PDFs often hide useful assets inside the file: product photos, charts exported as bitmaps, logos, or scanned figures. Designers and archivists sometimes need those images at their original embedded resolution — not a screenshot of the entire page. Online extractors usually require an upload. For catalogs with unpublished imagery or reports with confidential charts, that upload is avoidable.
LoveMyFile's Extract Images tool uses pdfjs-dist in the browser to scan the PDF for embedded image resources, show them in a gallery preview, and let you download selections or a ZIP. It is not the same as converting every page to JPG. Extraction targets image objects already stored in the document; page rasterization paints the whole visible page.
What image extraction does — and when you need it
Embedded images are XObject (or similar) resources inside the PDF. pdfjs-dist walks the document structure, finds those bitmap resources, and decodes them into standalone files you can preview and save. Rasterizing a page with PDF to Images instead paints text, vectors, and photos together into a new bitmap of the page as shown. Choose extraction when you want the original embedded asset; choose page rasterization when you want a picture of the layout.
- Reuse figures — pull a chart or photo for a slide without re-exporting from the authoring app.
- Asset audits — list what bitmaps a vendor PDF actually contains before you approve a delivery.
- Archival recovery — recover images from a pack when the source files are gone but the PDF remains.
- Private catalogs — unpublished product photography should not hit a free online extractor on the way to your disk.
If your goal is a shareable page snapshot (including text as pixels), read the PDF to JPG conversion guide and use PDF to Images instead. Extraction and rasterization solve different problems; mixing them up wastes time and produces the wrong files.
Upload extractor vs browser-only extractor
| Factor | Typical online extractor | Browser-only (LoveMyFile) |
|---|---|---|
| Where files go | Vendor servers | Stay in browser memory |
| Retention | Minutes to hours (claimed) | None — no server copy |
| How to verify | Trust the privacy policy | Network tab shows no PDF upload |
| Cost / account | Often freemium, daily caps | Free, no account, no watermark |
| Best for | Public marketing PDFs | Private catalogs and reports |
How it works locally
The tab loads your PDF and pdfjs-dist inspects its image resources. Embedded bitmaps are decoded and shown in a gallery so you can judge which assets are useful before downloading. You can grab individual images or package many of them as a ZIP built in the browser. Processing stays local with open-source PDF libraries — no paid extraction API and no temporary cloud copy of the document. Static JS/CSS may still load from the CDN; the sensitive signal is whether your PDF bytes are posted. Confirm with how to verify a PDF tool does not upload.
Steps and tips
- Open Extract Images and add your PDF to the dropzone.
- Run extraction and review the gallery of found images.
- Download the assets you need individually, or take the ZIP when you want the whole batch.
- Discard duplicates or tiny decorative bits; keep only the figures you will reuse.
- If you needed a full-page picture instead, switch to PDF to Images.
Tip: some "images" in a PDF are tiny icons or repeated backgrounds. Expect more files than the count of photos you see visually. Vector artwork (paths, not bitmaps) will not appear as extractable photos. Gallery preview is the fastest way to filter noise before you fill a folder with one-pixel sprites.
Limits unique to image extraction
- Not page rasterization— you will not get a JPG of every page's full layout unless images fill those pages. Use PDF to Images for that workflow.
- Vectors stay vectors — logos drawn as paths are not bitmap extracts; they never existed as JPEG/PNG objects inside the PDF.
- Compression artifacts — you get what was embedded (JPEG artifacts included); extraction does not magically enhance quality.
- Encrypted PDFs — unlock first when you have the right to do so, then extract.
- Browser memory — image-heavy magazines can produce large galleries; prefer desktop Chrome/Edge for big ZIPs.
Common mistakes
- Expecting one image per page — pages can contain many small resources.
- Using extract when you wanted a screenshot of the page including text — that is rasterization, not extraction.
- Uploading a confidential annual report just to grab one chart — local extraction with gallery preview avoids that leak.
- Assuming missing photos were "deleted" by the tool — they may never have been embedded as bitmaps.
- Downloading every resource without skimming the gallery — you end up with icon noise and duplicate backgrounds.
Related tools and bottom line
Rebuild a PDF from recovered photos with Images to PDF, or capture whole pages with PDF to Images when extraction is the wrong mental model. Trim a source first with Crop PDF if you only care about a region that contains the figure you need.
Bottom line: pdfjs-dist can scan a PDF for embedded images, preview them in a gallery, and ZIP the keepers — all without uploading. Pulling embedded images in the browser keeps catalogs, drafts, and research packs private while still giving you the bitmaps you actually need. Do not confuse that with full-page rasterization; pick the tool that matches the asset you want.