Requirements for an NI Vision Calibration Grid

Issue Details

• I want to use the NI Vision Calibration VIs in LabVIEW or the Image Calibration step in Vision Builder for Automated Inspection. I know I can calibrate using a grid of dots, where dots refer to the circles in the calibration grid. What are the requirements for the grid?
• Where can I find an example calibration grid for my vision application? 

Solution

There are several factors to take into account when choosing or creating a calibration grid of dots.

• Minimum Dot Size

The actual size of the dot will vary based on the number of pixels of the camera and field of view selected of the camera. An optimal size is at least 9 pixels (3 x 3), which will ensure the camera will detect the dot accurately and IMAQ Vision will correctly process the dots.

• Minimum Dot Spacing

The individual dots should be at least 4 pixels apart. This will prevent two or more dots from being processed as one dot.

• Quantity of Dots Consideration

• As the number of dots in the field of view increases, the calibration of the image (in software) becomes better. For irregular or curved surfaces, a greater number of dots will insure the contour of the field of view is more accurately represented.

 

• Grid Layout

The dots need to be in vertical and horizontal lines that are perpendicular to each other to make up the grid:

                 

Additional Information

Examples NI Vision ships with 3 calibration examples for LabVIEW that install by default in<labview>\examples\Vision\2. Functions\Calibration You may use the calibration grid installed with NI Vision by selecting Start»All Programs»National Instruments»Vision»Documentation»Calibration Grid. The dots have radii of 2 mm and center-to-center distances of 1 cm. If your field of view is too small to view a large grid of dots at this scale, you can print the grid at a smaller scale.

See NI Vision Concepts Help  and Perspective and Nonlinear Distortion Calibration  for more details on calibration grids and the calibration process.