CN108335317A - Shopping guide method and device under a kind of line - Google Patents
Shopping guide method and device under a kind of line Download PDFInfo
- Publication number
- CN108335317A CN108335317A CN201710041475.9A CN201710041475A CN108335317A CN 108335317 A CN108335317 A CN 108335317A CN 201710041475 A CN201710041475 A CN 201710041475A CN 108335317 A CN108335317 A CN 108335317A
- Authority
- CN
- China
- Prior art keywords
- customer
- movement target
- behavioral data
- designated movement
- commodity
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0281—Customer communication at a business location, e.g. providing product or service information, consulting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Abstract
The embodiment of the invention discloses shopping guide method and devices under a kind of line, are related to Video Analysis Technology field, and capable of solving under line commercial product recommending in sales process, to service the consumption brought to customer inconvenient.The embodiment of the present invention includes:In the video sequence acquired in specified region, designated movement target is identified and tracked, obtains the figure information of the designated movement target, the figure information is used to indicate the action that the designated movement target in the video sequence generates in the process in tracking;From customer behavior library, according to the action of generation, the corresponding behavioral data of the designated movement target is retrieved;According to the behavioral data retrieved, the recommendation results of the corresponding designated movement target are generated.The embodiment of the present invention is suitable for recommending under line.
Description
Technical field
The present invention relates to shopping guide method and devices under Video Analysis Technology field more particularly to a kind of line.
Background technology
With growing, the commercial podium behavioral data abundant since customer can be got on line of on-line selling,
And personalized commercial product recommending service can be provided to different customers, consumption experience on the line to improve customer, and then improve
The income of on-line selling.
In contrast, for being sold under line, due to the particularity of business place under line, it is similar on line if wanting to realize
Commercial product recommending service used by sale generally requires customer and installs corresponding application on mobile terminals, and carried out in customer
It under line in shopping process, needs to start the application, personalized commercial product recommending service could be provided for customer.It can be seen that line
Under commercial product recommending service generally require customer all various aspects carry out cooperation could realize.In addition, even if customer attempts to experience
Commercial product recommending service under line, commending system in order to customer provide be more suitable for customer's self-demand, hobby commodity,
Commending system is then needed accurately to obtain the identity information and behavioral data of current customer.However, this data acquisition gesture
Necessarily accurately human face analysis process, also means that, it is also necessary to ensure in lower sale scene online, be capable of providing sufficient
Photographic device, to complete the human face analysis etc. of current customer.
It is well known that although human face analysis can realize the certification of customer identification, it is online under true open scene
Under, the accuracy rate of human face analysis technology is not high, this will reduce the precision of customer identification identification.So, with customer
As the commending system for using object, due to excessively relying on the technologies such as human face analysis, lead to that covering surface is smaller, precision is relatively low,
To influence under line, commercial product recommending services the Facility Consumption brought to customer in sales process.
Invention content
The embodiment of the present invention provides shopping guide method and device under a kind of line, can solve that commodity push away in sales process under line
It is inconvenient to recommend the consumption that service is brought to customer.
In order to achieve the above objectives, the embodiment of the present invention adopts the following technical scheme that:
In a first aspect, the method that the embodiment of the present invention provides, including:
In the video sequence acquired in specified region, designated movement target is identified and tracked, the designated movement is obtained
The figure information of target, the figure information are used to indicate that the designated movement target in the video sequence to be produced during tracking
Raw action;
From customer behavior library, according to the action of generation, the corresponding behavioral data of the designated movement target is retrieved;
According to the behavioral data retrieved, the recommendation results of the corresponding designated movement target are generated.
With reference to first aspect, in the first possible realization method of first aspect, the method further includes:
In each video sequence currently acquired, identifies and track all moving targets;
Customer is identified in all moving targets, and determines the behavioral data for belonging to customer, wherein behavioral data at least wraps
The record of customer's Evaluation product, customer are included with other people interbehavior, the browsing route of customer, customers in specified commodity location
Any one in the interior residence time.
The possible realization method of with reference to first aspect the first, in second of possible realization method of first aspect
In, the method further includes:
The behavioral data for belonging to customer is stored to the customer behavior library.
Second of possible realization method with reference to first aspect, in the third possible realization method of first aspect
In, customer is identified in all moving targets described, and determination belongs to after the behavioral data of customer, the method is also wrapped
It includes:
According to the content of the behavioral data, all moving targets for belonging to customer are grouped, are obtained at least one
Customer gathers.
The third possible realization method with reference to first aspect, in the 4th kind of possible realization method of first aspect
In, the behavioral data that the basis retrieves generates the recommendation results of the corresponding designated movement target, specifically includes:
If in the presence of the customer's set for including the designated movement target, gather in customer where the designated movement target
In, obtain the behavioral data of other moving targets in addition to the designated movement target;
According to the behavioral data of other moving targets, the recommendation results are determined;
If described using hot item as recommendation results there is no the customer's set for including the designated movement target
Hot item includes at least the commodity that sales volume in specified time is more than first threshold.
The 4th kind of possible realization method with reference to first aspect, in the 5th kind of possible realization method of first aspect
In, the behavioral data of other moving targets described in the basis determines the recommendation results, specifically includes:
The commodity of other moving target deep browsings are determined as the recommendation results, the commodity of the deep browsing
It is more than the commodity of second threshold including at least number of visits in preset time, number on probation is more than the commodity of third threshold value, customer
Accumulated dwelling time is more than any one in the commodity of the 4th threshold value.
Any one in first with reference to first aspect to the 5th kind of possible realization method, the 6th of first aspect the
It is described in each video sequence currently acquired in the possible realization method of kind, after identifying and tracking all moving targets,
The method further includes:
Duplicate removal processing is carried out to all moving targets.
With reference to first aspect or any one in first to the 5th kind of possible realization method of first aspect,
In 7th kind of possible realization method of one side, after the recommendation results for generating the corresponding designated movement target, institute
The method of stating further includes:
The recommendation results are pushed into designated terminal and are shown, the designated terminal includes at least what shoppers' guide used
Terminal.
Second aspect, the device that the embodiment of the present invention provides, including:
Preprocessing module, in specifying the video sequence acquired in region, identifying and tracking designated movement target, obtain
To the figure information of the designated movement target, the figure information is used to indicate the designated movement target in the video sequence
In the action that tracking generates in the process;
Data and model module, for from customer behavior library, according to the action of generation, retrieving the preprocessing module and knowing
Not and track the obtained corresponding behavioral data of the designated movement target;
The data and model module, for according to the behavioral data retrieved, generating the corresponding designated movement target
Recommendation results.
In conjunction with second aspect, in the first possible realization method of second aspect, described device further includes:
The preprocessing module is additionally operable in each video sequence currently acquired, is identified and is tracked all movement mesh
Mark;
Macro or mass analysis module is additionally operable to identify customer in all moving targets, and determines the behavioral data for belonging to customer,
Wherein, behavioral data includes at least the record of customer's Evaluation product, customer and other people interbehavior, the browsing route of customer, Gus
Any one in residence time of the visitor in specified commodity location;
The Macro or mass analysis module is additionally operable to store the behavioral data for belonging to customer to the customer behavior library;
The Macro or mass analysis module, is additionally operable to the content according to the behavioral data, will belong to all movement mesh of customer
Mark is grouped, and obtains at least one customer's set.
In conjunction with the first possible realization method of second aspect, in second of possible realization method of second aspect
In, the data and model module are specifically used for:
If in the presence of the customer's set for including the designated movement target, gather in customer where the designated movement target
In, obtain the behavioral data of other moving targets in addition to the designated movement target;
According to the behavioral data of other moving targets, the recommendation results are determined;
If described using hot item as recommendation results there is no the customer's set for including the designated movement target
Hot item includes at least the commodity that sales volume in specified time is more than first threshold.
In conjunction with second of possible realization method of second aspect, in the third possible realization method of second aspect
In, the data and model module are specifically used for:
The commodity of other moving target deep browsings are determined as the recommendation results, the commodity of the deep browsing
It is more than the commodity of second threshold including at least number of visits in preset time, number on probation is more than the commodity of third threshold value, customer
Accumulated dwelling time is more than any one in the commodity of the 4th threshold value.
In conjunction with second aspect first to the third possible realization method in any one, the 4th of second aspect the
In the possible realization method of kind, described device further includes:
Authentication module, for carrying out duplicate removal processing to all moving targets.
In conjunction with second aspect or second aspect first to the third possible realization method in any one,
In 5th kind of possible realization method of two aspects, described device further includes:
Display module, for the recommendation results to be pushed to designated terminal and are shown, the designated terminal includes at least
The terminal that shoppers' guide uses.
It is compared in the prior art, using customer as the commending system using object, and excessively relies on human face analysis etc.
Technology, said program can will generate recommendation results and push to the terminal that shoppers' guide uses, for being carried for customer for shoppers' guide
For more targeted recommendation service.In actual application, the generating process of recommendation results is mainly specified by locking
Moving target, and the historical behavior data of designated movement target are analyzed, are counted, it obtains later and the designated movement target
The recommendation results to match.So, recommendation results can with for shoppers' guide provide abundant, sufficient Customer Resource and
The relevant information of corresponding customer, is also just more conducive to shoppers' guide and provides more targetedly recommendation service for customer.Cause
This, commercial product recommending services the consumption inconvenience brought to customer in sales process under solution line.
Description of the drawings
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to needed in the embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for ability
For the those of ordinary skill of domain, without creative efforts, it can also be obtained according to these attached drawings other attached
Figure.
Fig. 1 is the structural schematic diagram of commending system under a kind of line provided in an embodiment of the present invention;
Fig. 2 is the structural schematic diagram of shopper device under a kind of line provided in an embodiment of the present invention;
Fig. 3 is the structural schematic diagram of shopper device under another line provided in an embodiment of the present invention;
Fig. 4 is shopping guide method flow chart under a kind of line provided in an embodiment of the present invention.
Specific implementation mode
To make those skilled in the art more fully understand technical scheme of the present invention, below in conjunction with the accompanying drawings and specific embodiment party
Present invention is further described in detail for formula.Embodiments of the present invention are described in more detail below, the embodiment is shown
Example is shown in the accompanying drawings, and in which the same or similar labels are throughly indicated same or similar element or has identical or class
Like the element of function.It is exemplary below with reference to the embodiment of attached drawing description, is only used for explaining the present invention, and cannot
It is construed to limitation of the present invention.Those skilled in the art of the present technique are appreciated that unless expressly stated, odd number shape used herein
Formula " one ", "one", " described " and "the" may also comprise plural form.It is to be further understood that the specification of the present invention
The middle wording " comprising " used refers to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that
Other one or more features of presence or addition, integer, step, operation, element, component and/or their group.It should be understood that
When we say that an element is " connected " or " coupled " to another element, it can be directly connected or coupled to other elements, or
There may also be intermediary elements.In addition, " connection " used herein or " coupling " may include being wirelessly connected or coupling.Here make
Wording "and/or" includes any cell of one or more associated list items and all combines.The art
Technical staff is appreciated that unless otherwise defined all terms (including technical terms and scientific terms) used herein have
Meaning identical with the general understanding of the those of ordinary skill in fields of the present invention.It should also be understood that such as general
Term, which should be understood that, those of defined in dictionary has a meaning that is consistent with the meaning in the context of the prior art, and
Unless being defined as here, will not be explained with the meaning of idealization or too formal.
The embodiment of the present invention can be applied to commending system under a kind of line, and the method flow in the embodiment of the present invention specifically may be used
To execute under line as shown in Figure 1 in commending system.Wherein, it is included at least in commending system under the line:Camera, management
Server, employee's equipment.Wherein, management server can be considered as shopper device under a kind of possible line.
The following two kinds deployment way may be used in camera:Top set is installed or strabismus installation.No matter which kind of installation side is used
Formula will ensure merchandising location, for example, the functional areas such as channel or cashier near entire shop and shop all exist
In the visible range of camera.In embodiments of the present invention, the main function of camera is the acquisition of video sequence, specific real
Existing mode can refer to that this will not be repeated here later.
In addition, camera can carry out debugging networking with wireless router, to pass through wireless network and management server
Communication connection is established, the video sequence that multiple cameras are acquired can be summarized, be analyzed by such management server.Its
In, analytic process can specifically include the operations such as the pretreatment of video sequence, duplicate removal, not limit herein.
In embodiments of the present invention, digital camera or simulation camera specifically may be used in camera.Wherein, digital
The analog video signal of shooting can be converted into digital signal by camera, and then be transmitted to the management service being connect with camera
Device.The vision signal that simulation camera captures, converts analog signals into figure pattern, and pressed by video frequency collection card
The management server being connect with camera is transmitted to after contracting, common simulation camera can be with computer or PC phase
Even, from computer or PC record and to management server uploaded videos signal, i.e., camera is practical passes through computer
Or PC is communicated with management server.
Employee's equipment can be considered as terminal used in shoppers' guide, can specifically make an independent table apparatus in fact, or
It is integrated in a variety of different media data playing devices, such as smart mobile phone, tablet computer (Tablet Personal
Computer), laptop computer (Laptop Computer), personal digital assistant (personal digital
Assistant, PDA) or wearable device (Wearable Device) etc..The effect of employee's equipment is, by the finger of generation
The recommendation results for determining moving target are shown by employee's equipment, to which recommendation results are presented to shoppers' guide, for shopping guide people
Member understands the demand of customer, to provide recommendation service under more good line for customer.
In addition, being also deployed with database in commending system under above-mentioned line, which can be Gu mentioned hereinafter
Objective behavior library, the content stored in the database can propose that this will not be repeated here later.
Shopper device under line is included at least in commending system under above-mentioned possible line.Under the line in shopper device, example
Such as, under line as shown in Figure 2 in shopper device, preprocessing module 10 and data and model module 11 are included at least.
Wherein, preprocessing module 10, in specifying the video sequence acquired in region, identifying and tracking designated movement
Target obtains the figure information of designated movement target, and figure information is for reflecting designated movement clarification of objective.In data and mould
In pattern block 11, it is provided with customer behavior retrieval module 111, for from customer behavior library, retrieval preprocessing module 10 to identify
And track the obtained corresponding behavioral data of designated movement target.In data and model module 11, it is pre- to be provided with customer's purchase
Module 112 is surveyed, for according to the behavioral data retrieved, generating the recommendation results of corresponding designated movement target.
Other than above-mentioned module, shopper device further includes camera module 12 under line.In preprocessing module 10, setting
There are target detection and tracking module 101, specifically can be used in each video sequence currently acquired, identifies and track all
Moving target.Under the line in shopper device, it is additionally provided with authentication module 13, customer's deduplication module 131 is specifically included, is used for
Duplicate removal processing is carried out to all moving targets.
As shown in figure 3, for another part of recommendation apparatus under line, wherein equally include camera module 12, pretreatment mould
Block 10, data and model module 11.In addition, it is additionally provided with Macro or mass analysis module 14 in another part of lower recommendation apparatus online,
For identifying customer in all moving targets, and is determined and belonged to by customer's deduplication module 141 and customer behavior analysis module 142
The behavioral data of customer Yu, wherein behavioral data include at least the record of customer's Evaluation product, customer and other people interbehaviors,
Any one in residence time of the browsing route, customer of customer in specified commodity location;The row of customer will be belonged to
Summarized by data summarization module 143 for data and is stored to customer behavior library;According to the content of behavioral data, will belong to
All moving targets of customer are grouped, and obtain at least one customer's set.
In data and model module 11, customer behavior library and recommended models 113 are additionally provided with, if including referring to for existing
The customer's set for determining moving target obtains then in customer's set where designated movement target in addition to designated movement target
The behavioral data of other moving targets;According to the behavioral data of other moving targets, recommendation results are determined;If there is no include referring to
The customer's set for determining moving target, then using hot item as recommendation results, hot item includes at least sale in specified time
Commodity of the amount more than first threshold.In addition, customer behavior library and recommended models 113, are additionally operable to other moving target depth is clear
The commodity look at are determined as recommendation results, and the commodity of deep browsing include at least number of visits in preset time and are more than second threshold
Commodity, number on probation are more than the commodity of third threshold value, customer's accumulated dwelling time more than any one in the commodity of the 4th threshold value
.Finally, in fig. 2, the personalized recommendation list display module 151 being located in display module 15, for pushing away recommendation results
It send to designated terminal and shows, designated terminal includes at least the terminal that shoppers' guide uses.
It should be noted that above-mentioned Fig. 2 and shown in Fig. 3, the two parts for belonging to recommendation apparatus under same line can be only
It erects and sets on different devices, or be arranged on the same device by the way of integrated.Wherein, if being arranged in same equipment
On, then can be considered as the module that this two parts has overlapping can realize function needed for above-mentioned two parts.For shown in Fig. 3
For another part, main function is to establish customer behavior library, also means that, in order to collect sufficient data, needs
Ensure to correspond between camera module 13 and preprocessing module 10, and is at least set in shopper device under same line
Camera module 13 and a preprocessing module 10 there are one setting.And for it is shown in FIG. 1 for realizing customer behavior prediction and
For the part of shopping guide's recommendation function, for the object using unique moving target as recommendation service, therefore, it is necessary to single
For preprocessing module, the processes such as entire target detection, tracking, duplicate removal, prediction and follow-up displaying are realized.
Shopping guide method under a kind of line of offer of the embodiment of the present invention, as shown in figure 4, this method specifically can be just like Fig. 2 and Fig. 3
Shown under line recommendation apparatus realize that this method specifically includes:
In S1, the video sequence acquired in specified region, designated movement target is identified and tracked, designated movement mesh is obtained
Target figure information.
Wherein, figure information is used to indicate the action that the designated movement target in video sequence generates in the process in tracking.
Specified region can be merchandising location in functional area, for example, around cashier, fitting room nearby, showcase institute
Can also be to easily identify and track designated movement mesh target area in merchandising location, for example, each in merchandising location in region etc.
A shops is close to relatively small region of population density etc. in the region, merchandising location at gate.It should be noted that above-mentioned work(
Energy region can collect effective designated movement target in order to camera, also, past in the collected information of functional area institute
Toward can reflect the information such as the purchase trend of customer, purchasing demand, the price that can bear;The above-mentioned region convenient for recognition and tracking
Often can be smaller due to the density of population or closer apart from camera, to improve accuracy of identification and tracking accuracy rate.Also
It is meant that the above-mentioned specified region referred to can provide certain convenient item for the acquisition of camera, identification and follow-up tracking, analysis
Part.In embodiments of the present invention, specified region is not limited only to above-mentioned zone, can also include other regions, not limit herein.
Specifically, in the present embodiment track designated movement target, and obtain designated movement target figure information it is specific
Technology, the technical solution applied at present that may be used, such as:Have been applied in the progress of the public places such as airport, station
The shadow tracking, identification and the stream of people statistics technology (this kind of technical solution applied, can realize the stream of people count
In the process, the body image in the image taken by multiple cameras is distinguished with background, and by different human bodies
The function that image mutually distinguishes).In the present embodiment then to identify designated movement target (specifically recognize each one
Body image) into line trace, and record action caused by designated movement target.
S2, from customer behavior library, according to the action of generation, the corresponding behavioral data of retrieval designated movement target.
It should be noted that customer behavior library may include at least one general behavior library and at least one personal behavior
Library.Wherein, general behavior library is mainly used for identifying strange customer, the behavior that personal behavior library is established primarily directed to frequent customer
Library, for those frequently occur on merchandising location each customer or customer set establish independent behavior library.
In view of the merchandising location volume of the flow of passengers is larger, therefore, the data in the customer behavior library established in the embodiment of the present invention
Amount is often larger.It, then can be according to data for general behavior library in order to reduce the memory space that customer behavior library occupies
Generation time is stored, and in order to will subsequently be deleted apart from current time more long behavioral data, is deposited with vacateing depositor space
Store up the behavioral data of current time generation.So, can not only periodicity or the general behavior library of real-time implementation more
Newly, simultaneously, it is ensured that the management of data is orderly.And for personal behavior library, if wanting to be further reduced personal behavior library
The memory space occupied then needs one parameter for being similar to time threshold of setting, to control the reservation, clear in personal behavior library
It is empty or with the partial data pointedly emptied in personal behavior library.For example, same customer is directed to, if the customer is close
It is not again appeared in the merchandising location in two weeks, then can delete the behavior number that the customer in personal behavior library has generated
According to so, if the customer again appears at merchandising location, the behavior number described in general behavior library may be used
According to being analyzed the customer as strange customer.
In addition, for convenience to designated movement target into line trace, the figure information for recognizing designated movement target it
Afterwards, each designated movement target can be given to add label, using the mark as the different designated movement targets of differentiation.It later, can be with
The existing mark of direct basis carrys out the retrieval of consummatory behavior data.
The behavioral data that S3, basis retrieve generates the recommendation results of corresponding designated movement target.
The behavioral data retrieved can be one group or multigroup behavioral data, can be with during generating recommendation results
The similarities and differences of each behavioral data are effectively extracted, and appropriate recommendation results are generated for the similarities and differences of each behavioral data.It is pushed away in addition, generating
The mode for recommending result can also be the hot item etc. of the current each shops's brand of analysis, can also propose that a variety of generations are recommended hereinafter
As a result specific implementation is not specifically limited in embodiments of the present invention for the mode of generation recommendation results.
The embodiment of the present invention is mainly to be generated recommendation results using object, carried for customer for shoppers' guide with shoppers' guide
For more targeted recommendation service.In embodiments of the present invention, after identifying and tracking designated movement target, from customer behavior
The corresponding behavioral data of designated movement target is retrieved in library, and the recommendation of corresponding designated movement target is generated according to behavior data
As a result.It can be seen that being by locking designated movement target, and to the historical behavior of designated movement target in above-mentioned realization method
Data are analyzed, are counted, and obtain the recommendation results with the designated movement destination matches later.So, can be to lead
Purchase personnel provide the relevant information of abundant, sufficient Customer Resource and corresponding customer, and it is to care for also just to be more conducive to shoppers' guide
Visitor, which provides, has more targetedly recommendation service.Therefore, it solves under line commercial product recommending in sales process and services to bring to customer
Consumption is inconvenient.
In order to get effective behavioral data, in embodiments of the present invention, before stating S1 to S3 in realization, also
It needs to pre-establish customer behavior library, and the specific of customer behavior library establishes mode, it is specific as follows:
In each video sequence currently acquired, identifies and track all moving targets, later in all moving targets
Middle identification customer, and determine the behavioral data for belonging to customer.Wherein, behavioral data include at least customer's Evaluation product record,
Customer and other people interbehaviors, customer residence time in specified commodity location of browsing route, customer in it is arbitrary
One.
Specifically, in each video sequence currently acquired, background and movement are identified in taken image
Target, and customer is identified from moving target, such as:The uniform of employee's dressing uniform color in shop, or in body specific bit
(such as left chest) unified wearing chest card is set, whether there can be chest according to the chest of human body in the color and moving target of clothes
Board distinguishes customer and salesman, and in practical applications, the moving target in StoreFront region is customer substantially other than salesman,
The needs of actually identifying can be met by being distinguished using the dichotomy of customer/salesman, and also be saved to captured movement
Target carries out the cost of identity screening.
Analysis for moving target (especially customer) behavioral data, can first extract corresponding motion feature, adopt
Classified with advance trained grader.Such as using sequence image and corresponding light stream or motion vector image as input,
Feature vector is connected into extract characteristics of image, then by different images feature with a trained convolutional neural networks, is sent into
(such as chronicle recognition model based on Recognition with Recurrent Neural Network or shot and long term memory mould is identified in trained chronicle recognition model
Type).Wherein, in order to reduce operand, interval sampling can be carried out to original video image sequence, the image sequence that sampling is obtained
It is analyzed.
Can be following concrete condition it should be noted that for the type of enumerated behavioral data:Customer is on probation
The record of product can specifically include the type for the product that customer tries out, the phase of price, color and the several products being compared
Close information etc.;Customer can specifically be presented as duration, the conversation content etc. that customer engages in the dialogue with other people with other people interbehaviors;
The browsing route of customer may include complete video clip in customer be tracked generate track or customer in designated position
Stay the line etc. between multiple coordinate points that certain time is generated.
After determining above-mentioned behavioral data, the processes such as retrieval, analysis are realized in order to facilitate commending system under line, it is also necessary to
The behavioral data for belonging to customer is stored to customer behavior library.So, lower commending system needs query portion behavior online
It when data, can directly be searched from customer behavior library, later by analyzing, handling, corresponding recommendation results be obtained, for shopping guide
Personnel browse.
It should be noted that in above-mentioned storing process, all of customer can will be belonged to according to the content of behavioral data
Moving target is grouped, and is obtained at least one customer's set, is distinguished storage according to customer's set later.But once
Data volume is larger, then is likely to the case where there are intersections between multiple customers gather occur, then in order to reduce storing process
Complexity, the time sequencing that still can be generated according to data, stores successively.Customer behavior library is being accessed later, and according to Gu
During behavioral data described in objective behavior library generates recommendation results, then the moving target corresponding to behavioral data is carried out
Grouping obtains current desired customer's set, and for the corresponding all behavioral datas of this customer set, generates suitable later
It should be in the recommendation results of current customer.
In a realization method of the embodiment of the present invention, according to the behavioral data retrieved, corresponding designated movement is generated
The recommendation results of target, specific implementation during may include:If in the presence of the customer's set for including designated movement target,
In customer's set where designated movement target, the behavioral data of other moving targets in addition to designated movement target is obtained, and
According to the behavioral data of other moving targets, recommendation results are determined;If there is no the customer's set for including designated movement target,
Using hot item as recommendation results.Wherein, hot item includes at least the quotient that sales volume in specified time is more than first threshold
Product.
It can be seen that no matter commending system determines whether there is the set of the customer including designated movement target under line, all may be used
To generate the recommendation results of corresponding designated movement target.It is only directed to the case where there are customer's set, then needs analysis entire
The factors such as real consumption trend, real consumption demand of most of customer in customer's set, to estimate out currently assigned movement
The factors such as the possibility propensity to consume of target, consumption demand, and then obtain final recommendation results;And for corresponding customer is not present
The case where set, can select current hot item to give the designated movement target to provide more good recommendation service
As final recommendation results, recommend to the designated movement target for shoppers' guide.
It should be noted that hot item can be current, this season, when annual turnover is higher, i.e., sales volume is more than first
The commodity or sales volume of threshold value are in the commodity that sales volume is former in the merchandising location in a long time, or current leading
Merchandising, for example, most new product etc..Wherein, first threshold can be preset, and specific setting means is referred to current each
The sale ratio etc. of the total sales volume of a commodity or each commodity, does not limit herein.In embodiments of the present invention, for
The method of determination of hot item is not specifically limited, and the above situation is only several specific methods of determination, but is not limited only to this.
For the behavioral data according to other moving targets, recommendation results are determined, it in embodiments of the present invention, can be specific
It is embodied as:The commodity of other moving target deep browsings are determined as recommendation results.Wherein, the commodity of deep browsing include at least
Number of visits, which is more than the commodity of second threshold, in preset time, number on probation is more than the commodity of third threshold value, customer is accumulative stops
Time is more than any one in the commodity of the 4th threshold value.Wherein, second threshold, third threshold value and the 4th threshold value all can be through
Value is tested, is preset in commending system, specific setting means is similar with first threshold, and this will not be repeated here;Preset time
For the preset time parameter in commending system, for example, 2 hours etc..
It is in embodiments of the present invention, specified at this in order to allow recommendation results more to meet the demand of currently assigned moving target
Customer belonging to moving target gathers in corresponding behavioral data, can targetedly extract a part or total data, and
The behavioral data generated to all customers belonged in customer set is counted, is analyzed, and finds being total between each customer
Property or most of customer caused by identical behavioral data, be directed to this kind of data later and generate recommendation results.Such one
Come, is equivalent to and is extracted the multiple and currently assigned higher customer of moving target similarity behavioral data, and according to these rows
Recommendation results are generated for data, the recommendation results to make more are bonded the shopping need of the designated movement target.
In view of the acquisition limitation of camera in merchandising location, the either foundation in behavior library or subsequently to just walking
Enter the process that the customer in merchandising location is identified, often in order to improve processing accuracy, be required for, each of is currently acquiring
In video sequence, after identifying and tracking all moving targets, duplicate removal processing is carried out to all moving targets.
It should be noted that in merchandising location, two kinds of deployment way usually may be used in camera, for example, top set is pacified
Dress or strabismus installation.But which kind of either above-mentioned deployment way, inevitably be exactly in the same merchandising location,
In order to ensure channel, functional area near each shop and shop in entire merchandising location etc. all in the visible model of camera
In enclosing, therefore, generally require at least to dispose two cameras in merchandising location, for acquiring video clip.
In embodiments of the present invention, specific duplicate removal mode may be used existing, larger in volumes of the flow of passengers such as airport, stations
Region in, the duplicate removal mode for counting the volume of the flow of passengers.For the specific implementation of duplicate removal, can be identified by characteristics of human body
Etc. modes realize, be herein not specifically limited.
Recommendation service under targetedly line is provided for customer for the ease of shoppers' guide, in embodiments of the present invention,
After the recommendation results for generating corresponding designated movement target, it is also necessary to recommendation results be pushed to designated terminal and shown.Its
In, designated terminal includes at least the terminal that shoppers' guide uses.
It should be noted that recommendation results may include the essential information of customer, for example, the letters such as the name of customer, age
Breath, these information can customer's bankcard consumption or registration, using the member card of merchandising location when obtain together.Such one
Come, shoppers' guide can be facilitated to grasp the basic condition of customer as early as possible, so that it is determined that the consumption ability to bear of customer, for example, leading
Purchase personnel determine the commodity that the range of age where customer usually requires, such as mother and baby's product according to the age of customer;According to customer
Job category, the general income situation of customer is inferred to, to analyze the spending amount etc. that customer may bear.It needs
Bright, above-mentioned analytic process can be analyzed by shoppers' guide according to the essential information of presentation and the experience of oneself, or
After completing analytic process by commending system under line, directly present on the terminal.
Recommendation results can also include needing to information such as the types of merchandize, color, size of the customer recommendation.Above- mentioned information
It can not only be determined by the behavioral data in behavior library, it can also be according to the purchase situation of customer, dressing in historical time
Situation etc. is analyzed, is determined, is not limited herein.
So, when customer will appear in shoppers' guide region, shoppers' guide can understand this in advance
The demand of customer, to provide more good recommendation service for the customer.In addition, the customer for leaving a upper sales region
For, the shoppers' guide of next sales region can contact with into administrative staff as early as possible, reduce customer and be in self-selection for a long time
Process can also promote the desire to purchase of customer to a certain extent in this way, even if customer is there are one good shopping environment
In the case of, increase the desire for consumer goods.
Each embodiment in this specification is described in a progressive manner, identical similar portion between each embodiment
Point just to refer each other, and each embodiment focuses on the differences from other embodiments.Especially for equipment reality
For applying example, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to embodiment of the method
Part explanation.The above description is merely a specific embodiment, but protection scope of the present invention is not limited to
This, any one skilled in the art in the technical scope disclosed by the present invention, the variation that can readily occur in or replaces
It changes, should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with the protection model of claim
Subject to enclosing.
Claims (14)
1. shopping guide method under a kind of line, which is characterized in that the method includes:
In the video sequence acquired in specified region, designated movement target is identified and tracked, the designated movement target is obtained
Figure information, the figure information is used to indicate what designated movement target in the video sequence generated in the process in tracking
Action;
From customer behavior library, according to the action of generation, the corresponding behavioral data of the designated movement target is retrieved;
According to the behavioral data retrieved, the recommendation results of the corresponding designated movement target are generated.
2. according to the method described in claim 1, it is characterized in that, the method further includes:
In each video sequence currently acquired, identifies and track all moving targets;
Customer is identified in all moving targets, and determines the behavioral data for belonging to customer, wherein behavioral data, which includes at least, to be cared for
The record of objective Evaluation product, customer are with other people interbehavior, the browsing route of customer, customers in specified commodity location
Any one in residence time.
3. according to the method described in claim 2, it is characterized in that, the method further includes:
The behavioral data for belonging to customer is stored to the customer behavior library.
4. according to the method described in claim 3, it is characterized in that, identify customer in all moving targets described, and really
Surely belong to after the behavioral data of customer, the method further includes:
According to the content of the behavioral data, all moving targets for belonging to customer are grouped, at least one customer is obtained
Set.
5. according to the method described in claim 4, it is characterized in that, the behavioral data that the basis retrieves, generates corresponding institute
The recommendation results for stating designated movement target, specifically include:
If in the presence of the customer's set for including the designated movement target, in customer's set where the designated movement target,
Obtain the behavioral data of other moving targets in addition to the designated movement target;
According to the behavioral data of other moving targets, the recommendation results are determined;
If there is no the customer's set for including the designated movement target, using hot item as recommendation results, the fast sale
Commodity include at least the commodity that sales volume in specified time is more than first threshold.
6. according to the method described in claim 5, it is characterized in that, the behavioral data of other moving targets described in the basis,
It determines the recommendation results, specifically includes:
The commodity of other moving target deep browsings are determined as the recommendation results, the commodity of the deep browsing are at least
It is more than the commodity of second threshold including number of visits in preset time, number on probation is more than the commodity of third threshold value, customer is accumulative
Residence time is more than any one in the commodity of the 4th threshold value.
7. method as claimed in any of claims 2 to 6, which is characterized in that described to regard each of currently acquiring
In frequency sequence, after identifying and tracking all moving targets, the method further includes:
Duplicate removal processing is carried out to all moving targets.
8. method as claimed in any of claims 1 to 6, which is characterized in that generating the corresponding specified fortune
After the recommendation results of moving-target, the method further includes:
The recommendation results are pushed into designated terminal and are shown, the designated terminal includes at least the end that shoppers' guide uses
End.
9. shopper device under a kind of line, which is characterized in that described device includes:
Preprocessing module, in specifying the video sequence acquired in region, identifying and tracking designated movement target, obtain institute
State the figure information of designated movement target, the figure information be used to indicate the designated movement target in the video sequence with
The action generated during track;
Data and model module, for from customer behavior library, according to the action of generation, retrieving the preprocessing module identification simultaneously
Track the obtained corresponding behavioral data of the designated movement target;
The data and model module, for according to the behavioral data retrieved, generating pushing away for the corresponding designated movement target
Recommend result.
10. device according to claim 9, which is characterized in that described device further includes:
The preprocessing module is additionally operable in each video sequence currently acquired, is identified and is tracked all moving targets;
Macro or mass analysis module for identifying customer in all moving targets, and determines the behavioral data for belonging to customer, wherein
Behavioral data includes at least the record of customer's Evaluation product, customer and other people interbehaviors, the browsing route of customer, customer and is referring to
Determine any one in the residence time in commodity location;
The Macro or mass analysis module is additionally operable to store the behavioral data for belonging to customer to the customer behavior library;
The Macro or mass analysis module, is additionally operable to the content according to the behavioral data, will belong to all moving targets of customer into
Row grouping obtains at least one customer's set.
11. device according to claim 10, which is characterized in that the data and model module are specifically used for:
If in the presence of the customer's set for including the designated movement target, in customer's set where the designated movement target,
Obtain the behavioral data of other moving targets in addition to the designated movement target;
According to the behavioral data of other moving targets, the recommendation results are determined;
If there is no the customer's set for including the designated movement target, using hot item as recommendation results, the fast sale
Commodity include at least the commodity that sales volume in specified time is more than first threshold.
12. according to the devices described in claim 11, which is characterized in that the data and model module are specifically used for:
The commodity of other moving target deep browsings are determined as the recommendation results, the commodity of the deep browsing are at least
It is more than the commodity of second threshold including number of visits in preset time, number on probation is more than the commodity of third threshold value, customer is accumulative
Residence time is more than any one in the commodity of the 4th threshold value.
13. the device according to any one of claim 10 to 12, which is characterized in that described device further includes:
Authentication module, for carrying out duplicate removal processing to all moving targets.
14. the device according to any one of claim 9 to 12, which is characterized in that described device further includes:
Display module, for the recommendation results to be pushed to designated terminal and are shown, the designated terminal includes at least shopping guide
The terminal that personnel use.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710041475.9A CN108335317A (en) | 2017-01-20 | 2017-01-20 | Shopping guide method and device under a kind of line |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710041475.9A CN108335317A (en) | 2017-01-20 | 2017-01-20 | Shopping guide method and device under a kind of line |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108335317A true CN108335317A (en) | 2018-07-27 |
Family
ID=62922831
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710041475.9A Pending CN108335317A (en) | 2017-01-20 | 2017-01-20 | Shopping guide method and device under a kind of line |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108335317A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109194931A (en) * | 2018-10-25 | 2019-01-11 | 哈尔滨海能达科技有限公司 | A kind of notification method and device of the target object based on video pictures |
CN110070148A (en) * | 2019-03-15 | 2019-07-30 | 北京木业邦科技有限公司 | A kind of forestry products characteristic analysis method, device and computer-readable medium |
CN110110688A (en) * | 2019-05-15 | 2019-08-09 | 联想(北京)有限公司 | A kind of information analysis method and system |
CN110264219A (en) * | 2019-05-06 | 2019-09-20 | 浙江华坤道威数据科技有限公司 | A kind of client's monitoring analysis system based on big data |
CN110348943A (en) * | 2019-05-31 | 2019-10-18 | 浙江口碑网络技术有限公司 | Processing method, device, storage medium and the computer equipment of commercial product recommending information |
CN110472993A (en) * | 2019-07-04 | 2019-11-19 | 人加智能机器人技术(北京)有限公司 | Information recommendation system and method |
CN110738782A (en) * | 2019-10-24 | 2020-01-31 | 名创优品(横琴)企业管理有限公司 | cashier queuing analysis method and system |
CN110753102A (en) * | 2019-10-15 | 2020-02-04 | 浙江口碑网络技术有限公司 | Service information pushing method and device based on geomagnetism |
CN111028029A (en) * | 2018-10-10 | 2020-04-17 | 深圳云天励飞技术有限公司 | Offline commodity recommendation method and device and electronic equipment |
CN112184331A (en) * | 2020-10-23 | 2021-01-05 | 北京爱笔科技有限公司 | People and goods association method and system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040128196A1 (en) * | 2002-09-19 | 2004-07-01 | Masatsugu Shibuno | One-to-one business support system and program for implementing the function of the system |
CN104573619A (en) * | 2014-07-25 | 2015-04-29 | 北京智膜科技有限公司 | Method and system for analyzing big data of intelligent advertisements based on face identification |
CN105678591A (en) * | 2016-02-29 | 2016-06-15 | 北京时代云英科技有限公司 | Video-analysis-based commercial intelligent operation decision-making support system and method |
-
2017
- 2017-01-20 CN CN201710041475.9A patent/CN108335317A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040128196A1 (en) * | 2002-09-19 | 2004-07-01 | Masatsugu Shibuno | One-to-one business support system and program for implementing the function of the system |
CN104573619A (en) * | 2014-07-25 | 2015-04-29 | 北京智膜科技有限公司 | Method and system for analyzing big data of intelligent advertisements based on face identification |
CN105678591A (en) * | 2016-02-29 | 2016-06-15 | 北京时代云英科技有限公司 | Video-analysis-based commercial intelligent operation decision-making support system and method |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111028029A (en) * | 2018-10-10 | 2020-04-17 | 深圳云天励飞技术有限公司 | Offline commodity recommendation method and device and electronic equipment |
CN111028029B (en) * | 2018-10-10 | 2023-09-01 | 深圳云天励飞技术有限公司 | Off-line commodity recommendation method and device and electronic equipment |
CN109194931A (en) * | 2018-10-25 | 2019-01-11 | 哈尔滨海能达科技有限公司 | A kind of notification method and device of the target object based on video pictures |
CN110070148A (en) * | 2019-03-15 | 2019-07-30 | 北京木业邦科技有限公司 | A kind of forestry products characteristic analysis method, device and computer-readable medium |
CN110264219A (en) * | 2019-05-06 | 2019-09-20 | 浙江华坤道威数据科技有限公司 | A kind of client's monitoring analysis system based on big data |
CN110264219B (en) * | 2019-05-06 | 2021-11-30 | 浙江华坤道威数据科技有限公司 | Customer monitoring and analyzing system based on big data |
CN110110688A (en) * | 2019-05-15 | 2019-08-09 | 联想(北京)有限公司 | A kind of information analysis method and system |
CN110348943A (en) * | 2019-05-31 | 2019-10-18 | 浙江口碑网络技术有限公司 | Processing method, device, storage medium and the computer equipment of commercial product recommending information |
CN110472993A (en) * | 2019-07-04 | 2019-11-19 | 人加智能机器人技术(北京)有限公司 | Information recommendation system and method |
CN110753102A (en) * | 2019-10-15 | 2020-02-04 | 浙江口碑网络技术有限公司 | Service information pushing method and device based on geomagnetism |
CN110738782A (en) * | 2019-10-24 | 2020-01-31 | 名创优品(横琴)企业管理有限公司 | cashier queuing analysis method and system |
CN112184331A (en) * | 2020-10-23 | 2021-01-05 | 北京爱笔科技有限公司 | People and goods association method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108335317A (en) | Shopping guide method and device under a kind of line | |
CN106776619B (en) | Method and device for determining attribute information of target object | |
CN109558535B (en) | Personalized article pushing method and system based on face recognition | |
US10713670B1 (en) | Method and system for finding correspondence between point-of-sale data and customer behavior data | |
CN109165992A (en) | A kind of intelligent shopping guide method, apparatus, system and computer storage medium | |
CN108830251A (en) | Information correlation method, device and system | |
KR101779096B1 (en) | The object pursuit way in the integration store management system of the intelligent type image analysis technology-based | |
CN106164959A (en) | Behavior affair system and correlation technique | |
CN110033298A (en) | Information processing equipment and its control method, system and storage medium | |
CN102122346A (en) | Video analysis-based physical storefront customer interest point acquisition method | |
CN109993595A (en) | Method, system and the equipment of personalized recommendation goods and services | |
CA3014365C (en) | System and method for gathering data related to quality of service in a customer service environment | |
KR20170060740A (en) | Offline Fashion-sales Active System | |
CN108596730A (en) | Processing method, device and the smart machine of dress ornament information | |
CN107871111A (en) | A kind of behavior analysis method and system | |
JP2019020986A (en) | Human flow analysis method, human flow analysis device, and human flow analysis system | |
AU2017231602A1 (en) | Method and system for visitor tracking at a POS area | |
JP6593949B1 (en) | Information processing apparatus and marketing activity support apparatus | |
KR102077630B1 (en) | System and method for analyzing commercial based on pos and video | |
US20150365390A1 (en) | Method of creating preference image identification code, method of diagnosing preference image identification code, and method of providing information using preference image identification code | |
CN113706227A (en) | Goods shelf commodity recommendation method and device | |
CN113887884A (en) | Business-super service system | |
JP2017130061A (en) | Image processing system, image processing method and program | |
US11615430B1 (en) | Method and system for measuring in-store location effectiveness based on shopper response and behavior analysis | |
Kröckel et al. | Customer tracking and tracing data as a basis for service innovations at the point of sale |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180727 |