Supermarket cashier loss prevention method based on face recognition and intelligent weighing and related equipment

Abstract

The invention discloses a supermarket cashier loss prevention method and related equipment based on face recognition and intelligent weighing, and the method comprises the steps: receiving an entrance monitoring video and the entering weight; identifying customer face data in the entrance monitoring video, and binding the customer face data with the corresponding entrance weight; receiving the leaving weight sent by the exit weighing equipment in real time, and determining face data of the target customer according to an exit monitoring video shot by the exit monitoring equipment; determining corresponding target customer entering weight and settlement commodity information according to the target customer face data; determining the weight of the settled commodity according to the settled commodity information; judging whether the difference value between the leaving weight and the target customer entering weight is greater than the settled commodity weight or not; if not, determining that the settlement state of the target customer is a normal settlement commodity; if yes, alarm information is sent to a security terminal corresponding to the security personnel, and the target customer face data is sent to the security terminal. The method is used for improving the loss prevention rate of supermarkets.

Description

Technical Field

[0001] The present application belongs to the field of cashier loss prevention, and in particular, relates to a supermarket cashier loss prevention method and related equipment based on face recognition and intelligent weighing. Background Art

[0002] In the retail industry, especially in supermarket management, commodity loss has always been a problem that cannot be ignored. With the continuous expansion of supermarket scale and the popularization of self-service shopping mode, commodity theft and missed scanning are becoming more and more serious, causing huge economic losses to supermarket operations.

[0003] In the related art, a scanning machine is usually installed at the exit of a supermarket. When unchecked-out merchandise passes through the scanning machine, an alarm mechanism is triggered to remind staff that unchecked-out merchandise has left the supermarket.

[0004] However, some customers may damage the barcodes or other items used to identify unchecked items, and hide them in places that are difficult to detect by the human eye before taking them out of the supermarket. This makes it difficult for scanning machines to detect the unchecked items, so customers will take the unpaid items out of the supermarket, causing losses to the supermarket, thereby reducing the supermarket's loss prevention rate. Summary of the invention

[0005] The present application provides a supermarket checkout loss prevention method and related equipment based on face recognition and intelligent weighing, which are used to improve the supermarket's loss prevention rate.

[0006] In the first aspect, the present application provides a supermarket cashier loss prevention method based on face recognition and intelligent weighing, which receives an entrance monitoring video sent by an entrance monitoring device and an entry weight sent by an entrance weighing device, where the entry weight is the weight of the customer when entering the supermarket; Identify customer facial data in the entrance surveillance video and bind the customer facial data to the corresponding entry weight; Receive the departure weight sent by the exit weighing device in real time, and determine the target customer's facial data based on the exit monitoring video captured by the exit monitoring device; Determine the corresponding target customer's weight and checkout product information based on the target customer's facial data; Determine the weight of settled goods based on settlement goods information; Determine whether the difference between the departure weight and the target customer's entry weight is greater than the settled goods weight; If not, the settlement status of the target customer is determined to be a normal settlement product; If it is greater, an alarm message is sent to the security terminal corresponding to the security personnel, and the target customer's facial data is sent to the security terminal.

[0007] By adopting the above technical solution, the target customer's entry weight sent by the entrance weighing device is received, and the target customer's facial data is identified in the entrance monitoring video. The target customer's facial data and entry weight are bound, so that all the target customer's information can be integrated and convenient to call and read. When the target customer leaves the supermarket, his departure weight will be measured. The difference between the target customer's departure weight and the entry weight can be calculated to calculate the weight of the goods carried by the target customer. Then, combined with the settled goods information entered by the cash register, the settled goods weight is obtained. The settled goods weight is compared with the above difference. When the difference is greater than the settled goods weight, it means that the target customer is suspected of carrying unsettled goods. Then, an alarm is immediately sounded to remind the security personnel of the abnormality, which can reduce the supermarket's loss rate and thus improve the supermarket's loss prevention rate. If the difference is not greater than the settled goods weight, it means that the target customer is settled normally and no further inspection is required, which can improve the efficiency of supermarket loss prevention.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, receiving the entrance monitoring video sent by the entrance monitoring device and the entrance weight sent by the entrance weighing device specifically includes: Sending a first instruction to the entrance weighing device to enable the data acquisition unit to collect weighing data of the entrance weighing device at a preset frequency, wherein the entrance weighing device is equipped with a plurality of weighing sensors arranged along the direction of entering the supermarket; Receive weighing data and identify each customer using image recognition technology; Use gait analysis to determine each customer's location and movement trajectory; Calculate the impact of each customer on the weighing sensor using a spatial filtering algorithm based on location information and movement trajectory; The corresponding entry weight of each customer is calculated based on the weighing data and the impact value.

[0009] By adopting the above technical scheme, the entrance weighing equipment includes a plurality of weighing sensors arranged along the direction of entering the supermarket, which can weigh customers in the direction of entering the supermarket, and the data acquisition unit collects weighing data according to a preset frequency, and can collect a sufficient amount of weight information during the time period when the customer passes through the entrance weighing equipment, and then uses gait analysis to determine the position information and movement trajectory of each customer, so as to obtain the customer's walking habits and better calculate the customer's weight, and finally uses a spatial filtering algorithm to calculate the impact value of each customer on the weighing sensor, and calculates the corresponding entry weight of each customer based on the weighing data and the impact value, which can make the weighed weight data more accurate.

[0010] In combination with some embodiments of the first aspect, in some embodiments, if the target customer's face value is greater than , an alarm message is sent to a security terminal corresponding to the security personnel, and the target customer's face data is sent to the security terminal, the method further includes: When it is determined that the target customer passes through the eating area, the weight data of the eating area weighing device and the monitoring data of the eating area monitoring device corresponding to the eating area are obtained; Determine the weight difference of target customers based on weight data and monitoring data; Calculate the sum of the entry weight and the weight difference to obtain the updated entry weight of the target customer; Calculate whether the difference between the exit weight and the updated entry weight is greater than the settled goods weight; If it is greater than the settled commodity weight, the steps of sending an alarm message to the security terminal corresponding to the security personnel and sending the target customer's facial data to the security terminal are executed; If it is not greater than the weight of the settled goods, the settlement status of the target customer is determined to be a normal settlement product.

[0011] By adopting the above technical solution, it is also necessary to consider whether the customer eats in the eating area. Eating in the eating area will increase the customer's weight slightly, and the increased weight will be reflected in the leaving weight, which will increase the difference between the leaving weight and the entering weight. This error will be mistakenly regarded as the customer carrying unsettled goods. Therefore, it is necessary to first determine whether the customer has passed through the eating area. If it is determined that the customer has passed through the eating area, the increased weight is calculated. After calculating this variable, it is determined whether the difference is greater than the weight of the settled goods according to the previous calculation method. In this way, the error caused by the customer's eating can be reduced, making the weight data of the target customer more accurate.

[0012] In combination with some embodiments of the first aspect, in some embodiments, if it is greater than, an alarm message is sent to a security terminal corresponding to the security personnel, and the target customer's face data is sent to the security terminal, specifically including: Obtain cashier information of the cashier device corresponding to the eating area; Determine the food purchase information corresponding to the target customer in the cash register information; Determine weight data for purchased food information; Determine whether the difference between the exit weight and the target customer's entry weight is greater than the sum of the settled goods weight and the weight data; If not, the settlement status of the target customer is determined to be a normal settlement product; If it is greater, an alarm message is sent to the security terminal corresponding to the security personnel, and the target customer's facial data is sent to the security terminal.

[0013] By adopting the above technical solution, when it is determined that a customer passes through the eating area, the cashier information of the cashier device corresponding to the eating area is obtained, the purchased food information corresponding to the target customer is determined in the cashier information, and then the weight data of the purchased food information is determined, and finally the increased weight is calculated, which can make the weight calculation of the target customer more accurate, thereby reducing the error caused by the customer's eating.

[0014] In combination with some embodiments of the first aspect, in some embodiments, after determining that the settlement status of the target customer is a normal settlement commodity if the weight is not greater than the settled commodity weight, the method further includes: Determine whether the target customer is carrying additional items based on the entrance surveillance video; If there are additional items, the categories of the additional items are determined, including consumable items and container items; If the category is consumables, determine whether the target customer is carrying consumables based on the exit surveillance video; If the customer is carrying consumable items, determine whether the difference between the departure weight and the target customer's entry weight is greater than the weight of the settled goods; If it is not greater than the weight of the settled goods, the settlement status of the target customer is determined to be a normal settlement product; If it is greater than the settled commodity weight, the steps of sending an alarm message to the security terminal corresponding to the security personnel and sending the target customer's facial data to the security terminal are executed; If no consumable items are carried, determine whether the absolute value of the difference between the departure weight and the entry weight of the target customer and the difference between the weight of the settled goods is within a preset range; If it is within the preset range, the settlement status of the target customer is determined to be a normal settlement product; If it is not within the preset range, the steps of sending an alarm message to the security terminal corresponding to the security personnel and sending the target customer's face data to the security terminal are executed.

[0015] By adopting the above technical solution, the influence of the target customer's own items on the weight calculation result can be reduced. The target customer may carry some extra items when entering the supermarket. The extra items may be consumable items or items for storage. The consumable items may be discarded by the target customer inside the supermarket, so the difference between the leaving weight and the entering weight will be biased, and this error value is the weight of the consumable items. When it is judged that the extra items are consumable items, it is judged according to the exit monitoring video whether the customer still carries the consumable items when leaving the supermarket. If it is still carried, it is not necessary to consider the influence of the consumable items. If not carried, it means that the customer has discarded it inside the supermarket. At this time, it is judged whether the absolute value of the difference between the leaving weight and the target customer's entering weight and the difference between the weight of the settled goods is within the preset interval. If it is within the preset interval, it means that the error is within the allowable range, so the influence of the consumable items can be ignored. On the contrary, it means that the target customer has carried unsettled goods. In this way, the influence of the target customer's own items on the weight calculation result can be reduced, so that the step of judging whether the target customer is a normal settled commodity is more accurate, thereby reducing the loss rate of the supermarket.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after executing the step of sending an alarm message to a security terminal corresponding to the security personnel and sending the target customer's face data to the security terminal if the target customer is not within the preset interval, the method further includes: If the category is container items, all surveillance videos containing the target customer are obtained; Determine whether the target customer has a preset abnormal behavior based on the surveillance video, where the preset abnormal behavior includes placing goods into the container; If there is a preset abnormal behavior, determine whether the difference between the exit weight and the target customer's entry weight is equal to the settled goods weight; If it is not equal to the settled commodity weight, executing the step of sending an alarm message to the security terminal corresponding to the security personnel and sending the target customer's facial data to the security terminal; If it is equal to the weight of the settled goods, the settlement status of the target customer is determined to be a normal settlement product; If there is no preset abnormal behavior, the settlement status of the target customer is determined to be a normal settlement product.

[0017] By adopting the above technical solution, if the additional items are container items, the target customer may place the unsettled goods in the container items, which is difficult to detect visually. Then obtain all surveillance videos containing the target customer, and judge whether the target customer has preset abnormal behavior based on the surveillance video. The preset abnormal behavior includes placing goods in the container items. If this abnormal behavior exists, then judge whether the difference between the exit weight and the target customer's entry weight is equal to the weight of the settled goods. If not, it means that the target customer has smuggled unsettled goods. If equal, it means that the customer may have placed the goods in the container items first and took them out during settlement, so there is no behavior of smuggling unsettled goods. The impact of additional items on loss prevention detection is reduced, and the cashier loss prevention rate is improved.

[0018] In combination with some embodiments of the first aspect, in some embodiments, if the target customer's face value is greater than , an alarm message is sent to a security terminal corresponding to the security personnel, and the target customer's face data is sent to the security terminal, the method further includes: Determine the unsettled product information; Receive cashier information from cashier devices in real time to determine whether there is any unsettled merchandise information; If it exists, a clear alarm message is sent to the security terminal; If it does not exist, the unsettled commodity information is sent to the preset staff terminal, and the unsettled commodity information includes the original location information of the commodity.

[0019] By adopting the above technical solution, when the existence of unsettled goods is detected, it is determined whether the target customer has settled them. If they have not been settled yet, the unsettled goods information is sent to the preset staff terminal, so that the corresponding staff can put the unsettled goods back to the corresponding position. If they have been settled, the alarm is lifted, which can improve the work efficiency of cashier loss prevention.

[0020] An embodiment of the present application provides a supermarket checkout loss prevention related device based on face recognition and intelligent weighing, and the related device includes a system, a storage medium and a computer product program.

[0021] In the second aspect, an embodiment of the present application provides a supermarket checkout loss prevention system based on facial recognition and intelligent weighing, and the supermarket checkout loss prevention system based on facial recognition and intelligent weighing includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code includes computer instructions, and one or more processors call the computer instructions to enable the system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0022] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, comprising instructions, which, when executed on a system, causes the system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0023] In a fourth aspect, an embodiment of the present application provides a computer program product, characterized in that when the computer program product runs on a system, the system executes the method described in any possible implementation manner in the first aspect.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The present application provides a supermarket cashier loss prevention method based on face recognition and intelligent weighing, which receives the target customer's entry weight sent by the entrance weighing device, identifies the target customer's face data in the entrance monitoring video, and binds the target customer's face data to the entry weight, so that all the target customer's information can be integrated for easy calling and reading. When the target customer leaves the supermarket, the departure weight will be measured, and the difference between the target customer's departure weight and the entry weight can be calculated to calculate how much weight the target customer carries. Combined with the settled goods information entered by the cash register device, the settled goods weight is obtained, and the settled goods weight is compared with the above difference. When the difference is greater than the settled goods weight, it means that the target customer is suspected of carrying unsettled goods, then an alarm is immediately issued to remind security personnel that there is an abnormality, which can reduce the supermarket's loss rate and thus improve the supermarket's loss prevention rate. If the difference is not greater than the settled goods weight, it means that the target customer is settled normally, and no further inspection is required, which can improve the efficiency of supermarket loss prevention.

[0025] 2. This application provides a supermarket cashier loss prevention method based on face recognition and intelligent weighing. It is also necessary to consider whether the customer is eating in the eating area. Eating in the eating area will increase the customer's weight slightly, and the increased weight will be reflected in the leaving weight, which will increase the difference between the leaving weight and the entering weight. This error will be mistakenly regarded as the customer carrying unsettled goods. Therefore, it is necessary to first determine whether the customer has passed through the eating area. If it is determined that the customer has passed through the eating area, the increased weight is calculated. After calculating this variable, it is determined whether the difference is greater than the weight of the settled goods according to the previous calculation method. In this way, the error caused by the customer's eating can be reduced, making the weight data of the target customer more accurate.

[0026] 3. The present application provides a supermarket cashier loss prevention method based on face recognition and intelligent weighing, which can reduce the impact of the target customer's own items on the weight calculation result. The target customer may carry some extra items when entering the supermarket. The extra items may be consumable items or items for storage. The consumable items may be discarded by the target customer inside the supermarket, so the difference between the leaving weight and the entering weight will be biased, and this error value is the weight of the consumable items. When it is judged that the extra items are consumable items, it is judged according to the exit surveillance video whether the customer still carries the consumable items when leaving the supermarket. If it is still carried, there is no need to consider the impact of the consumable items. If not carried, it means that the customer has discarded it inside the supermarket. At this time, it is judged whether the absolute value of the difference between the leaving weight and the target customer's entering weight and the difference between the weight of the settled goods is within a preset range. If it is within the preset range, it means that the error is within the allowable range, so the impact of the consumable items can be ignored. On the contrary, it means that the target customer has carried unsettled goods. In this way, the impact of the target customer's own items on the weight calculation result can be reduced, so that the step of judging whether the target customer is a normal settled commodity is more accurate, thereby reducing the loss rate of the supermarket. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of a supermarket checkout loss prevention method based on face recognition and intelligent weighing in an embodiment of the present application.

[0028] Figure 2 It is another flow chart of a supermarket checkout loss prevention method based on face recognition and intelligent weighing in an embodiment of the present application.

[0029] Figure 3 It is another flow chart of a supermarket checkout loss prevention method based on face recognition and intelligent weighing in the embodiment of the present application.

[0030] Figure 4 It is a schematic diagram of the physical device structure of a supermarket checkout loss prevention system based on face recognition and intelligent weighing provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more listed items.

[0032] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0033] An application scenario of an embodiment of the present application is described below: With the popularity of self-service shopping and fast checkout systems, customers' shopping experience has been significantly improved. However, supermarkets are facing a growing problem of product loss. Despite the installation of advanced scanning equipment and anti-theft systems at the exits, some customers still manage to bypass these security measures.

[0034] For example, one day, during a routine inventory count, a supermarket discovered that the loss rate of goods in the past month was far higher than expected. After careful analysis, it was found that the traditional anti-theft system had obvious loopholes. Some customers began to take more covert measures, such as deliberately damaging the barcodes or anti-theft tags on the goods, hiding them inside clothes or in personal belongings, and even hiding small high-value goods inside large low-value goods. These behaviors made it difficult for the scanning equipment at the exit to detect the unchecked goods, resulting in a significant decline in the loss prevention effect.

[0035] In order to solve the above technical problems, the present application provides a supermarket checkout loss prevention method based on face recognition and intelligent weighing, which is used to improve the supermarket's loss prevention rate.

[0036] Combine the following Figure 1 , a supermarket checkout loss prevention method based on face recognition and intelligent weighing in an embodiment of the present application is described: See also Figure 1 , which is a flow chart of a supermarket checkout loss prevention method based on face recognition and intelligent weighing in an embodiment of the present application.

[0037] S101, receiving the entrance monitoring video sent by the entrance monitoring device and the entrance weight sent by the entrance weighing device; An entrance monitoring device and an entrance weighing device are installed at the entrance of the supermarket. The system receives the entrance monitoring video sent by the entrance monitoring device and the entrance weight sent by the entrance weighing device. Specifically: a first instruction is sent to the entrance weighing device so that the data acquisition unit collects the weighing data of the entrance weighing device at a preset frequency. The entrance weighing device is installed with a plurality of weighing sensors arranged along the direction of entering the supermarket; Receive weighing data and identify each customer using image recognition technology; Use gait analysis to determine each customer's location and movement trajectory; Calculate the impact of each customer on the weighing sensor using a spatial filtering algorithm based on location information and movement trajectory; The corresponding entry weight of each customer is calculated based on the weighing data and the impact value.

[0038] By installing high-precision monitoring equipment and weighing equipment at the entrance of the supermarket, the system can accurately capture video images and weight data of each customer as they enter the supermarket. Entrance monitoring equipment usually includes high-definition cameras and may also be equipped with infrared sensors to ensure clear images in various light conditions. Entrance weighing equipment consists of multiple precision weighing sensors arranged along the path of customers entering the supermarket.

[0039] The system first sends a command to the entrance weighing device, activating the data acquisition unit to collect weighing data at a preset frequency (for example, 10 times per second). As customers walk through the weighing area, multiple sensors continuously record weight changes. At the same time, the system uses real-time image recognition technology to analyze surveillance video and identify each customer entering the supermarket.

[0040] In order to accurately calculate the weight of each customer, the system also requires complex data processing. First, gait analysis technology is used to determine the precise location and movement trajectory of each customer. This involves analyzing the customer's walking pattern, speed and direction. Then, the system uses a spatial filtering algorithm to calculate the impact value of each customer on each weighing sensor based on the customer's location and movement trajectory. Finally, combining the original weighing data with these impact values, the system can accurately calculate the actual weight of each customer.

[0041] Let's take a specific example: suppose a customer weighing 70 kg enters a supermarket. The entrance monitoring device captures a clear image of the customer, and the five sensors of the entrance weighing device record the following data in sequence (in kg): 10.5, 35.2, 68.7, 34.8, 9.6. The system determines through gait analysis that the customer's moving speed is 1.2 m / s and the moving trajectory is a straight line. After applying the spatial filtering algorithm, the system calculates the customer's influence on each sensor as 15%, 50%, 98%, 49%, and 14%. Finally, the system concludes that the customer's accurate weight is 70.1 kg.

[0042] S102, identifying customer facial data in the entrance surveillance video, and binding the customer facial data with the corresponding entry weight; In this step, the system can use face recognition technology to extract the facial features of each customer from the entrance surveillance video and generate corresponding face data. This process involves computer vision algorithms, including face detection, feature point positioning, feature extraction and feature matching, which are not limited here.

[0043] The system can first pre-process the entrance surveillance video, including image enhancement, noise removal, and lighting compensation, to ensure the accuracy of subsequent recognition. Then, a deep learning model (such as a convolutional neural network) is used to detect the face area in the video frame. For each detected face, the system locates key facial feature points, such as eyes, nose, mouth, etc.

[0044] Next, the system extracts a series of digital features from these feature points to form a high-dimensional vector, which is the unique facial data of the customer. This process may use dimensionality reduction techniques such as principal component analysis (PCA) or linear discriminant analysis (LDA).

[0045] Finally, the system binds the facial data with the input weight calculated in the previous step to form a complete customer information package. This information package can be temporarily stored in the system's cache for subsequent rapid retrieval and comparison.

[0046] For example, for the customer with a weight of 70.1 kg, the system may extract a feature vector containing 128 floating point numbers as its face data. Assume that the first few values ​​of this vector are: [0.235, -0.189, 0.412, 0.067, ...]. The system will create a data structure to associate this feature vector with the weight information of 70.1 kg, such as: {face_id: 12345, face_features: [0.235, -0.189, 0.412, 0.067, ...], entry_weight: 70.1}. This data structure will be stored in the system's memory for subsequent fast matching and verification.

[0047] S103, receiving the departure weight sent by the exit weighing device in real time, and determining the target customer's facial data based on the exit monitoring video captured by the exit monitoring device; This step marks the beginning of the loss prevention system's monitoring of customers leaving the supermarket. The system deploys high-precision weighing equipment and monitoring equipment at the supermarket exit, similar to those at the entrance. The exit weighing equipment continuously sends real-time weight data to the system, while the exit monitoring equipment continuously captures video footage of customers leaving.

[0048] The system first processes data from the exit weighing device. Similar to the situation at the entrance, the exit weighing device also consists of multiple sensors arranged along the customer's exit path. The system uses the same data processing algorithms as the entrance, including gait analysis and spatial filtering, to accurately calculate the weight of each exiting customer.

[0049] S104, determining the corresponding target customer's weight and settlement product information according to the target customer's facial data; The system uses the target customer's facial data determined in step S103 to retrieve the customer's complete information from the database, including the weight when entering the supermarket and the settlement commodity information during the shopping process. This process involves efficient database query and information integration technology.

[0050] First, the system uses the target customer's facial data as an index to quickly locate the customer's store entry record in the cache or database. This is usually achieved using efficient data structures such as hash tables or B-trees to ensure that the target information can be quickly found in a large amount of customer data.

[0051] Once a matching record is found, the system can obtain the weight data of the customer when he entered the supermarket. At the same time, the system also needs to access the settlement information associated with the customer. This settlement information may be stored in the supermarket's sales system, including detailed information on each purchased item, such as product name, quantity, unit price, and weight.

[0052] The system needs to integrate information from multiple data sources in real time. This may involve distributed database queries or microservices architecture to ensure that all relevant information can be obtained quickly and accurately. At the same time, the system also needs to deal with possible data inconsistencies or delays, such as using data versioning or transaction management to ensure data consistency.

[0053] S105, determining the weight of the settled goods according to the settled goods information; In step S104, the settlement commodity information already includes the name, quantity, unit price and weight of the commodity, so the system can determine the weight of the settled commodity based on the settlement commodity information.

[0054] S106, determining whether the difference between the exit weight and the entry weight of the target customer is greater than the settled commodity weight; The system needs to compare the difference between the weight of the customer when he leaves and when he enters, and compare this difference with the total weight of the settled goods to determine whether there are any unsettled goods. Of course, in actual situations, there may be a variety of problems that lead to inaccurate weight calculations. For example, the customer may pass through the edible area in the supermarket and buy some goods in the edible area, which can be eaten directly. Eating will inevitably increase the customer's weight, so the final calculated difference is likely to be greater than the weight of the settled goods, but in fact the customer did not bring any unsettled goods. Of course, there may be some other situations that affect the final calculation results. These situations will be described below and will not be repeated here.

[0055] S107, determining that the settlement status of the target customer is a normal settlement product; If not, the target customer's settlement status is determined to be normal settlement goods. When the difference between the leaving weight and the target customer's entering weight is not greater than the settled goods weight, it means that the target customer does not carry any unsettled goods. However, in some cases, when it is less than the settled goods weight, there may be an abnormal situation. For example, the target customer brought extra items before entering the supermarket and discarded them after entering the supermarket, which will make the difference smaller. This situation will be described in detail below and will not be repeated here.

[0056] S108. Send an alarm message to the security terminal corresponding to the security personnel, and send the target customer's facial data to the security terminal.

[0057] If it is greater, an alarm message will be sent to the security terminal corresponding to the security personnel, and the target customer's facial data will be sent to the security terminal. This step will be triggered when the system detects that the difference between the customer's leaving weight and entering weight is greater than the weight of the settled goods. This is a key link in the loss prevention system, which requires quick and accurate notification of security personnel, while also ensuring the confidentiality of information and the protection of customer privacy.

[0058] For example: the system generates an alarm message. This alarm message may contain the following content: Alert level (e.g., slightly suspicious, highly suspicious, etc.); The specific location where the incident occurred (such as a certain exit number); Timestamp; A brief description of the anomaly (e.g., weight difference exceeds threshold); Summary of relevant data (such as the difference between inbound and outbound weights, weight of settled goods, etc.); The system will immediately send this alarm information to the security personnel's mobile terminal. This process usually uses encrypted real-time communication protocols such as HTTPS or more secure proprietary protocols to ensure the security of information transmission. At the same time, the system may use push notification technology to ensure that security personnel can receive alarms in time even if their devices are in standby mode.

[0059] In addition to text messages, the system can also send the target customer's facial data. The facial data here is usually not necessarily the original image, but can also be processed feature data or low-resolution thumbnails to protect customer privacy and reduce data transmission volume. The system may use the following measures to process facial data: Use facial feature vectors instead of full images; Blurring or pixelating images; Only the coordinate information of key facial feature points is sent; At the same time, the system will record the detailed information of this alarm in the background, including: Complete customer shopping records; specific reasons for triggering the alarm and related calculation process; The time the alert was sent and the time the confirmation was received; The ID of the security officer who handled the alarm.

[0060] The above embodiments have the following beneficial effects: Through high-precision entrance and exit monitoring equipment, combined with image recognition and gait analysis technology, the system can accurately identify each customer and record their entry and exit weight. The target customer's entry weight sent by the entrance weighing equipment is received, and the target customer's facial data is identified in the entrance monitoring video. The target customer's facial data and entry weight are bound, which can integrate all the target customer's information and facilitate call and reading. When the target customer leaves the supermarket, his departure weight will be measured. The difference between the target customer's departure weight and the entry weight can be calculated to calculate the weight of the target customer's goods. Combined with the settled goods information entered by the cash register, the settled goods weight is obtained. The settled goods weight is compared with the above difference. When the difference is greater than the settled goods weight, it means that the target customer is suspected of carrying unsettled goods. Then, an alarm is immediately sounded to remind security personnel of the abnormality, which can reduce the supermarket's loss rate and thus improve the supermarket's loss prevention rate. If the difference is not greater than the settled goods weight, it means that the target customer is settled normally and no further inspection is required, which can improve the efficiency of supermarket loss prevention.

[0061] In the above embodiment, step S106 is described, in actual situations, there may be a variety of problems that lead to inaccurate weight calculation. For example, a customer may pass by the edible area in the supermarket and buy some goods in the edible area, which can be eaten directly. Eating will inevitably increase the customer's weight, so the difference calculated at the end is likely to be greater than the weight of the settled goods, but in fact the customer did not carry any unsettled goods. Of course, there may be some other situations that affect the final calculation result. How to reduce the impact of this situation on weight weighing and subsequent judgment of whether unsettled goods are carried will be described in conjunction with the following embodiments.

[0062] Combine the following Figure 2 , another supermarket checkout loss prevention method based on face recognition and intelligent weighing in the embodiment of the present application is described: See also Figure 2 , is another flow chart of a supermarket checkout loss prevention method based on face recognition and intelligent weighing in an embodiment of the present application.

[0063] S201, when it is determined that the target customer passes through the eating area, obtaining weight data of the eating area weighing device and monitoring data of the eating area monitoring device corresponding to the eating area; In order to solve the problem of weight calculation errors that may be caused by customers eating in supermarkets, the system first needs to determine whether the target customer has passed the eating area, which can be achieved by analyzing the customer's movement trajectory. Once it is determined that the customer has passed the eating area, the system will start collecting data from specific devices in the area.

[0064] The eating area is usually equipped with special weighing equipment and monitoring equipment. The weighing equipment in the eating area may be an embedded floor scale or a specially designed chair that can record the weight changes of customers in real time. These devices usually use high-precision sensors that can capture tiny weight changes. Monitoring equipment may include high-definition cameras and depth sensors to capture customers' behavior and posture.

[0065] The system receives data transmitted by these devices in real time. Weight data may be recorded multiple times per second to form a time series. Monitoring data may include video streams and depth images, which need to be processed and analyzed in real time.

[0066] For example, suppose target customer A enters the eating area at 13:45, and the system starts collecting data. The weighing equipment in the eating area records the following weight data sequence (unit: kg): [70.2, 70.2, 70.3, 70.5, 70.8, 71.0, 71.2, 71.2, 71.1, 71.1] This sequence reflects the weight changes of the customer during the meal. At the same time, the monitoring equipment in the eating area captures the video and depth image data of a series of actions of customer A, such as sitting down, taking food, chewing, and drinking water.

[0067] The system uses computer vision technology to analyze this data in real time and identify customers' eating behaviors. For example, through the posture estimation algorithm, the system can identify whether the customer's hand movements are consistent with the eating pattern; through the object detection algorithm, the system can identify the existence and position changes of items such as plates, food, and drinks.

[0068] S202, determining the weight difference of the target customer according to the weight data and the monitoring data; In this step, the system will use the weight data and monitoring data obtained from step S201, through complex data analysis and algorithm processing, to accurately calculate the target customer's weight change in the edible area, that is, the weight difference.

[0069] S203, calculating the sum of the input weight and the weight difference to obtain the updated input weight of the target customer; Add the entry weight and the weight difference to get the updated entry weight.

[0070] S204, determining whether the difference between the exit weight and the updated entry weight is greater than the settled commodity weight; The calculation in the previous step is equivalent to compensation calculation, that is, eliminating the data that affects the difference calculation, and then calculating the difference between the exit weight and the updated entry weight, so as to reduce the adverse effects of the target customer's eating. At this time, the accuracy of the result obtained by comparing with the settled commodity weight will be improved.

[0071] S205, determining that the settlement status of the target customer is a normal settlement product; If it is not greater than the weight of the settled goods, the settlement status of the target customer is determined to be a normal settlement product.

[0072] The same as step S107 in the above embodiment, which will not be described again here.

[0073] S206, obtaining cashier information of a cashier device corresponding to the eating area; If it is greater than the settled weight of the goods, the cashier information of the cashier device corresponding to the eatable area is obtained. If the target customer passes through the eatable area, but the difference between the exit weight and the updated entry weight is greater than the settled weight of the goods, the customer may have purchased goods in the eatable area but did not settle the bill. Next, it is necessary to verify whether this situation exists.

[0074] S207, determining the food purchase information corresponding to the target customer in the cashier information; The cashier information generally includes information corresponding to the settled goods, including weight information, brand information, settlement person information, settlement time, etc. Then the system can determine the settlement person information based on the target customer's facial information. This step can be implemented because in the previous embodiment, the target customer's facial information has been obtained and bound to the target customer when the target customer enters the supermarket. Then search the cashier information for the purchased food information corresponding to the target customer.

[0075] Then there may be two situations. The first situation is described in the following step S208. The cashier information contains the purchased food information corresponding to the target customer, which will not be described here.

[0076] The second situation is that the food purchase information corresponding to the target customer does not exist in the cash register information. In this case, the difference between the exit weight and the updated entry weight is greater than the weight of the settled goods, indicating that the target customer does have the situation described in step S206 where the customer purchased goods in the eating area but did not settle the bill. In this case, the system will send an alarm message to the security terminal corresponding to the security personnel and send the target customer's facial data to the security terminal.

[0077] S208, determining the weight data of the purchased food information; This step is the first case described in step S207, and the system can determine the corresponding weight data based on the purchased food information.

[0078] S209, determining whether the difference between the exit weight and the entry weight of the target customer is greater than the sum of the settled commodity weight and the weight data; The system determines whether the difference between the exit weight and the target customer's entry weight is greater than the sum of the settled goods weight and the weight data.

[0079] S210. Send an alarm message to the security terminal corresponding to the security personnel, and send the target customer's facial data to the security terminal.

[0080] If it is greater, an alarm message is sent to the security terminal corresponding to the security personnel, and the target customer's facial data is sent to the security terminal.

[0081] If not, the target customer's settlement status is determined to be a normal settlement product.

[0082] This step is the same as steps S107 and S108 in the previous embodiment, and will not be described in detail here.

[0083] The above embodiments have the following beneficial effects: Eating in the eating area will slightly increase the weight of the customer, and the increased weight will be reflected in the leaving weight, which will increase the difference between the leaving weight and the entering weight. This error will be mistakenly considered as the customer carrying unsettled goods. Therefore, it is necessary to first determine whether the customer has passed through the eating area. If it is determined that the customer has passed through the eating area, the increased weight is calculated. After calculating this variable, it is determined whether the difference is greater than the weight of the settled goods according to the previous calculation method. In this way, the error caused by the customer's eating can be reduced, and the weight data of the target customer can be made more accurate. When it is determined that the customer has passed through the eating area, the cashier information of the cashier device corresponding to the eating area is obtained, and the purchase food information corresponding to the target customer is determined in the cashier information, and then the weight data of the purchase food information is determined. Finally, the increased weight is calculated, which can make the weight calculation of the target customer more accurate, thereby reducing the error caused by the customer's eating.

[0084] In step S107 of the above embodiment, it is described that: However, in some cases, when the weight is less than the settled goods weight, there may be an abnormal situation, for example, the target customer brought extra items before entering the supermarket and discarded them after entering the supermarket, which will make the difference smaller. In actual situations, the extra items may be consumable items (such as garbage, etc.) or container items (such as backpacks, shopping bags, etc.). In order to reduce the impact of extra items, a detailed description is given below in conjunction with another embodiment.

[0085] Combine the following Figure 3 , another supermarket checkout loss prevention method based on face recognition and intelligent weighing in the embodiment of the present application is described: See also Figure 3 , which is another flow chart of a supermarket checkout loss prevention method based on face recognition and intelligent weighing in an embodiment of the present application.

[0086] S301, judging whether the target customer carries additional items according to the entrance surveillance video; The system determines whether the target customer is carrying additional items based on the entrance surveillance video. The system uses computer vision technology and artificial intelligence algorithms to analyze the entrance surveillance video in real time. Specifically, the system uses object detection and image segmentation algorithms to identify items on the customer's body or in his hands. These algorithms are usually based on deep learning models such as YOLO (You Only Look Once) or Mask R-CNN, which can quickly and accurately identify and locate objects in the video.

[0087] The system will compare the identified items with the preset supermarket product database to determine which are the extra items brought by the customer. In this process, the system can also consider the size, shape and color of the items. To improve accuracy, the system may combine multiple frames for analysis to eliminate the errors that may exist in a single frame.

[0088] For example, suppose a customer, Mr. Zhang, enters a supermarket and the system captures his image through the entrance surveillance camera. The image shows that Mr. Zhang is holding a black backpack in his hand and a brown leather bag on his shoulder. The system's object detection algorithm quickly identifies these two items and marks them as potential additional items. The system then compares the features of these two items with the supermarket's product database and finds that they are not among the products sold in the supermarket, so it is determined that Mr. Zhang is carrying additional items.

[0089] S302, determining the category of the additional items; If the customer brings additional items, the system will determine the category of the additional items, which includes consumable items and container items. After determining that the customer brings additional items, the system needs to further determine the category of these items. The purpose of this step is to distinguish between consumable items and container items, because these two types of items have different effects on subsequent loss prevention judgments. The system mainly achieves this function through deep learning image classification algorithms.

[0090] Specifically, the system uses pre-trained convolutional neural network (CNN) models, such as ResNet, Inception, or EfficientNet, which have been trained on large-scale image datasets and can recognize thousands of different objects. The system fine-tunes these general models to make them particularly good at distinguishing between consumable items and container items.

[0091] Consumable items usually include water bottles, food packaging, tissues and other items that may be consumed during a stay in the supermarket. Container items include backpacks, shopping bags, handbags and other containers that can be used to hold other items. The system will analyze the shape, material, opening and other features of the items to classify them.

[0092] S303: If the category is consumables, determine whether the target customer is carrying consumables based on the exit surveillance video; When the system determines that the additional items brought by the customer are classified as consumable items, the next critical step is to determine whether these consumable items are still there when the customer leaves the supermarket. The purpose of this step is to confirm whether the customer consumes these items in the supermarket, as this may affect the calculation of weight differences and loss prevention judgments.

[0093] S304, determining whether the difference between the exit weight and the entry weight of the target customer is greater than the settled commodity weight; If the customer carries consumable items, the difference between the departure weight and the target customer's entry weight is determined to be greater than the weight of the settled goods. In this case, the additional items have no effect on the judgment result.

[0094] S305, determining that the settlement status of the target customer is a normal settlement commodity; If it is not greater than the weight of the settled goods, the settlement status of the target customer is determined to be a normal settlement goods. This step is the same as step S107 and will not be described in detail here.

[0095] S306, sending an alarm message to the security terminal corresponding to the security personnel, and sending the target customer's facial data to the security terminal; If it is greater than the settled commodity weight, an alarm message is sent to the security terminal corresponding to the security personnel, and the target customer's face data is sent to the security terminal. This step is the same as step S108 and will not be repeated here.

[0096] S307, determining whether the absolute value of the difference between the departure weight and the arrival weight of the target customer and the difference between the weight of the settled goods is within a preset range; If no consumable items are carried, the system determines whether the absolute value of the difference between the departure weight and the entry weight of the target customer and the difference between the weight of the settled goods is within a preset range.

[0097] If it is within the preset range, execute step S305; if it is not within the preset range, execute step S306.

[0098] This step is the same as step S107 and step S108, and will not be described in detail here.

[0099] S308, if the category is container items, obtain all surveillance videos containing the target customer; This step is triggered when the system determines that the additional items carried by the customer belong to the category of contained items (such as backpacks, shopping bags, etc.) The purpose of this step is to comprehensively collect video data of the target customer's activities in the supermarket and provide complete visual information for subsequent behavior analysis.

[0100] S309, judging whether the target customer has preset abnormal behavior according to the surveillance video; According to the surveillance video, it is determined whether the target customer has preset abnormal behavior, and the preset abnormal behavior includes placing goods into the container. The specific technology for implementing this step can be computer vision technology, behavior recognition model, etc., which are mature existing technologies and are not described here.

[0101] However, there are other situations in this step, such as: There are consumables in the packed items. After entering the supermarket, the target customer discards the consumables in the packed items inside the supermarket, and then takes unsettled goods of the same mass as the consumables inside the supermarket. In this way, the weight of the goods themselves is balanced, but the weighing equipment cannot find the problem. Then it is necessary to manually inspect the packed items to determine whether there are unsettled items.

[0102] S310, determining whether the difference between the exit weight and the entry weight of the target customer is equal to the settled commodity weight.

[0103] If there is a preset abnormal behavior, determine whether the difference between the exit weight and the target customer's entry weight is equal to the settled goods weight; If it is not equal to the settled commodity weight, execute step S306; If it is equal to the settled commodity weight, execute step S305; If there is no preset abnormal behavior, step S305 is executed.

[0104] This step is the same as step S107 and step S108, and will not be described in detail here.

[0105] The above embodiment has the following beneficial effects: reducing the impact of the target customer's own items on the weight calculation result. The target customer may carry some extra items when entering the supermarket. The extra items may be consumable items or items for storage. The consumable items may be discarded by the target customer inside the supermarket, so the difference between the leaving weight and the entering weight will be biased, and this error value is the weight of the consumable items. When it is judged that the extra items are consumable items, it is judged according to the exit monitoring video whether the customer still carries the consumable items when leaving the supermarket. If it is still carried, it is not necessary to consider the impact of the consumable items. If not carried, it means that the customer has discarded it inside the supermarket. At this time, it is judged whether the absolute value of the difference between the leaving weight and the target customer's entering weight and the difference between the weight of the settled goods is within a preset range. If it is within the preset range, it means that the error is within the allowable range, so the impact of the consumable items can be ignored. On the contrary, it means that the target customer has carried unsettled goods. In this way, the impact of the target customer's own items on the weight calculation result can be reduced, so that the step of judging whether the target customer is a normal settled commodity is more accurate, thereby reducing the loss rate of the supermarket. If the extra items are container items, the target customer may place the unsettled goods in the container items, which is difficult to detect visually. Then obtain all surveillance videos containing the target customer, and determine whether the target customer has preset abnormal behavior based on the surveillance video. The preset abnormal behavior includes placing goods in the container items. If this abnormal behavior exists, then determine whether the difference between the exit weight and the target customer's entry weight is equal to the weight of the settled goods. If it is not equal, it means that the target customer has smuggled unsettled goods. If it is equal, it means that the customer may have placed the goods in the container items first and took them out during settlement, so there is no behavior of smuggling unsettled goods. The impact of extra items on loss prevention detection is reduced, and the cashier loss prevention rate is improved.

[0106] In addition, after the above embodiment, if the target customer does have the behavior of smuggling unsettled goods, it is necessary to continue to monitor whether the target customer corrects the behavior, then the system will determine the unsettled goods information; Receive cashier information from cashier devices in real time to determine whether there is any unsettled merchandise information; If it exists, a clear alarm message is sent to the security terminal; If it does not exist, the unsettled commodity information is sent to the preset staff terminal, and the unsettled commodity information includes the original location information of the commodity.

[0107] Through the above embodiment, when unsettled goods are detected, it is determined whether the target customer has settled them. If they have not been settled yet, the unsettled goods information is sent to the preset staff terminal, so that the corresponding staff can put the unsettled goods back to the corresponding location. If they have been settled, the alarm is lifted, which can improve the work efficiency of cashier loss prevention.

[0108] The following describes the system in the embodiment of the present invention from the perspective of hardware processing. Figure 4 , which is a schematic diagram of the physical device structure of a supermarket checkout loss prevention system based on face recognition and intelligent weighing provided in an embodiment of the present application.

[0109] It should be noted that Figure 4 The structure of the system shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0110] like Figure 4 As shown, the system includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 402 or the program loaded from the storage part 408 to the random access memory (RAM) 403, such as executing the method in the above embodiment. Various programs and data required for system operation are also stored in RAM 403. CPU 401, ROM 402 and RAM 403 are connected to each other through bus 404. Input / output (I / O) interface 405 is also connected to bus 404.

[0111] The following components are connected to the I / O interface 405: an input section 406 including a camera, an infrared sensor, etc.; an output section 407 including a liquid crystal display (LCD) and a speaker, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed so that a computer program read therefrom is installed into the storage section 408 as needed.

[0112] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 409, and / or installed from a removable medium 411. When the computer program is executed by the central processing unit (CPU) 401, various functions defined in the present invention are performed.

[0113] It should be noted that the computer-readable medium shown in the embodiment of the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable computer program. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.

[0114] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Among them, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0115] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the system described in the above embodiment; or may exist independently without being assembled into the system. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of a system, the system implements the method provided in the above embodiment.

[0116] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0117] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0118] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk), etc.

[0119] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A supermarket cashier loss prevention method based on face recognition and intelligent weighing, characterized in that: include: Receiving the entrance monitoring video sent by the entrance monitoring device and the entry weight sent by the entrance weighing device, wherein the entry weight is the weight of the customer when entering the supermarket; Identify customer facial data in the entrance surveillance video, and bind the customer facial data to the corresponding entry weight; Receive the departure weight sent by the exit weighing device in real time, and determine the target customer's facial data based on the exit monitoring video captured by the exit monitoring device; Determine the corresponding target customer's weight and checkout product information according to the target customer's facial data; Determine the weight of the settled goods according to the settled goods information; Determine whether the difference between the exit weight and the entry weight of the target customer is greater than the settled commodity weight; If not, the settlement status of the target customer is determined to be a normal settlement product; If it is greater, an alarm message is sent to the security terminal corresponding to the security personnel, and the target customer's face data is sent to the security terminal.

2. The method according to claim 1, characterized in that: The receiving of the entrance monitoring video sent by the entrance monitoring device and the entrance weight sent by the entrance weighing device specifically includes: Sending a first instruction to an entrance weighing device to enable a data acquisition unit to collect weighing data of the entrance weighing device at a preset frequency, wherein the entrance weighing device is equipped with a plurality of weighing sensors arranged along a direction of entering the supermarket; receiving the weighing data and identifying each customer using image recognition technology; Using gait analysis to determine the location information and movement trajectory of each customer; Calculate the impact value of each customer on the weighing sensor using a spatial filtering algorithm according to the location information and the movement trajectory; The entry weight corresponding to each customer is calculated based on the weighing data and the influence value.

3. The method according to claim 1, characterized in that After the step of sending an alarm message to a security terminal corresponding to the security personnel if the target customer's face value is greater than the target customer's face value, and sending the target customer's face data to the security terminal, the method further comprises: When it is determined that the target customer passes through the eatable area, obtaining weight data of an eatable area weighing device and monitoring data of an eatable area monitoring device corresponding to the eatable area; Determining a weight difference of the target customer according to the weight data and the monitoring data; Calculating the sum of the input weight and the weight difference to obtain the updated input weight of the target customer; Calculate whether the difference between the exit weight and the updated entry weight is greater than the settled commodity weight; If it is greater than the weight of the settled goods, executing the step of sending an alarm message to the security terminal corresponding to the security personnel and sending the target customer's face data to the security terminal; If it is not greater than the weight of the settled commodity, it is determined that the settlement status of the target customer is a normal settlement commodity.

4. The method according to claim 1 or 3, characterized in that: If the target customer's face data is greater than the target customer's face data, an alarm message is sent to the security terminal corresponding to the security personnel, and the target customer's face data is sent to the security terminal, specifically including: Acquire cashier information of a cashier device corresponding to the edible area; Determining the food purchase information corresponding to the target customer in the cashier information; Determining weight data of the purchased food information; Determine whether the difference between the exit weight and the entry weight of the target customer is greater than the sum of the settled commodity weight and the weight data; If not, the settlement status of the target customer is determined to be a normal settlement product; If it is greater, an alarm message is sent to the security terminal corresponding to the security personnel, and the target customer's face data is sent to the security terminal.

5. The method according to claim 1 or 3, characterized in that: After determining that the settlement status of the target customer is a normal settlement commodity if the weight is not greater than the settled commodity weight, the method further includes: Determining whether the target customer carries additional items according to the entrance surveillance video; If the additional items are carried, then the categories of the additional items are determined, the categories including consumable items and container items; If the category is the consumables, determining whether the target customer carries the consumables based on the exit surveillance video; If the consumables are carried, determining whether the difference between the departure weight and the entry weight of the target customer is greater than the settled goods weight; If it is not greater than the weight of the settled goods, determining that the settlement status of the target customer is a normal settlement goods; If it is greater than the weight of the settled goods, executing the step of sending an alarm message to the security terminal corresponding to the security personnel and sending the target customer's face data to the security terminal; If the consumables are not carried, determining whether the absolute value of the difference between the departure weight and the entry weight of the target customer and the difference between the weight of the settled goods is within a preset range; If it is within the preset interval, determining that the settlement status of the target customer is a normal settlement product; If it is not within the preset interval, the steps of sending an alarm message to the security terminal corresponding to the security personnel and sending the target customer's face data to the security terminal are performed.

6. The method according to claim 5, characterized in that After executing the step of sending an alarm message to a security terminal corresponding to the security personnel and sending the target customer's face data to the security terminal if the target customer is not within the preset interval, the method further includes: If the category is the contained items, then all surveillance videos containing the target customer are obtained; Determining whether the target customer has a preset abnormal behavior according to the surveillance video, wherein the preset abnormal behavior includes placing goods into the container; If the preset abnormal behavior exists, determining whether the difference between the exit weight and the entry weight of the target customer is equal to the settled commodity weight; If it is not equal to the settled commodity weight, executing the step of sending an alarm message to the security terminal corresponding to the security personnel and sending the target customer's facial data to the security terminal; If it is equal to the weight of the settled goods, then the settlement status of the target customer is determined to be a normal settlement goods; If the preset abnormal behavior does not exist, the settlement status of the target customer is determined to be a normal settlement product.

7. The method according to claim 1, characterized in that After the step of sending an alarm message to a security terminal corresponding to the security personnel if the target customer's face value is greater than the target customer's face value, and sending the target customer's face data to the security terminal, the method further comprises: Determine the unsettled product information; Receiving the cashier information of the cashier device in real time, and determining whether there is the unsettled commodity information; If so, sending a clear alarm message to the security terminal; If not, the unsettled commodity information is sent to a preset staff terminal, and the unsettled commodity information includes the commodity original location information.

8. A supermarket cashier loss prevention system based on face recognition and intelligent weighing, characterized in that: The system comprises: One or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the system to execute the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a system, the system is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is run on a system, the system is caused to execute the method according to any one of claims 1 to 7.

Images