Smart: opening a new chapter in Internet of Things security awareness

Smart: opening a new chapter in Internet of Things security awareness As we all know, the Internet of Things is also called a sensor network and it is a connected Internet. It extends and extends the user end of the Internet to various types of objects through various types of sensors. The three major elements of the Internet of Things are awareness, connectivity and applications. If the Internet of Things technology is applied to the security industry to form a security solution, it is a security solution for the Internet of Things. This article will focus on the "perceived" part of the analysis of the Internet of Things technology and security industry applications closely linked.

Among the three major elements of the Internet of Things, the core is perception, which is the most front-end sensor. In general concept, sensors include RFID, various types of sensing probes, and data acquisition devices. Video capture devices with intelligent video analysis are essentially more intuitive and intelligent sensors that can be used for intelligent video analysis. Perceive the target and its behavior in the video scene. With the development of intelligent video analysis and automatic identification technology, video perception will become the most important sensory technology and become one of the most important technologies in the information-aware layer of the Internet of Things. Existing mature major applications include:

- Detecting, capturing and recognizing human faces and sensation of people's identities; - Analyzing the behavior of moving targets (people and objects) to prevent perimeter intrusion; - Perceiving the flow of people for use in passenger flow statistics and analysis, entertainment venues and other public places to stay Number of warnings; - The disappearance or appearance of people or things, used for property preservation, identification of suspicious relics, etc. - Perceive and capture the license plate in sports, which is used to capture vehicle license plates illegally occupying bus lanes - to sense information such as crowd gathering status, driving fatigue status, and smoke phenomenon.

In addition, various types of sensor technologies such as infrasound pressure, pan-infrared, vibration, and acceleration can be applied to security monitoring applications. The technology of the Internet of Things is used to solve security needs, and thus the IoT security solutions for various industries are introduced. Program.

The status of intelligent video analysis technology So far, most of the understanding and application of intelligent video analysis is still limited to moving target detection (perimeter intrusion prevention), a small number of applications related to target recognition and video optimization, the application has not yet been a breakthrough What kind of application is the market of intelligent video analysis technology? Looking at this issue from the perspective of IoT applications, it is easy to get the answer: Intelligent video analytics technology can perceive data and information and can use it to expand the application of this data and information.

As we mentioned earlier, the information that intelligent video analysis technology can sense and understand includes the face (user identity), the behavior of people and things, the flow of people, the disappearance of people and things, the state of crowded people, the state of human fatigue, and smog. Produce and spread etc. All applications that need this information can potentially be used for intelligent video analytics.

Practical applications of intelligent video internet of things Cooperative face recognition applications Cooperative face recognition uses embedded recognition schemes to form a typical terminal form of the Internet of Things.

The application of face recognition is a very intuitive application of intelligent video analysis technology, which solves the intuitive recognition of machine eyes to people. After the accumulation of technology in recent years, the technology of cooperative face recognition has been well-known. The Internet of Things is also called a sensor network and it is a connected Internet. It extends and extends the user side of the Internet through various types of sensors. Class object on target. The three major elements of the Internet of Things are awareness, connectivity, and applications. If the Internet of Things technology is applied to the security industry to form a security solution, it is an Internet of Things security solution. This article will focus on the "perceived" part of the analysis of the Internet of Things technology and security industry applications closely linked.

Among the three major elements of the Internet of Things, the core is perception, which is the most front-end sensor. In general concept, sensors include RFID, various types of sensing probes, and data acquisition devices. Video capture devices with intelligent video analysis are essentially more intuitive and intelligent sensors that can be used for intelligent video analysis. Perceive the target and its behavior in the video scene. With the development of intelligent video analysis and automatic identification technology, video perception will become the most important sensory technology and become one of the most important technologies in the information-aware layer of the Internet of Things. Existing mature major applications include:

- Detecting, capturing and recognizing human faces and sensation of people's identities; - Analyzing the behavior of moving targets (people and objects) to prevent perimeter intrusion; - Perceiving the flow of people for use in passenger flow statistics and analysis, entertainment venues and other public places to stay Number of warnings; - The disappearance or appearance of people or things, used for property preservation, identification of suspicious relics, etc. - Perceive and capture the license plate in sports, which is used to capture vehicle license plates illegally occupying bus lanes - to sense information such as crowd gathering status, driving fatigue status, and smoke phenomenon.

In addition, various types of sensor technologies such as infrasound pressure, pan-infrared, vibration, and acceleration can be applied to security monitoring applications. The technology of the Internet of Things is used to solve security needs, and thus the IoT security solutions for various industries are introduced. Program.

The status of intelligent video analysis technology So far, most of the understanding and application of intelligent video analysis is still limited to moving target detection (perimeter intrusion prevention), a small number of applications related to target recognition and video optimization, the application has not yet been a breakthrough What kind of application is the market of intelligent video analysis technology? Looking at this issue from the perspective of IoT applications, it is easy to get the answer: Intelligent video analytics technology can perceive data and information and can use it to expand the application of this data and information.

As we mentioned earlier, the information that intelligent video analysis technology can sense and understand includes the face (user identity), the behavior of people and things, the flow of people, the disappearance of people and things, the state of crowded people, the state of human fatigue, and smog. Produce and spread etc. All applications that need this information can potentially be used for intelligent video analytics.

Practical applications of intelligent video internet of things Cooperative face recognition applications Cooperative face recognition uses embedded recognition schemes to form a typical terminal form of the Internet of Things.

The application of face recognition is a very intuitive application of intelligent video analysis technology, which solves the intuitive recognition of machine eyes to people. After the accumulation of technology in recent years, the technology of cooperative face recognition has entered the stage of practical application.

Through years of technology research and development in the industry, especially the introduction of new light source recognition technology and Live Feature Model training, the face recognition terminal has achieved low cost (similar to the cost of fingerprint technology) and high Speed ​​(1 second recognition), low misrepresentation and rejection, and other advantages, and has the advantages of intuitive, verifiable, applications: attendance, access control, kindergarten transfer certification, visitor registration management, meeting attendance and so on.

Non-cooperative face capture and face-to-face non-cooperative face capture, which cannot match the captured person, face information accuracy is not high, and the angle change is large, making it difficult to make high-precision face recognition. Authentication. However, in the face comparison of small sample banks, it still has a high application value.

Non-cooperative face capture, through the front-end intelligent video analysis terminal, captures all face photos from the video scene and transmits them to the back-end platform. The back end compares the closest face data/data sets by comparing the faces of the sample database for the user to perform initial identity screening.

The biggest difference from cooperative face recognition is that the former belongs to accurate recognition, and the latter belongs to non-exact contrast. Therefore, the application site is mainly a non-critical authentication place, which plays a role of prompting and warning for the user and reminds the user. Conduct further identity verification.

Application: Preliminary identification and prompting of VIP/blacklist users in shops and star hotels; preliminary identification and prompting of VIP/blacklist users in financial establishments, airports, terminals, and customs.

Passenger Flow Statistics, Population Statistics Existing common passenger flow statistics technologies include infrared sensors, gravity sensors, mechanical barriers, etc. However, these technologies can only achieve about 70% accuracy, and there are many restrictions on the application sites.

The current passenger flow statistics terminal product realized through video intelligent analysis technology in the existing embedded platform is a very typical terminal of the Internet of Things. Directly embedded camera and intelligent analysis module, GPRS/3G and other transmission modules, through the non-contact analysis of the entrance and main channel, can achieve more than 95% accuracy, and does not require back-end analysis support, can be applied to all types Analysis of passenger flow dynamics in commercial venues, over-the-counter warnings of number of people in entertainment venues, passenger flow statistics on mobile vehicles, etc. (Customs flow statistics on mobile vehicles are an important part of intelligent transportation).

The demand for passenger flow statistics is becoming more and more prosperous. If we can fully consider the special needs of business users when technology is implemented, it is entirely possible to become a very important application branch of intelligent video analysis technology. Applications: Various types of commercial supermarkets, chain stores, small and medium sized shops, park attractions, Wenbo clubs, public transportation vehicles, and long-distance vehicles.

1. Perimeter intrusion prevention.

A single intelligent video analysis has many shortcomings in the application of perimeter intrusion prevention: changeable scenarios, high false alarm rate, and unsatisfactory recognition results. However, combining intelligent video sensors with other types of sensors for comprehensive detection with multiple defense detections can achieve very good results. The comprehensive application program for typical applications of homes, shops, and vehicles has now been well evaluated.

Infrared detection technology, such as infrasound pressure sensing technology, pan-infrared and microwave detection technologies using the Internet of Things technology are very effective for perimeter intrusion detection. However, after these technologies detect an alarm, there is no way to immediately confirm whether it is a false alarm. If these technologies are integrated with video analytics technologies, mobile video, etc., and dual-probe detection is performed, users can be provided with high efficiency, low false positives, and intuitive user experience.

Applications: Unattended small and medium-sized shops, warehouses, homes, small office space, etc.

2. The intelligent video analysis technology for fatigue detection of drivers and duty staff can perceive human fatigue status through analysis of human face characteristics and human eye status information. A device embedded in the camera and analysis module is a human fatigue state detection terminal, a typical Internet of things smart sensor terminal.

In key applications such as private cars, freight cars, long-distance passenger cars, dangerous goods transport vehicles, and special machinery operations, human fatigue status information is a very important indicator. At this time, human fatigue detection terminals have a very good application opportunity. For special applications, such as vehicle information collection, security precautions, oil circuit control, GPS positioning, etc., this Internet of Things terminal can provide a more excellent application mode.

Applications: private cars, freight cars, long-distance passenger cars, dangerous goods transport vehicles, special machinery operation posts.

3. The current plan for judging the degree of crowd aggregation is to use an embedded platform and adopt intelligent video analysis technology to detect and determine crowd gathering phenomenon in video scenes. The smart video analyzer is said to have embedded GPRS/3G. Communication module is also a typical Internet of Things application terminal. The intelligent video analyzer directly analyzes and judges the front-end on the cameras that need to be judged, and sends the results to the back-end application platform to provide real-time warning of the aggregation phenomenon.

Applications: Safe cities, squares, government entrances and exits and other special occasions.

In addition to the above application directions, smoke detection can be used for early warning of fires, face capture can be used for face photo capture and storage in various situations, etc. These technologies have good application prospects.

In the IoT solution, the embedded platform is adopted, which means that the video-aware analysis and completion are all implemented on the front-end device. In this way, all necessary elements of the Internet of Things terminal are independently provided in the product form, and all the functions of the perception can be performed independently. Coupled with the network transmission and back-end applications, that is, with the three elements of perception, connection and application, can be completely formed an IoT security application system.

Intelligent video analysis technology has lightened the application of the video surveillance industry, established a smooth bridge between traditional video surveillance and the Internet of Things, brought new vitality to the Internet of Things technology, and can also use the Internet of Things technology and ideas for security. Industrial development adds a good support.