Commit Graph

339 Commits

Author SHA1 Message Date
sz 74cad11d24 Not the prettiest CMakeLists.txt, but it builds...
will want to bring over some options from the zstd Makefile (or decide
this is a terrible idea and bail out)
2020-08-21 23:45:33 -05:00
sz 5d1fd05ebd Trying to include a minimal-ish zstd impl 2020-08-21 23:41:36 -05:00
sz 9d417b3094 wip zstd streams
We'll see how well this works before committing to it
2020-08-21 23:25:05 -05:00
sz3 4f0277cf14
Merge pull request #28 from sz3/bugfix-small-file
Bugfix: don't crash on undersized images
2020-08-21 23:20:06 -05:00
sz cb00ed1f91 how on earth... 2020-08-21 21:24:07 -05:00
sz c90a13b6d4 Use the config value for the constant 2020-08-21 19:09:56 -05:00
sz 1346d2623e The dash is kind of annoying to deal with, make it an underbar 2020-08-20 23:52:50 -05:00
sz a28d3f39bf Ignore local changes to the samples submodule. 2020-08-20 23:43:56 -05:00
sz 1c8f90b613 If you ask the CimbReader to decode a file that's too small, it blows up
so make it check!
2020-08-20 23:30:42 -05:00
sz3 7be1cf873f
Merge pull request #27 from sz3/samples-submodule
Add samples submodule
2020-08-20 20:18:34 -05:00
sz 2998ee3007 Add samples submodule 2020-08-20 20:16:26 -05:00
sz3 984ab3fc7e
Merge pull request #26 from sz3/more-docs-edits
More docs edits
2020-08-20 20:03:56 -05:00
sz 986e976d51 One more batch of docs edits, now with less scare quotes 2020-08-20 20:03:28 -05:00
sz3 bdd58c4bec
Merge pull request #25 from sz3/install
Add functional `make install` target, docs edits
2020-08-20 19:44:10 -05:00
sz 88662cfe79 make install 2020-08-20 18:23:07 -05:00
sz3 08a6ffd7a0
Merge pull request #24 from sz3/more-licensing-spam
More license spam
2020-08-20 03:34:09 -05:00
sz 68b2a9a686 Missed these 2020-08-20 03:32:37 -05:00
sz3 f3f936eca1
Merge pull request #23 from sz3/docs
More docs
2020-08-20 02:14:49 -05:00
sz d49e3d49f7 Big docs edits oh baby 2020-08-20 02:14:04 -05:00
sz 5afa713830 ordering 2020-08-19 07:14:31 -05:00
sz 9a459ab691 Add CFC link? 2020-08-19 06:01:15 -05:00
sz fb6d9c662b Docsing 2020-08-19 03:53:46 -05:00
sz3 fa78e3c48d
Merge pull request #22 from sz3/license-spam
MPL2.0 comment spam
2020-08-19 02:37:21 -05:00
sz f91a2b064b MPL2.0 comment spam
Some of these (test.cpp) are absurd, but it's easier to get all of them!
2020-08-19 02:36:10 -05:00
sz3 46fbbc7154
Merge pull request #21 from sz3/misc-fix
Fixes: scan logic
2020-08-19 02:29:59 -05:00
sz 460d288216 Scanner: Merging the results of the confirm step into the results
This seems to bring the depth-first and breadth-first implementations to
nearly identical results, which is compelling.

Overall success %s are very similar to how things were before --
evidently, "close enough" tends to also be good enough, usually.
2020-08-19 01:36:58 -05:00
sz 6a32df5cae Remove old CLAHE code, it's not needed 2020-08-18 23:42:44 -05:00
sz e1906b1a20 Forgot about catch! 2020-08-18 21:09:11 -05:00
sz 420639b1f6 Fill out concurrent_fountain_decoder_sink with more interface methods. 2020-08-14 02:52:05 -05:00
sz b689624a77 Validate confirms twice (once for x, once for y) 2020-08-14 02:35:51 -05:00
sz e945ee6959 Check rows/cols for confirms 2020-08-14 01:29:15 -05:00
sz 27128f2bd9 Fix link to ABOUT.md 2020-08-14 00:50:52 -05:00
sz3 18218da865
Merge pull request #20 from sz3/readme
Readme
2020-08-13 23:49:45 -05:00
sz3 9c44e218be
Merge pull request #19 from sz3/command-line
Update/improve(?) command line options
2020-08-13 23:47:27 -05:00
sz3 e38cfe79f9
Merge pull request #18 from sz3/auto-rotate
Change scanner pattern to allow auto-rotation
2020-08-13 23:43:35 -05:00
sz bd073e2c02 we do it 2020-08-13 23:18:42 -05:00
sz d2cf2fae5e workin on a readme 2020-08-12 20:53:58 -05:00
sz3 6f4cf72fce
Merge pull request #17 from sz3/more-colors
Support for 8 colors??? Maybe???
2020-08-12 00:14:32 -05:00
sz 236e8ea156 Update more tests 2020-08-12 00:11:04 -05:00
sz 7b671ebe40 Updating sample image paths, and tests 2020-08-11 22:32:26 -05:00
sz 7352dbd2b2 Put the scanner threshold back at 0 for now
I've been using this to play with how sensitive we want to be when there
is ghosting on part of a cimbar image (e.g. LED screen refresh). Higher
(0) == more sensitive == don't try to decode. The lower values give us
more extracted images to look at, but on average they might be more
trouble than they're worth...
2020-08-11 21:52:09 -05:00
sz 8b397ff815 Update/improve(?) command line options?
cxxopts actually does provide some decent flexibility, so try using it.
2020-08-09 23:03:44 -05:00
sz 5b5f1e58ee Don't try to cast things to this 2020-08-09 22:44:25 -05:00
sz 4cc3f0309a Adjust scanner threshold value?
we blur the image before the adaptive threshold, so it makes some degree
of sense that it might be an improvement if we require a stronger signal
to consider a pixel "on".

-20 is somewhat arbitrary -- and this will break a lot of tests. But I
do think some negative value is going to be the right call.
2020-08-08 02:48:58 -05:00
sz 4645d21bd9 Make a better guess for the 4th corner of the image.
We don't need to be 100% on the money, but we do need to be "close
enough".
2020-08-08 02:45:17 -05:00
sz 66ff1df7ba Oops 2020-08-07 22:55:16 -05:00
sz c03b0a80d7 WIP: handle skewed images better 2020-08-07 21:48:38 -05:00
sz 1786ba68fa Fix/update some tests 2020-08-07 03:00:13 -05:00
sz 477dd93b09 Update a bunch of tests with new sample images 2020-08-07 01:02:51 -05:00
sz 4d27600f0d Update command line help to match reality 2020-08-06 23:27:16 -05:00