Problems with highlighting text in a image

Asked by Yogesh Anand on 2017-05-02

Hi, I am working on the SikuliX and exploring its reading text feature (OCR) from a region (image). Sometimes it identifies the word and sometime it doesn't. Also, if its a phrase (combination of words), it fails to identify them. Is there any better and reliable way to do this task?

Question information

Language:
English Edit question
Status:
Answered
For:
Sikuli Edit question
Assignee:
No assignee Edit question
Last query:
2017-05-02
Last reply:
2017-05-12
RaiMan (raimund-hocke) said : #1

sorry no.

You have to live with its current weaknesses and find workarounds.

There will not be any improvements until end of this year with version 2.

You might have a look at Tess4J, which exposes the Tesseract API to the Java level and hence could also be used in the Jython context.

masuo (masuo-ohara) said : #2

How about trying the following way?
Run another OCR application that can be read highlighting text.
And by SikuliX, simulate the reading execution operation to the OCR application.

Find out OCR application that can use clipboard to get text.
SikuliX get text from the OCR application through the clipboard.

Can you help with this problem?

Provide an answer of your own, or ask Yogesh Anand for more information if necessary.

To post a message you must log in.