Enhance Edges enhances or identifies edges in an image, which can improve object identification or other downstream image processing.
This module enhances the edges (gradients) in a grayscale image. All methods other than Canny produce a grayscale image that can be used in an
Identify module or thresholded using the
ApplyThreshold module to produce a binary (black/white) mask of edges. The Canny algorithm produces a binary (black/white) mask image consisting of the edge pixels.
Settings:
Select the input image
What did you call the image in which you want to enhance the edges?
Name the output image
What do you want to call the image with edges enhanced?
Automatically calculate the threshold?
(Used only with the Canny option and automatic thresholding)
Select
Yes to automatically calculate the threshold using a three-category
Otsu algorithm performed on the Sobel transform of the image.
Select No to manually enter the threshold value.
Absolute threshold
(Used only with the Canny option and manual thresholding)
The upper cutoff for Canny edges. All Sobel-transformed
pixels with this value or higher will be marked as an edge.
You can enter a threshold between 0 and 1.
Threshold adjustment factor
(Used only with the Canny option and automatic thresholding)
This threshold adjustment factor is a multiplier that is applied to
both the lower and upper Canny thresholds if they are calculated
automatically. An adjustment factor of 1 indicates no adjustment.
The adjustment factor has no effect on any threshhold entered manually entered.
Select an edge-finding method
There are several methods that can be used to enhance edges:
- Sobel: Finds edges using the Sobel approximation to the derivative.
The Sobel method derives a horizontal and vertical gradient measure and returns the
square-root of the sum of the two squared signals.
- Prewitt: Finds edges using the Prewitt approximation to the derivative.
It returns edges at those points where the gradient of the image is maximum.
- Roberts: Finds edges using the Roberts approximation to the derivative.
The Roberts method looks for gradients in the diagonal and anti-diagonal directions
and returns the square-root of the sum of the two squared signals. This method is fast,
but it creates diagonal artifacts that may need to be removed by smoothing.
- LoG: Applies a Laplacian of Gaussian filter to the image
and finds zero crossings.
- Canny: Finds edges by looking for local maxima
of the gradient of the image. The gradient is calculated using the derivative
of a Gaussian filter. The method uses two thresholds to detect strong and weak
edges, and includes the weak edges in the output only if they are connected to
strong edges. This method is therefore less likely than the others to be fooled
by noise, and more likely to detect true weak edges.
Select edge direction to enhance
(Used only with Prewitt and Sobel methods)
The direction of the edges
are you are identifying in the image (predominantly horizontal, predominantly vertical,
or both).
Calculate value for low threshold automatically?
(Used only with the Canny option and automatic thresholding)
Select
Yes to automatically calculate the low / soft threshold cutoff for
the Canny method.
Select No to manually enter the low threshold value.
Low threshold value
(Used only with the Canny option and manual thresholding)
Enter the soft threshold cutoff for the Canny method.
The Canny method will mark all Sobel-transformed pixels with values
below this threshold as not being edges.