preproc/lib/hocr, branch v0.1.0 Miscellaneous OCR related utilities and packages Continue separating the repository; remove all but preproc, and move integralimg package under it 2019-10-08T14:37:07+00:00 Nick White git@njw.name 2019-10-08T14:37:07+00:00 787d63fc5d13c6250bd33da5a8e1eadbe86188cd 






Handle no words found error in a better way so any page that is actually 0 confidence is recognised
2019-09-05T21:42:30+00:00

Nick White
git@njw.name

2019-09-05T21:24:37+00:00

561d8461cbe19316762489cd7b04f95b9014bcda












Don't abort analysis if we encounter a hocr with no words, just skip it
2019-09-05T21:20:35+00:00

Nick White
git@njw.name

2019-09-05T21:20:35+00:00

60a198f7ee5843a0f77b6dfb845c3b0413e83705












Return an error if page average calculation cant be done with hocr
2019-05-15T15:03:29+00:00

Nick White
git@njw.name

2019-05-15T15:03:29+00:00

d7f07893d08d9c29f46e50c4f779b0e701f411e4












Rewrite pgconf to be more accurate by measuring average word confidence rather than average line confidence
2019-05-14T17:02:34+00:00

Nick White
git@njw.name

2019-05-14T17:02:34+00:00

f49a8a74a8ef2c96cc2bbf34461a8387f7e324d8












Add pgconf tool that prints the overall confidence for a whole page of hocr
2019-05-14T10:20:33+00:00

Nick White
git@njw.name

2019-05-14T10:20:33+00:00

6b4e704befb7f82627d2c9a4e3f4e2971fdaf883












Better error handling with hocr lines
2019-03-26T15:21:43+00:00

Nick White
git@njw.name

2019-03-26T15:21:43+00:00

7d4cc9f7dd297c08e02447dda1f8aad9db0b0768












Generalise get text from hocr lines
2019-02-25T13:01:28+00:00

Nick White
git@njw.name

2019-02-25T13:01:28+00:00

cd1fb1c9f6e1384ac0add8904425e6f92b17a704












Add tool to extract plain text from hocr
2019-02-25T12:29:59+00:00

Nick White
git@njw.name

2019-02-25T12:09:06+00:00

3c4c5f7c202b7c54ca8d23e7bd7bff4a4bb696cc












gofmt
2019-01-25T17:41:52+00:00

Nick White
git@njw.name

2019-01-25T17:41:52+00:00

988658936cbcd8e92b35a66e1943bea0f9eaf3bc