smart advertisement push method according to any one of claims 1 to 8 when executed by the processor.
10. A storage medium having an AI-based smart advertisement push program stored thereon, wherein the AI-based smart advertisement push program, when executed by a processor, implements the steps of the AI-based smart advertisement push method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011351491.6A CN112330371A (en) | 2020-11-26 | 2020-11-26 | AI-based intelligent advertisement pushing method, device, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011351491.6A CN112330371A (en) | 2020-11-26 | 2020-11-26 | AI-based intelligent advertisement pushing method, device, system and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112330371A true CN112330371A (en) | 2021-02-05 |
Family
ID=74308851
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011351491.6A Pending CN112330371A (en) | 2020-11-26 | 2020-11-26 | AI-based intelligent advertisement pushing method, device, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112330371A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114155019A (en) * | 2021-11-04 | 2022-03-08 | 广州市玄武无线科技股份有限公司 | Marketing message generation method, marketing message generation device, marketing message generation equipment and storage medium |
| WO2023060759A1 (en) * | 2021-10-11 | 2023-04-20 | 深圳创维－Rgb电子有限公司 | Video pushing method, device, and storage medium |
| CN116383620A (en) * | 2023-03-29 | 2023-07-04 | 北京鹅厂科技有限公司 | Method and device for applying multi-mode artificial intelligence |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104967885A (en) * | 2015-03-27 | 2015-10-07 | 哈尔滨工业大学深圳研究生院 | Advertisement recommending method and system based on video content |
| CN106600337A (en) * | 2016-12-22 | 2017-04-26 | 深圳Tcl数字技术有限公司 | Advertisement pushing method and apparatus |
| CN107277573A (en) * | 2017-06-12 | 2017-10-20 | 深圳Tcl数字技术有限公司 | Video-frequency advertisement put-on method, device and computer-readable recording medium |
| CN108040125A (en) * | 2017-12-28 | 2018-05-15 | 北京陌上花科技有限公司 | Content recognition and method for pushing and TV syndrome AI assistant devices |
| CN108765033A (en) * | 2018-06-08 | 2018-11-06 | Oppo广东移动通信有限公司 | Transmitting advertisement information method and apparatus, storage medium, electronic equipment |
| CN111416995A (en) * | 2020-03-25 | 2020-07-14 | 深圳创维－Rgb电子有限公司 | Content pushing method and system based on scene recognition and intelligent terminal |
| JP2020156001A (en) * | 2019-03-22 | 2020-09-24 | 株式会社三井住友銀行 | Advertisement recommendation method utilizing ai, program, and computer |
-
2020
- 2020-11-26 CN CN202011351491.6A patent/CN112330371A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104967885A (en) * | 2015-03-27 | 2015-10-07 | 哈尔滨工业大学深圳研究生院 | Advertisement recommending method and system based on video content |
| CN106600337A (en) * | 2016-12-22 | 2017-04-26 | 深圳Tcl数字技术有限公司 | Advertisement pushing method and apparatus |
| CN107277573A (en) * | 2017-06-12 | 2017-10-20 | 深圳Tcl数字技术有限公司 | Video-frequency advertisement put-on method, device and computer-readable recording medium |
| WO2018227887A1 (en) * | 2017-06-12 | 2018-12-20 | 深圳Tcl数字技术有限公司 | Video advertisement delivery method, video advertisement generation method, device, server and readable storage medium |
| CN108040125A (en) * | 2017-12-28 | 2018-05-15 | 北京陌上花科技有限公司 | Content recognition and method for pushing and TV syndrome AI assistant devices |
| CN108765033A (en) * | 2018-06-08 | 2018-11-06 | Oppo广东移动通信有限公司 | Transmitting advertisement information method and apparatus, storage medium, electronic equipment |
| JP2020156001A (en) * | 2019-03-22 | 2020-09-24 | 株式会社三井住友銀行 | Advertisement recommendation method utilizing ai, program, and computer |
| CN111416995A (en) * | 2020-03-25 | 2020-07-14 | 深圳创维－Rgb电子有限公司 | Content pushing method and system based on scene recognition and intelligent terminal |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023060759A1 (en) * | 2021-10-11 | 2023-04-20 | 深圳创维－Rgb电子有限公司 | Video pushing method, device, and storage medium |
| CN114155019A (en) * | 2021-11-04 | 2022-03-08 | 广州市玄武无线科技股份有限公司 | Marketing message generation method, marketing message generation device, marketing message generation equipment and storage medium |
| CN116383620A (en) * | 2023-03-29 | 2023-07-04 | 北京鹅厂科技有限公司 | Method and device for applying multi-mode artificial intelligence |
| CN116383620B (en) * | 2023-03-29 | 2023-10-20 | 北京鹅厂科技有限公司 | Method and device for applying multi-mode artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109819179B (en) | Video editing method and device | |
| CN107566907B (en) | Video clipping method, device, storage medium and terminal | |
| CN112330371A (en) | AI-based intelligent advertisement pushing method, device, system and storage medium | |
| CN113591523B (en) | Display device and experience value updating method | |
| CN111163274B (en) | Video recording method and display equipment | |
| CN110209879B (en) | Video playing method, device, equipment and storage medium | |
| CN110958465A (en) | Video stream pushing method and device and storage medium | |
| CN112333499A (en) | Method for searching target equipment and display equipment | |
| CN114071197A (en) | Screen projection data processing method and device | |
| CN106162357A (en) | Obtain the method and device of video content | |
| CN112788268A (en) | Information pushing method based on video recording, smart television and storage medium | |
| CN108401173B (en) | Mobile live broadcast interactive terminal, method and computer readable storage medium | |
| CN110971783B (en) | Television sound and picture synchronous self-tuning method, device and storage medium | |
| CN112258232A (en) | Promotion content display method and device based on live broadcast picture | |
| CN112312142B (en) | Video playing control method and device and computer readable storage medium | |
| CN111291219A (en) | Method for changing interface background color and display equipment | |
| US8483435B2 (en) | Information processing device, information processing system, information processing method, and information storage medium | |
| JP4289576B2 (en) | Imaging device and information addition device | |
| KR20120110941A (en) | A method for expressing whether of virtual world and an apparatus thereof | |
| CN111491124A (en) | Video processing method and device and electronic equipment | |
| CN105657510A (en) | Image display apparatus and information providing method thereof | |
| KR20170043944A (en) | Display apparatus and method of controlling thereof | |
| CN114210045A (en) | Intelligent eye protection method and device and computer readable storage medium | |
| CN113609387A (en) | Playing content recommendation method and device, electronic equipment and storage medium | |
| CN112399235A (en) | Method for enhancing photographing effect of camera of smart television and display device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |