My Kwiki

Here is the demo code for BorderDetection. Note that this is for a Blackfin EZ-KIT, not the Handy board.

Amount the different information extraction algortihms that can be of interest on image processing, we found those related to the extraction of data that be interpretated as information.

An example of such an information can be related to the object's border. Once a color filter is applied, the new image is providing a map of regiones where each one correspont to a possible object's element.

Mission

INCLUDE IMAGES

In this Lab we like you to write a program that can used a color filtered image and some 2D figure models as INPUT and a border extraction map that underlines the corresponding 2D models.

I) Sobel border detector

Note:

you should use the bigger granularity as possible for the color image processing. For the example in Lab 1 use jump=1.

Starter Code

The main function for the Sobel filters the next:

#define SOBEL_OUTPUT_MAXIMUM 240

#define SOBEL_OUTPUT_SCALE 2 . . .

for( y = 0; y < height; y++ )

{

for( x = 0; x < width; x ++) 

 {

  Sum_X = Sobel_In[y-1][x-1] + 2 * Sobel_In[y][x-1] +    

          Sobel_In[y+1][x-1] -(Sobel_In[y-1][x+1] + 2 *

          Sobel_In[y][x+1] + Sobel_In[y+1][x+1]);

  Sum_Y = Sobel_In[y-1][x-1] + 2 * Sobel_In[y-1][x] + 

          Sobel_In[y-1][x+1] - (Sobel_In[y+1][x-1] + 2 * 

          Sobel_In[y+1][x] + Sobel_In[y+1][x+1]);

  Sum = (abs(Sum_X) + abs(Sum_Y));

 }

}

An initial code for computing Sobel after have filtered a color is available here.

II) Canny Edge detector

Introduce the code Blake adapted to do Canny

Here you can find an introduction to the Canny Edge Detector

III) Susan Edge Detector

Introduce the code Ramon make in C and CHDL

Here you can find an introduction of the SUSAN Edge detector

Links & References

Here you have some Wiki links to Sobel Filter and 2D homogenous Trasnformations