Foreground detection
A computing system can receive an image including foreground pixels. The foreground pixels can be determined based on determining eccentricity ε k based on a sequence of images acquired by a stationary sensor. The vehicle can determine moving objects in the image based on the foreground pixels. The vehicle can be operated based on the moving objects in the image.
1. A method, comprising:
receiving an image including foreground pixels, the foreground pixels determined based on determining an eccentricity ε k that is a measure of a difference of intensities of respective pixels from past intensity values, for a sequence of images acquired by a stationary sensor;
determining moving objects in the image based on the foreground pixels; and
operating a vehicle based on the moving objects in the image.
2. The method of claim 1 , further comprising determining eccentricity ε k based on determining a mean μ k for pixels of the sequence of images, based on a previous mean μ k-1 according to equation μ k =(1−α)μ k-1 +αx k where α is an empirically determined constant.
3. The method of claim 2 , further comprising determining the eccentricity ε k based on determining a variance σ k 2 for pixels of the sequence of images, based on a previous variance σ k-1 2 and the mean μ k according to equation
σ
k
2
=
(
1
-
α
)
σ
k
-
1
2
+
α
(
x
k
-
μ
k
)
T
(
x
k
-
μ
k
)
1
-
α
.
4. The method of claim 3 , further comprising determining the eccentricity ε k based on the mean μ k and the variance σ k 2 according to equation
ɛ
k
=
α
+
α
(
x
k
-
μ
k
)
T
(
x
k
-
μ
k
)
σ
k
2
.
5. The method of claim 4 , further comprising determining the foreground pixels by comparing ε k to α times an empirically determined constant.
6. The method of claim 1 , further comprising determining moving objects in the image based on determining connected regions of foreground pixels with empirically determined minimum and maximum areas.
7. The method of claim 6 , wherein the connected regions of foreground pixels are determined based on having similar eccentricity ε k .
8. The method of claim 1 , wherein the stationary sensor is a red, green blue (RGB) color video camera included in a traffic infrastructure system.
9. The method of claim 8 , wherein the vehicle receives the image including foreground pixels from the traffic infrastructure system via a network based on a location of the vehicle.
10. A system, comprising a processor; and
a memory, the memory including instructions to be executed by the processor to:
receive an image including foreground pixels, the foreground pixels determined based on determining an eccentricity ε k , that is a measure of a difference of intensities of respective pixels from past intensity values, for a sequence of images acquired by a stationary sensor;
determining moving objects in the image based on the foreground pixels; and
operating a vehicle based on the moving objects in the image.
11. The system of claim 10 , further comprising determining the eccentricity ε k based on determining a mean μ k for pixels of the sequence of images, based on a previous mean μ k-1 according to equation μ k =(1−α)μ k-1 +αx k .
12. The system of claim 11 , further comprising determining the eccentricity ε k based on determining a variance based on σ k 2 for pixels of the sequence of images, based on a previous variance σ k-1 2 and the mean μ k according to equation
σ
k
2
=
(
1
-
α
)
σ
k
-
1
2
+
α
(
x
k
-
μ
k
)
T
(
x
k
-
μ
k
)
1
-
α
.
13. The system of claim 12 , further comprising determining the eccentricity ε k based on the mean μ k and the variance σ k 2 according to equation
ɛ
k
=
α
+
α
(
x
k
-
μ
k
)
T
(
x
k
-
μ
k
)
σ
k
2
.
14. The system of claim 13 , further comprising determining the foreground pixels by comparing ε k to α times an empirically determined constant.
15. The system of claim 10 , further comprising determining moving objects in the image based on determining connected regions of foreground pixels with empirically determined minimum and maximum areas.
16. The system of claim 15 , wherein the connected regions of foreground pixels are determined based on having similar eccentricity ε k .
17. The system of claim 10 , wherein the stationary sensor is a red, green blue (RGB) color video camera included in a traffic infrastructure system.
18. The system of claim 17 , wherein the vehicle receives the image including foreground pixels from the traffic infrastructure system via a network based on a location of the vehicle.
19. A system, comprising:
means for controlling vehicle steering, braking and powertrain;
computer means for:
receiving an image including foreground pixels the foreground pixels determined based on determining an eccentricity ε k that is a measure of a difference of intensities of respective pixels from past intensity values, for a sequence of images acquired by a stationary sensor;
determining moving objects in the image based on the foreground pixels; and
operating a vehicle based on the moving objects in the image and based on the means for controlling vehicle steering, braking and powertrain.
20. The system of claim 19 , further comprising determining the eccentricity ε k based on determining a mean μ k based on equation μ k =(1−α)μ k-1 +αx k .