Helps video surveillance: AI can identify thief's body language and movements

This article is produced by NetEase Smart Studio (public number smartman 163). Focus on AI and read the next big era!

[Netease smart news November 26 news] Review of the development of artificial intelligence technology in the past 10 years is fascinating, but in fact the best AI applications have not yet appeared. As we prepare to enter 2018, there are many exciting developments in the pipeline, especially with respect to AI-enhanced video surveillance.

Researchers have clearly documented the progression of machine learning and AI technology over the years. Whether IBM’s Watson won many of the world’s smartest people in quizzes, or whether the Chinese platform outperformed human performance in IQ testing, or whether Google AI wrote poetry on its own, they were all examples of AI technology being so great.

However, so far, the achievements of AI can only be said to be more interesting than useful. However, the times are changing. In addition to major breakthroughs in the pharmaceutical and medical fields, AI has also seen some positive developments in the security industry. Specifically, new progress will improve video surveillance. What specific progress will we see? Let's take a look at a few examples:

1. Overcoming human flaws in AI

Entrepreneur Bob?Beachler explained: "One problem with monitoring operators is attention fatigue. The brain naturally alternates between concentration and distraction. In monitoring, Distraction can have dire consequences, so if we can establish a monitoring system that will never be distracted and cooperate with humans to eliminate the wrong monitoring system, this is the promise of AI in video surveillance."

While video surveillance technology always recorded video, and allows people to watch it when you want, but it's never been able to tell a human operator: "Hey, this looks not quite right!" Motion detection often produce false alarms in addition to In addition, there is no function that can effectively make up for human brain defects.

Just last year, machine vision developer Movidius and security hardware maker Hikvision announced a partnership and joint development of a camera with a deep neural network AI that will notice real-time alerts when something suspicious occurs.

An example of the application of this product is to monitor a store. When the same person moves around within a certain period of time or returns to the same place multiple times, it will notice this and determine that burglary may have occurred. The camera will capture this behavior and this may be something that the human camera operator has not noticed and has issued an alert for further investigation.

2. Excellent facial recognition

After a crime occurs, people often painstakingly review the camera footage in order to find out the culprit. Unfortunately, criminals are only recognized when the user identifies someone on the screen. Advanced AI software like MegaFace hopes to change this situation.

The MegaFace project is a project from the University of Washington. It is trained on nearly 5 million images of about 672,000 people from Flickr. In the test, the accuracy rate of this system is 75%, and it can match two images of the same person in 1 million different facial image data sets. If you have a chance to guess 10 times, the accuracy rate will increase to 90%.

Obviously, facial recognition systems have great potential value and can cooperate with video surveillance. It can help law enforcement officers find the suspect's face in the crowd, which may help hunt criminals - and even prevent crime.

3. Real-time response to crime

People hope that these new AI monitoring systems can detect crimes in real time. For example, the Movidius system can identify unattended objects. A practical application is the airport and train station. Cameras can identify unattended packages, track the time they are left behind, and issue warnings to have onsite security personnel investigate suspicious activity.

Another example is a camera system that is trained to recognize the body language and movement used by thieves before and during the theft. When these behaviors are identified, the security team can deploy personnel to check the situation.

4. The Future of AI and Video Surveillance

AI has great potential in many industries and niche areas, but we can hardly ignore its huge change in security. Whether it's a corporate monitoring store, homeowners protecting personal property, or law enforcement agencies, AI has many applications. The development of AI in the coming years will bring more excitement to people. (From: VentureBeat Compilation: NetEase See Compilation Robot Review: Little)

Pay attention to NetEase smart public number (smartman163), obtain the latest report of artificial intelligence industry.

Outdoor Small Energy Storage System

Portable power supply humanized output port design: AC dual output port 220V output, to solve multi-channel power demand; DC 24V, 12V cigarette lighter, dual 5V USB output, more widely used. Can meet the needs of most electrical appliances, such as energy storage system for LED Light,energy storage system for outdoor, energy storage system for medical equipment,mobile phones, telephones, digital cameras, mobile hard drives, digital cameras, tablets, laptops, car starters, pumps, postal and telecommunications, environmental instruments Etc; can also be used in the following areas such as: finance, first aid, excavation, exploration, military, science, media, tourism, disaster relief, medical assistance, environmental protection and areas with widespread power shortages.

Solar Storage Battery,Small Solar System,Energy Storage System For Computer,Energy Storage System For Power Tools,Portable Small Electric Station,Multifunctional Lithium Generator

Shenzhen Enershare Technology Co.,Ltd , https://www.enersharepower.com