In addition to world files, GeoExpress embeds encoded MrSID and JPEG 2000 images with a header to preserve metadata from source images. This header is able to carry information on image location, projection, date produced, and RGB color lookup tables (CLUT) for indexed color images. Each tag is accompanied by a short description of the method used to write out the data, either ASCII, signed or unsigned 8-bit, signed or unsigned 16- or 32-bit integers, or single or double-precision floating point.
Certain input image types such as NITF, GeoTIFF, ERDAS IMAGINE and USGS DOQ may have additional metadata tags. Such tags are carried over from the source image as-is to the output file in addition to the standard LizardTech metadata tags. If the image being encoded is a mosaic, GeoExpress preserves the metadata values from the first image in the mosaic. If no information is available for a specific tag, the tag is left out of the header. Some tags apply to any MrSID or JPEG 2000 image, with or without georeferencing information; other tags apply specifically to georeferenced images.
NOTE: When encoding GeoTIFF imagery, all the native GeoTIFF metadata tags are stored as LizardTech metadata tags prepended with the string "GEOTIFF_CHAR::".
In addition there are special tags that contain statistical information about the pixel intensities in an encoded image. These tags contain the minimum, maximum, mean, and standard deviation of the pixel values in the image. For a grayscale image, single values are output. For an RGB image, values are output in an array where the first value represents the red band, the second band represents the green band, and the third band represents the blue band.
Both the mrsidgeoinfo command line tool and ExpressView Browser Plug-in can display LizardTech metadata and GeoTIFF tags. Both tools are included on the GeoExpress DVD.
The tags written into MrSID and JPEG 2000 metadata are listed below. If an image or mosaic is cropped while being encoded, the values apply to the cropped output image unless noted otherwise. There is no predetermined order for the tags.
| Tag | Description | Notes |
|---|---|---|
| IMAGE::BITS_PER_SAMPLE | Number of bits per sample (uint16) | |
| IMAGE::COLOR_SCHEME | Colorspace of image (uint32) | Values: 0 for RGB, 3 for grayscale, 2 for CMYK, 10 for multispectral |
| IMAGE::DATA_TYPE | Datatype of samples in image (uint32) | Values: 0 for unsigned 8-bit int, 1 for 32-bit float, 2 for unsigned 16-bit int |
| IMAGE::DEFAULT_DATA_VALUE | Sample values for background pixel |
Stored as an array of values, one for each band, in order. The values are stored in the datatype of the samples NOTE: With older images containing uint8 data, the tag |
| IMAGE::DYNAMIC_RANGE_WINDOW | Custom contrast setting (double) | Represents the size of the range of the data |
| IMAGE::DYNAMIC_RANGE_LEVEL | Custom brightness setting (double) | Represents the midpoint of the range of the data |
| IMAGE::EOM | End of metadata | |
| IMAGE::HEIGHT | Height of the image, in pixels (uint32) | |
| IMAGE::INPUT_FILE_SIZE | Size of the input image or mosaic in bytes (double) | |
| IMAGE::INPUT_FORMAT | Name of input image type (string) | |
| IMAGE::INPUT_LUT | Color lookup table | Stored as an array of 256*3 values, one value for each band (R,G,B) for each of the 256 entries in the table |
| IMAGE::INPUT_NAME | Filename of the input image (string) | |
| IMAGE::NO_DATA_VALUE | Sample values for background pixel | Used only by older MrSID images; see IMAGE::DEFAULT_DATA_VALUE |
| IMAGE::SOM | Start of metadata | |
| IMAGE::TRANSPARENT_DATA_VALUE | Sample values for the "no data" pixel | Stored as an array of values, one for each band, in order. Values are stored in the datatype of the samples |
| IMAGE::WIDTH | Width of the image, in pixels (uint32) | |
| IMAGE::X_RESOLUTION | Georeferencing pixel resolution in x-direction (double) | |
| IMAGE::XY_ORIGIN | Georeferencing (x,y) location for the center of the upper left corner pixel | Stored as an array of two doubles |
| IMAGE::Y_RESOLUTION | Georeferencing pixel resolution in x-direction (double) |
The tags in the table below apply only to images encoded with areas of interest:
| Tag | Description | Notes |
|---|---|---|
| IMAGE::AOI::n::REGION::VECTOROVERLAY | Name of vector overlay file, if any – (string) | |
| IMAGE::AOI::n::REGION::VECTOROVERLAY_LAYER | Layer number from vector overlay file (if one is used) – (integer) | |
| IMAGE::AOI::n::REGION::X | Upper left X pos of region – (integer) | |
| IMAGE::AOI::n::REGION::Y | Upper left Y pos of region – (integer) | |
| IMAGE::AOI::n::METHOD | The AOI method used – (string) | Values: "shift inner", "shift outer", "weight" |
| IMAGE::AOI::n::WEIGHT | Weight value used – (double) | |
| IMAGE::AOI::n::MAGNIFICATION | Magnification at which AOI was applied – (double) | |
| IMAGE::AOI::n::NAME | Optional name of AOI region – (string) | |
| IMAGE::AOI::n::COMMENT | Optional comment for AOI region – (string) | |
| IMAGE::AOI::n::URL | Optional URL referring to AOI region – (string) |
The classical tags in the table below apply to MG2 images only:
| Tag | Description | Notes |
|---|---|---|
| IMAGE::COMPRESSION_BLOCK_SIZE | Block size used in MrSID encoding (uint32) | |
| IMAGE::COMPRESSION_GAMMA | G-weight value used in MrSID encoding (float) | |
| IMAGE::COMPRESSION_VERSION | Version of encoder used (array of 3 sint32 values) | |
| IMAGE::COMPRESSION_WEIGHT | Weight value used in MRSID encoding (float) | |
| IMAGE::CREATION_DATE | Date and time of image encoding (string) | |
| IMAGE::COMPRESSION_NLEV | Number of zoom (resolution) levels in the image (uint32) | |
| IMAGE::STATISTICS:MAXIMUM | Maximum sample values for each band in the input image (array of values) | The number and datatype of the values correspond to the number of bands and sample type of the image |
| IMAGE::STATISTICS:MINIMUM | Maximum sample values for each band in the input image (array of values) | The number and datatype of the values correspond to the number of bands and sample type of the image |
| IMAGE::STATISTICS:MEAN | Average value of all samples for each band (array of doubles) | |
| IMAGE::STATISTICS:STANDARD_DEVIATION | Standard deviation of all samples for each band | |
| IMAGE::TARGET_COMPRESSION_RATIO | Compression ratio used for encoding (float) | For MG2, this only approximates the actual compression ratio achieved |
The tags in the table below apply to MG4 images only:
| Tag | Description | Notes |
|---|---|---|
| IMAGE::QUANTIZATION_SCALE | Precision value used for quantization | For floating point images, GeoExpress uses the quantization compression method. For more information on the precision value, see Floating Point Quantization Precision. |
When using GeoTIFF input images, the GeoTIFF metadata tags are copied directly into the MrSID or JPEG 2000 file. When using ERDAS IMAGINE and USGS DOQ metadata, certain other custom metadata tags are inserted as well.
For information about how GeoExpress handles existing NITF metadata, see NITF* Input Metadata.
*Support for encoding NITF images is available for GeoExpress. Contact your LizardTech representative to purchase the NITF Extension for GeoExpress.