how to remove clutters in Ground Penetrating Radar (GPR) data

7 views (last 30 days)
I'm trying filter a 3d data consisting of several 3d images. they contain data related to landmines detection. I've loaded the 3d data and selected one 3d image out of it and reduced to a 2d image. see below my code.
>> close all;
% Load a 3D GPR Image
I=load('sphere_clean.mat');
% Calculate Absolute Value of the 3D GPR Image
% Only Consider the Tx1-Rx1 for GPR
GPR= abs(I.T1R1);
% Obtain Size of the Image
[depth,x,y]=size(GPR);
% Create a 2D Array and Initialize with zero
E_yz=zeros(x,y);
E_xz=zeros(x,y);
b_size=10;
step=b_size-1;
map_yz= zeros(x-1,y-1,3);
%Define the window size
for i=1:x
% Select One Slice YZ
image=abs(GPR(:,i,:));
% Convert from 3D (GPR(depth,1,y)) to 2D Matrix
image=reshape(image,depth,y);
end
for i=1:x
% Select One Slice XZ
image2=abs(GPR(:,:,i));
% Convert from 3D (GPR(depth,1,y)) to 2D Matrix
image2=reshape(image2,depth,y);
end
>> figure,imagesc(image);colormap(gray);
GPR = double(image);
>> imshow(image2)
>> imshow(image)
>> BW1 = edge(image,'sobel');
BW2 = edge(image,'canny');
figure;
imshowpair(BW1,BW2,'montage')
title('Sobel Filter Canny Filter');
>> L = medfilt2(image,[3 3]);
figure, imshow(L)
>>
my problem is this gpr read images come with clutter and noise due to real signal mixed with the reflected signals (clutters) from the soil and nature of the equipment (noise). get rid of noise is not the problem, I've seen so many noise filters to pick from like median filters, Gaussian etc. my problem is removing clutters. kalman filters or wavelets, matching to polynomials were recommended but i dont know how to apply this to images and I'm not even sure if this will effectively remove clutters from gpr loaded data. how can i go about this? any recommendation will be highly appreciated.

Answers (1)

Image Analyst
Image Analyst on 7 Oct 2015

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!