Interpretation: Intelligent face capture technology in IoT security applications

On the other hand, based on the needs of the development of the Internet of Things, the application of security products and technologies in the process of realizing the Internet of Things is an extension of the security application field. Its essence and most important thing is the intelligence of security products.

In the security industry, video surveillance applications account for the vast majority of the market, so the first thing to be solved in the intelligentization of security products is the intelligentization of video surveillance technology. Through intelligent video analysis technology, some key information worthy of attention can be separated from the video. In this way, video acquisition equipment with intelligent video analysis technology becomes an intelligent sensor for collecting key information and becomes the front-end core of the Internet of Things application. As a result, the application brought about is no longer just video surveillance application, but a series of new applications centered on video surveillance applications and surrounding video surveillance applications have been generated.

This article will focus on face capture technology and its applications to analyze its IoT security applications.

Face capture technology refers to the technology that uses intelligent video analysis technology to detect and locate appropriate faces from video sources and then capture them as face photos. This technology does not require the cooperation of the monitored person, and can quickly and easily obtain face photos, and can be deployed in large quantities.

Undoubtedly, human information is one of the most critical information that needs to be paid attention to in the security field. Face photos have the characteristics of small storage space and large information capacity, and can establish a relationship with personal identity, so capturing and storing face photos is of great significance. Capturing a large number of photos can bring about the retention of traces of human behavior and activities, which leads to at least four types of excellent applications: face retrieval, blacklist deployment, VIP recognition, and stranger recognition.

How to capture face

To capture a face, there are three main ways to do it:

The front-end face capture device analyzes and captures faces in the video source;

By connecting the video to the face capture device at the front end, the face capture function is realized on the front-end embedded device, and the captured face photos are transmitted to the face database through the network and other interfaces. This method is generally used for new projects, small projects, and decentralized projects.

Intelligent video analysis platform to separate and capture faces from the backend centralized video source;

By introducing video sources at the back end of the centralized video surveillance platform, face capture can be achieved on the back end intelligent video analysis platform. This method is most suitable for intelligent transformation of large centralized video surveillance platforms such as safe cities and safe campuses.

Use a handheld device to capture faces while moving.

Using handheld video acquisition equipment with intelligent processing capabilities, face photos are collected on site, and the captured faces are transmitted to the back-end face database for relevant real-time applications. This is the simplest and most effective way to collect faces when mobile collection and mobile law enforcement are required.

After the front-end of face capture application completes face capture and collection, it generally needs to cooperate with the front-end or back-end face photo recognition and comparison technology to complete the comprehensive application. There are four main types of applications.

Face retrieval

Face retrieval refers to searching for records where the face has appeared by inputting photos or other face information and searching for facial features in a large number of face photos and video sources.

Generally, this application is used in conjunction with the intelligent transformation of safe cities. When a large number of face photos are collected by the front end, the traces of the behavior and activities that the photos have retained in the system can be quickly located through face retrieval, which can easily locate the activity tracks and places where the person concerned has appeared, etc., and can be very effective in conjunction with the application of various technical investigation methods.

Blacklist defense

In some critical situations, it is necessary to quickly deploy alarms for blacklisted people, such as various ports, airports, and petition offices. When the faces of people on the blacklist appear in the video scene, they can be quickly identified and alarmed, prompting manual intervention.

The face captured by the front end is automatically compared with the blacklist database in real time. When the similarity reaches a certain threshold, the alarm linkage is automatically triggered, thus realizing the blacklist defense process. It has great deterrence and application effect for wanted criminals, smugglers, etc. Shenzhen Huanggang Port, supported by Shenzhen Feiruis Technology, has deployed a similar system.

VIP customer identification

Imagine a scenario: when a VIP customer enters a business premises, the system can automatically identify the customer without the customer actively showing proof, and provide personal information, consumption history, consumption preferences and other information to the service staff, who will then provide corresponding personalized services. This is VIP customer identification.

The main application occasions include luxury stores, high-end car service points, high-end clubs, bank key customer management centers, sales offices, etc.

Stranger Recognition Stranger recognition refers to a method of identifying and prompting strangers outside a certain range. The main applications are private gardens, private clubs, etc. By automatically comparing the captured face with the faces of all owners/members, strangers are automatically prompted for manual intervention by staff.

Application Analysis

Through the analysis of face capture and its industry applications, it is not difficult to find that its application occasions are very wide, involving large-scale government applications such as safe cities, safe campuses, and urban management, as well as corporate applications such as commerce and banks. It can also be used in general application fields such as property management, and has very obvious application prospects.

Summarize

Through intelligent video analysis technology, ordinary video acquisition equipment will be transformed into intelligent IoT sensors that can perceive a lot of key information. By reusing key information, many advanced applications that were previously unimaginable in the field of video surveillance can be completed. In this way, video acquisition has broken through the framework of ordinary applications such as video surveillance and gained a larger market space with both breadth and height.