WO2018072207A1 - 一种信息推送方法、装置和*** - Google Patents

一种信息推送方法、装置和*** Download PDF

Info

Publication number
WO2018072207A1
WO2018072207A1 PCT/CN2016/102931 CN2016102931W WO2018072207A1 WO 2018072207 A1 WO2018072207 A1 WO 2018072207A1 CN 2016102931 W CN2016102931 W CN 2016102931W WO 2018072207 A1 WO2018072207 A1 WO 2018072207A1
Authority
WO
WIPO (PCT)
Prior art keywords
picture
information
push
target object
terminal
Prior art date
Application number
PCT/CN2016/102931
Other languages
English (en)
French (fr)
Inventor
高斌
Original Assignee
达闼科技（北京）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 达闼科技（北京）有限公司 filed Critical 达闼科技（北京）有限公司
Priority to PCT/CN2016/102931 priority Critical patent/WO2018072207A1/zh
Priority to CN201680002672.3A priority patent/CN107077507B/zh
Publication of WO2018072207A1 publication Critical patent/WO2018072207A1/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
    • G06F16/50—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
    • G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
    • 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 information push technology, and in particular, to an information push method, apparatus and system.
  • the existing push mode is formally set by the user to set a certain subscription type, and then the consultation of each subscription type obtained by the background is targeted to send the user to the user.
  • the subscription type For example, if the user sets the car subscription type, the background server sends the car electronic magazine to the subscriber after obtaining the car electronic magazine edited by the editor.
  • the existing information push is a broadcast-like push method. For the user, the content obtained may contain a lot of content that it does not care about, and therefore, it is a loose push method.
  • One of the technical problems to be solved by the embodiment of the present invention is how to actively send a push message to a user after uploading a picture of a certain location of the user and knowing that one or more target objects included in the picture have changed. .
  • an embodiment of the present invention provides an information push system, where the information push system includes a server and at least two terminals, where the at least two terminals include a first terminal and a second terminal;
  • the first picture uploaded by the first terminal specifically:
  • the second terminal is configured to upload a second picture
  • the server is further configured to acquire the second picture, analyze the first picture and the second picture, and confirm a similarity between target objects in the first picture and the second picture relative to a same location
  • the degree is less than the first preset threshold, and the push information corresponding to the target object is sent to the first terminal.
  • the server is further configured to:
  • the first terminal uploads the first picture
  • the uploaded content further carries the first picture related information, where the first picture related information includes content location information, picture resolution, and generation of the first picture.
  • the first picture related information includes content location information, picture resolution, and generation of the first picture.
  • the server When receiving the first picture, the server is further configured to collate the first picture and the first picture related information, and manage the first picture according to the collation result; wherein the classification management includes: Storing pictures according to one or more categories of the same type of position information, picture resolution and interval type, and generation time being the same time period type;
  • the server is further configured to obtain, according to one or more of content location information, image resolution, and generation time information of the second picture in the related information of the second picture, the first picture whose type matches.
  • the similarity between the target objects is composed of similarities between one or more regions constituting the target object
  • the server acquires the second picture, analyzes the first picture and the The second picture specifically includes:
  • the server filters the noise information in the first picture and the second picture to obtain an area corresponding to the target object in the first picture and the second picture;
  • the area similarity of the target object in the first picture and the second picture is obtained by using the feature recognition to compare the area corresponding to the target object in the first picture and the second picture.
  • an embodiment of the present invention provides an information pushing method, where the information pushing method includes:
  • the method further comprising:
  • the categorization management comprises: the same type and picture according to the location information
  • the resolution stores the picture in the same type as the interval type, and the generation time is one or more of the same time period types;
  • the method further includes:
  • the classification management specifically includes:
  • a calibration picture for calibrating the one or more target objects is confirmed from the picture library.
  • the analyzing the first picture and the second picture, and confirming that the similarity between the target object and the target object in the first picture and the second picture is less than a first preset threshold specifically:
  • the similarity between the target objects is composed of the similarity between the one or more regions constituting the target object
  • the analyzing the first picture and the second picture specifically includes:
  • the area similarity of the target object in the first picture and the second picture is obtained by using the feature recognition to compare the area corresponding to the target object in the first picture and the second picture.
  • the method further includes:
  • the generating the push information according to the one or more pictures comprises:
  • the combining the one or more push maps and the first image to obtain the push information specifically:
  • the time information of the respective pictures is attached to the push information.
  • an embodiment of the present invention provides an information pushing method, where the information pushing method includes:
  • the push message carrying the first picture, and one or more push pictures; wherein the one or more push pictures and one or more targets included in the first picture
  • the objects are in the same geographic location.
  • the method further includes:
  • the historical path data of the uploading terminal wherein the historical path data includes location information and time information of the historical path.
  • the method further includes:
  • a push message is set that does not receive the same location information as the historical track and that has a time difference from the historical track within a preset threshold.
  • the method further includes:
  • an information pushing apparatus including:
  • a first picture receiving module configured to receive a first picture uploaded by the first terminal
  • a second picture receiving module configured to receive a second picture uploaded by the second terminal
  • Pushing a message module configured to analyze the first picture and the second picture, confirm the first picture and In the second picture, the similarity between the target objects with respect to the same location is less than the first preset threshold, and the push information corresponding to the target object is sent to the first terminal.
  • the present invention further provides an information pushing device, where the information pushing device includes:
  • the uploading module is configured to upload the first picture and the first picture related information
  • Receiving a push message module configured to receive a push message, where the push message carries the first picture, and one or more push pictures; wherein the one or more push pictures and the first picture One or more of the included target objects are in the same geographic location.
  • the embodiment of the present invention further provides an electronic device, which is used to implement the steps of the information pushing method according to any one of the second aspect or the third aspect, including:
  • At least one processor and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being stored by the memory
  • An instruction program executed by at least one processor the program being set by the program to perform the information pushing method step of any one of the second aspect or the third aspect.
  • the embodiment of the present invention further provides a non-volatile computer storage medium, wherein the computer storage medium stores computer executable instructions that are executed by one or more processors.
  • the target object included in the second picture uploaded by the second terminal is compared with whether the first picture changes in the first picture.
  • the target object refers to an object that is in the same position (usually referred to as a geographical location) in reality. If the change occurs, the generated push message is sent to the first terminal, so that the first user can timely obtain the latest state after the target object changes in the first picture uploaded by the history.
  • the first image uploaded by the first terminal be associated with the target object of interest to the first user, but also the change of the target object can be identified according to the image uploaded by other users in the system, so that the user interested information can be accurately fed back. Helps improve the accuracy and effectiveness of push information.
  • FIG. 1 is a schematic structural diagram of an information push system according to an embodiment of the present invention.
  • FIG. 2 is a schematic structural diagram of a target object distribution diagram according to an embodiment of the present invention.
  • FIG. 3 is a schematic diagram showing a photographing azimuth angle according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of a target object obtained by taking azimuth angle of shooting according to an embodiment of the present invention
  • FIG. 6 is a schematic diagram showing a shooting azimuth angle according to an embodiment of the present invention.
  • FIG. 7 is a schematic diagram of a target object obtained by taking azimuth angle of shooting according to an embodiment of the present invention.
  • FIG. 8 is a flowchart of an information pushing method according to an embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 10 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 11 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 12 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 13 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 14 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 15 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • 16 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 17 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 18 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 19 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 20 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
  • the information push system includes a server 1, at least two terminals (for example, terminals 2, 3, and 4 shown in FIG. 1).
  • the at least two terminals include the first terminal 2 and the second terminal 3.
  • Different users can use different smart terminals to connect to the server 1.
  • the way to authenticate users is the personal account created by the user. Therefore, even the same user can use different smart terminals to complete personal account login. Therefore, in the embodiment, the first terminal 2 generally refers to one or more terminals that belong to the same user, and the message transmission between the server 1 and the user is completed by using the smart terminal. Sign in to the current user's personal account.
  • the information pushing system in this embodiment is specifically implemented as:
  • the server 1 stores a first picture uploaded by the first terminal 2;
  • the first picture may be obtained by taking a smart phone or a SLR as the first terminal 2, or may be downloaded from the network by the first terminal 2, or may be obtained by software synthesis in the first terminal 2.
  • the method for uploading the server 1 includes uploading the first picture to the circle of friends, backing up the first picture to the web server 1, loading the first picture into an album in its own space, and the like.
  • the server 1 receives a picture uploaded by each terminal, and the server 1 directly receives a picture uploaded by each terminal, for example, the server 1 serves as a data backup server of the user, and the server serves as a friend circle server of the user. and many more.
  • the server 1 may also indirectly obtain a picture uploaded by the terminal from a third-party server, where the third-party server may be a server for maintaining the microblog, a server for managing the circle of friends, a server for managing the intranet, and the like. Wait.
  • the terminal described in the embodiments of the present invention uploads a picture to the server 1 , and the server 1 directly receives the picture uploaded by the terminal and the server 1 indirectly obtains the picture uploaded by the terminal from the third-party server.
  • the second terminal 3 is configured to upload a second picture
  • the second terminal 3 may be completed by directly uploading the second picture to the server 1; or the second terminal 3 may first upload a second picture to the third-party server, and then the third terminal will perform the The second picture is forwarded to the server 1.
  • the server 1 is further configured to acquire the second picture, analyze the first picture and the second picture, and confirm that the first picture and the second picture are in a position relative to a target object in the same location
  • the similarity is less than the first preset threshold, and the push information corresponding to the target object is sent to the first terminal 2.
  • the first preset threshold may be set by a background operator, for example, the first preset threshold is set to 30%-50%.
  • the one or more pictures whose similarity is less than the first preset threshold represent a different state from the target object in the same position in the first picture.
  • the first preset threshold may also be adjusted according to the evaluation feedback given by the first terminal 2 after receiving the recommendation message (usually completed by the first user to operate the first terminal 2). For example, if the first terminal 2 feeds back the push message is valid, the first preset threshold may be retained or the parameter value of the first preset threshold may be further increased; if the first terminal 2 feeds back the push message is inaccurate, Then, the parameter value of the first preset threshold can be lowered.
  • the first preset threshold generally determines the size of the change of the target object. If the first preset threshold is set to be larger, it indicates that the change of the target object is smaller, and therefore, the existence of the false positive is easily caused.
  • the misjudgment refers to the fact that the picture of the target object that has not undergone substantial change may be pushed to the user because the picture pixel or the partial mask causes the judgment error.
  • the target object in the same location refers to an object at the same position in the real environment recorded by the first picture and the second picture.
  • the target object can be generally designated as a building in a picture, a sign in a picture, a street in a picture, a storefront in a picture, and the like having an invariant characteristic for a certain period of time; and for a character, a vehicle, an animal, a weather, etc.
  • the server 1 does not default.
  • a dynamic object that does not have an invariant characteristic for a certain period of time is used as the target object.
  • the target object and the dynamic object, as well as other recognizable content, may be collectively referred to as an object in various embodiments of the present invention.
  • the confirmation of the target object in the same location includes multiple manners.
  • the location information of the corresponding first image and the second image may be obtained according to the location.
  • the information determines the target object at the same position in the first picture and the second picture.
  • the location information includes not only latitude and longitude positioning information, but also azimuth information of the photographing.
  • the photographed azimuth angle information can further improve the recognition accuracy of the target object.
  • the image extraction method may be used to first perform contour extraction in the image to divide the image into one or more objects. (including the above target object and dynamic object, etc.) and filter out one or more target objects; then, according to the target objects of the one or more target objects (such as vertices, center points, inflection points, etc.)
  • a distribution map (the effect diagram thereof is shown in FIG. 2, wherein the irregular pattern is the target object distribution map), and the target object at the same position in the first picture and the second picture is identified.
  • the contour extraction in the image is first performed, the image is divided into one or more objects (including the above target object and dynamic object, etc.), and one or more target objects are filtered out, according to the image.
  • the target contour quickly matches one or more target objects in the first picture and the second picture whose similarity is greater than the second preset threshold (for example, two target object similarities of 1 and 2 shown in FIG. 2 Respectively larger than the corresponding target objects 1 and 2 in the other graph.
  • One or more target objects here are members of the target object distribution map, and then the target object is combined based on the target object.
  • the image processing mode and the target recognition algorithm may separately process a picture that does not carry location information, or may be used as an auxiliary verification method when carrying a picture analysis of location information.
  • the target object profile can complete the angular rotation of the target object profile in combination with the azimuth angle information of the captured position information, so that the target objects in the two images with different shooting angles can be more accurately identified. As shown in FIG. 3, the azimuth angle of the photographing is 30 degrees east and south, and the corresponding rotation method can refer to the existing algorithm formula, which will not be repeated here.
  • the embodiment of the present invention determines the second terminal by analyzing the pictures uploaded by different terminals (including the first terminal and the second terminal involved in the embodiment) according to the similarity between the target objects in the same location in the reality included in the image. 3 uploading the target object of the same location included in the second picture compared to whether the first picture changes. If the change occurs, the generated push message is sent to the first terminal 2, so that the first user can timely obtain the latest state after the target object changes in the first picture uploaded by the history.
  • the first image uploaded by the first terminal 2 can be associated with the target object that is of interest to the first user, and the target object can be identified according to the image uploaded by other users in the system, so that the user interested information can be accurately fed back. Helps improve the accuracy and effectiveness of push information.
  • the similarity between the target objects at the same location is smaller than the first preset threshold, and the push information corresponding to the target object is sent to the The first terminal 2 is described.
  • the push information may be a message including one or more of the target object name, the target object change time, and the link address of the browse second image.
  • an extension scheme is also provided in the embodiment of the present invention, specifically:
  • the server 1 searches for one or more pictures whose similarity is greater than the second preset threshold, compared with the feature information, according to the feature information of the target object included in the second picture.
  • the second preset threshold may be set by a background operator, for example, the second preset threshold is set to be 60%-80%.
  • the one or more pictures whose similarity is greater than the second preset threshold are usually in the same state as the target object in the second picture, and are uploaded to the server 1 by other terminals.
  • the simplest method is to set the one or more pictures in a corresponding order, and carry them in the push message to the first terminal 2, so that the first terminal 2 receives the push message after receiving the push message. Can be displayed in the display screen Displaying the one or more pictures to the first user.
  • the server 1 analyzes the first picture and the second picture, so how to locate the corresponding first picture in the huge picture library more efficiently is preferred. To overcome the problem.
  • a method of how to manage a picture library is also provided in connection with embodiments of the present invention. specific:
  • the first terminal 2 uploads the first picture, and the uploaded content further includes first picture related information, where the first picture related information includes content location information, picture resolution, and generation time information of the first picture.
  • first picture related information includes content location information, picture resolution, and generation time information of the first picture.
  • the server 1 When receiving the first picture, the server 1 is further configured to collate the first picture and the first picture related information, and manage the first picture according to the collation result; wherein the classification management includes : storing pictures according to one or more categories of the same type of position information, picture resolution and interval type, and generation time being the same time period type;
  • the server 1 is further configured to obtain, according to one or more of the content location information, the image resolution, and the generation time information of the second image in the related information of the second picture, the first picture whose type matches.
  • the proofreading may be performed on the first picture edited by the first terminal 2 in addition to one or more of content location information, picture resolution, and generation time information of the first picture.
  • a picture related information for example, a title of the first picture, location information of the first picture input by the first user (especially the manner in which the user takes a screenshot), and the like.
  • the proofreading process may further include confirming the location information in the picture, that is, confirming the location information in the picture by an operation similar to the matching of the target object described above.
  • the manner of the classification management includes classifying the first picture into a group of pictures with the same location information to complete storage.
  • the image can be stored together in a set of pictures having common location information through the proofreading process, so that the range of the image can be analyzed according to the effective reduction of the position information when the image is subsequently analyzed. (quantity).
  • the embodiment of the present invention further provides an information pushing method, which may be executed by the server 1 in the information pushing system in Embodiment 1.
  • the first terminal and the second terminal are merely named for convenience of distinguishing the description, and the first terminal and the second terminal corresponding to the home terminal are referred to as a first user and a second user.
  • the second terminal may represent the uploaded picture function in reality. Any terminal owned by the second user, and the first terminal has the function of uploading pictures described by the second terminal, and the first user is usually the subscriber of the server push message in this embodiment (the subscription may be the first user) Proactively initiated, it can also be that the server defaults to the first user's own).
  • the information pushing method includes:
  • step 201 the first picture uploaded by the first terminal is received.
  • the first picture uploaded by the first terminal received by the server may be a method for directly receiving the picture uploaded by the terminal by using the server described in Embodiment 1, and may also be indirectly from the third-party server by using the server described in Embodiment 1. Get the image uploaded by the terminal.
  • the first picture may be obtained by taking a smart phone or a DSLR as the first terminal; the first picture may also be downloaded by a personal computer, a tablet computer or a smart phone network as the first terminal; The first picture may also be generated by the first terminal by using a screenshot.
  • step 202 a second picture uploaded by the second terminal is received.
  • step 203 the first picture and the second picture are analyzed, and it is confirmed that the similarity between the target objects in the first picture and the second picture is less than the first preset threshold, and the corresponding correspondence is sent.
  • the push information of the target object is sent to the first terminal.
  • the target object in the same location refers to a target object in the same location in the real environment recorded by the first picture and the second picture.
  • the target object generally refers to a building in a picture, an identification card in a picture, a street in a picture, a storefront in a picture, and the like, and has an invariant characteristic for a certain period of time.
  • the server usually does not use the dynamic object that does not have the invariant characteristics for a certain time as the target object.
  • dynamic objects are collectively referred to as objects.
  • the dynamic object is also referred to as noise information with respect to the target object.
  • the confirmation of the target object in the same location includes multiple manners.
  • the location information may be obtained according to the location information. Determining the target object at the same position in the first picture and the second picture.
  • the location information includes not only the latitude and longitude positioning information, but also the azimuth angle information of the shooting, and the captured azimuth angle information can further improve the recognition accuracy of the target object, specifically: according to the azimuth angle information of the shooting. And adjusting a rotation angle of the target object distribution map in the image processing manner in the first picture and/or the second picture.
  • the azimuth angle information of the photographing may be calculated according to a GPS locator and/or a gyroscope of the smart phone.
  • the object area calculated by using the target recognition algorithm in the first picture and/or the second picture may be adjusted according to the captured azimuth angle information, for example, if the second picture is relative to the first picture.
  • the latitude and longitude in the position information is the same, and the azimuth angle information of the captured position is different by 30 degrees, and when the target object with the highest similarity and the same position in the first picture and the second picture is matched, it is taken from the first picture or Taking the feature area of the target object in the second picture, it is necessary to perform the ignoring process of the corresponding partial feature area, which is the target object and the second object in the first picture generated by the azimuth angle information of the shooting being different by 30 degrees.
  • the information contained in the target object in the image is not equal.
  • the shooting angle shown in FIG. 3 the target object effect is as shown in FIG. 5, and for the shooting angle shown in FIG. 6, the target object imaged as shown in FIG. 7 is shown in FIG.
  • the shooting angle is that when the target object matching of FIG. 5 and FIG. 7 is performed, the shadow feature area is ignored, and the matching process of the target object similarity is not participated. Thereby, the information of the target object between the first picture and the second picture due to the shooting angle is not equal.
  • the image processing manner may be specifically performed: first, contour extraction in the image is performed, and the image is divided into One or more objects (including the above-mentioned target object and dynamic object, etc.), and one or more target objects are filtered out; and then, according to the target object distribution map composed of the center points of the one or more target objects, the first A target object in the same position in a picture and a second picture.
  • the contour extraction in the image is first performed, the image is divided into one or more objects (including the above target object and dynamic object, etc.), and one or more target objects are filtered out, according to the image.
  • the target contour quickly matches the target object of the same position in the first picture and the second picture with the similarity greater than the preset threshold, and then based on the target object combined with the target object distribution map, the first picture and the first picture can be found faster.
  • the result of the target object of the same location in the second picture and the similarity is less than the first preset threshold.
  • the target object of the same location included in the second picture uploaded by the second terminal is compared with the first object. Whether the picture has changed.
  • Push generated if there is a change The message is sent to the first terminal, so that the first terminal can timely obtain the latest state after the target object changes in the first picture uploaded by the history.
  • the target object that is of interest to the user can be associated with the first picture uploaded by the user, and the change of the target object can be recognized according to the picture uploaded by the other terminal, so that the information of interest of the user can be accurately feedbacked, which helps to improve the push information. Accuracy and effectiveness.
  • the number of pictures that the server can receive from each terminal may be very large in a period of time. Therefore, it is extremely necessary to provide a step-by-step screening mechanism and a classification management method.
  • the method for the step-by-step screening mechanism and the categorization management is indispensable for the server to obtain the first picture related information while receiving the first picture uploaded by the first terminal, where the first picture related information includes The one or more of the content location information, the image resolution, and the generation time information of the first image, the method further includes:
  • the categorization management comprises: the same type and picture according to the location information
  • the resolution stores the picture in the same type as the interval type, and the generation time is one or more of the same time period types;
  • the method further includes:
  • the proofreading may be performed on the first picture edited by the first terminal for the first picture, in addition to one or more of the content location information, the picture resolution, and the generation time information of the first picture.
  • the picture related information for example, the title of the first picture, the location information of the first picture input by the first user (especially the manner in which the user takes a screenshot), and the like.
  • the proofreading process may further include confirming the location information in the picture, that is, confirming the location information in the picture by an operation similar to the matching of the target object described above.
  • the corresponding calibration picture is usually confirmed by combining the related information of the first picture, and is completed by matching the target object in the calibration picture and the target object of the first picture. It is equivalent to synchronizing the matching process in the proofreading process and the matching process in the method step.
  • the server side staff can also be added to complete the proofreading process.
  • the manner of the categorization management includes: if the first picture related information includes the content location information of the first picture, classifying the first picture to a group of pictures having the same location information And storing a picture library according to the feature information of the one or more target objects and the first picture related information included in the first picture, and confirming from the picture library for calibrating the one or more target objects Calibration picture.
  • the calibration picture contains more comprehensive information of the corresponding target object, and a simple understanding is that the pixel reaches a specified value, and the picture with a sufficiently small noise relative to the target object.
  • the calibration picture may be generated by a person shooting, or may be selected from a plurality of users taking pictures of the same location target object.
  • the calibration picture may be multiple, and each calibration picture includes one or more target objects in the first picture. Comparing the target object in the first picture, the target object information included in the corresponding calibration picture is more comprehensive, and the feature information of the target object using the calibration picture is used as the feature information of the target object in the first picture to participate in subsequent reception by the server.
  • the analysis of each picture (where the subsequent received pictures may include the second picture) will be more efficient and accurate.
  • the step-by-step screening mechanism includes: the server first determines whether the location information in the picture related information matches the location information of the first picture for the received picture, and if not, does not match; if the match, the call is performed. The first picture or the calibration picture of the first picture completes confirmation of the target object. Next, if it is determined that the picture related information does not include the location information, the target object is confirmed by the image processing method and/or the target recognition algorithm, and details are not described herein.
  • the similarity between the target objects and the target in the same position is less than the first preset threshold, and specifically includes:
  • step 2031 the location information corresponding to the first picture and the second picture is analyzed, and it is determined that the first picture and the second picture are in the same geographic location. If it is determined that the first picture and the second picture are in different geographical locations, the subsequent execution steps are skipped, and the process is directly ended.
  • step 2032 matching the second picture and calibrating the calibration picture of one or more target objects in the first picture, confirming that the first picture and the second picture are in a position relative to the target object in the same position
  • the similarity is less than the first preset threshold.
  • the use of the calibration image is introduced and the storage is completed according to the location information.
  • the method further includes: sending the push information corresponding to the target object to the first terminal, the method further includes:
  • step 204 the user corresponding to one or more pictures is sent to the calibration picture of the first picture to send the recommendation message.
  • the server uploads the picture uploaded by each terminal, usually after the school contains the location object related to the reality, and the server can associate one or more calibration pictures by the classification management.
  • the purpose is originally to be able to be more precise and efficient when subsequently analyzing pictures uploaded by other terminals like the second picture.
  • the association relationship between the terminal uploading picture and the calibration picture is further utilized, which is equivalent to calibrating a set of pictures that are in the same period as the target object state included in the first picture by the calibration picture.
  • the analysis of the first picture and the second picture (described by using the second picture as an example) in the set of pictures can complete the sending of the push message of the user corresponding to the picture in the entire group, which not only reduces the number of analysis (ie, It is not necessary to take each picture in the group and analyze the second picture, and it helps to improve the push efficiency.
  • the uploading of the image by the terminal is used as a basis for judging, since a user may complete uploading of different images in different locations through different terminals at different times, the user account is still used as a specific implementation process.
  • the identification basis of the picture uploaded by the terminal to the server that is, the picture in the server does not use different terminal IDs as the basis for distinguishing the picture attribution, but the user account registered by the terminal is used as the basis for distinguishing whether the picture is from the same user.
  • step 203 when performing the step of "sending the push information corresponding to the target object to the first user" in step 203, there is a preferred implementation, specifically:
  • step 301 based on the feature information of the target object included in the second picture, one or more pictures whose similarity is greater than the second preset threshold is compared with the feature information.
  • the second preset threshold may be set by a background operator, for example, the second preset threshold is set to be 60%-80%.
  • the one or more pictures whose similarity is greater than the second preset threshold are usually in the same state as the target object in the second picture, and are uploaded to the server by other terminals.
  • step 302 push information is generated based on the one or more pictures.
  • the preferred implementation can generate the push information together with the calibration picture in the manner of the above-mentioned classification management, that is, the one or more pictures and one or more calibration pictures are simultaneously included in the push information.
  • step 302 of the preferred implementation there is a method for generating an implementable recommendation message, which specifically includes:
  • step 3021 one or more push pictures whose picture generation time is separated by a specified duration is filtered out; and/or, the visual angle of the object included in the picture is selected to be the same object in the first picture.
  • the visual angle of the object included in each of the objects can be confirmed by the azimuth angle information of the above shooting, and details are not described herein again.
  • step 3022 the one or more push maps and the first picture are combined to obtain push information.
  • the information pushing method according to the second embodiment of the present invention is described from the perspective of the server side, and the third embodiment of the present invention is developed from the user side to explain how to implement the information pushing method proposed by the present invention.
  • Methods include:
  • step 401 the first picture and the first picture related information are uploaded.
  • the first picture includes one or more target objects, where the picture related information includes one or more of content location information, picture resolution, and generation time information of the first picture.
  • the uploading method includes uploading the first picture to a circle of friends, backing up the first picture to a web server, loading the first picture into an album in its own space, and the like.
  • a push message is received, where the push message carries the first picture and one or more push pictures; wherein the one or more push pictures and the first picture are included One or more target objects are in the same geographic location.
  • the optional information of the push information further includes time information of each push picture.
  • the method further includes step 403, specifically:
  • the one piece is arranged on the push interface according to the time information of each push picture. Or multiple push pictures.
  • the first picture is arranged at a starting position of the one or more push pictures, or the first picture is arranged in a top manner on the one or more push pictures.
  • the historical path data of the uploading terminal wherein the historical path data includes location information and time information of the historical path.
  • the terminal is one or more terminals that belong to the user, and the terminal includes the first terminal described in Embodiment 1-2, and the user is specifically the first user in Embodiment 1-2.
  • the user may set a push message that does not receive the same location information as the historical path and that has a time difference from the historical path within a preset threshold.
  • the preset threshold may be set by the user, and is usually set to 1 day, 1 week or 1 month. The longer the preset threshold set by the user, the smaller the degree of the user can receive the push message, so as to meet the push message requirements of different users.
  • a time parameter may also be set.
  • the time parameter is used by the server to determine that the current time meets the time parameter, one or more push messages corresponding to the user are packaged and sent, and the purpose of personalizing the push message can also be achieved.
  • Both of the above methods involve delayed transmission of a push message, and if the delayed transmission request is combined with the calibration picture described in Embodiment 2, a set of pictures that are in the same period as the target object included in the first picture are Completing the analysis of the first picture and the second picture in the set of pictures (in which the second picture is taken as an example), the generation of the push message of the user corresponding to the picture in the entire group can be completed.
  • the generated push message may be temporarily stored, and the corresponding push message is sent after the preset threshold or time parameter is met.
  • the user in addition to setting the number of push pictures carried in the push message, the user may also set one or more of the push picture resolution, the push picture source, and the target object angle in the push picture. item.
  • the user can set the resolution of the image pushed by the server to 1280*1960 or the resolution is less than 760*1280, etc. by logging in to the platform provided by the server.
  • the resolution setting of the picture is mainly based on the habits and requirements of the user browsing the picture, and the traffic status owned by the user.
  • the user intelligent terminal network status monitoring function uses a high-resolution recommendation map when confirming that the current network status of the user's smart terminal is connected to a wifi or a wired network; when confirming that the current network status of the user's intelligent terminal is a mobile network (for example, 3G)
  • 3G mobile network
  • 4G networks use a low resolution recommendation map.
  • the resolution of the one or more recommended pictures carried in the recommendation message received by the first terminal is not the resolution of the original picture stored in the server, but is compressed; the compressed picture is Forming part of the pixel information in the original picture, and being able to specify the recommended picture in the first user click recommendation message display interface, the recommended picture of the user selected by the server is sent to the user by the compressed part, thereby Clicking on the recommended image generates a complete picture of the pixel information.
  • the user can set to receive only the picture uploaded by the specified type object as a push picture source, the specified type object includes a third party company, a friend of the user, a person in the designated group, and a designated campus area. Personnel, personnel within the designated company, etc.
  • the user can set the server to preferentially select one or more pictures including the specified angle range of the target object as the push picture for generating the push message.
  • the embodiment of the present invention further provides an information pushing device, where the information pushing device can be used to run the information pushing method described in Embodiment 2.
  • the information pushing device may be a server, a home computer with sufficient computing power, and a virtual execution excuse in the big data cloud, as shown in FIG. Push devices include:
  • a first picture receiving module configured to receive a first picture uploaded by the first terminal
  • a second picture receiving module configured to receive a second picture uploaded by the second terminal
  • a push message module configured to analyze the first picture and the second picture, and confirm that the similarity between the target object and the target object in the first picture and the second picture is less than a first preset threshold, and send a corresponding The push information of the target object is sent to the first terminal.
  • the first picture receiving module and the second picture receiving module are respectively connected to the push message module.
  • the push message module is used to analyze the similarity of the target object in the same location in the reality image included in the image uploaded by the terminal, and the second terminal uploads the target object in the same location included in the second image to compare the first image. Whether it has changed. If the change occurs, the generated push message is sent to the first terminal, so that the first user can obtain the first time of the historical upload. The latest state of the target object after the change in the image.
  • the target object that is of interest to the first user can be associated with the first image uploaded by the first terminal, and the change of the target object can be identified according to the image uploaded by other terminals in the system, so that the first user interested information can be accurately fed back. To help improve the accuracy and effectiveness of push information.
  • the first picture receiving module obtains the first picture related information, and also obtains the first picture related information, where the first picture receiving module
  • the first picture related information includes one or more of content location information, picture resolution, and generation time information of the first picture.
  • the apparatus further includes a classification management module, where the The class management module is connected to the first picture receiving module and the second picture receiving module, and the first picture receiving module and the second picture receiving module are connected to the push message module via the categorization management module, specifically:
  • the categorization management module is configured to collate the first picture and the first picture related information, and manage the first picture according to the collation result.
  • the categorization management refer to the content described in Embodiment 2, and details are not described herein.
  • the categorization management module further includes a feature retrieval module and a target object calibration module, as shown in FIG.
  • the feature retrieval module is configured to retrieve a picture library according to the feature information of the one or more target objects and the first picture related information included in the first picture;
  • the target object calibration module is configured to confirm, from the picture library, a calibration picture for calibrating the one or more target objects.
  • the push message module further includes a location information confirmation module and a calibration picture matching module, and specifically includes:
  • the location information confirming module is configured to analyze location information corresponding to the first picture and the second picture, and determine that the first picture and the second picture are in the same geographic location;
  • the calibration picture matching module is configured to match the second picture and calibrate a calibration picture of one or more target objects in the first picture, and confirm that the first picture and the second picture are relative to the same location The similarity between the target objects is less than the first predetermined threshold.
  • the similarity between the target objects is composed of similarities between one or more regions constituting the target object
  • the push message module further includes:
  • the screening module is configured to filter noise information in the first picture and the second picture to obtain a An area corresponding to the target object in a picture and a second picture;
  • the area similarity calculation module obtains an area corresponding to the target object in the first picture and the second picture by using a feature recognition to obtain an area corresponding to the target object in the first picture and the second picture. Similarity.
  • the push message module further includes a push picture matching module and a push message generating module, as shown in FIG.
  • the push picture matching module is configured to search, according to the feature information of the target object included in the second picture, one or more pictures whose similarity is greater than a second preset threshold compared with the feature information. ;
  • the push message generating module is configured to generate push information according to the one or more pictures.
  • the push message generating module further includes a first screening module and/or a second screening module, as shown in FIG.
  • the first screening module is configured to filter one or more push pictures whose picture generation time is separated by a specified duration
  • the second screening module is configured to filter out one or more push maps whose visual angle of the object included in the image is the same as and/or complementary to the visual angle of the same object in the first image;
  • the one or more push maps are used to generate the push information.
  • the embodiment of the present invention further provides an information pushing device, and the information pushing device can be used to run the information pushing method described in Embodiment 3.
  • the information pushing device may be a smart phone, a tablet computer, a personal PC, or the like.
  • the information pushing device includes:
  • the uploading module is configured to upload the first picture and the first picture related information
  • Receiving a push message module configured to receive a push message, where the push message carries the first picture, and one or more push pictures; wherein the one or more push pictures and the first picture One or more of the included target objects are in the same geographic location.
  • the embodiment of the invention can ensure that the first user obtains the latest state after the target object changes in the first picture uploaded by the history in time by receiving the push message module. Not only through the first terminal The first picture transmitted is associated with the target object of interest to the first user, and the change of the target object can be recognized according to the picture uploaded by other terminals in the system, so that the information of the first user can be accurately fed back, which helps to improve the push. The accuracy and validity of the information.
  • the uploading module further includes:
  • the historical activity data uploading module is configured to upload history path data of the terminal, wherein the historical path data includes location information and time information of the historical path.
  • the device further includes a first setting module, specifically:
  • the first setting module is configured to set a push message that does not receive the same location information as the historical path and that has a time difference from the historical path within a preset threshold.
  • the apparatus further includes a second setting module, specifically:
  • the second setting module is configured to set one or more of a number of push pictures carried in the push message, a push picture resolution, a push picture source, and a target object angle in the push picture.
  • the embodiment of the present invention further provides an electronic device, which is used to implement the steps of the information pushing method according to any one of Embodiments 2-3. As shown in FIG. 20, the method includes:
  • At least one processor and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being stored by the memory
  • An instruction program executed by at least one processor the program being set by the program to perform the information pushing method step of any one of the items proposed in Embodiment 2-3.
  • FIG. 20 a processor is used as an example, and the processor and the memory may be connected by a bus or other manner, and FIG. 20 is taken as an example through a bus connection.
  • the memory is a nonvolatile computer readable storage medium and can be used for storing a nonvolatile software program, a nonvolatile computer executable program, and a module, such as the information pushing method and corresponding in the second and third embodiments.
  • Program instructions/modules (for example, the individual modules shown in Figure 9-15).
  • Processor pass The non-volatile software program, the command, and the module stored in the memory are executed to execute various functional applications and data processing of the information push device, that is, the information push method according to the second embodiment and the respective modules of the fourth embodiment are realized.
  • the memory may include a high speed random access memory, and may also include a non-volatile memory such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device.
  • the memory optionally includes a memory remotely located relative to the processor, the remote memory being connectable to the processor over a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
  • the program instructions/modules are stored in the memory, and when executed by the one or more processors, perform the information pushing method in the second embodiment described above, for example, executing each of the above-described FIG. Steps; can also implement the various modules and units described in FIG. 9-15; or perform the information pushing method in the third embodiment described above, for example, perform the steps shown in FIG. 8 described above; -19 each module, unit.
  • the social device of the embodiments of the present invention may exist in various forms, performing the steps illustrated in FIG. 4 described above; and the various modules, units illustrated in FIGS. 9-15, or the steps illustrated in FIG. 8; In the case of each module and unit described in 16-19, the social device includes but is not limited to:
  • Mobile communication devices These devices are characterized by mobile communication functions and are mainly aimed at providing voice and data communication.
  • Such terminals include: smart phones (such as iPhone), multimedia phones, functional phones, and low-end phones.
  • Ultra-mobile personal computer equipment This type of equipment belongs to the category of personal computers, has computing and processing functions, and generally has mobile Internet access.
  • Such terminals include: PDAs, MIDs, and UMPC devices, such as the iPad.
  • Portable entertainment devices These devices can display and play video content, and generally have mobile Internet access. Such devices include: video players, handheld game consoles, as well as smart toys and portable car navigation devices.
  • the server consists of a processor, a hard disk, a memory, a system bus, etc.
  • the server is similar to a general-purpose computer architecture, but because of the need to provide highly reliable services, processing power and stability High reliability in terms of reliability, security, scalability, and manageability.
  • the embodiment of the invention further provides a non-volatile computer storage medium, wherein the computer storage medium comprises a read only memory (ROM), a random access memory (RAM), a disk or CD, etc.
  • the computer storage medium stores computer executable instructions that are executed by one or more processors, such as one of the processors of FIG. 20, that can cause the one or more processors to perform the second embodiment
  • the information pushing method in, for example, performing the various steps shown in FIG. 4 described above; the various modules and units described in FIG. 9-15 may also be implemented; or the information pushing method in the third embodiment may be performed, for example, performing The various steps shown in FIG. 8 described above; the various modules and units described in FIGS. 16-19 can also be implemented.

Landscapes

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

Abstract

本发明涉及信息推送技术领域，提供了一种信息推送方法、装置和***。所述***包括第一终端历史上传有第一图片；服务器用于获取各终端上传的图片；第二终端上传第二图片；服务器获取第二图片，分析第一图片和第二图片，确认第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，则发送对应目标对象的推送信息给第一终端。本发明实施例通过分析不同终端上传的图片，根据图片中所包含的目标对象间的相似度，来判断第二终端上传第二图片包含的同一位置的目标对象相比较第一图片中的是否发生变化，对于发生变化的将推送消息发送给第一终端，从而保证第一用户能够及时的获取第一图片中目标对象发生变化后的状态。

Description

一种信息推送方法、装置和*** 【技术领域】
本发明涉及信息推送技术领域，特别是涉及一种信息推送方法、装置和***。
【背景技术】
随着社会的发展，一直伴随着有信息的推送，例如：传统的订阅报纸、订阅期刊等等。随着互联网的普及，信息推送的方式也从实体转换为类似微信里的公众订阅号、互联网平台中的咨询订阅等等。现有的推送方式从形式上来说，都是通过用户设定某一订阅类型，然后，由后台有针对性的将其获取到的各订阅类型的咨询有针对性的向所述用户发送其设定的该订阅类型。例如：用户设定了汽车订阅类型，则后台服务器在获取到编辑人员编辑完成的汽车电子杂志后，便将该汽车电子杂志发送给该订阅用户。现有的这种信息推送是一种类似广播的推送方式，对于用户来说其获取到的内容中可能包含许多其并不关心的内容，因此，是一种松散的推送方式。
用户出行时拍照留念或者拍照晒朋友圈是当前很流行的事，其中，除了晒出自己的美颜以外，还为了表达自己到过某处地方或则参观过某处景点。发明人发现，随着城市发展速度的加快，我们周边的环境也在日新月异，对于一个用户曾经拍照留念过的地方若发生重建或者改修，通常情况下该用户是无法获知到的。而现有的信息推送方法是无法将该地方重建或者改修后的环境信息，推送给历史上在该地方曾拍照的用户。
【发明内容】
本发明实施例要解决的其中一个技术问题是如何就用户上传的某一位置的图片，并在获知该图片中所包含的一个或者多个目标对象发生变化后，能够主动向该用户发送推送消息。
本发明实施例采用如下技术方案：
第一方面，本发明实施例提供了一种信息推送***，所述信息推送***包括服务器、至少两个终端，其中，所述至少两个终端包括第一终端和第二终端；所述服务器存储第一终端上传的第一图片，具体的：
所述第二终端，用于上传第二图片；
所述服务器，还用于获取所述第二图片，分析所述第一图片和所述第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，发送对应所述目标对象的推送信息给所述第一终端。
优选的，所述服务器还用于：
根据所述第二图片中所包含的所述目标对象的特征信息，搜索与之特征信息相比，相似度大于第二预设阈值的一张或者多张图片；
根据所述一张或者多张图片生成所述推送信息。
优选的，所述第一终端上传第一图片同时，上传的内容中还携带第一图片相关信息，其中所述第一图片相关信息包括所述第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，具体的：
所述服务器在接收到所述第一图片时，还用于校对所述第一图片与所述第一图片相关信息，并根据校对结果归类管理所述第一图片；其中归类管理包括：根据位置信息相同类型、图片分辨率同区间类型、生成时间为同一时段类型中的一项或者多项分类存储图片；
所述服务器，还用于根据第二图片的相关信息中的第二图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，获得类型相匹配的第一图片。
优选的，目标对象之间的相似度，是由构成所述目标对象的一个或者多个区域之间的相似度组成，所述服务器获取所述第二图片，分析所述第一图片和所述第二图片，具体包括：
服务器筛选所述第一图片和第二图片中的噪音信息，得到第一图片和第二图片中对应所述目标对象的区域；
利用特征识别比对所述第一图片和第二图片中对应所述目标对象的区域，得到所述第一图片和第二图片中，对应所述目标对象的区域相似度。
第二方面，本发明实施例提供了一种信息推送方法，所述信息推送方法包括：
接收第一终端上传的第一图片；
接收第二终端上传的第二图片；
分析所述第一图片和第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，发送对应所述目标对象 的推送信息给所述第一终端。
优选的，接收第一终端上传的第一图片同时，还获取到第一图片相关信息，其中所述第一图片相关信息包括所述第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，所述方法还包括：
接收到所述第一图片时，校对所述第一图片与所述第一图片相关信息，并根据校对结果归类管理所述第一图片；其中归类管理包括：根据位置信息相同类型、图片分辨率同区间类型、生成时间为同一时段类型中的一项或者多项分类存储图片；
所述分析所述第一图片和所述第二图片之前，还包括：
根据第二图片的相关信息中的第二图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，获得类型相匹配的第一图片。
优选的，所述归类管理具体包括：
根据第一图片中所包含的一个或者多个目标对象的特征信息和第一图片相关信息检索图片库；
从所述图片库中确认用于标定所述一个或者多个目标对象的标定图片。
优选的，所述分析所述第一图片和第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，具体包括：
分析所述第一图片和第二图片所对应的位置信息，确定所述第一图片和第二图片处于相同的地理位置；
匹配所述第二图片和标定所述第一图片中一个或者多个目标对象的标定图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值。
优选的，目标对象之间的相似度，是由构成所述目标对象的一个或者多个区域之间的相似度组成，所述分析所述第一图片和第二图片，具体包括：
筛选所述第一图片和第二图片中的噪音信息，得到第一图片和第二图片中对应所述目标对象的区域；
利用特征识别比对所述第一图片和第二图片中对应所述目标对象的区域，得到所述第一图片和第二图片中，对应所述目标对象的区域相似度。
优选的，所述方法还包括：
根据所述第二图片中所包含的所述目标对象的特征信息，搜索与之特征信息相比，相似度大于第二预设阈值的一张或者多张图片；
根据所述一张或者多张图片生成推送信息。
优选的，所述根据所述一张或者多张图片生成推送信息，具体包括：
筛选出图片生成时间相隔指定时长的一张或者多张推送图片；和/或，筛选出图片所包含的对象的视觉角度与所述第一图片中同一对象的视觉角度相同和/或互补的一张或者多张推送图；
组合所述一张或者多张推送图和第一图片得到推送信息。
优选的，所述组合所述一张或者多张推送图和第一图片得到推送信息，具体包括：
将所述一张或者多张推送图与所述第一图片添加到所述推送信息中；
在所述推送信息中增加所述一张或者多张推送图与所述第一图片的预设位置信息；
将相应各图片的时间信息附加到所述推送信息中。
第三方面，本发明实施例提供了一种信息推送方法，所述信息推送方法包括：
上传第一图片和第一图片相关信息；
接收推送消息，所述推送消息中携带所述第一图片，以及一张或者多张推送图片；其中，所述一张或者多张推送图片与所述第一图片所包含的一个或者多个目标对象处于相同的地理位置。
优选的，所述方法还包括：
上传终端的历史行径数据，其中所述历史行径数据包括历史行径的位置信息和时间信息。
优选的，所述方法还包括：
设置不接收与所述历史行径的位置信息相同，且与历史行径的时间差在预设阈值内的推送消息。
优选的，所述方法还包括：
设置接收推送消息中所携带的推送图片数量、推送图片分辨率、推送图片来源和推送图片中目标对象角度中的一项或者多项。
第四方面，本发明实施例提供了一种信息推送装置，包括：
第一图片接收模块，用于接收第一终端上传的第一图片；
第二图片接收模块，用于接收第二终端上传的第二图片；
推送消息模块，用于分析所述第一图片和第二图片，确认所述第一图片和 第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，发送对应所述目标对象的推送信息给所述第一终端。
第五方面，本发明还提供了一种信息推送装置，所述信息推送装置包括：
上传模块，用于上传第一图片和第一图片相关信息；
接收推送消息模块，用于接收推送消息，所述推送消息中携带所述第一图片，以及一张或者多张推送图片；其中，所述一张或者多张推送图片与所述第一图片所包含的一个或者多个目标对象处于相同的地理位置。
第六方面，本发明实施例还提供了一种电子设备，用于实现第二方面或者第三方面任一所述的信息推送方法步骤，包括：
至少一个处理器；以及，与所述至少一个处理器通信连接的存储器；其中，所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述存储器存储有可被所述至少一个处理器执行的指令程序，所述指令被程序设置为执行第二方面或者第三方面所提出的任意一项所述的信息推送方法步骤。
第七方面，本发明实施例还提供了一种非易失性计算机存储介质，其特征在于，所述计算机存储介质存储有计算机可执行指令，该计算机可执行指令被一个或多个处理器执行，用于完成第二方面中任一所述的信息推送方法，或者用于完成第三方面中任一所述的信息推送方法。
本发明实施例通过分析不同终端上传的图片，根据图片中所包含的目标对象间的相似度，来判断第二终端上传第二图片包含的目标对象相比较第一图片中的是否发生变化。其中，所述目标对象指现实中处于同一位置(通常指地理位置)的对象。若发生变化则将生成的推送消息发送给第一终端，从而保证第一用户能够及时的获取到其历史上传的第一图片中目标对象发生变化后的最新状态。不仅能够通过第一终端上传的第一图片关联上第一用户感兴趣的目标对象，还能够根据***中其他用户上传的图片识别出目标对象的改变，从而能够精准的反馈用户感兴趣信息，有助于提高推送信息的准确度和有效性。
【附图说明】
图1是本发明实施例提供的一种信息推送***架构示意图；
图2是本发明实施例提供的一种目标对象分布图结构示意图；
图3是本发明实施例提供的一种拍摄的方位角度的展示示意图；
图4是本发明实施例提供的一种信息推送方法的流程图；
图5是本发明实施例提供的一种拍摄的方位角度得到的目标对象示意图；
图6是本发明实施例提供的一种拍摄的方位角度的展示示意图；
图7是本发明实施例提供的一种拍摄的方位角度得到的目标对象示意图；
图8是本发明实施例提供的一种信息推送方法的流程图；
图9是本发明实施例提供的一种信息推送装置结构示意图；
图10是本发明实施例提供的一种信息推送装置结构示意图；
图11是本发明实施例提供的一种信息推送装置结构示意图；
图12是本发明实施例提供的一种信息推送装置结构示意图；
图13是本发明实施例提供的一种信息推送装置结构示意图；
图14是本发明实施例提供的一种信息推送装置结构示意图；
图15是本发明实施例提供的一种信息推送装置结构示意图；
图16是本发明实施例提供的一种信息推送装置结构示意图；
图17是本发明实施例提供的一种信息推送装置结构示意图；
图18是本发明实施例提供的一种信息推送装置结构示意图；
图19是本发明实施例提供的一种信息推送装置结构示意图；
图20是本发明实施例提供的一种电子设备的结构示意图。
【具体实施方式】
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
在本发明的描述中，术语“内”、“外”、“纵向”、“横向”、“上”、“下”、“顶”、“底”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而不是要求本发明必须以特定的方位构造和操作，因此不应当理解为对本发明的限制。
此外，下面所描述的本发明各个实施方式中所涉及到的技术特征只要彼此之间未构成冲突就可以相互组合。
实施例1:
本发明实施例提供了一种信息推送***，如图1所示，所述信息推送***包括服务器1、至少两个终端(例如：图1中所示终端2、3和4)，其中，所述至少两个终端包括第一终端2和第二终端3。其中，不同用户可以使用不同的智能终端连接服务器1，而对于服务器1其认证用户的方式为用户所创建的个人账号，因此，即便是同一个用户也可以使用不同的智能终端完成个人账号登录。 因此，本实施例中所述第一终端2泛指归属于同一个用户的一个或者多个终端，所述服务器1与用户之间的消息传递，则是通过智能终端来完成，该智能终端上登录有当前用户个人账号。本实施例所述信息推送***具体实现为：
所述服务器1存储第一终端2上传的第一图片；
所述第一图片可以是由作为第一终端2的智能手机或者单反拍照得到的，也可以是第一终端2从网络中下载得到的，还可以是通过第一终端2中的软件合成得到的。而上传服务器1的方式包括将所述第一图片上传朋友圈、第一图片备份到网络服务器1、第一图片加载到自身空间里的相册等等。
其中，所述服务器1接收各终端上传的图片的方式，包括所述服务器1直接接收各终端上传的图片，例如：所述服务器1作为用户的数据备份服务器、所述服务器作为用户的朋友圈服务器等等。所述服务器1还可以间接的从第三方服务器中获取终端上传的图片，所述第三方服务器可以是用于维护微博的服务器、用于管理朋友圈的服务器、用于管理校内网的服务器等等。在本发明各实施例中所描述的终端向服务器1上传图片，均包括所述服务器1直接接收终端上传的图片和服务器1间接的从第三方服务器中获取终端上传的图片。
所述第二终端3，用于上传第二图片；
如上所述，所述第二终端3可以是直接向所述服务器1上传的第二图片完成；也可以是第二终端3首先向第三方服务器上传第二图片，然后由第三方服务器将所述第二图片转发到所述服务器1。
所述服务器1，还用于获取所述第二图片，分析所述第一图片和所述第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，发送对应所述目标对象的推送信息给所述第一终端2。
其中，第一预设阈值可以由后台操作人员设置，例如：第一预设阈值设置为30％-50％。所述相似度小于第一预设阈值的一张或者多张图片，代表了与第一图片中处于同一位置的目标对象处于不同状态。优选的，所述第一预设阈值还可以根据第一终端2接收到推荐消息后给予的评价反馈(通常是由第一用户操作第一终端2完成)，做出相应的调整。例如：若第一终端2反馈推送消息是有效的，则可以保留所述第一预设阈值或者进一步提高该第一预设阈值的参数值；若第一终端2反馈推送消息是不准确的，则可以降低该第一预设阈值的参数值。所述第一预设阈值通常决定了目标对象的改变大小，若第一预设阈值设置的越大，则表明目标对象的改变相差越小，因此，也就容易造成误判的存在。这里 的误判指的是可能因为图片像素或者部分遮掩造成判断失误，而将未发生实质改变的目标对象所在的图片推送给了用户。
在本发明实施例中，所述同一位置的目标对象是指第一图片和第二图片所记录的现实环境中处于相同位置的对象。所述目标对象通常可以指定为图片中的楼房、图片中的标识牌、图片中的街道、图片中的店面等等具有一定时间内不变特性的对象；而对于人物、车辆、动物、天气等等动态对象来说，除非是在***中有特殊设定(例如：由用户登录主页或者登录APP方式设定，或者由操作人员在服务器1的***参数中设定等)，否则服务器1默认不会将此不具有一定时间内不变特性的动态对象作为所述目标对象。其中，在本发明各实施例中所述目标对象和动态对象，以及其他可辨识的内容均可统称为对象。在分析所述第一图片和第二图片时，优选的要做一轮目标对象的筛选，排除上述容易发生动态变化的动态对象，因此，除了所述目标对象以外的对象在筛选过程中也被称为噪音信息。
其中，对于同一位置的目标对象的确认包括多种方式，例如：服务器1在获取第一图片和第二图片时，还得到相应第一图片和第二图片的位置信息，则可以根据所述位置信息判断第一图片和第二图片中同一位置的目标对象。在优选的方案中，所述位置信息不仅包括经纬度定位信息，还包括拍摄的方位角度信息。所述拍摄的方位角度信息能够进一步提高目标对象的识别准确度。
结合本发明实施例，除了上述通过位置信息判断第一图片和第二图片中同一位置的目标对象外，还可以通过图像处理方式，首先进行图片中轮廓提取，将图片划分为一个或者多个对象(包括上述目标对象和动态对象等)，并筛选出一个或者多个目标对象；然后，根据所述一个或者多个目标对象的关键点(例如顶点、中心点、拐点等等)构成的目标对象分布图(其效果图如图2所示，其中的不规则图形便为所述目标对象分布图)，识别出第一图片和第二图片中同一位置的目标对象。除此以外，还可以通过目标识别算法，首先进行图片中轮廓提取，将图片划分为一个或者多个对象(包括上述目标对象和动态对象等)，并筛选出一个或者多个目标对象，根据图片之间目标轮廓快速匹配出第一图片和第二图片中相似度大于第二预设阈值的同一位置的一个或者多个目标对象(例如：图2所示的1和2两个目标对象相似度分别大于另一幅图中的相应1和2两个目标对象。此处的一个或者多个目标对象是上述目标对象分布图诸多目标对象中的成员)，然后基于该目标对象结合上述目标对象分布图，则能够高 效的找到第一图片和第二图片中的同一位置的目标对象的、且相似度小于第一预设阈值的结果(即通过目标对象分布图中个别点位，快速完成两张图片之间各自目标对象分布图相应点位的配对)。所述图像处理方式和所述目标识别算法可以单独对没有携带位置信息的图片进行处理，也可以作携带位置信息的图片分析时的辅助验证方法。其中，所述目标对象分布图在结合相应位置信息中的拍摄的方位角度信息，能够完成目标对象分布图的角度旋转，从而能够更精确的识别两张拍摄角度有差异的图片中的目标对象。以图3所示，所述拍摄的方位角度为东偏南30度，相应的旋转方法可以借鉴已有的算法公式，在此不一一赘述。
本发明实施例通过分析不同终端(包括实施例中涉及的第一终端和第二终端)上传的图片，根据图片中所包含的现实中同一位置的目标对象间的相似度，来判断第二终端3上传第二图片包含的同一位置的目标对象相比较第一图片中的是否发生变化。若发生变化则将生成的推送消息发送给第一终端2，从而保证第一用户能够及时的获取到其历史上传的第一图片中目标对象发生变化后的最新状态。不仅能够通过第一终端2上传的第一图片关联上第一用户感兴趣的目标对象，还能够根据***中其他用户上传的图片识别出目标对象的改变，从而能够精准的反馈用户感兴趣信息，有助于提高推送信息的准确度和有效性。
在本发明实施例中，在确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，发送对应所述目标对象的推送信息给所述第一终端2。所述推送信息可以是一条消息，所述消息包括该目标对象名称、目标对象发生改变时间和浏览第二图片的链接地址中的一项或者多项。除此以外，为了能够让第一用户查阅到所述同一位置的目标对象相关的其它图片，在本发明实施例中还提供了一种扩展方案，具体的：
所述服务器1根据所述第二图片中所包含的所述目标对象的特征信息，搜索与之特征信息相比，相似度大于第二预设阈值的一张或者多张图片。
其中，第二预设阈值可以由后台操作人员设置，例如第二预设阈值设置为60％-80％。所述相似度大于第二预设阈值的一张或者多张图片，通常是与第二图片中该目标对象处于相同状态的，由其它终端上传到服务器1中的。
根据所述一张或者多张图片生成所述推送信息。其中，最简单的生成方式就是将所述一张或者多张图片设置相应排列顺序，并携带在推送消息中发送给第一终端2，以便所述第一终端2在接收到所述推送消息后能后在显示屏幕中展 示所述一张或者多张图片给第一用户。
在本发明实施例中涉及服务器1分析所述第一图片和第二图片，那么如何才能够在庞大的图片库中，更为高效的为所述第二图片定位到相应的第一图片就是优选去克服的问题。因此，结合本发明实施例还提供了就如何管理图片库的方法。具体的：
所述第一终端2上传第一图片同时，上传的内容中还携带第一图片相关信息，其中所述第一图片相关信息包括所述第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，具体的：
所述服务器1在接收到所述第一图片时，还用于校对所述第一图片与所述第一图片相关信息，并根据校对结果归类管理所述第一图片；其中归类管理包括：根据位置信息相同类型、图片分辨率同区间类型、生成时间为同一时段类型中的一项或者多项分类存储图片；
所述服务器1，还用于根据第二图片的相关信息中的第二图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，获得类型相匹配的第一图片。
其中，校对除了针对上述第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项进行以外，还可以针对于由第一终端2对于其发送的第一图片编辑的第一图片相关信息，例如：第一图片的标题、由第一用户输入的第一图片的位置信息(尤其适用用户截图的方式)等等。除此以外，校对过程还可以包含对图片中位置信息的确认，即通过类似上述目标对象匹配的操作，完成图片中位置信息的确认。
其中，所述归类管理的方式包括将所述第一图片归类到与其拥有相同位置信息的一组图片完成存储。
由此，每一次接收到终端上传的照片时，便可以通过校对过程完成图片在拥有共性位置信息的一组图片中共同存储，以便后续分析图片时能够根据位置信息有效的限缩分析图片的范围(数量)。
实施例2:
本发明实施例还提供了一种信息推送方法，所述信息推送方法可以由实施例1中的信息推送***中服务器1来执行。第一终端和第二终端仅仅是为了方便区分描述而作的命名，所述第一终端和第二终端对应的归属用户被称为第一用户和第二用户。所述第二终端所代表的可以是现实中具有所述上传图片功能 的第二用户所拥有的任意终端，而第一终端除了具备第二终端所述的上传图片功能外，通常第一用户还是本实施例中服务器推送消息的订阅方(该订阅可以是第一用户主动发起的，还可以是服务器默认设定第一用户拥有的)。在本发明实施例中由于第一用户或者第二用户所使用的终端可以是不同类型的，因此，为了描述方便统一以第一终端和第二终端作为相应第一用户的智能终端和第二用户的智能终端的简称，如图4所示，所述信息推送方法包括：
在步骤201中，接收第一终端上传的第一图片。
其中，服务器所接收的第一终端上传的第一图片可以是通过实施例1中描述的服务器直接接收终端上传的图片的方式，还可以通过实施例1中描述的服务器间接的从第三方服务器中获取终端上传的图片。
所述第一图片可以是由作为第一终端的智能手机或者单反拍照得到的；所述第一图片也可以是由作为第一终端的个人电脑、平板电脑或者智能手机网络下载得到；除此以外，所述第一图片还可以是第一终端通过截图方式生成的。
在步骤202中，接收第二终端上传的第二图片。
所述第二终端的第二图片生成方式和服务器接收方式可以参考步骤201中第一终端的第一图片的生成方式，在此不再赘述。
在步骤203中，分析所述第一图片和第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，则发送对应所述目标对象的推送信息给所述第一终端。
在本发明实施例中，所述同一位置的目标对象是指第一图片和第二图片所记录的现实环境中处于相同位置的目标对象。其中，所述目标对象通常指图片中的楼房、图片中的标识牌、图片中的街道、图片中的店面等等具有一定时间内不变特性的对象。而对于人物、车辆、动物、天气等等动态对象来说，除非是在有设定情况下，通常服务器默认不将此不具有一定时间内不变特性的动态对象作为目标对象，其中，目标对象和动态对象统称为对象。因此，在分析所述第一图片和第二图片时，优先的要做一轮目标对象的筛选，排除上述容易发生动态变化的对象。在本发明实施例中，所述动态对象相对于目标对象来说也被称为噪音信息。
其中，对于同一位置的目标对象的确认包括多种方式，例如：服务器在获取第一图片和第二图片时，还得到相应第一图片和第二图片的位置信息，则可以根据所述位置信息判断第一图片和第二图片中同一位置的目标对象。在优选 的方案中，所述位置信息不仅包括经纬度定位信息，还包括拍摄的方位角度信息，所述拍摄的方位角度信息能够进一步提高目标对象的识别准确度，具体的：根据所述拍摄的方位角度信息，调整第一图片和/或第二图片中图像处理方式中生成目标对象分布图的旋转角度。其中，所述拍摄的方位角度信息可以根据智能手机的GPS***和/或陀螺仪计算得到。
在本发明实施例中还可以根据所述拍摄的方位角度信息，调整第一图片和/或第二图片中使用目标识别算法所计算的对象区域，例如：若第二图片相对于第一图片的位置信息中的经纬度相同，而其拍摄的方位角度信息相差30度，则在匹配第一图片和第二图片中相似度最高的、且同一位置的目标对象时，取自第一图片中的或者取自第二图片中的目标对象的特征区域，需要做相应部分特征区域的忽略处理，该部分特征区域是由于所述拍摄的方位角度信息相差30度产生的第一图片中目标对象和第二图片中目标对象所包含的信息不对等造成的。以图3所示的拍摄角度，其拍摄出来的目标对象效果如图5所示，而对于以图6所示的拍摄角度，其拍摄出来的目标对象效果如图7所示，因此，对应相应的拍摄角度，在进行图5和图7目标对象匹配时，将其中阴影特征区域忽略掉，不参与到目标对象相似度的匹配过程中。从而避免了由于拍摄角度造成的第一图片和第二图片之间目标对象的信息不对等。
结合本发明实施例，除了上述通过位置信息判断第一图片和第二图片中同一位置的目标对象外，还可以通过所述图像处理方式，具体的：首先进行图片中轮廓提取，将图片划分为一个或者多个对象(包括上述目标对象和动态对象等)，并筛选出一个或者多个目标对象；然后，根据所述一个或者多个目标对象的中心点构成的目标对象分布图，识别出第一图片和第二图片中同一位置的目标对象。除此以外，还可以通过目标识别算法，首先进行图片中轮廓提取，将图片划分为一个或者多个对象(包括上述目标对象和动态对象等)，并筛选出一个或者多个目标对象，根据图片之间目标轮廓快速匹配出第一图片和第二图片中相似度大于预设阈值的同一位置的目标对象，然后基于该目标对象结合上述目标对象分布图，则能够更快的找到第一图片和第二图片中的同一位置的目标对象的、且相似度小于第一预设阈值的结果。
本发明实施例通过分析不同终端上传的图片，根据图片中所包含的现实中同一位置的目标对象间的相似度，来判断第二终端上传第二图片包含的同一位置的目标对象相比较第一图片中的是否发生变化。若发生变化则将生成的推送 消息发送给第一终端，从而保证第一终端能够及时的获取到其历史上传的第一图片中目标对象发生变化后的最新状态。不仅能够通过用户上传的第一图片关联上用户感兴趣的目标对象，还能够根据其它终端上传的图片识别出目标对象的改变，从而能够精准的反馈用户感兴趣信息，有助于提高推送信息的准确度和有效性。
在服务器运行本实施例所述方法时，在一段时间内服务器可以接收到的来自各终端的图片可能非常多，那么提供一种逐级筛选机制和归类管理的方式则显得极为有必要。
对于所述逐级筛选机制和归类管理的方式必不可少的就是，服务器在接收第一终端上传的第一图片同时，还获取到第一图片相关信息，其中所述第一图片相关信息包括所述第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，所述方法还包括：
接收到所述第一图片时，校对所述第一图片与所述第一图片相关信息，并根据校对结果归类管理所述第一图片；其中归类管理包括：根据位置信息相同类型、图片分辨率同区间类型、生成时间为同一时段类型中的一项或者多项分类存储图片；
所述分析所述第一图片和所述第二图片之前，还包括：
根据第二图片的相关信息中的第二图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，获得类型相匹配的第一图片。
其中，校对除了针对上述第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项进行以外，还可以针对于由第一终端对于其发送的第一图片编辑的第一图片相关信息，例如：第一图片的标题、由第一用户输入的第一图片的位置信息(尤其适用用户截图的方式)等等。除此以外，校对过程还可以包含对图片中位置信息的确认，即通过类似上述目标对象匹配的操作，完成图片中位置信息的确认。
对于图片信息的校对，通常还是结合第一图片的相关信息确认相应标定图片，并通过匹配标定图片中目标对象和第一图片的目标对象来完成。相当于将校对过程中的匹配过程和方法步骤中的匹配过程同步进行。可选的，还可以增加服务器侧的工作人员来完成校对过程。
所述归类管理的方式包括：若第一图片相关信息包括所述第一图片的内容位置信息，则将所述第一图片归类到与其拥有相同位置信息的一组图片完成存 储；和/或根据第一图片中所包含的一个或者多个目标对象的特征信息和第一图片相关信息检索图片库，从所述图片库中确认用于标定所述一个或者多个目标对象的标定图片。
其中，所述标定图片包含了相应目标对象较为全面的信息，简单理解就是像素达到指定值，并且相对于目标对象来说噪音足够小的图片。对于服务器来说，所述标定图片可以是由专人拍摄生成，还可以是从诸多用户对于同一位置目标对象拍摄图片中挑选得到。
其中，所述标定图片可以是多张，并且各标定图片包含第一图片中的一个或者多个目标对象。相比较第一图片中的目标对象，其对应的标定图片所包含的目标对象信息更全面，使用所述标定图片的目标对象的特征信息作为第一图片中目标对象的特征信息参与服务器后续接收到的各图片(其中，后续接收到的各图片可以包括所述第二图片)的分析，将会更加高效和准确。
所述逐级筛选机制包括：服务器对于接收到的图片，首先确定图片相关信息中的位置信息与所述第一图片的位置信息是否相符，若不相符，则不做匹配；若相符，则调用所述第一图片或者所述第一图片的标定图片完成目标对象的确认。其次，若确定图片相关信息不包含位置信息，则利用上述图像处理方式和/或目标识别算法进行目标对象的确认，在此，不一一赘述。
基于上述归类管理的方式，且第一图片和第二图片的相关信息中包含位置信息，则在步骤203中执行的分析所述第一图片和第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，具体包括：
在步骤2031中，分析所述第一图片和第二图片所对应的位置信息，确定所述第一图片和第二图片处于相同的地理位置。若确定所述第一图片和第二图片处于不同的地理位置，跳过后续的执行步骤，直接结束流程。
在步骤2032中，匹配所述第二图片和标定所述第一图片中一个或者多个目标对象的标定图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值。
在上述归类管理的方式中引入了标定图像的使用和根据位置信息完成存储，然而，对于第一终端所发送的第一图片的标定图片来说，当确认步骤203中第一终端的第一图片中的目标对象发生了改变，则除了发送对应所述目标对象的推送信息给所述第一终端外，所述方法还包括：
在步骤204中，向与所述第一图片的标定图片建立有标定关系一张或者多张图片所对应的用户发送所述推荐消息。
服务器对于每一个终端所上传的图片，通常在校对其包含涉及现实中的位置对象后，服务器可以通过所述归类管理为其关联一个或者多个标定图片。其目的本来是为了能够在后续分析类似第二图片这种由其它终端上传的图片时，能够更加精准和高效。然而，在步骤204中则是进一步将终端上传图片和标定图片之间的关联关系给利用起来，相当于通过标定图片标定了一组和第一图片所包含的目标对象状态处于同一时段的图片，从而能够完成该组图片中的第一图片与第二图片(以第二图片为例描述)的分析，便能完成整个组内图片所对应用户的推送消息的发送，不仅减少了分析次数(即不用将该组内的每一张图片拿来和第二图片做分析)，而且有助于提高推送效率。在本发明各实施例中虽然是根据终端上传图片作为判断依据，但是，由于一个用户可能通过多个终端分不同时间、在不同地点完成不同图片的上传，因此具体实现过程中还是以用户账号作为该终端上传到服务器中图片的辨识依据，即服务器中图片不会使用不同的终端ID作为区分图片归属的依据，而是以该终端所登录的用户账号作为区分图片是否来源于同一用户的依据。
结合本发明实施例，在执行步骤203中的“发送对应所述目标对象的推送信息给所述第一用户”时，存在一种优选的实现方案，具体的：
在步骤301中，根据所述第二图片中所包含的所述目标对象的特征信息，搜索与之特征信息相比，相似度大于第二预设阈值的一张或者多张图片。
其中，第二预设阈值可以由后台操作人员设置，例如第二预设阈值设置为60％-80％。所述相似度大于第二预设阈值的一张或者多张图片，通常是与第二图片中该目标对象处于相同状态的，由其它终端上传到服务器中的。
在步骤302中，根据所述一张或者多张图片生成推送信息。
该优选的实现方式能够配合上述归类管理的方式中的标定图片一起来生成推送信息，即在所述推送信息中同时包含所述一张或者多张图片，以及一张或者多张标定图片。
结合该优选的实现方案中的步骤302，存在一种可实现的推荐消息的生成方法，具体包括：
在步骤3021中，筛选出图片生成时间相隔指定时长的一张或者多张推送图片；和/或，筛选出图片所包含的对象的视觉角度与所述第一图片中同一对象的 视觉角度相同和/或互补的一张或者多张推送图。
其中，相隔指定时长是根据用户或者***所设定的推送消息中所包含的图片数量，以及第一图片生成时间和第二图片生成时间差来决定。例如：第一图片生成时间为A，第二图片生成时间为B，用户设定的推送消息中携带的推送图片数量为N，则所述指定时长＝(B-A)/N。其中，若无法获取相隔指定时长位置的图片时，则取与所述相隔指定时长位置时间最邻近的图片。
其中，各自所包含的对象的视觉角度可以通过上述拍摄的方位角度信息确认，在此不再赘述。
在步骤3022中，组合所述一张或者多张推送图和第一图片得到推送信息。
本发明实施例中，还对于步骤3022的组合方式给予了一种可选的组合实现方案，具体包括：
将所述一张或者多张推送图片与所述第一图片添加到所述推送信息中；在所述推送信息中增加所述一张或者多张推送图与所述第一图片的预设位置信息；将相应各图片的时间信息附加到所述推送信息中。所述时间信息用于与所述一张或者多张推送图与所述第一图片共同显示。
实施例3:
本发明实施例2所述的信息推送方法是从服务器侧角度展开来描述的，而本发明实施例3则是从用户侧展开来阐述如何实现本发明所提出的信息推送方法，所述信息推送方法包括：
在步骤401中，上传第一图片和第一图片相关信息。
其中，所述第一图片中包括一个或者多个目标对象；其中，所述图片相关信息包括第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项。
其中，上传的方式包括将所述第一图片上传朋友圈、第一图片备份到网络服务器、第一图片加载到自身空间里的相册等等
在步骤402中，接收推送消息，所述推送消息中携带所述第一图片，以及一张或者多张推送图片；其中，所述一张或者多张推送图片与所述第一图片所包含的一个或者多个目标对象处于相同的地理位置。
其中，所述推送信息中可选的还包括各推送图片的时间信息，则在执行步骤402之后，优选的还包括步骤403，具体的：
在步骤403中，根据各推送图片的时间信息，在推送界面上排列所述一张 或者多张推送图片。可选的，所述第一图片排列在所述一张或者多张推送图片的起始位置，或者，所述第一图片以置顶的方式排列在所述一张或者多张推送图片之上。
结合本发明实施例，在所述步骤401中上传的内容除了第一图片和第一图片相关信息，存在一种可选的方案，具体的：
上传终端的历史行径数据，其中所述历史行径数据包括历史行径的位置信息和时间信息。
所述终端为归属于用户的一个或者多个终端，其中，所述终端包括实施例1-2中所述的第一终端，所述用户在实施例1-2中具体为第一用户。根据所述历史行径的位置信息和时间信息，用户可以设置不接收与所述历史行径的位置信息相同，且与历史行径的时间差在预设阈值内的推送消息。所述预设阈值可以由用户来设置，通常是设置为1天、1周或者1个月。其中，用户设置的预设阈值越长，则一定层度上可以减少用户收到推送消息的频率，以便满足不同用户的推送消息需求。因此，在可选的方案中，除了可以“设置不接收与所述历史行径的位置信息相同，且与历史行径的时间差在预设阈值内的推送消息”外，还可以设置一时间参数，该时间参数用于服务器判断当前时间满足该时间参数时，将对应该用户的一条或者多条推送消息打包发送，同样可以起到个性化完成推送消息的目的。上述两种方式都涉及推送消息的延迟发送，若该延迟发送需求结合实施例2中所述的通过标定图片标定了一组和第一图片所包含的目标对象状态处于同一时段的图片，从而能够完成该组图片中的第一图片与第二图片(其中，以第二图片为例描述)的分析，便能完成整个组内图片所对应用户的推送消息的生成。在本发明实施例中，对于设置了预设阈值或者时间参数的用户，对其生成的推送消息可以暂时存储起来，并在满足所述预设阈值或者时间参数后发送相应的推送消息。
在本发明实施例所述方法中，用户除了可以设置接收推送消息中所携带的推送图片数量外，还可以设置推送图片分辨率、推送图片来源和推送图片中目标对象角度中的一项或者多项。
对于设置推送图片分辨率，例如：用户可以通过登录服务器所提供的平台，设置服务器所推送的图片的分辨率要达到1280*1960或者分辨率要小于760*1280等等。所述图片的分辨率设置主要是根据用户浏览图片的习惯和要求，以及用户自身拥有的流量状态为依据完成的。在具体实现过程中，还可以加入 用户智能终端网络状态监测功能，当确认用户的智能终端当前网络状态为连接wifi或者有线网络时，则使用高分辨率的推荐图；当确认用户的智能终端当前网络状态为移动网络(例如：3G、4G网络)时，则使用低分辨率的推荐图。
除了上述通过判断网络状态来避免推送消息占用用户过多流量外，还存在另一种可选的实现方式。具体的：第一终端所接收到的推荐消息中所携带的一张或者多张推荐图片的分辨率并非服务器中存储的原始图片的分辨率，而是经过压缩处理的；所述经过压缩的图片由原始图片中的部分像素信息组成，并且，能够在第一用户点击推荐消息显示界面中指定推荐图片时，由所述服务器将用户选中的推荐图片被压缩部分发送给所述用户，从而和用户点击的推荐图片生成像素信息完整的图片。
对于推送图片来源，例如：用户可以设置只接收由指定类型对象上传的图片作为推送图片来源，所述指定类型对象包括第三方公司、所述用户的好友、指定群组内的人员、指定校区内的人员、指定企业内的人员等等。
对于推送图片中目标对象角度，例如：用户可以设置服务器优先挑选拍摄包含所述目标对象的指定角度范围的一张或者多张图片作为生成推送消息的推送图片。
实施例4:
本发明实施例还提供了一种信息推送装置，所述信息推送装置可用于运行实施例2所述的信息推送方法。在具体实现过程中，所述信息推送装置可以是一服务器，还可以是一具有足够计算能力的家用电脑，还可以大数据云中的某一虚拟执行借口，如图9所示，所述信息推送装置包括：
第一图片接收模块，用于接收第一终端上传的第一图片；
第二图片接收模块，用于接收第二终端上传的第二图片；
推送消息模块，用于分析所述第一图片和第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，发送对应所述目标对象的推送信息给所述第一终端。
其中，第一图片接收模块和第二图片接收模块分别连接所述推送消息模块。
本发明实施例通过推送消息模块分析不同终端上传的图片中所包含的现实中同一位置的目标对象的相似度，来判断第二终端上传第二图片包含的同一位置的目标对象相比较第一图片中的是否发生变化。若发生变化则将生成的推送消息发送给第一终端，从而保证第一用户能够及时的获取到其历史上传的第一 图片中目标对象发生变化后的最新状态。不仅能够通过第一终端上传的第一图片关联上第一用户感兴趣的目标对象，还能够根据***中其它终端上传的图片识别出目标对象的改变，从而能够精准的反馈第一用户感兴趣信息，有助于提高推送信息的准确度和有效性。
在本发明实施例实现过程中，还存在一种可选的实现方式，其中，第一图片接收模块在接收第一终端上传的第一图片同时，还获取到第一图片相关信息，其中所述第一图片相关信息包括所述第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，如图10所示，所述装置还包括归类管理模块，所述归类管理模块连接所述第一图片接收模块和第二图片接收模块，并且，所述第一图片接收模块和第二图片接收模块经由所述归类管理模块连接到推送消息模块，具体的：
所述归类管理模块，用于校对所述第一图片与所述第一图片相关信息，并根据校对结果归类管理所述第一图片。所述归类管理具体可以参考实施例2中所述内容，在此不一一赘述。
结合本发明实施例，还存在一种可选的实现方案，所述归类管理模块还包括特征检索模块和目标对象标定模块，如图11所示，具体包括：
所述特征检索模块，用于根据第一图片中所包含的一个或者多个目标对象的特征信息和第一图片相关信息检索图片库；
所述目标对象标定模块，用于从所述图片库中确认用于标定所述一个或者多个目标对象的标定图片。
结合本发明实施例，还存在一种可选的实现方案，如图12所示，所述推送消息模块还包括位置信息确认模块和标定图片匹配模块，具体包括：
所述位置信息确认模块，用于分析所述第一图片和第二图片所对应的位置信息，确定所述第一图片和第二图片处于相同的地理位置；
所述标定图片匹配模块，用于匹配所述第二图片和标定所述第一图片中一个或者多个目标对象的标定图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值。
结合本发明实施例，还存在一种可选的实现方案，目标对象之间的相似度，是由构成所述目标对象的一个或者多个区域之间的相似度组成，所述推送消息模块还包括筛选模块和区域相似度计算模块，如图13所示，具体包括：
所述筛选模块，用于筛选所述第一图片和第二图片中的噪音信息，得到第 一图片和第二图片中对应所述目标对象的区域；
所述区域相似度计算模块，利用特征识别比对所述第一图片和第二图片中对应所述目标对象的区域，得到所述第一图片和第二图片中，对应所述目标对象的区域相似度。
结合本发明实施例，还存在一种可选的实现方案，所述推送消息模块还包括推送图片匹配模块和推送消息生成模块，如图14所示，具体的：
所述推送图片匹配模块，用于根据所述第二图片中所包含的所述目标对象的特征信息，搜索与之特征信息相比，相似度大于第二预设阈值的一张或者多张图片；
所述推送消息生成模块，用于根据所述一张或者多张图片生成推送信息。
结合本发明实施例，还存在一种可选的实现方案，所述推送消息生成模块还包括第一筛选模块和/或第二筛选模块，如图15所示，具体的：
所述第一筛选模块，用于筛选出图片生成时间相隔指定时长的一张或者多张推送图片；
所述第二筛选模块，用于筛选出图片所包含的对象的视觉角度与所述第一图片中同一对象的视觉角度相同和/或互补的一张或者多张推送图；
其中，所述一张或者多张推送图用于生成所述推送信息。
值得说明的是，上述装置内的模块、单元之间的信息交互、执行过程等内容，由于与本发明的信息推送方法实施例基于同一构思，具体内容可参见本发明实施例2中的叙述，相应模块之间也可以依照实施例2所述方法进行组合，此处不再赘述。
实施例5:
本发明实施例还提供了一种信息推送装置，所述信息推送装置可用于运行实施例3所述的信息推送方法。在具体实现过程中，所述信息推送装置可以是智能手机、平板电脑、个人PC等等，如图16所述，所述信息推送装置包括：
上传模块，用于上传第一图片和第一图片相关信息；
接收推送消息模块，用于接收推送消息，所述推送消息中携带所述第一图片，以及一张或者多张推送图片；其中，所述一张或者多张推送图片与所述第一图片所包含的一个或者多个目标对象处于相同的地理位置。
本发明实施例通过接收推送消息模块够保证第一用户及时的获取到其历史上传的第一图片中目标对象发生变化后的最新状态。不仅能够通过第一终端上 传的第一图片关联上第一用户感兴趣的目标对象，还能够根据***中其它终端上传的图片识别出目标对象的改变，从而能够精准的反馈第一用户感兴趣信息，有助于提高推送信息的准确度和有效性。
结合本发明实施例，还存在一种可选的实现方案，如图17所示，所述上传模块还包括：
历史行径数据上传模块，用于上传终端的历史行径数据，其中所述历史行径数据包括历史行径的位置信息和时间信息。
结合本发明实施例，还存在一种可选的实现方案，如图18所示，所述装置还包括第一设置模块，具体的：
所述第一设置模块，用于设置不接收与所述历史行径的位置信息相同，且与历史行径的时间差在预设阈值内的推送消息。
结合本发明实施例，还存在一种可选的实现方案，如图19所示，所述装置还包括第二设置模块，具体的：
所述第二设置模块，用于设置接收推送消息中所携带的推送图片数量、推送图片分辨率、推送图片来源和推送图片中目标对象角度中的一项或者多项。
值得说明的是，上述装置内的模块、单元之间的信息交互、执行过程等内容，由于与本发明的信息推送方法实施例基于同一构思，具体内容可参见本发明实施例3中的叙述，相应模块之间也可以依照实施例3所述方法进行组合，此处不再赘述。
实施例6:
本发明实施例还提供了一种电子设备，用于实现实施例2-3任一所述的信息推送方法步骤，如图20所示，包括：
至少一个处理器；以及，与所述至少一个处理器通信连接的存储器；其中，所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述存储器存储有可被所述至少一个处理器执行的指令程序，所述指令被程序设置为执行实施例2-3所提出的任意一项所述的信息推送方法步骤。
其中，图20中以一个处理器为例，所述处理器和存储器可以通过总线或者其他方式连接，图20中以通过总线连接为例。
存储器作为一种非易失性计算机可读存储介质，可用于存储非易失性软件程序、非易失性计算机可执行程序以及模块，如第2、第3实施方式中的信息推送方法以及对应的程序指令/模块(例如，图9-15所示的各个模块)。处理器通 过运行存储在存储器中的非易失性软件程序、指令以及模块，从而执行信息推送装置的各种功能应用以及数据处理，即实现第2实施方式的信息推送方法以及第4实施方式的各个模块、单元的功能，亦或实现第3实施方式的信息推送方法以及第5实施方式的各个模块、单元的功能。
存储器可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在一些实施方式中，存储器可选包括相对于处理器远程设置的存储器，这些远程存储器可以通过网络连接至处理器。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
所述程序指令/模块存储在所述存储器中，当被所述一个或者多个处理器执行时，执行上述第2实施方式中的信息推送方法，例如，执行以上描述的图4所示的各个步骤；也可实现图9-15所述的各个模块、单元；亦或执行上述第3实施方式中的信息推送方法，例如，执行以上描述的图8所示的各个步骤；也可实现图16-19所述的各个模块、单元。
本发明实施方式的社交装置可以多种形式存在，在执行以上描述的图4所示的各个步骤；以及图9-15所述的各个模块、单元，或者图8所示的各个步骤；以及图16-19所述的各个模块、单元时，该社交装置包括但不限于：
(1)移动通信设备：这类设备的特点是具备移动通信功能，并且以提供话音、数据通信为主要目标。这类终端包括:智能手机(例如iPhone)、多媒体手机、功能性手机，以及低端手机等。
(2)超移动个人计算机设备:这类设备属于个人计算机的范畴，有计算和处理功能，一般也具备移动上网特性。这类终端包括：PDA、MID和UMPC设备等，例如iPad。
(3)便携式娱乐设备：这类设备可以显示和播放视频内容，一般也具备移动上网特性。该类设备包括：视频播放器，掌上游戏机，以及智能玩具和便携式车载导航设备。
(4)服务器：提供计算服务的设备，服务器的构成包括处理器、硬盘、内存、***总线等，服务器和通用的计算机架构类似，但是由于需要提供高可靠的服务，因此在处理能力、稳定性、可靠性、安全性、可扩展性、可管理性等方面要求较高。
(5)其他具有视频播放功能或连接互联网功能的电子设备。
本发明实施例还提供了一种非易失性计算机存储介质，其中，所述计算机存储介质包括只读存储器(ROM，Read Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁盘或光盘等。所述计算机存储介质存储有计算机可执行指令，该计算机可执行指令被一个或多个处理器执行，例如图20中的一个处理器，可使得上述一个或多个处理器可执行第2实施方式中的信息推送方法，例如，执行以上描述的图4所示的各个步骤；也可实现图9-15所述的各个模块、单元；或者执行第3实施方式中的信息推送方法，例如，执行以上描述的图8所示的各个步骤；也可实现图16-19所述的各个模块、单元。
值得说明的是，上述装置和***内的模块、单元之间的信息交互、执行过程等内容，由于与本发明的处理方法实施例基于同一构思，具体内容可参见本发明方法实施例中的叙述，此处不再赘述。
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (20)

  1. 一种信息推送***，其特征在于，所述信息推送***包括服务器、至少两个终端，其中，所述至少两个终端包括第一终端和第二终端；所述服务器存储第一终端上传的第一图片，具体的：
    所述第二终端，用于上传第二图片；
    所述服务器，还用于获取所述第二图片，分析所述第一图片和所述第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，发送对应所述目标对象的推送信息给所述第一终端。
  2. 根据权利要求1所述的信息推送***，其特征在于，所述服务器还用于：
    根据所述第二图片中所包含的所述目标对象的特征信息，搜索与之特征信息相比，相似度大于第二预设阈值的一张或者多张图片；
    根据所述一张或者多张图片生成所述推送信息。
  3. 根据权利要求1所述的信息推送***，其特征在于，所述第一终端上传第一图片同时，上传的内容中还携带第一图片相关信息，其中所述第一图片相关信息包括所述第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，具体的：
    所述服务器在接收到所述第一图片时，还用于校对所述第一图片与所述第一图片相关信息，并根据校对结果归类管理所述第一图片；其中归类管理包括：根据位置信息相同类型、图片分辨率同区间类型、生成时间为同一时段类型中的一项或者多项分类存储图片；
    所述服务器，还用于根据第二图片的相关信息中的第二图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，获得类型相匹配的第一图片。
  4. 根据权利要求1-3任一所述的信息推送***，其特征在于，目标对象之间的相似度，是由构成所述目标对象的一个或者多个区域之间的相似度组成，所述服务器获取所述第二图片，分析所述第一图片和所述第二图片，具体包括：
    服务器筛选所述第一图片和第二图片中的噪音信息，得到第一图片和第二图片中对应所述目标对象的区域；
    利用特征识别比对所述第一图片和第二图片中对应所述目标对象的区域，得到所述第一图片和第二图片中，对应所述目标对象的区域相似度。
  5. 一种信息推送方法，其特征在于，所述信息推送方法包括：
    接收第一终端上传的第一图片；
    接收第二终端上传的第二图片；
    分析所述第一图片和第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，发送对应所述目标对象的推送信息给所述第一终端。
  6. 根据权利要求5所述的信息推送方法，其特征在于，接收第一终端上传的第一图片同时，还获取到第一图片相关信息，其中所述第一图片相关信息包括所述第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，所述方法还包括：
    接收到所述第一图片时，校对所述第一图片与所述第一图片相关信息，并根据校对结果归类管理所述第一图片；其中归类管理包括：根据位置信息相同类型、图片分辨率同区间类型、生成时间为同一时段类型中的一项或者多项分类存储图片分类存储图片；
    所述分析所述第一图片和所述第二图片之前，还包括：
    根据第二图片的相关信息中的第二图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，获得类型相匹配的第一图片。
  7. 根据权利要求6所述的信息推送方法，其特征在于，所述归类管理还包括：
    根据第一图片中所包含的一个或者多个目标对象的特征信息和第一图片相关信息检索图片库；从所述图片库中确认用于标定所述一个或者多个目标对象的标定图片；
    则所述分析所述第一图片和第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，具体包括：
    分析所述第一图片和第二图片所对应的位置信息，确定所述第一图片和第二图片处于相同的地理位置；
    匹配所述第二图片和标定所述第一图片的标定图片，确认所述第一图片的标定图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值。
  8. 根据权利要求5-7任一所述的信息推送方法，其特征在于，目标对象之间的相似度，是由构成所述目标对象的一个或者多个区域之间的相似度组成，所述分析所述第一图片和第二图片，具体包括：
    筛选所述第一图片和第二图片中的噪音信息，得到第一图片和第二图片中对应所述目标对象的区域；
    利用特征识别比对所述第一图片和第二图片中对应所述目标对象的区域，得到所述第一图片和第二图片中，对应所述目标对象的区域相似度。
  9. 根据权利要求5-8任一所述的信息推送方法，其特征在于，所述方法还包括：
    根据所述第二图片中所包含的所述目标对象的特征信息，搜索与之特征信息相比，相似度大于第二预设阈值的一张或者多张图片；
    根据所述一张或者多张图片生成推送信息。
  10. 根据权利要求9所述的信息推送方法，其特征在于，所述根据所述一张或者多张图片生成推送信息，具体包括：
    筛选出图片生成时间相隔指定时长的一张或者多张推送图片；和/或，筛选出图片所包含的对象的视觉角度与所述第一图片中同一对象的视觉角度相同和/或互补的一张或者多张推送图；
    组合所述一张或者多张推送图和第一图片得到推送信息。
  11. 根据权利要求10所述的信息推送方法，其特征在于，所述组合所述一张或者多张推送图和第一图片得到推送信息，具体包括：
    将所述一张或者多张推送图与所述第一图片添加到所述推送信息中；
    在所述推送信息中增加所述一张或者多张推送图与所述第一图片的预设位置信息；
    将相应各图片的时间信息附加到所述推送信息中。
  12. 一种信息推送方法，其特征在于，所述信息推送方法包括：
    上传第一图片和第一图片相关信息；
    接收推送消息，所述推送消息中携带所述第一图片，以及一张或者多张推送图片；其中，所述一张或者多张推送图片与所述第一图片所包含的一个或者多个目标对象处于相同的地理位置。
  13. 根据权利要求12所述的信息推送方法，其特征在于，所述方法还包括：
    上传终端的历史行径数据，其中所述历史行径数据包括历史行径的位置信息和时间信息；
    设置不接收与所述历史行径的位置信息相同，且与历史行径的时间差在预设阈值内的推送消息。
  14. 根据权利要求12或13所述的信息推送方法，其特征在于，所述方法还包括：
    设置接收推送消息中所携带的推送图片数量、推送图片分辨率、推送图片来源和推送图片中目标对象角度中的一项或者多项。
  15. 一种信息推送装置，其特征在于，包括：
    第一图片接收模块，用于接收第一终端上传的第一图片；
    第二图片接收模块，用于接收第二终端上传的第二图片；
    推送消息模块，用于分析所述第一图片和第二图片，确认所述第一图片和第二图片中，相对于同一位置的目标对象之间的相似度小于第一预设阈值，发送对应所述目标对象的推送信息给所述第一终端。
  16. 根据权利要求15所述的信息推送装置，其特征在于，接收第一终端上传的第一图片同时，还获取到第一图片相关信息，其中所述第一图片相关信息包括所述第一图片的内容位置信息、图片分辨率、生成时间信息中的一项或者多项，所述装置还包括归类管理模块，所述归类管理模块连接所述第一图片接收模块，具体的：
    所述归类管理模块，用于校对所述第一图片与所述第一图片相关信息，并 根据校对结果归类管理所述第一图片。
  17. 一种信息推送装置，其特征在于，所述信息推送装置包括：
    上传模块，用于上传第一图片和第一图片相关信息；
    接收推送消息模块，用于接收推送消息，所述推送消息中携带所述第一图片，以及一张或者多张推送图片；其中，所述一张或者多张推送图片与所述第一图片所包含的一个或者多个目标对象处于相同的地理位置。
  18. 根据权利要求17所述的信息推送装置，其特征在于，所述上传模块还包括：
    历史行径数据上传模块，用于上传终端的历史行径数据，其中所述历史行径数据包括历史行径的位置信息和时间信息。
  19. 根据权利要求18所述的信息推送装置，其特征在于，所述装置还包括第一设置模块，具体的：
    所述第一设置模块，用于设置不接收与所述历史行径的位置信息相同，且与历史行径的时间差在预设阈值内的推送消息。
  20. 一种非易失性计算机存储介质，其特征在于，所述计算机存储介质存储有计算机可执行指令，该计算机可执行指令被一个或多个处理器执行，用于完成权利要求5-11任一所述的信息推送方法，或者用于完成权利要求12-14任一所述的信息推送方法。
PCT/CN2016/102931 2016-10-21 2016-10-21 一种信息推送方法、装置和*** WO2018072207A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2016/102931 WO2018072207A1 (zh) 2016-10-21 2016-10-21 一种信息推送方法、装置和***
CN201680002672.3A CN107077507B (zh) 2016-10-21 2016-10-21 一种信息推送方法、装置和***

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/102931 WO2018072207A1 (zh) 2016-10-21 2016-10-21 一种信息推送方法、装置和***

Publications (1)

Publication Number Publication Date
WO2018072207A1 true WO2018072207A1 (zh) 2018-04-26

Family

ID=59623723

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/102931 WO2018072207A1 (zh) 2016-10-21 2016-10-21 一种信息推送方法、装置和***

Country Status (2)

Country Link
CN (1) CN107077507B (zh)
WO (1) WO2018072207A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111444143A (zh) * 2020-02-26 2020-07-24 车智互联（北京）科技有限公司 图片集中的图片展示方法、装置及移动终端
US11073483B2 (en) * 2018-10-10 2021-07-27 Chengdu Boe Optoelectronics Technology Co., Ltd. Display device and bonding accuracy detection method
CN114710554A (zh) * 2022-03-30 2022-07-05 北京奇艺世纪科技有限公司 一种消息处理方法、装置、电子设备及存储介质
CN116150420A (zh) * 2023-04-23 2023-05-23 北京微应软件科技有限公司 一种图片任务推送结果的评估方法和***

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108549660B (zh) * 2018-03-12 2022-08-02 维沃移动通信有限公司 信息推送方法及装置
CN108763532A (zh) * 2018-05-31 2018-11-06 上海掌门科技有限公司 用于推送信息、展现信息的方法和设备
CN110474987A (zh) * 2019-08-21 2019-11-19 北京联创众升科技有限公司 一种事件信息推送方法及装置
CN111147469B (zh) * 2019-12-20 2022-10-11 航天信息股份有限公司 一种对终端设备上传数据进行内容验证的方法及***

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040239762A1 (en) * 2003-05-21 2004-12-02 Porikli Fatih M. Adaptive background image updating
CN102547090A (zh) * 2010-11-24 2012-07-04 三星电子株式会社 数字拍摄设备及其提供照片的方法
CN104767782A (zh) * 2014-01-08 2015-07-08 腾讯科技（深圳）有限公司 一种关联照片事件的方法和装置
CN105528444A (zh) * 2015-12-23 2016-04-27 北京奇虎科技有限公司 一种相册生成方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040239762A1 (en) * 2003-05-21 2004-12-02 Porikli Fatih M. Adaptive background image updating
CN102547090A (zh) * 2010-11-24 2012-07-04 三星电子株式会社 数字拍摄设备及其提供照片的方法
CN104767782A (zh) * 2014-01-08 2015-07-08 腾讯科技（深圳）有限公司 一种关联照片事件的方法和装置
CN105528444A (zh) * 2015-12-23 2016-04-27 北京奇虎科技有限公司 一种相册生成方法和装置

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11073483B2 (en) * 2018-10-10 2021-07-27 Chengdu Boe Optoelectronics Technology Co., Ltd. Display device and bonding accuracy detection method
CN111444143A (zh) * 2020-02-26 2020-07-24 车智互联（北京）科技有限公司 图片集中的图片展示方法、装置及移动终端
CN111444143B (zh) * 2020-02-26 2024-02-23 车智互联（北京）科技有限公司 图片集中的图片展示方法、装置及移动终端
CN114710554A (zh) * 2022-03-30 2022-07-05 北京奇艺世纪科技有限公司 一种消息处理方法、装置、电子设备及存储介质
CN114710554B (zh) * 2022-03-30 2024-04-26 北京奇艺世纪科技有限公司 一种消息处理方法、装置、电子设备及存储介质
CN116150420A (zh) * 2023-04-23 2023-05-23 北京微应软件科技有限公司 一种图片任务推送结果的评估方法和***
CN116150420B (zh) * 2023-04-23 2023-08-11 北京微应软件科技有限公司 一种图片任务推送结果的评估方法和***

Also Published As

Publication number Publication date
CN107077507B (zh) 2021-06-04
CN107077507A (zh) 2017-08-18

Similar Documents

Publication Publication Date Title
WO2018072207A1 (zh) 一种信息推送方法、装置和***
US11917037B2 (en) Methods, devices, and systems for determining a subset for autonomous sharing of digital media
US12001475B2 (en) Mobile image search system
US9805065B2 (en) Computer-vision-assisted location accuracy augmentation
US9565534B2 (en) Apparatus and method for identifying point of interest in contents sharing system
CN103635953B (zh) 使用用户特定内容增强可视数据流的***
US9830337B2 (en) Computer-vision-assisted location check-in
US20140343984A1 (en) Spatial crowdsourcing with trustworthy query answering
US9727565B2 (en) Photo and video search
CN110298269B (zh) 场景图像定位方法、装置、设备及可读存储介质
US9317600B2 (en) View of a physical space augmented with social media content originating from a geo-location of the physical space
US10346495B2 (en) System and method for large scale crowdsourcing of map data cleanup and correction
JP2019502152A (ja) 地図更新方法および車載端末
US20220383053A1 (en) Ephemeral content management
WO2014090034A1 (zh) 实现增强现实应用的方法及设备
WO2018152822A1 (zh) 一种生成相册的方法、装置和移动终端
US11416542B1 (en) System and method for uploading still images of matching plates in response to an alert hit list using distributed LPR acquisition
WO2020098431A1 (zh) 一种构建地图模型的方法及装置
CN105956091A (zh) 扩展信息获取方法及装置
CN105635318B (zh) 图像获取方法和***
TW201814553A (zh) 適地性空間物件遠距管理方法與系統
CN113063421A (zh) 导航方法及相关装置、移动终端、计算机可读存储介质
TWI613550B (zh) 相片及視頻分享
CN105069718B (zh) 基于移动物联网的智慧国土不动产自助监测方法和***
KR20190091214A (ko) 동영상으로부터 장소 정보를 추출하는 장치 및 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16919417

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 14/08/2019)

122 Ep: pct application non-entry in european phase

Ref document number: 16919417

Country of ref document: EP

Kind code of ref document: A1