Point Cloud Alignment via Database Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing measurement systems struggle to accurately and efficiently align measured point clouds with ideal point clouds in a short time, especially in manufacturing large structures, due to the difficulty in preparing measurement point clouds similar to ideal point clouds.
Innovation Solution
A measurement system that selects a data set from a database based on similarity comparison between ideal and measurement images, aligns the selected alignment point cloud with the measurement point cloud, and calculates shape deviations using the shape comparison point cloud, thereby visualizing shape deviations effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional point cloud alignment methods are used, then alignment accuracy can be achieved, but processing time becomes excessively long and computational load increases
Solution Approach 1:
The system performs preliminary actions by pre-processing images to generate ideal point clouds and storing them in a database before actual measurement occurs. When measurement data is acquired, the system quickly retrieves pre-generated point clouds that match the measured object, avoiding the need for real-time complex point cloud generation and alignment computations.
Solution Approach 2:
The system creates ideal point clouds as copies from images stored in the database, which represent the ideal shape of the measurement object. These copied point clouds are then used for alignment comparison with measured point clouds, eliminating the need to generate ideal point clouds from scratch during measurement processing.
2Measurement precision
If complex alignment algorithms are applied to improve alignment accuracy, then measurement precision improves, but device complexity and computational requirements increase
Solution Approach 1:
The system performs complex point cloud generation and ideal shape calculation in advance, storing results in the database. This preliminary computation simplifies the actual measurement process, as the system only needs to retrieve pre-computed data and perform simple comparisons, significantly reducing real-time computational complexity.
Solution Approach 2:
The system introduces an intermediate database that stores pre-processed ideal point clouds and corresponding images. This database acts as a mediator between the measurement system and the alignment process, providing ready-to-use reference data that simplifies the alignment algorithm requirements.
3Productivity
If the measurement point cloud is sparse, then measurement speed improves, but alignment accuracy deteriorates due to insufficient data points
Solution Approach 1:
The system generates dense ideal point clouds as copies from stored images, which serve as reference data. Even when measured point clouds are sparse, the dense ideal point clouds provide sufficient reference information for accurate alignment and shape deviation calculation.
Solution Approach 2:
The system changes the density parameter by generating ideal point clouds with higher point density than the measured point clouds. This parameter change ensures that even sparse measured data can be accurately aligned and compared against the denser ideal reference data.
Data Source
AI summary
According to one embodiment, a measurement system includes a processor. The processor selects a data set from a database, based on similarity obtained from a result of comparison between an ideal image and a measurement image. The database stores a data set for measurement objects each including the ideal image, an alignment point cloud, and a shape comparison point cloud, when viewed from a plurality of fields of view. The processor aligns the alignment point cloud included in the selected data set with a measurement point cloud. The processor calculates a shape deviation between the shape comparison point cloud and the measurement point cloud, which are included in the selected data set, based on a result of the alignment. The processor visualizes on a display device based on a result of the calculation of the shape deviation.