Bird Home Automation Enters Into A partnership With the AV-LinkPro | Security News – security informed

Bird Home Automation Enters Into A partnership With the AV-LinkPro | Security News – security informed

this May Be the KI video analysis Ever Really be Intelligent?

Video-surveillance is often in connection with the security. But in most cases, it is used for the recording of incidents and support the investigation after the fact rather than prevention of adverse events. Artificial intelligence–powered video analytics is a promising trend that fundamentally changes the nature of how things work. Extract the manageable data of a video stream can help you recognize dangerous situations at an early stage, minimising the damage and to prevent in the ideal case of complete emergencies. At the same time, AI significantly of video surveillance, security systems, extends the area of application.
AI significantly of video surveillance extends the range of applications on security systems
But the hype around this new, trendy technology prevents the potential users from the selection of high-quality solutions in a wide variety of products. This often leads to over-expectation, followed by a complete let-down. AI-powered video analytics really the key to a technological breakthrough in the field of video surveillance can? We’ll have a look, what can not afford the technology, what it can, and where it is from here can go.
A technological breakthrough or just another bubble?
It is often said that the video management software (VMS) is gaining market values is becoming increasingly tangible, and available everywhere. A lot of products with similar characteristics (or at least similar promises from the manufacturers) make it hard to choose. As a result, the name of the supplier and reputation are arguments in one of their most important sale. The manufacturers have two options available: you wrapped up in a price war and rely on the latest expenditure, or offer a product that is really innovative and revolutionary.
The manufacturers have two options: engage in a price war, or you offer a product that truly innovative and revolutionary
VMS developer, you choose the second route are gravitating in the direction of the manufacture of products, the artificial intelligence is based on neural networks and deep learning. Emerging two or three years ago, the KI-video analytics market is experiencing a boom in growth. This new tech wave is the still stagnant, stirred in addition to waters from the VMS world and the small, ambitious developers had to be a little optimistic. It seems, now you have the chance, as a market leader in the next few years.
But the hype surrounding the popular trend is the reasonable question, under experienced security industry professionals. These concerns come from customers in search of a solution to their problems, and by suppliers, the establishment of a long-term development strategy. This is very similar to another tech bubble, like the built in pre-AI video analysis, and bursting, as it became clear that the sensational promises were all around him of pure marketing hype (and pretty ruthless, so). However, there are a lot of factors that show that AI-powered video surveillance systems are not in another bubble.
The three factors
The first — and most important — comes from systems that are already on client sites. They fulfill the same promises made, the lessons during the previous bubble by hotheads in a hurry to the computer to analyze the events in real-time with a classical algorithmic approach.
The second is the fact that this new technology is seen, the investment of not only software and cloud startups, but also established VMS-developer. is Even giants such as Intel, the presented a full line of the neural network accelerator hardware, and a set of software tools optimised to work with you, especially in the area of computer vision.
This new technology is seen, the investment of not only software and cloud startups, but also established a VMS developer
The third factor is the artificial intelligence capabilities. AI is playing chess, driving cars, and works wonders in many other areas. Why should it not also be applied to video monitoring and analysis?
What can make the AI
What can artificial intelligence in video surveillance systems in this Phase of development? It can not quite a sequence of events and understand the “logic” of what is in the camera field of view to analyze. At least, not yet. But it is likely that the AI will learn to do this in the next few years. However, neural network analysis can already detect, classify and track objects very well, offers a high degree of accuracy, even in hectic scenes.
Artificial intelligence can be used in the real world:

Smoke and flame for the early fire-alarm in the case of open spaces (forest, open storage, Parking, etc.);
distinguish people and vehicles from animals and other moving objects, such as for the protection of the perimeter of the nature Park from poachers;
differentiate between a person with a helmet and protective clothing of a person, without to the prevention of accidents at hazardous production or on the construction site;
count of objects of a certain type, e.g., cars in a Parking lot, the people in the area to move goods on a conveyor belt, etc. in non-safety-relevant solutions.

These are just a few examples. After the training of a neural network, it can address other, similar tasks. In General, a neural network is trained, in certain conditions, is not reproducible. In other words, it will not work so well under different conditions. On the other hand, the developers have learned how to quickly train the AI for the needs of a specific project. The main requirement is that you have enough video material.
A little off, that the use of neural networks in the facial and automatic number plate recognition is. This is an example of reproducible neural networks (train station once, deploy everywhere), which makes them more attractive commercially. If not-reproducible neural networks is only made economically possible due to the rapid development of special hardware (the above-mentioned Intel product, for example), then the use of AI in face recognition and ANPR has been established for a long time.
The use of AI in face recognition and ANPR has been established for a long time
A different kind of AI-analytics, we will investigate the behavior of analytics. This feature, probably more than any other, brings to video-surveillance-system closer to what is happening in front of the camera. His potential is huge.
How Behavior Analytics Works
From a technical point of view, analytics behavior combines artificial intelligence with a classical algorithmic approach. A neural network trained on a variety of scenarios can determine that the position of the bodies, heads and limbs of people in the camera field of view. The algorithm gives an array with the data containing descriptions of their poses.
Conditions can be set, for the data to detect a particular pose, such as the raised hands, bowing or crouching people. Developers can use this to quickly create new recognition to identify tools that are potentially dangerous for the specified behavior of a government or business client. There is no need for additional training of the neural network.
How Behavioral Analytics Can Be Used
Someone crouched next to an ATM machine could be a technician, CIT guard, or burglar. Bank’s security are to be informed in any of the cases.
One person-shooter position, together with a Bank clerk or cashier with your hands raised is an indication of a robbery. The system can be configured to automatically send notifications with a surveillance snapshot of the police, so that they can judge the threat and, if necessary, take action. It is important that the police receive the alarm, also if the employee must turn in the alarm.
In many cases, the attention should be directed on an outstretched individual. This could be someone who needs emergency assistance, or it could be someone sleeping in an inappropriate public place, for example, a 24/7 ATM room.
Behavioral analytics can also be used to safety in the workplace ensure. For example, track whether the employees hold the handrails when on the stairs, in a factory or a construction site.
Now What?
Behavioral analysis can be used wherever your customer takes the fancy of you. With this function you can practically every pose has a potentially dangerous behavior is detected. The timely response to an alarm contributes to the prevention of damage to the material, or in other situations, losses.
Practically every pose has can be recognized on a potentially dangerous behavior
A range of possible development for the behavioral analytics is the ability to analyze a sequence of poses of the same person, or a combination of poses and relative positions of individuals. This is the next stage of evolution, in AI applications in video surveillance: the move from “identify” to “understand” behavior in real-time.
In its simplest form, can be used in this type of analysis, in order to detect deviations from the search procedure, in prisons, if a person is checked, you must assume there is a predefined sequence of poses. An extended form allows it to detect a type of abnormal behavior, such as a brawl breaks out in a public space. In the ideal case, conduct analytics to predict dangerous situations can tell based on almost imperceptible notes that he collected statistics and Big Data analysis.
At the moment, this sounds like pure fantasy, but the, what seemed like whimsy not too long ago is now a reality, with AI. It’s already beaten people in chess and the game of Go (Weiqi). To dub artificial intelligence will be able to people to charades a day? It is quite possible that we’ll soon see for yourself.

Released on Mon, 09 Dec 2019 05:38:03 +0000

Leave a Comment