Consider NSCT decompose?

[BenderBlog]

Refrence. https://www.mathworks.com/matlabcentral/fileexchange/10049-nonsubsampled-contourlet-toolbox

A. L. da Cunha, J. Zhou, M. N. Do, "The Nonsubsampled Contourlet Transform: Theory, Design, and Applications," IEEE Transactions on Image Processing, to appear 2005.

In short, my lab proj uses this, and that's the only reason we used MATLAB... It is used in SAR image fusion and CT picture (I assume...)

Great work, thanks.

[wwhenxuan]

Refrence. https://www.mathworks.com/matlabcentral/fileexchange/10049-nonsubsampled-contourlet-toolbox

A. L. da Cunha, J. Zhou, M. N. Do, "The Nonsubsampled Contourlet Transform: Theory, Design, and Applications," IEEE Transactions on Image Processing, to appear 2005.

In short, my lab proj uses this, and that's the only reason we used MATLAB... It is used in SAR image fusion and CT picture (I assume...)

Great work, thanks.

Obviously, NSCT is a very popular image processing method, which introduces more directionality and local detail capture capabilities based on traditional transform methods (such as fast Fourier transform and wavelet transform), making it more effective in processing complex features such as edges and textures of images or signals.
I have used this algorithm to process hyperspectral images in my computer vision and image understanding course, and I like the nonsubsampled pyramid decomposition method.

Thanks for your suggestion, we will certainly incorporate this algorithm into PySDKit. But since the project is developed by only two people, the speed is slow. We will first focus on the decomposition algorithm of one-dimensional signals. After reproducing almost all one-dimensional algorithms, we will further reproduce the decomposition and reconstruction algorithms of images similar to NSCT.

Although Scikit-image does not use the NSCT algorithm, it also has some other interesting image processing that can be used for the SAR image fusion and CT picture you described. I hope it will be helpful to you @BenderBlog .

Best regards 🥰🥰🥰

https://scikit-image.org/docs/stable/auto_examples/index.html