What the component does
Checkerboard corners, dot centres and grid intersections provide known coordinates. Choose the geometry that the calibration software can locate reliably.
Machine Vision Calibration & Metrology
Relate image coordinates to known target geometry so that a machine vision system can measure position, shape or scale.
Where the component works
Target, lens, sensor and software
- Known target geometry at the object plane
- Lens images the target
- Sensor samples; software fits the model
The lower dimension shows working distance. The rays illustrate the object field mapped onto the sensor; software uses detected feature coordinates.
Acquire varied target poses
- Tilted target view
- Frontal target view
- A further orientation and field position
For suitable camera models, varied poses and image coverage constrain calibration better than repeated identical views.
How the system works together
Target size, feature spacing and working distance must produce enough well-resolved features across the field. Calibration is tied to the chosen lens and focus state.
What an unsuitable choice changes
Poor flatness or inaccurate feature positions can be interpreted as camera distortion. A resolution target tests image detail and is not a substitute for a geometric calibration pattern.
Components for this application
Checkerboard Calibration Targets
Known internal corners for camera intrinsics, lens-distortion correction and stereo coordinate mapping.
Dot Array Calibration Targets
Known circular-feature coordinates for geometric calibration, field mapping and multi-camera correspondence.
Grid Calibration Targets
Distributed intersections for image scale and distortion mapping.
Other relevant products and configurations
Geometric Calibration Targets
Compare geometries for camera intrinsic and distortion calibration.
Multi-Feature Imaging Test Targets
Combine scale, alignment and detail checks in one fixture.
Fiducial Patterned Glass
Alignment features for repeat imaging and fixture registration.
Concentric Square Test Targets
Nested frames show image centring, edge straightness and apparent geometry across the camera field.
USAF 1951 Resolution Targets
Checks resolved image detail as a separate test after geometric calibration.
Optical Resolution Chart, R2 Grid
Known feature positions for image scale and geometric calibration.
Optical Resolution Chart, R76 Dot Array
Known feature positions for image scale and geometric calibration.
Optical Resolution Chart, R1 Grid
Known feature positions for image scale and geometric calibration.
Optical Resolution Chart, R3 Grid
Known feature positions for image scale and geometric calibration.
Optical Resolution Chart, R4 Grid
Known feature positions for image scale and geometric calibration.
Optical Resolution Chart, Ats01 Automotive
Known feature positions for image scale and geometric calibration.
Principles and selection details
System context
A camera-based measurement system needs known physical features to relate image coordinates to the object. Calibration patterns help estimate image scale, lens distortion and camera pose; fiducials support alignment between images, fixtures and parts. Choose a pattern that your software can locate reliably.
Selection considerations
Match target pitch and active area to the field of view and pixel sampling. Pattern-position errors and substrate flatness enter the measurement uncertainty, so a nominal pitch alone does not define calibration accuracy. Distortion calibration and sharpness evaluation are different tasks.
Camera geometry, stereo and robot vision
A camera target can support intrinsic calibration, stereo-camera relationships or a robot vision workflow, depending on the software and captures. A robot hand–eye calibration also requires corresponding robot poses; the target alone does not determine the robot-to-camera transform.
Sharpness and dimensional confidence are separate
A sharp image may still contain geometric distortion, while a calibrated camera can be out of focus. Combine geometric targets with a suitable resolution check when both dimensional measurement and small-feature visibility matter.
Connect target geometry to the camera model
Dot centres and checkerboard corners are alternative coordinate features. Match their count, pitch and ordering to the software, collect views across the usable field and validate independent dimensions after fitting. Use a resolution target separately when the question is whether fine detail can be resolved.
Related applications
Planning your project
Machine-vision calibration relates known physical target geometry to image coordinates and the measurement model.
Identify whether the task is intrinsic calibration, stereo geometry, hand–eye transformation or image scale before choosing a checkerboard, circles, grid or scale.
Does low reprojection error prove dimensional accuracy?
It describes fit to the captured observations. Verify real dimensions with independent geometry in the intended working configuration.