Unmanned store access control monitoring system

Abstract

The invention discloses an access control monitoring system for an unmanned store, which is characterized in that a video frame picture of a walking track is acquired by acquiring and preprocessing a video of a customer going in and out of the unmanned store and a video of the walking track in the unmanned store, and the video frame picture of the walking track is analyzed to acquire a result of judging whether an article is taken at a position where a commodity shelf stays; inquiring and pricing according to the type of the fetched article with a real fetched article judgment result in the customer fetched article judgment unit, and judging whether the payment amount of the customer is matched with the total price of the fetched article obtained by the customer fetched article pricing unit or not by the automatic charging equipment so as to judge whether to pass or not; and the effects of automatic analysis according to the articles taken by the customers and access management in cooperation with access control are achieved.

Description

Technical Field

[0001] The present invention belongs to the field of monitoring systems, and more specifically, relates to an access control monitoring system for unmanned stores. Background Art

[0002] The existing intelligent access control management system for unattended convenience stores has realized unmanned, intelligent, and efficient access management through modern technological means such as face recognition, iris recognition, and RFID technology. The system has the following functions: Face recognition: When a customer enters the store, the system automatically recognizes the customer's face through a camera and establishes a file to achieve fast passage. Iris recognition: For member customers, the system can directly verify their identities through iris recognition technology to realize functions such as member discounts and points redemption. RFID technology: When goods enter and exit, the system automatically records the goods information for convenient inventory management. Remote monitoring: The store owner can view the store monitoring images in real time through the mobile APP at any time to understand the store situation.

[0003] The defect of the existing technology is that although the existing intelligent access control management system for unattended convenience stores has realized automatically recognizing the customer's face through a camera and establishing a file to achieve fast passage, it lacks a system that can automatically analyze whether there are unpaid goods based on the items taken by the customer and cooperate with the access control for access management. Summary of the Invention

[0004] In view of the problem that the existing access control monitoring system for unmanned stores lacks automatic analysis based on the items taken by the customer and cooperation with the access control for access management, the present invention provides an access control monitoring system for unmanned stores.

[0005] To achieve the above technical objectives, the technical solution adopted by the present invention is as follows:

[0006] An access control monitoring system for unmanned stores includes a video monitoring unit inside and outside the unmanned store, a customer walking trajectory capturing unit inside the store, a customer item taking judgment unit, a customer item taking pricing unit, an access control unit, and a data storage unit;

[0007] The video monitoring unit inside and outside the unmanned store is used to collect videos of customers entering and leaving outside the unmanned store and videos of the walking trajectories inside the unmanned store;

[0008] The customer walking trajectory capturing unit inside the store is communicatively connected to the video monitoring unit inside and outside the unmanned store and is used to preprocess the videos of all walking trajectories inside the store of the customer to obtain video frame pictures of the walking trajectories;

[0009] The customer item taking judgment unit is communicatively connected to the customer walking trajectory capturing unit inside the store and is used to analyze based on the video frame pictures of the walking trajectories obtained by the customer walking trajectory capturing unit inside the store to obtain a judgment result on whether an item is taken at the position where the customer stays on the commodity shelf;

[0010] The customer item picking pricing unit is communicatively connected to the customer item picking judgment unit, and is used to query the price according to the types of picked items for which the item picking judgment result in the customer item picking judgment unit is true, and calculate the total price.

[0011] The access passage unit is communicatively connected to the customer item picking pricing unit, and judges whether to pass by determining whether the checkout amount of the customer matches the total price of the picked items obtained by the customer item picking pricing unit through an automatic toll collection device.

[0012] The data storage unit is used to store customer information, videos of customers outside the unmanned store, videos of customers walking inside the unmanned store, and item information.

[0013] Furthermore, the detailed steps for preprocessing the video of a customer walking inside the unmanned store are as follows:

[0014] Split the video of the walking trajectory inside the unmanned store into video streams to obtain all the video frame pictures of the customer's walking trajectory inside the unmanned store.

[0015] Compare all the video frame pictures of the customer's walking trajectory inside the unmanned store with the pictures of the commodity placement areas pre-stored in the data storage unit using the differential image comparison algorithm to obtain the video frame pictures of the commodity placement areas in the video frame pictures of the walking trajectory and the video frame pictures of invalid walking trajectories.

[0016] Delete the video frame pictures of invalid walking trajectories, and send the video frame pictures of the commodity placement areas in the walking trajectory to the customer item picking judgment unit as the judgment data for whether an item is picked at the position where the commodity shelf stays.

[0017] Furthermore, the detailed steps for judging whether an item is picked at the position where the commodity shelf stays are as follows:

[0018] Match and query the corresponding item information in the data storage unit through the video frame pictures of the commodity placement areas in the walking trajectory. The item information includes item weight, item quantity, item unit price, the position of the shelf where it is located, item code, and item appearance picture data.

[0019] Perform differential comparison between the previous frame picture and the next frame picture in the video frame pictures of the commodity placement areas in the walking trajectory to judge whether the quantity of the items at the position of the shelf where the passed items are located has decreased.

[0020] Query whether the weight of the items at the position of the shelf where the corresponding items are located has decreased during the time period between the previous frame picture and the next frame picture, and cooperate to judge whether the quantity of the items has decreased.

[0021] If the quantity of items at the shelf position where the item is located decreases and the weight of the item also decreases, it is determined that the item has been picked up. If the quantity of items at the shelf position where the item is located does not decrease, it is determined that the customer has not picked up any items during the walking trajectory in the item display area picture.

[0022] Furthermore, according to the item picking judgment results in the video frames of the traversed walking trajectory, obtain the item picking information with a true result, and query the corresponding item unit price of the quantity of items picked up by the customer, the item code, and the item appearance picture data in the data storage unit; and summarize and calculate the total price and total weight of the items picked up by the customer.

[0023] Furthermore, the automatic toll collection device is equipped with a camera, an item code collection device, a weight collector, a scanning toll collection device, and a display screen;

[0024] The automatic toll collection device collects pictures of all items that need to be settled for the items picked up by the customer through the camera, and at the same time collects the item codes of the items that need to be settled for the items picked up by the customer through the item code collection device, and matches and queries the item unit price in the data storage unit;

[0025] Summarize all item unit prices to calculate the customer's checkout amount and feedback it to the display screen;

[0026] After the customer pays the corresponding fee through the scanning toll collection device, compare it with the total price of the items picked up obtained by the item picking pricing unit of the customer; judge whether the fee paid by the customer through the scanning toll collection device is consistent with the total price of the items picked up obtained by the item picking pricing unit of the customer. If they are consistent, the access control unit releases. If they are inconsistent, the broadcast unit notifies the customer that there are items that have not been settled and need to be settled before passing.

[0027] Furthermore, collect data on the customer entering the store and leaving the store and perform analysis by collecting the videos of the customer entering and leaving outside the unmanned store.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] Through the collection of the videos of the customer entering and leaving outside the unmanned store and the videos of the walking trajectory inside the unmanned store, obtain the video frames of the walking trajectory after preprocessing, analyze the video frames of the walking trajectory, obtain the judgment result of whether an item is picked up at the position where the item stays on the commodity shelf, query and price according to the types of picked-up items with a true result in the customer item picking judgment unit, and the automatic toll collection device judges whether the customer's checkout amount matches the total price of the items picked up obtained by the item picking pricing unit of the customer, and then realizes the judgment of whether to pass; it achieves the function of automatically analyzing the items picked up by the customer and cooperating with the access control for entry and exit management. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1The present invention is an overall flow chart of an unmanned store access control monitoring system in an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to facilitate the understanding of those skilled in the art, the present invention is further described below in conjunction with embodiments and drawings. The contents mentioned in the implementation modes are not intended to limit the present invention.

[0032] like Figure 1 As shown, this embodiment provides an unmanned store access control monitoring system, including an unmanned store internal and external video monitoring unit, a customer store walking trajectory capture unit, a customer take item judgment unit, a customer take item pricing unit, an access control access unit and a data storage unit;

[0033] The video monitoring unit inside and outside the unmanned store is used to collect videos of customers entering and leaving the unmanned store and the videos of customers walking inside the unmanned store;

[0034] The customer's walking trajectory capture unit in the store is connected to the video monitoring unit inside and outside the unmanned store for preprocessing the video of all the walking trajectories of the customer in the store and obtaining the video frame images of the walking trajectories;

[0035] The customer picking up item judgment unit is in communication connection with the customer walking track capture unit in the store, and is used to obtain a judgment result of whether to pick up items at the stop position of the commodity shelf according to the video frame image analysis of the walking track acquired by the customer walking track capture unit in the store;

[0036] The customer-taken-item pricing unit is in communication with the customer-taken-item judgment unit, and is used to query and price the items according to the types of items for which the judgment result of the customer-taken-item judgment unit is true, and calculate the total price;

[0037] The access control unit is connected to the customer item taking pricing unit, and the automatic charging device determines whether the customer's checkout amount matches the total price of the items taken obtained by the customer item taking pricing unit, thereby realizing the pass judgment;

[0038] The data storage unit is used to store customer information, videos of customers entering and leaving the unmanned store, videos of customers walking inside the unmanned store, and item information.

[0039] Detailed steps for preprocessing videos of people walking in unmanned stores:

[0040] Split the video stream of the walking trajectory in the unmanned store to obtain all the video frame images of the customer's walking trajectory in the unmanned store;

[0041] All video frame pictures of the walking trajectory of customers in the unmanned store are compared with the pictures of the commodity placement areas pre-stored in the data storage unit using the differential image comparison algorithm to obtain the video frame pictures of the walking trajectory in the commodity placement area pictures and the video frame pictures of the invalid walking trajectory;

[0042] Differential image comparison algorithm: The principle is to compare the image to be detected with a standard image to find the obvious differences (i.e., defects) between the image to be detected and the standard image (ideal image). The standard image can be obtained by training several OK product images, or by processing a single OK product image. After training, a standard image and a variation image (variation image) are obtained. The variation image contains the allowable range of gray values of each pixel point Q in the image.

[0043] The standard image and the variation image are used to create a variation model. In this way, other images can be compared with the variation model.

[0044] Delete the video frame pictures of the invalid walking trajectory, and send the video frame pictures of the walking trajectory in the commodity placement area pictures to the customer item-taking judgment unit as the judgment data for whether an item is taken at the commodity shelf staying position.

[0045] Detailed steps for judging whether an item is taken at the commodity shelf staying position:

[0046] Match and query the corresponding item information in the data storage unit through the video frame pictures of the walking trajectory in the commodity placement area pictures. The item information includes item weight, item quantity, item unit price, the position of the shelf where it is located, item code, and item appearance picture data;

[0047] Perform differential comparison between the previous frame picture and the next frame picture in the video frame pictures of the walking trajectory in the commodity placement area pictures to judge whether the quantity of the items at the position of the shelf where the passed items are located has decreased;

[0048] Query whether the weight of the items at the position of the shelf where the corresponding items are located has decreased during the time period between the previous frame picture and the next frame picture, and cooperate to judge whether the quantity of the items has decreased;

[0049] If the quantity of the items at the position of the shelf where the items are located has decreased and the weight of the items has also decreased, it is judged that the commodity has been grabbed. If the quantity of the items at the position of the shelf where the items are located has not decreased, it is judged that the customer has not taken any items in the walking trajectory in the commodity placement area pictures.

[0050] According to the item picking judgment results in the video frame pictures of the traversed walking trajectory, obtain the item picking information with a true result, and query the corresponding item unit prices of the customer's picked item quantity, item code, and item appearance picture data in the data storage unit; and summarize and calculate the total price and total weight of the customer's picked items.

[0051] The automatic toll collection device is provided with a camera, an item code acquisition device, a weight collector, a scanning toll collection device, and a display screen;

[0052] The automatic toll collection device collects all item pictures that need to be settled for the items picked by the customer through the camera, and at the same time collects the item codes of the items that need to be settled for the items picked by the customer through the item code acquisition device, and matches and queries the item unit prices in the data storage unit;

[0053] Summarize all item unit prices to calculate the customer's checkout amount and feedback it to the display screen;

[0054] After the customer scans the toll collection device to pay the corresponding fee, it is compared with the total price of the picked items obtained by the picked item pricing unit of the customer; it is judged whether the corresponding fee paid by the customer scanning the toll collection device is consistent with the total price of the picked items obtained by the picked item pricing unit of the customer. If they are consistent, the access control unit releases. If they are inconsistent, the customer is notified through the broadcast unit that there are items that have not been settled and need to be settled before passing.

[0055] Collect data on the customer entering the store and leaving the store and analyze it by collecting the video of the customer entering and leaving outside the unmanned store.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] Through the video collection of the customer entering and leaving outside the unmanned store and the video of the walking trajectory inside the unmanned store, preprocess to obtain the video frame pictures of the walking trajectory, analyze the video frame pictures of the walking trajectory, obtain the judgment result of whether an item is picked at the position where the customer stays on the commodity shelf, query and price according to the types of picked items with a true judgment result in the customer picked item judgment unit, and the automatic toll collection device judges whether the customer's checkout amount matches the total price of the picked items obtained by the picked item pricing unit of the customer, so as to realize the judgment of whether to pass; it achieves the function of automatically analyzing the items picked by the customer and cooperating with the access control for access management.

[0058] The above provides a detailed introduction to an unmanned store access control monitoring system provided by the present application. The description of the specific embodiments is only used to help understand the method and its core idea of the present application. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the principle of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

Claims

1. An unmanned store access control and monitoring system, characterized in that, It includes a video monitoring unit inside and outside the unmanned store, a customer walking trajectory capturing unit inside the store, a customer item picking judgment unit, a customer item picking pricing unit, an access control unit, and a data storage unit; The video monitoring unit inside and outside the unmanned store is used to collect videos of customers entering and leaving outside the unmanned store and videos of the walking trajectories inside the unmanned store; The customer walking trajectory capturing unit inside the store is communicatively connected to the video monitoring unit inside and outside the unmanned store and is used to preprocess the videos of all walking trajectories of customers inside the store to obtain video frame pictures of the walking trajectories; The customer item picking judgment unit is communicatively connected to the customer walking trajectory capturing unit inside the store and is used to analyze the video frame pictures of the walking trajectories obtained by the customer walking trajectory capturing unit inside the store to obtain a judgment result on whether an item is picked at the position where the commodity shelf is stayed; The customer item picking pricing unit is communicatively connected to the customer item picking judgment unit and is used to query and price according to the types of picked items with the judgment result of item picking being true in the customer item picking judgment unit and calculate the total price; The access control unit is communicatively connected to the customer item picking pricing unit and judges whether the customer's checkout amount matches the total price of the picked items obtained by the customer item picking pricing unit through an automatic charging device, thereby realizing the judgment of whether to pass; The data storage unit is used to store customer information, videos of customers entering and leaving outside the unmanned store, videos of walking inside the unmanned store, and item information.

2. The access control monitoring system for unmanned stores according to claim 1, wherein Detailed steps for preprocessing the videos of walking inside the unmanned store: Split the videos of the walking trajectories inside the unmanned store into video streams to obtain all video frame pictures of the walking trajectories of customers inside the unmanned store; Compare all the video frame pictures of the walking trajectories of customers inside the unmanned store with the pictures of the commodity placement areas pre-stored in the data storage unit using the differential image comparison algorithm to obtain video frame pictures of the walking trajectories in the pictures of the commodity placement areas and video frame pictures of invalid walking trajectories; Delete the video frame pictures of invalid walking trajectories and send the video frame pictures of the walking trajectories in the pictures of the commodity placement areas to the customer item picking judgment unit as judgment data on whether an item is picked at the position where the commodity shelf is stayed.

3. The access control monitoring system for unmanned stores according to claim 2, characterized in that Detailed steps for judging whether an item is picked at the position where the commodity shelf is stayed: Match and query the corresponding item information in the data storage unit through the video frame pictures of the walking trajectories in the pictures of the commodity placement areas. The item information includes item weight, item quantity, item unit price, shelf position, item code, and item appearance picture data; Perform differential comparison between the previous frame picture and the next frame picture in the video frame pictures of the walking trajectories in the pictures of the commodity placement areas to judge whether the quantity of items at the shelf position where the passed items are located has decreased; Query whether the weight of the items at the shelf position where the corresponding items are located has decreased during the time period between the previous frame picture and the next frame picture, and cooperate to judge whether the quantity of items has decreased; If the quantity of items at the shelf position where the items are located decreases and at the same time the weight of the items decreases, it is judged that the commodity has been grabbed. If the quantity of items at the shelf position where the items are located does not decrease, it is judged that the customer has not picked up any items in the walking trajectories in the pictures of the commodity placement areas.

4. The unmanned store access control and monitoring system according to claim 3, characterized in that, Based on the item picking judgment results in the video frame pictures of the traversed walking trajectory, obtain the item picking information with a true result, and query the corresponding item unit prices of the customer's picked item quantity, item code, and item appearance picture data in the data storage unit; and summarize and calculate the total price and total weight of the items picked by the customer.

5. The unmanned store access control and monitoring system according to claim 4, characterized in that, The automatic toll collection device is equipped with a camera, an item code collection device, a weight collector, a scanning toll collection device, and a display screen; The automatic toll collection device collects all item pictures that need to be settled for the items picked by the customer through the camera, and at the same time collects the item codes of the items that need to be settled for the items picked by the customer through the item code collection device, and matches and queries the item unit prices in the data storage unit; Summarize all item unit prices to calculate the customer's checkout amount and feedback it to the display screen; After the customer scans the toll collection device to pay the corresponding fee, compare it with the total price of the picked items obtained by the item pricing unit for the items picked by the customer; judge whether the fee paid by the customer scanning the toll collection device is consistent with the total price of the picked items obtained by the item pricing unit for the items picked by the customer. If they are consistent, the access control unit will release. If they are inconsistent, the broadcast unit will notify the customer that there are items that have not been settled and need to be settled before passing.

6. The unmanned store access control and monitoring system according to claim 5, wherein Collect and analyze the data of the customer entering the store and leaving the store by collecting the video of the customer entering and leaving outside the unmanned store.

Images