qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2026-04-30 19:22:03 +02:00
Code Issues Projects Releases Wiki Activity

get_marginals(): improve peak point threshold criterion…

Browse source 
in search of valid peaks (gaps between text columns),
- drop absolute values for minimum gap depth
  (likely crafted for some fixed resolution examples)
- instead, use criterion relative to maximum column depth
  and page height (trying to loosely approximate the prior
  constants, albeit somewhat more permissive)
This commit is contained in:
Robert Sachunsky 2026-04-25 02:23:16 +02:00
parent 4bdea39c98
commit 6d55d0b87b
1 changed files with 1 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

11
src/eynollah/utils/marginals.py
View file

@ -37,16 +37,7 @@ def get_marginals(text_with_lines, text_regions, num_col, slope_deskew, kernel=N
text_with_lines_y_eroded = text_with_lines_eroded.sum(axis=0) text_with_lines_y_eroded = text_with_lines_eroded.sum(axis=0)
max_textline_thickness_percent = 100. * text_with_lines_y_eroded.max() / height max_textline_thickness_percent = 100. * text_with_lines_y_eroded.max() / height
min_textline_thickness = max_textline_thickness_percent / 100. * height / 20.
# rs: min_textline_thickness seems to be calibrated for some fixed resolution,
# but text_with_lines varies in size!
if max_textline_thickness_percent < 30:
min_textline_thickness = 8
elif max_textline_thickness_percent < 50:
min_textline_thickness = 20
else:
min_textline_thickness = 45
# min_textline_thickness = max_textline_thickness_percent / 100. * height / 20.
# plt.figure() # plt.figure()
# ax1 = plt.subplot(2, 1, 1, title="text_with_lines_eroded") # ax1 = plt.subplot(2, 1, 1, title="text_with_lines_eroded")