Surface filtering is a powerful tool in machine vision used to generate “cleaner” image data and/or segment features for inspection. Commonly found in 2D vision systems, this capability is now being used in 3D vision solutions where filtering on both intensity (2D) and height data (3D) is used in combination to achieve more accurate results.
Surface Filter Tool in Gocator®
The Surface Filter tool provides several common vision processing filters that you can apply to surface data, as well as two “cropping” filters that output a subset of the surface data, letting you pre-process scan data to get more repeatable measurements.
Surface filters are used to remove noise and generate a cleaner, more accurate 3D scan of the target (original image on left, filtered image on right)
Users can easily set the parameters of the Surface Filter tool based on their specific needs––including region, intensity vs. height map as data output, number of filters to be chained, filter types, kernel size (i.e., number of data points), and more.
There are fifteen different surface filters in Gocator®, including:
- Open
- Close
- Erode
- Dilation
- Morph Gradient
- Sobel Magnitude
- Laplacian
- Gaussian
2D view (sobel magnitude)
You can enable up to seven of the filters at once, in any order. Filters in the tool are chained together, and any Surface or Feature tool can use the resulting filtered surface data as input. Data from the tool can then be used with existing native tools for final measurement.
In Summary
The Surface Filter tool in Gocator® allows you to input scan data, execute multiple steps of data processing (in any sequence), and generate a new, cleaner or segmented surface as output to which you can then apply measurements. The tool also allows you to work with both 2D intensity and 3D height data, which makes it easier and more accurate to locate and identify parts and increases your overall measurement repeatability.
For more information on our LMI Products give us a call at 800-214-0322