Klemens Dickbauer

Hi Quan, thanks for your quick reaction! My code: private Tesseract instance; private String OCRResult; // .. instance = new Tesseract(); instance.setPageSegMode(1); instance.setOcrEngineMode(2); instance.setLanguage("deu"); // also tried with "eng" // i tried with and without the 'tessdata' at the end of the data path instance.setDatapath(new File("/usr/share/tesseract-ocr/4.00/tessdata/").getPath()); OCRResult = instance.doOCR(imageFile); I don't think that it can't read the traineddata file, because...

When invoking tesseract via tess4j on one of my linux boxes, I get a SIGSEV preceeded by the following two lines: Failed loading language 'deu.traineddata' Tesseract couldn't load any languages! No other messages at this position in the log. I think the file itself is found (I tried both using the TESSDATA_PREFIX variable and using instance.setDatapath() in the code); also when I download another deu.traineddata (e.g. the curernt one on github) I get parameter check errors, which is probably okay...