Method and apparatus for extracting outlines from 3D images
By generating 2D outlines from specified regions of 3D point clouds, the method addresses inefficiencies in processing large 3D data sets, enhancing machine vision applications through reduced complexity and improved performance.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- COGNEX CORP
- Filing Date
- 2026-01-08
- Publication Date
- 2026-05-26
AI Technical Summary
Existing machine vision techniques face inefficiencies in processing large 3D point clouds due to the lack of spatial relationship information, leading to complex and resource-intensive processing, especially when time or hardware constraints are present.
A method to generate a 2D outline from a 3D point cloud by specifying a region of interest, projecting 3D points onto a 2D plane, binning and quantizing these points, and connecting their representation positions to form a polyline.
This approach reduces data complexity, improving processing efficiency and enabling applications such as feature extraction, metrology, and defect inspection by converting 3D point clouds into manageable 2D outlines.
Smart Images

Figure 2026086411000001_ABST