A PNG is pixels. The words in it look like text to you, but to your computer they’re just colored dots — so you can’t search or copy them. To make PNGs searchable, you run OCR on them and save the result as a PDF with a text layer.
This guide is for when you already have a set of PNG files — screenshots, exported diagrams, saved images — and want them in one searchable PDF. It covers preparing the images, three free ways to convert them, and the PNG-specific problems that trip people up.
PNG vs searchable PDF
- PNG: an image. No searching, no selecting, no indexing.
- Searchable PDF: looks identical, but has real text behind the image.
- Result: Ctrl-F, copy, and index every word.
PNG is actually a good starting point for OCR. It’s lossless, so letter edges stay sharp, unlike heavily compressed JPEGs where text gets fuzzy halos.
Prepare the PNGs first
- Order: rename files with zero-padded numbers (img-001.png, img-002.png) so every tool sorts them the same way.
- Crop: remove taskbars, browser chrome, and side panels. Stray UI text ends up in your OCR results and in search hits.
- Size: don’t shrink images to save space. Small text is the number one cause of bad OCR. If text in the image is tiny, recapturing it larger works better than any setting.
- Rotation: make sure everything is upright before OCR.
Three free ways to convert PNG to searchable PDF
1. NAPS2 (easiest)
- Open NAPS2 and import your PNG files.
- Drag the thumbnails into the right order if needed.
- Enable OCR and pick the language.
- Save as PDF. Each PNG becomes a page with text behind it.
2. Tesseract (one image at a time)
Tesseract is the open-source OCR engine many other tools are built on. Its command-line program can write a searchable PDF directly from an image: you give it the PNG, an output name, and “pdf” as the output format. That produces one searchable PDF per image, which you’d then merge. It’s best when you want precise control or are converting a handful of files.
3. img2pdf + OCRmyPDF (best for big batches)
img2pdf combines a folder of PNGs into one PDF without re-compressing them. OCRmyPDF then adds the text layer to that PDF. Together they handle hundreds of images in one go and keep the original image quality.
PNG-specific pitfalls
- Transparency. PNGs can have transparent backgrounds. Some converters render transparency as black, which makes dark text disappear. If pages come out black, flatten the images onto a white background first.
- Dark mode screenshots. Light text on a dark background can recognize less reliably than dark on light. If results are poor, inverting the colors before OCR can help.
- Retina or high-DPI captures are large in pixels but may be marked with a different DPI. That can make PDF pages appear huge or tiny. It doesn’t affect searchability, but check the page size if you plan to print.
- Mixed sizes. A tall scrolling screenshot next to a small crop gives a PDF with very uneven pages. Group similar images into separate PDFs if that bothers you.
CapDrop does this whole chain in one pass on Windows: pick a region and a page key, set an interval, and walk away — it captures each page, auto-crops, runs OCR locally, and binds everything into one searchable PDF. 100% local, for content you own, 7-day free trial. Try a real searchable sample — Ctrl-F it yourself.
If you’re capturing the pages fresh rather than starting from existing PNGs, doing the capture, crop, OCR, and bind in one pass skips the renaming, cropping, and converting steps above.
How to check the text layer
- Open the PDF in any reader (Edge, Chrome, or a dedicated PDF app).
- Press Ctrl-F and search for a word from one of the last images, not just the first.
- Drag the cursor over a sentence: it should highlight word by word.
- Copy a passage into Notepad to see the exact OCR output and spot misread characters.
FAQ
Can I convert PNG to searchable PDF online? Many sites offer it, but they upload your images to process them. For screenshots of private messages, account pages, or work documents, a local tool is the safer choice.
Does the same process work for JPG? Yes. Everything above applies to JPEG files too. Just avoid heavily compressed JPEGs if you have the option, because compression blurs letter edges.
Can I get a Word document instead? OCR tools can also export plain text, and some can create editable documents. But layout is rarely preserved perfectly. A searchable PDF keeps the exact look while still letting you search and copy.
Why do some words come out wrong? Usually small text, low contrast, or unusual fonts. Recapturing at a larger size fixes most of it.
Should I delete the PNGs afterwards? Keep them until you’ve checked the PDF. If a page needs fixing later, rebuilding from the originals is much easier than editing the PDF.