Maybe this is a bit off, but my experience with PDFBox and looking for bar-codes led me to completely avoid rendering the PDF. Instead I iterate the resources in the pages, and look for images which I process one by one by calling PDImage.getImage(). This seems to avoid a few problems. Also that makes for easier deformation attempts (rescaling up, rescaling down, blurring, etc.) to increase success rate. Notably this does not work well with generated PDF files, where bar-codes are actually text with crafted fonts.
На пт, 5.04.2019 г. в 17:43 ч. STAMPF Lukas <[email protected]> написа: > > Hi, > > I am using PDFBox to generate BufferedImages which I feed to ZXing 3.3.3 to > do QR scanning. When I updated from 2.0.13 to 2.0.14 I noticed a severe > reduction in scanning reliability. I guess its related to this change > https://issues.apache.org/jira/browse/PDFBOX-4391 > > This is the render command I use: > bim = pdfRenderer.renderImageWithDPI(pageCounter, dpi, > ImageType.BINARY); > > I try multiple dpi settings between 150 and 300, to increase my changes to > find a QR code. > Is there some flag I can set to test out if the antialiasing causes this? > > Kind regards > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
