Remove supported embedded image objects
Remove Images from PDF
Create a new PDF without referenced image XObjects. Original page content streams remain in place, preserving text and vector artwork where present.
PDF page preview
This removes image XObject streams referenced by the PDF object graph. Inline images encoded directly inside content streams, external resources, and unusual malformed structures may not be detected. Output is refused if image streams remain after processing. Image-only pages can become blank.
How to remove embedded images from a PDF
- Select a PDF and review the first pages in the preview.
- Remove the supported image objects. If no removable images exist, no output file is created.
- The output is reopened with PDF-Lib and PDF.js; the page count is checked and remaining image streams prevent download.
What remains and limitations
PDF page content streams, including text and vector drawing operators, are not rasterized or rewritten. Image XObject references are removed from PDF dictionaries. Inline images inside content streams cannot be safely identified by this implementation and may remain; output validation checks indirect image XObject streams, not every possible image representation. Annotations, signatures, attachments, and metadata are not intentionally removed. Modifying a digitally signed PDF may invalidate its signature. Encrypted PDFs unsupported by PDF-Lib cannot be processed.
Related tools
Frequently asked questions
Will this erase text that happens to look like an image?
Text and vector drawing operators are preserved. Image XObjects are removed, so page backgrounds, scans, or image-only page content may disappear.
Does this securely sanitize the PDF?
No. The tool removes supported image objects only and is not a general PDF sanitizer or secure-redaction tool.