Releases: OCR-D/ocrd_olena
Releases · OCR-D/ocrd_olena
v1.2.0
Changed:
- use
sauvola-ms-split
as default impl
(method)
- default to
win-size=0
which now equals odd DPI value of image
- set
pcGtsId
in PAGE-XML, ht @mikegerber, #63
- images are written to the same
output_file_grp
as the PAGE-XML, OCR-D/core#505
Added:
- parameter
dpi
to override image meta-data DPI
v1.1.10
Changed:
- Use
ocrd log
for logging
- Implement
--overwrite
v1.1.9
- Check whether ImageMagick/GraphicsMagick and Boost are installed before compilation, #53
v1.1.8
Fixed:
- Use explicit
-format
in identify
call, #57
v1.1.7
Fixed:
- Use install-time Python binary when needed
Changed:
- Ignore
Page/PrintSpace
(as opposed to Page/Border
)
v1.1.6
Fixed:
- Quote all arguments in
scribo-cli
to allow file names with whitespace
- Pass on different
k
values at each scale for sauvola-ms*
Changed:
- Skip
AlternativeImage/@comments=binarized
in PAGE input
- Update
repo/olena
to upstream master
- Update
repo/assets
to upstream master
v1.1.5
Fixed:
- ignore/pass
file://
prefixes in absolute pathnames
- disable Qt when building Olena
v1.1.4
Changed:
- Update
repo/olena
to upstream master
- Update
repo/assets
to upstream master
Fixed:
- ignore/pass
file://
prefixes in absolute pathnames
- compensate for different
k
defaults for singh/niblack impl
- pass on correct
k
values for sauvola-ms-{fg,split} impl
v1.1.3
Changed:
- Update
repo/olena
to most recent upstream
make build-olena
configured not to build documentation
v1.1.2
1.1.2
Changed:
- Use bash from env instead of fixed shebang
- Ensure minimum bash version