CN107833363B - Automatic inventory management system and method for self-service vending machine based on image recognition - Google Patents
Automatic inventory management system and method for self-service vending machine based on image recognition Download PDFInfo
- Publication number
- CN107833363B CN107833363B CN201710895892.XA CN201710895892A CN107833363B CN 107833363 B CN107833363 B CN 107833363B CN 201710895892 A CN201710895892 A CN 201710895892A CN 107833363 B CN107833363 B CN 107833363B
- Authority
- CN
- China
- Prior art keywords
- commodity
- goods
- vending machine
- self
- image
- 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
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F9/00—Details other than those peculiar to special kinds or types of apparatus
- G07F9/002—Vending machines being part of a centrally controlled network of vending machines
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
-
- 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
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F9/00—Details other than those peculiar to special kinds or types of apparatus
- G07F9/006—Details of the software used for the vending machines
Abstract
The invention discloses a self-service vending machine automatic inventory management system and method based on image recognition. The system comprises a self-service vending machine provided with an automatic lifting platform, wherein a camera device capable of shooting commodities on a replenishing position or a goods position at the outermost end of a goods channel of a goods selling position is arranged on the automatic lifting platform; the vending machine, the camera device and the automatic lifting platform at each selling position are provided with a local control system, and each local control system can be connected with a cloud service system through the Internet; when the image of the goods at the outer end of the goods supplementing position or the goods selling position of the self-service vending machine is captured by the camera device, the image is identified and then the goods database of the self-service vending machine which is in sale is updated, namely the latest stock of the self-service vending machine is used. The invention can greatly increase the types of the sold commodities and realize the full-automatic intelligent identification and perception of inventory for consumers.
Description
Technical Field
The invention relates to a self-service vending machine, in particular to a self-service vending machine automatic inventory management system and method based on image recognition.
Background
Vending machine shopping has become a growing trend for new retail. At present, most self-service vending machine containers use a front visible glass storage mode, users can obtain the goods at the sight, and the goods are widely accepted by consumers.
The self-service vending machine container comprises a plurality of goods shelves, each goods shelf comprises a plurality of goods ways, and the common goods transfer mode of the goods ways comprises a spring rotation transfer mode and a conveyor belt transfer mode. Therefore, in any mode, most of goods on the outermost side of the goods channel can be clearly seen, and effective image recognition can be carried out.
The traditional delivery mode based on a spiral spring or based on a belt transmission simply pushes out the goods, then the goods freely fall into a goods taking bin, and some goods (such as glass bottles) cannot freely fall off or can be broken. Therefore, a form of adopting the lifting goods receiving platform appears, namely, goods are firstly pushed out from the goods channel to enter the automatic lifting goods receiving platform which arrives first, and then the goods are conveyed to the goods taking bin by the automatic lifting goods receiving platform, so that the goods falling event caused by a free falling mode from a high place can be avoided.
The selling mode of current self-service vending machine mainly has:
(1) the traditional self-service vending machine purchase mode is as follows: the customer selects a particular lane (e.g., a1, the first lane of the first shelf) on the keypad of the vending machine, and then selects either cash or cashless for payment, and the particular lane is shipped, i.e., a so-called lane-driven purchase of goods.
(2) The commodity driving purchase mode is that a touch screen is arranged on a control cabinet of the self-service vending machine, pictures of selectable commodities are displayed on the touch screen, customers click the pictures to select a payment mode, and the vending machine sells the commodities corresponding to specific pictures.
(3) One novel approach is: the consumer runs the relevant self-service vending machine APP with self-service vending machine operator on own cell-phone, and the self-service vending machine of specific position is demonstrateed to the APP that sells goods, consequently clicks the self-service vending machine picture of corresponding position, can show and can sell out commodity picture or goods way selection in the appointed vending machine, and the consumer can select goods according to (1) goods way driven mode, also can click the commodity picture and select goods according to (2) goods driven mode. After the goods are selected. The payment mode can be selected, and the goods can be delivered after the payment is successful.
From the above, the above three purchasing methods are integrated to obtain two purchasing methods (the lane driving type and the commodity driving type), and both the lane driving purchasing method and the commodity driving purchasing method need to know the inventory information in the self-service vending machine in advance: i.e., which lane has which goods available for sale. Otherwise, the user suffers a loss. For example, in a lane driven purchase, if the user inadvertently selects an empty lane, the user may pay for the money but not obtain the goods. However, how to determine whether any goods are available for purchase in the vending machine? How to determine on which lanes the purchasable items are respectively located?
The traditional solution is: when the replenishment workers put the goods, the replenishment workers must strictly put the same kind of goods according to the same goods channel in a fixed mode, because the prices of the goods in the same goods channel must be kept consistent, and inventory statistics is possible only when the same goods are put in the same goods channel.
The traditional inventory statistics method is as follows: the number of the specific goods shelf of the self-service vending machine in the specific goods channel is determined by a mode of manual input in sequence through a self-service vending machine keyboard or a self-service vending machine touch screen. The method is characterized in that 60 channels of one automatic sales counter are used for calculation, and a replenishment worker needs to interact with a machine for nearly 200 times through a keyboard or a touch screen for accurate entry. Moreover, after the replenishment personnel replenish the goods, the replenishment personnel need to remember which goods channel replenishes several goods by memory, and once the memory is wrong, the replenishment information is not correct. Certainly, the replenishment person can also mend a goods way and just type once, but reduced like this and type the mistake, but can lead to lower replenishment efficiency, longer unit replenishment time to, because most packing cabinets all are the freezer, the long-time loss that can lead to the refrigerating output greatly of opening the door replenishment, the energy consumption is very big. In addition, because each replenishment worker needs to manage the operation and replenishment of a plurality of self-service vending machines scattered in different places, firstly, the manual stocking mode of manual replenishment takes a lot of time and mistakes are easy to make, and in addition, the traditional replenishment and stocking mode requires that the same commodity is placed on the same commodity channel, so that the types of the commodity which can be sold are reduced, and on the other hand, the replenishment workers are required not to be mixed for replenishment, the difficulty of manual classification is increased, and the replenishment time is prolonged.
Disclosure of Invention
The invention aims to solve the technical problem of providing an automatic inventory management system and method of a self-service vending machine based on image recognition, which can improve the types of sellable commodities and has simple inventory entry aiming at the defects of the prior art.
In order to solve the technical problems, the invention provides an automatic inventory management system of a self-service vending machine based on image recognition, which comprises the self-service vending machine, wherein the self-service vending machine is internally provided with a plurality of goods shelves, each goods shelf is provided with a plurality of goods channels, each goods channel is provided with a plurality of goods positions, each goods channel adopts spring spiral goods discharging or conveyor belt goods discharging, and the self-service vending machine is internally provided with an automatic lifting platform which can run to the outermost end of each goods channel to receive goods, wherein:
the automatic lifting platform of the self-service vending machine is provided with a camera device which can shoot goods on the outermost goods position of the goods channel at the positioning position of the automatic lifting platform;
the vending machine, the camera device and the automatic lifting platform are respectively connected with a local control system, and a sample database containing commodity names, commodity numbers, commodity prices and commodity atlas of pre-sold commodities is arranged in the local control system;
after the self-service vending machine replenishes goods each time, the local control system controls the automatic lifting platform to traverse all goods channels of all goods shelves of the self-service vending machine once, and after the automatic lifting platform moves to a goods channel of one goods shelf each time, a camera device on the automatic lifting platform takes a snapshot of images of goods on goods positions at the outermost ends of the goods channels; then the image is associated with the lane number of the shelf, the commodity name, the commodity number and the commodity price of the corresponding commodity are obtained after the image identification is carried out by the local control system, and after the commodity information is associated with the lane number of the shelf, the commodity information is stored in the local control system and the stock of the corresponding lane of the corresponding shelf of the self-service vending machine is updated;
when the self-service vending machine sells goods each time, the local control system controls the automatic lifting platform to move to a specific goods channel of a specific goods shelf where the customer selects the goods and receives a goods sent from the specific goods channel of the specific goods shelf, and then the local control system controls the camera device on the automatic lifting platform to capture an image of the goods on the goods position at the outermost end of the goods channel for selling goods; then the image is associated with the lane number of the shelf, the commodity name, the commodity number and the commodity price of the corresponding commodity are obtained after the image identification is carried out by the local control system, and after the commodity information is associated with the lane number of the shelf, the commodity information is stored in the local control system, and the stock of the corresponding lane of the corresponding shelf of the self-service vending machine is updated.
The scheme is further improved in that a plurality of same or different commodities are placed on a plurality of goods positions of the same goods channel.
In a further improvement of the above solution, the vending machine has a plurality of local control systems, each local control system of the vending machine is connected to the cloud service system through a network, a sample database containing trade names, commodity numbers, commodity prices, and commodity atlas of pre-sold commodities is provided in the cloud service system, after the image is captured by the camera device, the local control system associates the image with the lane number of the shelf and transmits the image to the cloud service system, and then the cloud service system performs image recognition to obtain the trade name, commodity number, and commodity price of the corresponding commodity, and after the commodity information is associated with the lane number of the shelf, the commodity information is transmitted to the local control system, and the stock of the corresponding lane of the corresponding shelf of the vending machine is updated.
In order to solve the technical problem, the invention provides an automatic inventory management method of an automatic inventory management system of a self-service vending machine based on image recognition, which comprises the following steps:
(1) when the self-service vending machine is powered on and started, or when the self-service vending machine finishes replenishment, or the self-service vending machine sells goods for each time, the local control system commands the camera device to respectively shoot images of goods at the outermost ends of all goods channels of the self-service vending machine, or only shoots images of goods at the outermost ends of the goods channels of the self-service vending machine during each time of selling goods;
(2) after the camera device shoots an image, the local control system associates the image with the corresponding goods channel number of the goods shelf corresponding to the image, and the image is transmitted to the local control system or the cloud service system for image recognition;
(3) during image identification, similarity identification is carried out on each image and a commodity atlas stored in a sample database in a local control system or a cloud service system respectively to obtain commodity names, commodity numbers and commodity prices of corresponding commodities, the commodity information is associated with a shelf number and a commodity channel number corresponding to an image shot by a camera device, the identification result of each image is represented by a two-dimensional array, the row of the two-dimensional array represents the shelf number, the column represents the commodity channel number, the array element is a data structure containing the commodity information, and the two-dimensional array is stored in a commodity-selling database of the cloud service system or the local control system of the self-service vending machine and is stored locally as the latest stock of the self-service vending machine and is used as the basis for the next purchase of a consumer.
And the sample database stores commodity atlas of the pre-sold commodities in the self-service vending machine, the commodity atlas of the same kind of commodities corresponds to the same commodity name and the same commodity number, and the commodity number corresponds to the commodity price of each commodity.
The commodity atlas in the sample database is a plurality of orientation photo albums of each commodity taken in the vertical direction.
The plurality of orientation photo sets at least comprise pictures of the front side and the back side of the commodity which are arranged in the vertical direction respectively.
The goods in the self-service vending machine are placed in the vertical direction, and one or more kinds of goods are placed in each goods channel.
The image recognition steps are as follows:
storing commodity names, commodity atlas, commodity numbers and commodity price associations of pre-sold commodities in a sample database of a local control system or a cloud service system;
vertically placing the pre-sold commodities in each goods space of the self-service vending machine;
the automatic lifting platform moves to an appointed goods channel of an appointed goods shelf of the local control system under the control of the local control system, then the local control system commands a camera device on the automatic lifting platform to shoot an image of a commodity, the image is associated with the goods shelf number and the goods channel number of the goods channel where the commodity is located, and the image, the associated goods shelf number and the goods channel number are transmitted and stored to the local control system or the cloud service system;
and performing similarity analysis on each image and a commodity atlas in the sample database, identifying the commodity atlas with the highest similarity as a commodity number of a corresponding commodity, further obtaining the commodity name and the commodity price of the corresponding commodity through the sample database, and associating the commodity name and the commodity price with a shelf number and a commodity way number corresponding to the image shot by the camera device, wherein the identification result of each image is represented by a two-dimensional array, the rows of the two-dimensional array represent the shelf numbers, the columns represent the commodity way numbers, and the array elements are a data structure containing the commodity numbers, the commodity prices and the commodity names.
And the information of the commodity number field in the array structure is 0, which indicates that no commodity exists, namely the commodity channel is an empty commodity channel.
The automatic inventory management system and method of the self-service vending machine based on the image recognition are supported by the national science foundation project 61272147, the operator only needs to put the pre-sold commodity on the commodity channel in the vertical direction, the same commodity can be put on the same commodity channel during placement, or different commodities can be mixed (the same commodity channel is provided with different commodities), after the completion of the replenishment, the inventory information is not needed to be manually input, but the local control system instructs the camera device to shoot the image of the replenishment or the commodity at the outermost end of the commodity channel after the sale, and the local control system (or the cloud server) is compared with the commodity atlas in the sample database for image recognition, so that the commodity name, the commodity number and the corresponding commodity price on the outermost end commodity position of the self-service vending machine are recognized and used as the latest inventory information of the self-service vending machine. Therefore, the goods selected by the customer before purchasing the goods are ensured to be really available, the single-machine replenishment time of an operator is greatly reduced, the efficiency is improved, the service quality of the customer is improved, and the types of the goods which can be sold are increased (because the same goods channel can be used for placing not only one kind of goods).
The invention has the advantages that:
1: from the viewpoint of replenishment, an operator can randomly place the types of commodities without setting the prices of the commodities in the corresponding commodity channels, and because each commodity number in the sample database is associated with the commodity price of the corresponding commodity (the price is set in the database), the commodity price of each goods position in each commodity channel can be automatically updated after the commodity is identified, and the situation that the same commodity must be placed in the same commodity channel like a traditional replenishment method is avoided;
2: from the point of view of the types of the commodities, the invention can greatly increase the types of the commodities which can be sold, because the same commodity channel does not need to be placed with the same type of commodities as the traditional self-service vending machine, but can be placed with a plurality of different commodities.
3: according to the replenishment entry mode, an operator is thoroughly liberated from the traditional manual replenishment entry mode which is heavy, time-consuming and easy to make mistakes, and full-automatic intelligent identification and perception of inventory facing to consumers are realized. The operator can leave the house only by completing the goods and closing the gate without any operation.
4: the invention is not only suitable for the self-service vending machine adopting transparent front-view glass, but also suitable for the non-transparent self-service vending machine.
Drawings
Fig. 1 is a diagram of the hardware architecture of a vending machine inventory management system based on image recognition according to the present invention.
Fig. 2 is a diagram of an application example of the automatic inventory management system of the self-service vending machine based on image recognition.
In the figure: 100-a vending machine; 200-automatic lifting platform; 201-camera (video camera); 300-goods taking port; 400-shelf; 401 — article of commerce.
Detailed Description
As shown in fig. 1 and fig. 2, the automatic inventory management system of the self-service vending machine based on image recognition of the invention comprises the following parts:
1. multiple self-service vending machines distributed at different vending positions
The self-service vending machine can adopt a transparent panel storage cabinet and also can adopt an opaque panel storage cabinet (when the self-service vending machine is opaque, a light condition in the cabinet can be set with a light-emitting device of a controllable switch), the self-service vending machine is internally provided with a plurality of goods shelves from top to bottom, each goods shelf is provided with a plurality of goods channels from left to right, each goods channel is provided with a plurality of goods positions from inside to outside, each goods channel is provided with a plurality of same or different goods, and each goods channel adopts a spring spiral goods outlet mode or a conveyor belt goods outlet mode or other arbitrary goods outlet modes. Automatic lifting platforms which can run to the outermost end of each goods channel to receive goods are arranged in the vending machines respectively.
2: camera equipment (cam or camera)
The camera lens is arranged on the automatic lifting platform of each self-service vending machine, so that the lens of the camera device can shoot commodities on the outermost goods position of the goods channel at the positioning part of the automatic lifting platform.
3: local control system
Each vending machine, the camera device and the automatic lifting platform at each vending position are provided with a local control system, and control linkage is carried out through the local control system. The local control system can be independently arranged or arranged in the self-service vending machine (the camera device is directly connected to the local control system of the self-service vending machine, and can also be associated with the corresponding self-service vending machine through a cloud service system through a network).
4: cloud service system
The local control system (when the camera device is connected with the local control system) of each self-service vending machine is connected to the cloud service system with the cloud server through the Internet, or the respective vending machine and the camera device are respectively connected to the cloud service system through the network, and the cloud service system associates the vending machine and the camera device to form an organic whole in advance).
The automatic inventory management method of the self-service vending machine based on the image recognition comprises the following steps:
(1) and storing the commodity names, commodity atlas, commodity codes and commodity price association of the pre-sold commodities in a sample database of a local control system or a cloud service system, and vertically placing the pre-sold commodities in a goods space of the self-service vending machine. The commodity atlas for each pre-sold commodity is a collection of multiple orientation (pose) photographs taken of the commodity in a vertical direction. The plurality of orientation (posture) photo sets at least comprise front pictures and back pictures of the commodity taken in the vertical direction respectively, so that when an operator replenishes the self-service vending machine, the commodity is placed in the vertical direction without strictly placing the commodity according to a certain fixed orientation. In addition, because the types of the commodities placed in the same vending machine are limited and the quantity of the commodities is not too large (generally, the quantity is at most dozens), the sample database is known and the quantity of the commodities is limited, so that the comparison and identification of a large quantity of commodity pictures are avoided, and the identification rate can be improved and the identification time can be shortened.
In this embodiment, the automatic sales counter has 6 shelves from top to bottom, each shelf has 10 lanes, each lane can hold 8 items, and the items can be placed in a mixed manner, so that the maximum number of items stored in one sales counter is 6 × 10 × 8 — 480, the maximum type of stored items is also 480 (one each), the maximum type of the items that can be sold by the consumer at a time is 60, that is, the maximum type of the stock (total number of lanes) available in the automatic sales counter at a time is 60 (if each is different, each is a single), the minimum type is 1 (each lane holds the same single), and the maximum number of the single items is 60 (one single). Without loss of generality, based on the configuration of 6 shelves, 10 lanes and 8 lanes, the total number of stored commodities is 480, the total number of the commodities is 120, the number of each commodity is 4, and each lane can be arbitrarily arranged with at least one commodity (8 single commodities) to at most 8 commodities (one commodity). All shelves of a self-service vending machine are numbered A, B, C, D.E and F from top to bottom in sequence, and the goods channel of each shelf is numbered 1, 2, 3, 4, 5, 6, … … and 10 from left to right in sequence.
(2) After the self-service vending machine is powered on and started, or after the self-service vending machine finishes replenishing goods, or after the self-service vending machine sells for one time, the local control system or the cloud service system commands the camera device to shoot the outer end goods positions of all goods channels or the images of the outmost goods of the goods channels.
(3) After the image is shot by the camera device, the image is transmitted to the local control system or the cloud service system to be subjected to image recognition, and preferably transmitted to the cloud service system to be subjected to image recognition.
(4) And respectively carrying out similarity identification on each image and a commodity atlas stored in a sample database in a local control system to obtain the commodity name, the commodity number and the commodity price of the corresponding commodity, and associating the commodity name, the commodity name and the commodity price with the shelf number corresponding to the image shot by the camera device, wherein the identification result of each image is represented by a two-dimensional array, the row of the two-dimensional array represents the shelf number, the column represents the commodity passage number, the array element is a data structure, the data structure comprises the information of the commodity number, the commodity price, the commodity name and the like), and if the information of a commodity number field in the structure is 0, the situation that no commodity exists is indicated, namely the commodity passage is an empty commodity passage. The two-dimensional array is used for updating an on-sale commodity database stored in a local control system of the self-service vending machine or a cloud service system, and the updated on-sale commodity database is stored locally as the latest stock of the self-service vending machine and is used as the basis for the next purchase of a consumer.
In this embodiment, each image needs to be subjected to similarity analysis with 240 pictures (120 kinds of products, one picture for each of the front and back sides of each product) which are the total number of the product atlas in the sample database, and the 120 kinds of products are sequentially numbered 001,002,003. . . 120 (the product number is associated with a product name, such as cola (001), milk (002), etc., and the product number is also associated with the selling price of the product), the product number with the highest similarity (or the product number with the similarity higher than a certain threshold) in 240 pictures is identified and labeled. The maximum number of images to be recognized is 60 images at each replenishment, so that the maximum number of times of similarity recognition of each image is 60 × 240 — 1440 pairs. After each sale, when only the images of the just sold channels are selected to be taken, the identification times are 1 × 240 to 240.
The similarity recognition of each image can adopt any one of the mature static image similarity recognition methods (such as deep learning DP, support vector machine SVA, component analysis PCA and the like), and the image recognition is preferentially carried out in a cloud service system. Because the identification is carried out in the cloud service system, if the calculated amount is large, or because a plurality of self-service vending machines transmit request identification to the cloud service system at the same time, a plurality of cloud servers can be adopted, distributed multithreading parallel calculation is carried out, and 1440 images can be simultaneously identified in parallel, so that the requirement of real-time calculation is met to the maximum extent.
The recognition result of each image is represented by a two-dimensional array with 6 rows by 10 columns, wherein the rows represent the shelf numbers (A, B, C.,) from top to bottom, and the columns represent the lane numbers (1, 2, 3., 10) from left to right, the array element is a data structure which contains the information of commodity numbers, commodity prices, commodity names and the like, if the information of the commodity number field in the structure is 000,001,002., 120, wherein 000 represents that the lane is empty, 001 represents that the lane has a cola stock, and 002 represents that the lane has a milk stock and the like. The two-dimensional array is stored in an on-sale commodity database corresponding to the self-service vending machine in the cloud service system, meanwhile, the two-dimensional array is remotely transmitted to a local control system of the self-service vending machine which is actually photographed, the two-dimensional array is stored locally as the latest inventory of the self-service vending machine, corresponding price information is also transmitted, and therefore the inventory information and the corresponding price information are dynamically updated.
(5) When a customer purchases goods, if the corresponding goods channel has goods, the local control system commands the automatic lifting platform to move to the corresponding goods channel position, after the goods sent by the goods channel are received to the temporary storage bin of the lifting platform, the camera device on the automatic lifting platform immediately shoots an image of the goods channel, and then the image is transmitted to the local control system or the cloud service system to be identified and updated in the goods selling database, so as to determine new stock which can be selected by the customer next time (including updating the goods code, the goods atlas and the goods price of the corresponding goods so as to be displayed on a display screen of a self-service vending machine or displayed on a mobile terminal of the customer such as a mobile phone app).
Claims (9)
1. An automatic inventory management system of a self-service vending machine based on image recognition comprises the self-service vending machine, wherein the self-service vending machine is internally provided with a plurality of goods shelves, each goods shelf is provided with a plurality of goods channels, each goods channel is provided with a plurality of goods positions, each goods channel adopts spring spiral goods delivery or conveyor belt goods delivery, and the self-service vending machine is internally provided with an automatic lifting platform which can run to the outermost end of each goods channel to receive goods,
placing a plurality of same or different commodities on a plurality of cargo positions of the same cargo channel;
the automatic lifting platform of the self-service vending machine is provided with a camera device which can shoot goods on the outermost goods position of the goods channel at the positioning position of the automatic lifting platform;
the vending machine, the camera device and the automatic lifting platform are respectively connected with a local control system, and a sample database containing commodity names, commodity numbers, commodity prices and commodity atlas of pre-sold commodities is arranged in the local control system;
after the self-service vending machine replenishes goods each time, the local control system controls the automatic lifting platform to traverse all goods channels of all goods shelves of the self-service vending machine once, and after the automatic lifting platform moves to a goods channel of one goods shelf each time, a camera device on the automatic lifting platform takes a snapshot of images of goods on goods positions at the outermost ends of the goods channels; then the image is associated with the lane number of the shelf, the commodity name, the commodity number and the commodity price of the corresponding commodity are obtained after the image identification is carried out by the local control system, and the commodity information is stored in the local control system after being associated with the lane number of the shelf and the inventory of the corresponding lane of the corresponding shelf of the self-service vending machine is updated;
when the self-service vending machine sells goods each time, the local control system controls the automatic lifting platform to move to a goods channel of a goods shelf where the goods are selected by a customer and receives one goods sent out from the goods channel of the goods shelf, and then the local control system controls the camera device on the automatic lifting platform to capture an image of the goods on the goods position at the outermost end of the goods channel; then the image is associated with the lane number of the shelf, the local control system performs image recognition to obtain the commodity name, the commodity number and the commodity price of the corresponding commodity, and after the commodity information is associated with the lane number of the shelf, the commodity information is stored in the local control system, and the inventory of the corresponding lane of the corresponding shelf of the self-service vending machine is updated.
2. The image recognition based vending machine inventory management system of claim 1, it is characterized in that the self-service vending machine is provided with a plurality of self-service vending machines, the local control systems of the self-service vending machines are respectively connected to the cloud service system through the network, the cloud service system is internally provided with a sample database containing commodity names, commodity numbers, commodity prices and commodity atlas of pre-sold commodities, when the camera device takes a snapshot of an image, the local control system associates the image with the goods channel number of the goods shelf and transmits the image to the cloud service system, then the image recognition is carried out by the cloud service system to obtain the commodity name, the commodity number and the commodity price of the corresponding commodity, after being associated with the lane number of the shelf, the commodity information is transmitted to a local control system, and the stock of the corresponding lane of the corresponding shelf of the self-service vending machine is updated.
3. A method for automatic inventory management of a vending machine automatic inventory management system based on image recognition according to any of claims 1-2, characterized in that it comprises the following steps:
(1) when the self-service vending machine is powered on and started, or when the self-service vending machine finishes replenishment, or the self-service vending machine sells goods for each time, the local control system commands the camera device to respectively shoot images of goods at the outermost ends of all goods channels of the self-service vending machine, or only shoots images of goods at the outermost ends of the goods channels of the self-service vending machine during each time of selling goods;
(2) after the camera device shoots an image, the local control system associates the image with the corresponding goods channel number of the goods shelf corresponding to the image, and the image is transmitted to the local control system or the cloud service system for image recognition;
(3) during image identification, similarity identification is carried out on each image and a commodity atlas stored in a sample database in a local control system or a cloud service system respectively to obtain commodity names, commodity numbers and commodity prices of corresponding commodities, the commodity information is associated with a shelf number and a commodity channel number corresponding to an image shot by a camera device, the identification result of each image is represented by a two-dimensional array, the row of the two-dimensional array represents the shelf number, the column represents the commodity channel number, the array element is a data structure containing the commodity information, and the two-dimensional array is stored in a commodity-selling database of the cloud service system or the local control system of the self-service vending machine and is stored locally as the latest stock of the self-service vending machine and is used as the basis for the next purchase of a consumer.
4. The automatic inventory management method of the automatic inventory management system of the self-service vending machine based on the image recognition as claimed in claim 3, wherein a commodity atlas of the pre-sold commodities in the self-service vending machine is stored in the sample database, the commodity atlas of the same commodity corresponds to the same commodity name and commodity number, and the commodity number corresponds to the commodity price of each commodity.
5. The automatic inventory management method of an image recognition based vending machine automatic inventory management system according to claim 4, wherein the commodity atlas in the sample database is a plurality of orientation photo albums of each commodity taken in a vertical direction.
6. The method of claim 5, wherein the plurality of sets of oriented photographs include at least one each of front and back photographs of the goods in a vertical orientation.
7. The automatic inventory management method of the image recognition-based automatic inventory management system of the vending machine as recited in claim 3, wherein the goods in the vending machine are arranged in a vertical direction, and each goods channel is arranged with one or more goods.
8. The automatic inventory management method of an image recognition based vending machine automatic inventory management system according to claim 3, wherein the image recognition comprises the steps of:
storing commodity names, commodity atlas, commodity numbers and commodity price associations of pre-sold commodities in a sample database of a local control system or a cloud service system;
vertically placing the pre-sold commodities in each goods space of the self-service vending machine;
the automatic lifting platform moves to an appointed goods channel of an appointed goods shelf of the local control system under the control of the local control system, then the local control system commands a camera device on the automatic lifting platform to shoot an image of a commodity, the image is associated with the goods shelf number and the goods channel number of the goods channel where the commodity is located, and the image, the associated goods shelf number and the goods channel number are transmitted and stored to the local control system or the cloud service system;
and performing similarity analysis on each image and a commodity atlas in the sample database, identifying the commodity atlas with the highest similarity as a commodity number of a corresponding commodity, further obtaining the commodity name and the commodity price of the corresponding commodity through the sample database, and associating the commodity name and the commodity price with a shelf number and a commodity way number corresponding to the image shot by the camera device, wherein the identification result of each image is represented by a two-dimensional array, the rows of the two-dimensional array represent the shelf numbers, the columns represent the commodity way numbers, and the array elements are a data structure containing the commodity numbers, the commodity prices and the commodity names.
9. The automatic inventory management method of the automatic inventory management system of the self-service vending machine based on the image recognition as recited in claim 8, wherein the information of the goods number field in the data structure is 0, which indicates that no goods exist, i.e. the goods channel is an empty goods channel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710895892.XA CN107833363B (en) | 2017-09-28 | 2017-09-28 | Automatic inventory management system and method for self-service vending machine based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710895892.XA CN107833363B (en) | 2017-09-28 | 2017-09-28 | Automatic inventory management system and method for self-service vending machine based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107833363A CN107833363A (en) | 2018-03-23 |
| CN107833363B true CN107833363B (en) | 2020-11-10 |
Family
ID=61644022
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710895892.XA Active CN107833363B (en) | 2017-09-28 | 2017-09-28 | Automatic inventory management system and method for self-service vending machine based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107833363B (en) |
Families Citing this family (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108416901A (en) * | 2018-03-27 | 2018-08-17 | 合肥美的智能科技有限公司 | Method and device for identifying goods in intelligent container and intelligent container |
| CN108648334A (en) * | 2018-04-11 | 2018-10-12 | 合肥美的智能科技有限公司 | Self-service cabinet and its abnormal method for controlling reporting, self-service system |
| CN108945912B (en) * | 2018-04-24 | 2020-02-18 | 深圳奥比中光科技有限公司 | Intelligent replenishment system and method |
| CN108711021A (en) * | 2018-04-25 | 2018-10-26 | 广州葵翼信息科技有限公司 | A kind of merchandise control method of automatic selling counter |
| CN110491026A (en) * | 2018-05-15 | 2019-11-22 | 苏璇 | A kind of sales counter |
| CN108765704B (en) * | 2018-05-23 | 2020-07-24 | 济南每日优鲜便利购网络科技有限公司 | Replenishment control method, device and equipment for automatic vending system and storage medium |
| CN108922026B (en) * | 2018-07-09 | 2020-06-23 | 深圳码隆科技有限公司 | Replenishment management method and device for vending machine and user terminal |
| CN108986310A (en) * | 2018-07-10 | 2018-12-11 | 合肥美的智能科技有限公司 | Vending machine and management method thereof |
| CN109118200A (en) * | 2018-07-26 | 2019-01-01 | 上海凯景信息技术有限公司 | A kind of commodity identification and cash register system based on image recognition |
| CN109272638A (en) * | 2018-08-22 | 2019-01-25 | 湖南金码智能设备制造有限公司 | A kind of automatic vending machine, intelligent inventory monitoring system and method |
| CN109377650A (en) * | 2018-08-22 | 2019-02-22 | 湖南金码智能设备制造有限公司 | A kind of automatic vending machine, vending system and good selling method |
| CN109272647A (en) * | 2018-08-29 | 2019-01-25 | 北京华沁智联科技有限公司 | The update method and device of automatic vending warehouse item state |
| CN109389745A (en) * | 2018-08-29 | 2019-02-26 | 广州伊思高科技有限公司 | A kind of intelligent goods selling equipment and method based on image procossing |
| CN110910570A (en) * | 2018-09-18 | 2020-03-24 | 威海新北洋数码科技有限公司 | Goods channel configuration method, vending machine and mobile terminal |
| CN109147181A (en) * | 2018-09-30 | 2019-01-04 | 湖南金码智能设备制造有限公司 | A kind of automatically vending system of anti-theft type, out pallet piling up method and test method |
| CN109523694A (en) * | 2018-10-22 | 2019-03-26 | 南京云思创智信息科技有限公司 | A kind of retail trade system and method based on commodity detection |
| CN109544783A (en) * | 2018-11-26 | 2019-03-29 | 湖南金码智能设备制造有限公司 | A kind of vending machine intelligent inventory update method and system based on video content recognition |
| CN109711473A (en) * | 2018-12-29 | 2019-05-03 | 北京沃东天骏信息技术有限公司 | Item identification method, equipment and system |
| CN109767559A (en) * | 2019-01-21 | 2019-05-17 | 深圳市智莱科技股份有限公司 | The method that the cargo path of automatic vending equipment defines |
| CN109859384B (en) * | 2019-01-25 | 2021-08-31 | 深圳市乐科智控科技有限公司 | Information management method, device and storage medium |
| CN110084649A (en) * | 2019-04-29 | 2019-08-02 | 深圳正品创想科技有限公司 | Intelligent vending machine information processing method and its device, server |
| CN110321884B (en) * | 2019-06-13 | 2020-11-13 | 贝式计算(天津)信息技术有限公司 | Method and device for identifying serial number |
| CN110310414B (en) * | 2019-07-15 | 2023-09-08 | 深圳市道中创新科技有限公司 | Lifting device capable of conducting inspection, intelligent vending device and inspection method thereof |
| CN110379074A (en) * | 2019-07-24 | 2019-10-25 | 成都我搜挖信息技术股份有限公司 | A kind of system and method for the automatic cycle shipment of vending machine |
| CN110390763A (en) * | 2019-07-24 | 2019-10-29 | 成都我搜挖信息技术股份有限公司 | A kind of Self-help vending machine automated stock control system and method based on image recognition |
| CN111968300B (en) * | 2019-08-13 | 2022-04-19 | 上海爱餐机器人(集团)有限公司 | Automatic identification method, device, system, medium and vending machine for sold dish packaging boxes |
| CN111161346B (en) * | 2019-12-30 | 2023-09-12 | 北京三快在线科技有限公司 | Method and device for layering commodities in goods shelves and electronic equipment |
| CN111985875A (en) * | 2020-08-25 | 2020-11-24 | 北京意锐新创科技有限公司 | Method and device for obtaining sales information of vending machine and vending machine |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN206340058U (en) * | 2016-09-27 | 2017-07-18 | 王国峰 | A kind of solid shop/brick and mortar store merchandise sales management system based on video flowing and RFID |
| CN107098096A (en) * | 2017-05-03 | 2017-08-29 | 盐城晟阳电子科技有限公司 | A kind of automated three-dimensional storage facilities |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100775083B1 (en) * | 2005-11-02 | 2007-11-12 | 게이트마켓 주식회사 | Taking-out device of vending machine |
| US8189855B2 (en) * | 2007-08-31 | 2012-05-29 | Accenture Global Services Limited | Planogram extraction based on image processing |
| JP2010006557A (en) * | 2008-06-27 | 2010-01-14 | Toshiba Tec Corp | Article management system |
| CN102849386B (en) * | 2012-08-15 | 2014-08-06 | 杭州扬果科技有限公司 | Lift type touch screen intelligent warehouse |
| CN105190655B (en) * | 2013-03-04 | 2018-05-18 | 日本电气株式会社 | Article management system, information processing equipment and its control method and control program |
| US9349238B2 (en) * | 2013-03-13 | 2016-05-24 | Pantry Retail, Inc. | Vending kit and method |
| CN104112216A (en) * | 2013-04-22 | 2014-10-22 | 学思行数位行销股份有限公司 | Image identification method for inventory management and marketing |
| CN105913566B (en) * | 2016-04-10 | 2019-01-15 | 河南优谷智能科技有限公司 | A kind of large capacity automatic vending machine for avoiding kinds of goods from damaging |
| CN106251108A (en) * | 2016-08-03 | 2016-12-21 | 杭州图钉网络科技有限公司 | A kind of leaflet intelligent warehousing system and management method |
-
2017
- 2017-09-28 CN CN201710895892.XA patent/CN107833363B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN206340058U (en) * | 2016-09-27 | 2017-07-18 | 王国峰 | A kind of solid shop/brick and mortar store merchandise sales management system based on video flowing and RFID |
| CN107098096A (en) * | 2017-05-03 | 2017-08-29 | 盐城晟阳电子科技有限公司 | A kind of automated three-dimensional storage facilities |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107833363A (en) | 2018-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107833363B (en) | Automatic inventory management system and method for self-service vending machine based on image recognition | |
| CN107730168B (en) | Automatic vending machine inventory management system and method based on image recognition | |
| US11587035B2 (en) | System and method for order fulfillment | |
| CN109214751B (en) | Intelligent inventory management system based on inventory position change | |
| CN107240000B (en) | Self-service vending method, system and electronic equipment | |
| US9483790B2 (en) | Systems and methods for providing goods | |
| US10074083B2 (en) | Retail convenience market technology with enhanced vendor administration and user mobile app functionality | |
| US10185984B2 (en) | Delivery of remotely ordered items to the current location of a user when geographic information indicates that the user is within a predetermined area associated with a physical store | |
| CN108564444B (en) | A kind of automatic vending management system and management method | |
| US20120303154A1 (en) | Retail location robotic wall system | |
| US20100234986A1 (en) | Method and systems for collecting inventory and marketing data, providing data and video services | |
| US20210374815A1 (en) | An automated store | |
| US20060102645A1 (en) | Method and apparatus for defining and utilizing product location in a vending machine | |
| WO2019222246A1 (en) | Systems and methods for automated storage and retrieval | |
| CN110119914B (en) | Object replenishment processing method, device and system | |
| CN110503521B (en) | Rental information processing method, device, server and system of intelligent sales terminal | |
| WO2018048688A1 (en) | Apparatus and method for providing item interaction with a virtual store | |
| CN111512349A (en) | Unmanned shop system | |
| CN111742339A (en) | Shop | |
| US20140330682A1 (en) | Vending Machine for Selling Mobile Phone Accessories | |
| JP6745786B2 (en) | Store | |
| CN104169951A (en) | Context based scanning and reading systems and methods | |
| CN112950839B (en) | Commodity selling method | |
| CN106875578A (en) | Laser disc automatic lending machine and its system, method | |
| CN113554801A (en) | Unmanned vending system |
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 |