Tesseract don't detect legible text

[Jorge Infante]

I have a .tif with information about cadastral data.

There is a method to do better the recognition?

I'm attaching the .tif.

The output of command:

tesseract dont-detect.tif dont-detect -l spa  

give me in a dont-detect.txt:

Í s:cc¡on <---- This is the "SECCION" word
MANZANA
GRAFICO
sun. mv. *
6
253
5

Any idea to do better this?

Thanks