how does edge(image,'log') calculate threshold?

hi friends,

As a part of my Masters curriculum I have to publish a paper during my course. I am using gradient and morphological operations for fingerprint image segmentation.

For command img2=edge(img1,'log'), function edge calculates threshold automatically.

How do I represent this in my paper? Is it ok that I use ">>threshold using LoG"? Or I have to explain algorithm for threshold calculation so this remains programming language independent.

I opened edge.m editor but could not understand how threshold is calculated.

Answer 1

[Image Analyst]

The method is not specified. If you need to know you'll have to call imgradient() and then do the thresholding yourself.

[Sam]

I have specified the method. it is 'log'... "The Laplacian of Gaussian method finds edges by looking for zero crossings after filtering I with a Laplacian of Gaussian filter."

BW = edge(I,'log',THRESH) specifies the sensitivity threshold for the Laplacian of Gaussian method. edge ignores all edges that are not stronger than THRESH. If you do not specify THRESH, or if THRESH is empty ([]), edge chooses the value automatically .

[Image Analyst]

You specified the edge detection method, not the thresholding method. LoG mean Laplacian of Gaussian. It refers to a kernel , not a threshold. Basically it blurs the image twice and subtracts the result from the original image, though it can all be done in one single step. Anyway, you can see that if you don't specify the threshold, it does it automatically using an unspecified method.