Re: How to output multiple matches/results instead of just one?

[Tom Morris]

You'll need to use the API to get that level of control.

On Thursday, May 5, 2016 at 8:22:54 AM UTC-4, shutupyoudo...@gmail.com wrote:

I want to use tesseract to convert images of a single Chinese/Japanese character which will be handwritten by a client-side user on an html5 canvas.

After several tests, I find Tesseract's recognition of handwritten chinese characters to be very good, but sometimes the result is slightly off.  Sometimes I get the correct 木 character as output, but sometimes I get the slightly different 本 character.

How can I make Tesseract output something like a list of top 10 best matches, one of which will likely be the desired result, instead of just outputting one incorrect result?

I want it to do something like:

Command: tesseract tree.gif out -psm 10 -l jpn  ---->  Output (best matches):  本,　木,　休,　十,　八,　六,.......etc

An example image I'm using is below.

[Quan Nguyen]

Specifically, ChoiceIterator.

https://github.com/tesseract-ocr/tesseract/wiki/APIExample

[shutupyoudo...@gmail.com]

Thanks, Tom!

[shutupyoudo...@gmail.com]

Thanks, Quan!  I appreciate you elaborating and including a link.