Licensing options for this format begin with FME Desktop Professional Edition.
The HDF4 Hyperion Reader module provides FME with access to continuous numeric or classified color data in multiple bands.
HDF stands for Hierarchical Data Format, and is a container for several different datasets, including one or more raster images. It is used most often for storing Scientific Datasets (SDS), which are multidimensional arrays filled with data. One HDF file may contain several different SDS arrays. They may differ in size, number of dimensions and may represent data for different regions.
The Hyperion (HDF4) reader can read both HDF4 and HDF4-EOS files. The latter is a modification of HDF maintained by NASA’s EOS (Earth Observing System). HDF4-EOS is used for storing telemetry from NASA’s ‘Terra’ and ‘Aqua’ satellites. HDF5-EOS is used for storing telemetry from ‘Aura’ satellites.
Note: HDF5 is a newer version of the HDF format, and is completely different from HDF4. The Hyperion (HDF4) reader cannot be used for HDF5 datasets, since HDF4 and HDF5 are not compatible with one another.
Since HDF is a container for datasets rather than a dataset itself, it can contain various subdatasets, which are in most cases associated with a particular imaging sensor. Hyperion data is produced either as level 0, which is raw data with no corrections applied, or level 1R, which is radiometrically controlled data with no geometric corrections applied.
This reader supports Hyperion level 1R data products, which consist of a metadata file, a dataset file (whose subdatasets include image data, spectral center wavelengths, spectral bandwidths, gain coefficients and a flag mask), and an ENVI formatted header file. The image data itself is stored in band interleaved by line, 16-bit signed integer radiance values. There are always 242 bands in a Hyperion L1R raster.
GCPs (ground control points) present along with a projection in a Hyperion file being read can either be applied to the data as an affine transformation, or stored as properties on the raster geometry.
The FME considers a top-level Hyperion container file to be a dataset.
FME Raster Features
FME raster features represent raster data and use several concepts that are unlike those used in the handling of vector data. The topics below describe how FME processes raster data.