implementing gabor filter on FPGA

hi~~everyone,i'm a newbie here. i want to implement the gabor filter on FPGA, will it be a proper way to implement?
what difficult will i meet? any suggestions will be appriciated......



Aaronbo

Aaronbo said:

hi~~everyone,i'm a newbie here. i want to implement the gabor filter on FPGA, will it be a proper way to implement?
what difficult will i meet? any suggestions will be appriciated......



Aaronbo

Click to expand...

Hi, I have to implement a Gabor filter in FPGA. Could you implement it? any advice I will appreciated. Greetings.

Hi guys,
As already u have completed working gaborfilter.If u don't mind can u guys post the mat lab code for gaborfilter as i am doing my project on fingerprint images comparison.
Thanks in Advance..

Regards,
GANESH

The below wiki link has some Matlab code:
Gabor filter - Wikipedia, the free encyclopedia

Not sure whether its the complete code.

Possible difficulties during the VHDL implementation are:
1)Fixed point operation handling.
2) Calculating sine and cosine values, exponential values.
3)Large memory needed for the image,synthesis may never finish if you have to do so many operations on the different pixels of the image.

Hi vipinlal,
I am Thankful for your reply.I have one doubt about interridegedistance parameter that will be used for false minutiae removal in fingerptin images comparison.In one document i found that it is calaculated using the procedure given here,we first calculate the inter ridge distance D which is the average distance between two neighboring ridges. For this scan each row to calculate the inter ridge distance using the formula:
Inter ridge distance = sum all pixels with values 1/row length.
Finally an averaged value over all rows gives D. .for the above process we are considering the image obtained after Thinning operation.Can you give the matlab code for it.otherwise i have written some code can you please check it once....
BElow is the code which i have written::
clc;
clear all;
close all;
y=ones(3);
y=[1 1 1; 1 1 1; 0 0 0 ];
[m,n]=size;
avgrowsum=0;
finalavgrowsum=0;
k=0;
for r=1:m
rowsum=0;
for c=1:n
if (y(r,c)>0)
rowsum=rowsum+y(r,c);
disp(rowsum)
avgrowsum = (m/rowsum);
disp(avgrowsum)
end

end


if (avgrowsum > 0)
finalavgrowsum=avgrowsum+finalavgrowsum;
k=k+1;
disp(finalavgrowsum)
end

end
disp(finalavgrowsum)
disp(k)
ffinalavgrowsum= (finalavgrowsum/k);
disp(ffinalavgrowsum)

Thanks in advance.

Regards,
GANESH