Word search doesnt work sometime. what can i check to find out how to fix?

word search doesnt work sometime. what can i check to find out how to fix?

You could start by providing more details; ideally what specifically you are searching for, and where it isn’t being found despite your expectations; then we can take it from there.

i had a letter scanned into DT3. It was from a lady named Gerstenberger. I typed Gerstenberger in the DT3 search space in the bar at top right all database search.

Result…one other document with that name was found … the letter was not found. this has happened on several other documents but not all

Was OCR applied to the scan? Otherwise its contents aren’t searchable. Or is this a handwritten letter?

1 Like

how do I check to see if OCR IS on?

  • Select the document, then choose Tools > Inspectors > Concordance: List.
    Do you see words in there?
    • If so, there is a text layer in the document.

You said before that you scanned this document. If neither your scanner nor you (with DT or another software) performed OCR on it (which only you can know), there’s no text layer. In general, DT will mark PDFs with a text/OCR layer with the type “PDF+Text” in the list view, the info sidebar, and the info popup.

There are words in a list.

I dont know what to do with info. I dont know what a " text Layer" means. I was searching for the document by her last name Gerstenberger. I dont think her name is on that list. Some of the words? on the list are truncated. I found it on the list, but the spelling is Gcrstenbcrger. …That is really interesting.

  1. How do i make sure DT performs OCR?
  2. How do i see list view?
  3. How do I see info sidebar?
  4. How do I see info popup?
    to look for “PDF+Text”?

So obviously the OCR didn’t recognize the word correctly. Which explains why it can’t be found.

1 Like

I suggest you read the manual. It should answer most of your questions. Others have already been discussed thoroughly in the forum – searching here for “OCR” and “automatic” should reveal the relevant threads.

1 Like

I suggest that you consult the help menu.

1 Like

There is built-in help under Help > DEVONthink 3 Help.
Also there is a downloadable manual.

An OCR’d PDF is comprised of layers.
The bottom layer is the image.
The layer above that is the text layer containing words identified by the OCR engine. No OCR engine is 100% accurate and much of the results are determined by the quality of the original. Sufficient resolution, a maximum of 300dpi is more than sufficient for the majority of OCR operations. Also the lighting and contrast is important if you’re taking photos of a document.

thank you. the manual didnt show first time i looked in help?
i will look for … 1. How do i make sure DT performs OCR?
2. How do i see list view?
3. How do I see info sidebar?
4. How do I see info popup?
to look for “PDF+Text”?

in the manual

1 Like

Just type ocr into the search field at the top of the DEVONthink Help menu for a start.

Stephen