76
Contents
Falcon4 86M Cameras
172
ADCOffset
Used during FPN calibration
176
Sensor width
10752
180
nCoeffFileSize
Cofficient data size in bytes
184
spare
Reserved for future use
256
~ …
FFC data
Each pixel has 4 bytes. Bit0~bit8 is FPN Base, Bit9~bit16 is FPN delta, Bit17~Bit31
is PRNU.
The file downloaded to PC is a raw image file, which combines each FPN base, FPN delta, and PRNU
into a double word. Teledyne Dalsa provides a standalone command line application
(FFC_Codec.exe) to decode this raw image into 3 readable .tif files.
The usage of FFC_Codec.exe is as follows:
decode Source File(Binary in camera) Target Files(FPN_Base (tif), Delta
FPN(tif), PRNU FFC(tif))\n");
example: FFC_Codec.exe "decode" "FlatFieldCoefficients.tif" "fpn_base.tif"
"fpn_delta.tif" "prnu.tif";
encode Source File Files(FPN_Base (tif), Delta FPN(tif), PRNU FFC(tif))
Target File(Binary in camera)
example: FFC_Codec.exe "encode" "fpn_base.tif" "fpn_delta.tif"
"prnu.tif" "encoded.tif"
Note that when running encode, target file must exist. This is used to extract FFC header
info. The initial target file is the FFC raw file downloaded from camera.
After decode the source binary file, one can use other image processing software to view
this target tiff file
.
Camera Defect Map
The camera defect map file is a Cabernet Defect Map File is a text file that contains information on
row, column, and cluster defects found during the camera test.
There are two copies of the file loaded into the camera: a factory version that the user has read-
only access to, and a user version that can be overwritten.
The header section at the top of the file (see example below) contains the camera model number,
serial number, date when it was tested / the defects were found, and the device firmware in the
camera at the time.
The file reports all defects found under rolling shutter, followed by all defects found under global
shutter.
The information recorded for each row defect is the top row of the defect, the bottom row of the
defect, and the size (number of rows) of the defect. Most row defects are only a single row so the
top and bottom row will be the same and the size will be one.
Column defects are reported the same way as row defects except the leftmost and rightmost
defective columns are used instead of top and bottom rows for each defect.
For cluster defects the center X and Y coordinates of each defect are recorded along with the
number of defective pixels.
For all measurements the top left pixel of the image is (0, 0).
