Databases
StackOverflow2013
Postsdbo
POTesseract OCR Library - Learning Font
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POTesseract OCR Library - Learning Font
    primarykey
    data
    text
    <p>Well I'm using a complied .NET version of this OCR which can be found @ <a href="http://www.pixel-technology.com/freeware/tessnet2/" rel="noreferrer">http://www.pixel-technology.com/freeware/tessnet2/</a></p> <p>I have it working, however the aim of this is to translate license plates, sadly the engine really doesn't accurately translate some letters, for example here's an image I scanned to determine the character problems</p> <p><img src="https://i.stack.imgur.com/CDQ3n.jpg" alt="enter image description here"></p> <p>Result:</p> <p>12345B7B9U ABCDEFGHIJKLMNUPIJRSTUVHXYZ</p> <p>Therefore the following characters are being translated incorrectly:</p> <p><strong>1, O, Q, W</strong></p> <p>This doesn't seem too bad, however on my license plates, the result isn't so great:</p> <p><img src="https://i.stack.imgur.com/SEqNC.jpg" alt="enter image description here"> = H4 ODM</p> <p><img src="https://i.stack.imgur.com/Wh2KV.jpg" alt="enter image description here"> = LDH IFW</p> <p>Fake Test</p> <p><img src="https://i.stack.imgur.com/sNrQG.jpg" alt="enter image description here"> = NR4 y2k</p> <p>As you might be able to tell, I've tried noise reduction, increasing contrast, and remove pixels that aren't absolute black, with no real improvements.</p> <p>Apparently you can 'learn' the engine new fonts, but I think I would need to re-compile the library for .NET, also it seems this is performed on a Linux OS which I don't have.</p> <p><a href="http://www.scribd.com/doc/16747664/Tesseract-Trainingfor-Khmer-LanguageFor-Posting" rel="noreferrer">http://www.scribd.com/doc/16747664/Tesseract-Trainingfor-Khmer-LanguageFor-Posting</a></p> <p>So I'm stuck as what to try next, I've wrote a quick console application purely for testing purposes if anyone wants to try it. If anyone has any ideas/graphic manipulation/library thoughts, I'd appreciate hearing them.</p>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. This table or related slice is empty.
    1. USAsh
    plurals
    1. This table or related slice is empty.
    1. PL
      singulars
      1. LTLinked
    2. PL
      singulars
      1. LTLinked
    3. PL
      singulars
      1. LTLinked
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    2. PO
      singulars
      1. PTAnswer
    3. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POTesseract OCR Library - Learning Font
      1. USAsh
      1. VTBountyStart
    2. VO
      singulars
      1. POTesseract OCR Library - Learning Font
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. POTesseract OCR Library - Learning Font
      1. This table or related slice is empty.
      1. VTUpMod
    1. COHere is another Problem: O and 0 have excalty the same shape, so there is no way the OCR engine could recognize them correctly.
      singulars
      1. POTesseract OCR Library - Learning Font
      1. USMartin Thurau
    2. COYeah I noticed that, can't believe they are exactly the same shape. Anyway I've already thought of a method of dealing with this outside OCR, so it's np.
      singulars
      1. POTesseract OCR Library - Learning Font
      1. USAsh
    3. CO@Martin Thurau: wrong ! In a text font, 0 and O are different most of the time - happily ! So if you do a special training for this font, you should be able to differentiate them, this is my case. Of course that may be much more difficult if you don't know the fonts in advance.
      singulars
      1. POTesseract OCR Library - Learning Font
      1. USEmmanuel
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload