CN102982348A - Identification method of advertisement image - Google Patents
Identification method of advertisement image Download PDFInfo
- Publication number
- CN102982348A CN102982348A CN2012105723807A CN201210572380A CN102982348A CN 102982348 A CN102982348 A CN 102982348A CN 2012105723807 A CN2012105723807 A CN 2012105723807A CN 201210572380 A CN201210572380 A CN 201210572380A CN 102982348 A CN102982348 A CN 102982348A
- Authority
- CN
- China
- Prior art keywords
- image
- characteristic information
- advertising
- identified
- feature extraction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an identification method of an advertisement image, which comprises the steps that an image to be identified which is uploaded from a mobile terminal is obtained; the image to be identified contains a part of the advertisement image or the whole advertisement image; the image to be identified is compared with feature information in a preset feature information base; the advertisement image corresponding to the image to be identified is obtained; advertisement information corresponding to the advertisement image is obtained; and the advertisement information is fed back to the mobile terminal. The invention further provides a server. According to the identification method, a method of obtaining the advertisement information by adding an additional two-dimensional code or a bar code on the advertisement image is avoided, the whole advertisement image is within an available shooting range of the user, a method that the user have to shoot a specific position of the two-dimensional code or the bar code is avoided, the difficulty for the user to obtain the extended advertisement information is lowered, the advertisement information carried by the advertisement image is obtained by the user quickly and conveniently, and a user experience is further optimized.
Description
Technical field
The present invention relates to communication and areas of information technology, be specifically related to a kind of recognition methods of advertising image.
Background technology
Continuous progress along with technology, the mankind have entered the fast information age, in order to enhance competitiveness, many businessmans are chosen in the more positions of flow of the people such as market, the station board image that plasters posters, represent by the advertising message that will want the to publicize form with advertising image, to reach the purpose of advertising.
Advertising image is in order to win user's favor, design comparatively attractive in appearance more, the cost of thereupon bringing be the quantity of information that comprises seldom, the extra increase on advertising image of some advertising images comprises two-dimension code or the bar code of extend information for this reason, make the image user take or scan two-dimension code or the bar code that is positioned at ad-hoc location by portable terminal, to obtain the extend information beyond advertising image.
Because the restriction of the length of advertising image own, the extra two-dimension code that increases or bar code will affect this advertising image as being arranged on apparent position whole structure, and for guaranteeing that advertising image is attractive in appearance, with the two-dimension code that additionally increases or bar code printing on the more inclined to one side ad-hoc location of advertising image, image but because of two-dimension code or the shape of bar code own less, attractive force is not strong, the two-dimension code or the bar code that cause the user to be difficult to view being positioned on the ad-hoc location more inclined to one side on the advertising image, and be difficult to this two-dimension code or bar code are scanned, thereby increase the difficulty that the user obtains the advertising message of expansion.In addition, with regard to advertising image, the extra two-dimension code that increases or bar code also can cause its cost to increase.
Hence one can see that, adopt in the prior art the two-dimension code or the bar code that additionally are added on the advertising image to obtain advertising message, not only increased the cost of advertising image, and increased the difficulty that the user obtains the advertising message of expansion, cause the user can't obtain quickly and easily the advertising message that this advertising image carries, affect the user and experience.
Summary of the invention
The invention discloses a kind of recognition methods of advertising image, utilize this method can solve the user and can't obtain quickly and easily the advertising message that this advertising image carries, affect the problem that the user experiences.
A kind of recognition methods of advertising image comprises:
That obtains that portable terminal uploads is identified image, described part or all that contains advertising image in the image that be identified;
Characteristic information in the described characteristic information storehouse that is identified image and presets is compared, and determine to have and described advertising image corresponding to the identical characteristic information of image that be identified;
Obtain and the corresponding advertising message of described advertising image;
Described advertising message is fed back to described handover terminal.
Preferably, described the described image that is identified is compared with characteristic information in the default characteristic information storehouse, comprising:
The described image that is identified itself is compared with characteristic information in the described default characteristic information storehouse.
Preferably, described the described image that is identified is compared with characteristic information in the default characteristic information storehouse, comprising:
The described image that is identified is carried out feature extraction, and obtain the described Characteristic of Image information that is identified;
Characteristic information in the described characteristic information storehouse that is identified Characteristic of Image information and presets is compared.
Preferably, described the described image that is identified is carried out feature extraction, comprising: will be identified image and carry out Gaussian Blur and process;
Use the method for default feature extraction to carry out feature extraction to being identified image by default feature.
Preferably, the constructive process in described default characteristic information storehouse comprises:
Each the described advertising image and the corresponding advertising message of each described advertising image that provide in advance are provided;
Each described advertising image and each described advertising message are stored respectively;
Each advertising image is carried out respectively Gaussian Blur to be processed;
Use the method for default feature extraction that each described advertising image is carried out feature extraction by default feature, and obtain the characteristic information of each described advertising image;
The characteristic information of each described advertising image is carried out respectively packet memory;
With the characteristic information of packet memory respectively each described advertising message corresponding with each described advertising image set up and link, consist of each subcharacter information bank;
Adopt at least one sub-characteristic information storehouse to create described default characteristic information storehouse.
Preferably, it is characterized in that,
Described default feature comprises: color characteristic, textural characteristics, shape facility or space characteristics.
The method of described default feature extraction comprises: Fourier converter technique, window Fourier converter technique, Wavelet Transform, boundary direction histogram method, based on Tamura texture feature extraction method.
A kind of server comprises:
The first acquiring unit is used for obtaining that portable terminal uploads is identified image, described part or all that contains advertising image in the image that be identified;
The contrast unit is used for the characteristic information in the described characteristic information storehouse that is identified image and presets is compared, and determines the described affiliated advertising image with identical characteristic information of image that is identified;
Second acquisition unit obtains and the corresponding advertising message of described advertising image;
Feedback unit is used for described advertising message is fed back to described handover terminal.
Preferably, described contrast unit comprises:
The first contrast unit, the characteristic information that is used for describedly being identified image itself and described default characteristic information storehouse compares; Or
The second contrast unit is used for the described image that is identified is carried out feature extraction, and obtains the described Characteristic of Image information that is identified, and the characteristic information in the described characteristic information storehouse that is identified Characteristic of Image information and presets is compared.
Preferably, described the second contrast unit comprises:
Processing unit is used for being identified image and carries out the Gaussian Blur processing;
Feature extraction unit is used for using the method for default feature extraction to carry out feature extraction to being identified image by default feature.
Preferably, it is characterized in that, also comprise:
Acquiring unit is used for obtaining each described advertising image and the corresponding advertising message of each described advertising image that provides in advance;
The first storage unit is used for each described advertising image and each described advertising message are stored respectively;
The second processing unit is used for that each advertising image is carried out respectively Gaussian Blur and processes;
The Second Characteristic extraction unit is used for using the method for default feature extraction that each described advertising image is carried out feature extraction by default feature, and obtains the characteristic information of each described advertising image;
The second storage unit is used for the characteristic information of each described advertising image is carried out respectively packet memory;
Set up link unit, be used for the characteristic information of packet memory respectively each described advertising message corresponding with each described advertising image set up and link, consist of each subcharacter information bank;
Creating unit is used for adopting at least one sub-characteristic information storehouse to create described default characteristic information storehouse.
The present invention adopts above technological means to obtain beneficial effect:
The invention discloses a kind of recognition methods of advertising image, based on default characteristic information storehouse, contain all characteristic informations of advertising image in the default characteristic information storehouse, can obtain advertising image by the characteristic information in this characteristic information storehouse; In this method the user by portable terminal in the advertising image any part or all take and obtain being identified image, and will be identified image uploading to server, after server obtains and is identified image, characteristic information in the characteristic information storehouse that is identified image and presets is compared, because containing all characteristic informations of advertising image in the default characteristic information storehouse, therefore can obtain and be identified the corresponding advertising image of image by being identified image, and then obtain and the corresponding advertising message of advertising message, and advertising message fed back to portable terminal, so that the user obtains advertising message.The method that the present invention has avoided adopting the two-dimension code that additionally is added on the advertising image or bar code to obtain advertising message, eliminated on the one hand the expense of independent making two-dimension code or bar code, cost, the view picture advertising image all is the coverage that the user can use on the other hand, avoid the user must take the method for the ad-hoc location at two-dimension code or bar code place, therefore reduced the difficulty that the user obtains the advertising message of expansion, so that the user obtains the advertising message that this advertising image carries quickly and easily, further so that affect user's experience more excellent.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, the below will do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art, apparently, accompanying drawing in the following describes only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the process flow diagram of the recognition methods of the disclosed advertising image of the embodiment of the invention;
Fig. 2 is the process flow diagram of characteristic information storehouse constructive process default in the recognition methods of the disclosed advertising image of the embodiment of the invention;
Fig. 3 is the process flow diagram of the recognition methods of the disclosed another advertising image of the embodiment of the invention;
Fig. 4 is the picture view of advertising image of the recognition methods of the disclosed another advertising image of the embodiment of the invention;
Fig. 5 is the process flow diagram of the recognition methods of the disclosed another advertising image of the embodiment of the invention;
Fig. 6 is the structural representation of the disclosed server of the embodiment of the invention;
Fig. 7 is the structural representation of contrast unit in the disclosed server of the embodiment of the invention;
Fig. 8 is the structural representation of second unit in the disclosed server of the embodiment of the invention;
Fig. 9 is the structural representation of creating unit in the disclosed server of the embodiment of the invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that obtains under the creative work prerequisite.
Now along with the continuous progress of technology, the portable terminals such as smart mobile phone also obtain develop rapidly, transferring integrated increasing application function on the terminal, for example:
(1) takes pictures/record a video
Portable terminal has been equipped with camera and camera function, and the user can utilize camera function, takes pictures anywhere or anytime and stores, for the generation of image provides the entrance source.
(2) local storage
Portable terminal has been equipped with and can have inserted or extendible memory carrier, comprises storage card commonly used, and the capacity of these memory carriers is very large, and storing for this locality of image provides memory carrier
(3) wireless network
Portable terminal has been equipped with and can have used the wifi function, can be by the wifi wireless network, and stable access to LAN or internet at a high speed is for portable terminal and extraneous communication provide easily passage.Portable terminal has been equipped with 3G card or GMS card, and portable terminal utilizes 3G card or GMS cartoon to cross the carrier network accessing Internet, and the user can open wireless network whenever and wherever possible.
The user can be the content of the picture of taking or local storage, is delivered to any end of internet by wireless network, is the mobile interconnected condition that provides.
As shown in Figure 1, the process flow diagram for the first embodiment of the recognition methods of advertising image of the present invention comprises:
Step S101: obtain and be identified image, that obtains namely that portable terminal uploads is identified image, described part or all that contains advertising image in the image that be identified;
The user takes advertising image by the portable terminal shoot function, the picture that the user obtains by shooting is called and is identified image, be identified image two kinds of situations are arranged: the first situation, be identified and only contain advertising image in the image, in the first situation, being identified image is in the advertising image part or all, namely is identified image and is any part or view picture advertising image in the advertising image; The second situation, be identified and contain outside scenery and advertising image in the image, in the second situation, it is consistent with the first situation to be identified the advertising image that comprises in the image, namely can be in the advertising image part or all, no matter outside scenery and the advertising image ratio in being identified image, as long as be identified the characteristic information that contains in the image in the advertising image, namely can be used as and be identified image, characteristic information is the information of a series of representative advertising image contents such as the expression advertising image brightness variation that can identify of computing machine, dimensional variation, change color.
Portable terminal is stored in storage card or the inner memory carrier by the image that is identified that shooting is obtained, and will be identified image uploading to server by wireless network.
Step S102: compare and obtain the result with the characteristic image storehouse, the characteristic information that is about in the described characteristic information storehouse that is identified image and presets compares, and determines to have and described advertising image corresponding to the identical characteristic information of image that be identified;
Up to now, characteristic information does not have omnipotent and precise definition, feature is the part of " interesting " in the digital picture, and the most important characteristic of characteristic information is " repeatability ": the characteristic information that the different images of Same Scene extracts should be identical.The characteristic information storehouse obtains for by server advertising image being carried out feature extraction, and namely the characteristic information storehouse is the information bank that all characteristic informations by advertising image form, and has wherein comprised the various characteristic informations of advertising image.
As shown in Figure 2, the constructive process for default characteristic information storehouse comprises:
Step S201: obtain advertising image and advertising message, each the described advertising image and the corresponding advertising message of each described advertising image that provide in advance namely are provided;
Businessman will want in advance to carry out the advertising image of extraneous information expansion and the advertising message of wanting to expand provides to advertising company, advertising company uploads onto the server advertising image and the advertising message that businessman provides, wherein the quantity of advertising image is not limit, the content of advertising message can be the various information such as product information, company's site, performer's content, story introduction, server obtains advertising image and advertising message, and advertising image and advertising message are further processed.
Step S202: advertising image and advertising message are stored, namely each described advertising image and each described advertising message are stored respectively;
After server obtains each advertising image and corresponding each advertising message with advertising image that different businessmans provide, respectively each advertising image and each advertising message are stored respectively, in the process of storage, described each advertising image set up with the corresponding described advertising message of image separately respectively link, in order to obtain the corresponding advertising message of described advertising image according to advertising image.
Step S203: described advertising image is carried out feature extraction;
Feature extraction is an elementary computing in the image processing, is first calculation process that an image is carried out, and server checks that each pixel determines whether this pixel represents a feature.
Advertising image is carried out feature extraction to be comprised: each advertising image is carried out respectively Gaussian Blur process; Input picture is generally smoothed in metric space by Gaussian Blur nuclear, then uses the method for default feature extraction that each described advertising image is carried out feature extraction by default feature, and obtains the characteristic information of each described advertising image; Described default feature comprises: color characteristic, textural characteristics, shape facility or space characteristics, the method for described default feature extraction comprises: Fourier converter technique, window Fourier converter technique, Wavelet Transform, boundary direction histogram method, based on Tamura texture feature extraction method.
Characteristics of image commonly used has color characteristic, textural characteristics, shape facility, spatial relationship feature.When equation of light image, what usually see is the continuous texture zone similar to gray level, their formation object that combines.If but the small-sized or contrast of object is not high, usually to adopt higher resolution to observe: if the size of object is very large or contrast is very strong, only to need to reduce resolution.If dimension of object varies, or contrast has to have by force in the weak situation and exists simultaneously, at this moment extract Characteristic of Image and have superiority to carrying out image studies, feature extracting method commonly used has: Fourier converter technique, window Fourier conversion (Gabor), Wavelet Transform, least square method, boundary direction histogram method, based on the texture feature extraction of Tamura textural characteristics etc.
The present invention carries out feature extraction by Feature Extraction Technology to advertising image, and is concrete, the advertising image Feature Extraction Technology can be carried out refinement and cut apart, and obtains all characteristic informations of advertising image.For example: each advertising image has its feature, such as alphabetical A a point is arranged, P has a circle, and there is individual acute angle etc. at the center of Y, the Messages amount of image outline curvature maximum or contour direction flip-flop is maximum, the color that also has in addition picture, the many information of shading value etc., feature extraction namely is unique points all in the picture in its entirety to be extracted obtain characteristic information, because the element of feature extraction is very many and very concrete, therefore the characteristic information that obtains is except the image that fritter is arranged, even also have specific to a slight pixel, so the degree of accuracy of feature extraction is very high.
Step S204: characteristic information is carried out packet memory, and the characteristic information that is about to each described advertising image carries out respectively packet memory;
Server obtains the characteristic information of each advertising image by feature extraction, and the different characteristic information of the advertising image that these are different carries out respectively packet memory, namely is stored in the different storage unit.
Step S205: make up the subcharacter information bank, with the characteristic information of packet memory respectively each described advertising message corresponding with each described advertising image set up and link, consist of each subcharacter information bank;
The characteristic information of each storage unit that the server by utilizing above-mentioned steps obtains, respectively the storage information of different storage unit is set up with each advertising image and link, and the storage unit after each storage unit and the foundation of advertisement advertising image linked becomes the subcharacter information bank.
Step S206: create the characteristic information storehouse, adopt at least one sub-characteristic information storehouse to create described default characteristic information storehouse.
With one, the set of a plurality of or all subcharacter information bank in the above-mentioned steps, create default characteristic information storehouse, characteristic information and the advertising image of of namely containing in the characteristic information storehouse, a plurality of or all storage unit.
Server can find the advertising image corresponding with this characteristic information by the characteristic information in this default characteristic information storehouse, can obtain the advertising message that is linked with this advertising image by this advertising image.
In step S102, the server mobile terminal receive is uploaded is identified image, and will be identified image and compare with characteristic information in the default characteristic information storehouse, if comprise when being identified the advertising image that comprises in the image in the characteristic information storehouse, server can obtain and is identified the corresponding advertising image of image by contrast; If do not comprise in the characteristic information storehouse when being identified the advertising image that comprises in the image, server informs that this picture of terminal is not included in the server, and follow-up process does not then belong to invention which is intended to be protected, is not giving unnecessary details.
Step S103: obtain and the corresponding advertising message of described advertising image;
In default characteristic information storehouse, advertising image and advertising message have linking relationship, obtain being identified the corresponding advertising image of image according to above-mentioned steps, can obtain the corresponding advertising message of described advertising image by advertising image.
Step S104: described advertising message is fed back to described handover terminal.
In server, default characteristic information storehouse is stored in respectively different storage unit with advertising message, set up link between the two, utilize advertising image can obtain the corresponding advertising message of this advertising image by link, server obtains this advertising message, by wireless network advertising message is issued to portable terminal, portable terminal shows this advertising message, so that the user can see this information.
The invention discloses a kind of recognition methods of advertising image, based on default characteristic information storehouse, contain all characteristic informations of advertising image in the default characteristic information storehouse, can obtain advertising image by the characteristic information in this characteristic information storehouse.In this method the user by portable terminal in the advertising image any part or all take and obtain being identified image, and will be identified image uploading to server, after server obtains and is identified image, characteristic information in the characteristic information storehouse that is identified image and presets is compared, because containing all characteristic informations of advertising image in the default characteristic information storehouse, therefore can obtain and be identified the corresponding advertising image of image by being identified image, and then obtain and the corresponding advertising message of advertising message, and advertising message fed back to portable terminal, so that the user obtains advertising message.
The method that the present invention has avoided adopting the two-dimension code that additionally is added on the advertising image or bar code to obtain advertising message, eliminated on the one hand the expense of independent making two-dimension code or bar code, cost, the view picture advertising image all is the coverage that the user can use on the other hand, avoid the user must take the method for the ad-hoc location at two-dimension code or bar code place, therefore reduced the difficulty that the user obtains the advertising message of expansion, so that the user obtains the advertising message that this advertising image carries quickly and easily, further so that affect user's experience more excellent.
As shown in Figure 3, the process flow diagram for the second embodiment of the recognition methods of advertising image of the present invention comprises:
Step S301: obtain and be identified image, that obtains namely that portable terminal uploads is identified image, described part or all that contains advertising image in the image that be identified;
Step S302: determine to be identified the corresponding advertising image of image, the characteristic information that is about in the described characteristic information storehouse that is identified image and presets compares, and determines to have and described advertising image corresponding to the identical characteristic information of image that be identified;
Characteristic information in the characteristic information storehouse that is identified image and presets is compared, two kinds of situations is arranged:
The first situation is:
Feature in the characteristic information storehouse that directly will be identified image itself and preset compares, obtain and be identified the corresponding advertising image of image by contrast, but may contain outside scenery because being identified in the image, or a plurality of characteristic informations, therefore the degree of accuracy of the method is not high, in order to improve precision, can also carry out feature extraction, namely the second situation with being identified image;
The second situation: will be identified image and carry out feature extraction, carry out with being identified image that characteristic information compares in the characteristic information that obtains after the feature extraction and the default characteristic information storehouse, the described Characteristic of Image information that is identified is compared with characteristic information in the default characteristic information storehouse.
Server obtains being identified Characteristic of Image information by feature extraction, characteristic information is stored respectively, and be stored in the different storage unit.
Feature extraction is an elementary computing in the image processing, that is to say that it is first calculation process that an image is carried out.It checks that each pixel determines whether this pixel represents a feature.
Advertising image is carried out feature extraction to be comprised: each advertising image is carried out respectively Gaussian Blur process; Input picture is generally smoothed in metric space by Gaussian Blur nuclear, then uses the method for default feature extraction that each described advertising image is carried out feature extraction by default feature, and obtains the characteristic information of each described advertising image; Described default feature comprises: color characteristic, textural characteristics, shape facility or space characteristics, the method for described default feature extraction comprises: Fourier converter technique, window Fourier converter technique, Wavelet Transform, boundary direction histogram method, based on Tamura texture feature extraction method.
Carry out obtaining described all characteristic informations that comprise in the image that are identified after the feature extraction, and compare with characteristic information in the default characteristic information storehouse being identified Characteristic of Image information after this feature extraction, determine to have and described advertising image corresponding to the identical characteristic information of image that be identified;
In addition, because the characteristic information in the characteristic information storehouse is very many, in order to make server to searching more fast characteristic information, therefore when creating the characteristic information storehouse, characteristic information after the feature extraction can be set up index according to type, and according to index with the storage of classifying of the characteristic information of each advertising image, and each characteristic information set up with the advertising image of self link, in order to find according to index and be identified the corresponding characteristic information of image according to server, obtain being identified the corresponding advertising image of image according to this characteristic information, obtain being identified the required advertising message that obtains of image according to this advertising image, characteristic information in the characteristic information storehouse is classified according to type and set up index, and described type comprises color, gray scale, brightness etc.
Step S303: obtain and the corresponding advertising message of described advertising image;
Step S304: described advertising message is fed back to described handover terminal.
The present invention is by carrying out feature extraction to this recognition image, can be with the information of feature extraction more specifically to pixel, be more convenient for default characteristic information storehouse in the contrast of characteristic information, user of the present invention takes arbitrarily advertising image and obtains to be identified image, by being identified image uploading to network, can realize obtaining of extend information.The method of having avoided adopting the two-dimension code that additionally is added on the advertising image or bar code to obtain advertising message, eliminated on the one hand the expense of independent making two-dimension code or bar code, cost, the view picture advertising image all is the coverage that the user can use on the other hand, avoid the user must take the method for the ad-hoc location at two-dimension code or bar code place, therefore reduced the difficulty that the user obtains the advertising message of expansion, so that the user obtains the advertising message that this advertising image carries quickly and easily, further so that affect user's experience more excellent.
The below will describe the scene embodiment of the recognition methods of advertising image of the present invention in detail.
The information that advertising pictures of the present invention and advertising message provide for businessman, if businessman wants certain things, personage or scenery are advertised, the information of the expansion that then can carry out the image of advertising things, personage or scenery with wanting and want to publicize provides to advertising company, and the value-added service of purchase advertising company, so that the user can obtain the advertising message of extra expansion.Wherein things, personage or scenery comprise: things or the scenery of the every profession and trades such as cosmetics of household, the film of newly showing, performer, new listing, concrete situation can be as the criterion according to the information that businessman provides, and does not do restriction at this.
Advertising company obtains advertising pictures and the advertising message that each businessman provides, if businessman wishes that advertising pictures is more eye-catching more noticeable, the advertising pictures that then this businessman can be provided is made separately and is obtained advertising image, the advertising pictures that this businessman is only arranged on this advertising image, the user can demonstrate the corresponding advertising message of this advertising image by taking the optional position of this image on user's portable terminal.But this kind mode only can make the user obtain single information, and the user can not obtain the various information of other industry.
The present invention provides a kind of advertising image be used to being identified on the basis based on the recognition methods of advertising image, in order to make a width of cloth advertising image can express more advertising message, satisfy user's obtaining various advertising messages, simultaneously in order to reduce the cost of businessman, advertising company adopts spatial division advertising image technology, advertising image is divided into a plurality of little area of space, the different advertising image that different area of space provides for each businessman can be passed on different advertising messages by the zones of different of advertising image.
As shown in Figure 4, photo synoptic diagram for a kind of novel advertisement image, the various information that provide comprising businessman, such as: the every field such as amusement, music, novel, automobile, the user can be by taking the different spaces zone, obtain the extend information of zones of different, enriched the content of the required extend information of user.
The following examples are intended for a leading vehicle take a benz 4S shop and do value-added service as example, as shown in Figure 5, comprising:
Step S501: set up in advance and contain benz car Characteristic of Image information bank.
Advertising image and the advertising message of benz car at first are provided by the benz shop, benz car advertising image belongs to automotive field, therefore the advertising image of benz car can be positioned in the area of space of automobile, or directly the advertising image that provides of benz shop as the image that is identified of the area of space of automobile, as long as the information of expansion is provided for the benz shop by the extend information that provides.The shop of running quickly in addition also need provide wants the information expanded, shape and color that this car is for example generally only arranged on advertising image, the concrete information that does not have other, so the information that businessman provides can be parameters and the performance of this car, the video ads of this car, selling price, and the various information such as official website of Benz Co..
By server benz car advertising image is operated by step S201-S206, and when the construction feature information bank, adding is classified according to type to the characteristic information in the characteristic information storehouse and is set up index, in order to make server to searching more fast characteristic information, described type comprises color or gray scale etc., obtain thus benz car Characteristic of Image the characteristics of information information bank, so that after the user takes the image of benz car, in server, can search the characteristic information of benz car.
Step S502: the image of the car that will run quickly is positioned over the advertising position;
The advertising image that advertising company provides the benz shop is positioned over the more remarkable position of user's flow of the people, the place that flows of the people such as market or public transport stop board is more, so that the user takes advertising image, the advertising image of car of maybe will running quickly is positioned over the area of space that is positioned at automotive field, so that the user except obtaining automobile information, also can obtain other information.
Step S503: the user takes by portable terminal and obtains being identified image;
The user takes the advertising image of benz car by portable terminal, the user is by the shoot function of portable terminal, can take the optional position of benz car image, angle or direction difference because of the user, so what obtain is identified the optional position that image is advertising image, can be part or all of advertising image, and the image that is identified after will taking is stored in the mobile inner memory carrier, so that the user uses this to be identified image anywhere or anytime, carry out obtaining of extra advertising message by the advertising image recognition technology, so that the user obtains advertising message easily.
Step S504: the user will be identified image uploading to server by portable terminal;
The user comprises by wireless network: 3G network, wifi network etc. will be identified the server of image uploading to advertising company's appointment, utilize server that advertising image is processed.
Step S505: the described image that is identified is carried out feature extraction;
The present invention carries out feature extraction by Feature Extraction Technology to being identified image, concrete can carry out refinement and cut apart being identified image feature extraction techniques, all characteristic informations that obtain being identified because of the process of above embodiment with the detailed description feature extraction, do not repeat them here.
Step S506: obtain described be identified image in all characteristic informations of comprising;
Server obtains being identified Characteristic of Image information by feature extraction, characteristic information is stored respectively, and be stored in the different storage unit.
Step S507: create the characteristic image storehouse, namely create and be identified the Characteristic of Image image library, described characteristic image storehouse is contained part or all and is identified the described characteristic information that comprises in the image;
Whole or a part of characteristic informations that are identified the described characteristic information that comprises in the image that server obtains deposit into the row storage with characteristic information, and are stored in the storage unit, will store the characteristic information storehouse of the feature unit of characteristic information as advertising image.
In addition because the characteristic information in the characteristic information storehouse is very many, in order to make server to searching more fast characteristic information, therefore characteristic information is set up index, the characteristic information in the characteristic information storehouse is classified according to type and set up index, described type comprises color or gray scale etc.
Step S508: compare and obtain the result with the characteristic image storehouse, be about to described characteristic information in the described characteristic image storehouse and the described characteristic information in the described default characteristic information storehouse and compare, and obtain and the described corresponding advertising image of image that is identified;
Server compares the characteristic information in the described characteristic image storehouse and characteristic information in the default characteristic information storehouse, when containing consistent characteristic information, server is namely found character pair information, and further server is linked to the advertising pictures corresponding with this characteristic information by this characteristic information.
Step S509: obtain and the corresponding advertising message of described advertising image;
Step S510: described advertising message is fed back to described handover terminal.
The advertising message of acquisition for mobile terminal benz car, and advertising message shows, so that the user obtains advertising message timely.
Embodiment discloses a kind of recognition methods of advertising image by concrete scene embodiment, based on default characteristic information storehouse, contain all characteristic informations of advertising image in the default characteristic information storehouse, can obtain advertising image by the characteristic information in this characteristic information storehouse.In this method the user by portable terminal in the advertising image any part or all take and obtain being identified image, and will be identified image uploading to server, after server obtains and is identified image, characteristic information in the characteristic information storehouse that is identified image and presets is compared, because containing all characteristic informations of advertising image in the default characteristic information storehouse, therefore can obtain and be identified the corresponding advertising image of image by being identified image, and then obtain and the corresponding advertising message of advertising message, and advertising message fed back to portable terminal, so that the user obtains advertising message.
The method that the present invention has avoided adopting the two-dimension code that additionally is added on the advertising image or bar code to obtain advertising message, eliminated on the one hand the expense of independent making two-dimension code or bar code, cost, the view picture advertising image all is the coverage that the user can use on the other hand, avoid the user must take the method for the ad-hoc location at two-dimension code or bar code place, therefore reduced the difficulty that the user obtains the advertising message of expansion, so that the user obtains the advertising message that this advertising image carries quickly and easily, further so that affect user's experience more excellent.
As shown in Figure 6, the present invention also comprises a kind of server, comprising:
A kind of server comprises:
The first acquiring unit 600 is used for obtaining that portable terminal uploads is identified image, described part or all that contains advertising image in the image that be identified;
Further, as shown in Figure 7, described contrast unit 700 comprises:
The first contrast unit 701, the characteristic information that is used for describedly being identified image itself and described default characteristic information storehouse compares;
Perhaps
The second contrast unit 702 is used for the described image that is identified is carried out feature extraction, and obtains the described Characteristic of Image information that is identified, and the characteristic information in the described characteristic information storehouse that is identified Characteristic of Image information and presets is compared.
Further, as shown in Figure 8, described the second contrast unit 702 comprises:
As shown in Figure 9, described default characteristic information storehouse adopts creating unit 1000 to create, and described creating unit comprises:
The 3rd acquiring unit 1001 is used for obtaining each described advertising image and the corresponding advertising message of each described advertising image that provides in advance;
The first storage unit 1002 is used for each described advertising image and each described advertising message are stored respectively;
The second processing unit 1003 is used for that each advertising image is carried out respectively Gaussian Blur and processes;
Second Characteristic extraction unit 1004 is used for using the method for default feature extraction that each described advertising image is carried out feature extraction by default feature, and obtains the characteristic information of each described advertising image;
The second storage unit 1005 is used for the characteristic information of each described advertising image is carried out respectively packet memory;
Set up link unit 1006, be used for the characteristic information of packet memory respectively each described advertising message corresponding with each described advertising image set up and link, consist of each subcharacter information bank;
The second creating unit 1007 is used for adopting at least one sub-characteristic information storehouse to create described default characteristic information storehouse.
Above-mentioned creating unit also comprises:
Set up indexing units, be used for before described characteristic information storehouse and described advertising image foundation are linked, the characteristic information in the described characteristic information storehouse is classified according to type and set up index, described type comprises color or gray scale.
The present invention also provides a kind of system, comprising: above-mentioned server and described portable terminal, and both connect by wireless network;
If the described function of present embodiment method realizes with the form of SFU software functional unit and during as independently production marketing or use, can be stored in the computing equipment read/write memory medium.Based on such understanding, the part that the embodiment of the invention contributes to prior art or the part of this technical scheme can embody with the form of software product, this software product is stored in the storage medium, comprise that some instructions are with so that a computing equipment (can be personal computer, server, mobile computing device or the network equipment etc.) carry out all or part of step of the described method of each embodiment of the present invention.And aforesaid storage medium comprises: the various media that can be program code stored such as USB flash disk, portable hard drive, ROM (read-only memory) (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disc or CD.
Each embodiment adopts the mode of going forward one by one to describe in this instructions, and what each embodiment stressed is and the difference of other embodiment that same or similar part is mutually referring to getting final product between each embodiment.
To the above-mentioned explanation of the disclosed embodiments, make this area professional and technical personnel can realize or use the present invention.Multiple modification to these embodiment will be apparent concerning those skilled in the art, and General Principle as defined herein can in the situation that does not break away from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention will can not be restricted to these embodiment shown in this article, but will meet the widest scope consistent with principle disclosed herein and features of novelty.
Claims (10)
1. the recognition methods of an advertising image is characterized in that, comprising:
That obtains that portable terminal uploads is identified image, described part or all that contains advertising image in the image that be identified;
Characteristic information in the described characteristic information storehouse that is identified image and presets is compared, and determine to have and described advertising image corresponding to the identical characteristic information of image that be identified;
Obtain and the corresponding advertising message of described advertising image;
Described advertising message is fed back to described handover terminal.
2. the method for claim 1 is characterized in that, described the described image that is identified is compared with characteristic information in the default characteristic information storehouse, comprising:
The described image that is identified itself is compared with characteristic information in the described default characteristic information storehouse.
3. the method for claim 1 is characterized in that, described the described image that is identified is compared with characteristic information in the default characteristic information storehouse, comprising:
The described image that is identified is carried out feature extraction, and obtain the described Characteristic of Image information that is identified;
Characteristic information in the described characteristic information storehouse that is identified Characteristic of Image information and presets is compared.
4. method as claimed in claim 3 is characterized in that, described the described image that is identified is carried out feature extraction, comprising:
To be identified image and carry out the Gaussian Blur processing;
Use the method for default feature extraction to carry out feature extraction to being identified image by default feature.
5. such as the described method of any one in the claim 1, it is characterized in that the constructive process in described default characteristic information storehouse comprises:
Each the described advertising image and the corresponding advertising message of each described advertising image that provide in advance are provided;
Each described advertising image and each described advertising message are stored respectively;
Each advertising image is carried out respectively Gaussian Blur to be processed;
Use the method for default feature extraction that each described advertising image is carried out feature extraction by default feature, and obtain the characteristic information of each described advertising image;
The characteristic information of each described advertising image is carried out respectively packet memory;
With the characteristic information of packet memory respectively each described advertising message corresponding with each described advertising image set up and link, consist of each subcharacter information bank;
Adopt at least one sub-characteristic information storehouse to create described default characteristic information storehouse.
6. such as claim 4 or 5 described methods, it is characterized in that,
Described default feature comprises: color characteristic, textural characteristics, shape facility or space characteristics;
The method of described default feature extraction comprises: Fourier converter technique, window Fourier converter technique, Wavelet Transform, boundary direction histogram method, based on Tamura texture feature extraction method.
7. a server is characterized in that, comprising:
The first acquiring unit is used for obtaining that portable terminal uploads is identified image, described part or all that contains advertising image in the image that be identified;
The contrast unit is used for the characteristic information in the described characteristic information storehouse that is identified image and presets is compared, and determines the described affiliated advertising image with identical characteristic information of image that is identified;
Second acquisition unit obtains and the corresponding advertising message of described advertising image;
Feedback unit is used for described advertising message is fed back to described handover terminal.
8. server as claimed in claim 7 is characterized in that, described contrast unit comprises:
The first contrast unit, the characteristic information that is used for describedly being identified image itself and described default characteristic information storehouse compares; Or
The second contrast unit is used for the described image that is identified is carried out feature extraction, and obtains the described Characteristic of Image information that is identified, and the characteristic information in the described characteristic information storehouse that is identified Characteristic of Image information and presets is compared.
9. server as claimed in claim 8 is characterized in that, described the second contrast unit comprises:
Processing unit is used for being identified image and carries out the Gaussian Blur processing;
Feature extraction unit is used for using the method for default feature extraction to carry out feature extraction to being identified image by default feature.
10. such as claim 7 or 8 described servers, it is characterized in that, also comprise:
The 3rd acquiring unit is used for obtaining each described advertising image and the corresponding advertising message of each described advertising image that provides in advance;
The first storage unit is used for each described advertising image and each described advertising message are stored respectively;
The second processing unit is used for that each advertising image is carried out respectively Gaussian Blur and processes;
The Second Characteristic extraction unit is used for using the method for default feature extraction that each described advertising image is carried out feature extraction by default feature, and obtains the characteristic information of each described advertising image;
The second storage unit is used for the characteristic information of each described advertising image is carried out respectively packet memory;
Set up link unit, be used for the characteristic information of packet memory respectively each described advertising message corresponding with each described advertising image set up and link, consist of each subcharacter information bank;
Creating unit is used for adopting at least one sub-characteristic information storehouse to create described default characteristic information storehouse.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012105723807A CN102982348A (en) | 2012-12-25 | 2012-12-25 | Identification method of advertisement image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012105723807A CN102982348A (en) | 2012-12-25 | 2012-12-25 | Identification method of advertisement image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102982348A true CN102982348A (en) | 2013-03-20 |
Family
ID=47856330
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012105723807A Pending CN102982348A (en) | 2012-12-25 | 2012-12-25 | Identification method of advertisement image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102982348A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679506A (en) * | 2013-12-02 | 2014-03-26 | 小米科技有限责任公司 | Information acquiring method, device, terminal and server |
| CN104424485A (en) * | 2013-08-22 | 2015-03-18 | 北京卓易讯畅科技有限公司 | Method and device for obtaining specific information based on image recognition |
| CN104902318A (en) * | 2015-04-29 | 2015-09-09 | 小米科技有限责任公司 | Playing control method and terminal device |
| CN104933140A (en) * | 2015-06-18 | 2015-09-23 | 上海融视广告传媒有限公司 | Media interaction method based on image |
| CN105160561A (en) * | 2015-09-30 | 2015-12-16 | 北京中搜网络技术股份有限公司 | Situational picture advertising method and apparatus |
| CN105303217A (en) * | 2015-10-13 | 2016-02-03 | 北京工业大学 | Efficient implicit two-dimensional code system |
| CN105447534A (en) * | 2014-08-07 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Imaged-based information presenting method and device |
| CN106203406A (en) * | 2016-08-27 | 2016-12-07 | 李春华 | A kind of identification system based on cloud computing |
| CN106202188A (en) * | 2016-06-23 | 2016-12-07 | 广州筷子信息科技有限公司 | The method and apparatus of advertising creative element tags |
| CN108564087A (en) * | 2018-04-13 | 2018-09-21 | 腾讯科技（深圳）有限公司 | Risk Identification Method, device, terminal and the storage medium of small advertisement |
| CN108647549A (en) * | 2018-03-14 | 2018-10-12 | 福州迈新生物技术开发有限公司 | The processing method of bar code image, apparatus and system |
| CN110580640A (en) * | 2019-09-05 | 2019-12-17 | 大有智能科技(嘉兴)有限公司 | Information pushing and receiving method, terminal, data processing server and system |
| CN111275134A (en) * | 2020-02-26 | 2020-06-12 | 上海明略人工智能(集团)有限公司 | Image matching method and device, storage medium and electronic device |
| US10841482B1 (en) | 2019-05-23 | 2020-11-17 | International Business Machines Corporation | Recommending camera settings for publishing a photograph based on identified substance |
| US10943256B2 (en) | 2016-06-23 | 2021-03-09 | Guangzhou Kuaizi Information Technology Co., Ltd. | Methods and systems for automatically generating advertisements |
| CN115018565A (en) * | 2022-08-08 | 2022-09-06 | 长沙朗源电子科技有限公司 | Advertisement media image identification method, system, equipment and readable storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090313127A1 (en) * | 2008-06-11 | 2009-12-17 | Yahoo! Inc. | System and method for using contextual sections of web page content for serving advertisements in online advertising |
| CN101789111A (en) * | 2009-12-30 | 2010-07-28 | 中国科学院自动化研究所 | Site advertisement and system for real-time interaction of electronic advertisement data |
-
2012
- 2012-12-25 CN CN2012105723807A patent/CN102982348A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090313127A1 (en) * | 2008-06-11 | 2009-12-17 | Yahoo! Inc. | System and method for using contextual sections of web page content for serving advertisements in online advertising |
| CN101789111A (en) * | 2009-12-30 | 2010-07-28 | 中国科学院自动化研究所 | Site advertisement and system for real-time interaction of electronic advertisement data |
Non-Patent Citations (1)
| Title |
|---|
| 杨红颖等: "基于SVR分类的多特征彩色图像检索算法", 《小型微型计算机系统》, vol. 29, no. 10, 31 October 2008 (2008-10-31) * |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104424485A (en) * | 2013-08-22 | 2015-03-18 | 北京卓易讯畅科技有限公司 | Method and device for obtaining specific information based on image recognition |
| CN103679506A (en) * | 2013-12-02 | 2014-03-26 | 小米科技有限责任公司 | Information acquiring method, device, terminal and server |
| CN103679506B (en) * | 2013-12-02 | 2018-04-24 | 小米科技有限责任公司 | A kind of method, apparatus, terminal and server for obtaining information |
| CN105447534A (en) * | 2014-08-07 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Imaged-based information presenting method and device |
| CN104902318B (en) * | 2015-04-29 | 2018-09-18 | 小米科技有限责任公司 | Control method for playing back and terminal device |
| CN104902318A (en) * | 2015-04-29 | 2015-09-09 | 小米科技有限责任公司 | Playing control method and terminal device |
| CN104933140A (en) * | 2015-06-18 | 2015-09-23 | 上海融视广告传媒有限公司 | Media interaction method based on image |
| CN104933140B (en) * | 2015-06-18 | 2018-07-17 | 上海融视广告传媒有限公司 | A kind of Media method based on image |
| CN105160561A (en) * | 2015-09-30 | 2015-12-16 | 北京中搜网络技术股份有限公司 | Situational picture advertising method and apparatus |
| CN105303217A (en) * | 2015-10-13 | 2016-02-03 | 北京工业大学 | Efficient implicit two-dimensional code system |
| CN105303217B (en) * | 2015-10-13 | 2018-12-18 | 北京工业大学 | A kind of efficient implicit 2 D code system |
| US10943256B2 (en) | 2016-06-23 | 2021-03-09 | Guangzhou Kuaizi Information Technology Co., Ltd. | Methods and systems for automatically generating advertisements |
| CN106202188A (en) * | 2016-06-23 | 2016-12-07 | 广州筷子信息科技有限公司 | The method and apparatus of advertising creative element tags |
| CN106202188B (en) * | 2016-06-23 | 2021-05-11 | 广州筷子信息科技有限公司 | Method and device for tagging elements of advertisement creatives |
| CN106203406A (en) * | 2016-08-27 | 2016-12-07 | 李春华 | A kind of identification system based on cloud computing |
| CN108647549A (en) * | 2018-03-14 | 2018-10-12 | 福州迈新生物技术开发有限公司 | The processing method of bar code image, apparatus and system |
| CN108647549B (en) * | 2018-03-14 | 2022-10-21 | 福州迈新生物技术开发有限公司 | Bar code image processing method, device and system |
| CN108564087A (en) * | 2018-04-13 | 2018-09-21 | 腾讯科技（深圳）有限公司 | Risk Identification Method, device, terminal and the storage medium of small advertisement |
| CN108564087B (en) * | 2018-04-13 | 2023-05-12 | 腾讯科技（深圳）有限公司 | Risk identification method, device, terminal and storage medium for small advertisements |
| US10841482B1 (en) | 2019-05-23 | 2020-11-17 | International Business Machines Corporation | Recommending camera settings for publishing a photograph based on identified substance |
| CN110580640A (en) * | 2019-09-05 | 2019-12-17 | 大有智能科技(嘉兴)有限公司 | Information pushing and receiving method, terminal, data processing server and system |
| CN110580640B (en) * | 2019-09-05 | 2024-07-09 | 大有智能科技(嘉兴)有限公司 | Information pushing and receiving method, terminal, data processing server and system |
| CN111275134A (en) * | 2020-02-26 | 2020-06-12 | 上海明略人工智能(集团)有限公司 | Image matching method and device, storage medium and electronic device |
| CN115018565A (en) * | 2022-08-08 | 2022-09-06 | 长沙朗源电子科技有限公司 | Advertisement media image identification method, system, equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102982348A (en) | Identification method of advertisement image | |
| US9959903B2 (en) | Video playback method | |
| Rees-Roberts | After fashion film: social video and brand content in the influencer economy | |
| US8065200B2 (en) | Virtual web store with product images | |
| CN102930523B (en) | Picture processing system and picture splicing implementation method | |
| CN106610785B (en) | commodity object list information processing method and device | |
| CN104038705B (en) | Video creating method and device | |
| CN103218595A (en) | Terminal and method for recognizing two-dimensional codes | |
| CN103988202A (en) | Image attractiveness based indexing and searching | |
| WO2012024190A1 (en) | Method and system of sending information based on a user card | |
| CN104331515B (en) | A kind of method and system for automatically generating tourism diary | |
| US9818044B2 (en) | Content update suggestions | |
| US20130290871A1 (en) | Transmedia Storytelling Tracking and Mapping System | |
| CN107357934A (en) | A kind of frame animation processing system and frame animation processing method based on Web | |
| WO2023087990A1 (en) | Image display method and apparatus, computer device, and storage medium | |
| CN105786776A (en) | Animation display method and apparatus | |
| KR20220036016A (en) | Solution for making of art gallery employing virtual reality | |
| CN111680482B (en) | Title image-text generation method and computing device | |
| CN105247606A (en) | Image display method and user terminal | |
| CN105578035A (en) | Image processing method and electronic device | |
| CN111158647B (en) | Creative material self-adaptive generation method and device based on structural theory | |
| CN106779816A (en) | A kind of advertisement distributing system and method based on 3D maps | |
| US20160005177A1 (en) | Service provision program | |
| CN113989404B (en) | Picture processing method, apparatus, device, storage medium, and program product | |
| CN114445750A (en) | Video object segmentation method, device, storage medium and program product |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130320 |
|
| RJ01 | Rejection of invention patent application after publication |