Tag: tesseract

Cause: the package for English language is missing.

On Centos:

sudo yum install tesseract-en

On Ubuntu:

sudo apt-get install tesseract-ocr-eng

Error:

tesseract test.tif output.txt
Tesseract Open Source OCR Engine
check_legal_image_size:Error:Only 1,2,4,5,6,8 bpp are supported:16

Cause: tif image bits/sample value not in 1,2,4,5,6,8 or tif image is transparent with alpha value (Extra Samples: 1<unassoc-alpha>):

tiffinfo  test.tif
  TIFF Directory at offset 0x7b6 (1974)
  Image Width: 145 Image Length: 39
  Resolution: 72, 72 (unitless)
  Bits/Sample: 8
  Compression Scheme: AdobeDeflate
  Photometric Interpretation: min-is-black
  Extra Samples: 1<unassoc-alpha>
  FillOrder: msb-to-lsb
  Orientation: row 0 top, col 0 lhs
  Samples/Pixel: 2
  Rows/Strip: 39
  Planar Configuration: single image plane
  DocumentName: a.tif
  Software: ImageMagick 6.2.8 05/07/12 Q16
  Predictor: horizontal differencing 2 (0x2)

Solution: convert tif to acceptable bit sample and remove alpha value:

convert test.tif -background white -alpha remove -flatten -alpha off test.tif

or convert to jpg and reconvert to tif again should remove alpha:

convert test.tif test.jpg
convert test.jpg test.tif

Error:

tesseract test.png output.txt
Tesseract Open Source OCR Engine
name_to_image_type:Error:Unrecognized image type:test.png
IMAGE::read_header:Error:Can’t read this image type:test.png
tesseract:Error:Read of file failed:test.png

Cause: You are using tesseract version <=2.0.4

Solution:

1. convert png to tif:

convert test.png -background white -alpha remove -flatten -alpha off test.tif

2. Upgrade teeseract to latest version