CN103198424B - A kind of purchase method and device based on intelligent display device - Google Patents
A kind of purchase method and device based on intelligent display device Download PDFInfo
- Publication number
- CN103198424B CN103198424B CN201310108906.0A CN201310108906A CN103198424B CN 103198424 B CN103198424 B CN 103198424B CN 201310108906 A CN201310108906 A CN 201310108906A CN 103198424 B CN103198424 B CN 103198424B
- Authority
- CN
- China
- Prior art keywords
- commodity
- pixel
- sectional drawing
- value
- array
- 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.)
- Expired - Fee Related
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention is applied to multimedia application field there is provided a kind of purchase method based on intelligent display device and device, and methods described includes：Obtain the commodity sectional drawing that user chooses；Pixel is gathered from the commodity sectional drawing, the pixel value of the pixel gathered is matched with default pixel criterion value, product features information is extracted；By the product features information transmission to shopping website, merchandise news corresponding with the product features information is inquired from the shopping website, the merchandise news is fed back into user, so that user selects the corresponding commodity of purchase from the merchandise news.Thus no longer need to be manually entered product features information and done shopping, the commodity sectional drawing that system can be chosen according to user, from trend user recommended user commodity interested.
Description
Technical field
The invention belongs to multimedia application field, more particularly to a kind of purchase method and dress based on intelligent display device
Put.
Background technology
With the application of continuing to develop for technology, particularly internet in the electronic equipment such as TV and mobile terminal, make
A variety of internet interactive applications can be realized by intelligent display devices such as TVs by obtaining people, and such as online, object for appreciation is played, seen a film.
Meanwhile, with the development of ecommerce, the shopping at network based on intelligent display devices such as TVs is also being risen, however, TV
It has been merely possible to a simple display device to realize that shopping at network, i.e. user are connected to shopping website on TV, so
Input merchandise news is searched corresponding commodity and bought afterwards, it is impossible to which the system based on intelligent display devices such as TVs of realization is pushed away automatically
Recommend shopping, and need user to be actively entered the relevant informations of commodity, therefore process is more complicated, not enough facilitates.
The content of the invention
The purpose of the embodiment of the present invention is to provide a kind of purchase method and device based on intelligent display device, it is intended to solve
The certainly existing purchase method based on intelligent display device can not realize that the system based on intelligent display devices such as TVs is pushed away automatically
Recommend shopping, the problem of not enough facilitating.
The embodiment of the present invention is achieved in that a kind of purchase method based on intelligent display device, and methods described includes：
Obtain the commodity sectional drawing that user chooses；
Pixel is gathered from the commodity sectional drawing, by the pixel value of the pixel gathered and default pixel criterion value
Matched, extract product features information；
By the product features information transmission to shopping website, inquire and believe with the product features from the shopping website
Corresponding merchandise news is ceased, the merchandise news is fed back into user, so that user selects purchase correspondence from the merchandise news
Commodity.
The another object of the embodiment of the present invention is to provide a kind of purchasing article based on intelligent display device, described device
Including：
Sectional drawing unit, the commodity sectional drawing for obtaining user's selection；
Feature extraction unit, for gathering pixel from the commodity sectional drawing, by the pixel value of the pixel gathered
Matched with default pixel criterion value, extract product features information；
Commodity retrieve and shopping unit, for by the product features information transmission to shopping website, from the shopping website
In inquire merchandise news corresponding with the product features information, the merchandise news is fed back into user, so that user is from institute
State the corresponding commodity of selection purchase in merchandise news.
In embodiments of the present invention, the commodity sectional drawing chosen when watching TV programme to user is analyzed and processed automatically
The characteristic information of commodity in commodity sectional drawing is obtained, corresponding merchandise news and store information just can be inquired from shopping website simultaneously
It is shown to user so that user can advantageously complete shopping at network on intelligent display device very much.No longer need to be manually entered
Product features information search corresponding goods are done shopping, the commodity sectional drawing that system can be chosen according to user, are recommended from trend user
User may be interested commodity.
Brief description of the drawings
Fig. 1 is the flow chart of the purchase method provided in an embodiment of the present invention based on intelligent display device；
Fig. 2 is the schematic diagram of the quaternary tree image cutting method provided in an embodiment of the present invention for cutting image；
Fig. 3 is the leaf node that one has commodity image in Fig. 2 provided in an embodiment of the present invention（That is black leaf node）Generation
Color normal distribution；
Fig. 4 is the purchasing article structure chart provided in an embodiment of the present invention based on intelligent display device.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention,
It is not intended to limit the present invention.
In order to illustrate technical solutions according to the invention, illustrated below by specific embodiment.
Embodiment one：
A kind of purchase method flow based on intelligent display device provided as shown in Figure 1 for first embodiment of the invention
Figure, for convenience of description, illustrate only the part related to the embodiment of the present invention.
In step S101, the commodity sectional drawing that user chooses is obtained.
In embodiments of the present invention, user, if seeing oneself article interested, can lead to when watching TV programme
Cross remote control and sectional drawing is carried out on intelligent display device screen, to intelligent display device input sectional drawing instruction.Thus, for intelligence
For display device, after the sectional drawing instruction that user's input is received when playing TV programme, the article in TV programme is entered
Row sectional drawing, obtains the commodity sectional drawing that user chooses.It is preferred that, because in TV programme playing process, picture switching is too fast, because
And now selected sectional drawing may not be the commodity sectional drawing interested to user, thus, intelligent display device is playing electricity
If receiving the sectional drawing instruction of user's input during depending on program, playback receives the caching in preset time period before sectional drawing instruction
Television program image, and by the image feedback in time period to user, now user is just optional gets oneself business interested
Product sectional drawing, and intelligent display device obtains user's commodity sectional drawing selected when watching buffered television program video.Wherein, intelligence
Energy display device includes but is not limited to intelligent television.
It is preferred that, commodity sectional drawing is passed to sectional drawing Processing Interface by intelligent display device by asynchronous message transfer mechanism,
Sectional drawing Processing Interface processing image is then withouted waiting for, and can continue occur other requests, the processing of sectional drawing Processing Interface is completed
After can notify intelligent display device automatically.
In step s 102, gather pixel from commodity sectional drawing, by the pixel value of the pixel gathered with it is default
Respective pixel standard value is matched, and extracts product features information.
In embodiments of the present invention, after the commodity sectional drawing of user's selection is got, the commodity sectional drawing is analyzed, from
In extract product features information.Wherein, product features information includes but is not limited to：Trade name mark, merchandise classification, commodity
Price, Brand.
The step can be divided into two sub-steps.
The first step, cuts to the commodity sectional drawing that user chooses, and gathers the picture of commodity pixel in the commodity sectional drawing
Element value.
In embodiments of the present invention, the commodity sectional drawing that user chooses is cut first, cuts out commodity and background.This
Place, is cut using quaternary tree image cutting method to commodity sectional drawing, in the way of image to be cut into four quadrants pair
Commodity sectional drawing recurrence cuts quadrant, until the high density pixel in commodity sectional drawing is cut in each quadrant, and according to the business
The quadrant cut state generation quaternary tree of product sectional drawing.
Specifically, commodity sectional drawing is cut into four quadrants first, then four quadrants are entered according to image complexity
Row sequence, three higher quadrants of selection complexity（Three higher quadrants of complexity are selected i.e. in four quadrants）, judge
Whether the minimum range at each quadrant all pixels point to corresponding quadrant edge in above three quadrant is less than or equal to should
The preset percentage of quadrant side length, if then no longer splitting the quadrant, otherwise continues the quadrant being divided into four quadrants,
The quadrant for selecting three image complexities in four quadrants high, then judges each quadrant all pixels point in three quadrants
Whether the minimum range to corresponding quadrant edge is less than or equal to the preset percentage of the quadrant side length, until in each quadrant
The minimum range at all pixels point to quadrant edge be both less than or equal to the quadrant side length preset percentage when, just no longer to business
Product sectional drawing carries out quadrant segmented.Wherein, the preset percentage of quadrant side length can be set as needed, and this is not restricted herein,
Preferably, the 1/100 of quadrant side length is taken.
Finally, corresponding quaternary tree is generated according to final commodity sectional drawing quadrant cut state, in the quaternary tree,
Differentiation has the node of commodity image and the node without commodity image（There is no commodity image as background）, wherein, there is commodity figure
The node of picture is the quadrant for representing to have after cutting commodity image, and the node without commodity image does not have commodity image after representing cutting
Quadrant.
Illustrated below by Fig. 2.
In fig. 2, upper left commodity figure represents the commodity sectional drawing that user selectes, first by the figure be cut into four as
Limit, is then further continued for according to image complexity, after judgement, and three quadrants higher to image complexity carry out further
Analysis, judges whether the minimum range at each quadrant all pixels point to quadrant edge where it in three quadrants is less than
Or equal to the preset percentage of the quadrant side length, if continuing to cut more than if, until the pixel in all quadrants to its institute
Minimum range become less than or equal to quadrant.Finally, it is the portion for having commodity image dash area to be designated as in the commodity figure
Point, blank parts are background（As shown in Figure 2）.Then, it is shown below with a left side according to the cutting situation of image generation such as Fig. 2
The corresponding quaternary tree of top commodity figure, wherein, dark node is the node for having commodity image, and white nodes are no commodity image
Node.
After commodity sectional drawing is cut, there will be business in quaternary tree using the highly dense IMAQ strategy of normal distribution
The leaf node of product image（Node obtained by when i.e. last time is cut）Generate color normal distribution, and therefrom collection commodity institute
There is the pixel value of pixel.
Specifically, the leaf node for having commodity image in quaternary tree to be called to OpenCV API, the normal distribution of color is generated
Figure, sets color density threshold value, is background during less than the threshold value, and remove the low value of color density（Remove background）, retain
The high value of color density, so as to collect at least one pixel, each pixel has the pixel value of itself, as in commodity
The pixel value of each pixel.
Being illustrated in figure 3 one has the color normal distribution of leaf node generation of commodity image, and quadrant in Fig. 2 is set here
The length of side be L, then axis is exactly X=L/2, here gather quadrant comparatively dense region pixel pixel value, as L/
The pixel value of 4 to 3L/4 interval pixels.
Second step, carries out matching treatment by the pixel value of commodity pixel and default respective pixel standard value, obtains business
Product characteristic information.
In embodiments of the present invention, when obtaining the commodity sectional drawing of user's selection and analyzing, also need what is inputted into access customer
The commodity page, the commodity message that now correspondence is initiated during user's input commodity page can pass to backstage, and backstage disappears according to the commodity
The type search of the commodity value carried in breath goes out default correspondence standard information, and the standard information includes：Default multigroup pixel
The type of commodity value includes but is not limited to intelligent display device in standard value and corresponding multiple fixed reference feature arrays, commodity message
Model etc.（Because intelligent display device model is different, its standard information is different, thus in order to for different intelligent displays
Equipment handles commodity sectional drawing pixel pixel value, it is necessary to which the model for intelligent display device obtains the standard letter of respective model
Breath）.Wherein, the characteristic information of one commodity of each fixed reference feature array correspondence.Wherein, each pixel has a pixel
Value, we are represented with red, green, blue colour herein, and the fixed reference feature array carries at least one set of fixed reference feature value.
Get after corresponding pixel criterion value, pass through formula Y1=sqrt ((Xr0-Xr1)2+(Yg0-Yg1)2+ (Zb0-
Zb1)2) calculate respectively commodity each pixel pixel value and default respective pixel standard value standard deviation.Wherein,
Sqrt represents square root calculation, and Xr0 represents red standard value, and Yg0 represents green standard value, and Zb0 represents blue standard value, Xr1,
Yg1, Zb1 represent the red, green, blue colour of the one of pixel collected in commodity sectional drawing respectively.Therewith, to calculating
Multiple standard deviations are averaged, and the difference of each standard deviation and the average value are calculated respectively, and the multiple differences calculated are entered
Row sequence, forms array and simultaneously stores, wherein can it is ascending according to the size of the difference or it is descending be ranked up, herein
This is not restricted.Finally, fixed reference feature array corresponding with the pixel criterion value is recalled from database, and is called
OpenCVAPI, using remaining system angle algorithmic formulaCalculate successively above-mentioned array with it is default
Deviation between multiple fixed reference feature arrays（Due to a kind of each group of commodity of fixed reference feature array representation, therefore it will be calculated
Array is compared with each group of fixed reference feature array, when deviation therebetween is minimum, then it is assumed that representated by the array
Commodity and reference commodity are closest, can find out this commodity and refer to commodity）, the corresponding fixed reference feature of minimum deflection obtained by calculating
Product features information corresponding to array is the product features information of commodity sectional drawing.Wherein, (i=1,2,3 ... n) represent Xik
Difference in the array, (j=1,2,3 ... n) represent default fixed reference feature numerical value to Xjk, and Cij represents array and default ginseng
Examine the deviation between feature array.Each fixed reference feature array just represents a kind of product features information of commodity, passes through remaining system
The characteristic information of commodity representated by the minimum fixed reference feature array of distance that angle algorithm is calculated is closest to commodity sectional drawing
The product features information of middle commodity.
To be elaborated on how below with an example using remaining system angle algorithmic formula calculate array with it is default multiple
Deviation between fixed reference feature array.The multiple differences calculated are Y11, Y22 ... Ynn, then by Y11, and Y22 ... Ynn are by number
Preserved after being worth size sequence composition array.Then, multiple fixed reference feature arrays are found from database（Such as（r11,r12,…
R1n) with (k11, k12, k13 ... .k1n)）, then the remaining system angle algorithmic formula for calling OpenCVAPI to realizeArray is calculated respectively（Y11,Y22,…Ynn）With（R11, r12 ... r1n) and (k11,
K12, k13 ... .k1n) between deviation, wherein, Xik (i=1,2,3 ... n) represent the difference of the array, Xjk (j=1,2,
3 ... n) represent default fixed reference feature array, and Cij represents the deviation between array and default fixed reference feature array.If calculating
Go out（R11, r12 ... r1n) with（Y11,Y22,…Ynn）Deviation it is smaller, then（R11, r12 ... r1n) corresponding to commodity it is special
Reference breath is the product features information of commodity sectional drawing；If calculating（(k11, k12, k13 ... .k1n) with（Y11,Y22,…
Ynn）Deviation it is smaller, then the product features information corresponding to (k11, k12, k13 ... .k1n) be commodity sectional drawing commodity it is special
Reference ceases.
In step s 103, by product features information transmission to shopping website, the merchandise news is fed back into user, for
User selects the corresponding commodity of purchase from the merchandise news.
In embodiments of the present invention, due to present many shopping websites（Such as Taobao）API be all open, thus
Obtain after product features information, product features information just can be passed to the application programming interfaces of shopping website as parameter
（API, Application Programming Interface）, just can be inquired in shopping website with commodity sectional drawing
The corresponding merchandise news and store information of commodity, and these merchandise newss and store information are included in intelligent display terminal
Upper confession user selection, user just can log in shopping website on intelligent display terminal and complete shopping at network.
In embodiments of the present invention, the commodity sectional drawing chosen when watching TV programme to user is analyzed and processed automatically
The characteristic information of commodity in commodity sectional drawing is obtained, corresponding merchandise news and store information just can be inquired from shopping website simultaneously
It is shown to user so that user can advantageously complete shopping at network on intelligent display device very much.No longer need to be manually entered
Product features information search corresponding goods are done shopping, the commodity sectional drawing that system can be chosen according to user, are recommended from trend user
User may be interested commodity.
Embodiment two：
Fig. 4 is the purchasing article structural representation provided in an embodiment of the present invention based on intelligent display device, in order to interior
Appearance becomes simplified, illustrate only the part related to the embodiment of the present invention, other parts are omitted herein.
Sectional drawing unit 41, the commodity sectional drawing for obtaining user's selection.
In embodiments of the present invention, sectional drawing unit 41 includes：
Programme replay unit 411, if the sectional drawing instruction for receiving user's input when playing TV programme, playback
Receive the buffered television program video in preset time period before sectional drawing instruction；
Sectional drawing acquiring unit 412, the commodity sectional drawing selected when watching buffered television program video for obtaining user.
In embodiments of the present invention, user, can be if seeing oneself article interested when watching TV programme
Sectional drawing, user's input sectional drawing instruction are carried out on screen.Thus, such as described sectional drawing acquiring unit 412 is when playing TV programme
After the sectional drawing instruction for receiving user's input, sectional drawing is carried out to the article in TV programme, the commodity sectional drawing that user chooses is obtained.
It is preferred that, because in TV programme playing process, it may not be to use that picture, which switches too fast thus now selected sectional drawing,
Commodity sectional drawing interested to family, thus, if the programme replay unit 411 receives user's input when playing TV programme
Sectional drawing instruction, then playback receives the buffered television program video in preset time period before sectional drawing instruction, and the sectional drawing is obtained
Take unit 412 by the image feedback in time period to user, now user is just optional gets oneself commodity sectional drawing interested,
And sectional drawing acquiring unit 412 obtains user's commodity sectional drawing selected when watching buffered television program video.
Feature extraction unit 42, for gathering pixel from the commodity sectional drawing, by the pixel of the pixel gathered
Value is matched with default pixel criterion value, extracts product features information.
In embodiments of the present invention, product features information includes but is not limited to：Trade name mark, merchandise classification, commodity
Price, Brand.
In embodiments of the present invention, feature extraction unit 42 includes：
Graphics processing unit 421, cuts to the commodity sectional drawing for pitching image cutting method using four, gathers institute
State the pixel value of commodity pixel in commodity sectional drawing.
Wherein, graphics processing unit 421 includes again：
Image cutting 4211 is right according to the height of image complexity for commodity sectional drawing to be cut into four quadrants
Four quadrants are ranked up, three high quadrants of selection image complexity, judge each quadrant in three quadrants
Whether the minimum range at middle all pixels point to place quadrant edge is less than or equal to the preset percentage of the quadrant side length；It is then
Quaternary tree is generated according to the quadrant cut state of the commodity sectional drawing；Otherwise cutting is proceeded, until owning in each quadrant
Pixel be both less than to the minimum range at place quadrant edge or the preset percentage equal to the quadrant side length untill.
Specifically, commodity sectional drawing is cut into four quadrants by image cutting 4211 first, then complicated according to image
Degree is ranked up to four quadrants, three higher quadrants of selection complexity（Select complexity higher i.e. in four quadrants
Three quadrants）, judge each quadrant all pixels point in above three quadrant to the minimum range at corresponding quadrant edge
Whether the preset percentage of the quadrant side length is less than or equal to, if then no longer splitting to the quadrant, otherwise continuing should
Quadrant is divided into four quadrants, the quadrant for selecting three image complexities in four quadrants high, then judges in three quadrants
Whether the minimum range at each quadrant all pixels point to corresponding quadrant edge is less than or equal to default the hundred of the quadrant side length
Divide ratio, until the minimum range at all pixels point to quadrant edge in each quadrant is both less than or default equal to the quadrant side length
During percentage, with regard to no longer carrying out quadrant segmented to commodity sectional drawing.Wherein, the preset percentage of quadrant side length can as needed and
If be not restricted to this herein, it is preferable that take the 1/100 of quadrant side length.Further, according to final commodity sectional drawing quadrant
Cut state generates corresponding quaternary tree, in the quaternary tree, distinguishes the node and no commodity image for having commodity image
Node（There is no commodity image as background）, wherein, the node for having commodity image represent cutting after have commodity image as
Limit, the node without commodity image does not have the quadrant of commodity image after representing cutting.
Pixel extraction unit 4212, for the leaf node in the quaternary tree to be generated into color normal distribution, and from institute
State the pixel value that commodity pixel is gathered in normal distribution.Specifically, the leaf node for having commodity image in quaternary tree is called
OpenCV API, generates the normal distribution of color, sets color density threshold value, is background during less than the threshold value, and remove face
The low value of color density（Remove background）, the high value of retaining color density, so that at least one pixel is collected, each pixel
Point has the pixel value of itself, the pixel value of each pixel as in commodity.
Product features acquiring unit 422, for the pixel value of the commodity pixel to be entered with default pixel criterion value
Row matching treatment, obtains product features information.
Wherein, product features acquiring unit 422 includes again：
Standard deviation computing unit 4221, for calculate respectively the pixel value of each pixel of the commodity with it is default corresponding
The standard deviation of pixel criterion value；
In embodiments of the present invention, specifically, when obtaining the commodity sectional drawing of user's selection and analyzing, also need into access customer
The commodity page of input, the commodity message initiated now is corresponded to during user's input commodity page can pass to backstage, and backstage is according to this
The type search of the commodity value carried in commodity message goes out default correspondence standard information, and the standard information includes：It is default many
The type of commodity value includes but is not limited to intelligent aobvious in group pixel criterion value and corresponding multiple fixed reference feature arrays, commodity message
Show unit type etc.（Because intelligent display device model is different, its standard information is different, thus in order to for different intelligence
Energy display apparatus processes commodity sectional drawing pixel pixel value is, it is necessary to which the model for intelligent display device obtains respective model
Standard information）.Wherein, the characteristic information of one commodity of each fixed reference feature array correspondence.Wherein, each pixel has one
Pixel value, herein we represent that the fixed reference feature array carries at least one set of fixed reference feature value with red, green, blue colour.
Get after corresponding pixel criterion value, standard deviation computing unit 4221 passes through formula Y1=sqrt ((Xr0-Xr1)2
+(Yg0-Yg1)2+(Zb0-Zb1)2) calculate respectively commodity each pixel pixel value and default respective pixel standard value
Standard deviation.Wherein, sqrt represents square root calculation, and Xr0 represents red standard value, and Yg0 represents green standard value, and Zb0 is represented
Blue standard value, Xr1, Yg1, Zb1 represents the red, green, blue colour of the one of pixel collected in commodity sectional drawing respectively.
Difference value determining unit 4222, the average value for seeking multiple calculated standard deviations, calculates each standard deviation respectively
With the difference of the average value, and the multiple differences calculated are ranked up, by multiple differences formation array after sequence and deposited
Storage；Wherein can it is ascending according to the size of the difference or it is descending be ranked up, this is not restricted herein.
Characteristic matching unit 4223, for utilizing formulaThe array is calculated successively
Deviation between default fixed reference feature array, extracts fixed reference feature array corresponding with the minimum deflection calculated, by institute
The corresponding characteristic information of fixed reference feature array is stated as product features information, wherein, (i=1,2,3 ... n) represent institute to the Xik
The difference in array is stated, (j=1,2,3 ... n) represent default fixed reference feature value to Xjk, and Cij represents the array and default ginseng
Examine the deviation between feature array.Each fixed reference feature array just represents a kind of product features information of commodity, characteristic matching
The characteristic information of commodity representated by the fixed reference feature array for the distance minimum that unit 4223 is calculated by remaining system angle algorithm
The product features information of commodity closest in commodity sectional drawing.
To be elaborated on how below with an example using remaining system angle algorithmic formula calculate array with it is default multiple
Deviation between fixed reference feature array.The multiple differences calculated are Y11, Y22 ... Ynn, then by Y11, and Y22 ... Ynn are by number
Preserved after being worth size sequence composition array.Then, multiple fixed reference feature arrays are found from database（Such as（r11,r12,…
R1n) with (k11, k12, k13 ... .k1n)）, then the remaining system angle algorithmic formula for calling OpenCVAPI to realizeArray is calculated respectively（Y11,Y22,…Ynn）With（R11, r12 ... r1n) and (k11,
K12, k13 ... .k1n) between deviation, wherein, Xik (i=1,2,3 ... n) represent the difference of the array, Xjk (j=1,2,
3 ... n) represent default fixed reference feature array, and Cij represents the deviation between array and default fixed reference feature array.If calculating
Go out（R11, r12 ... r1n) with（Y11,Y22,…Ynn）Deviation it is smaller, then（R11, r12 ... r1n) corresponding to commodity it is special
Reference breath is the product features information of commodity sectional drawing；If calculating（(k11, k12, k13 ... .k1n) with（Y11,Y22,…
Ynn）Deviation it is smaller, then the product features information corresponding to (k11, k12, k13 ... .k1n) be commodity sectional drawing commodity it is special
Reference ceases.
Commodity retrieve and shopping unit 43, for by the product features information transmission to shopping website, from the shopping network
Merchandise news corresponding with the product features information is inquired in standing, the merchandise news is fed back into user, for user from
The corresponding commodity of selection purchase in the merchandise news.
In embodiments of the present invention, due to present many shopping websites（Such as Taobao）API be all open, thus
Obtain after product features information, product features information just as parameter can be passed to shopping network by commodity retrieval and shopping unit 43
The application programming interfaces stood（API, Application Programming Interface）, just can be looked into shopping website
The merchandise news and store information corresponding with the commodity in commodity sectional drawing is ask, and these merchandise newss and store information are shown
Show and selected on intelligent display terminal for user, user just can log in shopping website on intelligent display terminal and complete network purchase
Thing.
One kind of a kind of purchasing article and above-described embodiment based on intelligent display device in the present embodiment is based on intelligence
The purchase method of display device is corresponding, and realization principle is basically identical, and implementation process refer to above-described embodiment, no longer right herein
The process that implements of device is repeated.
In embodiments of the present invention, the commodity sectional drawing chosen when watching TV programme to user is analyzed and processed automatically
The characteristic information of commodity in commodity sectional drawing is obtained, corresponding merchandise news and store information just can be inquired from shopping website simultaneously
It is shown to user so that user can advantageously complete shopping at network on intelligent display device very much.No longer need to be manually entered
Product features information search corresponding goods are done shopping, the commodity sectional drawing that system can be chosen according to user, are recommended from trend user
User may be interested commodity.
Can be with it will appreciated by the skilled person that realizing that all or part of step in above-described embodiment method is
The hardware of correlation is instructed to complete by program, described program can be stored in a computer read/write memory medium
In, described storage medium, such as ROM/RAM, disk, CD.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
Any modifications, equivalent substitutions and improvements made within refreshing and principle etc., should be included in the scope of the protection.
Claims (4)
1. a kind of purchase method based on intelligent display device, it is characterised in that methods described includes：
S1, the commodity sectional drawing for obtaining user's selection；
S2, from the commodity sectional drawing pixel is gathered, by the pixel value of the pixel gathered and default pixel criterion value
Matched, extract product features information；
S3, by the product features information transmission to shopping website, inquired from the shopping website and the product features believe
Corresponding merchandise news is ceased, the merchandise news is fed back into user, so that user selects purchase correspondence from the merchandise news
Commodity；
The step S2 is specifically included：
S21, using four fork image cutting methods the commodity sectional drawing is cut, commodity pixel in the collection commodity sectional drawing
The pixel value of point；
S22, the pixel value of the commodity pixel and default pixel criterion value be subjected to matching treatment, obtain product features letter
Breath；
The step S21 is specifically included：
S210, commodity sectional drawing is cut into four quadrants, four quadrants is ranked up according to the height of image complexity,
Three quadrants for selecting complexity high；
S211, judge all pixels point in each quadrant in three quadrants to place quadrant edge minimum range whether
Less than or equal to the preset percentage of the quadrant side length；Be to go to S212, otherwise go to the step S210, until it is each as
Limit in all pixels point be both less than to the minimum range at place quadrant edge or the preset percentage equal to the quadrant side length untill；
S212, according to the quadrant cut state of the commodity sectional drawing generate quaternary tree；
S213, the leaf node in the quaternary tree generated into color normal distribution, and gather business from the normal distribution
The pixel value of product pixel；
The step S22 is specifically included：
S221, the pixel value that each pixel of the commodity is calculated respectively and the standard deviation of default respective pixel standard value；
S222, the average value for seeking multiple calculated standard deviations, calculate the difference of each standard deviation and the average value respectively, and will
The multiple differences calculated are ranked up, and multiple differences after sequence are formed into array and stored；
S223, the deviation for calculating the array and default fixed reference feature array, are extracted corresponding with the minimum deflection calculated
Fixed reference feature array, regard the corresponding characteristic information of the fixed reference feature array as product features information.
2. the method as described in claim 1, it is characterised in that in the step S223, utilize formula
The deviation between the array and default fixed reference feature array is calculated successively, wherein, the XikRepresent the difference in the array
Value, XjkRepresent default fixed reference feature value, CijRepresent the deviation between the array and default fixed reference feature array, i=1,
2,3 ... n, j=1,2,3 ... n.
3. a kind of purchasing article based on intelligent display device, it is characterised in that described device includes：
Sectional drawing unit, the commodity sectional drawing for obtaining user's selection；
Feature extraction unit, for gathering pixel from the commodity sectional drawing, by the pixel value of the pixel gathered and in advance
If pixel criterion value matched, extract product features information；
Commodity are retrieved and shopping unit, for the product features information transmission, to shopping website, to be looked into from the shopping website
Merchandise news corresponding with the product features information is ask, the merchandise news is fed back into user, so that user is from the business
The corresponding commodity of selection purchase in product information；
The feature extraction unit includes：
Graphics processing unit, cuts to the commodity sectional drawing for pitching image cutting method using four, gathers the commodity
The pixel value of commodity pixel in sectional drawing；
Product features acquiring unit, for carrying out matching place the pixel value of the commodity pixel and default pixel criterion value
Reason, obtains product features information；
Described image processing unit includes：
Image cutting, for commodity sectional drawing to be cut into four quadrants, according to the height of image complexity to described four
Quadrant is ranked up, three high quadrants of selection complexity, judges all pixels point in each quadrant in three quadrants
Whether the minimum range to place quadrant edge is less than or equal to the preset percentage of the quadrant side length；It is then according to the commodity
The quadrant cut state generation quaternary tree of sectional drawing；Otherwise continue to cut, until in each quadrant all pixels point to place quadrant
The minimum range at edge be both less than or the preset percentage equal to the quadrant side length untill；
Pixel extraction unit, for the leaf node in the quaternary tree to be generated into color normal distribution, and divides from the normal state
The pixel value of commodity pixel is gathered in Butut；
The product features acquiring unit includes：
Standard deviation computing unit, pixel value and default respective pixel standard for calculating each pixel of the commodity respectively
The standard deviation of value；
Difference value determining unit, the average value for seeking multiple calculated standard deviations calculates each standard deviation and is averaged with this respectively
The difference of value, and the multiple differences calculated are ranked up, multiple differences after sequence are formed into array and stored.
4. device as claimed in claim 3, it is characterised in that the product features acquiring unit also includes：
Characteristic matching unit, for utilizing formulaCalculate successively the array with it is default
Deviation between fixed reference feature array, extracts fixed reference feature array corresponding with the minimum deflection calculated, by described with reference to special
The corresponding characteristic information of array is levied as product features information, wherein, the XikRepresent the difference in the array, XjkRepresent
Default fixed reference feature value, CijRepresent the deviation between the array and default fixed reference feature array, i=1,2,3 ... n, j
=1,2,3 ... n.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310108906.0A CN103198424B (en) | 2013-03-29 | 2013-03-29 | A kind of purchase method and device based on intelligent display device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310108906.0A CN103198424B (en) | 2013-03-29 | 2013-03-29 | A kind of purchase method and device based on intelligent display device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103198424A CN103198424A (en) | 2013-07-10 |
| CN103198424B true CN103198424B (en) | 2017-09-08 |
Family
ID=48720943
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310108906.0A Expired - Fee Related CN103198424B (en) | 2013-03-29 | 2013-03-29 | A kind of purchase method and device based on intelligent display device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103198424B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103607383A (en) * | 2013-11-12 | 2014-02-26 | 苏州壹世通科技有限公司 | Data access method and system |
| CN103927348B (en) * | 2014-04-02 | 2018-09-07 | 华为技术有限公司 | Image processing method, information acquisition method and device |
| CN106170816A (en) * | 2014-12-05 | 2016-11-30 | 株式会社威舍里 | Comprise the Products Show system and methods based on social commercial affairs of automatic recommendation function |
| CN106202078A (en) * | 2015-04-30 | 2016-12-07 | 雅虎公司 | Object hunting system and method and use the mobile communications device of this object method for searching |
| WO2017002477A1 (en) * | 2015-06-30 | 2017-01-05 | ケーアンドエムエンタープライズ株式会社 | Method and program for processing images within display screen |
| CN107102786B (en) | 2016-02-19 | 2020-06-12 | 腾讯科技（北京）有限公司 | Information processing method and client |
| CN106162213B (en) * | 2016-07-11 | 2019-08-09 | 福建方维信息科技有限公司 | A kind of merchandise display method and system based on net cast shopping |
| CN107679175B (en) * | 2017-09-29 | 2020-04-07 | 深圳市聚宝汇科技有限公司 | Method and system for batch datamation of vegetarian gold products |
| CN108055589B (en) * | 2017-12-20 | 2021-04-06 | 聚好看科技股份有限公司 | Intelligent television |
| CN111212250B (en) | 2017-12-20 | 2023-04-14 | 海信视像科技股份有限公司 | Smart television and display method of graphical user interface of television picture screenshot |
| CN109963201A (en) * | 2017-12-26 | 2019-07-02 | 深圳Tcl新技术有限公司 | A kind of real-time shopping method, system, intelligent TV network and storage medium |
| CN108921048A (en) * | 2018-06-14 | 2018-11-30 | 深圳码隆科技有限公司 | A kind of shopping settlement method, device and user terminal |
| US11039196B2 (en) | 2018-09-27 | 2021-06-15 | Hisense Visual Technology Co., Ltd. | Method and device for displaying a screen shot |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101937549A (en) * | 2010-10-09 | 2011-01-05 | 姚建 | Network shopping guidance system |
| CN102982472A (en) * | 2012-11-29 | 2013-03-20 | 何松 | Method and device for automatically recommending similar commodities by terminal |
-
2013
- 2013-03-29 CN CN201310108906.0A patent/CN103198424B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101937549A (en) * | 2010-10-09 | 2011-01-05 | 姚建 | Network shopping guidance system |
| CN102982472A (en) * | 2012-11-29 | 2013-03-20 | 何松 | Method and device for automatically recommending similar commodities by terminal |
Non-Patent Citations (1)
| Title |
|---|
| "快速的分形码图像检索方法";王兴元 等;《计算机科学与探索》;20090430(第4期);第2.1、5节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103198424A (en) | 2013-07-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103198424B (en) | A kind of purchase method and device based on intelligent display device | |
| US10299011B2 (en) | Method and system for user interaction with objects in a video linked to internet-accessible information about the objects | |
| CN1538351B (en) | Method and computer for generating visually representative video thumbnails | |
| CN107404656B (en) | Live video recommended method, device and server | |
| US11914639B2 (en) | Multimedia resource matching method and apparatus, storage medium, and electronic apparatus | |
| US8554639B2 (en) | Method and system for managing and displaying product images | |
| US10524005B2 (en) | Facilitating television based interaction with social networking tools | |
| KR20190099027A (en) | Video content summary | |
| CN103365936A (en) | Video recommendation system and method thereof | |
| CN110874780A (en) | Scenic spot playing system and recommendation method based on big data statistics | |
| US20130314438A1 (en) | Interactive overlay for digital video | |
| CN103065261A (en) | Method, device and system of video shopping based on gesture operation | |
| CN106454431B (en) | TV programme suggesting method and system | |
| CN112507163B (en) | Duration prediction model training method, recommendation method, device, equipment and medium | |
| WO2016201800A1 (en) | Information acquisition method, server, terminal, and method and apparatus for constructing database | |
| CN111159555A (en) | Commodity recommendation method, commodity recommendation device, server and storage medium | |
| CN113408590B (en) | Scene recognition method, training method, device, electronic equipment and program product | |
| KR102215735B1 (en) | Character selling and purchasing device for service provision in virtual reality space | |
| KR20210059593A (en) | Method, apparatus and computer program for providing sale mediate service based on influencer's contents | |
| CN111737517A (en) | Instant recommendation method and system based on short video | |
| KR102215729B1 (en) | Service provision program by user avatar applied to virtual reality space | |
| CN110969155A (en) | Food propaganda system based on intelligence AR product album of paintings | |
| CN112104910B (en) | Video searching method, device and system | |
| CN109309875B (en) | Method for displaying user behavior characteristic model on smart television | |
| CN112819313A (en) | Target image generation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170908 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |