I am getting an error as "Check for incorrect argument data type or missing argument in call to function 'exp'." can anyone help me out?

a=imread("house.jpeg");
outimage=a;
k=.001;
iterations=30;
lambda=.25;
[m,n]=size(a);
rowc=[1:m];
colc= [1:n];
rown= [1  1:m-1];
cole= [1  1:n-1];
rows= [1:m   m];
colw=  [1:n  m];

for i = 1: iterations
    deltaN= outimage(rown,colc)- outimage(rowc,colc);
    deltaE= outimage(rowc,cole)- outimage(rowc,colc);
    
    fluxN=  deltaN * exp(-(1/k)*abs(deltaN));
    fluxE=  deltaE * exp(-(1/k)*abs(deltaE));
    
    outimage= outimage + lambda*(fluxN-fluxN(rows,rowc) + fluxE- fluxE(rowc,colw));
    
end

Answer 1

[Steven Lord]

Most likely your image data is of an integer type. The exp function is not defined for the integer types.

one = int8(1);
y = exp(one)

Check for incorrect argument data type or missing argument in call to function 'exp'.

There are few integer values that when used as the input to exp result in an integer valued result.

Convert your data to double before calling exp on it.

[yanqi liu]

deltaN= double(outimage(rown,colc)- outimage(rowc,colc));

deltaE= double(outimage(rowc,cole)- outimage(rowc,colc));

fluxN= deltaN * exp(-(1/k)*abs(deltaN));

fluxE= deltaE * exp(-(1/k)*abs(deltaE));

but

outimage= outimage + lambda*(fluxN-fluxN(rows,rowc) + fluxE- fluxE(rowc,colw));

may be dimension error

Answer 2

[Image Analyst]

Here are some improvements but I don't really know what the vectors are for and what the deltaN and deltaE represent. Are they the vertical and horizontal gradients like you'd get from imgradientxy()?

grayImage = imread("peppers.png");  % Read in image.
imshow(grayImage, []);
% Convert to double so we can do subtractions and get negative numbers.
grayImage = double(grayImage);
[rows, columns, numberOfColorChannels] = size(grayImage);
if numberOfColorChannels == 3
    % Convert to gray scale
    grayImage = rgb2gray(grayImage);
end
outimage = grayImage;
k = .001;
iterations = 30;
lambda = .25;
% Define some vectors.  I have no idea what these are for.
rowc = [1 : rows];
colc = [1 : columns];
rown = [1, 1:rows-1];
cole = [1, 1:columns-1];
rows = [1:rows, rows];
colw =  [1:columns, rows];
for k = 1: iterations
    fprintf('Starting iteration %d of %d.\n', k, iterations)
    % Probably need to index rown, colc, etc. below.
    deltaN = outimage(rown,colc)- outimage(rowc,colc);
    deltaE = outimage(rowc,cole)- outimage(rowc,colc);
    
    fluxN = deltaN * exp(-(1/k)*abs(deltaN));
    fluxE = deltaE * exp(-(1/k)*abs(deltaE));
    
    outimage= outimage + lambda*(fluxN-fluxN(rows,rowc) + fluxE- fluxE(rowc,colw));
end