Seoul to install 3,000 cameras with artificial intelligence to analyze movement patterns and detect possible crimes – FindNow
Connect with us

Artificial Intelligence

Seoul to install 3,000 cameras with artificial intelligence to analyze movement patterns and detect possible crimes



In ‘ Minority Report ‘, a film starring Tom Cruise in which he plays the chief of the Washington DC PreCrimen police force, they use young mutants who can see the future to predict the crimes that will happen next and stop the suspect sooner. to do anything. It is science fiction, of course, not like the 3,000 artificial intelligence cameras to be installed in Seoul, the capital of South Korea.

This initiative is carried out by the Electronic And Telecommunications Research Institute (ETRI) and the Seocho District, which according to ZDNet , will install 3,000 cameras throughout the district to process data from pedestrians and “measure the probabilities that a crime takes place . ” If positive, the camera will alert police stations to dispatch personnel to the location.

The cameras will use artificial intelligence to process the location, time and behavior patterns of people on the streets to predict possible crimes, see a robbery. In that sense, the cameras will analyze when someone is walking at a normal pace or is chasing someone , in addition to detecting if they are wearing glasses, a mask or a hat, and even if they are carrying a bag or suspect object. They will also know if it is day or night.

The software is still under development, but ETRI aims to have the final version ready by 2022 . Subsequently, the chambers will expand to other districts in Seoul as well as other provinces. It is a daunting task, since in Seoul there are 25 districts where 9.7 million people live (in Madrid there are 6.55 million inhabitants, to put it in context). In all of Seoul there are more than 55 million inhabitants.

To train the software, the Seocho District and ETRI will use 20,000 court judgments and video files associated with crimes for the AI ​​to learn to detect the patterns. Once the system is capable of it, it will be able “to compare if what is being recorded in the present coincides with past criminal patterns”, they explain in ZDNet.

The software will train with sentences and recordings of previous crimes

It is easy to see it with an example: if there are many videos in which you see how someone with a mask follows a person and steals a backpack, the AI ​​will understand that if that is happening again it is likely that a similar crime will be committed. In that case, the camera will analyze the probabilities and, if they are higher than a certain percentage, will notify the nearest police station and they will send a patrol to take a look.

Similar systems have already been tried on occasion and, in fact, some studies show positive results. A NIST study analyzed a database of 1.6 million people registered by the police and the result was a 95% success rate in identifying criminals . In another that took place in the UK , however, it was only successful in 45% of 200 tests. The Chinese police, for their part, have glasses with facial recognition that, apparently, have worked well in detecting certain individuals.