Home

Über Vimage

Vimage kennenlernen

Einsteiger-Tutorial

Vimage Praxis

Grundlagenwissen

Alle Tutorials

Technologien

Hintergrundwissen

• Dokumentation
•• Fiximage File Format Description

Informationen

Zentrale Download-Seite

Vimage erwerben

Vimage Leerseite

Anhang

Vimage Special

Fiximage File Format Description

Fiximage header hexdump

FIXIMAGE FILE FORMAT DOCUMENTATION
==================================
(Issue 03.01.09)
IN GENERAL
----------
 The Fiximage file format is a highly simple and precise raster image file format.
 Fiximages are in general raw data. The first 512 file bytes contains
 a header with few control parameters. At byte 512 is starting the image data
 matrix, each pixel - one grey value per band. The image matrix is stored
 in line by line order, band sequential with a 32 byte alignment.
HEADER DATA
-----------
 Header length: 512 bytes.
 The 1st 256 file bytes contains 32 8-byte words.
 The 2nd 256 file bytes contains 3 text or comment strings.
-------------------------------------------------------------------------------
 00   ImgType  String*8   X  Must be FIXIMAGE. Also little/big endian indicator
 08   ImgTyp2  String*8   O  (Fixlib-Version. 8 chars)
 16   ImgXXXX  Long       X  Number of columns
 24   ImgYYYY  Long       X  Number of rows
 32   ImgNofB  Long       X  Number of bands or channels
 40   ImgNofL  Long       O  Number of layers (Should be 1)
 48   ImgDTyp  String*8   X  Data type. BINARY ... BYTE ... COMPLEX
 56   ImgDefL  Long       O  Default layer (Should be 1)
 64   GeoType  String*8   .  [not used]
 72   GeoTyp2  String*8   .  [not used]
 80   GeoNUni  String*8   X  Reference unit: KM, M, CM, MM ...
 88   GeoMUni  Currency   O  Map unit. (Should be MM)
 96   GeoSWPX  Currency   X  SW pixel centre geo coordinate X
104   GeoSWPY  Currency   X  SW pixel centre geo coordinate Y
112   GeoNEPX  Currency   X  NE pixel centre geo coordinate X
120   GeoNEPY  Currency   X  NE pixel centre geo coordinate Y
128   RadMode  String*8   X  Color model: MONO, XY, RGB or CMYK
136   RadUnit  String*8   O  Vertical unit. Should be M
144   RadType  String*8   .  [not used]
152   RadTyp2  String*8   .  [not used]
160   RadBLev  Currency   X  Black level
168   RadWLev  Currency   X  White level
176   GeoScaX  Currency   X  Reference scale number X
184   GeoScaY  Currency   X  Reference scale number Y
192   AuxPar1  Long       .  [not used]
200   AuxPar2  Long       .  [not used]
208   AuxPar3  Long       .  [not used]
216   AuxPar4  Long       .  [not used]
224   AuxPar5  Long       .  [not used]
232   AuxPar6  Long       .  [not used]
240   HdrGenE  Byte*8     X  Header generation params, actual: 0 0 0 0 0 0 0 3
248   HdrLeng  Long       O  (Header length. Should be 512)
-------------------------------------------------------------------------------
256   ComTitl  String*64  .  Title
320   ComNote  String*64  .  Note
384   ComDesc  String*128 .  Description
-------------------------------------------------------------------------------
 Long:         64 bit signed integer.
 Currency:     64 bit signed integer multiplied with 10000 (Fixpoint).
 Byte*8:       8 x 8 bytes.
 String*n:     Fixed length strings, length n.
 X  ...    Must be set.
 O  ...    Should be set to future use.
 .  ...    Not in use or comment
IMAGE DATA
----------
 Per pixel (and band) is storaged one numeric value, the grey value of the
 pixel. The 1st pixel of the 1st line of the 1st band has file address 512.
 The data is storaged first line by line, than band sequential.
 The lines have a 32 byte alignment. Every new line startes on a
 32 byte limit.
 The 16 supported data types are:
 ¯VOID«       0 bit/value. All values are zero.
 ¯BINARY«     1 bit/value. 2 values, 0 or 255
 ¯NONARY«   3.2 bit/value. 9 values 0, 31.8, 63.7, 95.6, 127.5 ... 255
 ¯BYTE«       8 bit integers 0 ... 255
 ¯CHAR«      16 bit integers 0 ... 65535
 ¯SHORT«     16 bit integers -32768 ... +32767
 ¯RELIEF«    16 bit numbers -10268 ... +10267 (in range ±2500 0.1 precision)
 ¯TETRABYT«  32 bit integers 0 ... 4294967295
 ¯INTEGER«   32 bit integers -2147483548 ... 2147483647
 ¯FIXPOINT«  32 bit fixpoint numbers -214748.3648 ... 214748.3647
 ¯SINGLE«    32 bit single float. Up to 10E38, 7 digits precision
 ¯OCTABYTEG« 64 bit integers 0 ... 18446744073709551615
 ¯LONG«      64 bit integers  -9223372036854775808 ... 9223372036854775807
 ¯CURRENCY«  64 bit-fixpoint numbers, -922337203685477.5808-922337203685477.5807
 ¯DOUBLE«    64 bit double float. Up to 10E308, 14 digits precision
 ¯COMPLEX«   64 bit complex. Two 32 bit float numbers (real and imaginary)
 ¯NONARY« encodes the 9 values 0, 31.8, 63.7, 95.6, 127.50 ... 255 in
 3.2 bit per value. That is a compact data type especially for maps and
 drawings. The 9 values are codes as nonary digits 0 ... 8 and taken as a
 10-nonary-digit-word 0 ... 3'486'784'401 in a 32 bit integer.
 ¯RELIEF« encodes the values 0 ... ±2500 als 0 ... ±25000 with one digit
 after decimal point. Values > 2500 are integers and shifted in the
 range ±25000 ... ±32767. Such encoding allows a compact DEM
 storaging with a elevation range -10268 ... +10267 (experimental state).
 ¯FIXPOINT« and ¯CURRENCY« are fixpoint formats both with 4 digits after
 decimal point. Fixpoint also is a recommended DEM data storage format.
COORDINATES
-----------
 • Image coordinates: The left lower image corner (SW corner) has
   image coordinate (1, 1). Every pixel increments by 1.
 • Geo coordinates are unitless coordinates in the nature or a map.
   They are defined by the header parameters GeoSWPX, GeoSWPY, GeoNEPX and
   GeoNEPY. Each pixel geo coordinate gives the pixel centre position.
   To convert them into pixel corner coordinates add 1/2 pixel size.
 • Output coordinates are millimeter coordinates on a output media. To get
   them add the reference unit GeoNUni (km, m, cm, mm, µm, degree, inch ...)
   to the (unitless) geo coordinate and scale it with the reference scales
   GeoScaX (GeoScaY). Then shift coordinate (0, 0) to the left lower corner
   of the left lower pixel.
 • All coordinates are defined Cartesian. That means that x = right-, y = up-
   direction.
OTHER REMARKS
-------------
 • Vimage internal number format is single.
 • Grey value -9999.0000 often means ¯empty«.
 • The Vimage internal image matrices have 16 outer margin.
FIXLIB CONSTANTS
----------------
 FIXLIB_COMMENT  [Fixlib comment      ]   (String) Fixpoint Image File Format Library FIXLIB 03.01.09, 18.09.2009
 FIXLIB_VERSION  [Fixlib version      ]   (String) 03.01.09
 FIXLIB_GENERATION [Fixlib subversion ]   (Long)   3
 FIXMAXINTEGER   [Largest 32 bit int  ]   (Long)   2147483647
 FIXMININTEGER   [Smallest 32 bit int ]   (Long)   -2147483648
 FIXMAXUINTEGER  [Largest 32 bit uint ]   (Double) 4294967295
 FIXMAX          [Largest fixpoint    ]   (Double) 214748,3647
 FIXMIN          [Smallest fixpoint   ]   (Double) -214748,3648
 FIXMAXCOREX     [Max numb of columns ]   (Long)   262144
 FIXMAXCOREY     [Max numb of lines   ]   (Long)   262144
 FIXDIV          [Fixpoint divisor    ]   (Long)   10000
 FIX_EARTH_RADIUS        [Earth radius]   (Double) 6371004,2029572
 FIX_EARTH_RADIUS_POLE                    (Double) 6371004,2029572
 FIX_EARTH_RADIUS_EQUATOR                 (Double) 6371004,2029572
 FIX_DEGREE_IN_M [Earth degree length ]   (Double) 111195
 FIX_NAUTICAL_MILE_IN_M [Nautical mile]   (Double) 1853,25
[End of Fiximge documentation]
  

Zum Seitenanfang