Skip to content

Instantly share code, notes, and snippets.

@bikz05

bikz05/gist:27c50374616678ba17df

Created December 20, 2014 11:24
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save bikz05/27c50374616678ba17df to your computer and use it in GitHub Desktop.
Save bikz05/27c50374616678ba17df to your computer and use it in GitHub Desktop.
Download ZIP
Image Gradients
Raw
gistfile1.txt
Dervative is rate of change.
`Hello`
@bikz05
Copy link
Author

bikz05 commented Dec 20, 2014

We all have heard this clichéd line trillions of time right from the time of high school.

Derivative is rate of change.

Similarly, Image derivative represents the rate of the change in the intensity values of an image. But wait, we learnt that the derivatives can only be found out of functions that are continuous and images are discrete signals. So, we approximate this by using Derivatives Masks.
The various kinds of derivative masks that are available are -

  1. Backward Difference [-1, 1]
  2. Forward Difference [1, -1]
  3. Central Difference [-1, 0, 1]

Images are 2D, we need to find 2 partial derivatives. From the 2 partial derivatives, we get the magnitude and the direction of vector.

Related

  1. Youtube Lecture by Dr. Mubarak Shah

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment