An individual deviation and group behavior statistical field coupling-based target anomaly detection method, system, device and storage medium
By coupling individual behavior with the statistical field of group behavior, the anomaly detection process is dynamically adjusted, which solves the problems of high false alarm rate and insufficient robustness in the existing technology, and achieves high accuracy and real-time adaptive target anomaly detection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING ZHURUI TECHNOLOGY CO LTD
- Filing Date
- 2026-05-07
- Publication Date
- 2026-07-03
AI Technical Summary
Existing technologies lack the ability to incorporate the overall state of the group into the individual anomaly detection logic in dynamic environments, resulting in high false alarm rates, insufficient robustness, and poor real-time adaptability. Furthermore, static feature splicing methods lack dynamic mechanisms.
By acquiring multidimensional behavioral feature data, an individual behavior distribution model and a group behavior statistical field model are established. Individual deviation and group statistical variables are calculated. A modulation function is used to generate coupled anomaly indicators, and the anomaly judgment indicators and sensitivity are dynamically adjusted to achieve real-time detection.
It improves the accuracy of anomaly identification, reduces the false alarm rate, enhances the robustness and adaptability of the system, is suitable for real-time deployment on edge devices, and reduces the dependence on cloud computing power.
Smart Images

Figure CN122333282A_ABST