The projections support (read if a *.prj file is available) is quite limited. Additional sample .prj files may be sent to the maintainer, warmerdam@pobox.com.
The NODATA value for the grid read is also preserved when available.
If pixels being written are not square (the width and height of a pixel in georeferenced units differ) then DX and DY parameters will be output instead of CELLSIZE. Such files can be used in Golden Surfer, but not most other ascii grid reading programs. For force the X pixel size to be used as CELLSIZE use the FORCE_CELLSIZE=YES creation option or resample the input to have square pixels.
When writing floating-point values, the driver uses the "%6.20g" format pattern as a default. You can consult a reference manual for printf to have an idea of the exact behaviour of this ;-). You can alternatively specify the number of decimal places with the DECIMAL_PRECISION creation option. For example, DECIMAL_PRECISION=3 will output numbers with 3 decimal places.
The AIG driver is also available for Arc/Info Binary Grid format.
NOTE: Implemented as gdal/frmts/aaigrid/aaigriddataset.cpp.
Supported by GDAL for read access. Creation is possible, but it must be considered as experimental and a means of testing read access (although files created by the driver can be read successfully on another GIS software)
An ADRG dataset is made of several files. The file recognised by GDAL is the General Information File (.GEN). GDAL will also need the image file (.IMG), where the actual data is.
The Transmission Header File (.THF) can also be used as an input to GDAL. If the THF references more than one image, GDAL will report the images it is composed of as subdatasets. If the THF references just one image, GDAL will open it directly.
Overviews, legends and insets are not used. Polar zones (ARC zone 9 and 18) are not supported (due to the lack of test data).
This is an alternative to using the OGDI Bridge for ADRG datasets.
See also : the ADRG specification (MIL-A-89007)Support includes reading of an affine georeferencing transform, some projections, and a color table (.clr) if available.
This driver is implemented based on a reverse engineering of the format. See the format description for more details.
The projections support (read if a prj.adf file is available) is quite limited. Additional sample prj.adf files may be sent to the maintainer, warmerdam@pobox.com.
NOTE: Implemented as gdal/frmts/aigrid/aigdataset.cpp.
This driver should also support GEO/NOS format as supplied by Softchart. These files normally have the extension .nos with associated .geo files containing georeferencing ... the .geo files are currently ignored.
This driver is based on work by Mike Higgins. See the frmts/bsb/bsb_read.c files for details on patents affecting BSB format.
Starting with GDAL 1.6.0, it is possible to select an alternate color palette via the BSB_PALETTE configuration option. The default value is RGB. Other common values that can be found are : DAY, DSK, NGT, NGR, GRY, PRC, PRG...
NOTE: Implemented as gdal/frmts/bsb/bsbdataset.cpp.
The driver does not support reading or writting gzipped (.bt.gz) .bt files even though this is supported by the VTP software. Please unpack the files before using with GDAL using the "gzip -d file.bt.gz".
Projections in external .prj files are read and written, and support for most internally defined coordinate systems is also available.
Read/write imagery access with the GDAL .bt driver is terribly slow due to a very inefficient access strategy to this column oriented data. This could be corrected, but it would be a fair effort.
NOTE: Implemented as gdal/frmts/raw/btdataset.cpp.
See Also: The BT file format is defined on the VTP web site.
This driver is known to work with CEOS data produced by Spot Image, but will have problems with many other data sources. In particular, it will only work with eight bit unsigned data.
See the separate SAR_CEOS driver for access to SAR CEOS data products.
NOTE: Implemented as gdal/frmts/ceos/ceosdataset.cpp.
The imagery is in a distinct imagery file, often a TIFF file, but the DIMAP dataset handles accessing that file, and attaches geolocation and other metadata to the dataset from the metadata xml file.
From GDAL 1.6.0, the content of the <Spectral_Band_Info> node is reported as metadata at the level of the raster band. Note that the content of the Spectral_Band_Info of the first band is still reported as metadata of the dataset, but this should be considered as a deprecated way of getting this information. NOTE: Implemented as gdal/frmts/dimap/dimapdataset.cpp.
NOTE: Implemented as gdal/frmts/raw/doq1dataset.cpp.
This driver was implemented by Derrick J Brashear.
NOTE: Implemented as gdal/frmts/raw/doq2dataset.cpp.
See Also: USGS DOQ Standards
This driver does not always do well differentiating between floating point and integer data. The GDAL extension to the .hdr format to differentiate is to add a field named PIXELTYPE with values of either FLOAT, SIGNEDINT or UNSIGNEDINT. In combination with the NBITS field it is possible to described all variations of pixel types. eg.
ncols 1375 nrows 649 cellsize 0.050401 xllcorner -130.128639 yllcorner 20.166799 nodata_value 9999.000000 nbits 32 pixeltype float byteorder msbfirstThis driver may be sufficient to read GTOPO30 data.
NOTE: Implemented as gdal/frmts/raw/ehdrdataset.cpp.
See Also:
To open a dataset select the file with the header information. The driver finds the image file from the header information. Erdas documents call the header file the "raw" file and it may have the extension .raw while the image file that contains the actual raw data may have the extension .bl.
NOTE: Implemented as gdal/frmts/raw/eirdataset.cpp.
GDAL should support reading bil, bip and bsq interleaved formats, and most pixel types are supported, including 8bit unsigned, 16 and 32bit signed and unsigned integers, 32bit and 64 bit floating point, and 32bit and 64bit complex floating point. There is limited support for recognising map_info keywords with the coordinate system and georeferencing. In particular, UTM and State Plane should work.
Creation Options:
The control points of the GEOLOCATION GRID ADS dataset are read if available, generally giving a good coverage of the dataset. The GCPs are in WGS84.
Virtually all key/value pairs from the MPH and SPH (primary and secondary headers) are copied through as dataset level metadata.
NOTE: Implemented as gdal/frmts/envisat/envisatdataset.cpp.
See Also: Envisat Data Products at ESA.
FITS is a format used mainly by astronomers, but it is a relatively simple format that supports arbitrary image types and multi-spectral images, and so has found its way into GDAL. FITS support is implemented in terms of the standard CFITSIO library, which you must have on your system in order for FITS support to be enabled. Both reading and writing of FITS files is supported. At the current time, no support for a georeferencing system is implemented, but WCS (World Coordinate System) support is possible in the future.
Non-standard header keywords that are present in the FITS file will be copied to the dataset's metadata when the file is opened, for access via GDAL methods. Similarly, non-standard header keywords that the user defines in the dataset's metadata will be written to the FITS file when the GDAL handle is closed.
Note to those familiar with the CFITSIO library: The automatic rescaling of data values, triggered by the presence of the BSCALE and BZERO header keywords in a FITS file, is disabled in GDAL. Those header keywords are accessible and updatable via dataset metadata, in the same was as any other header keywords, but they do not affect reading/writing of data values from/to the file.
NOTE: Implemented as gdal/frmts/fits/fitsdataset.cpp.
NOTE: Implemented as gdal/frmts/raw/genbindataset.cpp.
NOTE: Implemented as gdal/frmts/gsg/gsagdataset.cpp.
NOTE: Implemented as gdal/frmts/gsg/gsbgdataset.cpp.
NOTE: Implemented as gdal/frmts/gsg/gs7bgdataset.cpp.
Details on the supporting code, and format can be found on the GXF-3 page.
NOTE: Implemented as gdal/frmts/gxf/gxfdataset.cpp.
Projection and georeferencing information is read though some projections (ie. Meteosat, and Hammer-Aitoff) are not supported. When writing IDA files the projection must have a false easting and false northing of zero. The support coordinate systems in IDA are Geographic, Lambert Conformal Conic, Lambert Azimuth Equal Area, Albers Equal-Area Conic and Goodes Homolosine.
IDA files typically contain values scaled to 8bit via a slope and offset. These are returned as the slope and offset values of the bands and they must be used if the data is to be rescaled to original raw values for analysis.
NOTE: Implemented as gdal/frmts/raw/idadataset.cpp.
See Also: WinDisp
Features:
These files are represented as having one 32bit floating band with elevation data. The georeferencing of the files is returned as well as the coordinate system (always lat/long on the Tokyo datum).
There is no update or creation support for this format.
NOTE: Implemented as gdal/frmts/jdem/jdemdataset.cpp.
See Also: Geographic Survey Institute (GSI) Web Site.
GDAL does read the map extents (geotransform) from LAN/GIS files, and attempts to read the coordinate system informaton. However, this format of file does not include complete coordinate system information, so for state plane and UTM coordinate systems a LOCAL_CS definition is returned with valid linear units but no other meaningful information.
The .TRL, .PRO and worldfiles are ignored at this time.
NOTE: Implemented as gdal/frmts/raw/landataset.cpp
Development of this driver was financially supported by Kevin Flanders of (PeopleGIS).
Reading lat/long GCPs (TOP_LEFT_CORNER, ...) is supported but there is no support for reading affine georeferencing or projection information.
Unrecognised keywords from the .hdr file are preserved as metadata.
All data types with GDAL equivelents are supported, including 8, 16, 32 and 64 bit data precisions in integer, real and complex data types. In addition tile organized files (as produced by the Vexcel SAR Processor - APP) are supported for reading.
On creation (with a format code of MFF) a simple, ungeoreferenced raster file is created.
MFF files are not normally portable between systems with different byte orders. However GDAL honours the new BYTE_ORDER keyword which can take a value of LSB (Integer -- little endian), and MSB (Motorola -- big endian). This may be manually added to the .hdr file if required.
NOTE: Implemented as gdal/frmts/raw/mffdataset.cpp.
The NDF driver only supports 8bit data. The only supported projection is UTM. NDF version 1 (NDF_VERSION=0.00) and NDF version 2 are both supported.
NOTE: Implemented as gdal/frmts/raw/ndfdataset.cpp.
See Also: NLAPS Data Format Specification.
The units information in the file will be ignored, but x_range, and y_range information will be read to get georeferenced extents of the raster. All netCDF data types should be supported for reading. Newly created files (with a type of GMT) will always have units of "meters" for x, y and z but the x_range, y_range and z_range should be correct. Note that netCDF does not have an unsigned byte data type, so 8bit rasters will generally need to be converted to Int16 for export to GMT.
NetCDF support in GDAL is optional, and not compiled in by default.
NOTE: Implemented as gdal/frmts/netcdf/gmtdataset.cpp.
See Also: Unidata NetCDF Page
The format type for creating new files is PAux. All PCI data types (8U, 16U, 16S, and 32R) are supported. Currently georeferencing, projections, and other metadata is ignored.
NOTE: Implemented as gdal/frmts/raw/pauxdataset.cpp.
See Also: PCI's .aux Format Description
GDAL includes support for reading and writing PCRaster raster files. PCRaster is a dynamic modelling system for distributed simulation models. The main applications of PCRaster are found in environmental modelling: geography, hydrology, ecology to name a few. Examples include rainfall-runoff models, vegetation competition models and slope stability models.
The driver reads all types of PCRaster maps: booleans, nominal, ordinals, scalar, directional and ldd. The same cell representation used to store values in the file is used to store the values in memory.
The driver detects whether the source of the GDAL raster is a PCRaster file. When such a raster is written to a file the value scale of the original raster will be used. The driver always writes values using UINT1, INT4 or REAL4 cell representations, depending on the value scale:
| Value scale | Cell representation |
|---|---|
| VS_BOOLEAN | CR_UINT1 |
| VS_NOMINAL | CR_INT4 |
| VS_ORDINAL | CR_INT4 |
| VS_SCALAR | CR_REAL4 |
| VS_DIRECTION | CR_REAL4 |
| VS_LDD | CR_UINT1 |
For rasters from other sources than a PCRaster raster file a value scale and cell representation is determined according to the folowing rules:
| Source type | Target value scale | Target cell representation |
|---|---|---|
| GDT_Byte | VS_BOOLEAN | CR_UINT1 |
| GDT_Int32 | VS_NOMINAL | CR_INT4 |
| GDT_Float32 | VS_SCALAR | CR_REAL4 |
| GDT_Float64 | VS_SCALAR | CR_REAL4 |
The driver can convert values from one supported cell representation to another. It cannot convert to unsupported cell representations. For example, it is not possible to write a PCRaster raster file from values which are used as CR_INT2 (GDT_Int16).
Although the de-facto file extension of a PCRaster raster file is .map, the PCRaster software does not require a standardized file extension.
NOTE: Implemented as gdal/frmts/pcraster/pcrasterdataset.cpp.
See also: PCRaster website at Utrecht University and PCRaster Environmental Software company website.PNG files are linearly compressed, so random reading of large PNG files can be very inefficient (resulting in many restarts of decompression from the start of the file).
Text chunks are translated into metadata, typically with multiple lines per item. World files with the extensions of .pgw, .pngw or .wld will be read. Single transparency values in greyscale files will be recognised as a nodata value in GDAL. Transparent index in paletted images are preserved when the color table is read.
PNG files can be created with a type of PNG, using the CreateCopy() method, requiring a prototype to read from. Writing includes support for the various image types, and will preserve transparency/nodata values. Georeferencing .wld files are written if option WORLDFILE setted. All pixel types other than 16bit unsigned will be written as eight bit.
Creation Options:
PNG support is implemented based on the libpng reference library. More information is available at http://www.libpng.org/pub/png.
Netpbm files can be created with a type of PNM.
Creation Options:
This is a read-only reader for RPF products, like CADRG or CIB, that use the table of content file - A.TOC - from a RPF exchange.
The driver will report a different subdataset for each subdataset found in the A.TOC file.
Result of a gdalinfo on a A.TOC file.Subdatasets: SUBDATASET_1_NAME=NITF_TOC_ENTRY:CADRG_GNC_5M_1_1:GNCJNCN/rpf/a.toc SUBDATASET_1_DESC=CADRG:GNC:Global Navigation Chart:5M:1:1 [...] SUBDATASET_5_NAME=NITF_TOC_ENTRY:CADRG_GNC_5M_7_5:GNCJNCN/rpf/a.toc SUBDATASET_5_DESC=CADRG:GNC:Global Navigation Chart:5M:7:5 SUBDATASET_6_NAME=NITF_TOC_ENTRY:CADRG_JNC_2M_1_6:GNCJNCN/rpf/a.toc SUBDATASET_6_DESC=CADRG:JNC:Jet Navigation Chart:2M:1:6 [...] SUBDATASET_13_NAME=NITF_TOC_ENTRY:CADRG_JNC_2M_8_13:GNCJNCN/rpf/a.toc SUBDATASET_13_DESC=CADRG:JNC:Jet Navigation Chart:2M:8:13
In some situations, NITF tiles inside a subdataset don't share the same palettes. The RPFTOC driver will do its best to remap palettes to the reported palette by gdalinfo (which is the palette of the first tile of the subdataset). In situations where it wouldn't give a good result, you can try to set the RPFTOC_FORCE_RGBA environment variable to TRUE before opening the subdataset. This will cause the driver to expose the subdataset as a RGBA dataset, instead of a paletted one.
It is possible to build external overviews for a subdataset. The overview for the first subdataset will be named A.TOC.1.ovr for example, for the second dataset it will be A.TOC.2.ovr, etc. Note that you must re-open the subdataset with the same setting of RPFTOC_FORCE_RGBA as the one you have used when you have created it. Do not use any method other than NEAREST resampling when building overviews on a paletted subdataset (RPFTOC_FORCE_RGBA unset)
A gdalinfo on one of this subdataset will return the various NITF metadata, as well as the list of the NITF tiles of the subdataset.
See Also:
This driver gives an equivalent functionnality to the RPF driver from the OGDI library, that can be accessed through the OGDI Bridge
NOTE: Implemented as gdal/frmts/nitf/rpftocdataset.cppThis driver works with most Radarsat and ERS data products, including single look complex products; however, it is unlikely to work for non-Radar CEOS products. The simpler CEOS driver is often appropriate for these.
This driver will attempt to read 15 lat/long GCPS by sampling the per-scanline CEOS superstructure information. It also captures various pieces of metadata from various header files, including:
CEOS_LOGICAL_VOLUME_ID=EERS-1-SAR-MLD CEOS_PROCESSING_FACILITY=APP CEOS_PROCESSING_AGENCY=CCRS CEOS_PROCESSING_COUNTRY=CANADA CEOS_SOFTWARE_ID=APP 1.62 CEOS_ACQUISITION_TIME=19911029162818919 CEOS_SENSOR_CLOCK_ANGLE= 90.000 CEOS_ELLIPSOID=IUGG_75 CEOS_SEMI_MAJOR= 6378.1400000 CEOS_SEMI_MINOR= 6356.7550000The SAR_CEOS driver also includes some support for SIR-C and PALSAR polarimetric data. The SIR-C format contains an image in compressed scattering matrix form, described here. GDAL decompresses the data as it is read in. The PALSAR format contains bands that correspond almost exactly to elements of the 3x3 Hermitian covariance matrix- see the ERSDAC-VX-CEOS-004A.pdf document for a complete description (pixel storage is described on page 193). GDAL converts these to complex floating point covariance matrix bands as they are read in. The convention used to represent the covariance matrix in terms of the scattering matrix elements HH, HV (=VH), and VV is indicated below. Note that the non-diagonal elements of the matrix are complex values, while the diagonal values are real (though represented as complex bands).
NOTE: Implemented as gdal/frmts/ceos2/sar_ceosdataset.cpp.
The driver supports reading the following SAGA datatypes (in brackets the corresponding GDAL types): BIT (GDT_Byte), BYTE_UNSIGNED (GDT_Byte), BYTE (GDT_Byte), SHORTINT_UNSIGNED (GDT_UInt16), SHORTINT (GDT_Int16), INTEGER_UNSIGNED (GDT_UInt32), INTEGER (GDT_Int32), FLOAT (GDT_Float32) and DOUBLE (GDT_Float64).
The driver supports writing the following SAGA datatypes: BYTE_UNSIGNED (GDT_Byte), SHORTINT_UNSIGNED (GDT_UInt16), SHORTINT (GDT_Int16), INTEGER_UNSIGNED (GDT_UInt32), INTEGER (GDT_Int32), FLOAT (GDT_Float32) and DOUBLE (GDT_Float64).
Currently the driver does not support zFactors other than 1 and reading SAGA grids which are written TOPTOBOTTOM.
NOTE: Implemented as gdal/frmts/saga/sagadataset.cpp.
SDTS datasets consist of a number of files. Each DEM should have one file with a name like XXXCATD.DDF. This should be selected to open the dataset.
The elevation units of DEMs may be feet or meters. The GetType() method on a band will attempt to return if the units are Feet ("ft") or Meters ("m"). NOTE: Implemented as gdal/frmts/sdts/sdtsdataset.cpp.
The driver currently supports 1, 2, 3, and 4 band images. The driver currently supports "8 bit per channel value" images. The driver supports both uncompressed and run-length encoded (RLE) images for reading, but created files are always RLE compressed..
The GDAL SGI Driver was based on Paul Bourke's SGI image read code.
See Also:
NOTE: Implemented as gdal/frmts/sgi/sgidataset.cpp.
ASRP (in a geographic coordinate system) and USRP (in a UTM/UPS coordinate system) products are single band images with a palette and georeferencing.
NOTE: Implemented as gdal/frmts/adrg/srpdataset.cpp.
The driver does support creating new files, but the input data must be exactly formatted as a SRTM-3 or SRTM-1 cell. That is the size, and bounds must be appropriate for a cell.
See Also:
NOTE: Implemented as gdal/frmts/srtmhgt/srtmhgtdataset.cpp.
pixel X size rotation about the Y axis (usually 0.0) rotation about the X axis (usually 0.0) negative pixel Y size X coordinate of upper left pixel center Y coordinate of upper left pixel centerFor example:
60.0000000000 0.0000000000 0.0000000000 -60.0000000000 440750.0000000000 3751290.0000000000You can construct that file simply by using your favorite text editor.
World file usually has suffix .wld, but sometimes it may has .tfw, tifw, .jgw or other suffixes depending on the image file it comes with.
The XPM support does not support georeferencing (not available from XPM files) nor does it support XPM files with more than one character per pixel. New XPM files must be colormapped or greyscale, and colortables will be reduced to about 70 colors automatically.
NOTE: Implemented as gdal/frmts/xpm/xpmdataset.cpp.
The GFF format was implemented based on the Matlab code provided by Sandia to read the data. The driver supports all types of data (16-bit or 32-bit complex, real bytes) theoretically, however due to a lack of data only 32-bit complex data has been tested.
Sandia provides some sample data at http://sandia.gov/RADAR/sar-data.html.
The extension for GFF formats is .gff.
NOTE: Implemented as gdal/frmts/gff/gff_dataset.cpp.
Full list of GDAL Raster Formats
$Id$