Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I compiled it from sources with leptonica. This is a png image with transparent background, which I edited adding a blue color and still this error:

Tesseract Open Source OCR Engine v3.02.02 with Leptonica
Empty page!!
Empty page!!

Here's the image input:

enter link description here

share|improve this question

try the psm option.

-psm N
    Set Tesseract to only run a subset of layout analysis and assume a certain form of image. The options for N are:

    0 = Orientation and script detection (OSD) only.
    1 = Automatic page segmentation with OSD.
    2 = Automatic page segmentation, but no OSD, or OCR.
    3 = Fully automatic page segmentation, but no OSD. (Default)
    4 = Assume a single column of text of variable sizes.
    5 = Assume a single uniform block of vertically aligned text.
    6 = Assume a single uniform block of text.
    7 = Treat the image as a single text line.
    8 = Treat the image as a single word.
    9 = Treat the image as a single word in a circle.
    10 = Treat the image as a single character.

Reference: http://hilojack.sinaapp.com/?p=866

share|improve this answer

Tesseract is not trained to recognize handwriting. Don't know what it does with those colours either.

You could try and train tesseract with that handwriting...

share|improve this answer

Process your image to threshold background color. Turn the text color to black (for better recognition. once you have threshholded the background it should be fairly simple to change color values.) turn your image into grayscale. Then convert to .tif format.

Now you might have a chance at processing that image (super user doesn't let me post images so I link them):

Processed Image

Now run the following command:

tesseract test.tif test_output -psm 7

and the result was:

Tist

Which is pretty good given that I have not used any additional training data beside standard eng.

Screenshot of the result

share|improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .