一种基于颜色特征与深度图像特征的交通标志牌定位方法
By combining HSV color space segmentation and LiDAR point cloud data, high-precision positioning and recognition of traffic signs in complex environments were achieved, solving the problems of low recognition accuracy and positioning precision in existing technologies, and improving the system's stability and computational efficiency in complex environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG EXPRESSWAY GRP CO LTD INNOVATION RES INST
- Filing Date
- 2025-01-22
- Publication Date
- 2026-07-17
AI Technical Summary
Existing traffic sign recognition methods have low accuracy and positioning precision in complex environments, and low computational efficiency, making it difficult to meet real-time processing requirements. In particular, they are easily affected by background interference and changes in lighting in panoramic road images.
A traffic sign localization method based on color and depth image features is adopted. It combines HSV color space segmentation and LiDAR point cloud data. The HSV color space is constructed using RGB color space data from panoramic images. The 3D spatial information of the point cloud data is used for feature matching and spatial localization. A classification model is constructed by combining local binary mode and histogram of directional gradients algorithm to achieve accurate localization and category discrimination of the signs.
It improves the recognition accuracy and positioning precision of traffic signs in complex traffic scenarios, reduces the amount of computation, enhances the stability and recognition performance of the system in complex environments, and meets the needs of real-time processing.
Smart Images
Figure CN120014602B_ABST