3D Scan Registration Using Discriminative Line-Pairs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for 3D scan registration are prone to misalignments due to unstable compass data, leading to inefficiencies and inaccuracies in aligning point clouds, particularly in applications like building construction where precise spatial alignment is crucial.
Innovation Solution
A computer-implemented method and system that extracts discriminative line-pairs from point clouds, identifies matching line-pair groups, and computes a global transformation matrix based on best orientation angles to align point clouds, thereby improving registration robustness and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional registration methods are used, then processing time is reduced, but alignment accuracy deteriorates due to unstable compass data
Solution Approach 1:
The patent extracts and removes the harmful influence of unstable compass data by introducing a virtual reference coordinate system that is independent of physical compass measurements. The method extracts geometric features (line-pairs) from point clouds and uses them to establish transformations without relying on magnetic compass data, thereby taking out the source of instability from the registration process.
Solution Approach 2:
The patent introduces a virtual reference coordinate system as an intermediary between the two point clouds to be registered. This virtual reference acts as a mediator that does not depend on unstable compass data from either scan. By transforming both point clouds to this virtual reference through geometric feature matching, the method achieves reliable alignment without being affected by compass instability.
2Measurement precision
If discriminative line-pair extraction and multiple transformation matrices are computed, then alignment accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent segments the complex registration problem into manageable parts by first extracting discriminative line-pairs from point clouds, then identifying matching line-pair groups, and finally computing transformations for each group. This segmentation allows the method to handle complexity systematically through multiple discrete steps rather than attempting a single complex transformation.
Solution Approach 2:
The patent computes multiple transformation matrices for different matching line-pair groups rather than seeking a single perfect transformation. By computing transformations for multiple groups and selecting the best one, the method uses partial actions (multiple attempts) to ensure accurate registration, accepting the computational overhead as necessary for high precision.
3Reliability
If multiple matching line-pair groups are identified with multiple criteria, then registration robustness is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary actions by pre-identifying discriminative line-pairs and their matching groups before the actual registration computation. By pre-processing the point clouds to extract and organize matching line-pair groups with multiple criteria (geometric constraints, compass angle criteria), the method prepares the data structure in advance, making the final transformation computation more efficient despite the initial processing overhead.
Data Source
AI summary
Computer implemented methods and computerized apparatus are provided for global registration between a first point cloud and a second point cloud obtained by a scanning device on an identical spatial scene at two separate instances. The method comprises extracting a first set of discriminative line-pairs from the first point cloud and a second set of discriminative line-pairs from the second point cloud, wherein a discriminative line-pair is a line-pair having high discriminative power compared to a randomly selected line-pair. In some embodiments, then a plurality of matching line-pair groups between the two sets of discriminative line-pairs are identified in accordance with one thresholding criterion related to between-line relationship, line geometry and line location; and a compass angle criterion related to compass errors of the scanning device. The method further comprises finding most reliable correspondence between the two point clouds by voting and then computing a global transformation matrix. Finally, the global transformation matrix is used to align the two point clouds. Embodiments of the present invention provide an accurate and efficient registration especially for building construction applications.