A South Korean research team has successfully developed a technology that can use real-time closed-circuit television and artificial intelligence technology to detect and predict signs of criminal activity. With this new technology, closed-circuit television is now moving beyond its original surveillance role to realize its potential as an effective way to “prevent” crime.
The Electronic Telecommunications Research Institute (ETRI) recently announced that it has completed the development of “Dejaiview”, an innovative technology that uses closed-circuit television footage, crime-related statistics, location technology and more to detect and predict potential criminal signs. The possibility of a crime occurring.
Researchers at “Dejview” revealed that they focus on the trend of recurring crimes based on certain patterns, such as type, method, location, time, etc. “Dejview” is an artificial intelligence-based technology that can measure and predict criminal behavior. Determine the risk of potential crime by comparing current environmental/social patterns to past criminal cases.
ETRI’s “Dejview” predicts the occurrence of criminal activity through an “opportunity-based approach”, which can be divided into two different technologies based on application goals and methods. First of all, time/space-oriented crime prediction technology analyzes criminal activities and their opportunities based on factors such as time and space.
For example, if a specific combination of environmental factors is observed at a remote/quiet location where crimes occurred late at night in the past, the system will consider the risk of crime occurring at that location to be very high. Through this, the system will prevent crimes in advance. The Dejview system is expected to be used mainly by police and local government integrated closed-circuit television control centers across the country.
By using a set of data related to past criminal records and crime-related statistics, Dejview’s artificial intelligence automatically analyzes real-time closed-circuit television footage to compare and measure current situations and their similarities to past crimes. In addition, users can better understand the current situation, such as immediately identifying and tracking suspected crimes/disasters (i.e., stalking, collapse, early fire, drug trafficking, smuggling, etc.)
ETRI partnered with Seocho District to develop a Predicted Crime Map (PCM) by analyzing a total of 32,656 CCTV footage, including incidents/criminal activity that occurred in the area over the past three years since 2018. Through relevant statistics, the system displays areas with high probability of criminal activity by classifying the time, place, type of event, etc. of the event.
In addition, PCM presents real-time crime risks on the home screen. Using this feature,”crime controllers” can grasp the current existing crime risk levels by region/region. In other words, selective surveillance can become a viable option, focusing more on specific areas where criminal activity is at high risk. According to performance testing conducted by the Telecommunications Technology Association (TTA), the crime prediction performance of the Predicted Crime Map is 82.8%
ETRI has also developed individual-centered recidivist prediction technology as a second key feature of Dejview. The technology is strictly suitable for high-risk individuals who are highly likely to commit another crime. It analyzes individual risks based on their exercise patterns.
Current electronic surveillance systems use location information to determine whether a specific individual violates restrictions. If ETRI’s technology can be applied to existing systems, it can regularly check whether individuals have violated movement restrictions due to work and other reasons, and use AI technology to conduct in-depth analysis of their behavioral patterns and risks. Recipients.
ETRI’s research team plans to develop Dejview’s source technology into a professional system focusing on security-related services. They plan to expand it to a variety of security-related services, including crime-prediction systems in specific areas, proactive response systems for major infrastructure/critical facilities (airports, energy facilities, factories, etc.), security systems for major national events, pre-emptive response systems Systems that target dangerous behavior/intentions for personal security purposes, etc.
To this end, it is necessary to develop a world-class video security artificial intelligence technology to accurately analyze, understand, and infer situations by collecting visual information from real-time closed-circuit television recordings/images.
Based on the developed Dejview technology, ETRI researchers will work with local governments and agencies to provide crime signal detection and prediction services for a variety of safe venues and purposes. In addition, ETRI plans to expand existing technology to develop artificial intelligence-based electronic surveillance technology that can detect and proactively respond to abnormal behavior by high-risk individuals under electronic surveillance. Commercialization of Dejview is expected to take place by the end of next year.
Kim Geon-woo, senior researcher at ETRI’s AI Fusion Security Research Division, said: “We have successfully developed a video security technology that uses closed-circuit television information to predict and prevent the likelihood of criminal activity, rather than simply detecting the occurrence of crimes. Through this technology, we have now laid the technical foundation that allows us to leap to a new level and develop future, advanced social security systems.”
There are links in the video below this video. If you are interested, you can open it and have a look.
Thank you for watching this video. If you like it, please subscribe and like it. thank
Original text:https://techxplore.com/news/2024-09-korean-ai-cctvs-criminal.html
Oil tubing:
