Scan to PDF

What the Scan to PDF tool does

Scan to PDF turns the camera that's already in your pocket into a one-tap document scanner. You open the page on your phone, tap Open camera, the browser asks permission for the rear lens, and from there each shutter tap captures one page. Photograph the receipt, photograph the contract, photograph the handwritten meeting notes — when you're done you tap Build PDF and the captures get assembled into a single file named scan-{today}.pdf. Download, and the scan is yours.

Here's the part most tools won't say plainly: the frames never leave your device. The browser API getUserMedia streams video straight into a canvas in your tab. Each capture becomes a JPG held in your browser's memory. pdf-lib stitches the JPGs into the PDF locally, the download link is a blob URL, and the whole pipeline closes when you close the tab. There is no upload step. There is no server. There is no account. There is nothing to leak because there is nothing in transit.

Why a browser scanner is the right shape for this job

The dominant phone scanners — CamScanner, Adobe Scan, Microsoft Lens, Genius Scan — all want you to install an app, sign in, and let them sync your scans to a cloud. That's a strange ask for what is, at heart, a one-shot job: I have a piece of paper in front of me, I want a PDF of it on my computer in the next sixty seconds. The app model trades a minute of installation, a forced account creation, and an unknown future of background syncing for a feature you'll use four times a year.

A browser does the job cleaner. You already have a browser open. You already trust your camera with it (banks use it, video chat uses it, telehealth uses it). The shutter button is right there, the assembly is right there, the download is right there. When the scan is done, close the tab. The browser doesn't remember the captures, because they were never persisted anywhere outside that single tab session.

The trade is honest: we're not as fancy as CamScanner. No automatic edge detection, no perspective deskew, no shadow removal, no filter that makes a phone photo look like a 1200dpi flatbed scan. You get the raw JPG of what the camera saw, framed how you framed it. That is the right tool for receipts, school permission slips, a page from a textbook, the back of a check, a meeting whiteboard. It is the wrong tool for archival document preservation or a legal-quality OCR pipeline — if that's the job, use a real scanner.

About CamScanner, specifically

This part should be on the page rather than buried in a footnote. CamScanner is the most widely-installed phone scanner in the world, with somewhere north of 500 million downloads. It has also been on the wrong side of several data-handling stories. In 2019 the Google Play version of the app was found shipping with a malware dropper called Necro inside one of its advertising SDKs; Google pulled it from the store until the SDK was removed. Independent security researchers have repeatedly flagged the volume of data the app sends back to its servers — including, depending on the version and the settings, the scans themselves.

You can read those stories yourself; this article isn't a takedown. The point is that "free phone scanner" has historically meant "the scanner is the product and you are the data." Even when CamScanner is behaving, the trust model is: you scan a contract or a passport, the bytes travel to a server in a jurisdiction whose data laws you may not have read, and you hope the company's privacy policy is honored. That's a reasonable trade if the document is a recipe. It's a worse trade if the document is a driver's license, a tax form, or a medical record.

Scan to PDF removes the trade. The bytes don't go anywhere. There's no privacy policy to honor because there's no path to honor it on.

How to scan a multi-page document

The common flow is a multi-page contract or a stack of receipts. Here's the exact sequence.

1. Open this page on your phone. Tap Open camera. Your browser shows a one-time permission prompt — tap Allow. iOS Safari and Chrome on Android both prefer the rear camera when the page asks for facingMode: 'environment', which we do.
2. Frame the first page in the live preview. Hold the phone roughly parallel to the page, far enough that all four corners are visible, close enough that the text reads sharply. Tap the round shutter button. The capture lands at the bottom of the page as a thumbnail.
3. Move to the next page. Frame, tap. Move to the next. Frame, tap. There's no limit to the number of captures — capture five pages, capture forty.
4. If a shot is blurry or the corners are cut off, tap the × on its thumbnail to drop it. If the captures got out of order, use the arrow buttons to reorder. The list order becomes the PDF page order.
5. Tap Build PDF. The captures get JPG-encoded at 92% quality and embedded into a single PDF at native pixel dimensions. Tap Download. The file is named scan-{today}.pdf — for example, scan-2026-05-13.pdf.

Total time for a five-page scan, end to end: under a minute. The bottleneck is your steadiness with the phone, not the software.

A worked example: scanning a six-page contract

A counterparty mailed you a paper consulting agreement and asked for it back signed. You don't own a scanner. You have a phone, a kitchen table, and a window with good light.

You open microapp.com/scan-to-pdf on your phone, lay the contract flat under the window, tap Open camera, allow. The live preview shows the first page. You hold the phone about twelve inches above the page, square the corners in the frame, tap the shutter. Thumbnail appears. You flip to page two, repeat. Pages three through six the same way — six shutter taps, six thumbnails.

You scroll through the thumbnail list. Page four came out tilted, so you tap × to drop it, lay the page flat again, retake. Now there are six thumbnails in correct order. You tap Build PDF. After about three seconds, the download button appears. The output is scan-2026-05-13.pdf, six pages, around 9 MB.

You AirDrop it to your laptop. Open it in Sign PDF, draw your signature on page six, download scan-2026-05-13-signed.pdf. Email it back. Round-trip from "this paper landed in the mailbox" to "I sent it back signed" is under five minutes, and not a single byte of that contract sat on a third-party server at any step.

The privacy story, in technical detail

If you want to verify the no-upload claim rather than take our word, here is exactly what happens when you scan.

• Open camera calls navigator.mediaDevices.getUserMedia({ video: { facingMode: 'environment' } }). The returned MediaStream is attached to an HTML video element. That stream lives entirely in your browser's process; the page's JavaScript has no API to send video frames anywhere.
• Shutter tap draws the current video frame to an HTML canvas at the video's native dimensions, then calls canvas.toBlob('image/jpeg', 0.92). The resulting JPG blob is stored in a JavaScript array in the tab's memory. It does not touch disk and it does not touch the network.
• Build PDF hands the array of JPG blobs to pdf-lib, which constructs the PDF in JavaScript memory and returns the bytes as another blob. The download link is a URL.createObjectURL(blob) pointing at that in-memory blob.
• Close tab: the blobs get garbage-collected. The camera stream releases. The thumbnails are gone from the DOM. There is no localStorage, no IndexedDB, no service worker, no background sync. Closing the tab is the deletion.

To audit any of this, open your browser's developer tools and watch the Network tab during a scan session. You'll see the page load, you'll see the JS bundle load, and then — through every camera open, every shutter tap, and the final Build PDF — zero outbound requests. The phone scanner that calls home does so on the network. The phone scanner that doesn't, doesn't.

When this is the right tool and when it isn't

Right tool:

• Expense receipts. The whole job is "turn six restaurant receipts into one PDF for the expense form." Sixty seconds, done.
• A page from a printed book or magazine for personal reference.
• Meeting whiteboards. Photograph the board at the end of the session, build a PDF, drop it in the meeting notes.
• School forms, permission slips, medical intake paperwork — fill out on paper, scan, email back.
• One-off scans where installing CamScanner for a single use feels absurd.
• Any document you'd rather not send to a third-party server because the document is private. Tax forms. Medical bills. Bank statements. IDs.

Wrong tool:

• Bulk scanning of hundreds of pages. A real scanner with a document feeder will be faster and cleaner.
• Scans that need OCR before they're useful (you want to search the text or copy paragraphs out). You can OCR our output afterward with Adobe Acrobat or Tesseract, but the workflow is heavier than starting with an OCR-native scanner app.
• Archival-quality reproduction of old photographs, fragile documents, or anything where you need 1200 dpi color-accurate output. Phone cameras are general-purpose; flatbed scanners exist for a reason.
• Perspective-corrected, edge-detected output that looks like a flat scan even when shot at an angle. CamScanner and Adobe Scan do that. We don't.

Tips for sharper captures

Most phone-scan quality is just lighting and framing. A few things that help.

• Use the rear camera, not the front. The rear sensor is much higher resolution. We default to the rear camera, but some browsers in some configurations default to whichever camera was used last — if the live preview looks low-res, check.
• Lay the page flat. A curled receipt or a creased contract page captures with distortion the tool can't correct. Press the page flat under a book if it won't lie still.
• Soft, even light beats bright spots. Direct sunlight on a glossy page creates glare that washes out text. Window light at a slight angle, or overhead room light, is usually fine.
• Frame square. Hold the phone parallel to the page rather than tilted. Tilted captures will save fine but will look skewed when assembled — there's no perspective correction.
• Steady hand or steady surface. The shutter freezes the current frame; motion blur at capture time can't be fixed after. If you're shooting handheld in low light, brace your elbows.

Where this fits in the rest of the PDF toolkit

Scan to PDF is the front door of a small workflow that lives entirely in the browser. Once you have the scan, you can:

• Sign PDF — sign the scanned contract without printing it.
• PDF Merger — combine the scan with a cover letter or context page.
• Compress PDF — phone scans can run 2-3 MB per page; compress before emailing to make file sizes friendlier.
• Crop PDF — trim the margins on a scan that captured too much table around the document.
• PDF Watermark — stamp CONFIDENTIAL or DRAFT on the scan before sending.
• Image to PDF — if you already photographed the document in your phone's camera app and want to assemble the JPGs into a PDF without re-shooting, that's the tool.

About Microapp

Microapp is a place on the internet where you can find every tool you need — small, focused tools that open in a tab, do their job, and let you leave. Scan to PDF is one of about 140 tools in the catalog. 10% of every dollar Microapp earns goes to charity, off the top, audited and published quarterly. The tools are picked, the quality is uniform, and the brand promise is the opposite of Big Software: open the tool, do the thing, leave.

Frequently asked questions

Does this work on a desktop or only on mobile?

It works anywhere the browser exposes a camera. On a phone — the intended case — it opens the rear camera and you tap to capture each page. On a laptop it'll use the built-in webcam, which is fine for a quick whiteboard snap but awkward for a sheet of paper. If you're on desktop, you'll usually be happier shooting photos with your phone and using our Image to PDF tool to combine them — same final PDF, more flexible workflow.

Why does my browser say camera permission denied?

When you tap Open camera the browser shows a one-time permission prompt. If you tapped Block (or did it accidentally), the prompt won't come back automatically — you have to grant permission manually in browser settings. Chrome: tap the lock icon in the address bar → Site settings → Camera → Allow. Safari iOS: Settings app → Safari → Camera → Allow. Then reload this page and tap Open camera again. We can't override that prompt; the browser owns the camera.

Can I scan multiple pages into one PDF?

Yes — that's the main use case. Capture page 1, reframe, capture page 2, repeat. Each tap of the shutter adds a page to the list. When you're done, tap Build PDF and all captured pages get assembled in list order into a single multi-page PDF. Use the arrow buttons to reorder if you captured out of sequence.

Is the quality as good as a dedicated scanner app like CamScanner or Adobe Scan?

Honest answer: no. CamScanner and Adobe Scan add automatic edge detection, perspective correction, contrast boosting, and shadow removal — all the tricks that make a phone photo look like a flatbed scan. We don't do any of that. You get the raw JPG of whatever the camera saw, assembled into a PDF. For casual scanning that's fine. For archival originals or legal-quality output, use a dedicated scanner.

Do my captures upload anywhere?

No. The video stream goes from getUserMedia directly into a canvas element in your tab — never through a network request. Each capture is a JPG blob held in browser memory. The PDF gets built locally with pdf-lib. Open your browser's network tab while you scan and you'll see zero outbound traffic during capture and build.

What resolution are the captured pages?

Whatever your camera natively delivers — typically 1280×720 on a basic laptop webcam, up to 1920×1080 or 4032×3024 on modern phones depending on the browser. The capture pulls the video frame at its full native size, encodes it as JPG at 92% quality, and embeds it in the PDF at original pixel dimensions. PDF pages are sharp but file sizes scale with the camera; a 10-page scan from an iPhone can run 15-20 MB.

Does it work on iPhone Safari?

Yes. Safari on iOS 14.3+ supports getUserMedia and prefers the rear camera when you pass facingMode: 'environment', which we do. The first time you tap Open camera, iOS asks permission via a system prompt. After that the camera should open every time. If you're in an in-app browser (Instagram, Facebook, etc.) some of them block camera access — open the page in real Safari if it won't start.

Can I edit the captured pages — crop, rotate, adjust contrast?

Not in this tool. Capture is one-tap-one-page and the page goes into the PDF as captured. If you need cropping or rotation, capture here, download the PDF, then run it through our Crop PDF or Rotate PDF tool. Or use Image to PDF and feed it photos you've already touched up.