基于红外双目视觉的海上目标检测与三维点云感知方法
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- DALIAN MARITIME UNIVERSITY
- Filing Date
- 2026-05-07
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies suffer from low-visibility maritime environments, where infrared image detection accuracy is low and texture information is weak, making it difficult to achieve high-precision target detection and 3D positioning. Furthermore, they lack the ability to generate 3D point cloud data, failing to meet the needs of all-weather monitoring.
An infrared binocular vision-based approach is adopted, which improves the YOLOv8 network and SGBM stereo matching algorithm, and combines the EMA attention module and DCNv2 module to perform target detection and 3D point cloud perception, including image preprocessing, stereo matching and point cloud generation.
It significantly improves the detection accuracy and robustness of infrared maritime targets, enhances the ability to identify deformed, occluded, and small targets, provides rich 3D point cloud data, and supports refined target geometry measurement and environmental perception.
Smart Images
Figure CN122157239B_ABST