Pdf-extract is an open source set of tools and libraries for identifying and extracting semantically significant regions of a scholarly journal article (or conference proceeding) PDF.