CN108564087A - Risk Identification Method, device, terminal and the storage medium of small advertisement - Google Patents

Risk Identification Method, device, terminal and the storage medium of small advertisement Download PDF

Info

Publication number
CN108564087A
CN108564087A CN201810333099.5A CN201810333099A CN108564087A CN 108564087 A CN108564087 A CN 108564087A CN 201810333099 A CN201810333099 A CN 201810333099A CN 108564087 A CN108564087 A CN 108564087A
Authority
CN
China
Prior art keywords
image
risk identification
risk
small advertisement
character string
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810333099.5A
Other languages
Chinese (zh)
Other versions
CN108564087B (en
Inventor
罗芳柠
高威煌
许振声
林钊武
宋平波
廖唯心
梁立安
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen Co Ltd
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 Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to CN201810333099.5A priority Critical patent/CN108564087B/en
Publication of CN108564087A publication Critical patent/CN108564087A/en
Application granted granted Critical
Publication of CN108564087B publication Critical patent/CN108564087B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10Character recognition
    • G06V30/14Image acquisition
    • G06V30/148Segmentation of character regions
    • G06V30/153Segmentation of character regions using recognition of characters or words
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/22Matching criteria, e.g. proximity measures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0635Risk analysis of enterprise or organisation activities

Abstract

This application discloses a kind of Risk Identification Methods of small advertisement, device, terminal and storage medium, in the method, terminal obtains the image capture instruction to small advertisement to be identified, to carry out Image Acquisition to the small advertisement of pending risk identification, and obtain the image file of the small advertisement of the pending risk identification, from target image of the extraction in cog region in the image file, to obtain and export the risk identification result for the target image, so that user can recognize institute's risk that may be present in the target image in time, to in time, it efficiently prompts the user with and whether there is risk in small advertisement, reduce the risk that user is swindled or misled.

Description

Risk Identification Method, device, terminal and the storage medium of small advertisement
Technical field
This application involves technical fields more particularly to a kind of Risk Identification Method of small advertisement, device, terminal and storage to be situated between Matter.
Background technology
In daily life, user often touches various types of small advertisement, e.g., on the wall of public place Various types of promotional pamphlets etc. that the paper-made advertising of stickup, street corner distribute, and in these small advertisements include much swindle, false a surname It passes, intermediary misleads equivalent risk information.
Due to user it is difficult to intuitively identify that small advertisement whether there is risk, so being highly susceptible to luring for small advertisement It leads, and participates in the activity that small advertisement is publicized, eventually lead to user and be spoofed or user is caused to dislike.Such as, Yong Huyi The small advertisement puted up according to public place, may dial into the phone of swindle advertisement, and lead to user's monetary losses.As it can be seen that such as Risk information in what identification small advertisement to reduce the case where user is misled by small advertisement is that those skilled in the art compel to be essential Technical problems to be solved.
Invention content
In view of this, this application provides a kind of Risk Identification Method of small advertisement, device, terminal and storage medium, with Allow user in time, efficiently to identify in small advertisement institute's risk information that may be present, reduce user swindled or The risk of misleading.
To achieve the above object, on the one hand, this application provides a kind of Risk Identification Methods of small advertisement, including:
The image capture instruction to small advertisement to be identified is obtained, is adopted with carrying out image to the small advertisement of pending risk identification Collection;
Obtain the image file of the small advertisement of the pending risk identification;
The target image in cog region in described image file is extracted, the target image is described image file A part;
The risk identification for the target image is obtained as a result, the risk identification result is for characterizing the target figure With the presence or absence of the risky character string of tool as in;
Export the risk identification result.
In one possible implementation, terminal can send the target image to risk identification module, with logical It crosses the risk identification module and identifies whether the character string for including in the target image belongs to the risky character string of tool, obtain Risk identification result.
In another possible realization method, the image text of the small advertisement for obtaining the pending risk identification Part, including:
Show the image file of the small advertisement of the collected pending risk identification in image preview window, it is described Cog region is shown in image preview window;
When meeting image recognition condition, the described image file currently showed in described image the preview window is obtained.
Another aspect, present invention also provides a kind of risk identification devices of small advertisement, including:
Instruction acquisition unit, for obtaining the image capture instruction to small advertisement to be identified, with to pending risk identification Small advertisement carry out Image Acquisition;
Image acquisition unit, the image file of the small advertisement for obtaining the pending risk identification;
Image extraction unit, for extracting the target image in described image file in cog region, the target figure A part as being described image file;
Risk acquiring unit, for obtaining the risk identification for the target image as a result, the risk identification result For characterizing in the target image with the presence or absence of the risky character string of tool;
As a result output unit, for exporting the risk identification result.
Another aspect, present invention also provides a kind of terminals, including:
Processor and memory;
Wherein, the processor is for executing the program stored in the memory;
For storing program, described program is at least used for the memory:
The image capture instruction to small advertisement to be identified is obtained, is adopted with carrying out image to the small advertisement of pending risk identification Collection;
Obtain the image file of the small advertisement of the pending risk identification;
The target image in cog region in described image file is extracted, the target image is described image file A part;
The risk identification for the target image is obtained as a result, the risk identification result is for characterizing the target figure With the presence or absence of the risky character string of tool as in;
Export the risk identification result.
It is executable to be stored with computer present invention also provides a kind of storage medium for another aspect in the storage medium Instruction, when the computer executable instructions are loaded and executed by processor, realizes the wind of any one of them small advertisement as above Dangerous recognition methods.
As it can be seen that during user wants to know about small advertisement whether include risk information when, it is only necessary to by terminal acquisition wait for into The image file of the small advertisement of row risk identification, and control and the parts of images for carrying out risk identification is needed to be in the image file In cog region, in this way, terminal can extract the target image in cog region in the image file, and the mesh is got In logo image with the presence or absence of the risk identification for having risky character string as a result, so that terminal can export it is small wide for this The risk identification of announcement is as a result, so that user can fast, timely get the risk in the presence of small advertisement, reduction quilt Fraud or the probability misled.
Description of the drawings
In order to more clearly explain the technical solutions in the embodiments of the present application, make required in being described below to embodiment Attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is only embodiments herein, for ability For the those of ordinary skill of domain, without creative efforts, it can also be obtained according to the attached drawing of offer other Attached drawing.
Fig. 1 shows a kind of composition block schematic illustration of risk recognition system of the embodiment of the present application;
Fig. 2 shows a kind of a kind of flows of the Risk Identification Method of small advertisement of the embodiment of the present application to interact schematic diagram;
Fig. 3 shows an exemplary plot of terminal realization indicating risk in the embodiment of the present application;
Fig. 4 shows the another exemplary plot of terminal realization indicating risk in the embodiment of the present application;
Fig. 5 shows the exemplary plot that risk information library is built in the embodiment of the present application;
Fig. 6 shows a kind of another flow of the Risk Identification Method of small advertisement interaction schematic diagram of the embodiment of the present application;
Fig. 7 shows a kind of exemplary plot that the target image in cog region is adjusted in the embodiment of the present application;
Fig. 8 and Fig. 9 shows that the embodiment of the present application realizes an exemplary plot of risk identification in a kind of application scenarios;
Figure 10 shows a kind of composition schematic diagram of risk identification device one embodiment of small advertisement of the embodiment of the present application;
Figure 11 shows a kind of composed structure schematic diagram of terminal of the embodiment of the present application.
Specific implementation mode
During the scheme of the application can browse the small advertisements such as advertisement, the promotional pamphlet of public place stickup in user, Can more in time, easily provide (also referred to as risk is believed about whether having risky character string in the small advertisement to the user Breath) prompt, to reduce the risk that user is misled by false propaganda or fraud information.
Wherein, in the embodiment of the present application, small advertisement i.e. small size advertisement, refer generally to be posted, spray to metope And the advertisement in the public domains such as road surface, or the handbill that is distributed in places such as community, roads.
In the embodiment of the present application, there are the character strings of risk may include:With exist swindle, fraud, intermediary mislead or There are associated character strings for the object of person's false propaganda equivalent risk behavior.Wherein, there are the object of risk behavior can be product, User, producer or shop etc..And with there are the associated character string of the object of risk behavior can be object identity characteristic mark Know, the trade mark etc. associated by contact method and object.
For example, such as, by be there are the object of risk behavior there are the user of fraudulent act for, then the surname of the user What the identity such as name, identification card number, the contact methods such as telephone number, the address of the user and the user were promoted Name of product all can be identified the character string belonged to there are risk.
For another example, by be there are the object of risk behavior in the presence of the producer of deception consumer for, then the title of the producer, factory Fraud product, the responsible person of the producer, the producer and the producer that the brand name held of family, the producer are sold it is negative The address for blaming people can be as there are the character strings of risk.
For the ease of understanding the indicating risk method of the application, below for the Risk Identification Method of the small advertisement of the application The system being applicable in is introduced.Referring to Fig. 1, it illustrates a kind of a kind of structure composed signals of risk recognition system of the application Figure.
As shown in Figure 1, risk recognition system provided by the embodiments of the present application includes:Terminal 10 and server 20.Terminal 10 It is realized and is communicated to connect by network 30 between server 20.
Wherein, terminal 10 can be the mobile terminals such as mobile phone, tablet computer, or be connected with the personal meter of camera The fixed terminals such as calculation machine.
In the embodiment of the present application, terminal 10 can the image acquisition components such as camera of terminal setting or connection adopt Collection needs to carry out the corresponding image of small advertisements such as handbill, promotional pamphlet of risk identification, and by network 30 by collected figure As being transferred to server 20.
Correspondingly, character string included in the image that the meeting identification terminal of server 20 is sent, and match what this was identified Whether character string belongs to the risky character string of tool, and matching result is returned to terminal, so that terminal can export risk knowledge Not as a result, to achieve the purpose that prompt in user's small advertisement with the presence or absence of the risky character string of tool.
In order to reduce the flow spent by terminal transmission image, terminal 10 can also collect image to be identified it Afterwards, parts of images is extracted from the image as the target image for needing to be transferred to server, wherein the target image belongs to this The part of terminal the image collected.
In the case of a kind of possible, in order to ensure that the target image is comprising the part figure for needing identification risk information Picture, then can be when terminal collects the preview image of small advertisement, in the preview image window for the preview image for showing small advertisement In add a cog region, e.g., increase an identification frame and identify the cog region, to extract the target being in cog region Image.In Fig. 1, the screen of terminal 10 may be considered a preview image window, be also set up in the preview image window There are one frame 11 is identified, the region in the identification frame is cog region.In this way, user can by mobile image acquisition component with wait for Acquire the relative position between image, it would be desirable to being partially moved in the cog region for risk identification is carried out, so that terminal acquires Target image in the cog region.
In the case of another is possible, after terminal collects the image (e.g., preview image) of small advertisement, Ke Yijian The coordinate points that user specifies in the images are surveyed, and terminal can determine a cog region using the coordinate points as datum mark, Such as, it constructs comprising the datum mark and the cog region attached most importance to the datum mark, for example, the cog region can be with the datum mark Centered on rectangle.Correspondingly, terminal can extract the target image in cog region from the image collected, with Target image is sent to server.
Optionally, application can have been run in terminal, which communicates to connect for being established with server, and terminal is by answering Information exchange is carried out with the server.
The interactive process between terminal and server describes in detail below.
Such as, referring to Fig. 2, it illustrates a kind of interactions of the flow of Risk Identification Method one embodiment of small advertisement of the application The method of schematic diagram, the present embodiment may include:
S201, terminal obtains the image capture instruction to small advertisement to be identified, with the small advertisement to pending risk identification Carry out Image Acquisition.
Wherein, which carries out Image Acquisition for triggering terminal to the small advertisement of pending risk identification. Image Acquisition herein is the image file that the camera of terminal acquires the small advertisement of pending risk identification, e.g., terminal scanning The image file of the small advertisement, and shown in the form of preview image;It can also be the image file for absorbing the small advertisement.
In the case of a kind of possible, after terminal operating application, if a terminal detects that user clicks in the application The function items for the risk of small advertisement for identification being arranged, then terminal get the image capture instruction to small advertisement to be identified, from And terminal is allow to carry out Image Acquisition, to acquire the small advertisement of pending risk identification.
In the case of another is possible, after starting small advertisement identification function in the application of terminal, application can call Camera acquires the image file of small advertisement to be identified, in that case, if detecting that the acquisition of user's input picture refers to It enables, then terminal can absorb collected small advertisement to be identified, to obtain the image text of the small advertisement of pending risk identification Part.
S202, terminal obtain the image file of the small advertisement of the pending risk identification.
S203, terminal extract the target image in cog region in the image file.
Wherein, which is a part for the image file got in step S202.
In the embodiment of the present application, after the image file of small advertisement of pending risk identification being got in terminal, and Image file server identification is not sent to not instead of directly, from extracting the part figure in cog region in image file Picture, using the parts of images in cog region as the target image for needing progress risk information identification.
Wherein, cog region may refer to the related introduction of front.Such as, cog region may be displayed in image preview window, In this way, the image file of collected small advertisement can be presented in preview form in the image preview window by terminal, and locate In the parts of images in the cog region of image preview window be considered as in the corresponding image file of small advertisement pending risk know Other target image.
Optionally, application can have been run in the terminal, and e.g., which can be using house keeper etc., and terminal can pass through The risk identification to small advertisement is realized in the application.Such as, a function choosing-item for risk identification can be configured in this application. Correspondingly, when terminal detects that user clicks the function choosing-item by application, then terminal confirmly detects the Image Acquisition and refers to It enables, and Image Acquisition is started by the application, to acquire the image file of small advertisement, and then execute the related behaviour of subsequent terminal side Make.
It is understood that the condition that triggering terminal obtains the image file of the small advertisement of pending risk identification can have A variety of, in a kind of possible situation, terminal obtains current collected pending risk identification in response to the image capture instruction Small advertisement image file.
In the case of another is possible, is determined in terminal and currently meet risk identification condition (also referred to as image recognition item Part) when, obtain the image file of the small advertisement of the collected pending risk identification.For example, applying it in terminal operating Afterwards, if a terminal detects that image capture instruction, then the terminal can start Image Acquisition, to acquire the small of pending risk identification The image file of advertisement, certainly, terminal can also be shown the image collected file in the form of preview image;Accordingly , when meeting risk identification condition, terminal can obtain the image text of the small advertisement of current collected pending risk identification Part, so from the image file extract cog region in target image.
Wherein, which can be set as needed.Such as, terminal can be touched when detecting preset instructions Hair obtains the image file of current collected small advertisement.For example, a key range can be shown in the interface of terminal, when When detecting that user clicks the key range, then can triggering terminal obtain current the image collected file, so as to image text Part caches and carries out the relevant treatments such as image zooming-out.
For another example, detect that the duration that the image in cog region remains static is more than preset duration in terminal, then terminal Current the image collected file can be obtained, and from the target image extracted in image file in the cog region.It is appreciated that It is in the case of showing to have cog region in image preview window, if user is determining that the parts of images in cog region is to need The image to be identified, just will not be in the position etc. of preview image or adjustment cog region in mobile image the preview window, at this In the case of kind, so that it may to obtain the image collected file, and extract the target image in the cog region.
Certainly, risk identification condition can also have other possible, not limit herein.
In order to make it easy to understand, in conjunction with a kind of interface operation example of Fig. 3, step S201 to S203 as above is illustrated.Such as Fig. 3 In (a) user click terminal in apply in risk identification function after show the interface for including " sweep and sweep " button.If User, which needs to identify, whether there is risk information in the advertisement pasted, promotional pamphlet or image, then can click " sweep and sweep " Button, so that camera is opened in the application in triggering terminal, to acquire the small advertisement of pending risk identification by camera Image file.Correspondingly, as shown in (b) in Fig. 3, while camera is opened in the application in terminal, image preview window can be showed Mouth 301, and identification frame 302 is shown in the image preview image, the region which includes is a cog region 302. In the Fig. 3 be by taking image preview window is the entire screen of terminal as an example, in practical applications, the image preview window 301 Or it is presented in a window on screen upper layer, which only occupies the area of screen.
In Fig. 3 shown in (c), user it is pre- can to adjust the image by constantly adjusting the image-capture field of camera The picture material presented look in image window, and finally so that the parts of images in the preview image in cog region is desirable Carry out the parts of images of risk identification.In the Fig. 3 in (b), preview image is a handbill pasted somewhere, and The string number number (e.g., phone number) for including in preview image by frame in the identification frame 302, in this way when terminal detects After user is static, then show the image file of small advertisement in the terminal buffers image preview window, and extract the image file In target image in the corresponding cog region of identification frame.
Optionally, the position of the cog region in image preview window can be adjusted, correspondingly, terminal is received to the figure As the adjustment operation of the position of cog region in the preview window, and adjustment operation is responded, adjusts the cog region in the preview image In position.Such as, in Fig. 3 (c) user can adjust identification frame can in image preview window position, to change identification The corresponding cog region of frame is in the position coordinates in preview image.
It should be noted that above step S201 to step S203 is with terminal-pair small advertisement progress Image Acquisition and to adopting The image collected is introduced for extracting.But it is understood that first passing through intake or screenshotss etc. in advance in terminal In the case that mode saves the image file of small advertisement, need carry out risk identification when, terminal can also receive user from The operation of the image file of small advertisement is chosen in the image of preservation, and the image file of the small advertisement of user's selection is extracted Processing.
Such as, it is illustrated in conjunction with another example of Fig. 4.It is the image preserved from terminal with user in the example of fig. 4 For the image file of middle selection small advertisement.After the image identification function for starting application, terminal can be shown in Fig. 4 (a) Interface further include for triggering local image in the Fig. 4 in the interface of (a) other than comprising " sweep and sweep " option 401 " selection has photo " option 402 of selection, terminal detects that user clicks and is somebody's turn to do " selection has photo ", then terminal gets figure As acquisition instructions, in that case, in terminal the application can show local picture selection interface 403, (b) is shown in Fig. 4. After user selects and needs the image of preview, in terminal the application can show image preview window 404, and in image preview Show the preview image of user's selection in window, meanwhile, identification frame 405, (c) institute in Fig. 4 are presented on the preview image Show.User adjusts the position of the identification frame in the image preview window, to outline the target image for needing risk identification, The static duration of frame is identified more than after preset duration, terminal extracts the target image in the corresponding cog region of identification frame.
Meanwhile target image and target image in the image file for the ease of distinguishing small advertisement in cog region Except parts of images will be in the image file of small advertisement except identification frame in (c) in (c) in figure 3 and Fig. 4 Image blurred, and will identify frame in image clearly show.
The target image is sent to server by S204, terminal.
Such as, which is transferred to clothes by terminal based on the application in the terminal and the communication connection between server Business device.
Optionally, terminal can generate the request of the risk identification comprising the target image, so that server responds the risk Identification request, and the character to including in the target image carries out risk identification.
S205, server identify target string included in the target image.
It is understood that comprising characters such as some numbers, letter and words in target image, and in order to detect the mesh The content for whether belonging to risk information in logo image comprising some needs first to carry out character recognition to picture material, to obtain mesh The character string for including in logo image, to so that server can know the content information described in target image.
Such as, server can by optical character identification (Optical Character Recognition, OCR) technology, Identify the character string for including in the target image.For example, the character string presented in target image is converted to by OCR technique The character string of textual form, so that server can get the character content for including in the target image.
For the ease of distinguishing, in the embodiment of the present application, the character string identified from the target image is known as target word Symbol string.
It optionally, can also be from the character string identified after the character string for including in identifying the target image Extract the target string of specified type.Wherein, which can be preassigned character types, e.g., the specified class Type may include:The character types such as number, letter, word.
It is understood that in different application scene, user needs to pay close attention to the class of the risk information identified in other words Type also can difference.It is largely that telephone number is shown in handbill for example, for the handbill of stickup, with Other users are made a phone call to seek advice from the content of handbill, in that case, it may be necessary to pay close attention to advertisement Whether the telephone number in list belongs to the risky telephone number of tool.Correspondingly, server needs are identified from target image The character string (that is, telephone number) that continuous multiple numbers are constituted.
S206, server is by the target string and the risky string matching of preset tool.
Wherein, having risky character string can analyze in advance, belong to swindle, induction user or falseness Publicize the character string of equivalent risk feature.
Such as, can be able to include multiple with feature of risk in the risk information library with a preset risk information library Character string, in this way, server can match the target string with the character string in the risk information library, if matching To the same or similar satisfactory character string of degree, it is determined that the target string belongs to the risky character string of tool.
It illustrates, it is assumed that target string is telephone number, then at least can include more in preset risk information library A telephone number for belonging to the properties such as swindle, advertisement or intermediary, in this way, can identify the target figure based on the risk information library Whether the telephone number as in also belongs to the telephone number with properties such as swindles.For another example, it is assumed that target string is a quotient The name of an article claims, then can include the trade name of multiple false commodity in preset risk information library, correspondingly, if believing from risk Breath identifies trade name identical with the trade name in the target image in library, it is determined that the trade name in the target image Title belongs to risk information.
Optionally, it is contemplated that there are many risk classifications, and e.g., risk classifications may include:Falseness is promoted, is swindled, in advertisement The type that is situated between etc. can be with the preset risky word of tool for the concrete type of risk possessed by the clear target string Correspondence set between symbol string and risk classifications, correspondingly, can match whether there is and be somebody's turn to do from the correspondence set The corresponding at least one risk classifications of target string.If there is the corresponding at least one risk classifications of the target string, Then illustrate that the target string belongs to the risk character string at least one risk classifications.
Such as, the correspondence set between the character string and risk classifications can be stored in the risk information library, accordingly , the risky character string of a variety of tools is contained in the risk information library and each has corresponding to risky character string At least one risk classifications.
Wherein, the information such as the risky character string of the above preset tool, risk classifications can be by server from the network platform The data got can also be to be obtained from the risk information of user's report.Such as, referring to Fig. 5, it illustrates server generations The exemplary plot in risk information library.
As shown in Figure 5, on the one hand, server can receive the risk information that user is reported by terminal, e.g., risk letter Breath may include:Have risky character string and risk classifications, for example, address A, risk classifications:Multiple level marketing.Correspondingly, service Device can count the report number that different user reports each character string, when the report number of some character string is more than default time The character string, then is determined as having risky character string by number, and by character string, the corresponding risk classifications of character string and report Number is added to risk information library.On the other hand, server can obtain the risk information of network platform offer, by risk information It is added to risk information library.Wherein, the network platform can be by big data analysis in network etc., obtaining risk information.
S207 generates indicating risk when server matches to the target string, which belong to, has risky character string.
Wherein, which is used to indicate the target string in the target image included and belongs to the risky character of tool String.
Optionally, belong to the risky character string of tool in server matches to the target string, and be matched to the target In the case of at least one risk classifications that character string has, the indicating risk which generates can carry the target word The corresponding at least one risk classifications of symbol string.Such as, which can serve to indicate that the target string has risk, and wind Dangerous type is at least one risk classifications matched.
Optionally, in the case where presetting the report number corresponding to the risky character string of tool, which may be used also When belonging to the risky character string of tool to be matched to the target string, the report that the target string is reported can also be obtained Number.Correspondingly, the indicating risk, which can also carry the target string, is reported report number for risk information.
Risk instruction is sent to the terminal by S208, server.
S209, terminal export the indicating risk.
Wherein, terminal export the indicating risk mode can there are many may, e.g., terminal by voice mode output should Indicating risk;Can also be by exporting the indicating risk in the terminal with written form.
Such as referring to Fig. 3, after the target image in cog region is sent to server by terminal, if server identifies Include risk information in the target image, then can return to an indicating risk.And terminal can as in Fig. 3 (d), this using defeated The upper layer of the image preview window gone out shows there are one prompting frame 303, is shown in the prompting frame in identification frame 302 Telephone number " 13603****12 " belongs to the indicating risk of advertisement promotion phone.Meanwhile the electricity is also shown in the prompting frame Talk about the report number of number, that is, " 13603****12 is advertisement promotion phone by 200 user's marks ".For Fig. 4 examples In, (d) is identical as (d) in Fig. 3 in Fig. 4, repeats no more.
It should be noted that the present embodiment be with cog region be one be introduced, but it is understood that, identification Area can also there are many, e.g., the multiple coordinate points that can be pointed out according to user construct multiple cog regions or terminal in image Multiple cog regions etc. are showed in the preview window.The case where for multiple cog regions, terminal can be by the mesh in multiple cog regions Logo image is sent to server identification, and the process of target image in one cog region of the process and processing is similar, herein not It repeats again.
It is understood that being risky there is tool in server recognition target image in the embodiment of fig. 2 In the case of character string, server illustrates for returning to risk instruction to terminal, but it is understood that, server can Based on the preset risky character string of tool, to come with the presence or absence of the risky character string of tool in recognition target image, and obtain Risk identification result.The risk identification result for characterize the character string for including in the target image whether belong to tool it is risky Character string.Correspondingly, the risk identification result can be returned to terminal by the server, so that terminal exports the risk identification knot Fruit.
It should be noted that Fig. 1 be by risk recognition system include server for illustrate, correspondingly, be in fig. 2 with Terminal is illustrated for risky character string to identify in the target image with the presence or absence of having by server.But it is understood that , terminal can also complete the target image by being arranged beyond the clouds for carrying out the risk identification module of risk identification Risk identification or terminal in risk identification module is set, with the character string in the terminal Direct Recognition target image, And judge whether the character string in target image belongs to the risky character string of tool.In that case, risk recognition system Can not include above-mentioned server, and the application is for terminal is identified by which kind of mode class in the target image The no concrete mode that there is the risky character string of tool does not limit, as long as terminal can be got for the target image Risk identification result.
Optionally, in terminal, server or high in the clouds, setting can be useful for the character string in identification image, and to word Symbol string carries out the matched risk identification module of risk and is e.g. provided with risk identification module in the server, the risk identification module It can complete the relevant operation of front server side.It may alternatively be correspondingly, target image is sent to server by terminal: Terminal sends target image to risk identification module, to identify the word for including in the target image by the risk identification module Whether symbol string belongs to the risky character string of tool, obtains risk identification result.
As it can be seen that during user wants to know about small advertisement whether include risk information when, it is only necessary to by terminal acquisition wait for into The image file of the small advertisement of row risk identification, and control and the parts of images for carrying out risk identification is needed to be in the image file In cog region, in this way, terminal can extract the target image in cog region in the image file, and the mesh is got With the presence or absence of the risk identification for having risky character string as a result, so that terminal output user is small wide for this in logo image The risk identification of announcement is as a result, so that user can fast, timely get the risk in the presence of small advertisement, reduction quilt Fraud or the probability misled.
Meanwhile compared with situations such as image file of small advertisement is directly passed to server or high in the clouds by terminal, terminal from Target image of the extraction in cog region is transmitted in the image collected file, not only improves the number for reducing transmission image According to amount, reduces flow and expend;The risk identification mould in risk identification module, server or the server of terminal can be reduced again Block carries out the data volume of character recognition, is also beneficial to that the character string for needing risk identification is accurately positioned, improves the essence of risk identification Accuracy.
It is understood that in order to improve the accuracy of risk information library risk information, in server or high in the clouds Risk identification module etc. needs the report situation in conjunction with multiple users, updates the risk information library.Particularly, for some curbsides The small advertisements such as the psoriasis advertisement with harm, terminal are also required to the source of the target image provided to risk identification module (e.g., geographical location, web page address etc.) provides foundation with follow-up manual cleaning these psoriasis advertisements etc..Based on this, terminal is also After receiving user for the report request of target image report information can be sent to risk identification module.
In addition, in order to further decrease volume of transmitted data, the risk identification module transmission in terminal to server or high in the clouds Before target image, it is also necessary to carry out the processing such as compressing to target image.
It describes in detail with reference to Fig. 6.For ease of description, target is completed by server with terminal in figure 6 It is introduced for the risk identification of image.Such as Fig. 6, it illustrates a kind of Risk Identification Method of small advertisement of the application another The flow of embodiment interacts schematic diagram, and the method for the present embodiment may include:
S601, terminal operating application.
The application has the function of identifying the risk information for including in image and exports prompt.Such as, by taking mobile phone as an example, this is answered With that can be mobile phone pipe man, the component that part is used for risk identification have been increased newly in mobile phone house keeper.
S602, this in terminal are applied when detecting risk identification request, and identification operation interface is showed.
Such as, terminal operating application, to detect the behaviour for the different function option that user provides the application by application Make.
S603 calls camera when applying when detecting the image capture instruction to small advertisement in identifying operation interface Image is acquired, the image file of current collected small advertisement is presented to image preview window.
Cog region is presented in the image preview window.
Wherein, there is the option for triggering Image Acquisition, the interface that (a) is shown in Fig. 3 in the identification operation interface In " sweep and sweep " option.
Wherein, the image file of the small advertisement of the collected pending risk identification might not in step S603 To need to extract and the image file of risk identification, therefore, in step S603 can also will it is collected it is small extensively The image file of announcement is known as the preview image of small advertisement.
It is understood that step S602 and S603 are only a kind of triggering of image file of triggering terminal acquisition small advertisement Mode, in practical applications, terminal when detecting risk identification request, can also directly execute the camera of calling terminal simultaneously The preview image of the current collected small advertisement of camera is obtained, and includes the behaviour in image preview window by preview image Make.
S604, when meeting image recognition condition, using the figure for obtaining the small advertisement currently showed in image preview window As file, and determine coordinate position range of the cog region in image preview window.
Such as, the image file for the small advertisement that application cache image preview window currently shows, so as to subsequently to the small advertisement Image file cut.
Wherein, which may refer to the related introduction of preceding embodiment.Such as, image preview window is detected The duration that middle preview image remains static is more than preset duration.For example, opening camera when terminal carries out Image Acquisition, and It determines after completing image focusing, if sensing that the duration that user maintains camera to remain static is more than preset duration, Then illustrate that active user completes image alignment in cog region, can determine that the image in cog region is that user wishes identification at this time Partial graphical, so that it is determined that currently meeting image recognition condition.
It is understood that before meeting the image recognition condition, user can adjust camera as needed, with most Eventually so that is showed in cog region in image preview window is the parts of images of the pending risk identification of user.For the ease of reason Solution, is introduced in conjunction with Fig. 7.(a) is a flyer in the figure 7, such as the advertisement being pasted onto on wall.In user to end After this in end applies input picture acquisition instructions, as triggered application according to mode shown in (b) in (a) in prior figures 3 and Fig. 3 Start Image Acquisition, then terminal initially collects the image of a part of content comprising the flyer, as schemed in Fig. 7 (b) As showing preview image in the preview window 701, and cog region 702 shows the parts of images of the preview image.Assuming that user Need to identify the seller addresses in the flyer whether be false propaganda seller addresses, it is clear that cog region in (b) in Fig. 7 Portioned product effect advertisement information shown in 702 is simultaneously undesirable.Therefore, user adjusts terminal camera and is passed with the advertisement Relative position relation between list, to change the content of preview image in image preview window 701, and finally so that cog region What is presented in 702 is the parts of images for including seller addresses, in Fig. 7 shown in (c).(b) and (c) in comparison diagram 7 can be seen Go out the variation of target image in preview image and cog region.
Wherein, fixed in position of the cog region in the image preview window, then it can directly execute follow-up Step S605, without determining the corresponding coordinate position range of the cog region.If the cog region is in the image preview window Position can be adjusted, then the application can obtain the current cog region in the preview image (or figure in terminal As the preview window) in coordinate position range.
It is understood that after the preview image for showing collected small advertisement by image preview window, it is based on The identification location point that user specifies in preview image, the mode that cog region is created in preview image are applied equally to this Shen Please embodiment identification zone position, and the mistake of target image of the final extraction in the cog region are determined for this kind of situation Journey is similar, and details are not described herein.
S605 is cut using the coordinate position range based on the cog region from the image file of the small advertisement got Go out the target image in the cog region.
S606 rotates the target image using according to preset rotation mode, and compresses the postrotational target image.
In view of there is a situation where that terminal the image collected is different from the practical angle presentation of object to be absorbed, obtaining It to after the target image, needs first to rotate target image, so that postrotational target image and object before shooting Presentation effect it is identical.Such as, word is perpendicular straight arranged in actual advertisement list, but word may in the target image extracted It will present 90 degree of deflections.Certainly, if terminal be not present referring herein to angle presentation different problems, can not also execute Step S605.
The preset rotation mode can be set according to the imaging angle of the camera of terminal, such as, it is assumed that terminal taking arrives Image there are 90 degree clockwise deflections with the visual effect that actually shows, then can be by the target image along counterclockwise Direction rotated ninety degrees.
It illustrates, it is assumed that the color coded format of preview image is yuv format, and assumes the small advertisement got The coded format of preview image is:
Y1, Y2, Y3, Y4;
Y5, Y6, Y7, Y8;
U1, V1, U2, V2;
Then, it is assumed that need to cut out left half of image of the preview image as target image, then the coding of target image Format can be expressed as:
Y1, Y2;
Y5, Y6;
U1, V1.
And the coded format of the postrotational target image can be:
Y2, Y6;
Y1, Y5;
V1, U1.
Wherein, the purpose of compression goal image is in order to reduce the data volume of transmission, e.g., by the target image of yuv format Then the image of boil down to jpeg format again compresses the image of jpeg format.
S607 sends identification request using to server, and identification request is carried by rotation and compressed target Image.
S608, server carry out OCR identifications to the target image after decompression, identify the character in the target image String.
Such as, server compresses the target image through overcompression of application transport, the target figure after being decompressed Picture.
S609, server extract multiple continuous target strings of specified type from the character string identified.
Such as, in the case of the risk of the telephone number in needing recognition target image, server needs to first pass through OCR knowledges Do not go out after the word in the target image and number, is then extracted from the word and number identified again continuous multiple Number, to obtain the telephone number in the target image.
Optionally, in server from the target image it is unidentified go out character string, or to identify the mesh of specified type Character string is marked, then miscue can be returned to terminal, so that user readjusts the acquisition angles etc. of camera.
S610 whether there is the target string in server matches risk information library, if so, thening follow the steps S611;It is reminded if it is not, then returning to devoid of risk to terminal, the prompting so as to terminal output for prompting that risk is not detected.
S611, server obtain corresponding at least one risk classifications of the target string and every from risk information library The report number of kind risk classifications.
Such as, the report record that different target character string belongs to different risk classifications, report note are stored in risk information library Record can report that the target string belongs to the information such as the informer of each risk classifications, report number.Server can also be from Target string is obtained in risk information library to be reported as the report number with various risk classifications.
S612, server application into terminal send indicating risk.
The indicating risk carries at least one risk classifications and each risk classifications pair that the target string has The report number reported answered.
S613 exports at least one risk classifications and lifted that the target string has using by indicating risk frame The report number of report.
Such as, the example of terminal output indicating risk may refer in Fig. 3 in (d) or Fig. 4 shown in (d).
S614 shows report interface when using receiving report request of the user for the target string.
Such as, in user's checking, the target string really belongs to the character string there are risk, alternatively, user empirically or The target string is identified by other means and belongs to the risky character string of tool, then can initiate to be directed to the target to application The report of character string is asked.For example, when target string is telephone number, if user needs to report that the telephone number is swindle When number, then report request can be initiated.
In conjunction with illustrating for Fig. 3, e.g., also there are one " key is lifted for display in the prompting frame 303 of (d) display in figure 3 Report " option 304.It needs to report the target string in user, (not show in Fig. 3 to which triggering shows the report interface Go out).
S615 is inputted and the relevant report information of the target string using user is obtained in the report interface.
The report information may include the target string, the risk classifications and the target which is reported One or more of source of character string.
Wherein, the source of target string can be there are many situation, and e.g., the source of the target string can navigate to this The appearance position of target string, e.g., source can be the paste position of the advertisement belonging to the target string, show the page Page address etc..
It should be noted that after the report request for receiving user, it is only a kind of realization to show the report interface Mode.In practical applications, terminal can ask to determine report information according to report when receiving report request.
Such as, the report can be obtained and ask targeted target string, and using the target string as report information, And the report for carrying the target string is sent to server, so that the target string is labeled as having risk by server Character string.Meanwhile terminal can be by the indicating risk of output, the corresponding risk classifications of the target string are as the report Ask risk classifications possessed by the target string carried.
For another example, acquired in real time by terminal in user stickup advertisement or promotional pamphlet preview image in the case of, use The position that family is currently located be actually the granting of advertisement paste position or promotional pamphlet etc., in that case, user When initiating report request, the location information for reporting the terminal current can be given tacit consent to, then terminal, which can be got, carries position letter The report of breath is asked, and the position that the location information is occurred as the user's report target string.
S616 is asked using the report for carrying the report information is sent to server.
S617, server are asked according to the report, update the corresponding report record of the target string.
Such as, server is asked according to the report, risk classifications possessed by the target string in update risk information library, Adjust the report number etc. that the target string recorded in the risk information library is reported.
S618, server are reported to terminal return and are successfully indicated, exported in the terminal for prompting user's report success Prompt message.
(e) shows a kind of prompt user's report successfully prompting interface in Fig. 3.
It is understood that being so that after output indicating risk in terminal, terminal is in response to user's in the present embodiment Report request illustrates for sending report request to server.But it is understood that the mesh is not detected in terminal Exist in logo image in the case of having risky character string, the user of terminal still can be by clicking in prompting frame " key act The options such as report ", triggering terminal send the report request for the target string in target image to server.
It is understood that after increasing the report function of terminal to server, server can be according to each end The report at end is asked, and the risk information there are risk is more accurately got.Meanwhile providing risk letter in terminal to server In the case of ceasing corresponding source, it is also beneficial to source of the server according to risk information, is provided to related personnel and needs to clear up Risk information source, so as to promptly and accurately clear up some influence people life psoriasis advertisement, and shielding have wind The website etc. of dangerous information.
It should be noted that Fig. 6 is illustrated so that terminal obtains risk identification result by server as an example, but such as The server is replaced with the risk identification module in terminal, high in the clouds or server by fruit, is applied equally to the present embodiment.
In order to make it easy to understand, a kind of application scenarios in conjunction with this programme are introduced.Such as Fig. 8, it illustrates the application's A kind of example of the Risk Identification Method of small advertisement in application scenarios.In the application scenarios, user can be viscous for curbside The content of the small advertisement of patch carries out risk identification, and report that curbside pastes there are risk or influence the appearance of the city psoriasis it is wide It accuses.
Such as Fig. 8 and Fig. 9, selected by the first width figure in Fig. 8 it is found that being additionally arranged " psoriasis advertisement report " in the application of terminal , user is by selecting " psoriasis advertisement report " in the application can open the report of psoriasis advertisement and the boundary of identification Face prompts user to scan the psoriasis advertisement of curbside by " sweep and sweep ", subsequently to identify that psoriasis is wide in the interface Risk or triggering psoriasis advertisement in announcement etc..
After user clicks " sweep and sweep ", terminal shows an image preview window 801, and in image preview window A cog region 802 is presented in 801, to prompt user that camera is directed at psoriasis advertisement, and indicates that user is wide by psoriasis Risk identification is needed in announcement is partially placed into the cog region, as shown in the second width image in Fig. 8.Correspondingly, user passes through shifting Dynamic terminal, can make the key message in the psoriasis advertisement, the i.e. telephone number " 13603****123 " of telephone number It is put into the cog region 802, as shown in third width image in Fig. 8.Terminal is when detecting the image freeze in the cog region When length is more than preset duration, then the target image in the cog region is extracted from the preview image of the current psoriasis advertisement, To which server will be sent to comprising the target image of " 13603****123 ", the terminal as shown in third width image in Fig. 8 with Interaction between server.
Correspondingly, server carries out the identifications such as OCR to target image, the telephone number in the target image is extracted " 13603****123 ", and by the telephone number matches in the telephone number and risk information library, such as it is matched to identical phone Number then extracts the corresponding risk classifications of telephone number " 13603****123 " and report number, and e.g., risk classifications can be Swindle or intermediary etc., and the corresponding report number of the telephone number and risk classifications are sent to terminal, correspondingly, terminal Middle application interface is become in Fig. 9 from third width image in Fig. 8 in an intermediate width terminal interface.
Further, if the user thinks that the psoriasis advertisement there are harmfulness, then can be with first terminal circle in fig.9 " key report " is clicked in interface shown in face, then the terminal can be by the telephone number " 13603****123 " and the electricity The position (e.g., can be the current position of terminal) of psoriasis advertisement belonging to words number is sent to server, and server can be deposited The report information for storing up terminal transmission provides foundation subsequently to clear up psoriasis advertisement for staff, is also update risk Information bank provides foundation.Return to report successfully after instruction in server, the interface of terminal can from Fig. 9 first terminal circle Face becomes second interface (terminal interface of the rightmost side) of the terminal in Fig. 9 (in intermediate terminal interface).
Another aspect, present invention also provides a kind of risk identification devices of small advertisement, and such as referring to Figure 10, it illustrates this Apply for that a kind of composition schematic diagram of risk identification device one embodiment of small advertisement, the device of the present embodiment can be applied to end End, the device may include:
Instruction acquisition unit 1001, for obtaining the image capture instruction to small advertisement to be identified, with to pending risk The small advertisement of identification carries out Image Acquisition;
Image acquisition unit 1002, the image file of the small advertisement for obtaining the pending risk identification;
Image extraction unit 1003, for extracting the target image in described image file in cog region, the mesh Logo image is a part for described image file;
Risk acquiring unit 1004, for obtaining the risk identification for the target image as a result, the risk identification As a result it is used to characterize in the target image with the presence or absence of the risky character string of tool;
As a result output unit 1005, for exporting the risk identification result.
In the case of a kind of possible, the risk acquiring unit, including:
Image transmits subelement, for sending the target image to risk identification module, to be known by the risk Other module identifies whether the character string for including in the target image belongs to the risky character string of tool, obtains risk identification knot Fruit;
As a result subelement is obtained, the risk identification result identified for obtaining the risk identification module.
Optionally, described image transmission subelement is specifically used for, and sends the target image to risk identification module, To identify the multiple continuation character strings for the specified type for including in the target image by the risk identification module, and identify Whether the multiple continuation character string belongs to the risky character string of tool.
Optionally, which can also include:
Report receiving unit, after exporting the indicating risk result in the result output unit, acquisition is directed to The report for the character string for including in the target image is asked, and the report request is wrapped for asking to report in the target image The character string contained belongs to the risky character string of tool;
Transmission unit is reported, for transmitting the report request to the risk identification module, to know by the risk Character string in the target image is stored as having risky character string by other template.
Optionally, in the embodiment of one device of any of the above, the risk that the risk acquiring unit is got Recognition result is specially the indicating risk for the target image, and the indicating risk is deposited for characterizing in the target image Having risky character string.
In another possible realization method, described image acquiring unit may include:
Image preview subelement, for showing the small of the collected pending risk identification in image preview window The image file of advertisement shows cog region in described image the preview window;
Image obtains subelement, for when meeting image recognition condition, obtaining and currently being opened up in described image the preview window Existing image file.
Optionally, which can also include:
Receiving unit is adjusted, the adjustment of the position of cog region described in image preview window is operated for receiving user;
Response unit is adjusted, for being operated in response to the adjustment, adjusts the cog region in the preview image Show position.
In another possible realization method, described image acquiring unit may include:
Image preview subelement, for showing the small of the collected pending risk identification in image preview window The image file of advertisement;
Identification creates subelement, the identification location point for being specified in described image the preview window based on user, in institute It states and creates cog region in image preview window;
Image obtains subelement, for when meeting image recognition condition, obtaining and currently being opened up in described image the preview window Existing image file.
On the other hand, present invention also provides a kind of terminals, and such as referring to Figure 11, it illustrates one kind of the terminal of the application The terminal 1100 of composed structure schematic diagram, the present embodiment may include:Processor 1101 and memory 1102.
Optionally, which can also include that communication interface 1103, input unit 1104 and display 1105 and communication are total Line 1106.
Processor 1101, communication interface 1103, input unit 1104, display 1105, passes through communication at memory 1102 Bus 1106 completes mutual communication.
In the embodiment of the present application, the processor 1101, can be central processing unit (CentralProcessingUnit, CPU), application-specific integrated circuit, digital signal processor, ready-made programmable gate array or other programmable logic device Deng.
The processor can call the program stored in memory 1102.Specifically, processor can execute following message Operation in the embodiment of sending method performed by application server side.
For storing one or more than one program in memory 1102, program may include program code, the journey Sequence code includes computer-managed instruction, in the embodiment of the present application, is at least stored with for realizing following work(in the memory The program of energy:
The image capture instruction to small advertisement to be identified is obtained, is adopted with carrying out image to the small advertisement of pending risk identification Collection;
Obtain the image file of the small advertisement of the pending risk identification;
The target image in cog region in described image file is extracted, the target image is described image file A part;The risk identification for the target image is obtained as a result, the risk identification result is for characterizing the target figure With the presence or absence of the risky character string of tool as in;
Export the risk identification result.
In one possible implementation, which may include storing program area and storage data field, wherein Storing program area can be needed for storage program area and at least one function (such as image player function etc.) application program Deng;Storage data field can store the data created during the use according to computer, for example, user data and image data Etc..
In addition, memory 1102 may include high-speed random access memory, can also include nonvolatile memory, example Such as at least one disk memory or other volatile solid-state parts.
The communication interface 1103 can be the interface of communication module, such as the interface of gsm module.
The application can also include display 1104 and input unit 1105 etc..
Certainly, the structure of terminal shown in Figure 11 does not constitute the restriction to terminal in the embodiment of the present application, is actually answering May include than more or fewer components shown in Figure 11, or the certain components of combination with middle terminal.
On the other hand, the embodiment of the present application also provides a kind of storage medium, computer is stored in the storage medium Executable instruction when the computer executable instructions are loaded and executed by processor, is realized in as above any one embodiment The Risk Identification Method of small advertisement performed by end side.
It should be noted that each embodiment in this specification is described in a progressive manner, each embodiment weight Point explanation is all difference from other examples, and the same or similar parts between the embodiments can be referred to each other. For device class embodiment, since it is basically similar to the method embodiment, so fairly simple, the related place ginseng of description See the part explanation of embodiment of the method.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning Covering non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only that A little elements, but also include other elements that are not explicitly listed, or further include for this process, method, article or The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged Except there is also other identical elements in the process, method, article or equipment including element.
The foregoing description of the disclosed embodiments enables those skilled in the art to realize or use the present invention.To this A variety of modifications of a little embodiments will be apparent for a person skilled in the art, and the general principles defined herein can Without departing from the spirit or scope of the present invention, to realize in other embodiments.Therefore, the present invention will not be limited It is formed on the embodiments shown herein, and is to fit to consistent with the principles and novel features disclosed in this article widest Range.
It the above is only the preferred embodiment of the present invention, it is noted that those skilled in the art are come It says, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications also should be regarded as Protection scope of the present invention.

Claims (14)

1. a kind of Risk Identification Method of small advertisement, which is characterized in that including:
The image capture instruction to small advertisement to be identified is obtained, to carry out Image Acquisition to the small advertisement of pending risk identification;
Obtain the image file of the small advertisement of the pending risk identification;
The target image in cog region in described image file is extracted, the target image is one of described image file Point;
The risk identification for the target image is obtained as a result, the risk identification result is for characterizing in the target image With the presence or absence of the risky character string of tool;
Export the risk identification result.
2. the Risk Identification Method of small advertisement according to claim 1, which is characterized in that described obtain is directed to the target The risk identification of image is as a result, include:
The target image is sent to risk identification module, to be identified in the target image by the risk identification module Including character string whether belong to the risky character string of tool, obtain risk identification result;
Obtain the risk identification result that the risk identification module identifies.
3. the Risk Identification Method of small advertisement according to claim 1, which is characterized in that described obtain is directed to the target The risk identification of image is as a result, include:
The indicating risk for the target image is obtained, the indicating risk has for characterizing to exist in the target image The character string of risk.
4. the Risk Identification Method of small advertisement according to claim 1, which is characterized in that described to obtain the pending wind The image file of the small advertisement nearly identified, including:
Show the image file of the small advertisement of the collected pending risk identification, described image in image preview window Cog region is shown in the preview window;
When meeting image recognition condition, the described image file currently showed in described image the preview window is obtained.
5. the Risk Identification Method of small advertisement according to claim 4, which is characterized in that further include:
It receives user and operation is adjusted to the position of cog region described in image preview window;
It adjusts and operates in response to the position, adjust the cog region and show position in described image the preview window.
6. the Risk Identification Method of small advertisement according to claim 1, which is characterized in that described to obtain the pending wind The image file of the small advertisement nearly identified, including:
Show the image file of the small advertisement of the collected pending risk identification in image preview window;
Based on the identification location point that user specifies in described image the preview window, identification is created in described image the preview window Area;
When meeting image recognition condition, the described image file currently showed in described image the preview window is obtained.
7. the Risk Identification Method of small advertisement according to claim 2, which is characterized in that described to pass through the risk identification Module identifies whether the character string for including in the target image belongs to the risky character string of tool, including:
The multiple continuation character strings for the specified type for including in the target image are identified by the risk identification module, and are known Whether not the multiple continuation character string belongs to the risky character string of tool.
8. the Risk Identification Method of small advertisement according to claim 2, which is characterized in that carried in the output risk After showing result, further include:
The report request for the character string for including in the target image is obtained, the report request is for asking described in report The character string for including in target image belongs to the risky character string of tool;
The report is transmitted to the risk identification module to ask, with will be in the target image by the risk identification template Character string be stored as having risky character string.
9. a kind of risk identification device of small advertisement, which is characterized in that including:
Instruction acquisition unit, for obtaining the image capture instruction to small advertisement to be identified, with to the small of pending risk identification Advertisement carries out Image Acquisition;
Image acquisition unit, the image file of the small advertisement for obtaining the pending risk identification;
Image extraction unit, for extracting the target image in described image file in cog region, the target image is A part for described image file;
Risk acquiring unit, for obtaining the risk identification for being directed to the target image as a result, the risk identification result is used for It characterizes in the target image with the presence or absence of the risky character string of tool;
As a result output unit, for exporting the risk identification result.
10. the risk identification device of small advertisement according to claim 9, which is characterized in that the risk acquiring unit, packet It includes:
Image transmits subelement, for sending the target image to risk identification module, to pass through the risk identification mould Block identifies whether the character string for including in the target image belongs to the risky character string of tool, obtains risk identification result;
As a result subelement is obtained, the risk identification result identified for obtaining the risk identification module.
11. the risk identification device of small advertisement according to claim 9, which is characterized in that described image acquiring unit, packet It includes:
Image preview subelement, the small advertisement for showing the collected pending risk identification in image preview window Image file, show cog region in described image the preview window;
Image obtains subelement, for when meeting image recognition condition, currently showing in acquisition described image the preview window Described image file.
12. the risk identification device of small advertisement according to claim 10, which is characterized in that further include:
It reports receiving unit, after exporting the indicating risk result in the result output unit, obtains for described The report for the character string for including in target image is asked, and the report request includes for asking to report in the target image Character string belongs to the risky character string of tool;
Transmission unit is reported, for transmitting the report request to the risk identification module, to pass through the risk identification mould Character string in the target image is stored as having risky character string by plate.
13. a kind of terminal, which is characterized in that including:
Processor and memory;
Wherein, the processor is for executing the program stored in the memory;
For storing program, described program is at least used for the memory:
The image capture instruction to small advertisement to be identified is obtained, to carry out Image Acquisition to the small advertisement of pending risk identification;
Obtain the image file of the small advertisement of the pending risk identification;
The target image in cog region in described image file is extracted, the target image is one of described image file Point;
The risk identification for the target image is obtained as a result, the risk identification result is for characterizing in the target image With the presence or absence of the risky character string of tool;
Export the risk identification result.
14. a kind of storage medium, which is characterized in that be stored with computer executable instructions, the calculating in the storage medium When machine executable instruction is loaded and executed by processor, the risk of claim 1 to 8 any one of them small advertisement as above is realized Recognition methods.
CN201810333099.5A 2018-04-13 2018-04-13 Risk identification method, device, terminal and storage medium for small advertisements Active CN108564087B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810333099.5A CN108564087B (en) 2018-04-13 2018-04-13 Risk identification method, device, terminal and storage medium for small advertisements

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810333099.5A CN108564087B (en) 2018-04-13 2018-04-13 Risk identification method, device, terminal and storage medium for small advertisements

Publications (2)

Publication Number Publication Date
CN108564087A true CN108564087A (en) 2018-09-21
CN108564087B CN108564087B (en) 2023-05-12

Family

ID=63535012

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810333099.5A Active CN108564087B (en) 2018-04-13 2018-04-13 Risk identification method, device, terminal and storage medium for small advertisements

Country Status (1)

Country Link
CN (1) CN108564087B (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109447751A (en) * 2018-10-25 2019-03-08 苏州商信宝信息科技有限公司 A kind of exchange side's identity based on big data is counter to cheat recognition methods and its system
CN109842631A (en) * 2019-03-21 2019-06-04 安徽威尔信通信科技有限责任公司 A kind of network information security intelligent analysis system
CN110009370A (en) * 2018-12-27 2019-07-12 阿里巴巴集团控股有限公司 Information processing method, device, electronic equipment and readable storage medium storing program for executing
CN110503460A (en) * 2019-07-26 2019-11-26 秒针信息技术有限公司 Monitoring of the advertisement method, device, system, electronic equipment and storage medium
CN110569743A (en) * 2019-08-19 2019-12-13 广东中凯智慧政务软件有限公司 advertisement information recording method, storage medium and management system
CN110705364A (en) * 2019-09-06 2020-01-17 武汉美格科技股份有限公司 Malicious advertisement eliminating method and system
CN110847104A (en) * 2019-10-25 2020-02-28 深圳市宝政通环境有限公司 Road cleaning vehicle for cleaning small advertisements
CN111368124A (en) * 2020-02-28 2020-07-03 北京百度网讯科技有限公司 Risk identification method, risk identification device and electronic equipment
CN111639333A (en) * 2020-05-28 2020-09-08 维沃移动通信有限公司 Information management method and device and electronic equipment
CN113033588A (en) * 2019-12-24 2021-06-25 中移雄安信息通信科技有限公司 Image processing method and electronic equipment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102982348A (en) * 2012-12-25 2013-03-20 百灵时代传媒集团有限公司 Identification method of advertisement image
CN104156490A (en) * 2014-09-01 2014-11-19 北京奇虎科技有限公司 Method and device for detecting suspicious fishing webpage based on character recognition
CN104573616A (en) * 2013-10-29 2015-04-29 腾讯科技(深圳)有限公司 Information identification method and related device and system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102982348A (en) * 2012-12-25 2013-03-20 百灵时代传媒集团有限公司 Identification method of advertisement image
CN104573616A (en) * 2013-10-29 2015-04-29 腾讯科技(深圳)有限公司 Information identification method and related device and system
CN104156490A (en) * 2014-09-01 2014-11-19 北京奇虎科技有限公司 Method and device for detecting suspicious fishing webpage based on character recognition

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109447751A (en) * 2018-10-25 2019-03-08 苏州商信宝信息科技有限公司 A kind of exchange side's identity based on big data is counter to cheat recognition methods and its system
CN110009370A (en) * 2018-12-27 2019-07-12 阿里巴巴集团控股有限公司 Information processing method, device, electronic equipment and readable storage medium storing program for executing
CN109842631A (en) * 2019-03-21 2019-06-04 安徽威尔信通信科技有限责任公司 A kind of network information security intelligent analysis system
CN110503460A (en) * 2019-07-26 2019-11-26 秒针信息技术有限公司 Monitoring of the advertisement method, device, system, electronic equipment and storage medium
CN110569743A (en) * 2019-08-19 2019-12-13 广东中凯智慧政务软件有限公司 advertisement information recording method, storage medium and management system
CN110705364A (en) * 2019-09-06 2020-01-17 武汉美格科技股份有限公司 Malicious advertisement eliminating method and system
CN110705364B (en) * 2019-09-06 2021-04-30 武汉美格科技股份有限公司 Malicious advertisement eliminating method and system
CN110847104A (en) * 2019-10-25 2020-02-28 深圳市宝政通环境有限公司 Road cleaning vehicle for cleaning small advertisements
CN113033588A (en) * 2019-12-24 2021-06-25 中移雄安信息通信科技有限公司 Image processing method and electronic equipment
CN111368124A (en) * 2020-02-28 2020-07-03 北京百度网讯科技有限公司 Risk identification method, risk identification device and electronic equipment
CN111368124B (en) * 2020-02-28 2023-05-26 北京百度网讯科技有限公司 Risk identification method, risk identification device and electronic equipment
CN111639333A (en) * 2020-05-28 2020-09-08 维沃移动通信有限公司 Information management method and device and electronic equipment

Also Published As

Publication number Publication date
CN108564087B (en) 2023-05-12

Similar Documents

Publication Publication Date Title
CN108564087A (en) Risk Identification Method, device, terminal and the storage medium of small advertisement
CN107093066A (en) Service implementation method and device
US20110264527A1 (en) Apparatuses, Methods and Systems for a Code-Mediated Content Delivery Platform
US20130325567A1 (en) System and method for creating a virtual coupon
US20140122220A1 (en) Apparatuses, methods and systems for measuring, and following up on user preferences among offer types and response options in a marketing campaign
CN109471657A (en) Gray scale dissemination method, device, computer equipment and computer storage medium
US20120205436A1 (en) System for enhanced barcode decoding and image recognition and method therefor
WO2013120064A1 (en) System and method for sending messages to a user in a capture environment
CN107704932A (en) A kind of article reclaimer, device and electronic equipment
CN108090206A (en) Sort method and device, the electronic equipment of comment information
CN104820855A (en) Generation and identification method of dynamic two-dimensional codes based on mobile environment perception technology
TWM467132U (en) Image recognition mobile-commerce system for mobile phone
CN107729791A (en) Information processing method and electronic equipment
WO2011146776A1 (en) Apparatuses,methods and systems for a voice-triggered codemediated augmented reality content delivery platform
CN108805577A (en) Information processing method, device, system, computer equipment and storage medium
CN104537049B (en) A kind of picture browsing method and device
WO2013123482A1 (en) System and method for mobile marketing, advertising, or interaction with a user
WO2013126382A1 (en) System and method for linking media expressions for purchasing a product or other actionable events
CN107580064A (en) 2 D code information sending method, device and storage medium
KR20130107040A (en) System and method for information providing service using contents combined with 2-dimensional code
CN110889057B (en) Service data visualization method and service object visualization device
CN107463853B (en) Method and system for audience tag analysis
CN112262385A (en) Information processing method, information display method, program, terminal, and server
CN112989299A (en) Interactive identity recognition method, system, device and medium
CN109614547A (en) The method and apparatus of preferred watermark for electric terminal

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant