Seocho-gu, the Nation's First to Use 'Predictive Crime Map' for Targeted Prevention

Joint Development by Former Korea Electronics and Telecommunications Research Institute...CCTV Analyzes Past Crime Big Data to Indicate Crime-Prone Areas by Time Slot Over 32,000 Cases Linked with CCTV in Last 3 Years, Displayed by Date, Location, and Incident Type...Focused Monitoring After Identifying Crime-Prone Areas by Time Slot Expected to Improve Control Center Efficiency

[Asia Economy Reporter Jongil Park] Seocho-gu has launched a meticulous smart network administration to eliminate crime blind spots.

Seocho-gu announced that it has developed and is operating the nation’s first ‘Predictive Crime Map’ system, which analyzes CCTV incident and accident data to predict high-crime areas using big data and conducts focused surveillance by time zone.

Developed jointly with the Electronics and Telecommunications Research Institute (ETRI), the ‘Predictive Crime Map’ system analyzes 32,656 CCTV incident and accident big data cases within the district over the past three years (2018?2021). It structures the data by occurrence date and time, location, and incident type (including the five major violent crimes, traffic accidents, fires, etc.) to display areas with a high likelihood of crime occurrence.

The system’s operation principle is simple. Operators working at the Seocho Smart Hub Center check the frequency and types of crimes by time and neighborhood on the Predictive Crime Map, identify high-risk areas marked in dark colors, and conduct focused monitoring through CCTV. For example, around 11 p.m., if the map shows a dense distribution of past thefts (10 cases) in a certain neighborhood and street, operators closely monitor pedestrians passing through that area via CCTV.

The district expects this system to improve work efficiency through selective and focused monitoring by operators and contribute to eliminating crime blind spots through meticulous surveillance like a hawk’s eye.

The reason for introducing the Predictive Crime Map was to address the difficulty in adequately responding to crime prevention, as the existing Seocho Smart Hub Center managed about 4,080 CCTVs, with each operator monitoring an average of 650 cameras.

To this end, the district has continuously updated various functions such as address search on the predictive crime map, camera search, and CCTV control features, developed jointly with the Korea Information and Communications Research Institute, incorporating big data from various incidents and accidents.

Future technological advancements of the district’s Predictive Crime Map system will continue. The district plans to introduce a ‘Crime Possibility Prediction Technology’ that uses artificial intelligence (AI) to compare real-time CCTV footage with the predictive crime map to detect situations similar to past crimes and predict the risk rate (%). This system is being promoted as part of the Ministry of Science and ICT’s project for ‘Development of Predictive Video Security Technology for Proactive Risk Response.’

For example, if a scene is captured in the early morning where a man is following a woman alone in a residential alley, the Seocho Smart Hub Center will analyze the man’s hand movements, time, facial expressions, and gestures using past crime big data. If the behavior is suspicious, the control center and police station will cooperate to contribute to proactive responses such as apprehending the suspect.

The Seocho Smart Hub Center has made significant contributions to preventing crime and ensuring residents’ safety. Last year, through CCTV video patrols, it detected 137 crimes including theft and sexual harassment, aiding in suspect arrests and responding to 9,653 incidents and accidents.

Acting Mayor Jeongwook Cheon said, “We will continue to strive to ensure there are no blind spots in the safety of our residents by utilizing smart technologies such as the Predictive Crime Map system.”