US 8,896,625 B2Granted

Method and system for fusing images

AssigneeHuawei Technologies Co., Ltd.

InventorsHai Chen, Lingzhi Liu

Life of this patent

Estimating the term…

Loading patent…

Abstract

A method and a system for fusing images in an image processing field are described. The method includes: capturing at least two frames of images having different intensity under different exposure time in the same scene; fusing the images having different intensity values according to an intensity mapping relation between every two frames of images in the images having different intensity values, and obtaining a fused image is, in which the intensity mapping relation represents a corresponding relation between an intensity value of a pixel location in a bright image and that at the corresponding location in a dark image. Through the method and the system for fusing the images according to the present invention, the dynamic range of the image is enhanced, and the fusing process has a desired result and is easy to implement.

Claims14

Dependent claims

A method for fusing images, comprising:

capturing by an image capturing device at least two frames of the images having different intensity values under different exposure time in a same scene;
calculating an intensity mapping function between each two frames of the images according to an intensity histogram and an intensity cumulative histogram of each frame of the images, the intensity mapping function is a function between x and f(x), wherein x represents the intensity value of a pixel located at a certain position in a bright image of the two frames, f(x) represents the intensity value of the pixel located at the certain position in a dark image of the two frames, each group of x and f(x) in the intensity mapping function is referred to as a mapping pair and a plurality of mapping pairs are obtained, and a value obtained by subtracting f(x) from x is referred to as a mapping difference;
selecting the mapping pair from the plurality of the mapping pairs according to the maximum mapping difference EDRmax and using the selected mapping pair as an intensity mapping relation, wherein EDRmax=max{x−f(x)};
and
calculating an intensity value of each frame of the mapped images according to the determined mapping pair, calculating a fusing weight of each frame of the images according to the determined mapping pair by the formula
s
k
′
=
{
exp
(
-
4
*
I
k
2
x
max
⁢
⁢
1
2
)
k
=
1
exp
(
-
4
*
(
2
⁢
⁢
I
k
-
x
max
⁢
⁢
k
-
f
(
k
-
1
)
⁡
(
x
max
⁡
(
k
-
1
)
)
)
2
(
x
max
⁢
⁢
k
-
f
(
k
-
1
)
⁡
(
x
max
⁡
(
k
-
1
)
)
)
2
)
k
=
2
⁢
⁢
…
⁢
⁢
(
n
-
1
)
exp
(
-
4
*
(
255
-
I
k
)
2
f
(
n
-
1
)
⁡
(
x
max
⁡
(
n
-
1
)
)
2
)
k
=
n
;
wherein Ik is the intensity value of kth image in n exposure images;
and
fusing a plurality of the frames of the images by the intensity value of each frame of the mapped images and the corresponding fusing weight of each frame of the images, and obtaining the fused image.

140 more elements

A system for fusing images, comprising:

an image capturing device, configured to capture at least two frames of the images having different intensity values under different exposure time in a same scene;
a statistics module, configured to calculate an intensity mapping function between each—two frames of the images according to an intensity histogram and an intensity cumulative histogram of each frame of the images, the intensity mapping function is a function between x and f(x), where x represents the intensity value of a pixel located at a certain position in a bright image of the two frames, f(x) represents the intensity value of the pixel located at the certain position in a dark image of the two frames, each group of x and f(x) is referred to as a mapping pair and a plurality of mapping pairs are obtained, and a value obtained by subtracting f(x) from x is referred to as a mapping difference;
and configured to select the mapping pair from the plurality of the mapping pairs according to the maximum mapping difference EDRmax and use the selected mapping pair as an intensity mapping relation, wherein EDRmax=max{x−f(x)};
and
a fusing module, configured to fuse the images having different intensity values according to the intensity mapping relation and obtain a fused image;
wherein the fusing module comprises:
a mapping unit configured to calculate an intensity value of each frame of the mapped images according to the mapping pair determined by the statistics module;
a synthesis weight calculating unit, configured to calculate a synthesis weight of each frame of the images according to the mapping pair by the formula
s
k
′
=
{
exp
(
-
4
*
I
k
2
x
max
⁢
⁢
1
2
)
k
=
1
exp
(
-
4
*
(
2
⁢
⁢
I
k
-
x
max
⁢
⁢
k
-
f
(
k
-
1
)
⁡
(
x
max
⁡
(
k
-
1
)
)
)
2
(
x
max
⁢
⁢
k
-
f
(
k
-
1
)
⁡
(
x
max
⁡
(
k
-
1
)
)
)
2
)
k
=
2
⁢
⁢
…
⁢
⁢
(
n
-
1
)
exp
(
-
4
*
(
255
-
I
k
)
2
f
(
n
-
1
)
⁡
(
x
max
⁡
(
n
-
1
)
)
2
)
k
=
n
;
wherein Ik is the intensity value of kth image in n exposure images;
and
a superposing unit, configured to perform a weighting superposing according to the intensity value of each frame of the images obtained by the mapping unit and the corresponding synthesis weight calculated by the synthesis weight calculating unit, and obtain the fused image.

143 more elements

Dependencies are inferred from the claim text.

Facts

Application: 12/455,930Filed 8 Jun 2009
Priority: CN 2008 1 022506127 Oct 2008
Published: US 2010/0103194 A129 Apr 2010
Granted: 25 Nov 2014
Term ends: Not estimatedThe source records don’t support an estimate.

Classification14 codes

H04N 23/71
Circuitry for compensating brightness variation in the scene: circuitry for evaluating the brightness variation
G06T 5/50
Image enhancement or restoration: using two or more images, e.g. averaging or subtraction
H04N 23/741
Circuitry for compensating brightness variation in the scene: by increasing the dynamic range of the image compared to the dynamic range of the…
H04N 23/743
Circuitry for compensating brightness variation in the scene: bracketing, i.e. taking a series of images with varying exposure conditions