On-Machine Workpiece Measurement Using Rotary Multi-Angle Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine tool measurement methods using single-lens cameras lack depth data, leading to inaccurate workpiece shape understanding, and methods requiring multiple cameras are prone to errors if optical axis intersections are not accurately calculated.
Innovation Solution
A machine tool on-machine measurement method that uses a rotary feed axis to orient the workpiece surfaces in the line-of-sight direction of a single imaging device, allowing for accurate measurement of positions in the X-axis and Y-axis directions and thickness in the Z-axis direction, minimizing errors related to camera optical axes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single-lens camera is used to image the workpiece, then the device complexity is reduced, but the measurement precision deteriorates due to lack of depth data
Solution Approach 1:
The patent rotates the workpiece around the optical axis to capture images from multiple angular positions (different dimensions), then reconstructs three-dimensional shape information from these two-dimensional images. This allows depth information to be obtained without using multiple cameras positioned at different spatial locations.
Solution Approach 2:
The patent introduces dynamic rotation of the workpiece during the measurement process. By rotating the workpiece to different angular positions and capturing images at each position, the system dynamically gathers multi-view data that enables accurate 3D reconstruction using a single stationary camera.
2Measurement precision
If the workpiece is imaged from two viewpoints to obtain depth information, then the measurement precision improves, but the device complexity increases due to requiring two imaging devices
Solution Approach 1:
Instead of using two cameras at different spatial positions, the patent uses a single camera and rotates the workpiece to capture images from different angular positions. This transforms the problem from a spatial multi-camera configuration to a temporal multi-angle configuration, achieving the same measurement goal with reduced device complexity.
Solution Approach 2:
The patent creates multiple virtual viewpoints by rotating the workpiece and capturing images, then uses image processing to synthesize depth information. This approach copies the effect of multiple cameras through sequential single-camera imaging and computational reconstruction.
3Measurement precision
If two imaging devices are used to capture the workpiece from different viewpoints, then depth information can be obtained, but reliability deteriorates if optical axis intersection is not accurately calculated
Solution Approach 1:
The patent uses the workpiece itself as the measurement target and rotates it to multiple positions, using its own geometry to generate multiple viewing angles. The known geometry of the workpiece fixture and workpiece is used to automatically calculate position and orientation, making the system self-calibrating and less sensitive to camera optical axis errors.
Solution Approach 2:
The patent incorporates feedback by using the captured images and known workpiece geometry to calculate and refine the position and orientation of the workpiece. This iterative process compensates for potential errors and improves measurement reliability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
When measuring the position of a workpiece (W) fixed to a table (16) of a machine tool: rotary feed axes (A, C) of the machine tool are pivoted such that an upper surface of the table (16) faces in the direction of a line of sight (O0) of a camera (120), and an image of the workpiece (W) is captured; the rotary feed axes (A, C) are pivoted such that a side surface of the table (16) faces in the direction of the line of sight (O0) of an image capturing device, and an image of the workpiece (W) is captured; and the position of the workpiece (W) is measured on the machine tool on the basis of the captured images.