Skip to content
PDFGem
Why Free?

Extract Compatible Raster Images from PDF as PNG

PDFGem

PDFGem extracts compatible raster images and re-exports them as new PNG files. It does not recover the original embedded JPEG, PNG, or other source file byte for byte, and it cannot promise to find every visual element on a PDF page.

Use Extract Images when you want separate photographs or bitmap graphics that the PDF stores as compatible image objects. If you need a picture of the complete page, including text and vectors, use PDF to PNG instead.

How PDF image extraction works

  1. Open Extract Images and select one PDF up to 100 MB.
  2. The browser scans the drawing operations on every page for compatible image XObjects and inline images.
  3. Supported pixel data is decoded and drawn to a canvas using the image’s stored width and height.
  4. A new PNG is created. The preview shows its pixel dimensions and the page where it was encountered.
  5. Download a result individually. If several images are found, you can also download a ZIP.

The processing happens locally in the browser. Your PDF is not sent to PDFGem for extraction. Available memory still matters because the source document, decoded pixels, previews, PNG blobs, and ZIP can exist in memory during the same session.

What the tool can find

The extractor is designed for compatible raster data: for example, a photograph, screenshot, scanned bitmap, or logo stored as a discrete image object. It also handles some inline images found directly in a page’s drawing instructions.

A visible element is not always a separate raster image. Charts may be drawn from vector lines and text. A logo may be built from paths. A background may be assembled from several objects, transparency masks, or layers. Those constructions cannot be assumed to appear as one downloadable image.

The tool also ignores images below 10×10 pixels and skips objects it cannot decode into supported bitmap data. A result count of zero means no compatible images were extracted; it does not prove that the page contains no visual content.

Stored dimensions are not an “original file” guarantee

The preview reports the width and height supplied by the decoded image object. The newly created PNG uses those dimensions. That is useful for knowing the pixel grid you received, but it is not a DPI claim and does not reveal the image’s effective size on the page.

PDFGem does not copy the original compressed stream out of the PDF. It decodes the pixels and encodes a new PNG through the browser canvas. The original filename, JPEG compression, metadata, color profile, and file bytes are not retained. Masks or other objects used by the PDF renderer may also be separate from the extracted result.

Extraction versus rendering the whole page

  • Extract Images: tries to recover compatible raster objects as separate PNG files.
  • PDF to PNG: renders the complete visible page, including text, vectors, backgrounds, and layout, as one raster image.

If an expected chart or composed graphic is missing, render the page with PDF to PNG. If you only need certain pages, use Split PDF first to reduce the document, then run extraction on the smaller PDF.

Open Extract Images, review the actual PNG previews, and keep the source PDF until you have confirmed that the results contain what you need.

Frequently Asked Questions

What images can PDFGem extract?

The tool looks for compatible raster image XObjects and some inline raster images in each page’s drawing operations. It does not extract vector paths, text, or every visual composition as a standalone image.

Is the downloaded PNG the original embedded file?

No. PDFGem decodes compatible image pixels, draws them to a canvas at the stored pixel dimensions, and creates a new PNG. It does not return the original PDF stream, source filename, compression, metadata, color profile, or byte-identical file.

Does the tool find every image visible in the PDF?

Not necessarily. It skips images smaller than 10 by 10 pixels and anything that cannot be decoded in a supported form. Vectors, masks, layered compositions, page backgrounds, and unsupported PDF constructions may not appear as separate results.

Can I download images individually or as a ZIP?

Yes. Every result has an individual download button. When more than one image is found, the main download creates a ZIP; when only one is found, it downloads that PNG directly.

What limits apply to large PDFs?

The input file limit is 100 MB. Scanning, decoding images, keeping previews, and creating a ZIP all use browser memory, so a large or image-heavy PDF may fail on a constrained device.