Can PP help to get the embedded molecules(smiles string or 2D draw) from pdf document?

Hi everyone, I want to collect molecules from some papers. But the all structurial information is embedded in pdf document. Some papers contains smiles in pdf, but others only provides 2d plot in pdf. Can pp help me to get the structures of molecules from pdf document? Or do you have other suggestion to achieve my goal? These pdf documents always have thousands of structures. I do not want to draw them one by one.