Commit Graph

  • 9c8decec5f Adapt Planta to recent Koo changes: setModelGroup() -> setRecordGroup() main Albert Cervera i Areny 2009-04-17 00:48:37 +0200
  • c79a7894ef Adapt Planta to recent Koo changes: ModelRecordGroup -> RecordGroup. Albert Cervera i Areny 2009-04-17 00:46:41 +0200
  • 0828afec76 Adapted to renamed 'ModelRecordGroup' -> 'RecordGroup' Albert Cervera i Areny 2009-03-25 02:02:48 +0100
  • 2d2910d49a Improvements in date and page number detection. Albert Cervera i Areny 2009-03-25 02:02:17 +0100
  • 642cf462e7 Call callback function only once per Tag. Added several words for invoice number. Albert Cervera i Areny 2009-03-25 02:01:29 +0100
  • 201109c913 Remove commented code in InvoiceRecognizer. Albert Cervera i Areny 2009-03-23 23:17:06 +0100
  • ccee97f718 Added an algorithm to detect fixed pitch fonts. Now the algorithm to format text (adding spaces) takes into account if font is fixedPitch. Detection seems to work reasonably well. Format text algorithm needs to take fixedPitch on a per bloc basis instead of using the same criteria for the same line. Albert Cervera i Areny 2009-03-23 23:14:58 +0100
  • 9cb1e8870a Remove a couple of print's. Albert Cervera i Areny 2009-03-23 23:14:33 +0100
  • 41224310da Several improvements in invoice recognition: Added new types, improved date recognition, improved performance. Albert Cervera i Areny 2009-03-23 16:09:11 +0100
  • 4119ec747c Important performance improvement: store images as BMP instead of PNG before processing them with external tools. A sample image took 13 seconds to be stored as PNG while BMP took less than a second. Albert Cervera i Areny 2009-03-16 23:45:12 +0100
  • 02241981b6 Improved Block class implementation. Make Ocr use it internally. First steps towards Block finding in documents. Albert Cervera i Areny 2009-03-16 23:18:56 +0100
  • 4cafa912ce Added Block, PdfReader, Range and TextPatterns. Albert Cervera i Areny 2009-03-14 18:11:20 +0100
  • 0eb944c512 - Fixed doxygen file. - Added invoice recognition module. Still missing block detection. Albert Cervera i Areny 2009-03-14 18:10:23 +0100
  • 2cbba682f0 Make region optional in textInRegion functions. Albert Cervera i Areny 2009-03-14 18:08:32 +0100
  • 49e5d3d40c Use new plugin loading mechanism introduced in Koo r508. Albert Cervera i Areny 2009-01-10 16:57:24 +0100
  • fe3f8c1cb8 Use Levenshtein module when available for performance reasons (it's 300 times faster). Albert Cervera i Areny 2008-12-30 20:25:59 +0100
  • f06e0db54c Removed item from TODO list because it's already implemented. Albert Cervera i Areny 2008-12-30 13:47:32 +0100
  • 03c316f47a Remove old commented line. Albert Cervera i Areny 2008-12-30 02:34:00 +0100
  • 7d8111e99c Implemented and fully working Ocr Cuneiform backend. Albert Cervera i Areny 2008-12-30 02:30:59 +0100
  • 53f96bbb8b Remove unnecessary 'return True'. Albert Cervera i Areny 2008-12-30 00:45:41 +0100
  • 021d5df12d More uppercase renaming. Albert Cervera i Areny 2008-12-30 00:44:53 +0100
  • c00ce1cabe Push all changes I missed a couple of commits ago. Albert Cervera i Areny 2008-12-29 01:55:57 +0100
  • 1f55cfd529 Added Koo plugin. Albert Cervera i Areny 2008-12-29 01:55:36 +0100
  • c2fb42ebbd Renamed: - Unwritable NaNScaN To NanScan. - Capitalized file names. Albert Cervera i Areny 2008-12-29 01:53:29 +0100
  • 2906e6392d Fix typo in auto_attach module description. Albert Cervera i Areny 2008-12-29 01:15:08 +0100
  • 8cf46ea005 Remove unnecessary constructor. Albert Cervera i Areny 2008-12-29 01:14:56 +0100
  • 6bd9b496e4 Added dataMatrix as recognizer in auto_attach module. Albert Cervera i Areny 2008-12-29 00:43:08 +0100
  • ec3b13fac2 Implemented DataMatrix analyzer. Other architecture enhancements involving generalization of analyzers to create new once more easily, improved spawn() function to allow it to be run in several simultaneous threads. Albert Cervera i Areny 2008-12-23 20:29:28 +0100
  • 327542e74d Implemented background processing in auto_attach module. Albert Cervera i Areny 2008-12-23 20:29:08 +0100
  • 100fc8b3bc Code cleanup and workflow improvements. - Removed '_id' postfix in relation fields. - Added relation with templatebox in document boxes. - Improved workflow with background alternative to scan process. - Background scanning process sends a request to the user when it's finished. Albert Cervera i Areny 2008-10-31 00:33:15 +0100
  • 67edd150b1 Do not add new line characters in the list line for formatedText. Albert Cervera i Areny 2008-10-29 16:49:51 +0100
  • 69e0e78d4e Renamed some files to fit with naming conventions (capitalization). Use new Koo API. OpenTemplateDialog not working yet but it could be Koo's fault so I just commit and I'll try to find the problem later. Added Deskew action too. Albert Cervera i Areny 2008-10-27 22:54:57 +0100
  • a4b8ccf707 Fixed Levenshtein distance for empty strings. Also create array more efficiently. Albert Cervera i Areny 2008-09-15 16:04:18 +0200
  • 907ccb94b2 New template finder by text now works. Not that it's perfect at all, it only works for most obvious cases, but it's a good start. Albert Cervera i Areny 2008-09-15 00:01:28 +0200
  • ea1541a331 Improve comment of 'region.isValid()' bug. Albert Cervera i Areny 2008-09-15 00:00:58 +0200
  • fa654c1837 Fix bug: using 'if region:' is not the same as 'if region != None' in this context. The first one checks if region isValid(). Albert Cervera i Areny 2008-09-14 23:59:13 +0200
  • 7cff77ef94 Fix bug in hamming distance: Difference in length between two strings must be multiplied by 2. Albert Cervera i Areny 2008-09-14 23:57:35 +0200
  • 9cc4af299d Show mouse position & findMatchingTemplateByText: - Show current mouse position in template designer. - New 'findMatchingTemplateByText()' still not working properly but most code is already there including the action to test it from template designer. Albert Cervera i Areny 2008-09-14 17:32:40 +0200
  • c0e881db5d Improve barcode displaying in template designer. Albert Cervera i Areny 2008-09-12 16:36:03 +0200
  • 1552729785 Show offset info when finding matching template. Albert Cervera i Areny 2008-09-12 16:26:59 +0200
  • bbb5784e08 Added Find Matching Template in template designer. This actions searches the best matching template using NaNScaN Recognier.findMatchingTemplate() function and loads it. Albert Cervera i Areny 2008-09-12 13:47:47 +0200
  • dd3b1a698c Change module description. Albert Cervera i Areny 2008-09-11 21:05:18 +0200
  • f425f17d73 Ease logging in and add login information in statusbar. Albert Cervera i Areny 2008-09-11 21:04:20 +0200
  • a5506050fa Added feature rect to template. Appropiate handling added to Planta template designer, barcode, ocr and recognizer as well as to the added auto_attach module. Storing and retrieval hasn't been tested though. Albert Cervera i Areny 2008-09-02 00:01:04 +0200
  • 8ac6afa164 Added Planta application. A template designer. This is the old 'playground' nano-client and thus not only depends on NaNScaN but in KTiny too. Albert Cervera i Areny 2008-09-01 16:42:41 +0200
  • e1be884d22 Added .gitignore file to repository. Albert Cervera i Areny 2008-09-01 16:13:14 +0200
  • ec5bbe011c Moved all source files to the new NaNScaN directory. Albert Cervera i Areny 2008-09-01 16:12:21 +0200
  • f4ba142626 Documented Ocr class. Updated doxygen header. Albert Cervera i Areny 2008-09-01 00:55:29 +0200
  • 63a380c442 Added doxgen documentation. Added docs in the Hamming class. Added a Translator class. Albert Cervera i Areny 2008-09-01 00:13:37 +0200
  • bffb921bf3 Added test-learn.py to test gamera learning probes. Albert Cervera i Areny 2008-08-31 23:20:32 +0200
  • 021030b26c Let test-scandialog be cross platform. Albert Cervera i Areny 2008-08-31 23:18:17 +0200
  • a94e41fe17 Added TODO file. Albert Cervera i Areny 2008-08-30 19:23:58 +0200
  • 48f4b0b627 Started some learning tests using gamera. Very sloooow by now. Albert Cervera i Areny 2008-08-30 19:23:09 +0200
  • 0b5450d982 Remove empty line. Albert Cervera i Areny 2008-08-30 19:22:25 +0200
  • 3e1433d5a4 Fix bug: rename variable. Albert Cervera i Areny 2008-08-30 19:21:40 +0200
  • 665dad1799 Write formated text in an output file in test-ocr.py Albert Cervera i Areny 2008-08-30 19:19:20 +0200
  • 97de6ad8bb Insert new Character() objects instead of rendering text directly in formatedText() function. This allows us to use the formated text algorithm in other places, not only for 'printing'. Albert Cervera i Areny 2008-08-30 19:16:29 +0200
  • bc533aae90 Added GPL header and fix a typo. Albert Cervera i Areny 2008-08-30 19:05:53 +0200
  • 99a09a4411 Added a loop on distances tests to measure speed. Albert Cervera i Areny 2008-08-30 19:05:17 +0200
  • d5daccecc8 Fix bug of (wrong var name). Albert Cervera i Areny 2008-08-30 18:59:28 +0200
  • 2b84e27b76 Added hamming distance implementation. Albert Cervera i Areny 2008-08-05 10:52:40 +0200
  • 1a3dc76ccf Added levenshtein distance implementation. Albert Cervera i Areny 2008-08-05 10:37:59 +0200
  • 8cce5313fe Merge branch 'master' of ssh://albert@192.168.1.2/home/albert/d/git/nanscan unknown 2008-06-22 12:45:13 +0200
  • ce112dfd02 Added scandialog. unknown 2008-06-22 12:42:41 +0200
  • 7f713765d0 - Added TemporaryFile class which will help handling temporary files for external commands. - Scan functions now expect a QImage() instead of a file name which is more appropiate and API friendly. - Reestructured deskew() function into slope(), deskewOnce() and deskew(). Albert Cervera i Areny 2008-06-19 20:45:22 +0200
  • efb1d33258 Added comments in text formating function (explain algorithm). Changed convertToGray for convertToGrayScale. Albert Cervera i Areny 2008-06-18 02:26:09 +0200
  • bd015179be Added deskew function which also introduces the required 'linearRegression' function. Albert Cervera i Areny 2008-06-18 02:03:27 +0200
  • 3588ba1a9b Added function formatedText(region) in ocr.py which puts characters one after the other and in its line with spaces in between trying to emulate standard tesseract output. Albert Cervera i Areny 2008-06-17 20:30:50 +0200
  • 301c0a7f32 Remove commented code from last commit. Albert Cervera i Areny 2008-06-16 23:54:05 +0200
  • 1f1744edbb Several improvements in recognizer.py - Refactored template selection code. Templates are translated 5 millimeters in each direction to find the best one (make it much slower, though). - Don't import everything in __init__.py - Changed names of several functions. Albert Cervera i Areny 2008-06-16 23:53:19 +0200
  • 243ab2e684 Added duplex capabilities to Twain backend. Albert Cervera i Areny 2008-06-15 18:57:02 +0200
  • b9ef19a44d Set proper resolution in scanned images (QImage). Albert Cervera i Areny 2008-06-15 17:25:36 +0200
  • 6fbe42e63c Fixed Trigram implementation. Albert Cervera i Areny 2008-06-15 11:17:32 +0200
  • b0320e055d Added our own trigram implementation. Now it's used in recognizer. Recognizer usage of trigram still untested but should most possibly work. Albert Cervera i Areny 2008-06-13 20:31:14 +0200
  • 6cb2c3b1bd Added __init__.py file needed for backends submodule. Some fixes in sane backend, not working yet due to a sane segfault (which also happens in sane examples). Albert Cervera i Areny 2008-05-24 01:40:11 +0200
  • 6805ba78e7 Merge branch 'master' of ssh://albert@192.168.1.2/home/albert/d/git/nanscan unknown 2008-05-24 01:12:09 +0200
  • 79db4254bb Add missing backends/common.py unknown 2008-05-24 01:10:48 +0200
  • 44618101d5 Revert "Commit windows merge with twain backend working." Albert Cervera i Areny 2008-05-24 01:09:54 +0200
  • 1326a5cf02 Commit windows merge with twain backend working. Albert Cervera i Areny 2008-05-24 01:08:09 +0200
  • cd6559cb30 nanscan scanning module working on windows. twain.pyd (twain python wrapper) has been added to the repository so it's easier to deploy. unknown 2008-05-24 01:04:29 +0200
  • dca04721f4 Initial scanner support. Nothing working yet. Created: - Generic scanner.py which delegates to backends depending on platform - Initial sane backend more or less created but nothing tested yet waiting for twain merge. Albert Cervera i Areny 2008-05-24 00:16:28 +0200
  • d965700280 Added change-resolution.py utility application. Albert Cervera i Areny 2008-05-12 21:42:24 +0200
  • 7bca063e2e Now we work with millimeters instead of pixels. This should allow us work with images scanned at different resolutions. Albert Cervera i Areny 2008-05-12 17:55:51 +0200
  • 32bb3f1aa8 Initial commit of nanscan library. Albert Cervera i Areny 2008-05-12 12:01:22 +0200