WO2019024003A1 - 一种用于智能设备的应用程序内容的推送方法及推送装置 - Google Patents

一种用于智能设备的应用程序内容的推送方法及推送装置 Download PDF

Info

Publication number
WO2019024003A1
WO2019024003A1 PCT/CN2017/095659 CN2017095659W WO2019024003A1 WO 2019024003 A1 WO2019024003 A1 WO 2019024003A1 CN 2017095659 W CN2017095659 W CN 2017095659W WO 2019024003 A1 WO2019024003 A1 WO 2019024003A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
image
information
database
recommended content
Prior art date
Application number
PCT/CN2017/095659
Other languages
English (en)
French (fr)
Inventor
顾海元
Original Assignee
深圳传音通讯有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳传音通讯有限公司 filed Critical 深圳传音通讯有限公司
Priority to PCT/CN2017/095659 priority Critical patent/WO2019024003A1/zh
Priority to CN201780094620.8A priority patent/CN111433761A/zh
Publication of WO2019024003A1 publication Critical patent/WO2019024003A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation

Definitions

  • the present invention relates to the field of smart devices, and in particular, to a push method and a push device for application content of a smart device.
  • an object of the present invention is to provide a push method and a push device for application content of a smart device.
  • the invention discloses a method for pushing application content of a smart device, characterized in that the pushing method comprises:
  • the recommended content of the application is displayed in the smart device.
  • constructing a database storing preset expression features and a correspondence between at least one set of emotion information and recommended content of the application comprises:
  • the database is built in a server
  • capturing an image of a user and acquiring a current expression feature in the image comprises:
  • the recommended content of the application corresponding to the current emotion information is obtained in the database, including:
  • the recommended content of the application includes at least one of a mobile phone theme, a screen saver, and music;
  • the application When the emotion information is included in the database, the application is launched in the smart device and the recommended content of the application is prompted.
  • displaying the recommended content of the application in the smart device further includes:
  • the recommended content is applied or restored to the original state according to the confirmation action.
  • the present invention also discloses a push device for an application content of a smart device, wherein the push device includes: a storage module, a photographing module, a matching module, an obtaining module, and a pushing module, wherein
  • the storage module is configured to construct a database storing preset expression features and a correspondence between at least one set of emotion information and recommended content of the application;
  • the photographing module is configured to capture an image of a user, and acquire a current expression feature in the image
  • the matching module is connected to the shooting module for comparing the current expression feature and the preset expression feature to determine current emotion information of the user;
  • the obtaining module is configured to be in communication with the matching module, and configured to acquire, in the database, recommended content of the application corresponding to the current emotional information;
  • the push module is in communication with the acquisition module, and is configured to display recommended content of the application in the smart device.
  • the storage module includes:
  • a database unit configured to build the database into a server
  • a communication unit in communication with the server, configured to establish a communication connection between the smart device and the server Connect
  • the loading unit is in communication connection with the communication unit, and the communication unit is configured to acquire the database in the server and load it.
  • the shooting module comprises:
  • An image obtaining unit configured to detect location information of a face of the user in the image
  • An image location unit in communication with the image acquisition unit, configured to determine a partial image of the facial image of the user according to the location information
  • An emotion extraction unit is communicatively coupled to the image location unit for extracting emotional characteristic information of the user from the partial image.
  • the acquiring module includes:
  • the recommended content of the application includes at least one of a mobile phone theme, a screen saver, and music;
  • an obtaining unit configured to: when the emotion information is included in the database, launch the application in the smart device and prompt the recommended content of the application.
  • the pushing module comprises:
  • a display unit configured to display a determination interface that requests the user to determine whether to use the recommended content
  • a receiving unit in communication with the display unit, configured to receive a confirmation action of the user
  • an execution unit in communication with the receiving unit, configured to apply the recommended content or restore to an original state according to the confirming action.
  • the current emotional state of the user can be obtained, and then the music corresponding to the user's current emotional state is automatically recommended to the user, and the user's emotion is adjusted.
  • the present invention automatically and accurately recommends to the user without manual search by the user.
  • the user's current emotional state matches the music, the application is more intelligent, and the user experience is higher;
  • FIG. 1 is a schematic flow chart of a method for pushing an application content for a smart device according to a preferred embodiment of the present invention
  • FIG. 2 is a schematic flow chart of step 101 of a pushing method according to a preferred embodiment of the present invention.
  • step 102 of a pushing method in accordance with a preferred embodiment of the present invention
  • FIG. 4 is a flow chart showing the step 104 of the pushing method in accordance with a preferred embodiment of the present invention.
  • FIG. 5 is a flow chart showing the step 104 of the pushing method in accordance with a preferred embodiment of the present invention.
  • FIG. 6 is a block diagram showing the structure of a push device for an application content of a smart device in accordance with a preferred embodiment of the present invention.
  • FIG. 7 is a block diagram showing the structure of a module in accordance with a preferred embodiment of the present invention.
  • FIG. 8 is a block diagram showing the construction of a module in accordance with a preferred embodiment of the present invention.
  • FIG. 9 is a block diagram showing the structure of a module in accordance with a preferred embodiment of the present invention.
  • FIG. 10 is a block diagram showing the structure of a module in accordance with a preferred embodiment of the present invention.
  • 10-pushing device 20-storage module, 30-shooting module, 40-matching module, 50-acquisition module, 60-pushing module:
  • the smart device used herein includes a device of a wireless signal receiver, a device having only a wireless signal receiver without a transmitting capability, and a device for receiving and transmitting hardware, which has the capability to A device that performs bidirectional communication reception and transmission hardware on a two-way communication link.
  • Such devices may include cellular or other communication devices having a single line display or a multi-line display or a cellular or other communication device without a multi-line display; PCS (Personal Communications Service), which may combine voice, data Processing, fax, and/or data communication capabilities; PDA (Personal Digital Assistant), which can include radio frequency receivers, pagers, Internet/Intranet access, web browsers, notepads, calendars, and/or GPS (Global Positioning System (Global Positioning System) receiver; conventional laptop and/or palmtop computer or other device having a conventional laptop and/or palmtop computer or other device that includes and/or includes a radio frequency receiver.
  • PCS Personal Communications Service
  • PDA Personal Digital Assistant
  • device may be portable, transportable, installed in a vehicle (aviation, sea and/or land), or adapted and/or configured to operate locally, and/or Run as a distribution Run in any other location on the earth and/or space.
  • the "device” and “smart device” used herein may also be a communication device, an internet device, a music/video playback device, such as a PDA, a MID (Mobile Internet Device), and/or have a music/video playback.
  • Functional mobile phones can also be smart TVs, set-top boxes and other devices. And the smart device, the smart terminal, the mobile terminal, and the mobile device described in this specification are all equivalent.
  • first, second, third, etc. may be used in the present disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other.
  • first information may also be referred to as second information without departing from the scope of the present disclosure.
  • second information may also be referred to as first information.
  • word "if” as used herein may be interpreted as "when” or “when” or “in response to a determination.”
  • first, second, third, etc. may be used in the present disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other.
  • first information may also be referred to as second information without departing from the scope of the present disclosure.
  • second information may also be referred to as first information.
  • word "if” as used herein may be interpreted as "when” or “when” or “in response to a determination.”
  • the database is built in a server; establishing the smart device a communication connection with the server; acquiring the database in the server and loading, including:
  • the recommendation method further includes: inputting an expression image, and correlating the entered at least one expression image with the recommended application information to form a relationship between the preset emotion feature and the recommendation information.
  • An expression image recognizer is used to collect, correlate and pre-store the corresponding device application matching the expression image, and store the expression image in the device to form a database.
  • the expression image recognizer may be a camera of a smart device, a camera connected to the smart device, or a camera module included in the smart device, etc., based on the stored database, may adopt a relatively direct storage method (using an input emoticon image or the emoticon image)
  • the relationship between the standard template and the recommendation information is used as the storage format of the data in the database. The user can associate, pre-record and modify the relationship between the recommendation information and the expression image, and the operation is convenient.
  • the smart device When the currently acquired emotional feature information does not exist in the database, that is, there is no relationship between the emotional feature information and the recommended information in the database, the smart device prompts the user whether to select the currently acquired emotional feature information and the user. After the specified recommendation information is established, it is stored in the database. After the user confirms, the corresponding relationship is saved in the database.
  • the database can be located in the memory of the smart device or in a remote server, and when the database is located in the remote server, the smart device can read the data stored in the server through the remote connection.
  • the user's facial image data mainly includes data of several areas that are most obvious when the facial expression occurs, and are easy to analyze, such as the user's eyebrow shape data, eye shape data, and mouth shape data.
  • the user's facial image data can be obtained from the user's facial photo or video image.
  • the smart device's camera is activated to perform video recording, and the user's video is obtained.
  • a static image of several frames is randomly selected from the video recording, the image includes an image frame of the user's facial image, and the facial image of the user in the image frame is identified and scanned, thereby extracting facial image data of the user.
  • the facial image of the user may be divided into a plurality of non-overlapping regions, and the facial image data of the user is obtained by performing non-coincident scanning on each region.
  • the method of obtaining the face image data of the user from the face image of the user is similar to the above-described method of acquiring the face image data from the image frame.
  • the specific judging method may be: first acquiring a facial image, and then dividing the image of the user's mouth and the image of the eye. When the degree of the corner of the user's mouth and the degree of frowning are greater, the probability that the user is in a negative mood is greater. The result of the user being in a negative mood can be identified with greater probability.
  • the user's mouth and eyes are not the only way to identify emotions, and other image data based on the user's face can also be a good way to determine the user's emotions.
  • the user face can be collected in advance.
  • the way of image data is created by the user's own database that conforms to the user's personalization.
  • the database stores: facial image data corresponding to positive emotions that are obtained according to historical characteristics of the user, and facial image data corresponding to negative emotions.
  • the obtained user face image data may be directly compared with the corresponding data in the database to obtain a more accurate recognition result.
  • the emotions corresponding to each emotion recognition result may be compared, and each emotion recognition result is obtained from each emotion recognition result.
  • the highest accuracy of the emotion recognition result is obtained as the current emotion of the user.
  • the user may obtain the corresponding at least two user emotion information by acquiring the emotion feature information in the facial image data of the user and performing the emotion recognition based on the acquired feature information, thereby determining the user according to the obtained at least two user emotion information.
  • the emotions achieve accurate recognition of user emotions.
  • the mood changes from time to time, generally according to each user's own operating habits, for example, when the user frequently inputs an error message when inputting information, the negative emotion rises;
  • the screen is found to be slow, the corresponding emotion is slow, and the negative emotion rises.
  • the emotion of the user is disclosed.
  • these speeds will change accordingly. For example, when the user's emotion is excited, the speed of inputting information, the speed of tapping the screen will become faster, and the communication information communicated with others is likely to be emotionally excited.
  • the Emotion API can be used as an input to the facial expression in the image
  • the Face API is used to return the confidence of a set of emotions of each face in the image and the bounding box of the face, and finally obtain an sentiment analysis graph, the sentiment analysis graph It includes: anger, depression, disgust, fear, happiness, calmness, sadness, surprise, etc.), and determines the final emotional feature value according to the magnitude of the various emotions in the sentiment analysis graph.
  • the recommendation information includes at least one of a mobile phone theme, a screen saver, and music; and when the emotional feature information is included in the database, the recommendation information is displayed in the smart device.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • the corresponding desktop theme that has been downloaded to the terminal is recommended according to the user's emotional feature information, and the corresponding desktop theme may be downloaded and recommended according to the user's emotional feature information.
  • this step can quickly filter and recommend the appropriate user's desktop theme from the downloaded desktop theme.
  • the recommended way is to recommend through the notification bar or the status bar push message form, the user can click on the notification bar or the status bar push message to jump to the corresponding desktop theme display page.
  • the desktop theme display page shows desktop themes related to emotional feature information.
  • some user information such as age, occupation, etc., can be collected and recommended based on the information.
  • the step of recommending a desktop theme corresponding to the user's emotional feature information according to the determined user's emotional feature information includes: downloading the emotional feature information of the current recommendation and the user according to the user's emotional feature information Corresponding desktop theme information; display the desktop theme information through a menu bar or a status bar. Users need to click on the menu bar or the desktop topic information in the status bar to further expand the specific content.
  • the content displayed in the menu bar or status bar includes text and links. This operation allows users to easily browse information about recommended desktop themes, and save resources and streamline the process.
  • the desktop theme recommendation method further includes: if the user's entry request is obtained within a preset time period, the entry includes the The online interface of the recommended desktop theme preview image, the online interface displays a desktop theme preview corresponding to the displayed desktop theme information; if the user's entry request is not obtained within the preset time period or the user's abandonment is obtained When requested, the desktop topic information for this presentation is revoked.
  • the text information of the desktop theme is displayed in the menu bar or the status bar, and the text information is accompanied by a desktop theme link stored on the server.
  • the user clicks on the text link the user enters the request and obtains the server's consent to enter the online interface of the desktop theme preview with detailed information.
  • the online interface displays more theme-related or similar desktop theme previews; for example: for happy moods, you can recommend sports themes, colors tend to be darker, icons tend to be business, or some NBA and other types of themes; Sad mood, you can recommend the theme of shopping, color is biased towards bright colors, icons and other updates are bright, or more stars or Korean drama types. Displayed desktop themes for users to view and select.
  • the recommended theme it can be further classified according to different types of cars, beauty, plants, weather, etc.
  • the desktop theme information of the present presentation is automatically revoked and returned to the desktop.
  • the user may select a favorite desktop theme and click a related button of the desktop theme, and a selection box pops up for the user to select.
  • the user can choose to apply or not to apply the selected desktop theme. Automatic download when the user selects the app
  • the desktop application is automatically installed; when the user chooses to cancel, the recommended desktop theme information is canceled, and the original state of the desktop application is returned.
  • Entering the online interface allows the user to have more choices, and avoids the problem of high memory usage of the terminal caused by downloading too much desktop theme information, thereby achieving the purpose of saving hardware resources.
  • Embodiment 2 is a diagrammatic representation of Embodiment 1:
  • the user's current emotional state is calm, the user is recommended to mix some tracks, that is, various styles of music, at this time the user's mood is calm, suitable for enjoying the music of various styles; if the user's current emotional state is happy, then Recommend some cheerful songs for the user, in line with the user's happy mood at this time, make people feel more comfortable; if the user's current emotional state is sad, recommend some slow, gentle music for the user, can comfort and cure the user at this time Sad mood; if the user's current emotional state is angry, recommend some rock venting emotions for the user, and recommend some calm songs, which can subconsciously appease the user's emotions; if the user's current emotional state is excited, It also recommends cheerful and dynamic songs for the user, which is also in line with the user's mood at this time; if the user's current mood is depressed, the user is recommended some lyrics for more inspirational and soothing music, so that the user's emotions are slowly calmed down.
  • the corresponding music that has been downloaded to the terminal is recommended according to the emotional feature information of the user, or the corresponding music may be downloaded online for recommendation according to the emotional feature information of the user.
  • this step can quickly filter and recommend the music of the appropriate user from the downloaded music.
  • the recommended way is to recommend through the notification bar or the status bar push message form, the user can click on the notification bar or the status bar push message to jump to the corresponding music play page.
  • the music page shows the music associated with the emotional feature information.
  • the step of recommending music corresponding to the emotional feature information of the user according to the determined emotional feature information of the user includes: downloading, according to the emotional feature information of the user, corresponding to the emotional feature information of the user in the current recommendation Music; display the music through the menu bar or status bar. Users need to click on the menu bar or the music in the status bar to further expand the specific content.
  • the content displayed in the menu bar or status bar includes text and links. This operation allows the user to easily browse information related to the recommended music, and saves resources and implements a simple process.
  • the user may select the favorite music and click the related button of the music, and a selection box pops up for the user to select, the user may Select the app or not to apply the selected music.
  • the user may select the music information recommended this time is canceled and the original state of the music application is returned. .
  • Entering the online interface allows users to have more choices while avoiding downloading too much music and causing the terminal's memory.
  • the problem of high resource occupancy has achieved the goal of saving hardware resources.
  • Embodiment 3 is a diagrammatic representation of Embodiment 3
  • the recommended wallpaper that has been downloaded to the terminal is recommended according to the user's emotional feature information, or the corresponding wallpaper is downloaded and recommended according to the user's emotional feature information, and the recommendation is obtained by receiving the recommended wallpaper request sent by the smart device.
  • the screen resolution of the smart device When the user has downloaded a large number of wallpapers to the terminal in advance, this step can quickly filter and recommend the wallpaper of the appropriate user from the downloaded wallpaper.
  • the recommended way is to recommend through the notification bar or the status bar push message form, the user can click on the notification bar or the status bar push message to jump to the corresponding wallpaper play page.
  • the wallpaper page shows the wallpaper associated with the emotional feature information.
  • the step of recommending a wallpaper corresponding to the user's emotional feature information according to the determined user's emotional feature information includes: downloading, according to the user's emotional feature information, the corresponding corresponding to the user's emotional feature information in the current recommendation Wallpaper; display the wallpaper through the menu bar or status bar. Users need to click on the menu bar or the wallpaper in the status bar to further expand the specific content.
  • the content displayed in the menu bar or status bar includes text and links. This operation allows the user to easily browse related information of the recommended wallpaper, and saves resources and realizes a simple process.
  • the user can select the favorite wallpaper and click the relevant button of the wallpaper, and a selection box pops up for the user to select, the user can select Apply or not apply the selected wallpaper.
  • the desktop application is automatically downloaded and automatically installed; when the user selects to cancel, the recommended wallpaper is canceled and the original state of the wallpaper is returned.
  • Entering the online interface allows the user to have more choices, while avoiding the problem of high memory usage of the terminal caused by downloading too many wallpapers, and achieving the purpose of saving hardware resources.
  • the device according to the present invention can be implemented by a push device 10, and includes: a storage module 20, a photographing module 30, a matching module 40, an obtaining module 50, and a pushing module 60:
  • the database is built in a server; a communication connection between the smart device and the server is established; and the database in the server is acquired and loaded, including:
  • the database unit provides: inputting an expression image, and correlating the input at least one expression image with the recommended application information to form a relationship between the preset emotion feature and the recommendation information.
  • An expression image recognizer is used to collect, correlate and pre-store the corresponding device application matching the expression image, and store the expression image in the device to form a database.
  • the expression image recognizer may be a camera of a smart device, a camera connected to the smart device, or a camera module included in the smart device, etc., based on the stored database, may adopt a relatively direct storage method (using an input emoticon image or the emoticon image)
  • the relationship between the standard template and the recommendation information is used as the storage format of the data in the database. The user can associate, pre-record and modify the relationship between the recommendation information and the expression image, and the operation is convenient.
  • the smart device When the currently acquired emotional feature information does not exist in the database, that is, there is no relationship between the emotional feature information and the recommended information in the database, the smart device prompts the user whether to select the currently acquired emotional feature information and the user. After the specified recommendation information is established, it is stored in the database. After the user confirms, the corresponding relationship is saved in the database.
  • the database can be located in the memory of the smart device or in a remote server, and when the database is located in the remote server, the smart device can remotely connect data stored in the server through the loading unit through the communication unit.
  • the user's facial image data mainly includes data of several areas that are most obvious when the facial expression occurs, and are easy to analyze, such as the user's eyebrow shape data, eye shape data, and mouth shape data.
  • the image acquisition unit acquires facial image data of the user, which may be obtained from a facial photo or a video image of the user.
  • the camera of the smart device is activated to perform video. Recording, after obtaining the video recording of the user, randomly extracting a static image of several frames from the video recording, the image containing an image frame of the user's facial image, and the image location unit identifying and scanning the facial image of the user in the image frame, Thereby, the face image data of the user is extracted.
  • the facial image of the user may be divided into a plurality of non-overlapping regions, and the facial image data of the user is obtained by performing non-coincident scanning on each region.
  • the method of obtaining the face image data of the user from the face image of the user is similar to the above-described method of acquiring the face image data from the image frame.
  • the specific judging method adopted by the emotion extracting unit may be: first acquiring a facial image, and then dividing the image of the user's mouth and the image of the eye, and when the corner of the user's mouth is bent and the degree of frowning is larger, the user is The greater the probability of negative emotions, the greater the probability that the user is in a negative mood.
  • the user's mouth and eyes are not the only way to identify emotions, and other image data based on the user's face can also be a good way to determine the user's emotions.
  • the user's own database that conforms to the user's personalization may be established by collecting the user's facial image data in advance.
  • the database stores: facial image data corresponding to positive emotions that are obtained according to historical characteristics of the user, and facial image data corresponding to negative emotions.
  • the obtained user face image data may be directly compared with the corresponding data in the database to obtain a more accurate recognition result.
  • the emotions corresponding to each emotion recognition result may be compared, and each emotion recognition result is obtained from each emotion recognition result.
  • the highest accuracy of the emotion recognition result is obtained as the current emotion of the user.
  • the user may obtain the corresponding at least two user emotion information by acquiring the emotion feature information in the facial image data of the user and performing the emotion recognition based on the acquired feature information, thereby determining the user according to the obtained at least two user emotion information.
  • the emotions achieve accurate recognition of user emotions.
  • the mood changes from time to time, generally according to each user's own operating habits, for example, when the user frequently inputs an error message when inputting information, the negative emotion rises;
  • the screen is found to be slow, the corresponding emotion is slow, and the negative emotion rises.
  • the emotion of the user is disclosed.
  • these speeds will change accordingly. For example, when the user's emotion is excited, the speed of inputting information, the speed of tapping the screen will become faster, and the communication information communicated with others is likely to be emotionally excited.
  • the Emotion API can be used as an input to the facial expression in the image
  • the Face API is used to return the confidence of a set of emotions of each face in the image and the bounding box of the face, and finally obtain an sentiment analysis graph, the sentiment analysis graph It includes: anger, depression, disgust, fear, happiness, calmness, sadness, surprise, etc.), and determines the final emotional feature value according to the magnitude of the various emotions in the sentiment analysis graph.
  • the recommendation information includes at least one of a mobile phone theme, a screen saver, and music; when the emotional feature information includes When the database is in the database, the recommendation information is displayed in the smart device.
  • the obtaining unit acquires the recommendation information in the database constructed by the storage module 20 based on the emotion feature information of the current user that the matching module 40 matches from the capturing module 30.
  • the recommendation information includes at least one of a mobile phone theme, a screen saver, and music; when the emotional feature information is included in the database, the display unit displays the recommended information in the smart device.
  • the display unit displays an online interface of the preview information related to the recommendation information
  • the user can select the favorite recommendation information and click the relevant button of the recommendation information, and a selection box pops up for the user to select, and the receiving unit provides the user. You can choose to apply or not to apply the selected recommendation information.
  • the execution unit automatically downloads the recommendation information application and automatically loads; when the user selects to cancel, the execution unit cancels the recommended information of the recommendation and returns to the original state.
  • Entering the online interface allows users to have more choices, while avoiding the problem of high memory usage of the terminal caused by downloading too many wallpapers, achieving the goal of saving hardware resources.
  • an embodiment of the present invention further provides a computer readable storage medium having stored therein computer executable instructions, such as a non-volatile memory such as an optical disk, a hard disk, or a flash memory.
  • computer executable instructions are for causing a computer or the like to perform various operations in the above-described recommended method of application information.
  • the present invention includes apparatus related to performing one or more of the operations described herein. These devices may be specially designed and manufactured for the required purposes, or may also include known devices in a general purpose computer. These devices have computer programs stored therein that are selectively activated or reconfigured.
  • Such computer programs may be stored in a device (eg, computer) readable medium or in any type of medium suitable for storing electronic instructions and separately coupled to a bus, including but not limited to any Types of disks (including floppy disks, hard disks, optical disks, CD-ROMs, and magneto-optical disks), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory) , EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic card or light card.
  • a readable medium includes any medium that is stored or transmitted by a device (eg, a computer) in a readable form.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • Information Transfer Between Computers (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种用于智能设备的应用程序内容的推送方法及推送装置，其特征在于，所述推送方法包括：构建一存储有预设表情特征及至少一组情绪信息与应用程序的推荐内容之间对应关系的数据库（101）；拍摄一用户的图像，获取所述图像内的当前表情特征（102）；比较所述当前表情特征及所述预设表情特征，判断所述用户的当前情绪信息（103）；于所述数据库中获取所述当前情绪信息对应的所述应用程序的推荐内容（104）；于所述智能设备中显示所述应用程序的推荐内容（105）。本方法和装置不需要用户手动搜索即可自动、准确地向用户推荐与用户当前的情绪状态相符的音乐，应用更智能化，用户体验更高。

Description

一种用于智能设备的应用程序内容的推送方法及推送装置 技术领域
本发明涉及智能设备领域，尤其涉及一种用于智能设备的应用程序内容的推送方法及推送装置。
背景技术
主观情绪在人们生活和工作的各个方面起到了不可或缺的重要作用，而且在生活压力越来越大的今天，更多的人在不同程度上受到负面情绪的影响，尤其是一些特殊工作的人，比如从事航空航天、军事、心理医生、客服等工作人员，长时间工作环境单一、枯燥，容易产生消极情绪，而消极的情绪不但会降低工作效率，还会对身心健康造成不利的影响，因此，对情绪的自动识别研究显得尤为重要，且具有实际意义。
发明内容
为了克服上述技术缺陷，本发明的目的在于提供一种用于智能设备的应用程序内容的推送方法及推送装置。
本发明公开了一种用于智能设备的应用程序内容的推送方法，其特征在于，所述推送方法包括：
构建一存储有预设表情特征及至少一组情绪信息与应用程序的推荐内容之间对应关系的数据库；
拍摄一用户的图像，获取所述图像内的当前表情特征；
比较所述当前表情特征及所述预设表情特征，判断所述用户的当前情绪信息；
于所述数据库中获取所述当前情绪信息对应的所述应用程序的推荐内容；
于所述智能设备中显示所述应用程序的推荐内容。
优选地，构建一存储有预设表情特征及至少一组情绪信息与应用程序的推荐内容之间对应关系的数据库，包括：
所述数据库构建于一服务器中；
建立所述智能设备与所述服务器的通讯连接；
获取所述服务器内的所述数据库并加载。
优选地，拍摄一用户的图像，获取所述图像内的当前表情特征，包括：
在所述图像中检测所述用户的面部的位置信息；
根据所述位置信息确定所述用户的面部图像的局部图像；
从所述局部图像中提取所述用户的情绪特征信息。
优选地，于所述数据库中获取所述当前情绪信息对应的所述应用程序的推荐内容，包括：
所述应用程序的推荐内容包括手机主题、屏保、音乐中的至少一种；
当所述情绪信息包含于所述数据库中时，于所述智能设备中启动所述应用程序并提示所述应用程序的推荐内容。
优选地，于所述智能设备中显示所述应用程序的推荐内容，还包括：
显示一请求所述用户判断是否采用所述推荐内容的判断界面；
接收用户的确认动作；
根据所述确认动作应用所述推荐内容或还原至原始状态。
本发明还公开了一种用于智能设备的应用程序内容的推送装置，其特征在于，所述推送装置包括：存储模块、拍摄模块、匹配模块、获取模块、推送模块，其中，
所述存储模块，用于构建一存储有预设表情特征及至少一组情绪信息与应用程序的推荐内容之间对应关系的数据库；
所述拍摄模块，用于拍摄一用户的图像，获取所述图像内的当前表情特征；
所述匹配模块，与所述拍摄模块通讯连接，用于比较所述当前表情特征及所述预设表情特征，判断所述用户的当前情绪信息；
所述获取模块，与所述匹配模块通讯连接，用于于所述数据库中获取所述当前情绪信息对应的所述应用程序的推荐内容；
所述推送模块，与所述获取模块通讯连接，用于于所述智能设备中显示所述应用程序的推荐内容。
优选地，所述存储模块，包括：
数据库单元，用于将所述数据库构建于一服务器中；
通讯单元，与所述服务器通讯连接，用于建立所述智能设备与所述服务器的通讯连 接；
加载单元，与所述通讯单元通讯连接，通过所述通讯单元用于获取所述服务器内的所述数据库并加载。
优选地，所述拍摄模块，包括：
图像获取单元，用于在所述图像中检测所述用户的面部的位置信息；
图像位置单元，与所述图像获取单元通讯连接，用于根据所述位置信息确定所述用户的面部图像的局部图像；
情绪提取单元，与所述图像位置单元通讯连接，用于从所述局部图像中提取所述用户的情绪特征信息。
优选地，所述获取模块，包括：
所述应用程序的推荐内容包括手机主题、屏保、音乐中的至少一种；
获取单元，用于当所述情绪信息包含于所述数据库中时，于所述智能设备中启动所述应用程序并提示所述应用程序的推荐内容。
优选地，所述推送模块，包括：
显示单元，用于显示一请求所述用户判断是否采用所述推荐内容的判断界面；
接收单元，与所述显示单元通讯连接，用于接收用户的确认动作；
执行单元，与所述接收单元通讯连接，用于根据所述确认动作应用所述推荐内容或还原至原始状态。
采用了上述技术方案后，与现有技术相比，具有以下有益效果：
1.根据分析结果可以获取到用户当前的情绪状态，然后自动向用户推荐与用户当前的情绪状态相应的音乐，调节用户情绪，本发明不需要用户手动搜索即可自动、准确地向用户推荐与用户当前的情绪状态相符的音乐，应用更智能化，用户体验更高；
2.通过利用摄像头获取用户的面部信息，然后根据获取的面部信息识别确定用户的情绪，进而根据确定的用户情推荐与用户情对应的桌面主题；使得用户能够快速挑选出适合自己的桌面主题，大大缩短了用户挑选主题的时间，提高了用户的使用体验。
附图说明
图1为符合本发明一优选实施例中用于智能设备的应用程序内容的推送方法的流程示意图；
图2为符合本发明一优选实施例中，推送方法的步骤101的流程示意图；
图3为符合本发明一优选实施例中，推送方法的步骤102的流程示意图；
图4为符合本发明一优选实施例中，推送方法的步骤104的流程示意图；
图5为符合本发明一优选实施例中，推送方法的步骤104的流程示意图；
图6为符合本发明一优选实施例中用于智能设备的应用程序内容的推送装置的结构示意图。
图7为符合本发明一优选实施例中推送装置的模块的结构示意图。
图8为符合本发明一优选实施例中推送装置的模块的结构示意图。
图9为符合本发明一优选实施例中推送装置的模块的结构示意图。
图10为符合本发明一优选实施例中推送装置的模块的结构示意图。
附图标记：
10-推送装置、20-存储模块、30-拍摄模块、40-匹配模块、50-获取模块、60-推送模块：
具体实施方式
以下结合附图与具体实施例进一步阐述本发明的优点。
这里将详细地对示例性实施例进行说明，其示例表示在附图中。下面的描述涉及附图时，除非另有表示，不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本公开相一致的所有实施方式。相反，它们仅是与如所附权利要求书中所详述的、本公开的一些方面相一致的装置和方法的例子。
本技术领域技术人员可以理解，这里所使用的智能设备，既包括无线信号接收器的设备，其仅具备无发射能力的无线信号接收器的设备，又包括接收和发射硬件的设备，其具有能够在双向通信链路上，进行双向通信的接收和发射硬件的设备。这种设备可以包括：蜂窝或其他通信设备，其具有单线路显示器或多线路显示器或没有多线路显示器的蜂窝或其他通信设备；PCS(Personal Communications Service，个人通信系统)，其可以组合语音、数据处理、传真和/或数据通信能力；PDA(Personal Digital Assistant，个人数字助理)，其可以包括射频接收器、寻呼机、互联网/内联网访问、网络浏览器、记事本、日历和/或GPS(Global Positioning System，全球定位系统)接收器；常规膝上型和/或掌上型计算机或其他设备，其具有和/或包括射频接收器的常规膝上型和/或掌上型计算机或其他设备。这里所使用的“设备”、“智能设备”可以是便携式、可运输、安装在交通工具(航空、海运和/或陆地)中的，或者适合于和/或配置为在本地运行，和/或以分布形式，运行 在地球和/或空间的任何其他位置运行。这里所使用的“设备”、“智能设备”还可以是通信设备、上网设备、音乐/视频播放设备，例如可以是PDA，MID(Mobile Internet Device，移动互联网设备)和/或具有音乐/视频播放功能的移动电话，也可以是智能电视、机顶盒等设备。且本说明书中描述的智能设备、智能终端、移动终端、移动设备都是等同的。
在本发明的描述中，需要理解的是，术语“纵向”、“横向”、“上”、“下”、“前”、“后”、“左”、“右”、“竖直”、“水平”、“顶”、“底”“内”、“外”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本发明和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本发明的限制。
应当理解，尽管在本公开可能采用术语第一、第二、第三等来描述各种信息，但这些信息不应限于这些术语。这些术语仅用来将同一类型的信息彼此区分开。例如，在不脱离本公开范围的情况下，第一信息也可以被称为第二信息，类似地，第二信息也可以被称为第一信息。取决于语境，如在此所使用的词语“如果”可以被解释成为“在......时”或“当......时”或“响应于确定”。
在本发明的描述中，除非另有规定和限定，需要说明的是，术语“安装”、“相连”、“连接”应做广义理解，例如，可以是机械连接或电连接，也可以是两个元件内部的连通，可以是直接相连，也可以通过中间媒介间接相连，对于本领域的普通技术人员而言，可以根据具体情况理解上述术语的具体含义。
应当理解，尽管在本公开可能采用术语第一、第二、第三等来描述各种信息，但这些信息不应限于这些术语。这些术语仅用来将同一类型的信息彼此区分开。例如，在不脱离本公开范围的情况下，第一信息也可以被称为第二信息，类似地，第二信息也可以被称为第一信息。取决于语境，如在此所使用的词语“如果”可以被解释成为“在......时”或“当......时”或“响应于确定”。
参阅图1-5，为符合本发明一优选实施例中的流程示意图：
构建一存储至少一组情绪特征信息与推荐信息之间对应关系的数据库；
接收指向于所述智能设备的一用户的表情图像，提取所述照片中包含的情绪特征信息；
于所述数据库中获取所述情绪特征信息对应的所述推荐信息；
于所述智能设备中显示所述推荐信息；
在本发明的一个实施例中，其中所述数据库构建于一服务器中；建立所述智能设备 与所述服务器的通讯连接；获取所述服务器内的所述数据库并加载，包括：
该推荐方法还包括：录入表情图像，将录入的至少一表情图像与推荐的应用信息相互关联形成预设情绪特征与推荐信息的关系。通过一表情图像识别器用于采集、关联和预存与表情图像向匹配的相应设备应用，将表情图像存储在设备中形成数据库。该表情图像识别器可以是智能设备的摄像头，连接智能设备的相机，或者是智能设备包含的照相模块等，基于已存储的数据库，可以采用相对直接的存储方法(利用输入表情图像或该表情图像的标准模板的之间的关系与推荐信息的关系作为数据库内数据的存储格式)，用户对推荐信息与表情图像之间的关系能够进行关联、预录和修改，操作方便。
当当前获取的情绪特征信息不存在于所述数据库内时，也就是说该数据库内不存在一情绪特征信息与推荐信息的关系，智能设备提示用户是否将该当前获取的情绪特征信息与用户选择指定的一推荐信息建立对应关系后，存储于该数据库内，用户确认后，即在该数据库内保存该对应关系。
该数据库能够位于智能设备的存储器中或者位于一远端服务器中，当该数据库位于远端服务器中时，该智能设备能够通过远程连接读取该服务器中存储的数据。
用户的面部图像数据主要包括用户在发生面部表情时最明显，且便于分析的几个区域的数据，例如用户的眉毛形态数据、眼睛形态数据以及嘴部形态数据等。
实际应用中，用户的面部图像数据可以从用户的面部照片或视频图像中获得，以视频图像为例，当对用户的情绪进行识别时，启动智能设备的摄像头进行视频录像，在获得用户的视频录像后，从该视频录像中随机抽取若干帧的静态图像，该图像包含用户面部图像的图像帧，并对该图像帧中用户的面部图像进行识别和扫描，从而提取出用户的面部图像数据。其中，对于图像帧中用户面部图像的扫描，可以将用户的面部图像划分为若干个不重叠的区域，通过对各区域进行不重合的扫描来获得用户的面部图像数据。另外，当从用户的面部图像中获得用户的面部图像数据的方法与上述从图像帧中获取面部图像数据的方法类似。
具体的判断方法可以是，首先获取一面部图像，接着划分获取用户嘴部的图像以及眼部的图像，当用户嘴角的下弯程度和皱眉程度越大时，用户处于消极情绪的概率越大，就能较大概率地识别出用户处于消极情绪的结果。通过用户的嘴角以及眼部并不是唯一识别情绪的方法，而根据用户的面部的其他图像数据也能是很好地确定用户的情绪的方法。
进一步，为了使情绪识别跟更加符合用户的个性化特定，可以通过事先采集用户面 部图像数据的方式建立属于用户自己的，符合用户个性化的数据库。该数据库中存储有：根据历史获得的用户面部图像数据，分析获得的符合用户个性特点的积极情绪对应的面部图像数据，以及消极情绪对应的面部图像数据。当确定在进行用户情绪识别时，可以直接将获取到的用户面部图像数据分别与数据库中的对应数据进行对比，以获得更准确的识别结果。
实际应用中，在获得各类特征数据对应的情绪识别结果以及情绪结果对应的准确度后，可选的，可以通过将各情绪识别结果对应的准确度进行对比的方式，从各情绪识别结果中，获得准确度最高的情绪识别结果作为用户当前的情绪。从而达到降低误识别率，提高情绪识别准确率的目的。
优选地，可以通过获取用户的面部图像数据中的情绪特征信息，并基于获取到的特征信息进行情绪识别，获得对应的至少两个用户情绪信息，从而根据获得的至少两个用户情绪信息确定用户的情绪，实现了用户情绪的精准识别。
可选地，用户在使用智能设备，例如手机时，会有时不时地情绪变化，一般根据每个用户有自己的操作习惯，例如在输入信息时用户经常输入错误信息后，导致负面情绪上升；点击屏幕时，发现智能设备的相应速度很慢后，导致负面情绪上升，当用户同时在与他人进行通信的过程中，会将自身的该情绪公开。当用户的情绪发生变化时，这些速度会相应发生一定的变化，例如当用户的情绪激动时，输入信息的速度、点击屏幕的速度会变快，与他人交流的通信信息中很可能出现情绪激动相关的用词；反之，当用户情绪平静时，这些速度仍会保持在一定的数值范围内；而当用户的情绪较低落时，这些速度一般会变小，与他人交流的通信信息中很可能出现情绪低落相关的用词，因此，分析这类特征数据也可以获知用户的情绪状态。
优选地，可以使用Emotion API将图像中的面部表情作为输入，并使用Face API返回图像中每个面部的一组情绪的置信度以及面部的边界框，最后获得一个情感分析图，该情感分析图中包括：愤怒、消沉、厌恶、恐惧、开心、平静、悲伤、惊讶等)，根据该情感分析图中各类情绪的数值大小来确定最终的情绪特征值。
所述推荐信息包括手机主题、屏保、音乐中的至少一种；当所述情绪特征信息包含于所述数据库中时，于所述智能设备中显示所述推荐信息。
实施例一：
本实施例中，根据用户的情绪特征信息推荐己下载至终端的对应的桌面主题，也可以是根据用户的情绪特征信息在线下载对应的桌面主题进行推荐。当用户己提前下载大 量的桌面主题至终端时，本步骤可以从己下载的桌面主题中快速筛选并推荐合适用户的桌面主题。推荐的方式为通过通知栏或者状态栏的推送消息形式进行推荐，用户点击通知栏或者状态栏的推送消息后就可以跳转到相应的桌面主题展示页面。桌面主题展示页面展示了与情绪特征信息相关的桌面主题。
为了达到更精确的推荐效果，可以收集一些用户的信息，比如年龄、职业等信息，根据这些信息进行推荐。
为了达到更好的使用效果，该根据确定的用户的情绪特征信息推荐与用户的情绪特征信息对应的桌面主题的步骤包括：根据该用户的情绪特征信息下载本次推荐中与用户的情绪特征信息对应的桌面主题信息；将该桌面主题信息通过菜单栏或者状态栏进行展示。用户需要点击菜单栏或者状态栏的桌面主题信息才能进一步展开具体的内容。菜单栏或者状态栏展示的内容包括文字和链接。该操作可以使得用户简易浏览推荐的桌面主题的相关信息，并且节省资源、实现流程简便。
优选地，在所述将所述桌面主题信息通过菜单栏或者状态栏进行展示之后，所述桌面主题推荐方法还包括：若在预设时间段内获取到用户的进入请求，则进入包含有此次推荐的桌面主题预览图的在线界面，所述在线界面显示与展示的桌面主题信息相对应的桌面主题预览图；若在预设时间段内未获取到用户的进入请求或获取到用户的放弃请求时，则撤销本次展示的桌面主题信息。
具体地，在菜单栏或者状态栏进行展示的是桌面主题的文字信息，且文字信息附带有存储到服务器上的桌面主题链接。当用户点击文字链接时，即发出进入请求，获得服务器同意后，进入到具有详细信息的桌面主题预览图的在线界面。在线界面显示更多与主题相关的或者同类的桌面主题预览图；比如：对于高兴的心情，可以推荐运动主题，颜色更加倾向于暗色，图标倾向于商务，或者是一些NBA等类型的主题；对于悲伤的心情，可以推荐购物类的主题，颜色偏向于亮色，图标等更新鲜艳，或者多一些明星或者韩剧类型的主题。展示的桌面主题供用户查看和选择。对于推荐的主题，还可以进一步根据汽车、美女、植物、天气等不同类型分类展示。为了减少功耗，在预设时间段内未获取到用户的进入请求或获取到用户的放弃请求时，自动撤销本次展示的桌面主题信息，回到桌面。
更进一步的，在所述进入包含有此次推荐的桌面主题预览图的在线界面之后，用户可以选择喜欢的桌面主题，并点击该桌面主题的相关按钮，此时会弹出一个选择框供用户选择，用户可以选择应用或不应用所选择的桌面主题。当用户选择应用时，自动下载 该桌面应用并自动安装；当用户选择取消时，取消本次推荐的桌面主题信息，返回到桌面应用程序的原始状态。
进入在线界面可以让用户具有更多的选择，同时避免下载过多桌面主题信息造成终端的内存资源占用率高的问题，达到了节省硬件资源的目的。
实施例二：
如果用户当前情绪状态是平静的，则为用户推荐一些混合曲目，即各种风格的音乐，此时用户的心情平静，适合欣赏各种曲风的音乐；如果用户当前情绪状态是开心的，则为用户推荐一些欢快的歌曲，符合用户此时高兴的心情，让人心情更加舒畅；如果用户当前情绪状态是悲伤的，则为用户推荐一些缓慢的、温柔的音乐，可以安抚与治愈用户此时伤心的心情；如果用户当前情绪状态是愤怒的，则为用户推荐一些摇滚的发泄情绪类的歌曲，以及推荐一些平静的歌曲，可以潜意识中安抚用户的情绪；如果用户当前情绪状态是兴奋的，则为用户推荐欢快的、动感的歌曲，同样顺应用户此时的心情；如果用户当前情绪是消沉的，则为用户推荐一些歌词比较励志的和舒缓的音乐，让用户的情绪缓缓归于平静。
根据用户的情绪特征信息推荐己下载至终端的对应的音乐，也可以是根据用户的情绪特征信息在线下载对应的音乐进行推荐。当用户己提前下载大量的音乐至终端时，本步骤可以从己下载的音乐中快速筛选并推荐合适用户的音乐。推荐的方式为通过通知栏或者状态栏的推送消息形式进行推荐，用户点击通知栏或者状态栏的推送消息后就可以跳转到相应的音乐播放页面。音乐页面展示了与情绪特征信息相关的音乐。
为了达到更好的使用效果，该根据确定的用户的情绪特征信息推荐与用户的情绪特征信息对应的音乐的步骤包括：根据该用户的情绪特征信息下载本次推荐中与用户的情绪特征信息对应的音乐；将该音乐通过菜单栏或者状态栏进行展示。用户需要点击菜单栏或者状态栏的音乐才能进一步展开具体的内容。菜单栏或者状态栏展示的内容包括文字和链接。该操作可以使得用户简易浏览推荐的音乐的相关信息，并且节省资源、实现流程简便。
更进一步的，在所述进入包含有此次推荐的音乐预览图的在线界面之后，用户可以选择喜欢的音乐，并点击该音乐的相关按钮，此时会弹出一个选择框供用户选择，用户可以选择应用或不应用所选择的音乐。当用户选择取消时，取消本次推荐的音乐信息，返回到音乐应用程序的原始状态。。
进入在线界面可以让用户具有更多的选择，同时避免下载过多音乐造成终端的内存 资源占用率高的问题，达到了节省硬件资源的目的。
实施例三：
根据用户的情绪特征信息推荐己下载至终端的对应的壁纸，也可以是根据用户的情绪特征信息在线下载对应的壁纸进行推荐，通过在接收到智能设备发送的推荐壁纸请求的情况下，获取该智能设备的屏幕分辨率。当用户己提前下载大量的壁纸至终端时，本步骤可以从己下载的壁纸中快速筛选并推荐合适用户的壁纸。推荐的方式为通过通知栏或者状态栏的推送消息形式进行推荐，用户点击通知栏或者状态栏的推送消息后就可以跳转到相应的壁纸播放页面。壁纸页面展示了与情绪特征信息相关的壁纸。
为了达到更好的使用效果，该根据确定的用户的情绪特征信息推荐与用户的情绪特征信息对应的壁纸的步骤包括：根据该用户的情绪特征信息下载本次推荐中与用户的情绪特征信息对应的壁纸；将该壁纸通过菜单栏或者状态栏进行展示。用户需要点击菜单栏或者状态栏的壁纸才能进一步展开具体的内容。菜单栏或者状态栏展示的内容包括文字和链接。该操作可以使得用户简易浏览推荐的壁纸的相关信息，并且节省资源、实现流程简便。
更进一步的，在该进入包含有此次推荐的壁纸预览图的在线界面之后，用户可以选择喜欢的壁纸，并点击该壁纸的相关按钮，此时会弹出一个选择框供用户选择，用户可以选择应用或不应用所选择的壁纸。当用户选择应用时，自动下载该桌面应用并自动安装；当用户选择取消时，取消本次推荐的壁纸，返回到壁纸的原始状态。
进入在线界面可以让用户具有更多的选择，同时避免下载过多壁纸造成终端的内存资源占用率高的问题，达到了节省硬件资源的目的。
现在将参考附图描述实现本发明各个实施例的智能设备。在后续的描述中，使用用于表示元件的诸如“模块”、“部件”或“单元”的后缀仅为了有利于本发明的说明，其本身并没有特定的意义。因此，“模块”与“部件”可以混合地使用。
在本发明的描述中，除非另有规定和限定，需要说明的是，术语“安装”、“相连”、“连接”应做广义理解，例如，可以是机械连接或电连接，也可以是两个元件内部的连通，可以是直接相连，也可以通过中间媒介间接相连，对于本领域的普通技术人员而言，可以根据具体情况理解上述术语的具体含义。
参阅图6-10，为符合本发明装置，可以通过一推送装置10实现，在该推送装置10内，包括：存储模块20、拍摄模块30、匹配模块40、获取模块50、推送模块60：
存储模块20：
在本发明的一个实施例中，其中所述数据库构建于一服务器中；建立所述智能设备与所述服务器的通讯连接；获取所述服务器内的所述数据库并加载，包括：
数据库单元提供：录入表情图像，将录入的至少一表情图像与推荐的应用信息相互关联形成预设情绪特征与推荐信息的关系。通过一表情图像识别器用于采集、关联和预存与表情图像向匹配的相应设备应用，将表情图像存储在设备中形成数据库。该表情图像识别器可以是智能设备的摄像头，连接智能设备的相机，或者是智能设备包含的照相模块等，基于已存储的数据库，可以采用相对直接的存储方法(利用输入表情图像或该表情图像的标准模板的之间的关系与推荐信息的关系作为数据库内数据的存储格式)，用户对推荐信息与表情图像之间的关系能够进行关联、预录和修改，操作方便。
当当前获取的情绪特征信息不存在于所述数据库内时，也就是说该数据库内不存在一情绪特征信息与推荐信息的关系，智能设备提示用户是否将该当前获取的情绪特征信息与用户选择指定的一推荐信息建立对应关系后，存储于该数据库内，用户确认后，即在该数据库内保存该对应关系。
该数据库能够位于智能设备的存储器中或者位于一远端服务器中，当该数据库位于远端服务器中时，该智能设备能够通过通讯单元远程连接该服务器中通过加载单元存储的数据。
拍摄模块30：
匹配模块40：
用户的面部图像数据主要包括用户在发生面部表情时最明显，且便于分析的几个区域的数据，例如用户的眉毛形态数据、眼睛形态数据以及嘴部形态数据等。
实际应用中，图像获取单元获取用户的面部图像数据，具体地可以从用户的面部照片或视频图像中获得，以视频图像为例，当对用户的情绪进行识别时，启动智能设备的摄像头进行视频录像，在获得用户的视频录像后，从该视频录像中随机抽取若干帧的静态图像，该图像包含用户面部图像的图像帧，图像位置单元对该图像帧中用户的面部图像进行识别和扫描，从而提取出用户的面部图像数据。其中，对于图像帧中用户面部图像的扫描，可以将用户的面部图像划分为若干个不重叠的区域，通过对各区域进行不重合的扫描来获得用户的面部图像数据。另外，当从用户的面部图像中获得用户的面部图像数据的方法与上述从图像帧中获取面部图像数据的方法类似。
情绪提取单元采用的具体的判断方法可以是，首先获取一面部图像，接着划分获取用户嘴部的图像以及眼部的图像，当用户嘴角的下弯程度和皱眉程度越大时，用户处于 消极情绪的概率越大，就能较大概率地识别出用户处于消极情绪的结果。通过用户的嘴角以及眼部并不是唯一识别情绪的方法，而根据用户的面部的其他图像数据也能是很好地确定用户的情绪的方法。
进一步，为了使情绪识别跟更加符合用户的个性化特定，可以通过事先采集用户面部图像数据的方式建立属于用户自己的，符合用户个性化的数据库。该数据库中存储有：根据历史获得的用户面部图像数据，分析获得的符合用户个性特点的积极情绪对应的面部图像数据，以及消极情绪对应的面部图像数据。当确定在进行用户情绪识别时，可以直接将获取到的用户面部图像数据分别与数据库中的对应数据进行对比，以获得更准确的识别结果。
实际应用中，在获得各类特征数据对应的情绪识别结果以及情绪结果对应的准确度后，可选的，可以通过将各情绪识别结果对应的准确度进行对比的方式，从各情绪识别结果中，获得准确度最高的情绪识别结果作为用户当前的情绪。从而达到降低误识别率，提高情绪识别准确率的目的。
优选地，可以通过获取用户的面部图像数据中的情绪特征信息，并基于获取到的特征信息进行情绪识别，获得对应的至少两个用户情绪信息，从而根据获得的至少两个用户情绪信息确定用户的情绪，实现了用户情绪的精准识别。
可选地，用户在使用智能设备，例如手机时，会有时不时地情绪变化，一般根据每个用户有自己的操作习惯，例如在输入信息时用户经常输入错误信息后，导致负面情绪上升；点击屏幕时，发现智能设备的相应速度很慢后，导致负面情绪上升，当用户同时在与他人进行通信的过程中，会将自身的该情绪公开。当用户的情绪发生变化时，这些速度会相应发生一定的变化，例如当用户的情绪激动时，输入信息的速度、点击屏幕的速度会变快，与他人交流的通信信息中很可能出现情绪激动相关的用词；反之，当用户情绪平静时，这些速度仍会保持在一定的数值范围内；而当用户的情绪较低落时，这些速度一般会变小，与他人交流的通信信息中很可能出现情绪低落相关的用词，因此，分析这类特征数据也可以获知用户的情绪状态。
优选地，可以使用Emotion API将图像中的面部表情作为输入，并使用Face API返回图像中每个面部的一组情绪的置信度以及面部的边界框，最后获得一个情感分析图，该情感分析图中包括：愤怒、消沉、厌恶、恐惧、开心、平静、悲伤、惊讶等)，根据该情感分析图中各类情绪的数值大小来确定最终的情绪特征值。
所述推荐信息包括手机主题、屏保、音乐中的至少一种；当所述情绪特征信息包含 于所述数据库中时，于所述智能设备中显示所述推荐信息。
获取模块50：
获取单元基于匹配模块40从拍摄模块30匹配出的当前用户的情绪特征信息后，在存储模块20构建的数据库中，获取推荐信息。
推送模块60：
所述推荐信息包括手机主题、屏保、音乐中的至少一种；当所述情绪特征信息包含于所述数据库中时，显示单元于所述智能设备中显示所述推荐信息。
更进一步地，显示单元显示一个推荐信息相关预览图的在线界面之后，用户可以选择喜欢的推荐信息，并点击该推荐信息的相关按钮，此时会弹出一个选择框供用户选择，接收单元提供用户可以选择应用或不应用所选择的推荐信息。当用户选择应用时，执行单元自动下载该推荐信息应用并自动加载；当用户选择取消时，执行单元取消本次推荐的推荐信息，返回到原始状态。
进入在线界面可以让用户具有更多的选择，同时避免下载过多壁纸造成终端的内存资源占用率高的问题，达到了节省硬件资源的目的
此外，本发明实施例还提供一种计算机可读存储介质，其内存储有计算机可执行指令，上述的计算机可读存储介质例如为非易失性存储器例如光盘、硬盘、或者闪存。上述的计算机可执行指令用于让计算机或者类似的运算装置完成上述的应用信息的推荐方法中的各种操作。
本技术领域技术人员可以理解，本发明包括涉及用于执行本申请中所述操作中的一项或多项的设备。这些设备可以为所需的目的而专门设计和制造，或者也可以包括通用计算机中的己知设备。这些设备具有存储在其内的计算机程序，这些计算机程序选择性地激活或重构。这样的计算机程序可以被存储在设备(例如，计算机)可读介质中或者存储在适于存储电子指令并分别藕联到总线的任何类型的介质中，所述计算机可读介质包括但不限于任何类型的盘(包括软盘、硬盘、光盘、CD-ROM、和磁光盘)，ROM(Read-Only Memory，只读存储器)，RAM(Random Access Memory，随即存储器)，EPROM(Erasable Programmable Read-Only Memory，可擦写可编程只读存储器)，EEPROM(Electrically Erasable Programmable Read-Only Memory，电可擦可编程只读存储器)、闪存、磁性卡片或光线卡片。也就是，可读介质包括由设备(例如，计算机)以能够读的形式存储或传输信息的任何介质。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示 例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本发明的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不一定指的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。
应当注意的是，本发明的实施例有较佳的实施性，且并非对本发明作任何形式的限制，任何熟悉该领域的技术人员可能利用上述揭示的技术内容变更或修饰为等同的有效实施例，但凡未脱离本发明技术方案的内容，依据本发明的技术实质对以上实施例所作的任何修改或等同变化及修饰，均仍属于本发明技术方案的范围内。

Claims (10)

  1. 一种用于智能设备的应用程序内容的推送方法，其特征在于，所述推送方法包括：
    构建一存储有预设表情特征及至少一组情绪信息与应用程序的推荐内容之间对应关系的数据库；
    拍摄一用户的图像，获取所述图像内的当前表情特征；
    比较所述当前表情特征及所述预设表情特征，判断所述用户的当前情绪信息；
    于所述数据库中获取所述当前情绪信息对应的所述应用程序的推荐内容；
    于所述智能设备中显示所述应用程序的推荐内容。
  2. 如权利要求1所述的推送方法，其特征在于，构建一存储有预设表情特征及至少一组情绪信息与应用程序的推荐内容之间对应关系的数据库，包括：
    所述数据库构建于一服务器中；
    建立所述智能设备与所述服务器的通讯连接；
    获取所述服务器内的所述数据库并加载。
  3. 如权利要求1所述的推送方法，其特征在于，拍摄一用户的图像，获取所述图像内的当前表情特征，包括：
    在所述图像中检测所述用户的面部的位置信息；
    根据所述位置信息确定所述用户的面部图像的局部图像；
    从所述局部图像中提取所述用户的情绪特征信息。
  4. 如权利要求1所述的推送方法，其特征在于，于所述数据库中获取所述当前情绪信息对应的所述应用程序的推荐内容，包括：
    所述应用程序的推荐内容包括手机主题、屏保、音乐中的至少一种；
    当所述情绪信息包含于所述数据库中时，于所述智能设备中启动所述应用程序并提示所述应用程序的推荐内容。
  5. 如权利要求1所述的推送方法，其特征在于，于所述智能设备中显示所述应用程序的推荐内容，还包括：
    显示一请求所述用户判断是否采用所述推荐内容的判断界面；
    接收用户的确认动作；
    根据所述确认动作应用所述推荐内容或还原至原始状态。
  6. 一种用于智能设备的应用程序内容的推送装置，其特征在于，所述推送装置包括：存储模块、拍摄模块、匹配模块、获取模块、推送模块，其中，
    所述存储模块，用于构建一存储有预设表情特征及至少一组情绪信息与应用程序的推荐内容之间对应关系的数据库；
    所述拍摄模块，用于拍摄一用户的图像，获取所述图像内的当前表情特征；
    所述匹配模块，与所述拍摄模块通讯连接，用于比较所述当前表情特征及所述预设表情特征，判断所述用户的当前情绪信息；
    所述获取模块，与所述匹配模块通讯连接，用于于所述数据库中获取所述当前情绪信息对应的所述应用程序的推荐内容；
    所述推送模块，与所述获取模块通讯连接，用于于所述智能设备中显示所述应用程序的推荐内容。
  7. 如权利要求6所述的推送装置，其特征在于，所述存储模块，包括：
    数据库单元，用于将所述数据库构建于一服务器中；
    通讯单元，与所述服务器通讯连接，用于建立所述智能设备与所述服务器的通讯连接；
    加载单元，与所述通讯单元通讯连接，通过所述通讯单元用于获取所述服务器内的所述数据库并加载。
  8. 如权利要求6所述的推送装置，其特征在于，所述拍摄模块，包括：
    图像获取单元，用于在所述图像中检测所述用户的面部的位置信息；
    图像位置单元，与所述图像获取单元通讯连接，用于根据所述位置信息确定所述用户的面部图像的局部图像；
    情绪提取单元，与所述图像位置单元通讯连接，用于从所述局部图像中提取所述用户的情绪特征信息。
  9. 如权利要求6所述的推送装置，其特征在于，所述获取模块，包括：
    所述应用程序的推荐内容包括手机主题、屏保、音乐中的至少一种；
    获取单元，用于当所述情绪信息包含于所述数据库中时，于所述智能设备中启动所述应用程序并提示所述应用程序的推荐内容。
  10. 如权利要求6所述的推送装置，其特征在于，所述推送模块，包括：
    显示单元，用于显示一请求所述用户判断是否采用所述推荐内容的判断界面；
    接收单元，与所述显示单元通讯连接，用于接收用户的确认动作；
    执行单元，与所述接收单元通讯连接，用于根据所述确认动作应用所述推荐内容或还 原至原始状态。
PCT/CN2017/095659 2017-08-02 2017-08-02 一种用于智能设备的应用程序内容的推送方法及推送装置 WO2019024003A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2017/095659 WO2019024003A1 (zh) 2017-08-02 2017-08-02 一种用于智能设备的应用程序内容的推送方法及推送装置
CN201780094620.8A CN111433761A (zh) 2017-08-02 2017-08-02 一种用于智能设备的应用程序内容的推送方法及推送装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/095659 WO2019024003A1 (zh) 2017-08-02 2017-08-02 一种用于智能设备的应用程序内容的推送方法及推送装置

Publications (1)

Publication Number Publication Date
WO2019024003A1 true WO2019024003A1 (zh) 2019-02-07

Family

ID=65232098

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/095659 WO2019024003A1 (zh) 2017-08-02 2017-08-02 一种用于智能设备的应用程序内容的推送方法及推送装置

Country Status (2)

Country Link
CN (1) CN111433761A (zh)
WO (1) WO2019024003A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112883280B (zh) * 2021-03-25 2023-08-04 贵阳货车帮科技有限公司 用户推荐内容的处理系统以及方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103577516A (zh) * 2013-07-01 2014-02-12 北京百纳威尔科技有限公司 内容显示方法和装置
CN104298682A (zh) * 2013-07-18 2015-01-21 广州华久信息科技有限公司 一种基于人脸表情图像的信息推荐效果的评价方法及手机
CN104462468A (zh) * 2014-12-17 2015-03-25 百度在线网络技术（北京）有限公司 信息提供方法及装置
CN105163139A (zh) * 2014-05-28 2015-12-16 青岛海尔电子有限公司 信息推送方法、信息推送服务器和智能电视
CN106250553A (zh) * 2016-08-15 2016-12-21 珠海市魅族科技有限公司 一种服务推荐方法及终端

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103716702A (zh) * 2013-12-17 2014-04-09 三星电子（中国）研发中心 电视节目推荐装置和方法
CN105426404A (zh) * 2015-10-28 2016-03-23 广东欧珀移动通信有限公司 一种音乐信息推荐方法、装置和终端
CN105956059A (zh) * 2016-04-27 2016-09-21 乐视控股（北京）有限公司 基于情绪识别的信息推荐方法和装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103577516A (zh) * 2013-07-01 2014-02-12 北京百纳威尔科技有限公司 内容显示方法和装置
CN104298682A (zh) * 2013-07-18 2015-01-21 广州华久信息科技有限公司 一种基于人脸表情图像的信息推荐效果的评价方法及手机
CN105163139A (zh) * 2014-05-28 2015-12-16 青岛海尔电子有限公司 信息推送方法、信息推送服务器和智能电视
CN104462468A (zh) * 2014-12-17 2015-03-25 百度在线网络技术（北京）有限公司 信息提供方法及装置
CN106250553A (zh) * 2016-08-15 2016-12-21 珠海市魅族科技有限公司 一种服务推荐方法及终端

Also Published As

Publication number Publication date
CN111433761A (zh) 2020-07-17

Similar Documents

Publication Publication Date Title
US11460983B2 (en) Method of processing content and electronic device thereof
CN109857327B (zh) 信息处理装置、信息处理方法以及存储介质
US9608954B2 (en) Digital device for providing text messaging service and method for controlling the same
KR102252072B1 (ko) 음성 태그를 이용한 이미지 관리 방법 및 그 장치
CN108494947B (zh) 一种图像分享方法及移动终端
KR102369686B1 (ko) 미디어 아이템 부착 시스템
US11595489B2 (en) Selecting content for high velocity users
CN111970577A (zh) 字幕编辑方法、装置和电子设备
KR101626874B1 (ko) 이동 단말기 및 이동 단말기에서의 컨텐츠 전송 방법
US11782740B2 (en) Interface to configure media content
CN112154411A (zh) 使用音频数据进行设备控制
US11886966B2 (en) Generating data in a messaging system for a machine learning model
KR20140119611A (ko) 애플리케이션을 실행하는 방법 및 디바이스
KR20180109499A (ko) 사용자의 음성 입력에 대한 답변을 제공하는 방법 및 장치
US20180196885A1 (en) Method for sharing data and an electronic device thereof
CN109085982B (zh) 内容识别方法、装置及移动终端
US9330301B1 (en) System, method, and computer program product for performing processing based on object recognition
US20220182558A1 (en) Subtitle presentation based on volume control
CN110929176A (zh) 一种信息推荐方法、装置及电子设备
CN111385336A (zh) 页面通信方法、装置、计算机设备及存储介质
WO2019024003A1 (zh) 一种用于智能设备的应用程序内容的推送方法及推送装置
CN106776990B (zh) 一种信息处理方法、装置及电子设备
US20170024442A1 (en) Electronic device and method of acquiring user information in electronic device
CN110932964A (zh) 一种信息的处理方法及装置
US11908489B2 (en) Tap to advance by subtitles

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17919942

Country of ref document: EP

Kind code of ref document: A1