Six security related technologies for unmanned retail applications

Security technology has a very broad application space in the unmanned economy. It can even be said that the support behind security technology is the foundation of unmanned economic development. For example, if an unmanned store is to operate effectively for a long time, it must ensure the smooth completion of the transaction and the good shopping process experience, as well as the security of the shopping environment. So what is important in this process?

It must be the application of security monitoring and burglar alarm technology. In fact, monitoring technology is indispensable in various unmanned economic application scenarios, especially intelligent monitoring technology. In addition, there are applications such as image recognition technology, face recognition technology, RFID technology, etc. These technologies play an important role in applications such as unmanned shops, unmanned storage, and unmanned driving. Character.

First, intelligent monitoring technology

With the development of intelligent video analysis technology, in addition to the traditional security protection, the video surveillance system has become a new development trend in the application of industry business management, such as in chain stores, specialty stores, supermarkets, business halls and other commercial applications. It can assist business operations and provide efficient and accurate data for business analysis.

In the unmanned convenience store, the role of video surveillance intelligent analysis technology has been brought to a greater extent. Different from the security field, in the unmanned convenience store, the video intelligent analysis is more to achieve the refined daily management of the convenience store, and also includes the function of anti-theft and loss prevention for the unmanned convenience store.

From the user's scanning code into the store, first there will be a video surveillance system, that is, the front-end camera at the door of the convenience store. Through the camera target detection and recognition algorithm, the customer can be identified, and the identity of the customer can be confirmed, and the camera at the door can also be realized. The number of people statistics function, timely understanding of the number of people entering the store, to ensure the experience of the application within the store.

When customers enter the store to purchase goods, the front camera of multiple video surveillance systems can track and trace the customer's whereabouts and behaviors. In this process, it includes target detection and tracking, which is mainly to track the behavior and movement trajectory of consumers in the store. This function mainly relies on multi-channel surveillance cameras. Judging the attitude of a consumer by judging the behavior of the consumer, or by instructing the merchant to adjust the manner in which the merchandise is displayed, by traversing the movement trajectory of many consumers in the store or staying in front of a particular shelf.

For example, video intelligence analysis realizes the real-time heat map in the store, tracking which route the guest likes most, which shelf has the most intensive traffic, which shelf has the longest stay, and which shelf is relatively deserted. Biometrics can also be used to determine the proportion of males and females, whether the average size is too fat or too thin; the guests are standing in front of the shelves, the eyes are most accustomed to where to look first (to calculate the real golden position of the shelves).

Second, image recognition technology

In the dining area and rest area of ​​unmanned convenience stores, image recognition technology is applied. There is no doubt that image recognition technology can be said to be the strength of security companies. In convenience stores, the application of image recognition technology is actually relatively basic, only applied to face recognition technology, and only used as an access port, the actual participation feeling is too low.

No matter what kind of unmanned retail solution, you must first solve the C-side user experience problem, and then reduce the labor cost for the B-end customer. Because entering the store does not depend on the consumer, no consumer is empty talk. The future of unmanned goods should remove all media, consumers only need to complete the shopping by themselves. Even if consumers walk into the store empty-handed, they can successfully buy goods. To do this, we must first address the identity issue. Face recognition, fingerprint recognition, iris recognition technology, etc. can be realized. Which kind of identification technology is selected depends on the specific situation, and anyway, it is related to security. In the unmanned convenience store, in addition to accurate identification, we must continue to track consumers and analyze consumer behavior. This can be achieved by image recognition techniques for the analysis of moving images. After the user selects the product, he enters the specific settlement area. At this time, it is the turn of RFID technology to play, calculate the total amount of goods through RFID, and then match the user payment account completed by the biometric technology certification, and easily complete the shopping.

Third, face recognition technology

Face recognition is one of the most common technologies in unmanned supermarkets. In the process of consumption, consumers use mobile phones to enter the store like a subway card. At the same time, the camera at the entrance will perform face recognition. This is an artificial intelligence for deep learning that analyzes photos, details the content of the photos and related objects, and also scans the gender, facial expressions, emotional state, clothing, and even the beard of the characters in the photos. Make judgments.

Fourth, RFID technology

RFID technology is a vital part of the unmanned retail sector.

In one sentence, the technical aspects of shopping in an unmanned supermarket are summarized. That is, by adding an RFID tag (commonly known as an electronic tag) to each item, and then using a technique such as a visual sensor or a pressure sensor, the target object is automatically recognized by the RF signal and related. Data, non-contact automatic identification, and sales of goods.

For the identification of goods, there are currently no more supermarkets using RFID tags. RFID, called radio frequency identification, is a communication technology that uses radio signals to identify specific targets and read and write related data without having to identify the physical contact between the system and a particular target. Wireless radio is generally microwave and is suitable for short-distance identification communication.

Each item is affixed with an RFID tag to identify and settle the item purchased by the customer in the "Payment Tunnel". RFID is a mature and inexpensive solution, and retail giant Wal-Mart has used similar technologies to improve the efficiency of supply chain management.

When the customer leaves the store, the sensor scans the RFID tag and reconfirms the item purchased by the consumer, and automatically settles the amount on the consumer's account.

Of course, it also has some problems, it can not identify the special materials such as glass, so there are merchants such as wheat smart convenience store, then use the mobile phone to scan the QR code on the shelf or directly scan the barcode on the product package to complete the mobile payment of the goods. Thereby avoiding the drawbacks of RFID.

But at the same time, RFID technology also has problems such as non-uniform standards, high cost, and data collision. If you want to embark on the fast track of the Internet of Things, RFID manufacturers need to catch up.

Compared to traditional barcodes, the RFID recognizer can simultaneously recognize and read multiple RFID tags; in addition, if covered, RFID can penetrate non-metallic or non-transparent materials such as paper, wood and plastic, and Penetrating communication; and RFID is not limited in size and shape constraints. This is why many unmanned retail stores currently use RFID technology.

5. Artificial intelligence and deep learning technology

Artificial intelligence and deep learning are gradually integrated into video security monitoring. With the help of video intelligent analysis technology, store remote patrol shop, passenger flow statistical analysis, regional heat analysis, POS cashier supervision function can be realized, and the management efficiency of unmanned supermarkets can be improved. The face recognition technology can realize the passenger flow and customer face recognition information in each area of ​​the retail place. Through the passenger flow analysis and the business super heat map, the data information of the store operation is analyzed and statistically generated, and reports are generated to provide data for operation management. Decision basis.

Sixth, sensor technology

In addition to the application of AI technology such as computer vision, unmanned retail also involves large-scale sensors and smart devices, throughout the store, shopping to settlement.

For gravity sensors, goods are taken, goods are reduced, and weight is reduced; goods are returned, goods are increased, and weight is increased. Therefore, the change in weight value can mean taking or putting back.

For the infrared, in a specific place, the goods are put back, the infrared will be blocked, the goods are taken, and the infrared will not be blocked. It can be taken or replaced by the infrared occlusion status.

In summary, from the biometric identification technology that enters the unattended retail store--the image recognition technology purchased by the product--the FRID technology of product settlement, the interlocking and interlocking technologies are inseparable from the security technology.

Unmanned retail is a testing ground for various intelligent security technologies. From the initial anti-theft and loss prevention, to the later support for retail refinement management, to the new goal of plasticization, intelligentization and synergy of retail infrastructure, all of them are IoT technology, image intelligence analysis technology, video surveillance technology, biometric identification. The new application and new extension of technology based on the original anti-theft and loss prevention.

The artificial intelligence and deep learning are gradually integrated into the video security monitoring. With the video intelligent analysis technology, the store remote patrol shop, passenger flow statistical analysis, regional heat analysis, POS cashier supervision function can be realized, and the management efficiency of the unmanned supermarket can be improved. With the help of face recognition technology, the passenger flow and customer face identification information in each area of ​​the retail location can be realized. Through the analysis of passenger flow and the super-heat map, the data and information of the store operation are analyzed and statistically generated, and reports are generated to provide data for operation management. Decision basis.

In short, unattended retail stores will bring the ever-experienced Internet of Things technology, video intelligence analysis technology, face recognition technology and mobile payment technology to the fullest.