My hemispherical images had a lot of 'ground' rather than 'sky' which was leading to errors in calculating the thresholds. So, I got the polar projections and cropped them to eliminate the non-sky parts. I then ran Cano.py in pick-up mode and got the following error:
Traceback (most recent call last):
File "/Users/bonniewaring/Desktop/Cano.py", line 388, in main
imgpath_out, lai_out = process_image_batch(
^^^^^^^^^^^^^^^^^^^^
File "/Users/bonniewaring/Desktop/Cano.py", line 237, in process_image_batch
outlist = [process_image_single(x, threshold=threshold, slicepoint=slicepoint, save_files=save_files, mode=mode, outpath=outpath, fileext=fileext, quality=quality) for x in tqdm(imagepath_list, leave=False)]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/bonniewaring/Desktop/Cano.py", line 237, in <listcomp>
outlist = [process_image_single(x, threshold=threshold, slicepoint=slicepoint, save_files=save_files, mode=mode, outpath=outpath, fileext=fileext, quality=quality) for x in tqdm(imagepath_list, leave=False)]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/bonniewaring/Desktop/Cano.py", line 217, in process_image_single
thresh, lai = threshold_and_lai(polar, threshold)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/bonniewaring/Desktop/Cano.py", line 171, in threshold_and_lai
gap_fractions = calc_gap_fractions(thresh, (cx, cy, cr))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/bonniewaring/Desktop/Cano.py", line 122, in calc_gap_fractions
gap_fractions[i-1] = gap_fractions[i-1] + im_segment[int(y[j-1]) - 1, int(x[j-1]) - 1]
~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
IndexError: index 1130 is out of bounds for axis 1 with size 1129
I have attached an example of the original polar image and the cropped one. I also tried the no_slice option, wondering if it had to do with the dimensions of the image, but no dice!
From user:
My hemispherical images had a lot of 'ground' rather than 'sky' which was leading to errors in calculating the thresholds. So, I got the polar projections and cropped them to eliminate the non-sky parts. I then ran Cano.py in pick-up mode and got the following error:
I have attached an example of the original polar image and the cropped one. I also tried the no_slice option, wondering if it had to do with the dimensions of the image, but no dice!