Sooner or later almost everyone needs to turn a paper document into a PDF in a hurry: a signed contract, a receipt, a homework page, a form someone asked for "as a PDF, please." If you don't own a flatbed scanner, your phone camera is honestly good enough — the problem is that a raw photo usually comes out skewed, with the desk or your fingers visible around the edges, and uneven lighting that makes the text harder to read than the original paper. This guide walks through turning that rough photo into something that looks like it came out of an actual scanner, using Dagochim's free Scan to PDF tool, which runs entirely in your browser.
When you'd want this
- Submitting paperwork: photograph a contract, application or consent form, straighten it, and hand over a clean PDF instead of a crooked photo.
- Personal scans of books or course material: save a few pages you need for study without carrying the physical book around.
- Organizing receipts: flatten a crumpled receipt into something presentable for an expense report.
- Searchable archives: add OCR so you can later search the text inside the PDF with Ctrl+F instead of scrolling through images.
How to scan with Dagochim
- Open the Scan to PDF tool.
- Tap 'Take a photo' or 'Choose files' to add a document photo — or several pages at once.
- Four green dots snap automatically onto the page's corners. If any of them are off, drag each dot onto the real corner and press 'Apply to this page' to flatten it.
- Pick Original, Black & White document, or Sharper, whichever looks best, and reorder or rotate pages in the list if you scanned more than one.
- Choose a page size and image quality, turn on OCR if you want searchable text, then press 'Create PDF'.
Tips for taking a scan-friendly photo
Automatic corner detection works best when there's clear contrast between the document and whatever is behind it. Put a light sheet of paper on a dark desk, or a dark book cover on a light background, rather than photographing white paper on a white table. Light the page from the side or above instead of from directly behind the camera, so shadows don't fall across the text. Hold the phone as parallel to the page as possible and shoot from directly above; the less perspective distortion you start with, the more natural the automatic flattening will look. Leave a little room around the edges so all four corners are fully visible in the frame — if a corner gets cropped out of the photo itself, no amount of software correction can bring it back.
When automatic corner detection gets it wrong
If the background is close in color to the page, or another object overlaps the document, the automatic detection can grab the wrong points. When that happens, drag the four green dots with your finger or mouse until they sit exactly on the document's actual corners. You can press 'Auto-detect corners' again to retry, or 'Reset to full image' to start over from scratch. Whichever way you adjust them, remember to press 'Apply to this page' afterward — that's the step that actually performs the perspective warp using your corrected corners.
Black & White document mode vs. Sharper: which to pick
The 'Black & White document' filter shines when paper has yellowed with age or the lighting across the page is uneven, since it analyzes brightness locally rather than using one global threshold, which separates text from background cleanly and produces that crisp, photocopier-like look while also shrinking the file size considerably. 'Sharper' keeps the original colors but boosts contrast a little, which is the better choice when you need to preserve a stamp, a signature in colored ink, or a highlighted section. For pages that are genuinely photographic — a diagram, a photo pasted into a notebook — leaving the filter on 'Original' usually looks the most natural.
Making a searchable PDF with OCR
Turning on 'Add text recognition (OCR)' layers an invisible text version of the page underneath the scanned image, so once the PDF is created you can select and copy text with your mouse, or search the whole document with Ctrl+F, exactly like you could with a native PDF. The image you actually see never changes — if OCR misreads a word here or there, it only affects the invisible, searchable layer, not what's printed on the page. The engine recognizes Korean and English together by default, and the language data it needs is downloaded once and cached, so later scans start noticeably faster. For the best accuracy, straighten the corners and apply a filter first, then turn OCR on last; a crisp, flat image gives the recognizer far more to work with than a blurry, skewed one.
A word about copyright when scanning books
Technically, scanning any physical page — including one from a book, magazine or newspaper — is simple. Legally, it is a very different story: scanning an entire copyrighted work and then sharing it, uploading it, or distributing the file to other people can infringe copyright, even if making the scan itself only took a few minutes. Digitizing a whole book and passing the file around a class chat or posting it to a shared drive usually goes well beyond what's considered personal use under most copyright laws. Use this tool for documents you own or have the right to copy — contracts, your own notes, receipts, forms you need to submit — or for scanning a handful of pages from a book strictly for your own study or reference. Even for coursework or research, avoid distributing a full scan of someone else's copyrighted book without permission.
How this compares to a dedicated scanner app
Dedicated scanner apps usually require an install, and many push ads or an account sign-up before you can export a PDF. This tool works straight from the browser tab you already have open, and nothing you photograph is ever sent to a server — corner detection, warping, filtering and OCR all run locally on your device. Because the corner detector here is a small, custom piece of JavaScript rather than a heavyweight, professionally tuned computer-vision pipeline, it may occasionally need a manual nudge that a polished scanner app wouldn't require; the manual corner-dragging feature exists specifically to close that gap, so the final output quality ends up very similar in practice. It's a particularly good fit when you just need to scan one or two pages quickly and don't want to install yet another app for it.
Common issues
- The photo looks too dark or has glare: light the page from the side or above rather than straight-on, and angle any light source slightly if the page has a glossy cover or plastic sleeve that's reflecting it.
- The corner handles keep landing outside the page: press 'Reset to full image' and then drag each of the four handles into place one at a time, rather than all at once.
- OCR feels slow the first time: the first run has to download the recognition language data; later runs on the same browser reuse the cached copy and are much faster.
- The finished PDF is too large: lower the image quality slider, switch to the Black & White document filter, or run the result through Dagochim's Compress PDF tool afterward for extra savings.
Frequently asked questions
How does this compare to a dedicated scanner app?
You don't need to install anything, and your photos never leave your browser. The trade-off is that automatic corner detection may not be quite as polished as a dedicated scanner app, so it's worth double-checking the corners by hand.
Is it okay to scan a book and save it as a PDF?
Scanning for personal use or study is fine, but scanning an entire book and sharing or uploading it for others can infringe copyright.
Why is my OCR recognition coming out wrong?
Blurry or heavily skewed photos reduce accuracy. Line up the corners precisely and apply the Black & White document mode to sharpen the text before trying OCR again.