Feature Detection

This function can be used for corner and edge detection for any RGB or grayscale image.

Description

Feature detection is an importatnt aspect of any image or video processing application. The given code can be used for detecting cornera and edges in a RGB or grayscale image.It is based upon the famous paper on the topic titled "A Combined Corner and Edge Detector" by Harris and stephens.

• Required Products: Image Processing Toolbox
• MATLAB release: MATLAB 7.7 (R2008b)

[Tolga Birdal]

Hey Gagan,

Good submission. Simple to read.

Here are my quick recommendations:

1) For color images, the tensor is specified in a different manner. Precisely as,

T=[

Rx.^2 + Gx.^2 + Bx.^2 RxRy + GxGy + BxBy
RxRy + GxGy + BxBy Ry.^2 + Gy.^2 + By.^2

]

(Of course don't forget the channels are gaussian smoothed).

Here are 2 references:
http://lear.inrialpes.fr/people/vandeweijer/papers/cic04.pdf

ftp://ftp.diku.dk/diku/image/publications/riber.000413/Gradient_norm_of_multichannel_images.html

2) For such a tensor, there is analytical solution to the eigenvalues and eigenvectors. Probably you could also make use of that (for speed and accuracy). The references above include that as well.

Congrats,
Bests,

[GAGAN]

Thank you Tolga.