Filters

NI Vision Assistant

Filters

Prepares an image for processing so you can extract only the information you need from the image. Most of these filters apply a kernel across the image. A kernel represents a pixel and its relationship to neighboring pixels. The weight of the relationship is specified by the coefficients of each neighbor. The following controls are available.

  • Step Name—Name to give the step.
  • Filters
    Image Source Original input image.
    Smoothing-Lowpass Lowpass filtering. Smoothes images by eliminating details and blurring edges.
    Smoothing-Local Average Local averaging of the image pixels based on the kernel.
    Smoothing-Gaussian Gaussian filtering based on the kernel. Attenuates the variations of light intensity in the neighborhood of a pixel. The Gaussian kernel has the following model: a d c b x b c d a

    where a, b, c and d are integers and x < 1.

    Smoothing-Median Median filtering. Each pixel is assigned the median value of its neighborhood.
    Edge Detection-Laplacian Laplacian filtering. Extracts the contour of objects and outlines details. The Laplacian filter kernel has the following model: a d c b x b c d a

    where a, b, c, and d are integers and x is greater than or equal to the sum of the absolute values of the outer coefficients.

    Edge Detection-Diff. Differentiation filtering. Produces continuous contours by highlighting each pixel where an intensity variation occurs between itself and its three upper left neighbors.
    Edge Detection-Prewitt Prewitt filtering. A highpass filter that extracts the outer contours of objects.
    Edge Detection-Sobel Sobel filter. A highpass filter that extracts the outer contours of objects.
    Edge Detection-Roberts Roberts filter to detect edges. Outlines the contours that highlight pixels where an intensity variation occurs along the diagonal axes.
    Convolution-Highlight Details Convolution kernel that highlights the edges of an image.
    Convolution-Custom Custom filtering using the kernel coefficients and size that you specify.
  • Kernel Size—Size of the structuring element. Valid values include 3 x 3, 5 x 5, and 7 x 7.
  • Filter Size—Sets the size of the filter for the Lowpass and Median functions.
  • Kernel—Specifies the kernel coefficients.
Tip  The kernel coefficients for each filter are set to default values. You may need to experiment with different coefficients and the kernel size to get the results you want.