We are currently evaluating PDTron PDFNet, to replace another component in one of our internal document workflows.
Evaluation wen fine so far, however I could not find any information about the PDF Compression module (http://www.pdftron.com/pdfnet/addons.html#Optimizer); is this included in the evaluation, and if not is there a way for us to evaluate it?
What we are really interested in is whether you supports MRC compression at a high quality. We are looking for an MRC compression algorithm that provides 80-90% compression of image-over-text PDF documents, with a low number of compression artifacts. (That is, an 80 page text document would be between 4 and 5 MB. While we have found a similar compression in another SDK software product we have not yet found an acceptably low number of artifacts).
Currently PDFNet doesn’t offer built-in support for MRC (Mixed Raster Content) compression. MRC is really only a way of splitting original full color pages to a monochrome mask and a color layer which are compressed with JBIG2 and JPEG or JPEG2000.
Because PDFNet supports both JBIG2 and JPEG2000/JPEG compression which are needed for MRC, it could be used to implement MRC compression. Please keep in mind that MRC applies only to color scanned images whereas most scanned documents are monochrome. The main issue with MRC in context of PDF is related to quality and diminishing results if significant quality loss is not acceptable. For this reason, at the moment we don’t have specific plans to add MRC option to PDFNet. If you still want MRC, please let us know and we’ll work on it
PDFNet offers a class called pdftron.PDF.Optimizer, as shown in Optimizer sample project:
The sample is included as part of the SDK and you don’t need to download any additional components. With optimizer you could sometimes significantly decrease PDF file size without quality loss (e.g. by removing redundant objects, recompressing images with better compression methods, sub-setting fonts, etc.). Optimizer also supports shrinking of PDFs with large images via sub-sampling.