CN113674037B - Data acquisition and recommendation method based on shopping behaviors - Google Patents
Data acquisition and recommendation method based on shopping behaviors Download PDFInfo
- Publication number
- CN113674037B CN113674037B CN202111226921.6A CN202111226921A CN113674037B CN 113674037 B CN113674037 B CN 113674037B CN 202111226921 A CN202111226921 A CN 202111226921A CN 113674037 B CN113674037 B CN 113674037B
- Authority
- CN
- China
- Prior art keywords
- data
- customer
- shopping
- intelligent
- 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.)
- Active
Links
- 230000006399 behavior Effects 0.000 title claims abstract description 102
- 238000000034 method Methods 0.000 title claims abstract description 69
- 230000008569 process Effects 0.000 claims abstract description 17
- 230000002159 abnormal effect Effects 0.000 claims description 31
- 238000001514 detection method Methods 0.000 claims description 27
- 230000002265 prevention Effects 0.000 claims description 16
- 238000012544 monitoring process Methods 0.000 claims description 15
- 238000004458 analytical method Methods 0.000 claims description 13
- 230000000007 visual effect Effects 0.000 claims description 13
- 238000010195 expression analysis Methods 0.000 claims description 11
- 230000001815 facial effect Effects 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 9
- 230000007246 mechanism Effects 0.000 claims description 7
- 230000002452 interceptive effect Effects 0.000 claims description 6
- 230000002093 peripheral effect Effects 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000005728 strengthening Methods 0.000 claims description 4
- 238000005315 distribution function Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000002123 temporal effect Effects 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims 2
- 230000009286 beneficial effect Effects 0.000 description 12
- 238000012549 training Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 3
- 230000006378 damage Effects 0.000 description 3
- 238000013480 data collection Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000008921 facial expression Effects 0.000 description 2
- 230000036544 posture Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- 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
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a data acquisition and recommendation method based on shopping behaviors, which comprises the steps of acquiring a face image of a customer through a preset intelligent camera, and acquiring shopping behavior data through the face image; the intelligent camera comprises a front camera, a rear camera and a side camera; the method comprises the steps that commodity data in the shopping process are collected through a preset intelligent device, meanwhile, the intelligent device is located, and information of an interest area is determined; and generating marketing feedback data through the shopping behavior data and the information of the interest area, constructing a recommendation strategy based on the marketing feedback data, and pushing the recommendation strategy to intelligent equipment.
Description
Technical Field
The invention relates to the technical field of artificial intelligence and data acquisition, in particular to a data acquisition and recommendation method based on shopping behaviors.
Background
As a new economic society development form, digital economy profoundly influences various industries, and with the combination of increasingly prosperous digital intelligence and industrialization, the accumulated dividends are continuously emerging. For the retail industry, under such a "new and old interweaving, coexistence and destruction" form, comprehensive digital transformation has become a great trend choice, and is a core engine which needs to be pushed for realizing capital increment. Under the impact of epidemic situations, enterprises have more challenges, want to break through business boundaries, expand and open new cycles, and definitely need to overturn places, so that the necessity of digital transformation of traditional retail is clear, and the traditional retail suffers from little attack influence since the rise of the merchants. The purchase potential of the users of the channels on the internal line of business is urgently mined, and the importance is self-evident. The traditional marketing plan of the retail industry is changed from current time to current time, and in the life with extremely high internet coverage, the lack of high-frequency and high-coverage marketing makes it difficult to establish the sense of existence on the user psychological level, let alone to cultivate the faithful users. The method brings the results that the repurchase rate is difficult to promote, the new user expansion is not enough, the traditional retail industry is difficult to generate the effects of cultivation and guidance on users, and the data insight is far from shortage. There are great difficulties in data acquisition, processing and final application. The collection of the relevant data of the offline distribution channel users is complex, the data quality and the fineness are not enough, and the user portrait is difficult to analyze. The traditional retail business operation mode tends to be stable, and the thin profit and the multiple marketing are the way for the survival. Under the pinch of the economic impact caused by the serious interception of the online retail business state and the epidemic situation of the traditional retail, the offline flow is divided by multiple parties, and the sales volume is reduced; the rent and labor cost of stores are rising continuously, the traditional profit is spread out, and the total cost is rising inevitably.
Therefore, how to collect and coil offline data to achieve accurate and effective operation is an urgent problem to be solved offline. However, some off-line data acquisition applications cannot form a complete data link, and further cannot realize instant personalized marketing.
Disclosure of Invention
The invention provides a data acquisition and recommendation method based on shopping behaviors, and aims to solve the problems by an intelligent shopping algorithm which integrates intelligent acquisition of full-period data in the shopping process of a customer and personalized recommendation to the customer based on user image/shopping decision and the like.
The invention provides a data acquisition and recommendation method based on shopping behaviors, which is characterized by comprising the following steps:
the method comprises the steps that a preset intelligent camera is used for collecting a face image of a customer, and shopping behavior data are obtained through the face image; wherein,
the intelligent camera comprises a front camera, a rear camera and a side camera;
the method comprises the steps that commodity data in the shopping process are collected through a preset intelligent device, meanwhile, the intelligent device is located, and information of an interest area is determined;
and generating marketing feedback data through the shopping behavior data and the interest area information, constructing a recommendation strategy based on the marketing feedback data, and pushing the recommendation strategy to intelligent equipment.
As an embodiment of this technical solution, the acquiring a face image of a customer through a preset intelligent camera, and acquiring shopping behavior data through the face image includes:
the method comprises the steps that through a front camera, portrait acquisition is conducted on a customer, a face image of the customer is recorded based on a preset face detection algorithm, and identity information of the customer is recognized; performing expression analysis on the facial image of the customer in real time through the identity information of the customer and the corresponding facial image to determine an expression analysis result;
and evaluating the shopping behavior of the customer according to the expression analysis result, and determining the shopping behavior data of the customer.
As an embodiment of the present technical solution, the recording a face image of a customer based on a preset face detection algorithm, and identifying customer identity information includes:
acquiring a face image of a face of a customer;
recording and extracting target rectangular features of the face image based on a preset Adaboost algorithm, and generating weak separator features according to the target rectangular features;
strengthening the weak separator characteristics according to a preset weighted voting mechanism, and constructing strong classifier characteristics;
connecting the strong classifier features in series to generate a laminated classifier feature, identifying the face feature through the laminated classifier feature, and determining a face identification image;
based on a preset big data center, carrying out preprocessing operation on the face recognition image to determine a preprocessed image; wherein,
the preprocessing operation at least comprises one or more of light compensation, gray scale transformation, histogram equalization, normalization, geometric correction, filtering and sharpening;
extracting geometric features of the preprocessed image, extracting face feature points, modeling face features through the face feature points, and building a face feature model;
extracting face feature data in a face image to be recognized based on the face feature model;
and searching and matching the face feature data with a face feature template in a preset storage database, and identifying the identity information of the customer.
As an embodiment of the present technical solution, the extracting geometric features of the preprocessed image, extracting face feature points, and performing face feature modeling through the face feature points to construct a face feature model includes:
extracting human face features based on a preset geometric feature algorithm, and determining local geometric lines of the human face; wherein,
the geometric feature algorithm is used for identifying line changes of the eyes, the nose, the mouth and the chin;
analyzing the local shape and distance characteristics of the geometric lines, determining face classification characteristic data, and extracting face characteristic points according to the expression face characteristic data;
generating a corresponding face characteristic component according to the face characteristic points; wherein,
the face feature component at least comprises Euclidean distances, curvatures and angles among face feature points;
and constructing a human face feature model through the human face feature components, the human face feature points and the human face classification feature data.
As an embodiment of this technical solution, the collecting, by a preset smart device, commodity data in a shopping process further includes:
acquiring commodity bar code information through an intelligent terminal PAD preset on an intelligent shopping cart;
acquiring commodity data through a weight sensor and an intelligent camera preset in the intelligent shopping cart; wherein,
the commodity data comprises commodity type information, commodity putting weight and commodity putting time;
according to the commodity bar code information, the intelligent shopping cart detects corresponding commodity data and the commodity putting weight, judges whether the commodity putting weight is in accordance with the standard weight corresponding to the commodity bar code information or not, and determines a judgment result;
when the judgment result is in accordance with the judgment result, the intelligent shopping cart automatically adds the corresponding code scanning commodity to a shopping list;
and when the judgment result is inconsistent, monitoring and judging abnormal information of the commodities in the intelligent shopping cart based on a rear camera arranged on the intelligent shopping cart in advance, and determining a first judgment result.
As an embodiment of the present technical solution, when the determination result is inconsistent, monitoring and determining abnormal information of the commodity in the intelligent shopping cart based on a rear camera pre-installed on the intelligent shopping cart, and determining the first determination result includes:
when the judgment result is not in accordance with the judgment result;
monitoring commodities in the intelligent shopping cart based on a rear camera of the intelligent shopping cart, and acquiring the quantity of the commodities through a weight sensor of the intelligent shopping cart;
when the commodity bar code information is correct but does not accord with the corresponding commodity number, uploading the commodity number, simultaneously carrying out voice reminding on a customer, and generating abnormal information and sending the abnormal information to a preset control terminal after the reminding time exceeds a preset time threshold;
when the commodity bar code information and the corresponding commodity number are wrong, acquiring loss prevention information of the commodity;
acquiring a basket image before code scanning, a basket image before commodity putting and a basket image after the commodity putting through a rear camera, a code scanner and a weight sensor which are preset on the intelligent shopping cart;
comparing the basket image before code scanning, the basket image before commodity placing and the basket image after commodity placing, analyzing the commodity placed in the basket by the customer without code scanning, and determining the commodity without code scanning;
the abnormal commodity is determined through the loss prevention information and the unscanned code commodity, voice reminding is carried out on the customer, and after the reminding time exceeds a preset time threshold value, abnormal information is generated and sent to a preset control terminal.
As an embodiment of the present technical solution, the positioning the smart device and determining the information of the interest area at the same time includes:
the method comprises the steps that a position sensor device preset in an intelligent terminal PAD is used for collecting the motion state and direction of an intelligent shopping cart in real time, and the staying and browsing time of a customer is obtained; wherein,
the position sensor device comprises at least a triangular gyroscope, an accelerometer and a level meter;
detecting a beacon preset in a peripheral area through a Bluetooth device preset in an intelligent terminal PAD, and identifying the area position of an intelligent shopping cart through the beacon;
acquiring and recording a shopping path of a customer according to the staying and browsing time, the area position and the staying duration, and analyzing a customer hot area and a customer flow line of a store through the shopping path;
and determining the interest area information of the customer through the customer hot area and the customer flow line.
As an embodiment of the present technical solution, the generating marketing feedback data through the shopping behavior data and the information of the area of interest, and constructing a recommendation policy based on the marketing feedback data, and meanwhile, pushing the recommendation policy to an intelligent device includes:
generating marketing feedback data according to the shopping behavior data and the interest area information, and generating corresponding push content according to the marketing feedback data;
acquiring a scene video on the intelligent shopping cart based on a preset marketing opportunity and a preset visual detection behavior data method, and acquiring an acquired customer image by acquiring the scene video;
by the visual detection behavior data method, the customer behavior data in the image is analyzed and extracted, the customer behavior data is processed and analyzed and the semantics is extracted, the analysis result is determined,
performing intelligent learning through an analysis result to generate customized push content, pushing the customized push content to an intelligent terminal PAD, and determining a push result;
and acquiring interactive information in a pushing result in real time, acquiring behavior feedback data when a customer receives the interactive information in real time through a front-facing camera, generating an individual recommendation strategy, pushing the individual recommendation strategy and storing the individual recommendation strategy on intelligent equipment.
As an embodiment of the present technical solution, the generating marketing feedback data according to the shopping behavior data and the information of the area of interest, and generating corresponding push content according to the marketing feedback data further includes:
step 1: acquiring user information of a target customer;
step 2: obtaining historical shopping behavior data through the user information, and generating personalized shopping behavior data according to the historical shopping behavior data
(ii) a Wherein,
and step 3: counting and recording the personalized shopping behavior data based on a preset big data center to generate interest group consumption data;
wherein,
a consumption function representing the interest group is provided,
representing the consumption data of the interest group after the personalized shopping behavior data are counted,
representing statistical personalized shopping behaviorThe distribution function of the data population is,
representing a screening function for interestingness;
and 4, step 4: constructing an interest recommendation algorithm according to the interest group consumption data;
wherein,
a recommendation algorithm is represented for the interest, and,
in respect of
An interest recommendation algorithm for the batch personalized shopping data,
，
representing the total number of batches of personalized shopping data,
represents to the first
The consumption data of interest groups after the batch personalized shopping data statistics,
representing a recommendation function regarding consumption data of a group of interest,
representing about a target user
The transpose of the forward interest weight matrix of the temporal interest group consumption data,
a forward interest weight matrix representing consumption data about the interest group,
a negative interest weight matrix representing consumption data about the interest group,
effective parameters representing positive interest weight and negative interest weight consumed by interest groups;
and 5: and obtaining the interested commodities through the interest recommendation algorithm, receiving marketing feedback data, and generating corresponding push contents through the interested commodities and the marketing feedback data.
As an embodiment of the technical scheme, the intelligent device further comprises an outlet all-in-one machine; wherein,
the exit all-in-one machine is used for detecting an RFID beacon on the intelligent shopping cart based on a preset RFID identification mechanism, identifying a target vehicle, detecting whether an order is abnormal or not by collecting shopping data of the target vehicle, and determining a detection result; wherein,
and when the detection result is that the order is abnormal, the abnormal order is closed, the channel outlet cannot be opened, the vehicle number is broadcasted in a voice mode to inform an outlet loss prevention worker, meanwhile, commodity loss prevention data and management and inspection data of the order are obtained, and visual order information is generated.
The invention has the following beneficial effects:
according to the technical scheme, a preset intelligent camera is used for collecting a face image of a customer, and shopping behavior data are obtained through the face image; the intelligent camera comprises a front camera, a rear basket monitoring camera and a side camera; through the preset intelligent device, commodity data in the shopping process are collected, meanwhile, the intelligent device is located, interest area information is determined, the intelligent device at least comprises an intelligent terminal PAD and an intelligent shopping cart, and the intelligent shopping cart accompanies a customer in the whole process, so that the intelligent device has the natural data collection capacity. The method comprises the steps that intelligent elements on an intelligent shopping cart are utilized to collect members, commodities, user behaviors and position information, so that the shopping data of a customer in a store can be completely recorded, collecting hardware devices in the intelligent shopping cart at least comprise a camera, a code scanner, a weight sensor, an RFID reader, a gyroscope and an accelerometer, the behavior data of the customer can be collected by the whole intelligent shopping cart based on a data collecting method of visual detection, mobile positioning, code scanning laser identification, weight sensing and screen touch, marketing feedback data are generated through the shopping behavior data and information of interest areas, a recommendation strategy is determined based on the marketing feedback data, and the recommendation strategy is pushed to intelligent equipment.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a flow chart of a method for data collection and recommendation based on shopping behaviors in an embodiment of the present invention;
FIG. 2 is a flowchart of a shopping behavior data acquisition method based on a shopping behavior data acquisition and recommendation method according to an embodiment of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
It will be understood that when an element is referred to as being "secured to" or "disposed on" another element, it can be directly on the other element or be indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly or indirectly connected to the other element.
It will be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like, as used herein, refer to an orientation or positional relationship indicated in the drawings that is solely for the purpose of facilitating the description and simplifying the description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and is therefore not to be construed as limiting the invention.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
According to the embodiment of the invention, as shown in fig. 1, a data acquisition and recommendation method based on shopping behaviors is provided, which is characterized by comprising the following steps:
the method comprises the steps that a preset intelligent camera is used for collecting a face image of a customer, and shopping behavior data are obtained through the face image; wherein,
the intelligent camera comprises a front camera, a rear camera and a side camera;
the method comprises the steps that commodity data in the shopping process are collected through a preset intelligent device, meanwhile, the intelligent device is located, and information of an interest area is determined;
and generating marketing feedback data through the shopping behavior data and the interest area information, constructing a recommendation strategy based on the marketing feedback data, and pushing the recommendation strategy to intelligent equipment.
The working principle and the beneficial effects of the technical scheme are as follows:
acquiring a face image of a customer through a preset intelligent camera, and acquiring shopping behavior data through the face image; the intelligent camera comprises a front camera, a rear basket monitoring camera and a side camera; through the preset intelligent device, commodity data in the shopping process are collected, meanwhile, the intelligent device is located, interest area information is determined, the intelligent device at least comprises an intelligent terminal PAD and an intelligent shopping cart, and the intelligent shopping cart accompanies a customer in the whole process, so that the intelligent device has the natural data collection capacity. The method comprises the steps that intelligent elements on an intelligent shopping cart are utilized to collect members, commodities, user behaviors and position information, so that the shopping data of a customer in a store can be completely recorded, collecting hardware devices in the intelligent shopping cart at least comprise a camera, a code scanner, a weight sensor, an RFID reader, a gyroscope and an accelerometer, the behavior data of the customer can be collected by the whole intelligent shopping cart based on a data collecting method of visual detection, mobile positioning, code scanning laser identification, weight sensing and screen touch, marketing feedback data are generated through the shopping behavior data and information of interest areas, a recommendation strategy is determined based on the marketing feedback data, and the recommendation strategy is pushed to intelligent equipment.
In one embodiment, the acquiring, by a preset intelligent camera, a face image of a customer, and acquiring shopping behavior data by the face image includes:
the method comprises the steps that through a front camera, portrait acquisition is conducted on a customer, a face image of the customer is recorded based on a preset face detection algorithm, and identity information of the customer is recognized;
performing expression analysis on the facial image of the customer in real time through the identity information of the customer and the corresponding facial image to determine an expression analysis result;
and evaluating the shopping behavior of the customer according to the expression analysis result, and determining the shopping behavior data of the customer.
The working principle and the beneficial effects of the technical scheme are as follows:
through the leading camera of predetermineeing, draw a portrait to the customer and gather, there are three cameras on the intelligent shopping cart, leading camera detects the people's face, rearmounted camera monitoring basket, and the peripheral goods shelves of side camera discernment. And based on a preset face detection algorithm, recording a face image of the customer, wherein the front-facing camera is used for acquiring face data, the face data comprises user ID, age, gender, expression and the like, and the identification information of the customer is specifically applied to: after a customer enters a shopping cart and a preset event for starting shopping is triggered, a front face camera on a PAD (intelligent display terminal) is started, the face information of the customer is collected, a face detection algorithm is started at the same time, and the identity of the customer is identified and recorded; in the shopping process, the front face camera collects the commodities selected by the customer in real time, meanwhile, expression analysis of customized marketing content pushed on a screen is seen, expression analysis is carried out on the face image of the customer in real time through the identity information of the customer and the corresponding face image, an analysis result is obtained, the shopping behavior of the customer is evaluated, the shopping behavior data of the customer is determined, and the feedback of the customer on the marketing is analyzed.
In one embodiment, the recording of the facial image of the customer and the identification of the customer identity information based on a preset face detection algorithm includes:
acquiring a face image of a face of a customer;
recording and extracting target rectangular features of the face image based on a preset Adaboost algorithm, and generating weak separator features according to the target rectangular features;
strengthening the weak separator characteristics according to a preset weighted voting mechanism, and constructing strong classifier characteristics;
connecting the strong classifier features in series to generate a laminated classifier feature, identifying the face feature through the laminated classifier feature, and determining a face identification image;
based on a preset big data center, carrying out preprocessing operation on the face recognition image to determine a preprocessed image; wherein,
the preprocessing operation at least comprises one or more of light compensation, gray scale transformation, histogram equalization, normalization, geometric correction, filtering and sharpening;
extracting geometric features of the preprocessed image, extracting face feature points, modeling face features through the face feature points, and building a face feature model;
extracting face feature data in a face image to be recognized based on the face feature model;
and searching and matching the face feature data with a face feature template in a preset storage database, and identifying the identity information of the customer.
The working principle and the beneficial effects of the technical scheme are as follows:
acquiring customer identity information, receiving a face image or a video stream continuously input by a camera by a program based on a preset Adaboost algorithm, selecting some rectangular features which can represent the face most in each frame by using the Adaboost algorithm, recording and extracting target rectangular features of the face image, and determining the features of a weak separator according to the target rectangular features; the weak separator features are used for carrying out primary recognition on the face features of the customer; strengthening weak separator characteristics according to a preset weighted voting mechanism, and constructing strong classifier characteristics; the weak separator features are used for accurately identifying the face features of the customer; constructing the weak classifier into a strong classifier according to a weighted voting mode, and then connecting a plurality of strong classifiers obtained by training in series to form a cascade-structured stacked classifier, so that the detection speed of the classifier is effectively improved, and finally, the facial expression is confirmed and recognized; because the original image cannot be directly used due to the limitation of various conditions and random interference, such as noise, illumination and the like, the image quality is reduced, the identification effect is influenced, and the situation identification image is subjected to preprocessing operations based on a preset big data center, wherein the preprocessing operations at least comprise light compensation, gray level transformation, histogram equalization, normalization, geometric correction, filtering and sharpening to complete preprocessing of the image, so that the image quality is enhanced.
In one embodiment, the extracting geometric features of the preprocessed image, extracting face feature points, and performing face feature modeling through the face feature points to construct a face feature model includes:
extracting human face features based on a preset geometric feature algorithm, and determining local geometric lines of the human face; wherein,
the geometric feature algorithm is used for identifying line changes of the eyes, the nose, the mouth and the chin;
analyzing the local shape and distance characteristics of the geometric lines, determining face classification characteristic data, and extracting face characteristic points according to the expression face characteristic data;
generating a corresponding face characteristic component according to the face characteristic points; wherein,
the face feature component at least comprises Euclidean distances, curvatures and angles among face feature points;
and constructing a human face feature model through the human face feature components, the human face feature points and the human face classification feature data.
The working principle and the beneficial effects of the technical scheme are as follows:
the technical scheme adopts a geometric feature method to extract human face features for modeling. The facial features are formed by combining local changes such as changes of eyes, a nose, a mouth and a chin, characteristic data of facial expression classification is obtained according to expression shape description of each local part of a face and distance characteristics among the local part of the face, and characteristic components of the characteristic data comprise Euclidean distances among characteristic points, curvatures, angles and the like. These facial feature data are modeled to form a geometric description of the face. The shooting posture of the face affects the appearance of the face image because the face is three-dimensional, and the image is two-dimensional, and different information of the face is shot by the image in different postures. The method comprises the steps of performing cross-posture face recognition, performing image preprocessing, recovering a front face image through three-dimensional modeling, extracting geometric features of the preprocessed image, performing face feature modeling, extracting face feature data in the face image to be recognized, and obtaining feature data of face classification according to shape description of each part of a face and distance characteristics among the parts of the face, wherein feature components comprise Euclidean distances, curvatures, angles and the like among feature points. Searching and matching the face feature data with a face feature template in a preset storage database, searching and matching the acquired feature data of the face image to be recognized with the acquired face feature template stored in the database, and outputting a result obtained by matching when the similarity exceeds a threshold value by setting the threshold value.
In one embodiment, the collecting, by a preset smart device, commodity data in a shopping process further includes:
acquiring commodity bar code information through an intelligent terminal PAD preset on an intelligent shopping cart;
acquiring commodity data through a weight sensor and an intelligent camera preset in the intelligent shopping cart; wherein,
the commodity data comprises commodity type information, commodity putting weight and commodity putting time;
according to the commodity bar code information, the intelligent shopping cart detects corresponding commodity data and the commodity putting weight, judges whether the commodity putting weight is in accordance with the standard weight corresponding to the commodity bar code information or not, and determines a judgment result;
when the judgment result is in accordance with the judgment result, the intelligent shopping cart automatically adds the corresponding code scanning commodity to a shopping list;
and when the judgment result is inconsistent, monitoring and judging abnormal information of the commodities in the intelligent shopping cart based on a rear camera arranged on the intelligent shopping cart in advance, and determining a first judgment result.
The working principle and the beneficial effects of the technical scheme are as follows:
acquiring commodity bar code information through a preset intelligent terminal PAD; the commodity data are collected through a preset weight sensor of the intelligent shopping cart, and the commodity data in the shopping process of a customer are collected by the aid of the weight sensor of the cart body, a commodity bar code recognizer on an intelligent terminal PAD and a rear basket monitoring camera. After a user takes goods from a goods shelf, the goods are scanned by a code scanner on an intelligent terminal PAD, then the goods are placed into a bicycle basket, the code scanner identifies a goods bar code, the purchasing action of the customer and the purchased goods can be acquired, the goods data comprises goods type information, goods weight and goods placing time, and whether the goods are placed abnormally is judged according to the goods bar code information and the goods data; according to the commodity bar code information, the intelligent shopping cart detects corresponding commodity data and determines the weight of the commodity; judging whether the commodity bar code information and the corresponding commodity weight are in accordance, and determining a first judgment result; when the judgment result is in accordance with the judgment result, the intelligent shopping cart automatically adds the corresponding code scanning commodity to the shopping list; and when the judgment result is inconsistent, monitoring the commodities in the intelligent shopping cart based on the rear camera of the intelligent shopping cart, judging the abnormal information of the commodities, and determining a second judgment result, wherein the intelligent shopping cart acquires the commodity data of the customer in the shopping process by utilizing a weight sensor of the cart body, a commodity bar code recognizer on the intelligent terminal PAD and the rear basket monitoring camera. After the user takes off goods from the goods shelves, can sweep the sign indicating number through the bar code scanner on the intelligent terminal PAD earlier, then put into the bicycle basket. The bar code of the commodity is identified by the bar code scanner, so that the purchasing action of the customer and the information of the purchased commodity can be acquired.
In one embodiment, when the determination result is non-compliance, monitoring and determining abnormal information of the goods in the intelligent shopping cart based on a rear camera of a pre-device on the intelligent shopping cart, and determining the first determination result includes:
when the judgment result is not in accordance with the judgment result;
monitoring commodities in the intelligent shopping cart based on a rear camera of the intelligent shopping cart, and acquiring the quantity of the commodities through a weight sensor of the intelligent shopping cart;
when the commodity bar code information is correct but does not accord with the corresponding commodity number, uploading the commodity number, simultaneously carrying out voice reminding on a customer, and generating abnormal information and sending the abnormal information to a preset control terminal after the reminding time exceeds a preset time threshold;
when the commodity bar code information and the corresponding commodity number are wrong, acquiring loss prevention information of the commodity;
acquiring a basket image before code scanning, a basket image before commodity putting and a basket image after the commodity putting through a rear camera, a code scanner and a weight sensor which are preset on the intelligent shopping cart;
comparing the basket image before code scanning, the basket image before commodity placing and the basket image after commodity placing, analyzing the commodity placed in the basket by the customer without code scanning, and determining the commodity without code scanning;
the abnormal commodity is determined through the loss prevention information and the unscanned code commodity, voice reminding is carried out on the customer, and after the reminding time exceeds a preset time threshold value, abnormal information is generated and sent to a preset control terminal.
The working principle and the beneficial effects of the technical scheme are as follows:
through the induction equipment on the intelligent shopping cart, the abnormal commodity is detected in time, the cost of manual detection is avoided when a user buys a large amount of articles, the efficiency of detection in the event of errors is improved, and the probability of commodity misplacing is reduced.
In one embodiment, the simultaneously locating the smart device and determining the information of the region of interest includes:
the method comprises the steps that a position sensor device preset in an intelligent terminal PAD is used for collecting the motion state and direction of an intelligent shopping cart in real time, and the staying and browsing time of a customer is obtained; wherein,
the position sensor device comprises at least a triangular gyroscope, an accelerometer and a level meter;
detecting a beacon preset in a peripheral area through a Bluetooth device preset in an intelligent terminal PAD, and identifying the area position of an intelligent shopping cart through the beacon;
acquiring and recording a shopping path of a customer according to the staying and browsing time, the area position and the staying duration, and analyzing a customer hot area and a customer flow line of a store through the shopping path;
and determining the interest area information of the customer through the customer hot area and the customer flow line.
The working principle and the beneficial effects of the technical scheme are as follows:
the method comprises the steps that the motion state and the direction of the intelligent shopping cart are collected in real time through a position sensor device preset in an intelligent terminal PAD, the staying and browsing time of a customer is obtained, the motion state and the direction of the intelligent shopping cart are judged and collected in real time through a triangular gyroscope, an accelerometer and other sensors in the intelligent terminal PAD, and whether the customer stays and browses or not is judged; the position of the intelligent shopping cart is identified by detecting Beacon beacons on the peripheral goods shelf through a Bluetooth module in the intelligent terminal PAD, and the area where a customer stays and the residence time are determined, so that the interest categories of the customer are known, the shopping path of the customer is collected and recorded, and the hot area and the passenger flow line of the customer in a store are analyzed finally. In the research of macroscopic shopping behaviors of customers, it is important to acquire position data and time data in a retail scene of the customer online. LBS location based services determine the geographic location of a user's mobile device by acquiring its communications, in most cases the mobile device signals we acquire come from a handset. The geographic position of the mobile phone user can be acquired by using GSM, CDMA or GPS through collecting the sent signals, and the position of the mobile phone user is inferred under the further processing of the GIS. The positioning technologies adopted according to different characteristics of service objects are different, and indoor positioning comprises multiple technologies such as WIFI, Bluetooth, RFID, geomagnetism and UWB, wherein the Bluetooth and RFID technologies are considered from the comprehensive aspects of cost, positioning precision and maintenance cost, and the requirements of indoor mobile equipment position data acquisition precision can be met with lower cost and better performance.
In one embodiment, the simultaneously locating the smart device and determining the information of the region of interest includes:
the method comprises the steps that a position sensor device preset in an intelligent terminal PAD is used for collecting the motion state and direction of an intelligent shopping cart in real time, and the staying and browsing time of a customer is obtained; wherein,
the position sensor device comprises at least a triangular gyroscope, an accelerometer and a level meter;
detecting a beacon preset in a peripheral area through a Bluetooth device preset in an intelligent terminal PAD, and identifying the area position of an intelligent shopping cart through the beacon;
acquiring and recording a shopping path of a customer according to the staying and browsing time, the area position and the staying duration, and analyzing a customer hot area and a customer flow line of a store through the shopping path;
and determining the interest area information of the customer through the customer hot area and the customer flow line.
The working principle and the beneficial effects of the technical scheme are as follows:
the customized marketing information is pushed to a customer to a screen of a PAD (intelligent terminal) based on the current marketing opportunity and the user portrait, the customer checks, clicks and purchases pushed contents on the screen, and the behavior data acquisition method based on visual detection is used for acquiring scene videos and images. To the customer's image that gathers, utilize computer vision analysis to draw customer behavior data in the image to further process analysis to customer behavior data, the mutual information of these customers and screen can be gathered by real time, and simultaneously, leading camera also gathers customer behavior feedback data in real time, and the result that lets wisdom marketing is more and more accurate, and data information volume is abundant, the analysis precision is high.
In one embodiment, the generating marketing feedback data through the shopping behavior data and the interest area information, and constructing a recommendation strategy based on the marketing feedback data, and meanwhile, pushing the recommendation strategy to the intelligent device includes:
generating marketing feedback data according to the shopping behavior data and the interest area information, and generating corresponding push content according to the marketing feedback data;
acquiring a scene video on the intelligent shopping cart based on a preset marketing opportunity and a preset visual detection behavior data method, and acquiring an acquired customer image by acquiring the scene video;
by the visual detection behavior data method, the customer behavior data in the image is analyzed and extracted, the customer behavior data is processed and analyzed and the semantics is extracted, the analysis result is determined,
performing intelligent learning through an analysis result to generate customized push content, pushing the customized push content to an intelligent terminal PAD, and determining a push result;
and acquiring interactive information in a pushing result in real time, acquiring behavior feedback data when a customer receives the interactive information in real time through a front-facing camera, generating an individual recommendation strategy, pushing the individual recommendation strategy and storing the individual recommendation strategy on intelligent equipment.
The working principle and the beneficial effects of the technical scheme are as follows:
and aiming at the collected customer images, the customer behavior data in the images are extracted by utilizing computer vision analysis, so that the customer behavior data are further processed and analyzed. After a long-time research, a relatively mature technical framework and an object-oriented technical framework are established on the human behavior recognition level, and the video-oriented technical framework analyzes related behavior data by extracting behavior characteristics of observed objects in a video sequence and a video. The method is mainly realized by utilizing image pixel difference and optical flow change among different frames in a video, and real-time observation and analysis of commodity change and user behavior data in the bicycle basket are realized. The image-oriented technical framework needs to extract semantic features of a large number of image data sets, obtain a training model in a machine learning training mode, and provide reference for user behavior data analysis in a real scene. The behavior data acquisition method based on visual detection has the advantages of abundant data information amount and high analysis accuracy, and can be accurate to within 1 meter.
In one embodiment, the generating marketing feedback data through the shopping behavior data and the interest area information, and generating corresponding push content according to the marketing feedback data further includes:
step 1: acquiring user information of a target customer;
step 2: obtaining historical shopping behavior data through the user information, and generating personalized shopping behavior data according to the historical shopping behavior data
(ii) a Wherein,
and step 3: counting and recording the personalized shopping behavior data based on a preset big data center to generate interest group consumption data;
wherein,
a consumption function representing the interest group is provided,
representing the consumption data of the interest group after the personalized shopping behavior data are counted,
a distribution function representing a population of statistically personalized shopping behavior data,
representing a screening function for interestingness;
and 4, step 4: constructing an interest recommendation algorithm according to the interest group consumption data;
wherein,
a recommendation algorithm is represented for the interest, and,
in respect of
An interest recommendation algorithm for the batch personalized shopping data,
，
representing the total number of batches of personalized shopping data,
represents to the first
The consumption data of interest groups after the batch personalized shopping data statistics,
representing a recommendation function regarding consumption data of a group of interest,
representing about a target user
The transpose of the forward interest weight matrix of the temporal interest group consumption data,
a forward interest weight matrix representing consumption data about the interest group,
a negative interest weight matrix representing consumption data about the interest group,
effective parameters representing positive interest weight and negative interest weight consumed by interest groups;
and 5: and obtaining the interested commodities through the interest recommendation algorithm, receiving marketing feedback data, and generating corresponding push contents through the interested commodities and the marketing feedback data.
The working principle and the beneficial effects of the technical scheme are as follows:
the technical scheme adopts AI intelligent terminal equipment (intelligent shopping cart, outlet all-in-one machine) and intelligent algorithm application on the terminal equipment to provide a brand-new off-line shopping tool for consumers, collects a large amount of original data related to artificial intelligence such as RFID data, Bluetooth data, camera image data, various sensor data, mobile internet data and the like, and carries out support of algorithm training after marking treatment, so that the off-line shopping cart has the same digitalization capability as the on-line merchants, deeply explores user value, provides personalized service for the consumers through intelligent marketing and intelligent management, has the greatest advantages that the intelligent shopping cart can collect data of the whole shopping period of the users and actively make personalized recommendation according to user figures and marketing opportunities, namely, when the consumers use the intelligent shopping cart after entering a store each time, the recommendation method can make personalized recommendation according to the preference degree of target customers, obtaining user information of a target customer, obtaining historical shopping data through the user information, and generating personalized shopping data
Based on a preset big data center, counting and recording the personalized shopping data to generate interest group consumption data
And the recommendation given by the system is updated in real time, that is, when the marketing resources and the user data in the system are changed or different marketing opportunity information is corresponded in the process of travel, the given recommendation sequence can be automatically changedEstablishing an interest recommendation algorithm according to the interest group consumption data
(ii) a Greatly facilitating the customers and improving the service level of the enterprise.
In one embodiment, the intelligent device further comprises an outlet all-in-one machine; wherein,
the exit all-in-one machine is used for detecting an RFID beacon on the intelligent shopping cart based on a preset RFID identification mechanism, identifying a target vehicle, detecting whether an order is abnormal or not by collecting shopping data of the target vehicle, and determining a detection result; wherein,
and when the detection result is that the order is abnormal, the abnormal order is closed, the channel outlet cannot be opened, the vehicle number is broadcasted in a voice mode to inform an outlet loss prevention worker, meanwhile, commodity loss prevention data and management and inspection data of the order are obtained, and visual order information is generated.
The working principle and the beneficial effects of the technical scheme are as follows: the customer cart arrives at an exit, and the exit all-in-one machine detects RFID beacons on the approaching intelligent shopping cart through an RFID identification technology, so that the approaching vehicle is identified; after the order is obtained, the background damage prevention system calculates whether the order is abnormal or not according to shopping data in the process, and the abnormal order can be verified by informing an exit damage prevention person of opening a door and broadcasting the car number in a voice mode. And after the loss prevention personnel check, submitting the checking result on a screen, and collecting commodity loss prevention data and management checking data of the order by the outlet all-in-one machine.
In a word, AI intelligent terminal equipment (intelligent shopping cart, outlet all-in-one machine) and intelligent algorithm application on the terminal equipment are adopted to provide a brand-new off-line shopping tool mode for consumers, a large amount of RFID data, Bluetooth data, camera image data, various sensor data, mobile internet data and other artificial intelligence related original data are collected and are subjected to marking processing and then are used as support for algorithm training, so that off-line equipment has the same digitalization capacity as on-line power suppliers, the user value is deeply explored, and personalized service is provided for the consumers through intelligent marketing and intelligent management.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (8)
1. A data acquisition and recommendation method based on shopping behaviors is characterized by comprising the following steps:
the method comprises the steps that a preset intelligent camera is used for collecting a face image of a customer, and shopping behavior data are obtained through the face image; wherein,
the intelligent camera comprises a front camera, a rear camera and a side camera;
the method comprises the steps that commodity data in the shopping process are collected through a preset intelligent device, meanwhile, the intelligent device is located, and information of an interest area is determined;
generating marketing feedback data through the shopping behavior data and the information of the interest area, constructing a recommendation strategy based on the marketing feedback data, and meanwhile pushing the recommendation strategy to intelligent equipment;
generating marketing feedback data through the shopping behavior data and the information of the interest area, constructing a recommendation strategy based on the marketing feedback data, and pushing the recommendation strategy to intelligent equipment, wherein the method comprises the following steps:
generating marketing feedback data according to the shopping behavior data and the interest area information, and generating corresponding push content according to the marketing feedback data;
acquiring a scene video on the intelligent shopping cart based on a preset marketing opportunity and a preset visual detection behavior data method, and acquiring an acquired customer image by acquiring the scene video;
by the visual detection behavior data method, the customer behavior data in the image is analyzed and extracted, the customer behavior data is processed and analyzed and the semantics is extracted, the analysis result is determined,
performing intelligent learning through an analysis result to generate customized push content, pushing the customized push content to a PAD of the intelligent terminal, and determining a pushing result;
acquiring interactive information in a pushing result in real time, acquiring behavior feedback data when a customer receives the interactive information in real time through a front-facing camera, generating an individual recommendation strategy, pushing the individual recommendation strategy and storing the individual recommendation strategy on intelligent equipment;
generating marketing feedback data through the shopping behavior data and the interest area information, and generating corresponding push content according to the marketing feedback data further comprises:
step 1: acquiring user information of a target customer;
step 2: obtaining historical shopping behavior data through the user information, and generating personalized shopping behavior data according to the historical shopping behavior data
(ii) a Wherein,
and step 3: counting and recording the personalized shopping behavior data based on a preset big data center to generate interest group consumption data;
wherein,
a consumption function representing the interest group is provided,
representing the consumption data of the interest group after the personalized shopping behavior data are counted,
a distribution function representing a population of statistically personalized shopping behavior data,
representing a screening function for interestingness;
and 4, step 4: constructing an interest recommendation algorithm according to the interest group consumption data;
wherein,
a recommendation algorithm is represented for the interest, and,
in respect of
An interest recommendation algorithm for the batch personalized shopping data,
，
representing the total number of batches of personalized shopping data,
represents to the first
The consumption data of interest groups after the batch personalized shopping data statistics,
representing a recommendation function regarding consumption data of a group of interest,
representing about a target user
The transpose of the forward interest weight matrix of the temporal interest group consumption data,
a forward interest weight matrix representing consumption data about the interest group,
a negative interest weight matrix representing consumption data about the interest group,
effective parameters representing positive interest weight and negative interest weight consumed by interest groups;
and 5: and obtaining the interested commodities through the interest recommendation algorithm, receiving marketing feedback data, and generating corresponding push contents through the interested commodities and the marketing feedback data.
2. The shopping behavior-based data acquisition and recommendation method as claimed in claim 1, wherein the acquiring of the face image of the customer through a preset intelligent camera and the acquisition of the shopping behavior data through the face image comprises:
the method comprises the steps that through a front camera, portrait acquisition is conducted on a customer, a face image of the customer is recorded based on a preset face detection algorithm, and identity information of the customer is recognized;
performing expression analysis on the facial image of the customer in real time through the identity information of the customer and the corresponding facial image to determine an expression analysis result;
and evaluating the shopping behavior of the customer according to the expression analysis result, and determining the shopping behavior data of the customer.
3. The shopping behavior-based data acquisition and recommendation method as claimed in claim 2, wherein the recording of the facial image of the customer and the identification of the identity information of the customer based on a preset face detection algorithm comprises:
acquiring a face image of a face of a customer;
recording and extracting target rectangular features of the face image based on a preset Adaboost algorithm, and generating weak separator features according to the target rectangular features;
strengthening the weak separator characteristics according to a preset weighted voting mechanism, and constructing strong classifier characteristics;
connecting the strong classifier features in series to generate a laminated classifier feature, identifying the face feature through the laminated classifier feature, and determining a face identification image;
based on a preset big data center, carrying out preprocessing operation on the face recognition image to determine a preprocessed image; wherein,
the preprocessing operation at least comprises one or more of light compensation, gray scale transformation, histogram equalization, normalization, geometric correction, filtering and sharpening;
extracting geometric features of the preprocessed image, extracting face feature points, modeling face features through the face feature points, and building a face feature model;
extracting face feature data in a face image to be recognized based on the face feature model;
and searching and matching the face feature data with a face feature template in a preset storage database, and identifying the identity information of the customer.
4. The data acquisition and recommendation method based on shopping behaviors as claimed in claim 3, wherein the extracting geometric features of the preprocessed image, extracting face feature points, and modeling face features through the face feature points to build a face feature model comprises:
extracting human face features based on a preset geometric feature algorithm, and determining local geometric lines of the human face; wherein,
the geometric feature algorithm is used for identifying line changes of the eyes, the nose, the mouth and the chin;
analyzing the local shape and distance characteristics of the geometric lines, determining face classification characteristic data, and extracting face characteristic points according to the face classification characteristic data;
generating a corresponding face characteristic component according to the face characteristic points; wherein,
the face feature component at least comprises Euclidean distances, curvatures and angles among face feature points;
and constructing a human face feature model through the human face feature components, the human face feature points and the human face classification feature data.
5. The shopping behavior-based data acquisition and recommendation method as claimed in claim 1, wherein the commodity data during shopping is acquired through a preset intelligent device, further comprising:
acquiring commodity bar code information through an intelligent terminal PAD preset on an intelligent shopping cart;
acquiring commodity data through a weight sensor and an intelligent camera preset in the intelligent shopping cart; wherein,
the commodity data comprises commodity type information, commodity putting weight and commodity putting time;
according to the commodity bar code information, the intelligent shopping cart detects corresponding commodity data, judges whether the commodity weight is in accordance with the standard weight corresponding to the commodity bar code information, and determines a judgment result;
when the judgment result is in accordance with the judgment result, the intelligent shopping cart automatically adds the corresponding code scanning commodity to a shopping list;
and when the judgment result is inconsistent, monitoring and judging abnormal information of the commodities in the intelligent shopping cart based on a rear camera arranged on the intelligent shopping cart in advance, and determining a first judgment result.
6. The shopping behavior-based data acquisition and recommendation method as claimed in claim 5, wherein when the determination result is that the data acquisition and recommendation method does not conform to the predetermined reference, monitoring and determining abnormal information of the goods in the intelligent shopping cart based on a rear camera of the intelligent shopping cart comprises:
when the judgment result is not in accordance with the judgment result;
monitoring commodities in the intelligent shopping cart based on a rear camera of the intelligent shopping cart, and acquiring the quantity of the commodities through a weight sensor of the intelligent shopping cart;
when the commodity bar code information is correct but does not accord with the corresponding commodity number, uploading the commodity number, simultaneously carrying out voice reminding on a customer, and generating abnormal information and sending the abnormal information to a preset control terminal after the reminding time exceeds a preset time threshold;
when the commodity bar code information and the corresponding commodity number are wrong, acquiring loss prevention information of the commodity;
acquiring a basket image before code scanning, a basket image before commodity putting and a basket image after the commodity putting through a rear camera, a code scanner and a weight sensor which are preset on the intelligent shopping cart;
comparing the basket image before code scanning, the basket image before commodity placing and the basket image after commodity placing, analyzing the commodity placed in the basket by the customer without code scanning, and determining the commodity without code scanning;
the abnormal commodity is determined through the loss prevention information and the unscanned code commodity, voice reminding is carried out on the customer, and after the reminding time exceeds a preset time threshold value, abnormal information is generated and sent to a preset control terminal.
7. The shopping behavior-based data acquisition and recommendation method according to claim 1, wherein the simultaneously locating the smart device and determining the interest area information comprises:
the method comprises the steps that a position sensor device preset in an intelligent terminal PAD is used for collecting the motion state and direction of an intelligent shopping cart in real time, and the staying and browsing time of a customer is obtained; wherein,
the position sensor device comprises at least a triangular gyroscope, an accelerometer and a level meter;
detecting a beacon preset in a peripheral area through a Bluetooth device preset in an intelligent terminal PAD, and identifying the area position of an intelligent shopping cart through the beacon;
acquiring and recording a shopping path of a customer according to the staying and browsing time and the area position, and analyzing a customer hot area and a customer flow line of a store through the shopping path;
and determining the interest area information of the customer through the customer hot area and the customer flow line.
8. The shopping behavior-based data acquisition and recommendation method as claimed in claim 1, wherein the smart device further comprises an exit kiosk; wherein,
the exit all-in-one machine is used for detecting an RFID beacon on the intelligent shopping cart based on a preset RFID identification mechanism, identifying a target vehicle, detecting whether an order is abnormal or not by collecting shopping data of the target vehicle, and determining a detection result; wherein,
and when the detection result is that the order is abnormal, the abnormal order is closed, the channel outlet cannot be opened, the vehicle number is broadcasted in a voice mode to inform an outlet loss prevention worker, meanwhile, commodity loss prevention data and management and inspection data of the order are obtained, and visual order information is generated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111226921.6A CN113674037B (en) | 2021-10-21 | 2021-10-21 | Data acquisition and recommendation method based on shopping behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111226921.6A CN113674037B (en) | 2021-10-21 | 2021-10-21 | Data acquisition and recommendation method based on shopping behaviors |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113674037A CN113674037A (en) | 2021-11-19 |
| CN113674037B true CN113674037B (en) | 2022-01-25 |
Family
ID=78550778
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111226921.6A Active CN113674037B (en) | 2021-10-21 | 2021-10-21 | Data acquisition and recommendation method based on shopping behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113674037B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114220140B (en) * | 2021-11-23 | 2022-07-22 | 慧之安信息技术股份有限公司 | Image recognition-based market passenger flow volume statistical method and device |
| TWI831287B (en) * | 2022-07-12 | 2024-02-01 | 財團法人商業發展研究院 | A target customer consumption preference behavior observation system and method |
| CN115564472B (en) * | 2022-09-22 | 2024-05-31 | 珠海众志智能科技有限公司 | Equipment interactive information service system and method based on big data |
| CN116205555B (en) * | 2023-04-28 | 2023-09-05 | 宁夏金丝路大数据科技有限责任公司 | Logistics information visual management system and method based on big data |
| CN116362848B (en) * | 2023-06-03 | 2023-10-27 | 广州爱特安为科技股份有限公司 | Electronic commerce recommendation method, system, equipment and medium based on artificial intelligence |
| CN116682071B (en) * | 2023-08-04 | 2023-11-10 | 浙江大华技术股份有限公司 | Commodity interest information analysis method, equipment and storage medium |
| CN116720899B (en) * | 2023-08-09 | 2023-11-03 | 山东商软信息科技有限公司 | Super-intelligent business monitoring management method, device, electronic equipment and medium |
| CN117135380B (en) * | 2023-10-26 | 2023-12-26 | 环球数科集团有限公司 | Travel product live broadcast marketing system based on AIGC technology |
| CN118470870A (en) * | 2024-05-31 | 2024-08-09 | 深圳市易联达商用设备有限公司 | Cash register management method and system for unmanned market |
| CN118397568B (en) * | 2024-06-26 | 2024-09-03 | 汉朔科技股份有限公司 | Abnormal shopping behavior detection method, device and system of shopping cart and shopping cart |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809479A (en) * | 2016-03-07 | 2016-07-27 | 海信集团有限公司 | Item recommending method and device |
| CN106022865A (en) * | 2016-05-10 | 2016-10-12 | 江苏大学 | Goods recommendation method based on scores and user behaviors |
| CN107633430A (en) * | 2017-09-20 | 2018-01-26 | 哈尔滨工业大学 | A kind of Method of Commodity Recommendation based on community of colony |
| CN111784372A (en) * | 2019-04-03 | 2020-10-16 | Tcl集团股份有限公司 | Store commodity recommendation method and device |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040093274A1 (en) * | 2002-11-08 | 2004-05-13 | Marko Vanska | Method and apparatus for making daily shopping easier |
| CN102122346A (en) * | 2011-02-28 | 2011-07-13 | 济南纳维信息技术有限公司 | Video analysis-based physical storefront customer interest point acquisition method |
| CN102402765B (en) * | 2011-12-27 | 2017-07-28 | 北京京东尚科信息技术有限公司 | Electronic commerce recommending method based on user's Expression analysis |
| CN106408369B (en) * | 2016-08-26 | 2021-04-06 | 西安超嗨网络科技有限公司 | Method for intelligently identifying commodity information in shopping cart |
| WO2019147815A1 (en) * | 2018-01-24 | 2019-08-01 | Nike Innovate C.V. | System, platform and method for personalized shopping using a virtual shopping assistant |
| US20190236676A1 (en) * | 2018-01-30 | 2019-08-01 | Wal-Mart Stores, Inc. | Intelligent recommendation engine |
| CN109064270B (en) * | 2018-07-23 | 2022-05-17 | 京东方科技集团股份有限公司 | Intelligent shopping cart, server, intelligent shopping system and method |
| CN111932339A (en) * | 2020-08-06 | 2020-11-13 | 治略资讯整合股份有限公司 | Commodity recommendation method and system based on consumer groups and computer storage medium |
-
2021
- 2021-10-21 CN CN202111226921.6A patent/CN113674037B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809479A (en) * | 2016-03-07 | 2016-07-27 | 海信集团有限公司 | Item recommending method and device |
| CN106022865A (en) * | 2016-05-10 | 2016-10-12 | 江苏大学 | Goods recommendation method based on scores and user behaviors |
| CN107633430A (en) * | 2017-09-20 | 2018-01-26 | 哈尔滨工业大学 | A kind of Method of Commodity Recommendation based on community of colony |
| CN111784372A (en) * | 2019-04-03 | 2020-10-16 | Tcl集团股份有限公司 | Store commodity recommendation method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113674037A (en) | 2021-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113674037B (en) | Data acquisition and recommendation method based on shopping behaviors | |
| CN108198052B (en) | User selected commodity identification method, device and intelligent goods shelf system | |
| CN112348640B (en) | Online shopping system and method based on facial emotion state analysis | |
| CN110033298B (en) | Information processing apparatus, control method thereof, system thereof, and storage medium | |
| US8351647B2 (en) | Automatic detection and aggregation of demographics and behavior of people | |
| CN109726759B (en) | Unmanned vending method, device, system, electronic equipment and computer readable medium | |
| CN105678591A (en) | Video-analysis-based commercial intelligent operation decision-making support system and method | |
| CN107844990A (en) | A kind of approaches to IM and its system, terminal device for intelligent shops | |
| CN110472870A (en) | A kind of cashier service regulation detection system based on artificial intelligence | |
| CN108537166B (en) | Method and device for determining shelf browsing amount and analyzing browsing amount | |
| KR101779096B1 (en) | The object pursuit way in the integration store management system of the intelligent type image analysis technology-based | |
| JP5002441B2 (en) | Marketing data analysis method, marketing data analysis system, data analysis server device, and program | |
| JP6498900B2 (en) | Advertisement evaluation system, advertisement evaluation method | |
| US12093889B2 (en) | Method, a device and a system for checkout | |
| CN113222582B (en) | Face payment retail terminal | |
| CN110889719A (en) | Advertisement pushing system and method based on user consumption characteristics | |
| CN113630721A (en) | Method and device for generating recommended tour route and computer readable storage medium | |
| CN113887884A (en) | Business-super service system | |
| CN111222870A (en) | Settlement method, device and system | |
| CN110689389A (en) | Computer vision-based shopping list automatic maintenance method and device, storage medium and terminal | |
| CN109074498A (en) | Visitor's tracking and system for the region POS | |
| CN113781058A (en) | Payment method and device for unmanned vehicle, electronic equipment and storage medium | |
| CN116823378A (en) | Visual marketing method for unmanned retail equipment | |
| CN112257654A (en) | Face recognition device for smart city | |
| CN112989988A (en) | Information integration method, device, equipment, readable storage medium and program product |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A Data Collection and Recommendation Method Based on Shopping Behavior Effective date of registration: 20230926 Granted publication date: 20220125 Pledgee: Xi'an Caijin Financing Guarantee Co.,Ltd. Pledgor: XIAN CHAOHI NET TECHNOLOGY Co.,Ltd. Registration number: Y2023980059227 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |