Method and apparatus for identifying an imaging device
View Patent ↗A new technique for identifying from images a camera, or other imaging device such as a scanner, is based on the device's reference noise pattern, a unique stochastic characteristic of all common digital imaging sensors, including CCD, CMOS (Foveon™ X3), and JFET. First, one determines from images the sensor's reference pattern noise, which uniquely identifies each sensor. To identify the sensor from a given image, the presence or absence of the reference pattern noise in the image under examination is established using a correlation detector or other means.
1. A method for identifying an electronic image sensor from at least one electronic image comprising the steps of:
determining from said electronic image sensor a reference noise pattern
searching said at least one electronic image for said reference noise pattern; and
identifying presence or absence of said reference noise pattern in said at least one electronic image, whereby presence of said reference noise pattern identifies said electronic image sensor as having acquired said at least one electronic image,
wherein said determining:
makes use of dark frame subtraction and flat fielding,
employs a plurality of electronic images, or
comprises a step of denoising said at least one electronic image with a denoising filter.
2. The method of claim 1 , wherein said step of searching makes use of a correlation detector.
3. The method of claim 1 , wherein said step of determining employs a plurality of electronic images.
4. The method of claim 1 , wherein said step of determining comprises a step of denoising said at least one electronic image with a denoising filter.
5. The method of claim 4 , wherein said step of denoising extracts from said at least one electronic image a Gaussian noise with known variance σ 2 and σ>1.
6. The method of claim 4 , wherein said step of denoising extracts from said at least one electronic image a Gaussian noise with known variance σ 2 and σ=5.
7. The method of claim 2 , wherein a correlation ρ c between image noise Y-Fσ(Y) and said reference noise pattern P c is calculated by the formula:
ρ
c
(
Y
)
=
corr
(
Y
-
F
σ
(
Y
)
,
(
P
c
)
=
(
Y
-
F
σ
(
Y
)
-
E
{
Y
-
F
σ
(
Y
)
}
)
·
(
P
c
-
E
{
P
c
}
)
Y
-
F
σ
(
Y
)
-
E
{
Y
-
F
σ
(
Y
)
}
P
c
-
E
{
P
c
}
.
8. The method of claim 1 , wherein said step of determining is applied to said electronic image sensor that is based on semiconductor technology.
9. The method of claim 1 , wherein said step of determining is applied to said digital imaging sensor that comprises at least one of a CCD, CMOS, and JFET sensor.
10. Apparatus for identifying at least one electronic image sensor from at least one electronic image comprising:
means for determining a reference noise pattern from said at least one electronic image sensor;
means for searching said at least one electronic image for said reference noise pattern;
means for identifying presence or absence of said reference noise pattern in said at least one electronic image, whereby presence of said reference noise pattern identifies said at least one electronic image sensor as having acquired said at least one electronic image,
wherein said means for determining comprises a denoising filter, said means for searching comprises a correlation detector, or said means for searching further comprises means for applying dark frame subtraction and flat fielding.
11. Apparatus as in claim 10 , wherein said means for determining comprises a denoising filter.
12. Apparatus as in claim 10 , wherein said means for searching comprises a correlation detector.
13. Apparatus as in claim 12 , wherein said correlation detector calculates a correlation ρ c between image noise Y-Fσ(Y) and said reference noise pattern P c according to the formula:
ρ
c
(
Y
)
=
corr
(
Y
-
F
σ
(
Y
)
,
(
P
c
)
=
(
Y
-
F
σ
(
Y
)
-
E
{
Y
-
F
σ
(
Y
)
}
)
·
(
P
c
-
E
{
P
c
}
)
Y
-
F
σ
(
Y
)
-
E
{
Y
-
F
σ
(
y
)
}
P
c
-
E
{
P
c
}
.
14. Apparatus as in claim 10 , wherein said electronic image sensor is based on semiconductor technology.
15. Apparatus as in claim 10 , wherein said at least one digital imaging sensor is a one of a CCD, CMOS, and JFET sensor.
16. Apparatus as in claim 10 , wherein said at least one digital imaging sensor functions as an integral part of a one of a scanner, a digital videocamera, and a digital still camera.
17. Apparatus as in claim 10 , wherein:
said at least one digital imaging sensor comprises at least two digital imaging sensors adapted to together produce an electronic image;
said determining means determining a respective reference noise pattern from each of said electronic image sensors;
said searching means searching said at least one electronic image for said respective reference noise patterns; and
said identifying means identifying presence or absence of said respective reference noise patterns in said at least one electronic image, whereby presence of said respective reference noise patterns identifies said set of at least two digital imaging sensors as having acquired said at least one electronic image.