3D Vision Sensor Reference Plane Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional three-dimensional vision sensors face challenges in accurately measuring objects when the supporting surface is not horizontal, as the reference plane defined during calibration may not align with the object's surface, leading to noise in recognition and inefficient processing.
Innovation Solution
A three-dimensional vision sensor system that includes a stereo camera, parameter calculation, and recognition units, allowing for the extraction of characteristic points, calculation of three-dimensional coordinates, and adjustment of the reference plane to set a height of zero, enabling flexible reference plane selection and correction of measurement parameters for accurate height recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the reference plane is fixed during calibration, then the calibration process is simple and fast, but the measurement accuracy deteriorates when the supporting surface is not horizontal or does not coincide with the calibration plane
Solution Approach 1:
The reference plane is made dynamic rather than fixed. The system automatically determines the supporting surface plane through image processing and calculates the transformation matrix dynamically based on the actual object position, allowing the reference plane to adapt to different supporting surfaces without manual recalibration
Solution Approach 2:
The system changes the reference plane parameters automatically by detecting the supporting surface and calculating the transformation matrix that maps the calibration plane to the actual supporting surface, thereby adjusting measurement parameters to match the actual measurement conditions
2Ease of operation
If the calibration plane is set to height zero, then height measurement is convenient, but noise is introduced when objects below the target object are captured in the image
Solution Approach 1:
The system extracts only the relevant supporting surface plane information from the image and excludes other objects below the target object. By setting the supporting surface plane as the reference and using plane equation filtering, it removes noise from objects that should not be measured while maintaining convenience for height measurement
3Adaptability or versatility
If the supporting surface is not horizontal, then the measurement adapts to various object positions, but the measurement accuracy deteriorates due to misalignment with the calibration plane
Solution Approach 1:
The system dynamically adapts to different supporting surface orientations by automatically detecting the plane equation of the supporting surface and calculating the appropriate transformation matrix, enabling accurate measurement on non-horizontal surfaces without sacrificing precision
Solution Approach 2:
The system achieves universal adaptability to various supporting surfaces (horizontal, inclined, vertical) while maintaining measurement accuracy through automatic plane detection and transformation matrix calculation, making the system multi-functional for different measurement scenarios
Data Source
AI summary
Enabling height recognition processing by setting a height of an arbitrary plane to zero for convenience of the recognition processing. A parameter for three-dimensional measurement is calculated and registered through calibration and, thereafter, an image pickup with a stereo camera is performed on a plane desired to be recognized as having a height of zero in actual recognition processing. Three-dimensional measurement using the registered parameter is performed on characteristic patterns (marks m1, m2 and m3) included in this plane. Based on a positional relationship between a plane defined as having a height of zero through the calibration and the plane expressed by the calculation equation, a transformation parameter (a homogeneous transformation matrix) for displacing points in the former plane into the latter plane is determined, and the registered parameter is changed using the transformation parameter.