Convert registered atlas to tiff stack or tiff sequence

[DJESTRIN]

Hi,
Thanks for the great code.

I was wondering if anyone might have a suggestion for converting the outputs of map_nonuniform_multiscale_v02_mouse_gauss_newton.m to a tiff stack instead of neuroglancer's precomputed format.

Currently, I am taking the data from downloop_1_labels_to_target_highres.img and converting it to a numpy array via SimpleITK. Then, I am saving the numpy array as an image via cv2. Although this image stack looks pretty good, I am wondering whether it needs to be run through transform_points.m or transform_points.py. Is downloop_1_labels_to_target_highres.img the final registered atlas?

Also, although this atlas image stack contains the same number of z planes as my target image stack, the image dimensions are much smaller (atlas=921x920, target=7210x7059). I am currently thinking of resampling my atlas image stack to match the dimensions of my target image, however, I would appreciate any advice.

Thank you in advance for your help and time!

[tathey1]

You are correct that downloop_1_labels_to_target_highres.img is atlas labels that were transformed to the shape of the target. For me, this file is also at a lower resolution than the target data. It should be (approximately, due to rounding) whichever multiple of the target resolution is above 10microns - which is the resolution of the atlas. It seems that for you (as for me), this is a factor of 8.

What you are describing is also how I would approach the conversion into tif stacks. As long as you can find a package to read the img file, and write tifs. I never have read img's before in python but I have written tifs, usually with skimage.io.imwrite