Spherical Grid Visibility Analysis for Point Cloud Image Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The visibility problem arises when point clouds from lidar sensors and images from cameras are combined, as different viewing angles can cause points obscured by objects to be incorrectly labeled, leading to inaccurate mapping and navigation.
Innovation Solution
A method to determine a target quantity of point groups visible or not visible from a specified viewing point by transforming point coordinates into spherical coordinates, using a spherically curved grid to assign points to cells, and applying threshold values to identify visible points based on their distance from the viewing point.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If points from lidar sensor are mapped onto image without visibility analysis, then mapping efficiency is improved, but labeling accuracy deteriorates due to incorrect labeling of obscured points
Solution Approach 1:
The patent applies preliminary action by performing visibility analysis before mapping points onto the image. The method determines which points are visible from the camera's viewing point and filters out obscured points before they are incorrectly labeled. This preprocessing step ensures that only visible points are mapped, preventing labeling errors while maintaining efficient processing.
Solution Approach 2:
The patent extracts and removes obscured points from the point cloud before mapping. By identifying points that are hidden behind objects from the camera's perspective and excluding them from the mapping process, the method ensures that only visible points are assigned labels, thereby maintaining labeling accuracy without sacrificing overall mapping efficiency.
2Measurement precision
If all points in point cloud are processed for visibility determination, then visibility accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the point cloud processing into distinct stages: transforming points to camera coordinate system, determining visibility based on camera viewing point, and then mapping visible points to the image. This segmentation allows the system to process only relevant visible points for labeling, reducing unnecessary computational complexity while maintaining accurate visibility determination.
Solution Approach 2:
The patent applies local quality by focusing computational resources on determining visibility for points that are actually visible from the camera's perspective. Rather than uniformly processing all points with the same computational intensity, the method adapts the processing depth to the local visibility characteristics, reducing overall computational complexity while maintaining accuracy for visible points.
Data Source
AI summary
A method for determining a target quantity of point groups in a surrounding area. The point groups being visible or not visible from a specified viewing point in the surrounding area, from a quantity of point groups. The method includes: providing the quantity of point groups in the surrounding area, with coordinates in an origin coordinate system, transforming the coordinates of the quantity of point groups into spherical coordinates in a target coordinate system, wherein the viewing point lies in the surrounding area in the origin of the target coordinate system, determining a spherically curved grid in spherical coordinates in the target coordinate system, assigning each point group of at least some of the quantity to a cell or a grid point.