CN115965249A - Video network customer intelligent analysis management system based on artificial intelligence technology - Google Patents
Video network customer intelligent analysis management system based on artificial intelligence technology Download PDFInfo
- Publication number
- CN115965249A CN115965249A CN202211622630.3A CN202211622630A CN115965249A CN 115965249 A CN115965249 A CN 115965249A CN 202211622630 A CN202211622630 A CN 202211622630A CN 115965249 A CN115965249 A CN 115965249A
- Authority
- CN
- China
- Prior art keywords
- commodity
- customer
- retail store
- unmanned retail
- target unmanned
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 31
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 11
- 238000012544 monitoring process Methods 0.000 claims description 115
- 238000000034 method Methods 0.000 claims description 44
- 238000012216 screening Methods 0.000 claims description 19
- 238000004806 packaging method and process Methods 0.000 claims description 12
- 238000011156 evaluation Methods 0.000 claims description 10
- 238000012937 correction Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 230000006855 networking Effects 0.000 claims description 6
- 238000006467 substitution reaction Methods 0.000 claims description 5
- 230000000007 visual effect Effects 0.000 claims description 3
- 230000006399 behavior Effects 0.000 description 21
- 239000000047 product Substances 0.000 description 3
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention relates to the field of intelligent analysis and management of video-networking clients, and particularly discloses an intelligent analysis and management system of video-networking clients based on an artificial intelligence technology.
Description
Technical Field
The invention relates to the field of intelligent analysis and management of clients in a video network, in particular to an intelligent analysis and management system of clients in the video network based on an artificial intelligence technology.
Background
Along with the development of the Internet and the unmanned vending technology, the intelligent retail is pursued by more and more people, the video AI analysis provided by the intelligent video networking platform helps the enterprise responsible person or the shop owner to remotely patrol the shop, the cost is low, the manpower is not needed, one person can manage multiple shops, and the enterprise or the shop is further helped to realize more efficient operation.
However, while the unmanned retail store has the intelligent and efficient characteristic, some problems also exist: on the one hand, the security monitoring of the existing unmanned retail store is not strong enough, especially the stealing behavior of personnel, and a set of effective analysis monitoring method for the stealing behavior is lacked, and on the other hand, the favorite commodities of customers are not analyzed, so that the types of the marketable commodities in the store can not be obtained and the targeted replenishment can not be carried out, blind replenishment and resource waste are easy to occur, or supply and demand shortage occurs, the experience of customers is reduced, the customer loss is caused, and the normal operation of the store is influenced.
Disclosure of Invention
Aiming at the problems, the invention provides an intelligent analysis management system for the clients of the video networking based on the artificial intelligence technology, and the intelligent analysis management function of the clients of the video networking is realized.
The technical scheme adopted by the invention for solving the technical problems is as follows: the invention provides a visual networking client intelligent analysis management system based on artificial intelligence technology, comprising: the customer basic information acquisition module: the system is used for acquiring basic information of each customer in a target unmanned retail store in a preset monitoring period, wherein the basic information comprises shopping information and stealing information, the shopping information comprises the stay time of each type of commodity, the touch frequency of each type of commodity in each type of commodity and the purchase quantity of each type of commodity in each type of commodity, and the stealing information comprises the quantity of the evasion commodities and the times of plagiarism tendency actions.
Customer theft monitoring module: and the method is used for analyzing and obtaining the theft behavior tendency coefficient of each customer in the target unmanned retail store according to the theft information of each customer in the target unmanned retail store in the preset monitoring period.
Customer theft behavior evaluation module: and the system is used for judging whether each customer in the target unmanned retail store steals or not according to the steal behavior tendency coefficient of each customer in the target unmanned retail store and carrying out early warning.
The customer preference commodity type analysis module: and the preference coefficient of each type of commodity of each customer in the target unmanned retail store in the preset monitoring period is obtained according to the shopping information of each customer in the target unmanned retail store in the preset monitoring period.
The customer preference commodity type processing module: and the remote control terminal is used for analyzing and obtaining the hot sales degree of each commodity type in the target unmanned retail store according to the preference coefficient of each type of commodity of each customer in the target unmanned retail store in a preset monitoring period, and sending the hot sales degree of each commodity type in the target unmanned retail store to the target unmanned retail store.
A database: the system is used for storing standard package images of various commodities in the target unmanned retail store, taking a commodity suspicious action image set, putting back the commodity suspicious action image set, commodity types corresponding to various commodity placing areas and various commodities in various types of commodities.
On the basis of the embodiment, the customer basic information acquisition module acquires the theft information of each customer in the target unmanned retail store in a preset monitoring period, and the specific process comprises the following steps: the method comprises the steps of obtaining shopping videos of all customers in a target unmanned retail store in a preset monitoring period, obtaining commodity taking videos and commodity placing videos of all the customers according to the shopping videos of all the customers in the target unmanned retail store in the preset monitoring period, further obtaining commodity package images in the commodity taking videos and the commodity package images in the commodity placing videos of all the customers, extracting standard package images of all the commodities in the target unmanned retail store stored in a database, comparing the commodity package images in the commodity taking videos and the commodity package images in the commodity placing videos of all the customers with the standard package images of all the commodities to obtain commodities taken by all the customers and placed commodities, obtaining the taking times and the placing times of all the commodities corresponding to all the customers according to the commodities taken by all the customers and the commodities placed back by all the customers, comparing the taking times and the placing times of all the commodities corresponding to all the customers to obtain selected commodities of all the customers, and screening the commodities.
The method comprises the steps of obtaining videos of cashier areas in target unmanned retail stores in a preset monitoring period through a high-definition camera, obtaining payment videos of all customers through interception, obtaining payment time periods of all the customers according to starting time and ending time corresponding to the payment videos of all the customers, obtaining checkout information of the target unmanned retail stores in the preset monitoring period through a checkout platform of the target unmanned retail stores, screening out checkout information corresponding to the payment time periods of all the customers according to the payment time periods of all the customers, recording the checkout information as checkout information of all the customers, and obtaining checkout commodities of all the customers according to the checkout information of all the customers.
Comparing each shopping commodity of each customer with each checkout commodity, if a certain shopping commodity of a certain customer is not in the checkout commodities, marking the shopping commodity of the customer as an order-escaping commodity of the customer, screening to obtain each order-escaping commodity of each customer, counting to obtain the number of the order-escaping commodities of each customer in the target unmanned retail store in a preset monitoring period, and marking the number as alpha i And i represents the target unmanned retail store in the preset monitoring periodThe number of the ith customer, i =1,2.
On the basis of the above embodiment, the customer basic information acquisition module acquires theft information of each customer in the target unmanned retail store within a preset monitoring period, and the specific process further includes: according to each commodity taking video and each commodity putting back video of each customer, obtaining each action image corresponding to each commodity taking time and each action image corresponding to each commodity putting back time of each customer, extracting a suspicious action image set of the taken commodity and a suspicious action image set of the put commodity stored in a database, comparing each action image corresponding to each commodity taking time of each customer with the suspicious action image set of the taken commodity, if a certain action image corresponding to a certain commodity taking time of a certain customer belongs to the suspicious action image set of the taken commodity, recording the action corresponding to the commodity taking time of the certain customer as a suspicious taking action, screening each suspicious taking action corresponding to each commodity taking time of each customer, counting the number of the suspicious taking actions corresponding to each commodity taking time of each customer, and recording the suspicious taking actions
b represents the number of times of fetching goods for the b-th time, b =1,2,.. And c, and similarly, according to the analysis method of the number of suspicious fetching actions corresponding to the goods fetched by each customer for each time, the number of suspicious fetching actions corresponding to the goods put back by each customer for each time is obtained and recorded as ∑ and ∑>
f denotes the number of times the item was put back f =1,2.
The suspicious taking action quantity corresponding to each time of taking the commodity by each customer
And the number of suspicious playback actions corresponding to each time of the returned goods>
Substituted into formula>
Obtaining the times beta of plagiarism tendency actions of each customer in the target unmanned retail store within a preset monitoring period i Therein x 1 、χ 2 And weighting factors respectively representing the number of suspicious fetching actions corresponding to the preset fetched commodities and the number of suspicious retrieving actions corresponding to the retrieved commodities.
On the basis of the above embodiment, the customer basic information acquisition module acquires shopping information of each customer in the target unmanned retail store within a preset monitoring period, and the specific process includes: the method comprises the steps of obtaining the standing time of each customer in each commodity placing area in a target unmanned retail store according to a shopping video of each customer in the target unmanned retail store within a preset monitoring period, comparing the standing time of each customer in each commodity placing area in the target unmanned retail store with a preset standing time threshold, recording the commodity placing area as the staying commodity placing area of the customer if the standing time of a certain customer in a certain commodity placing area in the target unmanned retail store is larger than or equal to the preset standing time threshold, recording the standing time of the customer in the staying commodity placing area as the staying commodity placing area of the customer, and counting to obtain each staying commodity placing area of each customer and the staying time of each customer in each staying commodity placing area.
The method comprises the steps of extracting commodity types corresponding to each commodity placing area stored in a database, screening and obtaining the commodity types corresponding to each commodity placing area of each customer according to each commodity placing area of each customer, obtaining the stay time of each type of commodity of each customer in a target unmanned retail store in a preset monitoring period according to the commodity types corresponding to each commodity placing area of each customer and the stay time of each customer in each commodity placing area, and recording the stay time as the stay time of each type of commodity of each customer in the target unmanned retail store in the preset monitoring period
j denotes the number of the jth product type, j =1,2.
On the basis of the embodiment, the customer basic information acquisition module acquires shopping information of each customer in the target unmanned retail store in a preset monitoring period, and the specific processFurther comprising: extracting each commodity in each type of commodity in the target unmanned retail store stored in the database, screening and obtaining a commodity type style corresponding to each commodity taken by each customer according to each commodity taken by each customer, comparing the commodity type styles corresponding to each commodity taken by each customer with each other, counting the number of times of taking each commodity in each type of commodity by each customer, and recording the number as n ijp P denotes the number of item p, p =1,2.
And recording the time length of each commodity taking video of each customer as the touch time length of each commodity taking of each customer according to each commodity taking video of each customer, and obtaining the touch time length of each commodity taking corresponding to each commodity in each type of commodity of each customer according to the commodity type style corresponding to each commodity taking of each customer and the touch time length of each commodity taking of each customer.
Substituting the times of taking the commodities of various types by each customer and the touch duration of the commodities of various types corresponding to the commodities of various types by each customer into a formula
Obtaining the touch frequency delta of each commodity in each type of commodity of each customer in the target unmanned retail store in the preset monitoring period ijp Wherein phi represents a correction factor of the frequency of touch of the customer goods in the target unmanned retail store within a preset monitoring period, epsilon 1 、ε 2 Respectively representing the preset times of taking the commodities and the weight factors of the touch duration of the taken commodities, wherein delta n represents the preset threshold value of the times of taking the commodities, and the judgment result is obtained>
The method comprises the steps of representing the touch time of the ith customer for fetching the commodities at the u time corresponding to the pth commodity in the jth type commodities, wherein u =1,2.
According to each checkout commodity of each customer, comparing each checkout commodity of each customer with each commodity in each type of commodities in the target unmanned retail store, and counting to obtainAnd recording the purchase quantity of each commodity in each type of commodity of each customer in the target unmanned retail store as kappa according to the purchase quantity of each commodity in each type of commodity in the preset monitoring period ijp 。
On the basis of the embodiment, the specific process of the customer theft monitoring module is as follows: the number alpha of the order-escaping commodities of each customer in the target unmanned retail store within a preset monitoring period i And number of plagiarism-prone actions β i Substitution formula
Obtaining theft propensity coefficients for individual customers in a targeted unmanned retail store>
Wherein η represents a predetermined target steal propensity factor correction factor, γ, for a customer in an unmanned retail store 1 、γ 2 Weight factors respectively representing the preset number of the evasion commodities and the times of plagiarism tendency actions of the customers>
Represents a preset theft action influence factor corresponding to a single evasion commodity, and>
the method represents a theft behavior influence factor corresponding to a preset single plagiarism tendency action.
On the basis of the above embodiment, the specific process of the customer theft behavior evaluation module is as follows: the method comprises the steps of comparing a theft tendency coefficient of each customer in a target unmanned retail store with a preset theft tendency coefficient threshold value, if the theft tendency coefficient of a certain customer in the target unmanned retail store is larger than the preset theft tendency coefficient threshold value, the customer in the target unmanned retail store has a theft behavior, carrying out early warning through voice equipment in the target unmanned retail store, counting the customers with the theft behavior in the target unmanned retail store, and sending images of the customers with the theft behavior in the target unmanned retail store to a remote control terminal of the target unmanned retail store.
On the basis of the above embodiment, the specific process of the customer preference commodity type analysis module is as follows: stay time of each type of commodity of each customer in the target unmanned retail store within a preset monitoring period
Frequency of touch δ of each item of each type of item ijp And the purchase quantity kappa of each type of goods in each type of goods ijp Substituting into formula
Obtaining preference coefficients lambda of various types of commodities of various customers in the target unmanned retail store in a preset monitoring period ij Where m represents the total number of types of items.
On the basis of the above embodiment, the specific process of the customer preference product type processing module is as follows: comparing preference coefficients of various types of commodities of each customer in the target unmanned retail store within a preset monitoring period, taking the commodity type corresponding to the maximum preference coefficient as a favorite commodity type of the customer, counting to obtain the favorite commodity type of each customer, obtaining the number of favorite customers corresponding to each commodity type according to the favorite commodity type of each customer, and recording as w j Substituting the number of favorite customers corresponding to each commodity type into a formula
Obtaining the hot sales xi of each commodity type in the target unmanned retail store j Where ψ denotes a hot sales correction factor for a type of goods in a preset target unmanned retail store, and e denotes a natural constant.
And sorting the commodity types in the target unmanned retail store according to the sequence of the hot sales degree from high to low, and sending the sorted result to a remote control terminal of the target unmanned retail store.
Compared with the prior art, the visual networking client intelligent analysis management system based on the artificial intelligence technology has the following beneficial effects: 1. according to the video-networking client intelligent analysis management system based on the artificial intelligence technology, the shopping information and the stealing information of each client in the target unmanned retail store in the preset monitoring period are obtained, the stealing behavior tendency coefficient of each client in the target unmanned retail store and the hot sales degree of each commodity type are obtained through analysis, whether each client in the target unmanned retail store steals or not is judged, early warning is carried out, the hot sales degree of each commodity type is sent to the remote control terminal of the target unmanned retail store, the stealing behaviors of the unmanned retail store can be monitored remotely, the hot commodity type of the unmanned retail store can be identified rapidly, and therefore intelligent, efficient and accurate analysis management of the unmanned retail store is achieved.
2. According to the invention, the number of the order-escaping commodities and the times of plagiarism tendency actions of each customer in the target unmanned retail store in the preset monitoring period are obtained, whether each customer in the target unmanned retail store steals or not is further judged, early warning is carried out, the steals are deeply analyzed from multiple dimensions, an effective evaluation method aiming at the steals is established, the monitoring strength of the security of the store is further enhanced, and thus the security of the store is guaranteed.
3. According to the method, the hot sales degree of each commodity type in the target unmanned retail store is obtained by obtaining the shopping information of each customer in the target unmanned retail store in the preset monitoring period, the favorite commodities of the customers are analyzed, the types of the good-selling commodities in the store are further obtained, the targeted replenishment is carried out, the blind replenishment or the supply shortage is prevented, the experience of the customers is further improved, and the guarantee is provided for the normal operation of the store.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a system module connection diagram of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the invention provides an intelligent analysis management system for a client in a video network based on an artificial intelligence technology, which comprises a client basic information acquisition module, a client theft monitoring module, a client theft evaluation module, a client preference commodity type analysis module, a client preference commodity type processing module and a database.
The customer basic information acquisition module is respectively connected with the database, the customer theft monitoring module and the customer preference commodity type analysis module, the customer theft monitoring module is connected with the customer theft evaluation module, and the customer preference commodity type analysis module is connected with the customer preference commodity type processing module.
The customer basic information acquisition module is used for acquiring basic information of each customer in a target unmanned retail store in a preset monitoring period, wherein the basic information comprises shopping information and stealing information, the shopping information comprises the stay time of each type of commodity, the touch frequency of each type of commodity in each type of commodity and the purchase quantity of each type of commodity in each type of commodity, and the stealing information comprises the quantity of the order evasion commodities and the times of plagiarism tendency actions.
Further, the customer basic information acquisition module acquires the stealing information of each customer in the target unmanned retail store within a preset monitoring period, and the specific process comprises the following steps: the method comprises the steps of obtaining a shopping video of each customer in a target unmanned retail store in a preset monitoring period, obtaining a commodity taking video and a commodity putting-back video of each customer according to the shopping video of each customer in the target unmanned retail store in the preset monitoring period, further obtaining a commodity packaging image in the commodity taking video and a commodity packaging image in the commodity putting-back video of each customer, extracting standard packaging images of commodities in the target unmanned retail store stored in a database, respectively comparing the commodity packaging image in the commodity taking video and the commodity packaging image in the commodity putting-back video of each customer with the standard packaging image of each commodity to obtain the commodity taken by each customer and the commodity put-back of each customer, respectively obtaining the taking times and the putting-back times of the commodities corresponding to each customer according to the commodity taken by each customer and the commodity put-back of each customer, comparing the taking times and the putting-back times of the commodities corresponding to each customer, and screening the selected commodities of each customer.
The method comprises the steps of obtaining videos of cashier areas in target unmanned retail stores in a preset monitoring period through a high-definition camera, obtaining payment videos of all customers through interception, obtaining payment time periods of all the customers according to starting time and ending time corresponding to the payment videos of all the customers, obtaining checkout information of the target unmanned retail stores in the preset monitoring period through a checkout platform of the target unmanned retail stores, screening out checkout information corresponding to the payment time periods of all the customers according to the payment time periods of all the customers, recording the checkout information as checkout information of all the customers, and obtaining checkout commodities of all the customers according to the checkout information of all the customers.
Comparing each shopping commodity of each customer with each checkout commodity, if a certain shopping commodity of a certain customer is not in the checkout commodities, marking the shopping commodity of the customer as an order-escaping commodity of the customer, screening to obtain each order-escaping commodity of each customer, counting to obtain the number of the order-escaping commodities of each customer in a target unmanned retail store in a preset monitoring period, and marking the number as alpha i And i represents the number of the ith customer in the target unmanned retail store within the preset monitoring period, and i =1,2.
As a preferred scheme, the acquiring of the shopping videos of each customer in the target unmanned retail store within the preset monitoring period specifically includes: dividing the commodity placing areas of the target unmanned retail store according to a preset equal-area principle to obtain each commodity placing sub-area in the target unmanned retail store, and respectively arranging the cameras in each commodity placing sub-area in the target unmanned retail store according to a preset camera arrangement principle.
Setting the duration of a monitoring period, acquiring a monitoring video of each commodity placing sub-region of a target unmanned retail store in a preset monitoring period through a camera of each commodity placing sub-region of the target unmanned retail store, analyzing the monitoring video of each commodity placing sub-region of the target unmanned retail store in the preset monitoring period to obtain a video segment corresponding to each customer in each commodity placing sub-region monitoring video of the target unmanned retail store in the preset monitoring period, comparing the video segments corresponding to each customer in each commodity placing sub-region monitoring video of the target unmanned retail store in the preset monitoring period with each other to obtain a video segment corresponding to each customer in each commodity placing sub-region monitoring video of the target unmanned retail store in the preset monitoring period, splicing the video segments corresponding to each customer in each commodity placing sub-region monitoring video of the target unmanned retail store in the preset monitoring period from morning to evening to obtain a track video of each customer in the target unmanned retail store in the preset monitoring period, and recording the track video of each customer in each commodity placing sub-region monitoring video of the target unmanned retail store in the preset monitoring period as the shopping video of each customer in the target unmanned retail store.
As a preferred scheme, the specific acquisition method of each purchased commodity of each customer is as follows: comparing the picking times and the putting times of the commodities corresponding to the customers, if the picking times of the commodities corresponding to the customers are equal to the putting times, the commodities are not selected by the customers, if the picking times of the commodities corresponding to the customers are larger than the putting times, the commodities are selected by the customers, the difference value of the picking times and the putting times of the commodities is the number of the selected commodities, the number of the selected commodities corresponding to the customers is obtained by screening, the number of the selected commodities corresponding to the customers is collected, and the selected commodities of the customers are obtained.
Further, the customer basic information acquisition module acquires the stealing information of each customer in the target unmanned retail store within a preset monitoring period, and the specific process further comprises the following steps: according to the commodity taking videos and the commodity putting videos of the customers, action images corresponding to the commodities taken by the customers and action images corresponding to the commodities put back by the customers are obtained, and suspicious action image sets of the taken commodities stored in a database are extractedCombining and putting back a commodity suspicious action image set, comparing action images corresponding to commodities taken by each customer and the commodity suspicious action image set, if a certain commodity taken by a certain customer corresponds to a certain action image and belongs to the commodity suspicious action image set, marking the action corresponding to the commodity taken by the certain customer as a suspicious taking action, screening out the suspicious taking action corresponding to the commodities taken by each customer, counting the number of the suspicious taking actions corresponding to the commodities taken by each customer, and marking the suspicious taking actions as
b represents the number of times of fetching goods for the b-th time, b =1,2,.. And c, and similarly, according to the analysis method of the number of suspicious fetching actions corresponding to the goods fetched by each customer for each time, the number of suspicious fetching actions corresponding to the goods put back by each customer for each time is obtained and recorded as ∑ and ∑>
f denotes the number of times the item was put back for the f, f =1,2.
The suspicious taking action quantity corresponding to each time of taking the commodity by each customer
And the number of suspicious putting actions corresponding to each put-back commodity>
Substituted into formula>
Obtaining the times beta of plagiarism tendency actions of each customer in the target unmanned retail store within the preset monitoring period i Therein x 1 、χ 2 And weighting factors respectively representing the number of suspicious fetching actions corresponding to the preset fetched commodities and the number of suspicious retrieving actions corresponding to the retrieved commodities.
Further, the customer basic information acquisition module acquires shopping information of each customer in a target unmanned retail store within a preset monitoring period, and the specific process comprises the following steps: the method comprises the steps of obtaining the standing time of each customer in each commodity placing area in a target unmanned retail store according to a shopping video of each customer in the target unmanned retail store within a preset monitoring period, comparing the standing time of each customer in each commodity placing area in the target unmanned retail store with a preset standing time threshold, recording the standing time of a certain customer in a certain commodity placing area in the target unmanned retail store as the commodity placing area of the customer if the standing time of the certain customer in the certain commodity placing area is larger than or equal to the preset standing time threshold, recording the standing time of the customer in the commodity placing area as the staying time of the customer in the commodity placing area, and counting the staying time of each commodity placing area of each customer and the staying time of each customer in each commodity placing area.
The method comprises the steps of extracting commodity types corresponding to each commodity placing area stored in a database, screening and obtaining the commodity types corresponding to each commodity placing area of each customer according to each commodity placing area of each customer, obtaining the stay time of each type of commodity of each customer in a target unmanned retail store in a preset monitoring period according to the commodity types corresponding to each commodity placing area of each customer and the stay time of each customer in each commodity placing area, and recording the stay time as the stay time of each type of commodity of each customer in the target unmanned retail store in the preset monitoring period
j denotes the number of the jth product type, j =1,2.
Further, the customer basic information obtaining module obtains shopping information of each customer in a target unmanned retail store within a preset monitoring period, and the specific process further comprises the following steps: extracting each commodity in each type of commodity in the target unmanned retail store stored in the database, screening and obtaining a commodity type style corresponding to each commodity taken by each customer according to each commodity taken by each customer, comparing the commodity type styles corresponding to each commodity taken by each customer with each other, counting the number of times of taking each commodity in each type of commodity by each customer, and recording the number as n ijp P denotes the number of item p, p =1,2.
According to the commodity taking videos of the customers, the commodity taking time length of the customers is recorded as the commodity taking touch time length of the customers, and according to the commodity type style corresponding to the commodity taking of the customers and the commodity taking touch time length of the customers, the commodity taking touch time length corresponding to the commodities of the types of the customers is obtained.
Substituting the times of taking the commodities of various types in various types of commodities by each customer and the touch duration of the commodities of various types corresponding to various types of commodities by each customer into a formula
Obtaining the touch frequency delta of each commodity in each type of commodity of each customer in the target unmanned retail store in the preset monitoring period ijp Wherein φ represents a correction factor for the frequency of touch of customer merchandise in the targeted unmanned retail store within a predetermined monitoring period, ε 1 、ε 2 Weight factors respectively representing preset times of taking the commodities and the touch time length of the taken commodities, wherein delta n represents a preset commodity taking time threshold value, and the number of times of taking the commodities is greater than or equal to>
The method comprises the steps of representing the touch time of the ith customer for fetching the commodities at the u time corresponding to the pth commodity in the jth type commodities, wherein u =1,2.
According to the checkout commodities of the customers, comparing the checkout commodities of the customers with the commodities in the commodities of various types in the target unmanned retail store, counting the purchase quantity of the commodities of various types of the customers in the target unmanned retail store within a preset monitoring period, and recording the purchase quantity as k ijp 。
The customer theft monitoring module is used for analyzing and obtaining the theft tendency coefficient of each customer in the target unmanned retail store according to the theft information of each customer in the target unmanned retail store in the preset monitoring period.
Further, the specific process of the customer theft monitoring module is as follows: leading the target in a preset monitoring period to be free of zeroNumber of order-evading items alpha for each customer in the store i And number of plagiarism-prone actions β i Substitution formula
Obtaining theft propensity coefficients for individual customers in a targeted unmanned retail store>
Wherein η represents a predetermined theft propensity coefficient correction factor, γ, for the customer in the targeted unmanned retail store 1 、γ 2 Weight factors respectively representing preset number of order evasion commodities and times of plagiarism tendency actions of customers>
Represents a preset theft action influence factor corresponding to a single evasion commodity, and>
showing the influence factor of the theft corresponding to the preset single plagiarism tendency action.
The customer stealing behavior evaluation module is used for judging whether each customer in the target unmanned retail store steals or not according to the stealing behavior tendency coefficient of each customer in the target unmanned retail store and carrying out early warning.
Further, the specific process of the customer theft behavior evaluation module is as follows: the method comprises the steps of comparing a theft tendency coefficient of each customer in a target unmanned retail store with a preset theft tendency coefficient threshold value, if the theft tendency coefficient of a certain customer in the target unmanned retail store is larger than the preset theft tendency coefficient threshold value, the customer in the target unmanned retail store has a theft behavior, carrying out early warning through voice equipment in the target unmanned retail store, counting the customers with the theft behavior in the target unmanned retail store, and sending images of the customers with the theft behavior in the target unmanned retail store to a remote control terminal of the target unmanned retail store.
It should be noted that the invention further judges whether each customer in the target unmanned retail store has the stealing behavior by acquiring the number of the order-escaping commodities and the times of plagiarism tendency actions of each customer in the target unmanned retail store within a preset monitoring period, and performs early warning, and performs deep analysis on the stealing behavior from multiple dimensions, so as to establish an effective evaluation method aiming at the stealing behavior, thereby increasing the monitoring strength of the security of the store, and further providing guarantee for the security of the store.
The customer preference commodity type analysis module is used for obtaining preference coefficients of various types of commodities of various customers in the target unmanned retail store in the preset monitoring period according to shopping information of various customers in the target unmanned retail store in the preset monitoring period.
Further, the specific process of the customer preference commodity type analysis module is as follows: stay time of each type of commodity of each customer in the target unmanned retail store within a preset monitoring period
Frequency of touch δ of each item among each type of item ijp And the purchase quantity kappa of each item in each type of item ijp Substitution formula
Obtaining preference coefficients lambda of various types of commodities of various customers in the target unmanned retail store in a preset monitoring period ij Where m represents the total number of types of items.
The customer preference commodity type processing module is used for analyzing and obtaining the hot sales degree of each commodity type in the target unmanned retail store according to preference coefficients of each type of commodity of each customer in the target unmanned retail store in a preset monitoring period, and sending the hot sales degree of each commodity type in the target unmanned retail store to the remote control terminal of the target unmanned retail store.
Further, the specific process of the customer preference commodity type processing module is as follows: comparing preference coefficients of various types of commodities of various customers in the target unmanned retail store in a preset monitoring period, taking the commodity type corresponding to the maximum preference coefficient as the favorite commodity type of the customer, and counting to obtain the favorite commodity type of the customersThe favorite commodity types of the customers are obtained according to the favorite commodity types of the customers, the favorite customer number corresponding to each commodity type is obtained and recorded as w j Substituting the number of favorite customers corresponding to each commodity type into a formula
Obtaining the hot sales xi of each commodity type in the target unmanned retail store j Where ψ denotes a hot sales correction factor for a type of goods in a preset target unmanned retail store, and e denotes a natural constant.
And sorting the commodity types in the target unmanned retail store according to the sequence of the hot sales degree from high to low, and sending the sorted result to a remote control terminal of the target unmanned retail store.
It should be noted that the hot sales degree of each commodity type in the target unmanned retail store is obtained by obtaining the shopping information of each customer in the target unmanned retail store within the preset monitoring period, the favorite commodities of the customers are analyzed, then the types of the best-selling commodities in the store are obtained and the goods are supplemented in a targeted manner, so that blind goods supplement or supply shortage is prevented, the experience of the customers is improved, and the normal operation of the store is guaranteed.
The database is used for storing standard package images of various commodities in the target unmanned retail store, a commodity suspicious action image collection of taking commodities, a commodity suspicious action image collection of putting back, commodity types corresponding to various commodity putting areas and various commodities in various types of commodities.
The foregoing is illustrative and explanatory only of the present invention, and it is intended that the present invention cover modifications, additions, or substitutions by those skilled in the art, without departing from the spirit of the invention or exceeding the scope of the claims.
Claims (9)
1. Visual networking customer intelligence analysis management system based on artificial intelligence technique, its characterized in that includes:
the customer basic information acquisition module: the system comprises a monitoring system and a monitoring system, wherein the monitoring system is used for acquiring basic information of each customer in a target unmanned retail store in a preset monitoring period, wherein the basic information comprises shopping information and stealing information, the shopping information comprises the stay time of each type of commodity, the touch frequency of each type of commodity in each type of commodity and the purchase quantity of each type of commodity in each type of commodity, and the stealing information comprises the quantity of evasion commodities and the times of plagiarism tendency actions;
customer theft monitoring module: the system comprises a monitoring center, a monitoring center and a monitoring center, wherein the monitoring center is used for analyzing and obtaining the stealing behavior tendency coefficient of each customer in a target unmanned retail store according to the stealing information of each customer in the target unmanned retail store in a preset monitoring period;
customer theft behavior evaluation module: the system is used for judging whether each customer in the target unmanned retail store steals or not according to the steal behavior tendency coefficient of each customer in the target unmanned retail store and carrying out early warning;
the customer preference commodity type analysis module: the system comprises a monitoring center, a monitoring center and a display center, wherein the monitoring center is used for acquiring shopping information of each customer in a target unmanned retail store in a preset monitoring period;
customer preferred item type processing module: the remote control terminal is used for analyzing and obtaining the hot sales degree of each commodity type in the target unmanned retail store according to the preference coefficient of each type of commodity of each customer in the target unmanned retail store in a preset monitoring period, and sending the hot sales degree of each commodity type in the target unmanned retail store to the remote control terminal of the target unmanned retail store;
a database: the system is used for storing standard package images of various commodities in the target unmanned retail store, taking a commodity suspicious action image set, putting back the commodity suspicious action image set, commodity types corresponding to various commodity placing areas and various commodities in various commodities.
2. The system according to claim 1, wherein the system comprises: the customer basic information acquisition module acquires the stealing information of each customer in the target unmanned retail store in a preset monitoring period, and the specific process comprises the following steps:
the method comprises the steps of obtaining a shopping video of each customer in a target unmanned retail store in a preset monitoring period, obtaining a commodity taking video and a commodity placing-back video of each customer according to the shopping video of each customer in the target unmanned retail store in the preset monitoring period, further obtaining a commodity packaging image in the commodity taking video and a commodity packaging image in the commodity placing-back video of each customer, extracting standard packaging images of commodities in the target unmanned retail store stored in a database, respectively comparing the commodity packaging image in the commodity taking video and the commodity packaging image in the commodity placing-back video of each customer with the standard packaging image of each commodity to obtain the commodity taken by each customer and the commodity placed back of each customer, respectively obtaining the taking times and the placing-back times of the commodities corresponding to each customer according to the commodity taken by each customer and the commodity placed back of each customer, comparing the taking times and the placing-back times of the commodities corresponding to each customer, and screening the selected commodities of each customer;
acquiring videos of cashier desk areas in target unmanned retail stores in a preset monitoring period through a high-definition camera, intercepting payment videos of all customers, acquiring payment time periods of all customers according to start time and end time corresponding to the payment videos of all customers, acquiring checkout information of the target unmanned retail stores in the preset monitoring period through a checkout platform of the target unmanned retail stores, screening out checkout information corresponding to the payment time periods of all customers according to the payment time periods of all customers, recording the checkout information as the checkout information of all customers, and acquiring checkout commodities of all customers according to the checkout information of all customers;
comparing each shopping commodity of each customer with each checkout commodity, if a certain shopping commodity of a certain customer is not in the checkout commodities, marking the shopping commodity of the customer as an order-escaping commodity of the customer, screening to obtain each order-escaping commodity of each customer, counting to obtain the number of the order-escaping commodities of each customer in a target unmanned retail store in a preset monitoring period, and marking the number as alpha i And i represents the number of the ith customer in the target unmanned retail store within the preset monitoring period, i =1,2.
3. The system according to claim 2, wherein the system comprises: the customer basic information acquisition module acquires the stealing information of each customer in the target unmanned retail store in a preset monitoring period, and the specific process further comprises the following steps:
according to each commodity taking video and each commodity putting back video of each customer, obtaining each action image corresponding to each commodity taking time and each action image corresponding to each commodity putting back time of each customer, extracting a suspicious action image set of the taken commodity and a suspicious action image set of the put commodity stored in a database, comparing each action image corresponding to each commodity taking time of each customer with the suspicious action image set of the taken commodity, if a certain action image corresponding to a certain commodity taking time of a certain customer belongs to the suspicious action image set of the taken commodity, recording the action corresponding to the commodity taking time of the certain customer as a suspicious taking action, screening each suspicious taking action corresponding to each commodity taking time of each customer, counting the number of the suspicious taking actions corresponding to each commodity taking time of each customer, and recording the suspicious taking actions
b represents the number of times of taking the goods for the b-th time, b =1,2,.. And c, and similarly, the number of suspicious putting back actions corresponding to the goods put back by each customer is obtained according to the analysis method of the number of the suspicious taking actions corresponding to the goods taken by each customer, and is recorded as &>
f denotes the number of times the item was put back for the f-th time, f =1,2, · g;
corresponding each time of taking goods by each customer to the number of suspicious taking actions
And the number of suspicious putting actions corresponding to each put-back commodity>
Substituted into formula>
Obtaining the times beta of plagiarism tendency actions of each customer in the target unmanned retail store within the preset monitoring period i Therein x 1 、χ 2 And weighting factors respectively representing the number of suspicious fetching actions corresponding to the preset fetched commodities and the number of suspicious retrieving actions corresponding to the retrieved commodities.
4. The system according to claim 2, wherein the system comprises: the customer basic information acquisition module acquires shopping information of each customer in a target unmanned retail store in a preset monitoring period, and the specific process comprises the following steps:
according to shopping videos of all customers in a target unmanned retail store in a preset monitoring period, obtaining standing time of all customers in all commodity placing areas in the target unmanned retail store, comparing the standing time of all customers in all commodity placing areas in the target unmanned retail store with a preset standing time threshold, if the standing time of a certain customer in a certain commodity placing area in the target unmanned retail store is greater than or equal to the preset standing time threshold, recording the commodity placing area as the staying commodity placing area of the customer, recording the standing time of the customer in the staying commodity placing area as the staying time of the customer in the staying commodity placing area, and counting to obtain all staying commodity placing areas of all customers and the staying time of all customers in all staying commodity placing areas;
the method comprises the steps of extracting commodity types corresponding to each commodity placing area stored in a database, screening and obtaining the commodity types corresponding to each staying commodity placing area of each customer according to each staying commodity placing area of each customer, obtaining the staying time length of each type of commodity of each customer in a target unmanned retail store in a preset monitoring period according to the commodity types corresponding to each staying commodity placing area of each customer and the staying time length of each customer in each staying commodity placing area, and recording the staying time length as the staying time length of each type of commodity of each customer in the target unmanned retail store in the preset monitoring period
j denotes the number of the jth product type, j =1,2.
5. The system according to claim 2, wherein the system comprises: the customer basic information acquisition module acquires shopping information of each customer in a target unmanned retail store in a preset monitoring period, and the specific process further comprises the following steps:
extracting each commodity in each type of commodity in the target unmanned retail store stored in the database, screening and obtaining a commodity type style corresponding to each commodity taken by each customer according to each commodity taken by each customer, comparing the commodity type styles corresponding to each commodity taken by each customer with each other, counting the number of times of taking each commodity in each type of commodity by each customer, and recording the number as n ijp P represents the number of the pth item, p =1,2.
According to the commodity taking video of each customer, recording the commodity taking time of each customer as the commodity taking touch time of each customer, and according to the commodity type style corresponding to each commodity taking of each customer and the commodity taking touch time of each customer, obtaining the commodity taking touch time corresponding to each commodity in each type of commodity of each customer;
substituting the times of taking the commodities of various types by each customer and the touch duration of the commodities of various types corresponding to the commodities of various types by each customer into a formula
Obtaining the touch frequency delta of each commodity in each type of commodity of each customer in the target unmanned retail store in the preset monitoring period ijp Wherein phi represents a correction factor of the frequency of touch of the customer goods in the target unmanned retail store within a preset monitoring period, epsilon 1 、ε 2 Respectively representing the preset times of taking the commodities and the weight factors of the touch duration of the taken commodities, wherein delta n represents the preset threshold value of the times of taking the commodities, and the judgment result is obtained>
Indicating the touch time length of the ith customer for picking up the commodity for the u time corresponding to the pth commodity in the jth type of commodity, wherein u =1,2.
According to the checkout goods of the customers, comparing the checkout goods of the customers with the goods in the goods of various types in the target unmanned retail store, counting the number of the goods in the goods of various types of the customers in the target unmanned retail store in a preset monitoring period, and marking the number as kappa ijp 。
6. The system according to claim 1, wherein the system comprises: the specific process of the customer theft behavior monitoring module is as follows:
the number alpha of the order-escaping commodities of each customer in the target unmanned retail store in the preset monitoring period is set i And number of plagiarism-prone actions β i Substituting into formula
Obtaining theft propensity coefficients for individual customers in a targeted unmanned retail store>
Wherein η represents a predetermined target steal propensity factor correction factor, γ, for a customer in an unmanned retail store 1 、γ 2 Weight factors, theta, representing the number of the predetermined escape commodities and the number of plagiarism tendency actions of the customer 1 Shows a predetermined single theft influence factor theta corresponding to the single order-escaping commodity 2 Showing the influence factor of the theft corresponding to the preset single plagiarism tendency action.
7. The system according to claim 1, wherein the system comprises: the specific process of the customer theft behavior evaluation module is as follows:
the method comprises the steps of comparing a theft tendency coefficient of each customer in a target unmanned retail store with a preset theft tendency coefficient threshold value, if the theft tendency coefficient of a certain customer in the target unmanned retail store is larger than the preset theft tendency coefficient threshold value, then the customer in the target unmanned retail store has a theft behavior, carrying out early warning through voice equipment in the target unmanned retail store, counting the customers with the theft behavior in the target unmanned retail store, and sending images of the customers with the theft behavior in the target unmanned retail store to a remote control terminal of the target unmanned retail store.
8. The system according to claim 1, wherein the system comprises: the specific process of the customer preference commodity type analysis module is as follows:
the stay time of each type of commodity of each customer in the target unmanned retail store in the preset monitoring period
Frequency of touch δ of each item among each type of item ijp And the purchase quantity kappa of each type of goods in each type of goods ijp Substitution formula
Obtaining preference coefficients lambda of various types of commodities of various customers in the target unmanned retail store in a preset monitoring period ij Where m represents the total number of types of items.
9. The customer intelligent analysis management system of the video network based on the artificial intelligence technology as claimed in claim 1, wherein: the specific process of the customer preference commodity type processing module is as follows:
comparing preference coefficients of various types of commodities of various customers in the target unmanned retail store in a preset monitoring period, and taking the commodity type corresponding to the maximum preference coefficient as a favorite merchant of the customerCounting the favorite commodity types of each customer, obtaining the favorite customer number corresponding to each commodity type according to the favorite commodity types of each customer, and recording the favorite customer number as w j Substituting the number of favorite customers corresponding to each commodity type into a formula
Obtaining the hot sales xi of each commodity type in the target unmanned retail store j Wherein ψ represents a hot sales correction factor for a type of goods in a preset target unmanned retail store, and e represents a natural constant;
and sorting the commodity types in the target unmanned retail store in the order of high hot sales degree from low hot sales degree, and sending the sorted result to a remote control terminal of the target unmanned retail store.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211622630.3A CN115965249B (en) | 2022-12-16 | 2022-12-16 | Visual networking customer intelligent analysis management system based on artificial intelligence technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211622630.3A CN115965249B (en) | 2022-12-16 | 2022-12-16 | Visual networking customer intelligent analysis management system based on artificial intelligence technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115965249A true CN115965249A (en) | 2023-04-14 |
| CN115965249B CN115965249B (en) | 2024-01-23 |
Family
ID=87362859
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211622630.3A Active CN115965249B (en) | 2022-12-16 | 2022-12-16 | Visual networking customer intelligent analysis management system based on artificial intelligence technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115965249B (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102881100A (en) * | 2012-08-24 | 2013-01-16 | 济南纳维信息技术有限公司 | Video-analysis-based antitheft monitoring method for physical store |
| CN105894332A (en) * | 2016-04-22 | 2016-08-24 | 深圳市永兴元科技有限公司 | Commodity recommendation method, device and system based on user behavior analysis |
| CN107330684A (en) * | 2017-07-06 | 2017-11-07 | 广州联业商用机器人科技股份有限公司 | A kind of intelligent unmanned shop of management and control in high in the clouds and its automatic settlement method |
| CN108364422A (en) * | 2018-02-24 | 2018-08-03 | 广州逗号智能零售有限公司 | Self-service method and device |
| CN108551566A (en) * | 2018-04-03 | 2018-09-18 | 深圳市丰宜科技有限公司 | Shelf monitor, payment monitoring, account identifying system and method |
| CN109271847A (en) * | 2018-08-01 | 2019-01-25 | 阿里巴巴集团控股有限公司 | Method for detecting abnormality, device and equipment in unmanned clearing scene |
| CN110147723A (en) * | 2019-04-11 | 2019-08-20 | 苏宁云计算有限公司 | The processing method and system of customer's abnormal behaviour in a kind of unmanned shop |
| CN110428532A (en) * | 2019-08-13 | 2019-11-08 | 创新奇智(北京)科技有限公司 | A kind of unmanned shop management system and method claimed based on gravity with vision |
| CN110647783A (en) * | 2018-06-26 | 2020-01-03 | 杭州海康威视数字技术股份有限公司 | Unmanned goods shelf payment detection method, device and system |
| CN110689389A (en) * | 2018-07-04 | 2020-01-14 | 上海霖罕信息科技有限公司 | Computer vision-based shopping list automatic maintenance method and device, storage medium and terminal |
| CN111127174A (en) * | 2020-01-06 | 2020-05-08 | 鄂尔多斯市东驿科技有限公司 | Intelligent unmanned supermarket control system |
| CN111192425A (en) * | 2019-12-31 | 2020-05-22 | 新石器慧通（北京）科技有限公司 | Unmanned vehicle retail anti-theft method and system |
| US20210209575A1 (en) * | 2017-01-04 | 2021-07-08 | Xian Chaohi Net Technology Co., Ltd. | Method and system for identifying goods of intelligent shopping cart |
| US20220058609A1 (en) * | 2020-08-21 | 2022-02-24 | Toshiba Tec Kabushiki Kaisha | Image analysis apparatus, monitoring system, and checkout system |
| US20220076227A1 (en) * | 2018-11-20 | 2022-03-10 | Beijing Jingdong Shangke Information Technology Co, Ltd. | Self-service settlement method, apparatus and storage medium |
| CN114897123A (en) * | 2022-06-02 | 2022-08-12 | 邵鸣 | Perception recognition control system based on big data |
-
2022
- 2022-12-16 CN CN202211622630.3A patent/CN115965249B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102881100A (en) * | 2012-08-24 | 2013-01-16 | 济南纳维信息技术有限公司 | Video-analysis-based antitheft monitoring method for physical store |
| CN105894332A (en) * | 2016-04-22 | 2016-08-24 | 深圳市永兴元科技有限公司 | Commodity recommendation method, device and system based on user behavior analysis |
| US20210209575A1 (en) * | 2017-01-04 | 2021-07-08 | Xian Chaohi Net Technology Co., Ltd. | Method and system for identifying goods of intelligent shopping cart |
| CN107330684A (en) * | 2017-07-06 | 2017-11-07 | 广州联业商用机器人科技股份有限公司 | A kind of intelligent unmanned shop of management and control in high in the clouds and its automatic settlement method |
| CN108364422A (en) * | 2018-02-24 | 2018-08-03 | 广州逗号智能零售有限公司 | Self-service method and device |
| CN108551566A (en) * | 2018-04-03 | 2018-09-18 | 深圳市丰宜科技有限公司 | Shelf monitor, payment monitoring, account identifying system and method |
| CN110647783A (en) * | 2018-06-26 | 2020-01-03 | 杭州海康威视数字技术股份有限公司 | Unmanned goods shelf payment detection method, device and system |
| CN110689389A (en) * | 2018-07-04 | 2020-01-14 | 上海霖罕信息科技有限公司 | Computer vision-based shopping list automatic maintenance method and device, storage medium and terminal |
| CN109271847A (en) * | 2018-08-01 | 2019-01-25 | 阿里巴巴集团控股有限公司 | Method for detecting abnormality, device and equipment in unmanned clearing scene |
| US20220076227A1 (en) * | 2018-11-20 | 2022-03-10 | Beijing Jingdong Shangke Information Technology Co, Ltd. | Self-service settlement method, apparatus and storage medium |
| CN110147723A (en) * | 2019-04-11 | 2019-08-20 | 苏宁云计算有限公司 | The processing method and system of customer's abnormal behaviour in a kind of unmanned shop |
| CN110428532A (en) * | 2019-08-13 | 2019-11-08 | 创新奇智(北京)科技有限公司 | A kind of unmanned shop management system and method claimed based on gravity with vision |
| CN111192425A (en) * | 2019-12-31 | 2020-05-22 | 新石器慧通（北京）科技有限公司 | Unmanned vehicle retail anti-theft method and system |
| CN111127174A (en) * | 2020-01-06 | 2020-05-08 | 鄂尔多斯市东驿科技有限公司 | Intelligent unmanned supermarket control system |
| US20220058609A1 (en) * | 2020-08-21 | 2022-02-24 | Toshiba Tec Kabushiki Kaisha | Image analysis apparatus, monitoring system, and checkout system |
| CN114897123A (en) * | 2022-06-02 | 2022-08-12 | 邵鸣 | Perception recognition control system based on big data |
Non-Patent Citations (5)
| Title |
|---|
| CARLO COLOMBO: "Behavior monitoring through automatic analysis of video sequences", CIVR \'07: PROCEEDINGS OF THE 6TH ACM INTERNATIONAL CONFERENCE ON IMAGE AND VIDEO RETRIEVAL * |
| 林付春;张荣芬;何倩倩;刘宇红;: "机器视觉中的智能无人零售系统设计", 电子技术应用, no. 09 * |
| 舒渤予;: "无人超市识别技术及其应用进展", 科技与创新, no. 01 * |
| 袁路妍;李锋;: "无人值守智能商业融合平台技术框架设计", 物流技术, no. 02 * |
| 金剑锋;姚元;王明晓;曹雅楠;张磊;: "新零售时代,基于5G和AIoT技术的智慧门店探索与应用", 电信科学, no. 1 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115965249B (en) | 2024-01-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10242267B2 (en) | Systems and methods for false alarm reduction during event detection | |
| US8351647B2 (en) | Automatic detection and aggregation of demographics and behavior of people | |
| US20150006243A1 (en) | Digital information gathering and analyzing method and apparatus | |
| CN108537166B (en) | Method and device for determining shelf browsing amount and analyzing browsing amount | |
| CN103229169B (en) | Content providing and system | |
| CN111222389B (en) | Analysis method and system for commodity on super commodity shelf | |
| JP2008533543A (en) | Marketing method and system for retail environment | |
| CN109191197A (en) | Video passenger flow statistical analysis based on block chain technology | |
| CN108335317A (en) | Shopping guide method and device under a kind of line | |
| CN107179827A (en) | The intelligent interactive method and system of a kind of finance device | |
| CN113191845A (en) | Online live shopping platform data analysis processing method, system, equipment and computer storage medium | |
| WO2021129342A1 (en) | Data processing method, apparatus and device, storage medium, and computer program | |
| CN108648338A (en) | The shopping saving system method, apparatus and vending machine of automatic vending machine | |
| CN109712324A (en) | A kind of automatic vending machine image-recognizing method, good selling method and vending equipment | |
| CN113887884A (en) | Business-super service system | |
| CN115965249A (en) | Video network customer intelligent analysis management system based on artificial intelligence technology | |
| CN109978672A (en) | A kind of marketing system and method based on image recognition | |
| CN115294525A (en) | Market commodity supervisory systems | |
| CN114493673A (en) | Commodity price changing method, system, device and storage medium based on user behavior | |
| CN114821729B (en) | Commodity shopping guide method and device, cloud server and storage medium | |
| JP2010003145A (en) | Fashion analysis device, commodity trend analysis device and commodity recommendation device | |
| CN109697645A (en) | Information generating method and device for shelf | |
| CN116645375B (en) | Cigarette retail merchant inventory checking method based on three-dimensional reconstruction algorithm | |
| CN116823361B (en) | Jewelry collocation detection and pushing method based on artificial intelligence | |
| CN113128927B (en) | Double-camera-based optimization method for unmanned retail terminal hangar damage |
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 |