search

AiPacino / tesseract-ocr

Automatically exported from code.google.com/p/tesseract-ocr
Other
2 stars 0 forks source link

OCR works better when image is #1212

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago 
What steps will reproduce the problem?
1. Run an image through tesseract
2. Scale the image by 3x, cubic interpolation in GIMP
3. Run the bigger image through tesseract

What is the expected output? What do you see instead?
Step 1 produces garbage:

"Is very small The quesllon naturally zlrlscs. whethcr lhls dlsplacemenl IS"

Step 3 produces decent output:

"is very small. The question naturally arises whether this displacement is"

What version of the product are you using? On what operating system?

3.02.02 on Windows

Please provide any additional information below.

We should not have to manually resize images, the tesseract software should 
automatically internally resample the image to a suitable resolution before 
running OCR on it.

Original issue reported on code.google.com by omegat...@gmail.com on 23 May 2014 at 11:55