CalibrateGrid

Utilizing a user-configurable calibration grid pattern, CalibrateGrid creates a 2D transformation to convert between pixel and real-world coordinate systems. CalibrateGrid also accounts for linear, non-linear and lens distortion.

About CalibrateGrid

CalibrateGrid computes a 2D transformation between the set of points identified by the vertices of a checkerboard calibration grid pattern. During analysis, CalibrateGrid finds the pixel locations of the grid cell intersections, and then maps them back onto a model of the grid pattern.

In addition, CalibrateGrid uses a reference to a set of calibration points and a lens distortion model to compute a transformation mapping between image and real-world coordinates.

Note: To view or use vision tools on the undistorted image, use the TransformImage function.

You can configure the following with the CalibrateGrid:

  • The type of calibration grid pattern to be used.
  • The spacing of the calibration grid pattern.
  • The type of real-world, physical measurement units.

Optionally, you can print a calibration plate grid. For more information, see Printing the Calibration Pattern Using the Print Grid Dialog. Once the image is acquired, CalibrateGrid automatically identifies as many feature points as possible. As soon as you press the Calibration button, CalibrateGrid begins to compute the calibration and reports a calibration point based on the spacing of the feature points.

After the calibration is performed, the CalibrateGrid cell stores the calibration data. This allows you to copy the CalibrateGrid function from one job and paste to another, or export and import as an external file using Export Cell/Import Cell. It is possible to have a different calibration for each job, as well as multiple calibrations in a single job, if necessary. Other functions that use Calibrate functions as arguments can reference the CalibrateGrid data structure.

Note: When using the Calibrate function, the physical and optical setup (the lens, sensor and the physical relationship between the sensor and the scene) must be the same for both the calibration and run-time operation. If you alter any of these items, you have to recalibrate the system. It is recommended that you use the same sensor and lens for both calibration and processing the run-time images. The sensor and lens must also retain their original setup and calibration settings. For example, altering the resolution or moving the sensor invalidates the computed 2D transformation that maps pixel to real-world coordinates.