I'm trying to execute tesseract from command line in Ubuntu 17.10. I want the output in a .tsv
file because I need the confidence rate. As explained here, I execute:
tesseract testing_img.png out tsv
but I'm getting the following error:
read_params_file: Can't open tsv
Tesseract Open Source OCR Engine v3.05.00 with Leptonica
and the output is written correctly in a out.txt
file. It seems that it reads that tsv
parameter as a file to read, but I don't know why.
I've compiled Tesseract from source because I need Tesseract 3.05 in order to have a .tsv
file as output, so I can't use the version in the Ubuntu repository because it has Tesseract 3.04.
I'm running Ubuntu 17.10.
Here's some information about my Tesseract installation:
$ tesseract --version
tesseract 3.05.00
leptonica-1.75.3
libpng 1.6.34 : zlib 1.2.11
$ ls /usr/share/tesseract-ocr/tessdata/
configs eng.traineddata ita.traineddata osd.traineddata pdf.ttf tessconfigs
$ echo $TESSDATA_PREFIX
/usr/share/tesseract-ocr/
I had the same problem, in my case a file called tsv
in the directory
/usr/share/tesseract-ocr/tessdata/configs
is missing. I downloaded the source code of tesseract from:
https://github.com/tesseract-ocr/tesseract/archive/3.05.00.tar.gz
and replaced the content of the folder configs
.