To address the subjectivity in visual color inspection of product packaging bags, we can implement a MATLAB-based image processing solution. We have already been provided with three reference images, a target shade and its acceptable high and low color variants, against which a sample bag is compared.
We can formulate an approach which evaluates the sample by analysing the mean RGB values within a central region of interest (ROI) in the image. It checks whether the sample's RGB values lie within the range defined by the high and low reference shades. If so, the sample is accepted; otherwise, it is rejected. You can leverage ‘mean’ function of MATLAB to compute mean RGB values. Refer to the documentation of ‘mean’ function to know more about the syntax and its usage: https://in.mathworks.com/help/matlab/ref/double.mean.html
Below is a code snippet demonstrating the above approach:
% Compute mean RGB values
mean_low = mean(reshape(low_roi, [], 3));
mean_high = mean(reshape(high_roi, [], 3));
mean_target = mean(reshape(target_roi, [], 3));
mean_sample = mean(reshape(sample_roi, [], 3));
% Check if sample RGB is within acceptable bounds
R_in = mean_sample(1) >= min(mean_low(1), mean_high(1)) && mean_sample(1) <= max(mean_low(1), mean_high(1));
G_in = mean_sample(2) >= min(mean_low(2), mean_high(2)) && mean_sample(2) <= max(mean_low(2), mean_high(2));
B_in = mean_sample(3) >= min(mean_low(3), mean_high(3)) && mean_sample(3) <= max(mean_low(3), mean_high(3));
if R_in && G_in && B_in
result = 'ACCEPTED';
else
result = 'REJECTED';
end
I hope it helps.
