Deleting an image from a PDF in Adobe Acrobat Pro is done from the Edit PDF panel, where you select a placed picture and choose Delete; the browser-based Remove Images from PDF tool applies the same idea to every eligible embedded JPG in a single local pass. Adobe's desktop product is documented for rotating, moving, and deleting individual pictures one click at a time, but a single-image workflow is slow when a document has dozens of photos or scans on every page. A no-upload tool that runs entirely in your browser can replace each eligible embedded JPG with a 1 by 1 white JPEG while keeping the page drawing command and its resource reference intact, so the file stays valid and the rest of the document — text, links, page geometry, vector content — remains unchanged. This article walks through both routes: the quick Edit PDF flow inside Acrobat Pro for one-off deletions, and a local browser method that handles every eligible image at once without sending the file to a server.
What Adobe Acrobat Pro Offers for Deleting Images
Adobe's help pages for Acrobat Pro describe image deletion as a point-and-click action inside the Edit PDF workspace. You open the file, switch into Edit mode, click on a picture to select it, and either drag a handle to resize, rotate from the context menu, crop with the cropping handles, or pick Delete to remove the picture from that page. The same panel also lets you move or replace the image. The change applies only to the single object you clicked, which is fast for one or two pictures but requires repeated clicking and saving for a document with many embedded photographs or scans.
Acrobat Pro can handle most image encodings because it is a full PDF editor. CMYK JPEGs, JPX (JPEG 2000), JBIG2, images with soft masks, and indexed-color images all behave normally during a Delete action, because Acrobat simply removes the picture object from the page. The browser-based alternative described below is narrower: it only handles a specific subset of embedded JPGs, and it refuses the whole file if any single image falls outside that subset.
Both routes preserve the rest of the page. Acrobat Pro's Delete keeps text, links, form fields, and other vector content on the edited page; the local browser tool keeps the original resource reference and only swaps the image stream, which has the same effect for ordinary pages where images are placed as separate XObjects.
Why a Local Browser Tool Replaces the Acrobat Pro Image Step
There are three situations where opening Acrobat Pro just to remove a few pictures is heavier than the task itself: you do not have a paid license on the machine you are using, the PDF lives on a device where installing desktop software is not allowed, or the document contains dozens of identical-format embedded JPGs and clicking each one in turn would take a long time. A browser-only method that processes every eligible embedded image in one pass covers all three cases without changing the end result on screen.
The Remove Images from PDF tool is built for this exact task. It runs locally: you choose one PDF up to 25 MB from your device, the tool enumerates every indirect image XObject inside the file, and it checks each one against a strict compatibility rule. If every image passes, the tool generates a tiny 1 by 1 white JPEG in the browser, assigns it to each existing reference as a PDFRawStream, and saves a new PDF that opens cleanly in the same viewer used by other PDF tools on the site. No file is uploaded to a server at any point.
Two design choices make the output safer than a naive "strip image XObjects" approach. First, the tool does not delete the page resource name, because removing an object name while leaving the page's drawing command behind can produce a damaged PDF that some viewers refuse to open. Second, instead of producing a partial file, the tool refuses the entire job if even one image uses an encoding outside the supported subset. You never end up with a downloaded PDF whose button says images were removed while a complex image was silently left in place.
How to Delete an Image From a PDF in the Local Tool
Run the whole job from a single browser tab. No install, no account, no upload.
- Open the Remove Images from PDF page in your browser and choose one local PDF up to 25 MB from your device.
- Click Remove eligible images to start the all-or-nothing compatibility check. The tool inspects every indirect image XObject in the file before deciding whether to produce output.
- If the check passes, the tool generates a browser-local 1 by 1 white JPEG, swaps each eligible image stream for it while keeping the original resource reference, and saves the result as a new PDF on your device.
- The tool then reopens the saved PDF with the same PDF editor library used to write it and again with the browser's PDF viewer to confirm the page count and image-object counts match the source.
- Inspect the new PDF by scrolling through every page. Verify that text, links, page geometry, and non-image vector content look the same and that the picture areas are now blank.
- Download the new PDF only after the inspection confirms the white replacements are appropriate for your use case.
If you prefer the Acrobat Pro route for a single image, the equivalent sequence inside the desktop app is: open the PDF, choose Edit PDF from the right-hand panel, click the picture to select it, pick Delete from the context menu, and save the file. Repeat for every other picture you want to remove, then re-save once.
What Stays the Same vs. What Changes
The table below compares the two methods against the same task: removing every eligible embedded picture from a multi-page PDF while keeping the rest of the document intact.
| Aspect | Acrobat Pro Edit PDF | Remove Images from PDF (browser) |
|---|---|---|
| Where it runs | Desktop application | Browser tab, fully local |
| File upload | Not required for local files | None — file stays on your device |
| Granularity | One picture at a time | Every eligible image in one pass |
| Image encodings accepted | Most PDF image types | Unmasked 8-bit DCTDecode DeviceRGB or DeviceGray JPEGs only |
| Behavior with unsupported images | Edits what you select; ignores the rest | Refuses the entire file, saves no output |
| Text, links, form fields, vectors | Preserved on the edited page | Preserved across the document |
| Page geometry | Preserved | Preserved |
| Legal-grade redaction | No | No — white replacement only |
| Encrypted or signed PDFs | Handled by Acrobat's own auth flow | May be rejected; tool does not bypass passwords or signatures |
Limits and Edge Cases Worth Knowing
Knowing which images the browser tool will accept up front saves a failed run. The table below shows which embedded image types pass the compatibility check and which trigger an all-or-nothing refusal.
| Embedded image type | Accepted by the tool? |
|---|---|
| Unmasked 8-bit DCTDecode JPEG, DeviceRGB, no DecodeParms | Yes — replaced with 1 by 1 white JPEG |
| Unmasked 8-bit DCTDecode JPEG, DeviceGray, no DecodeParms | Yes — replaced with 1 by 1 white JPEG |
| JPEG with a soft mask or stencil mask | No — entire file refused |
| CMYK JPEG | No — entire file refused |
| JPXDecode (JPEG 2000) | No — entire file refused |
| JBIG2 | No — entire file refused |
| LZW or FlateDecode (e.g., PNG-style streams) | No — entire file refused |
| Indexed-color images | No — entire file refused |
| Inline images (in the content stream, not as XObjects) | Not claimed as removable by this version |
| Encrypted, malformed, or signed PDFs | May be rejected; tool does not bypass passwords or signatures |
Because the replacement is a 1 by 1 white JPEG, the visible result is a small white square wherever an eligible image used to sit, unless other page content covers that area. Anything drawn on top of the picture in the source — annotation boxes, form fields, vector shapes — is unchanged, so the page layout around the image is preserved. Original document metadata, XMP history, and any inaccessible prior revisions are not cleared by this tool; if those matter to your workflow, use a dedicated redaction process instead.
For a deeper walkthrough of the local Acrobat route and what to keep an eye on, the guide How to Remove Images From a PDF With Acrobat (Locally) covers the same idea in a slightly different order.
For the underlying mechanics, the tool's image-stream swap uses a PDFRawStream object as defined in pdf-lib's source, and the verification step relies on the same PDF.js viewer that the site's other conversion tools use to render output.
Keep the source PDF until you have inspected every page of the output. A white replacement is a visual blank, not a guaranteed privacy-grade erasure, so for legal, security, or personal-data workflows, run a proper redaction pass with a dedicated tool before sharing.
If you're weighing options, Delete Pages from PDF on Android Without an App covers this in detail.