3D Scan Data Subsampling for CAD Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Processing large 3D scan data files is labor-intensive and resource-intensive due to their large size and noise, leading to hardware overload and increased labor time, with existing subsampling methods producing unstructured data that is not efficiently usable by CAD software.
Innovation Solution
A subsampling method that aligns 3D scan data with x, y, z axes, splits it into smaller volumes, and uses weighted averages of nearest neighboring points to create structured data points at the intersection of pre-defined planes, reducing file size and noise, and can be run on standard hardware without requiring CAD software or advanced hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If 3D scan data is processed directly in CAD software, then processing accuracy is maintained, but hardware performance is overloaded and processing time increases significantly
Solution Approach 1:
The patent divides the 3D scan point cloud into multiple spatial blocks or regions, processing each block independently to generate subsampled data. This segmentation reduces the overall data volume while preserving essential geometric information, enabling faster processing without significant loss of accuracy.
Solution Approach 2:
The patent extracts and removes redundant points from the original point cloud through subsampling techniques. By selecting only the most representative points from each spatial block, the system reduces data volume while maintaining the essential geometric features needed for accurate CAD processing.
2Productivity
If 3D scan data is subsampled to reduce file size, then processing speed improves, but data structure becomes unstructured and noise remains
Solution Approach 1:
The patent applies different subsampling strategies to different spatial blocks based on their local characteristics. Each block is processed independently with appropriate sampling density, allowing the system to maintain structured data output while reducing overall file size. This local quality approach ensures that noise is minimized and data structure is preserved.
Solution Approach 2:
The patent performs preliminary processing steps including point cloud filtering, spatial block division, and structured subsampling before the data is used in CAD software. These preliminary actions organize the data into a structured format that is directly compatible with CAD requirements, eliminating the need for additional structuring steps later.
3Manufacturing precision
If manual filtering of point cloud noise is performed, then data quality improves, but labor time increases significantly
Solution Approach 1:
The patent implements automated subsampling algorithms that automatically filter noise and structure the point cloud data without requiring manual intervention. The system self-processes the data through computational geometry operations, eliminating the need for manual filtering while maintaining high data quality. This automated approach reduces labor time from hours to minutes.
4Productivity
If specialized hardware is used for processing large 3D data files, then processing performance improves, but system complexity and cost increase
Solution Approach 1:
The patent replaces the need for specialized hardware with optimized software algorithms. By implementing efficient data structures, spatial indexing, and parallel processing techniques in software, the system achieves high processing performance on standard hardware. This substitution eliminates the need for expensive specialized equipment while maintaining productivity.
Data Source
AI summary
The point cloud scan image data first is structured with a x-y-z coordinate system, then split into smaller Cluster Bounds (CB) volumes and a CB x-y-z is imposed on the CB centroid. The Euclidean x-y-z distance (ED) is calculated for each data point. Only points within Radial Distance (RD cylinder) are retained/processed. Data points within RD are weighted; points closer to centroid having higher weights. Each data point has Weighted Value WV (square of the inverse of the x-y-z distance). Each x,y,z has Weighted Average (WAV) per axes. WAV is sum of each original point coordinate times WV, divided by sum all WVs. Resultant is one subsampled point per axis. One WAV per CB is selected based upon axis nearly orthogonal to x,y,z WAV points. All CBs subsampled in parallel. Subsampled output comma separated coordinate file for CAD program.