A high-precision map updating method and device based on binocular vision
By acquiring and processing image data using binocular vision technology and combining it with satellite positioning, high-precision maps are automatically updated, solving the problem of high cost in updating high-precision maps and achieving efficient updating of traffic elements.
CN115757447BActive Publication Date: 2026-07-21WUHAN ZHONGHAITING DATA TECH CO LTD
View PDF 3 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- WUHAN ZHONGHAITING DATA TECH CO LTD
- Filing Date
- 2022-11-16
- Publication Date
- 2026-07-21
AI Technical Summary
Technical Problem
Updating existing high-precision maps is costly and complex, making it difficult to efficiently update changes in traffic elements.
Method used
Images are acquired using binocular vision technology. Through image segmentation, stereo matching, and depth map calculation, point cloud reconstruction and registration are performed in conjunction with satellite positioning data. Differences in map elements are compared, and high-precision maps are automatically updated.
Benefits of technology
It reduced the cost of updating high-precision maps, improved update efficiency, and enabled efficient updating of traffic elements.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN115757447B_ABST
Abstract
The application discloses a high-precision map updating method and device based on binocular vision, and the method comprises the following steps: acquiring at least a group of binocular images, segmenting the binocular images by using an image segmentation model to obtain map element information of each image, performing stereo matching on the map element information of the binocular images, and calculating a depth map according to a matching result, calibrating and registering the depth map and satellite positioning data to obtain absolute position information of each pixel in the depth map, reconstructing a point cloud with a preset density according to the absolute position information of each pixel in the depth map, registering the point cloud and a high-precision map corresponding to the position of the point cloud, comparing corresponding map elements of the two after registration, and updating the high-precision map according to a comparison result. The application reduces the cost of high-precision map updating and the complexity of data processing through matching of semantic information of binocular images and satellite positioning data.
Need to check novelty before this filing date? Find Prior Art