graphing an MTF from an image

Hello everyone,

If I have a captured image of a thermal imager with halfmoon target and I want to find the MTF of that image.

What would be the right code to that.

I tried graphing it but i didnt succeed.

Let's consider the below graph as the edge that I want to find the MTF of.

What I want to do is as follow:

1- To find the size (resolution of that image)

2- To find the edge response function (ESF)

3- To find the line spread function (LSF) from the edge response.

4- To graph the MTF from the line spread function (LSF)

It should seems like the below picture

Your help will be so much appreciated.

Thank you in advance.

Answer 1

[Abhishek Chakram]

Hi Abdulrahman Alsalmi,

To calculate the Modulation Transfer Function (MTF) from an image of a thermal imager with a half-moon target, you need to follow a series of steps that involve image processing techniques. Below is a high-level outline of the MATLAB code for the same:

% Assuming 'thermalImage' is your loaded thermal image with a half-moon target

% Step 1: Find the size (resolution) of the image
[rows, columns, ~] = size(thermalImage);

% Step 2: Find the edge response function (ESF)
% This usually involves finding the edge of the half-moon target and then
% extracting the pixel values across that edge.

% Convert the image to grayscale if it is not already
grayImage = rgb2gray(thermalImage);

% Find edges using an edge detection algorithm like 'Sobel'
edges = edge(grayImage, 'sobel');

% You may need to process 'edges' to find the actual edge of the half-moon
% For simplicity, let's assume 'edgeRow' is the row index of the half-moon edge
edgeRow = findEdgeRow(edges); % You need to implement this function

% Extract the edge response function (ESF)
ESF = double(grayImage(edgeRow, :));

% Step 3: Find the line spread function (LSF) from the ESF
% The LSF is the derivative of the ESF
LSF = diff(ESF);

% Step 4: Graph the MTF from the LSF
% The MTF is the magnitude of the Fourier Transform of the LSF
MTF = abs(fft(LSF));

% Normalize the MTF by its maximum value
MTF = MTF / max(MTF);

% Create a frequency axis for plotting
frequencyAxis = linspace(0, columns/2, numel(MTF)/2);

% Plot the MTF
figure;
plot(frequencyAxis, MTF(1:numel(MTF)/2));
xlabel('Spatial Frequency (cycles per pixel)');
ylabel('MTF');
title('Modulation Transfer Function');

You can refer to the following documentation to know more about the functions used:

• rgb2gray: https://in.mathworks.com/help/matlab/ref/rgb2gray.html
• edge: https://in.mathworks.com/help/images/ref/edge.html

Best Regards,

Abhishek Chakram