Mixing the democratization of artificial intelligence and the availability of metadata from image analysis pass away cards are new in terms of video surveillance – and they require a lot of thought when it comes to data protection. Surveillance cameras have long been part of many people’s everyday lives – mostly without even noticing it. Worldwide are 770 Millions of surveillance cameras in use. And: until the end 2021 over 1 billion surveillance cameras are expected to be in operation worldwide.
Video surveillance is a widely used technology
Despite its widespread use, the security market is currently in a decisive stage of technological change. Because artificial intelligence is becoming more and more popular – the availability of metadata from video surveillance is also increasing steadily. These developments are important from both a technological and an ethical perspective – and Europe is setting standards on both points.
Pass away development of CCTV cameras from the history
From military use to pass away distribution in private use to networked video surveillance, we know today: The The video surveillance market has seen several small revolutions to date. At first it was just a matter of setting up a camera and recording a defined area. One of the first major changes was the introduction of the IP camera – or network camera, connected via an Ethernet network, from a simple sensing unit to a full-fledged computer tool. At that time, in the 2000 emergency room Years, when the market for video security was booming, the first artificial intelligence algorithms were developed and used in the systems. They made it possible to analyze video flows and thus established the image analysis industry in the surveillance cameras. Thanks to the ability to analyze the amount of moving pixels in the image, these systems can, for example, detect abnormal movement in the video clip – for example, that an individual is present and report the information.
Quality of metadata: accuracy of the analysis as a result of the environment
The only disadvantage is that the environment in which the system is located determines the accuracy of this analysis. For example, heavy rain or wind can make detection more difficult by triggering false alarms or even not detecting a break-in. Another obstacle for these technologies are the costs of the infrastructure: Analyzing videos and delivering a relevant result requires computer servers with a large computing capacity. And those are not the only challenges that have become apparent over time. Surveillance cameras are often criticized because they pose another operational problem: because all recorded video clips have to be processed. Passes away requires considerable human resources, especially in urban security centers. Passes away, however, does not take into account the latest developments in video surveillance systems, but, reinforced by advances in artificial intelligence, are now becoming more and more efficient and autonomous.
Deep understanding and metadata: leap in development in video technology
The more precise the pass away Algorithms, the more proactive cameras equipped with image analysis. Passes away is particularly useful when it comes to feeding information back to the user of a video surveillance system. Pass away the new generation of artificial intelligence makes it possible today to carry out more precise preliminary analyzes of situations. In the case of a city, for example, new algorithms will be able to draw operators‘ attention to crowds or abnormal movements of crowds.
How is that possible? Quite simply with a new method for designing these image analysis algorithms: Deep Knowing. This means that large amounts of data can be used to train artificial intelligence. If we train artificial intelligence systems with all this data, they can recognize different forms: a man, a woman, a cat. But you can also recognize colors, for example a man with red tea and a mustache, a child on a scooter, a certain face. All passes away is made possible by deep learning algorithms that are able to process an exponential amount of data. In addition, there are advances in computing power and miniaturization of computer processors that are integrated directly into surveillance cameras. In this way, the costs for the use of these technologies become more affordable for end users.
Pass away future lies in deep knowing – an issue for data protection?
Today’s video surveillance systems are therefore experts in recognition of patterns has become. On the basis of what is recognized, they create a meta database that takes care of itself and enables an ever finer analysis of the images. Passes away is the basis for biometric and face recognition systems. Scary when it comes to protecting personal data? Not really, because the legal framework in Europe strictly protects individual freedoms (GDPR). From a security point of view, the goal of metadata is not to identify citizens (which is still not allowed in Europe anyway), but to optimize the use of video protection and make it proactive and proactive. Networked in-car cameras to avoid accidents, intelligent cities, optimize flows and limit incidents or cameras, recognize microexpressions of an individual with a stroke.
Pass away possibilities are unlimited and some are already a reality. Today’s security systems are designed from the ground up in such a way that they go beyond a protective logic. To 2024 are estimated 30% of cameras sold in the market to be able to embed deep knowledge. Image analysis based on artificial intelligence is booming. It is now only a matter of time: the time that is needed is needed to provide infrastructures and develop AI applications, to change society for the better.
Photo: I-Pro EMEA
The author of the technical article is Emmanuel Berthelot from I-Pro EMEA.
NachrichtenJetRead MoreSicherheit, „the, metadata, Nachrichten, potential, technology, video