Edge detection using Satellite Rectified Multispectral Imagery [closed]

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reas开发者_JS百科onably answered in its current form. For help clarifying this question so that it can be reopened, visit the help center. Closed 11 years ago.

I am trying to apply sobel and canny edge detection methods using MATLAB to a rectified Multispectral Satellite imagery, but it couldn't detect the edges. The results were just not as edges but as dotted noise all over.

What should I do to detect the delicate boundaries from the rectified multispectral imagery? Is there any special technique needed?

---

As an option, you can save your RGB images and try to open them with the Outliner utility http://outliner.codeplex.com/ This utility uses an alternative method of edge detection that works best on color images.

---

Assuming you're doing everything else right (with how you're calling the edge command etc.) this can happen if the images are particularly noisy. You might consider smoothing (such as filtering with a Gaussian or other kernel.) This should make the noise go away in favor of other more meaningful edges.

You should also try play with the thresholds.