Flash detection
View Patent ↗There is provided a method of pre-processing video data to detect flashes, comprising calculating a normalized difference index value for a Luminance [Y] component of the video data, calculating a normalized difference index value for Chrominance [U and V] components of the video data, calculating a normalized luminance [Y] mean gradient index for the video data, and providing an indication that a flash has been detected when all the following conditions are satisfied: the normalized difference index value for the Luminance [Y] component is greater than a first predetermined threshold (T Y ); the normalized difference index value for the Chrominance [U and V] components is less than a second predetermined threshold (T C ); and a difference between the normalized difference index value for the Chrominance [U and V] components and the normalized luminance [Y] mean gradient index is less than or equal to a third pre-determined threshold (T F ).
1. A method of pre-processing video data to detect flashes,
comprising:
calculating a normalized difference index value for a Luminance [Y] component of the video data;
calculating a normalized difference index value for Chrominance [U and V] components of the video data;
calculating a normalized luminance [Y] mean gradient index for the video data, wherein the normalized luminance [Y] mean gradient index is calculated by:
G
t
=
∇
M
(
t
)
V
max
=
(
m
t
-
m
t
-
1
)
V
max
wherein m t is the value mean of a luminance picture at time t, V max is the maximal value of luminance pixels in a picture in the video data, G t is the normalized luminance [Y] mean gradient index, and ΔM(t) is the difference between m t and m t−1 ; and
providing an indication that a flash has been detected when all the following conditions are satisfied:
the normalized difference index value for the Luminance [Y] component is greater than a first predetermined threshold (T Y );
the normalized difference index value for the Chrominance [U and V] components is less than a second predetermined threshold (T C ); and
a difference between the normalized difference index value for the Chrominance [U and V] components and the normalized luminance [Y] mean gradient index is less than or equal to a third pre-determined threshold (T F ).
2. The method of claim 1 , further comprising:
ceasing to provide an indication that a flash has been detected when:
either all of the following conditions are satisfied:
the normalized difference index value for the Luminance [Y] component is greater than a fourth predetermined threshold (T N Y );
the normalized difference index value for the Chrominance [U and V] components is less than a fifth predetermined threshold (T N C ); and
the normalized luminance [Y] mean gradient index is less than or equal to a sixth pre-determined threshold (T G );
or a time period spent in the flash detected state exceeds a pre-determined time limit (T L ).
3. The method of claim 1 , wherein the normalized difference index value for the Luminance [Y] component is calculated over consecutive pictures i and j using the equation:
D
Y
(
H
i
Y
,
H
j
Y
)
=
P
Y
-
∑
k
=
1
B
min
(
H
i
Y
(
k
)
,
H
j
Y
(
k
)
)
P
Y
wherein P Y is the total number of pixels in the luminance plane of a picture in the video data, D Y is the normalized difference index value for the Luminance [Y] component, H Y i is a histogram value for an ith frame or field of a video sequence, H Y j is a histogram value for an jth frame or field of a video sequence, H Y i (k) is a histogram value in bin k for a ith frame or field of a video sequence H Y j (k) is a histogram value in bin k for a jth frame or field of a video sequence, and B is the bin number of a given histogram.
4. The method of claim 1 , wherein the normalized difference index value for Chrominance [U and V] components is calculated over consecutive pictures i and j using the equation:
D
C
(
H
i
C
,
H
j
C
)
=
P
C
-
∑
k
=
1
B
min
(
H
i
C
(
k
)
,
H
j
C
(
k
)
)
P
C
wherein P C is the total number of pixels in the chrominance plane of a picture in the video data D C is the normalized difference index value for the chrominance [U and V] component, H C i is a histogram value for an ith frame or field of a video sequence, H C j is a histogram value for a jth frame or field of a video sequence, H C i (k) is a histogram value in bin k for an ith frame or field of a video sequence, H C j (k) is a histogram value in bin k for a jth frame or field of a video sequence, and B is the bin number of a given histogram.
5. The method of claim 4 , wherein the normalized index value for the Chrominance [U and V] components are combined in the same calculation.
6. The method of claim 1 wherein the first predetermined threshold is in the range [0, 1].
7. The method of claim 1 wherein the second predetermined threshold is in the range [0, 1].
8. The method of claim 1 wherein the third predetermined threshold is in the range [−1, 2].
9. The method of claim 1 wherein the fourth predetermined threshold is in the range [0, 1].
10. The method of claim 1 wherein the fifth predetermined threshold is in the range [0, 1].
11. The method of claim 1 wherein the sixth predetermined threshold is in the range [0, 1].
12. The method of claim 1 , further comprising:
adapting a system rate control of a subsequent video data encoding operation dependent upon the indication that a flash has been detected.
13. The method of claim 1 , wherein the method is carried out prior to a video data encoding step.
14. A digital signal processor adapted to carry out the method of claim 1 .
15. An apparatus adapted to carry out the method of claim 1 .
16. A non-transitory computer readable medium, containing instructions, which, when executed by a processor, carries out the method of claim 1 .
17. A method of pre-processing video data to detect flashes, comprising:
calculating a normalized difference index value for a Luminance [Y] component of the video data;
calculating a normalized difference index value for Chrominance [U and V] components of the video data;
calculating a normalized luminance [Y] mean gradient index for the video data, wherein the normalized luminance [Y] mean gradient index is calculated by:
G
t
=
∇
M
(
t
)
V
max
=
(
m
t
-
m
t
-
1
)
V
max
wherein m t is the value mean of a luminance picture at time t, V max is the maximal value of luminance pixels in a picture in the video data, and ΔM(t) is the difference between m t and m t−1 ; and
transitioning to a flash detected state when all the following conditions are satisfied:
the normalized difference index value for the Luminance [Y] component is greater than a first predetermined threshold (T Y );
the normalized difference index value for the Chrominance [U and V] components is less than a second predetermined threshold (T C ); and
a difference between the normalized difference index value for the Chrominance [U and V] components and the normalized luminance [Y] mean gradient index is less than or equal to a third pre-determined threshold (T F ).