An image PDF is a PDF whose pages are stored as photos or scanned images.
A regular PDF may contain copyable text information, but an image PDF only shows visible letters on screen and does not contain the text data this tool reads.
This tool can extract text from PDFs that already contain a text layer. PDFs made only of images are not supported in the current MVP.
Support for reading text from images may be considered later with a different technique, but it is not available now.