2

I downloaded a 4-band 1m spatial resolution NAIP image from Earth Explorer. The default image format is jpeg2000. You can see the trees in the attached image appear grainy, more so than what you would expect from a TIFF image. I believe this may be a result of the compression algorithm. To test, I used the Copy Raster tool in ArcGIS to convert the jpeg2000 to TIFF. I specified "NONE" for the compression in the tool environment. Unfortunately, the image appeared the same, although I noticed that there is LZW compression associated with the TIFF. Is there a way to essentially "unpack" a jpeg2000 image to remove all compression? I am open to any software-based approach or programmatic (Python, R, or Matlab) approach.

enter image description here

  • 3
    Lossy compression algorithms are lossy. Once lost, you need to go back further in the image history. – Vince Jun 26 '15 at 19:04
  • @Vince I was under the impression that there was a reversible integer wavelet transformation (en.wikipedia.org/wiki/JPEG_2000). – Aaron Jun 26 '15 at 19:09
  • That seems to be an option, but I don't think I've ever used a lossless jp2. It would certainly reduce the compression (and therefore attractiveness of the format for the purpose). What does the image metadata report? – Vince Jun 26 '15 at 19:21
  • I've used lossless compressed JP2, they're smaller than GeoTiff with LZW or Deflate compression, which would be a viable alternative for lossless compression, but (obviously) not smaller than the lossy compression. As with the compression ratio it needs to be specified at the time of compression to be reversible. Perhaps the USGS has retained the original data in uncompressed format, is there any way to contact them and request this data? or do they only provide data through their portal? – Michael Stimson Jun 28 '15 at 22:54
4

According to the Earth Explorer metadata on NAIP JPEG2000:

The U.S. Geological Survey (USGS) Earth Resources Observation and Science (EROS) Center manages and distributes NAIP products in JPEG2000 format, which is a compressed file with embedded georeferencing information. The 10:1 lossy compression makes the file size smaller by reorganizing the data, but it also slightly degrades the imagery.

In other words, there is no way to recover the data lost in the lossy compression.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.