How to remove all background noise due to inappropraite lighting from my pre-processed image and get better edge defination??

I have done some process on image and got the edge defined but unable to remove background noise.

I have used laplacian, sobel operator, average filter and power law for getting better edge defination

Are there any other methods for better edge defination and any way of removing background noise???

clc,clear
I=imread('warped_image.jpg');
aa=rgb2gray(I);
a=double(aa);

% Kernel for laplacian
kernel = [0,2,0; 2,-8,2; 0,2,0];
b = conv2(double(a), kernel, 'same');
c=b-a;
d=edge(a,'sobel');

% Kernel for smmothening by ave filter of 5*5
kernel1 = [1/9,1/9,1/9; 1/9,1/9,1/9; 1/9,1/9,1/9];
e = conv2(double(aa), kernel1, 'same');
f=c.*e;
g=a+f;

% transform law
[m,n]=size(a);
gamma=5;
cc=1;

for i=1:m
    for j=1:n
        h(i,j)=cc*(g(i,j)^gamma);
    end
end

imshow(h)

Any suggestions will help??

Should I try removing my bacground while it is in RGB format??? and then do processing of image.

Answer 1

[J. Alex Lee]

try smoothing before you apply your edge detector, not after

same goes for your power law (gamma correction?)

alternatively/also try increasing the window size of your gradient/laplacian kernels