Image format

J2K converter

A .j2k file is a bare JPEG 2000 codestream: the same wavelet-coded image data a .jp2 carries, but without the surrounding container boxes. It starts at the SOC marker and ends at EOC, which makes it the form used when JPEG 2000 data is embedded inside another format rather than stored on its own.

Private for supported formats — processed in your browser

Convert supported files

Runs on your device

Drop J2K files here

Batch files can each use a different output. Nothing uploads for local conversions.

Working inputs include camera RAW, browser-local audio/video, PDF, CBZ/CBR comics, office documents, ebooks, markup, 3D models, structured text, images, and archives.

Where J2K comes from

The codestream is defined by Annex A of ISO/IEC 15444-1, the same 2000 standard that defines the JP2 container. Because embedding systems supply their own metadata, the formats that carry JPEG 2000 internally — DICOM, PDF, MXF and digital cinema packages — store the naked codestream, and encoders expose it under the .j2k, .j2c and .jpc extensions.

How J2K works

  • Marker-segment structure: SOC, then the SIZ, COD and QCD headers, then tile-parts, terminated by EOC
  • Identical wavelet and EBCOT coding to a .jp2 file; only the container differs
  • No colour-specification box, so the colour space must be inferred or supplied by the embedding format
  • Progression order, quality layers, and resolution levels are declared in the COD marker segment
  • Commonly named .j2c or .jpc by encoders; the bytes are the same

When to use J2K

  • JPEG 2000 picture essence inside MXF for digital cinema distribution
  • Compressed pixel data embedded in DICOM instances
  • Image streams inside PDF using the JPXDecode filter
  • Pipeline intermediates where a wrapper would be redundant

Strengths and limitations

Strengths

  • Slightly smaller than the boxed form and trivial for a host format to embed
  • Every JPEG 2000 quality and resolution feature is available
  • Widely produced by broadcast and cinema encoders

Limitations

  • Carries no colour space, ICC profile, or metadata of its own
  • Extension naming is inconsistent across tools, which confuses file handling
  • No browser support at all, so it always needs conversion for display

Compatibility

OpenJPEG, Kakadu, ImageMagick, GDAL and the cinema and medical toolchains all read raw codestreams, usually under .j2k, .j2c or .jpc. No browser decodes them. This converter decodes .j2k in the page with the native JPEG 2000 decoder inside the self-hosted FFmpeg core, and because boxed files are sometimes named .j2k as well, the JP2 container signature is accepted here too.

None. A raw codestream has no metadata boxes at all — colour space, ICC profile, and EXIF have to come from the format that embeds it, which is one reason .jp2 exists.