llocarnini
2c39ffbcdd
changed kernel and iteration for better text removal
2022-04-27 11:12:23 +02:00
llocarnini
19fe6965fb
added line in display so the visual logger doesn't open too many plots
...
changes to fig_detection_with_layout.py so tables are getting parsed as well
reusage of adding external contour in table_parsing.py
2022-04-26 11:19:27 +02:00
llocarnini
17f5b22443
Merge branch 'master' of ssh://git.iqser.com:2222/rr/cv-analysis into fig-detection-scanned-pdfs
...
Conflicts:
cv_analysis/figure_detection.py
cv_analysis/layout_parsing.py
cv_analysis/table_parsing.py
scripts/annotate.py
2022-04-22 10:24:09 +02:00
llocarnini
11a2465789
few corrections for including smaller figures
2022-04-22 10:12:28 +02:00
Isaac Riley
88bb8dbddf
added visual logger for development
2022-04-21 15:10:35 +02:00
Isaac Riley
0ea556a7e0
slightly refactored table parsing and deleted unneeded file
2022-04-21 09:17:12 +02:00
llocarnini
3669b6b341
fig_detection_with_layout.py: approach to label the content of a page through layout detection, table parsing for detected tables needs to be added and overall codes needs to be reviewed
...
layout_parsing.py added condition so fig_detection_with_layout.py works
table_parsing.py uncommented line for better table parsing
text.py changed kernel sizes
2022-04-20 09:43:30 +02:00
llocarnini
420e484896
the thresholds deciding weather a countour is likely a primary text structure can be set better, as text structures are not always removed. this leads to over detection of figures
2022-04-12 16:48:29 +02:00
Isaac Riley
8730b34018
change name from vidocp to cv-analysis
2022-03-23 13:46:57 +01:00
