Convert Signed Pixel Range

[cameraNolan]

When working with raster data, we oftentimes produce signed 16bit floating point images. These images tend to be quite large in memory storage size when the pixel resolution is high.

Saving the images as floating point is ideal as it preserves the bit depth precision, but often times a lower bit depth is acceptable. We can convert from signed floating point to unsigned int, say a range of -1.0 to +1.0 converted to 0 - 65535. A translation scalar (formula) can be calculated to convert from one range to the other.

I was wondering if there was a standard exif/xmp tag that is used to save such a scalar that is universally supported in common GIS programs? Obviously we can create custom exif/xmp tags specific to our software, but we are looking for a more standardized method, if one exists.

[StarGeek]

I haven't heard of anything along these lines, and nothing in the XMP tags page seems to apply.

If you do make your own tags, EXIF would not be the place to put them, as XMP is much more flexible. See the example.config file for details.