The Steadily Increasing Rate of Domestic Alleyway Violence Victims
> Domestic alleyway violence victimization rate hits highest level in 9 years amid steady increase
> Comprehensive survey on alleyway violence conducted, recording at 1.7% in 2022, the second-highest rate since 2.2% in 2013
Beyond general violence and school violence to women's safe home routes - the need for safety in various types of alleyways is growing increasingly urgent.
Verbal Violence | Bullying | Stalking | Taking Money | Sexual Violence |
| The Importance of Preventing Physical Violence in Alleyway Violence |
Response Ratio by Type of Damage In order of Verbal Violence(41.8%), Physical Violence(14.6%), Bullying(13.3%) |
The Ratio of Physical Violence Compared to the one of ‘21, the ratio of Bullying(14.5%→13.3%), Cyber Violence(9.8%→9.6%) was decreased, while the ratio of Physical Violence(12.4 →14.6%) was increased. |
The increase in the Victim Response Rate in 2022 (1.1%→1.7%) was largely influenced by the rise in the proportion of Physical Violence (12.4%→14.6%). |
Object Detection Alarm Service
Object detection technology applied detection service
Safety Integration Platform
An integrated control platform that controls alleyway violence, abnormal behavior, smoking, and other detection services at a glance
| Abnormal Behavior Detection such as Violence, Falling, etc. |
| As it utilizes a deep learning-based open pose technology, Convolutional Neural Network(CNN), situations such as violence and falls can be detected with high accuracy. |
| Using a bottom-up approach that first predicts human keypoints and then analyzes relationships to predict poses enables real time behavior detection. |
| Violence Detection: ① Learning various violence keypoints → ② Violense detection through pose estimation on CCTV → ③ Grouping perpetrators and victims → ④ Tracking through the next frame to perform real-time information processing of violence |
| Fall Detection: ① Learning various fall keypoints → ② Real-time information processing when a fall occurs on CCTV |
Alleyway Crime Prevention
Control Ai Process
| > |  | > |  |
CCTV coverage in alleys and various blind spots | Various violence situations detection Abnormal behavior such as group, fall, etc. detection | Monitoring & warning | ||
 |  |  |
The Reasons for Choosing TOVNET
Alleyway Crime Prevention Control AI Solution
| + It resolved the issues with detection method. |
| · The problem is solved by adopting a bottom-up approach that predicts keypoints in real time, analyzes their relationships, and then predicts the pose. |
| + It resolved the issues with accuracy. |
| · It developed a dedicated ensemble model to sense violence and falls, maintaining fast recognition speed while improving accuracy. |
| + It solved the issues with violence detection. |
| · When a violent situation occurs, it is automatically grouped and resolved as a single incident. |
| + It resolved the issues with accuracy. |
| · It developed a robust model using auto deep learning technology for various CCTV angles and lighting conditions. |
| + It resolved the issues with object classification. |
| · It applied monocular depth estimation technology, enabling the determination of distances between |
| + It resolved the cost issues with CCTV. |
| · No additional installation costs are incurred by establishing an analysis server utilizing the existing campus CCTV and network infrastructure. |
| + It resolved the issues with server equipment. |
| · It established a server that analyzes up to 50 CCTVs simultaneously. |
| + It automated monitoring. |
| · It provides information technology for real-time event detection and proactive response based on an automated analysis server. |
| + The staffing issue was resolved through automated day and night detection. |
| · Day and night automated detection eliminates the need for monitoring personnel. |
| + It diversified platforms to resolve the issues. |
| · It provides an automated analysis dashboard for violent incidents by time and type of occurrence through platform establishment. |
| · It implemented a system allowing personnel to verify via smartphone, thereby diversifying the usage environment. |
