OCR of documents retrieved from USPTO
JonahP1621 last edited by
USPTO outputs its patent correspondence as images in PDF files, encoded without searchable text.
As my standard procedure, I run OCR on all patent correspondence from the USPTO. It has the two benefits of allowing me to search the documents while working and allowing my computer's file system to index the documents so that I can quickly search for relevant documents in the future.
I usually multitask while OCR is running, but it still takes about 15 seconds out of my day for each document and probably a bit more for some other users. It would increase productivity a bit if AppColl would automatically run OCR on all patent correspondence that it retrieves from the USPTO.
Yup, I agree, this is a little task that I do, but I would love it if it were already completed when I received the doc!