We are trying to switch to using PDFBox to create the thumbnail images of PDF Pages in our application. (The older product we currently use fails on OS 11).
I'm running into a problem if there is text on the page, the thumbnail image makes it hard to make any sense at all of the text. (yes, these are thumbnails, and don't need to be readable - but should be recognizeable.) The older renderer created thumbnails that, while the text ws not legible, it was definitely visible, and you could tell the shapes of the words. PDFBox creates thumbnails where the text is more of a scattering of random pixels, an you have to guess that it might be text. Is there any way when generating the raster to have it treat Text pixels as the higher priority for the color of a pixel? It seems to be only allowing the text color to control the pixel if the entire pixel is part of the character.
--------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
