meanTextConfidence is 0 for doc with empty last page but non-empty others in createDocumentsWithResults method

nguyenq / tess4j

Java JNA wrapper for Tesseract OCR API

Apache License 2.0

1.58k stars 372 forks source link

meanTextConfidence is 0 for doc with empty last page but non-empty others in createDocumentsWithResults method #242

Closed yshyman closed 1 year ago

yshyman commented 1 year ago

I am doing PDF ocr and found a bug that meanTextConfidence reports 0 in case last page of a PDF is empty while other are not. This returns empty info in OCRResult object.

Attaching sample document res.pdf

Best regards, Yurii

nguyenq commented 1 year ago

We're unable to reproduce the issue. A valid meanTextConfidence value was returned for the PDF file.

nguyenq commented 1 year ago

Not an issue.