Free · Fast · Privacy-first

Split Scanned PDF Documents

Scanned PDFs are typically image-based documents where every page is essentially a high-resolution photograph of a physical paper original.

⚡

Preserves original scan resolution

🔒

No image re-compression

✨

Works on image-based PDFs

⚡

Free, no sign-up

Cost: Free forever
Sign-up: Not required
Processing: In your browser
Privacy: Files stay local

FreeNo signupWhite-label

Add this PDF Splitter to your website

Drop the PDF Splitter into any page — blog post, product docs, intranet, school portal — with a single line of HTML. Your visitors get the full tool, processed entirely in their browser. No backend, no uploads, no signup.

Files stay 100% in the visitor's browser
Responsive — adapts to any container width
Free forever, no API key needed

Embed code

<iframe
  src="https://www.fixtools.io/pdf/pdf-splitter?embed=1"
  width="100%"
  height="780"
  frameborder="0"
  style="border:0;border-radius:16px;max-width:900px;"
  title="PDF Splitter by FixTools"
  loading="lazy"
  allow="clipboard-write"
></iframe>

<iframe
  src="https://www.fixtools.io/pdf/pdf-splitter?embed=1"
  width="100%"
  height="780"
  frameborder="0"
  style="border:0;border-radius:16px;max-width:900px;"
  title="PDF Splitter by FixTools"
  loading="lazy"
  allow="clipboard-write"
></iframe>

Attribution-friendly: a small "Powered by FixTools" link appears in the embed footer.

Image-heavy PDFs, scanned document structure, and resolution preservation during splitting

A scanned PDF differs fundamentally from a text-based PDF in its internal structure, and understanding the difference helps explain how the splitter handles each type. A text PDF contains vector fonts, structured text strings positioned at specific coordinates on the page, and compressed image objects for any embedded photographs that appear alongside the text. A scanned PDF, by contrast, typically contains one raster image object per page, encoded as a JPEG photograph for color or grayscale scans, a JPEG2000 image for higher-compression archival scans, or a CCITT Group 4 compressed bitmap for black and white document scans. The text visible in the scanned image is not selectable or searchable in the file unless optical character recognition has been applied separately to embed a text layer behind the image. When FixTools splits a scanned PDF, it is separating those page-sized image objects into new PDF containers with no text processing or font subsetting required, just clean image extraction.

Resolution is the single most important quality parameter for scanned PDFs because it determines how much detail the digital image can preserve from the physical original. Standard office scanning produces 200 to 300 DPI for everyday documents and 600 DPI for signature pages or fine-detail pages where line work and small text matter. Legal and archival scanning programs often specify 300 DPI color or 400 DPI grayscale as their baseline standard for evidentiary or compliance reasons. A single 300 DPI color letter-size page measuring 8.5 by 11 inches contains 3300 by 2550 pixels of image data and typically compresses to between 300 kilobytes and 1.5 megabytes as JPEG depending on the complexity of the page content. When FixTools extracts that page into a split output file, it copies the image data byte-for-byte from the source PDF without re-encoding the image, so the output has exactly the same pixel dimensions, DPI metadata, and compression format as the original scan.

For high-volume scanned document archives such as medical records, legal discovery productions, insurance claim files, and government records management systems, splitting is often a necessary prerequisite step before downstream OCR processing can run efficiently. Many OCR systems perform better on smaller input files and can be configured to process a batch of individual page PDFs in parallel rather than one large multi-page scan in a single sequential job. Splitting a 200-page scanned record into 200 individual page PDFs before running OCR allows the work to be parallelized across multiple CPU cores or worker processes, and it makes it considerably easier to identify and reprocess specific pages where OCR accuracy is low without having to rerun the whole document. After OCR completes on each page, the pages can optionally be merged back together into a single searchable document using the FixTools PDF Merger.

A useful detail about scanned PDF structure worth knowing is how the image objects relate to the page tree. Each page in a scanned PDF references one or more XObject image resources defined in the document's shared resource dictionary. When FixTools splits the document, it copies the page tree node along with the specific XObject references that page uses, and it includes only those referenced XObjects in the output file. Shared resources that happen to be referenced by other pages outside the split range are not copied into the output, which keeps the resulting file sizes as small as possible without duplicating data. This is especially helpful for scanned documents where image XObjects are typically the largest objects in the file and account for the vast majority of overall file size.

CCITT Group 4 compression standard for scanned documents JPEG compression specification (ISO 10918)PDF reference: image XObjects and raster data

How to use this tool

💡

Upload your scanned PDF and choose how to split it. FixTools extracts the scanned image pages without altering their resolution or quality.

How It Works

Step-by-step guide to split scanned pdf documents:

1
Upload your scanned PDF
Open the PDF Splitter in your browser and upload your scanned PDF by either dragging it onto the upload area or using the file picker. Scanned files are often large because of the embedded image data, so allow extra time for loading on slower connections or for parsing large documents in the browser. The total page count and approximate file size appear after the initial parse step.
2
Preview page thumbnails
The tool shows thumbnail previews of each scanned page so you can visually identify the natural split points in the document. Boundaries between distinct documents within a bundle often show up as blank separator pages, page numbering resets, or visibly different header formatting. Scrolling through the thumbnails quickly is the fastest way to map the structure of a multi-document scan.
3
Set your split parameters
Enter a page range or select multiple ranges to extract from the scanned document depending on your needs. For a bundle of separate documents, define one range per logical document so each becomes its own output file. For a single long document where you only need a section, enter a single range covering the pages of interest and ignore the rest of the file.
4
Split and download
Click the Split PDF button and download the resulting scanned-page PDFs from the links that appear after processing completes. The output files retain the original scan resolution byte-for-byte without any re-compression. Save them under descriptive names that reflect their content rather than the default sequential numbers, especially for bundles where each output represents a distinct document.

Real-world examples

Common situations where this approach makes a real difference:

Medical records coordinator

A hospital records coordinator receives a 120-page scanned patient chart bundle as a single consolidated PDF from a referring clinic that scanned its paper records in bulk. She needs to split the bundle into 6 individual patient charts of approximately 20 pages each based on the blank separator sheets the referring clinic inserted between patients. She uses FixTools to define 6 page ranges identified by visually scanning the thumbnail preview for blank separator pages, producing 6 standalone scanned PDFs at the original 300 DPI quality ready for upload to the hospital electronic health record system without any loss of evidentiary integrity.

Law firm paralegal

A litigation paralegal at a commercial law firm receives a 450-page discovery production PDF from opposing counsel consisting entirely of scanned correspondence files. She needs to split the production into individual document files where each scanned letter is 2 to 8 pages long, for upload to the firm's e-discovery review platform where each document gets a separate Bates number assignment. She identifies the document boundaries by reviewing thumbnails and defines page ranges for each individual letter. The FixTools split preserves the original scan quality required for court admissibility under the relevant federal rules of evidence.

Insurance claims processor

An insurance claims processor at a regional carrier has a 60-page scanned claim file submitted by a customer that needs to be split into separate components for routing to different internal departments. The claim form lives on pages 1 through 4 for the underwriting team, the attached police report runs on pages 5 through 12 for the special investigations unit, repair estimates take pages 13 through 28 for the appraiser, and damage photos occupy pages 29 through 60 for the field adjuster. She defines 4 ranges in FixTools in a single operation, and the image quality of the scanned damage photos is preserved exactly at the original 200 DPI JPEG compression.

Library archivist

A university library archivist has a 180-page scanned historical manuscript file weighing 145 megabytes at 400 DPI grayscale that needs to be divided into 4 thematic sections for separate online exhibition pages on the library digital collections website. She splits the manuscript in FixTools running in Chrome on a desktop workstation with 16 gigabytes of RAM, which is needed for the large image data. Each output section PDF retains the full 400 DPI grayscale resolution required to meet archival quality standards for digitized historical materials. Total processing time is about thirty-five seconds end to end.

Pro tips

Get better results with these expert suggestions:

Use desktop for scanned PDFs over 80MB

Scanned PDFs at 300 DPI color average between 1 and 3 megabytes per page in JPEG-compressed form depending on content complexity. A 100-page scanned file therefore typically weighs between 100 and 300 megabytes on disk. Processing a file of that size in a browser requires roughly three to four times the file size in active JavaScript heap memory for parsing and output construction, which works out to 300 megabytes to 1.2 gigabytes of working memory. On a desktop with 8 gigabytes of RAM or more, this is manageable. On a phone with 4 gigabytes of RAM, the browser tab will likely crash before completing the split.

Identify document boundaries using thumbnail view before splitting

For scanned bundles containing multiple documents merged into one file, use the thumbnail preview to visually identify where one document ends and the next begins before defining any ranges. Common visual boundary indicators include blank separator pages inserted intentionally by the scanning operator, distinctive document header pages that look different from surrounding content, and visible page number resets where a printed page footer suddenly restarts at one. Note the page numbers of each boundary before defining your ranges. This visual review step prevents range errors that would produce output files spanning two unrelated documents.

Compress output files after splitting for sharing

Scanned PDF sections retain the full resolution of the original scan after splitting, which means a 20-page section extracted from a 300 DPI color scanned document may still weigh between 20 and 40 megabytes by itself. Before emailing the section to a recipient with an attachment size limit or uploading it to a web system with a per-file size cap, run each section output file through the FixTools PDF Compressor as a follow-up step. For 300 DPI color scans, medium compression typically reduces file size by 70 to 80 percent with no visible quality change at normal reading magnification on a computer screen.

Keep originals before splitting for OCR pre-processing

If you plan to run OCR on the split sections after the splitting step, archive the original unsplit scanned PDF first in a known location before performing the split. OCR quality is sometimes higher when the engine has access to the full multi-page document at original resolution because it can use cross-page context to disambiguate ambiguous characters. Some OCR engines can process multi-page PDFs directly without needing single-page splits. Splitting into individual pages before OCR is useful for parallel batch processing across many cores, but always keep the original intact multi-page scan in case OCR needs to be rerun on specific pages later.

FAQ

Frequently asked questions

Yes, FixTools handles both text-based PDFs and image-based scanned PDFs using the same underlying splitting workflow. A scanned PDF typically contains one raster image object per page instead of structured text strings and vector graphics. FixTools reads the page tree structure of the PDF and copies each page image object along with any referenced resources into the appropriate output file without re-processing, re-rasterizing, or re-encoding the image data along the way. The scanned content of each page is preserved exactly as it appears in the original source document, which is essential for archival, legal, and medical use cases where bit-perfect image integrity is a hard requirement.

No, FixTools copies the image data from each scanned page directly into the split output file without re-encoding the image bytes in any way. A page that was scanned at 300 DPI in the original source PDF will be a 300 DPI page in the split output with identical pixel content. The pixel dimensions, DPI metadata, color depth, and compression format whether JPEG, JPEG2000 for archival scans, or CCITT Group 4 for black and white documents are all preserved byte-for-byte through the operation. There is no quality loss between source and output for the pages that you choose to extract into the split file regardless of the resolution.

Each page of a scanned PDF contains a raster image, which is a grid of colored or grayscale pixels covering the full page area at the scanned resolution. A standard letter-size page scanned at 300 DPI in color contains 3300 by 2550 pixels equal to about 8.4 million pixels of color data per page. Even with JPEG compression, a single such page typically takes between 300 kilobytes and 1.5 megabytes of storage in the file. A 100-page scanned document at 300 DPI color is therefore typically between 30 and 150 megabytes total. A text-only PDF of the same 100 pages might be only 500 kilobytes to 3 megabytes because text is stored compactly as character codes and font vectors rather than full pixel grids.

Splitting itself does not add or remove OCR text layers from the output files. If the original scanned PDF had an OCR text layer that was added by a scanner with built-in OCR capability or by software such as Adobe Acrobat Pro, ABBYY FineReader, or Google Drive's automatic OCR pipeline, that text layer is preserved in each split section for the pages where it was present in the source. If the original document had no text layer at all because the scan was never run through OCR, the split output files are also image-only and not searchable in any PDF viewer. To make scanned PDFs searchable after splitting, run OCR on the output files separately using any dedicated OCR tool of your choice.

Use the FixTools PDF Compressor as a follow-up step after splitting to reduce file size. The compressor reduces overall file size by re-encoding the embedded raster images at a lower JPEG quality setting that produces smaller output bytes at the cost of some image fidelity. For 300 DPI color scans, medium compression typically reduces file size by between 60 and 80 percent while maintaining acceptable visual quality at normal reading magnification on a computer screen. High compression produces even smaller files but may show visible JPEG artifacts on finely detailed text or on handwritten signatures, so use it cautiously when image quality matters.

There is no hard upper file size limit coded into FixTools itself, but the practical maximum depends on your specific device's available RAM and the number of other applications competing for memory at the time. A 100-page scanned PDF at 300 DPI color is approximately 100 to 200 megabytes on disk. Processing a file of that size in a browser requires roughly three to four times its file size in active JavaScript memory, which works out to between 300 and 800 megabytes of working memory during the split. On a modern desktop with 8 to 16 gigabytes of RAM, this is manageable in under a minute. On a phone with 3 to 4 gigabytes of RAM, files over 50 to 80 megabytes may cause the browser tab to crash.

Yes, PDFs produced by office multifunction printers from manufacturers such as Canon, HP, Ricoh, Konica Minolta, Sharp, and Xerox are all standard ISO 32000 compliant PDF files containing image XObjects with the scanned page data inside. FixTools handles these identically to any other image-based PDF regardless of which manufacturer created them. The only potential complication arises if a specific printer model uses a proprietary PDF variant or applies device-specific encryption to its scanned output. In those rare cases, you may need to print the file to a clean PDF first using a virtual PDF printer to normalize the file format before splitting.

Most scanned PDFs do not have accessibility features such as tagged PDF structure, alt text descriptions for images, or specified reading order because the bulk scanning process typically does not add this metadata automatically. If the original scanned PDF had accessibility tags added later either manually by a remediation team or by post-scan processing software, FixTools preserves those tags in the split output files for the specific pages that are included in each split segment. Tags that reference pages not included in a particular split segment are necessarily excluded from that output file, since the destinations they point to are no longer present in the same document.

Yes, this is one of the most common use cases for the scanned PDF splitter. Open the source bundle in the splitter, use the thumbnail preview to identify the page boundaries between separate documents inside the bundle, then define one page range per logical document and run the split. The tool produces one output file per range in a single zip archive ready for download. Bulk-scanned bundles from legal discovery productions, medical records imports, and government records transfers all benefit from this approach. After downloading, rename each output file to reflect the actual document content rather than the default sequential numbers.

Yes, scanned PDFs created by mobile scanning apps such as CamScanner, Adobe Scan, Microsoft Lens, Apple Notes built-in scanner, and Genius Scan all produce standard PDF files with image XObjects per page that FixTools splits the same way as scans from any other source. The image resolution from mobile scanning is typically lower than from a flatbed scanner because phone cameras have practical limits on capture detail, but the splitting operation works identically regardless of capture device. Mobile scans are often smaller files because of the lower resolution, which means they process faster in the browser than equivalent flatbed scans.