principal component analysis pca

Hello everybody

I have infrared images (144) for one day for a specific specimen. Anyhow, I transferred these images into 3D matrix that has thousands of signals. I want to subject the PCA approach so I can change the frequencies to get best results for these images.

I uploaded only one signal, please any help on doing that on my signals.

Thanks

Answer 1

[Neelanshu]

Hello Mohaneed,

I understand that you want to use PCA approach on the dataset of images. The signal shared can't be used for PCA as PCA cant be performed on a single observation.

To apply PCA to infrared images stored in a 3D matrix :

1. You need to reshape the data such that each row represents the flatenned image.
2. Then you can use the 'pca' function to perform pca on the data.

% Reshape the 3D matrix into a 2D matrix
[height, width, num_images] = size(images);
reshaped_images = reshape(images, height * width, num_images)';

% Perform PCA on the reshaped data
[coeff, score, latent, tsquared, explained] = pca(reshaped_images);

You may refer to the following documentation to learn more about the 'pca' function :

https://www.mathworks.com/help/stats/pca.html

Hope this helps!