ICR/OCR help wanted

[brunocechet]

I'm working with python and trying to recognize these images (imgur.com/a/HsvUv), but without satisfied results yet.

I've tried pytesseract and the Google Vision API following this tutorial (github.com/GoogleCloudPlatform/cloud-vision/tree/master/python/text).

I don't know what I could do to get better results

Someone could try to help me ?

Tnks in advance

---

Clickable links:

http://imgur.com/a/HsvUv

https://github.com/GoogleCloudPlatform/c...ython/text

[Larz60+]

Very skimpy information, please show what code & packages you have so far
and a sampling of your results and any errors

[brunocechet]

I'm using the default implementation based on an example of Google Vision API, like follows: https://github.com/GoogleCloudPlatform/c...xtindex.py

For the pytesseract, I'm just using the pytesseract.image_to_string(), passing the cropped image as parameter.

Do you guess that isolating the letters I could get better results ?