US 8,285,033 B2Granted

Bi-affinity filter: a bilateral type filter for color images

AssigneeSeiko Epson Corporation

InventorsMithun Das Gupta, Jing Xiao

Life of this patent

Estimating the term…

Loading patent…

Abstract

An edge preserving filter that works on the principle of matting affinity allows a better representation of the range filter term in bilateral class filters. The definition of the affinity term can be relaxed to suit different applications. An approximate bi-affinity filter whose output is shown to be very similar to the traditional bilateral filter is defined. The present technique has the added advantage that no color space changes are required and hence an input image can be handled in its original color space. This is a big benefit over the traditional bilateral filter, which needs conversion to perception based spaces, such as CIELAB, to generate results close to the present invention. The full bi-affinity filter preserves very minute details of the input image, and thus permits an enhanced zooming functionality.

Claims24

Dependent claims

A method of processing a digital, input image, comprising:

providing said digital input image, I, in physical memory, said input image having color information in a native color space for each image pixel;
providing a processing device coupled to said physical memory to access said input image I and create an output image, h, by implementing the following steps:
(a) applying the following sequence to said input image I,
h
σ
,
ɛ
⁡
(
x
)
=
∑
y
∈
Ω
x
⁢
⁢
(
f
s
σ
⁡
(
x
,
y
)
⁢
L
xy
ɛ
⁢
I
⁡
(
y
)
)
∑
y
∈
Ω
x
⁢
⁢
(
f
s
σ
⁡
(
x
,
y
)
⁢
L
xy
ɛ
)
where x and y are pixel locations over an image grid, and Ωx is the neighborhood induced around the central pixel x;
fs(x,y) measures the spatial affinity between pixels at x and y;
parameter σ controls an amount of spatial blurring;
Lxyε is a laplacian matrix that provides positive color affinity value for two examined pixels, x and y, that have the same color and provides a zero color affinity value for pixels with different color;
parameter ε is a regularization term whose relative weight determines the smoothness of output image h.

64 more elements

A method of enlarging an input image I, comprising:

providing said digital input image, I, in physical memory, said input image having color information in a native color space for each image pixel;
providing a processing device coupled to said physical memory to access said input image I and create an output image, h, by implementing the following steps:
(a) applying the following sequence to said input image I,
h
σ
,
ɛ
⁡
(
x
)
=
∑
y
∈
Ω
x
⁢
⁢
(
f
s
σ
⁡
(
x
,
y
)
⁢
L
xy
ɛ
⁢
I
⁡
(
y
)
)
∑
y
∈
Ω
x
⁢
⁢
(
f
s
σ
⁡
(
x
,
y
)
⁢
L
xy
ɛ
)
where:
(i) x and y are pixel locations over an image grid;
(ii) Ωx is the neighborhood induced around the central pixel x;
(iii) fs(x,y) measures the spatial affinity between pixels at x and y;
(iv) parameter σ controls an amount of spatial blurring;
(v) Lxyε a laplacian matrix that provides positive color affinity value for two examined pixels, x and y, that have the same color and provides a zero color affinity value for pixels with different color;
(vi) parameter ε is a regularization term whose relative weight determines the smoothness of output image h;
(vii) laplacian matrix L is defined as a diagonal matrix, D, and a weight matrix, W, with the formulation L=D−W, where diagonal matrix D is further defined as Dii=#[k|iεwk] at its diagonal, which represents the cardinality of the number of windows of which the pixel i is a member, individual terms of the weight matrix W are given by
W
ij
=
∑
k
❘
(
i
,
j
)
∈
w
k
⁢
1

w
k

⁢
(
1
+
(
I
i
-
μ
k
)
⁢
(
σ
k
+
ɛ

w
k

⁢
I
3
)
-
1
⁢
(
I
j
-
μ
k
)
)
and weight matrix Wij is evaluated over all possible overlapping windows that contain the center pixel;
and
(b) interpolating said input image I to a resolution higher than its native resolution;
(c) combining the results of step (a) with the interpolated, higher resolution image of step (b).

127 more elements

Dependencies are inferred from the claim text.