How to find confidence level of OCR text extracted using tesseract in java?

Asked Apr 27 '16 at 15:25

Active Apr 27 '16 at 16:34

Viewed 1,011 times

import java.io.File;
import com.googlecode.tesseract.android.TessBaseAPI;
public class ConfidenceLevelFinder 
{
    public static void main(String[] args) 
    {
        TessBaseAPI baseApi=new TessBaseAPI();
        baseApi.setPageSegMode(TessBaseAPI.PageSegMode.PSM_SINGLE_BLOCK);
        baseApi.setImage(new File("zoomed.jpg"));
        baseApi.meanConfidence();

        baseApi.end();
    }
}

I am using above code.
Using com.googlecode.tesseract.android.TessBaseAPI; library
But geeting Exception in thread "main" java.lang.UnsatisfiedLinkError: no lept in java.library.path
Which dll is needed for this using tesseract android jar?

edited Apr 27 '16 at 16:34

asked Apr 27 '16 at 15:25

sharmishtha kulkarni

1

Welcome to stackoverflow. Please fix the problem in your last point (5). Also post the error trace. – Robert Apr 27 '16 at 16:09
Which platform are you developing for? Windows? Android? – rmtheis Apr 27 '16 at 21:25
Try using the Tess4J project. The com.googlecode.tesseract.android.* classes are for Android only. – rmtheis Apr 28 '16 at 13:29
okay. @rmtheis can you provide code to determine Confidence level of output OCR text generated by tesseract? – sharmishtha kulkarni Apr 29 '16 at 04:51
The code to do that will be specific to the library that you end up using, so no, not without more info. – rmtheis Jun 15 '16 at 04:14

How to find confidence level of OCR text extracted using tesseract in java?

0 Answers0