基于雷视融合的触发式车辆识别与跟踪方法、装置及其可读存储介质
By employing a transient asynchronous fusion mechanism and trigger-based computation, the problems of time mismatch and wasted computing power in radar and video fusion are solved, improving the robustness and accuracy of vehicle recognition and adapting to edge deployment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HANGZHOU FEELING TECH CO LTD
- Filing Date
- 2026-05-11
- Publication Date
- 2026-07-17
AI Technical Summary
Existing radar and video fusion technologies suffer from the loss of high-frequency dynamic information from radar or distortion of spatial semantic features in video due to forced alignment of time scales. They also cannot adaptively adjust fusion weights, and the all-time computing mode leads to a waste of computing resources, making it difficult to meet the lightweight deployment requirements at the edge.
The radar-visual transient asynchronous fusion mechanism is adopted, which uses video frames as time anchors and is based on radar data-driven time-series recursive updates. During the fusion process, fusion weights are adaptively allocated, and a trigger-based calculation mechanism is combined to achieve on-demand recognition and trajectory optimization.
It improves the robustness and accuracy of vehicle recognition in complex road scenarios, reduces computing power consumption, and adapts to edge deployment environments.
Smart Images
Figure CN122176662B_ABST