Calibrate Images

Vision Assistant can calibrate images containing linear, perspective, or nonlinear distortions in order to return accurate measurements. The Image Calibration step calibrates images so that inspection results are returned in real-world units.

Calibrating an image is a two-step process. The first is an offline step during which you specify the type of calibration, the calibration parameters, and the real-world unit in which you want to express measurements. The second step of the calibration process applies the computed calibration to the image during the inspection process. This step is represented in the inspection as the Image Calibration step.

Vision Assistant supports three types of calibration, which depend on the position of the camera in relation to the object under inspection and the type of lens used in the application: Simple Calibration, Calibration Using User-Defined Points, and Grid Calibration.

• Select Simple Calibration when the camera angle is perpendicular to the image plane and lens distortion is negligible. Simple Calibration transforms a pixel coordinate to a real-world coordinate through scaling in the x (horizontal) and y (vertical) directions.
• Select Calibration Using User-Defined Points when lens distortion is negligible and when the camera is not perpendicular to the image plane, which causes perspective distortion in the image. This calibration type uses a set of known mappings between points in the image and their corresponding locations in the real world to compute the pixel-to-real-world mapping for the entire image.
• Select Grid Calibration when the image exhibits either perspective or nonlinear distortion. Nonlinear distortion usually occurs when you use a wide-angle lens. Grid Calibration uses the distances between dots in a calibration grid and the distances between dots in an image of the grid to generate a list of pixel-to-real-world mappings for the calibration process.

Tip  Whenever possible, position the camera perpendicular to the object to minimize perspective distortion in the image. Use a high quality lens or telecentric lens to minimize nonlinear distortion in the image.