Merge JPG Images into PDF on Linux

Converting a series of JPG images into a single PDF is useful for creating documents, presentations, or archiving scanned photos. Linux offers several tools for this task, each with different trade-offs between simplicity, quality preservation, and features.

Using img2pdf (Recommended)

For most cases, img2pdf is the best choice because it merges images without re-encoding or re-compressing them, preserving original quality:

img2pdf *.jpg -o output.pdf

To control the order of images, specify them explicitly:

img2pdf image1.jpg image2.jpg image3.jpg -o output.pdf

You can also set page size and margins:

img2pdf *.jpg -o output.pdf --imgsize 210x297mm --margin 10

The --imgsize flag accepts common sizes like A4, letter, or custom dimensions. If your images are already the right size, you can fit them to the page:

img2pdf *.jpg -o output.pdf --fit shrink

Install img2pdf on Debian/Ubuntu:

sudo apt install img2pdf

On RHEL/CentOS/Fedora:

sudo dnf install img2pdf

Using ImageMagick

ImageMagick’s convert command works but re-compresses images, reducing quality:

convert *.jpg output.pdf

This approach is slower and larger in file size compared to img2pdf. However, ImageMagick is useful if you need to manipulate images during conversion (resize, rotate, adjust colors):

convert *.jpg -resize 50% -quality 85 output.pdf

Using Ghostscript

For more control over PDF compression and settings, use Ghostscript:

convert *.jpg output.pdf && gs -sDEVICE=pdfwrite -dCompatibilityLevel=1.4 \
  -dPDFSETTINGS=/ebook -o compressed.pdf output.pdf

Using PDFtk (for manipulation after creation)

If you need to merge PDFs that you’ve already created, or reorder pages:

pdftk page1.pdf page2.pdf output merged.pdf

Handling Specific Formats

For mixed image formats (JPG, PNG, TIFF):

img2pdf *.jpg *.png -o output.pdf

For images with rotation metadata:

img2pdf --auto-orient *.jpg -o output.pdf

Processing Large Image Sets

For directories with many images, sort them numerically:

img2pdf $(ls -v *.jpg) -o output.pdf

Or use find with numerical sorting:

find . -maxdepth 1 -name "*.jpg" -print0 | sort -z -V | xargs -0 img2pdf -o output.pdf

Adding OCR (Optional)

To make PDFs searchable, use OCRmyPDF after creation:

img2pdf *.jpg -o output.pdf
ocrmypdf output.pdf output-ocr.pdf

This requires ocrmypdf and Tesseract. Install on Debian/Ubuntu:

sudo apt install ocrmypdf tesseract-ocr

Performance Considerations

• img2pdf: Fastest, lossless, smallest output file
• ImageMagick: Slower, lossy compression, useful for pre-processing
• Ghostscript: Good for compression control after creation

For archival or quality-critical work, always use img2pdf. Reserve ImageMagick and Ghostscript for cases where you need image manipulation or aggressive compression.