Calibrating The Camera: Flat Field Correction - Teledyne 4M User Manual

40

Calibrating the Camera: Flat Field Correction

N ote: FFC calibration can only be p erform ed on the fu ll w ind ow (4 : 3 and 1 :1 asp ect ratios).
This cam era has the ability to calcu late correction coefficients in ord er to rem ove non -u niform ity in the
im age. This vid eo correction op erates on a p ixel-by-p ixel basis and im p lem ents a tw o p oint correction for
each p ixel. This correction can red u ce or elim inate im age d istor tion cau sed by the follow ing factors:



Correction is im p lem ented as follow s:
V =[(V
output
w here
The algorithm is p erform ed in tw o step s. The fixed offset (FPN ) is d eterm ined first by p erform ing a n
averaging w ithou t any light. This calibration d eterm ines exactly how m u ch offset to su btract p er p ixel in
ord er to obtain flat ou tp u t w hen the sensor is not exp osed .
The w hite light (PRN U) calibration is p erform ed next to d eterm ine the m u ltip lication factors requ ired to
bring each p ixel to the requ ired valu e (target) for flat, w hite ou tp u t.
It is im p ortant to d o the FPN correction first. Resu lts of the FPN correction are u sed in the PRN U
p roced u re. We recom m end that you rep eat the correction w hen a tem p eratu re change greater than 10 °C
occu rs.
03-032-20107-01
Fixed Pattern N oise (FPN )
Photo Resp onse N on Uniform ity (PRN U)
Lens and light sou rce non -u niform ity
– FPN (per pixel) – global FPN) * PRNU (per pixel) – Background Subtract] x System Gain
input
V
output
V
input
PRN U( p ixel)
FPN ( p ixel )
Backgrou nd
Su btract
System Gain
Falcon2 4M, 8M, and 12M Camera User's Manual
= d igital ou tp u t p ixel valu e
= d igital inp u t p ixel valu e from the sensor
= PRN U correction coefficient p er p ixel
= FPN correction coefficient p er p ixel
= backgrou nd su btract valu e
= d igital gain valu e
Teledyne DALSA