For teachers, trainers, and students working from paper

Scanned PDF to Quiz: Turn a Scanned Document, Photocopied Handout, or Textbook Page Into a Test

Upload the scan exactly as it came off the copier or your phone camera. The page image is read directly, so an image-only PDF with no selectable text still produces a quiz with an answer key. No OCR step, no retyping.

PDF, JPG, PNG, HEIC, TIFF, and BMP accepted. Files are processed securely and deleted automatically.

Upload your scan and generate the quiz

Click to upload or drag and drop

PDF, DOCX, PPTX, TXT, JPG, JPEG, PNG, HEIC, ODP, ODT, BMP, or TIFF

up to 20MB

Please wait, your quiz is being created...

Uploading...

The short answer

You can make a quiz from a scanned PDF without running OCR first. A scanned PDF stores a picture of the page instead of selectable text, so the page has to be read visually before any question can be written from it. Upload the file as it is and the page image is read directly, then questions and an answer key are generated from what the page says. Scan quality, not file format, is what decides how good the questions are: 300 DPI, flat and square in the frame, is the practical target.

Last updated July 2026.

What each kind of scan actually gives you

Every scanned page falls somewhere on a quality ladder, and where it lands decides how much of the text is recovered and therefore how good the questions are. This is the honest version, including the cases where you should fix the scan before uploading rather than hoping for the best.

Source	Has a text layer	How well it reads	What to do first
Flatbed scan at 300 DPI	No, image only	Very well. Close to a native text PDF	Nothing. Upload as is
Scanner app on a phone	Usually no	Well, if the app squared the page	Check the crop caught all four corners
Plain phone photo of a page	No	Well in the center, weaker near a curved spine	Shoot straight down, press the book flat
Third-generation photocopy	No	Mixed. Faint serifs and subscripts drop out	Rescan in grayscale, not black and white
Fax or 150 DPI archive scan	No	Poor on small print, fine on headings	Rescan the original if you still have it
Searchable PDF already run through OCR	Yes	As well as the OCR pass was, no better	Spot-check the OCR text for garbled words
Native digital PDF	Yes	Best case. Nothing is inferred	Nothing

A quick way to place your own file on this ladder: run it through the PDF word counter. A native or OCR'd PDF returns a real word total. A pure image scan returns zero or close to it, which tells you immediately that the page has no text layer and has to be read visually.

Why a scanned PDF breaks most quiz generators

Most tools that promise to build a quiz from a document do one thing when you upload a file: they pull the text layer out and hand that text to a language model. That works perfectly on a file exported from Word or Google Docs, because the text layer is right there. Hand the same tool a scan and it extracts nothing, because there is nothing to extract. The page is a picture. What you get back is an error, or worse, a set of generic questions invented from the filename and a few stray characters.

This is why so many teachers conclude that AI quiz tools do not work on their materials. The materials are fine. The pipeline was never built for paper. Anything that came from a copier, a departmental scanner, a phone camera, or a 1990s course pack lands in this category, and in a lot of school and training environments that is most of what people actually have.

Reading the page as an image instead solves it at the source. The page is interpreted the way a person reads it, so layout survives: a numbered list stays a numbered list, a table keeps its rows, a heading is still recognizably a heading. That structure matters more than it sounds, because questions written from a document that kept its shape follow the document's own logic rather than a flattened stream of words.

No OCR step

Upload the scan as it is. You do not need Acrobat, a separate OCR service, or a searchable-PDF conversion first.

No retyping

The alternative to a tool that reads scans is typing the handout back in by hand. On a ten page packet that is an evening.

Layout survives

Tables, numbered steps, and headings are read in place, so questions follow the structure of the original page.

Mixed packets work

A packet where some pages are scanned and some are digital is handled in one upload, not two workflows.

How to turn a scanned document into a quiz

Upload the scan

Drop in the scanned PDF or the photo, up to 20 MB. Grayscale scans stay well under that even for a long chapter.

The page is read

Each page is interpreted visually, including pages with no text layer at all, so nothing has to be extracted first.

Pick the format

Choose multiple choice, true or false, or short answer, and set how many questions you want from the material.

Review and hand out

Skim the questions against the page, drop any that missed, then share the link or print with the answer key.

Step four matters more on scanned material than on digital files. On a faint photocopy, the one thing worth checking is any question that turns on a number, a subscript, or a date, because those are exactly the characters a poor scan degrades first.

Who works from scans every day

Teachers with a course packet

The reading is a photocopied chapter from a book the department never bought digitally. It has been copied so many times the margins are gray. It is still the assigned reading, and a quiz on it still has to exist by Friday.

Trainers with legacy manuals

The equipment manual and the written safety program predate the current file server. They exist as scans in a shared folder. Comprehension checks still have to come from those exact documents to be worth anything on an audit.

Students with borrowed material

The library copy could not leave the building, so what you have is twenty phone photos of the chapter. Testing yourself on them beats rereading them, and photos upload as readily as a PDF does.

Frequently asked questions

Can you make a quiz from a scanned PDF?

Yes. A scanned PDF holds a picture of a page rather than selectable text, so the file has to be read visually before questions can be written from it. PDFQuiz reads the page image directly and generates multiple choice, true or false, and short answer questions with an answer key, so you do not need to run a separate OCR step or retype the handout.

How do I know if my PDF is scanned or searchable?

Open the file and try to select a sentence with your cursor. If the text highlights word by word, the PDF has a text layer and is searchable. If your cursor draws a box over the whole page or selects nothing at all, the page is an image and the PDF is scanned. Pressing Ctrl+F and searching for a word you can plainly see is the same test in reverse.

Do I need to OCR a PDF before making a quiz from it?

Not with this tool. Running Acrobat or another OCR program first is an extra step that mainly helps when your scan is poor, because it lets you proofread the recovered text before anything is generated. For a clean scan at 300 DPI, uploading the original file directly gives equal or better results, since nothing is lost to a second conversion.

What scan quality do I need for good quiz questions?

Scan at 300 DPI, keep the page flat and square in the frame, and use grayscale or color rather than hard black and white for pages with faint print. Third-generation photocopies, heavy skew, and dark shadows across the gutter are what actually cause missed lines, not the file format. A page a person can read comfortably usually reads fine.

Can it read a photo of a textbook page taken with my phone?

Yes. JPG, PNG, HEIC, TIFF, and BMP images upload the same way a PDF does, so a phone photo of a textbook page or a worksheet works. Fill the frame with the page, avoid your own shadow, and shoot straight down rather than at an angle. Curved pages near the spine are the most common cause of dropped words.

Does a scanned PDF produce worse questions than a text PDF?

On a clean scan the difference is small. The questions follow whatever text is recovered, so a crisp 300 DPI scan of a textbook chapter produces questions very close to what the same chapter would produce as a native PDF. Quality drops when the scan does: faint photocopies, skewed pages, and handwriting in the margins are where accuracy falls off.

How large can the scanned file be?

Each upload takes a single file of up to 20 MB, which comfortably covers a scanned chapter or a packet of handouts. Scanned pages are much heavier than text pages, so a long scanned chapter can approach that limit where a text PDF never would. Splitting a very large scan into two uploads, or rescanning in grayscale rather than color, is the simplest fix.

What about handwriting on the scanned page?

Handwriting is read too, though less reliably than print, and neatness matters more than anything else. Margin notes on a printed handout usually come through. If the page is entirely handwritten, the handwritten notes to quiz page covers what to expect and how to photograph the page so it reads.

Upload the scan you already have

Photocopied handout, scanned chapter, phone photo of a page. It does not need a text layer and it does not need retyping.

Generate a quiz from your scan

Handwritten notes to quiz Photograph your own notes and test yourself on them. Image to quiz Turn a JPG or PNG of any page into questions. Textbook to quiz Build a chapter quiz from an assigned reading. Worksheet to quiz Reuse an existing worksheet as a graded quiz.