Hi @gogopotter90​ 

It's hard to say without data but I have seen similar issues where there appears to be no output written to the PDF file when there is a mismatch between the input data's coordinate system and the output page units mismatch. The features are usually written but the file appears to be empty as the features are very small.

And yes, it is possible to read a PDF file as raster. To do so, check Non-Spatial and Read Rasterized Pages in the PDF reader parameters. A raster should be created for each PDF page.

thanks for ur reply,i have tried now only to read the pdf as raster then save it as again pdf without doing any thing in workspace but i got insufficient data for an image .i am not sure what should i do ? even i could see the output in inspector of fme .could u provide me an adice,this data belongs to company and it will be hard for me to share it .thanks for help

Hi @gogopotter90​ 

Unfortunately, it is hard to say what the issue is for your workflow without seeing the data.

I have attached an example workspace where the PDF reader outputs a rasterized page and the page is written back to PDF. I hope it helps.

thanks alot for ur workspace but i got error @ReinterpretRaster: Converting from 4 bands to the 'RGB24' interpretation in RASTER mode requires an input RGBA raster. Consider converting to the 'RGBA32' interpretation firstRasterInterpretationCoercer

i am using fme 2018 because this workspace for higher version

The workspace I shared uses the RasterInterpretationCoercer to convert the RGBA32 output from the rasterized PDF to RGB24 (since the PDF writer does not work with alpha bands).

If your rasterized input does not have an alpha band, you do not need to use a RasterInterpretationCoercer.