am I calculate the gradients in point (i,j) correctly? Core point detection angorithm

Hello:)

I'm writing code of algorithm included in PDF i foune here: http://www.google.pl/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCMQFjAA&url=http%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fdownload%3Fdoi%3D10.1.1.106.6655%26rep%3Drep1%26type%3Dpdf&ei=OX2KUJDcC-3P4QSl-ICQAQ&usg=AFQjCNFW9awru-5mL6Mz0IqYgvGiullxJQ&sig2=AODiEjZR3CkR-6WMsoryLA

I'm in point 2.2 and i have problem that my program returnd me NaN when calculating gradient in specified point (i,j). Am I doing it right?

• [FX1, FY1]=gradient(O); %O is loaded image converted to double
• delta_x=interp2(FX1,u,v); %u,v are indexes like in eqations (4) and (5)
• delta_y=interp2(FY1,u,v);

I don't know why but whed indexes are inside matrix FX/Y1 interp2 returns me NaN like indexes would be out of range...

My program works from eq (11) to (4). Simple: 1. divide image in blocks 2. find center of block - (i,j) 3. calculate (11) of i,j 4. need to calculate (10) and (9) so calculate it for i and j and so on

I could add a zip file with my code but i don't know if you don't have problems with understanding polish names of variables.

I hope You can help me with my problem.

P.S. I search forum and find some topics about Core point detection but it didnt helped me. I found cool working program too but i'd rather write my own than buy it.

Cheers:)

[Sean de Wolski]

Perhaps you could post the data and a relevant line of code, i.e. call to interp2?