Gradient Magnitude

Node Icon

The Gradient Magnitude Recursive Gaussian (GMRG) filter is a type of edge detection filter that calculates the gradient magnitude of an image in a computationally efficient way. Unlike other gradient magnitude filters, such as Sobel and Scharr, the GMRG filter uses a recursive implementation of the Gaussian filter to reduce computational complexity. The Gaussian filter is used to smooth the image, reducing noise and preserving edges. The gradient magnitude is then calculated using the smoothed image. The GMRG filter is often used in computer vision and image processing tasks, such as edge detection, feature extraction, and image segmentation, where the gradient magnitude is required. The GMRG filter is particularly useful when processing large images, as it provides faster computation compared to other gradient magnitude filters. This particular implementation computes the magnitude of the gradient of an image by convolution with the first derivative of a Gaussian.



Input image(s).

Type: Image, List, Required, Single



Output image(s).

Type: Image, List


Std Dev [px] Float

The standard deviation of the Gaussian kernel in units of voxels.

Normalize Across Scale Boolean

The filter will be normalized in such a way that the values of derivatives are not biased by the size of the object. That is to say the maximum value a feature reaches across scale is independent of the scale of the object.

See also