地图中红绿灯信息的构建方法、装置、设备及存储介质
By combining image detection and LiDAR point cloud information in autonomous driving, the three-dimensional position information of traffic lights is screened out, solving the problems of high labor costs and low efficiency in existing technologies, and realizing efficient and accurate three-dimensional position construction of traffic lights.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NEOLITHIC ZHIHUA (BEIJING) TECHNOLOGY CO LTD
- Filing Date
- 2023-01-13
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for constructing 3D location information of traffic lights in autonomous driving suffer from high labor costs, low efficiency, and difficulty in rapidly improving accuracy. In particular, traffic light matching methods based on high-precision maps require a large amount of data for training and are difficult to achieve good results in a short period of time.
By obtaining the detection box location information of traffic lights from the image, combining it with LiDAR point cloud information, and using the external parameter matrix transformation from camera to LiDAR to convert the data range, point cloud cluster segmentation and scoring are performed to filter out the three-dimensional location information of traffic lights.
It significantly reduced labor costs, improved the efficiency of constructing 3D location information for traffic lights, achieved an accuracy rate of over 90% in most scenarios, and greatly improved mapping efficiency.
Smart Images
Figure CN115984820B_ABST