How to recognise transparent images?

[jukkaharkki]

Using OpenPDF library I can scan through document objects, see images and even extract these images. However, I have not found a way to check if an image is transparent. The information is not within an individual image but rather in relationship of two images or layers I assume. But how to check this out?

I don't need to flatten transparency (I know there are tools for it), just identifying is enough. The big picture is about printing a PDF and there transparent images are not supported by the service provider. I'm trying to identify PDF documents containing transparent images and reject printing.

Any hint appreciated

[mkl-public]

When you inspect the objects a PDF consists of, you should have a copy of the current specification at your hand, ISO 32000-2. You can get a free copy at the PDF association, see https://www.pdfa-inc.org/product/iso-32000-2-pdf-2-0-bundle-sponsored-access/

In section 8.9.5 you'll find the image dictionaries explained. In particular the entries ImageMask, Mask, and SMask. Beware, though, transparency can also be achieved for regions of a PDF page which by chance may contain images, e g. by clip paths and general graphics state SMasks.

[jukkaharkki]

Thanks, that led me to the right direction. Eventually I did very rudimentary test: if image has a mask, then it is part of transparency. Probably not 100% accurate but serves my purpose and worked with few test cases.