🖼️ Extract Images from PDF Client

Original images. We read the actual embedded image bytes from the PDF's object stream — the exact data the PDF uses to render those images. If a photograph was 3000×2000 when the PDF was created, you get a 3000×2000 file out. We never screenshot pages. Contrast this with a "PDF to JPG" tool, which rasterizes the whole page including text and backgrounds.

Four common reasons:

• The PDF is text-only — no embedded images to extract.
• Images are smaller than your Min Width/Height filter. Try lowering to 10 or 20 px.
• Images are vector graphics (charts drawn with lines and fills) rather than raster images. Vectors aren't images — they can't be extracted as PNG/JPG without rendering.
• Rare image formats like CMYK JPEGs or CCITT fax may fail to decode.

In many PDFs, the same image — a company logo, a watermark, a page header — is used on every page. Without dedup you'd get 50 identical logos. With dedup ON, we compute a hash of each image's data and keep only one copy. Note this only catches byte-identical images; visually similar but re-encoded versions will still appear separately.

PNG is lossless and preserves transparency — best for logos, diagrams, and screenshots. JPG is smaller but lossy and doesn't support transparency (we flatten to white or black). WebP is modern, small, and supports transparency — great if your target supports it. If in doubt, use PNG.

No. We extract only the embedded images — not the page rendering. So if your PDF has a photo with a caption "Figure 1" below it, you get the photo alone, not the caption. If you need whole-page images, use PDF to JPG instead.

No. Everything runs in your browser using pdf.js and JSZip. Your file never leaves your device.