1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
|
<html>
<head>
<title>MFF2 -- Vexcel MFF2 Image</title>
</head>
<body bgcolor="#ffffff">
<h1>MFF2 -- Vexcel MFF2 Image</h1>
GDAL supports MFF2 Image raster file format for read, update, and creation.
The MFF2 (Multi-File Format 2) format was designed to fit into
Vexcel Hierarchical Key-Value (HKV) databases, which can store
binary data as well as ASCII parameters.
This format is primarily used internally to the Vexcel InSAR
processing system.<p>
To select an MFF2 dataset, select the directory containing the <tt>attrib</tt>,
and <tt>image_data</tt> files for the dataset.<p>
Currently only latitude/longitude and UTM projection are supported
(georef.projection.name = ll or georef.projection.name = utm),
with the affine transform computed from the lat/long control points. In
any event, if GCPs are available in a georef file, they are returned with
the dataset.<p>
Newly created files (with a type of <tt>MFF2</tt>) are always just raw rasters
with no georeferencing information. For read, and creation all data types
(real, integer and complex in bit depths of 8, 16, 32) should be supported. <p>
IMPORTANT: When creating a new MFF2, be sure to set the projection before setting the geotransform (this is necessary because the geotransform is stored internally as 5 latitude-longitude ground control points, and the projection is needed to do the conversion).<p>
NOTE: Implemented as <tt>gdal/frmts/raw/hkvdataset.cpp</tt>.<p>
<h2>Format Details</h2>
<h3>MFF2 Top-level Structure</h3>
An MFF2 "file" is actually a set of files stored in a directory
containing an ASCII header file entitled "attrib", and binary
image data entitled "image_data". Optionally, there may be
an ASCII "georef" file containing georeferencing and projection
information, and an "image_data_ovr" (for "image_data" binary image data)
file containing tiled overviews of the image in TIFF format.
The ASCII files are arranged in key=value pairs. The allowable
pairs for each file are described below.
<h3>The "attrib" File</h3>
As a minimum, the "attrib" file must specify the image extents,
pixel size in bytes, pixel encoding and datatype, and pixel
byte order. For example,
<pre>
extent.cols = 800
extent.rows = 1040
pixel.size = 32
pixel.encoding = { unsigned twos_complement *ieee_754 }
pixel.field = { *real complex }
pixel.order = { lsbf *msbf }
version = 1.1
</pre>
specifies an image that is 1040 lines by 800 pixels in extent. The
pixels are 32 bits of real data in "most significant byte first" (msbf)
order, encoded according to the ieee_754 specification. In MFF2, when
a value must belong to a certain subset (eg. pixel.order must be either
lsbf or msbf), all options are displayed between curly brackets, and
the one appropriate for the current file is indicated with a "*".
<p>
The file may also contain the following lines indicating the number of
channels of data, and how they are interleaved within the binary data
file.
<pre>
channel.enumeration = 1
channel.interleave = { *pixel tile sequential }
</pre>
<h3>The "image_data" File</h3>
The "image_data" file consists of raw binary data, with extents,
pixel encoding, and number of channels as indicated in the "attrib" file.
<h3>The "georef" File</h3>
The "georef" file is used to describe the geocoding and projection
information for the binary data. For example,
<pre>
top_left.latitude = 32.93333333333334
top_left.longitude = 130.0
top_right.latitude = 32.93333333333334
top_right.longitude = 130.5
bottom_left.latitude = 32.50000000000001
bottom_left.longitude = 130.0
bottom_right.latitude = 32.50000000000001
bottom_right.longitude = 130.5
centre.latitude = 32.71666666666668
centre.longitude = 130.25
projection.origin_longitude = 0
projection.name = ll
spheroid.name = wgs-84
</pre>
describes an orthogonal latitude/longitude (ll) projected image, with
latitudes and longitudes based on the wgs-84 ellipsoid.
<p>
Since MFF2 version 1.1,
top_left refers to the top left corner of the top left pixel.
top_right refers to the top right corner of the top right pixel.
bottom_left refers to the bottom left corner of the bottom left pixel.
bottom_right refers to the bottom right corner of the bottom right pixel.
centre refers to the centre of the four corners defined above (center
of the image).
<p>
Mathematically, for an Npix by Nline image, the corners and centre in
(pixel,line) coordinates for MFF2 version 1.1 are:
<pre>
top_left: (0,0)
top_right: (Npix,0)
bottom_left: (0,Nline)
bottom_right: (Npix,Nline)
centre: (Npix/2.0,Nline/2.0)
</pre>
These calculations are done using floating point arithmetic (ie.
centre coordinates may take on non-integer values).
<p>
Note that the corners are always expressed in latitudes/longitudes, even
for projected images.
<p>
<h3>Supported projections</h3>
ll- Orthogonal latitude/longitude projected image, with latitude
parallel to the rows, longitude parallel to the columns. Parameters:
spheroid name, projection.origin_longitude (longitude at the origin of
the projection coordinates). If not set, this should default to the
central longitude of the output image based on its projection boundaries.
<p>
utm- Universal Transverse Mercator projected image. Parameters:
spheroid name, projection.origin_longitude (central meridian for the
utm projection). The central meridian must be the meridian at the
centre of a UTM zone, ie. 3 degrees, 9 degrees, 12 degrees, etc. If
this is not specified or set a valid UTM central meridian, the reader
should reset the value to the nearest valid central meridian based on
the central longitude of the output image. The latitude at the origin
of the UTM projection is always 0 degrees.
<p>
<h3>Recognized ellipsoids</h3>
MFF2 format associates the following names with ellipsoid equatorial
radius and inverse flattening parameters:
<pre>
airy-18304: 6377563.396 299.3249646
modified-airy4: 6377340.189 299.3249646
australian-national4: 6378160 298.25
bessel-1841-namibia4: 6377483.865 299.1528128
bessel-18414: 6377397.155 299.1528128
clarke-18584: 6378294.0 294.297
clarke-18664: 6378206.4 294.9786982
clarke-18804: 6378249.145 293.465
everest-india-18304: 6377276.345 300.8017
everest-sabah-sarawak4:6377298.556 300.8017
everest-india-19564: 6377301.243 300.8017
everest-malaysia-19694:6377295.664 300.8017
everest-malay-sing4: 6377304.063 300.8017
everest-pakistan4: 6377309.613 300.8017
modified-fisher-19604: 6378155 298.3
helmert-19064: 6378200 298.3
hough-19604: 6378270 297
hughes4: 6378273.0 298.279
indonesian-1974: 6378160 298.247
international-1924: 6378388 297
iugc-67: 6378160.0 298.254
iugc-75: 6378140.0 298.25298
krassovsky-1940: 6378245 298.3
kaula: 6378165.0 292.308
grs-80: 6378137 298.257222101
south-american-1969: 6378160 298.25
wgs-72: 6378135 298.26
wgs-84: 6378137 298.257223563
ev-wgs-84: 6378137 298.252841
ev-bessel: 6377397 299.1976073
</pre>
<h3>Explanation of fields</h3>
<pre>
channel.enumeration: (optional- only needed for multiband)
Number of channels of data (eg. 3 for rgb)
channel.interleave = { *pixel tile sequential } : (optional- only
needed for multiband)
For multiband data, indicates how the channels are interleaved. *pixel
indicates that data is stored red value, green value, blue value, red
value, green value, blue value etc. as opposed to (line of red values)
(line of green values) (line of blue values) or (entire red channel)
(entire green channel) (entire blue channel)
extent.cols:
Number of columns of data.
extent.rows:
Number of rows of data.
pixel.encoding = { *unsigned twos-complement ieee-754 }:
Combines with pixel.size and pixel.field to give the data type:
(encoding, field, size)- type
(unsigned, real, 8)- unsigned byte data
(unsigned, real, 16)- unsigned int 16 data
(unsigned, real, 32)- unsigned int 32 data
(twos-complement, real, 16)- signed int 16 data
(twos-complement, real, 32)- signed int 32 data
(twos-complement, complex, 64)- complex signed int 32 data
(ieee-754, real, 32)- real 32 bit floating point data
(ieee-754, real, 64)- real 64 bit floating point data
(ieee-754, complex, 64)- complex 32 bit floating point data
(ieee-754, complex, 128)- complex 64 bit floating point data
pixel.size:
Size of one pixel of one channel (bits).
pixel.field = { *real complex }:
Whether the data is real or complex.
pixel.order = { *lsbf msbf }:
Byte ordering of the data (least or most significant byte first).
version: (only in newer versions- if not present, older version is
assumed) Version of mff2.
</pre>
</body>
</html>
|
