# US 2015/0104112 A1 · Large Radius Edge-Preserving Low-Pass Filtering

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2015/0104112 A1
- Document ID: US20150104112A1
- Status: Published application
- Kind: A1
- Application number: 14/054,087
- Filed: 2013-10-15
- Published: 2015-04-16
- Earliest claimed: 2013-10-15
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Nathan Henri LEVY
- Web page: https://patentgrep.com/patent/US20150104112A1

## CPC

- G06T 5/10: using non-spatial domain filtering
- G06T 3/4015: Image demosaicing, e.g. colour filter arrays [CFA] or Bayer patterns
- G06T 5/94: based on local image properties, e.g. for local contrast enhancement
- G06T 2207/20028: Bilateral filtering
- G06T 2207/20192: Edge enhancement; Edge preservation
- G06T 2207/20208: High dynamic range [HDR] image processing
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method of large-radius edge-preserving low-pass filtering a digital signal having data points, including computing a weighted average of a signal layer at vertices spaced a distance apart such that an amount of the vertices is less than an amount of the data points; and producing, for each of the data points, a large-radius edge-preserving low-pass filtered signal based on the weighted average of the signal layer at vertices neighboring the data point.

## Claims

### Claim 1

A method of edge-preserving low-pass filtering a digital signal having data points, the method comprising:
computing a weighted average of a signal layer at vertices spaced a distance apart such that an amount of the vertices is less than an amount of the data points; and
producing, for each of the data points, a large-radius edge-preserving low-pass filtered signal based on the weighted average of the signal layer at vertices neighboring the data point.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
determining values of an object layer at the vertices, the object layer being a blurred version of the signal layer; and
storing the values of the object layer at the vertices in a memory.

### Claim 3 (depends on 2)

The method of claim 2, wherein
the digital signal is a digital image and,
the determining values of the object layer, the computing the weighted average of the signal layer and the producing the large-radius edge-preserving low-pass filtered signal each include performing a raster scan a respective layer associated with the digital image.

### Claim 4 (depends on 2)

The method of claim 2, wherein the received digital signal is an image signal, the signal layer is a y-channel component of a YUV formatted version of the image signal and the object layer is a blurred version of the signal layer.

### Claim 5 (depends on 1)

The method of claim 1, wherein the computing the weighted average of the signal layer at each of the vertices includes,
determining, for each vertex, an influence zone surrounding the vertex; and
computing, for each vertex, the weighted average of the signal layer of the data points in the signal layer within the influence zone.

### Claim 6 (depends on 5)

The method of claim 5, wherein, for each vertex, the influence zone includes the data points surrounding the vertex within a two-dimensional space having an area equal to twice a square of the distance the vertices are spaced apart.

### Claim 7 (depends on 1)

The method of claim 1, wherein the vertices of the object layer represent volume points in a digital audio signal and the large-radius edge-preserving low-pass filtered signal is a signal emphasizing differences in volume levels in the digital audio signal.

### Claim 8 (depends on 1)

The method of claim 1, wherein the weighted average of the signal layer is computed using the equation:
S

(
X
,
Y
)
=
Σ
{
(
x
,
y
)

s
.
t
.

x
-
X

<
Δ
&&

y
-
Y

<
Δ
}

s

(
x
,
y
)

w

(
x
,
y
,
X
,
Y
)
Σ
{
(
x
,
y
)

s
.
t
.

x
-
X

<
Δ
&&

y
-
Y

<
Δ
}

w

(
x
,
y
,
X
,
Y
)
where (x,y) are coordinates of the data points, (X,Y) are coordinates of the vertices, S(X,Y) is the weighted average of the signal layer at the coordinate of the vertices, Δ is the distance the vertices are spaced apart, s(x,y) is the signal layer, and w(x,y,S,Y) is a function that varies directly with a similarity between a value at data point (x,y) and a value at the vertex (X,Y).

### Claim 9 (depends on 1)

The method of claim 1, wherein the large-radius edge-preserving low-pass filtered signal is computed using the equation:
s
_

(
x
,
y
)
=
Σ
{
(
X
,
Y
)

s
.
t
.

x
-
X

<
Δ
&&

y
-
Y

<
Δ
}

S

(
X
,
Y
)

w

(
x
,
y
,
X
,
Y
)

b

(
x
,
y
,
X
,
Y
)
Σ
{
(
X
,
Y
)

s
.
t
.

x
-
X

<
Δ
&&

y
-
Y

<
Δ
}

w

(
x
,
y
,
X
,
Y
)

b

(
x
,
y
,
X
,
Y
)
where (x,y) are coordinates of the data points, (X,Y) are coordinates of the vertices, S(X,Y) is the weighted average of the signal layer at the coordinate of the vertices, Δ is the distance the vertices are spaced apart, s(x,y) is the signal layer, w(x,y,S,Y) is a function that varies directly with a similarity between a value at data point (x,y) and a value of at the vertex (X,Y), and b(x,y,X,Y) is a function that varies directly with a spatial distance between the data points and the vertices.

### Claim 10

An edge-preserving device configured to perform edge-preserving low-pass filtering on a digital signal having data points, the device comprising:
a memory configured to store data; and
a processor configured to,
compute a weighted average of a signal layer at vertices spaced a distance apart such that an amount of the vertices is less than an amount of the data points, and
produce, for each of the data points, a large-radius edge-preserving low-pass filtered signal based on the weighted average of the signal layer at vertices neighboring the data point.

### Claim 11 (depends on 10)

The device of claim 10, wherein
the processor is further configured to,
determine values of an object layer at the vertices; and
store the values of the object layer at the vertices in a memory.

### Claim 12 (depends on 11)

The device of claim 11, wherein the received digital signal is an image signal, the signal layer is a y-channel component of a YUV formatted version of the image signal and the object layer is a blurred version of the signal layer.

### Claim 13 (depends on 10)

The device of claim 10, wherein the processor is configured to compute the weighted average of the signal layer at each of the vertices by,
determining, for each vertex, an influence zone surrounding the vertex; and
computing, for each vertex, the weighted average of the signal layer of the data points in the signal layer within the influence zone.

### Claim 14 (depends on 13)

The device of claim 13, wherein, for each vertex, the influence zone includes the data points surrounding the vertex within a two-dimensional space having an area equal to a square of the distance the vertices are spaced apart.

### Claim 15

A digital signal processor configured to receive a digital signal having data points, the digital signal processor comprising:
an edge-preserving processor configured to generate an edge-preserving low-pass filtered image by,
computing a weighted average of a signal layer at vertices spaced a distance apart such that an amount of the vertices is less than an amount of the data points, and
producing, for each of the data points, a large-radius edge-preserving low-pass filtered signal based on the weighted average of the signal layer at vertices neighboring the data point; and
a dynamic range compression (DRC) processor configured to produce a compressed signal by performing dynamic range compression on the digital signal using the large-radius edge-preserving low-pass filtered signal.

### Claim 16 (depends on 15)

The digital signal process of claim 15, wherein the edge-preserving processor is further configured to,
determine values of an object layer at the vertices, the object layer being a blurred version of the signal layer; and
store the values of the object layer at the vertices in a memory.

### Claim 17 (depends on 15)

The digital signal processor of claim 15, wherein the edge-preserving processor computes the weighted average of the signal layer at each of the vertices by,
determining, for each vertex, an influence zone surrounding the vertex; and
computing, for each vertex, the weighted average of the signal layer of the data points in the signal layer within the influence zone.

### Claim 18 (depends on 17)

The digital signal processor of claim 17, wherein, for each vertex, the influence zone includes the data points surrounding the vertex within a two-dimensional space having an area equal to a square of twice the distance the vertices are spaced apart.

### Claim 19 (depends on 15)

The digital signal processor of claim 15, further comprising:
a bayer-to-yuv processor configured to generate the signal layer such that the signal layer represents a y-channel component of the received digital signal.

### Claim 20 (depends on 15)

The digital signal processor of claim 15, wherein the DRC processor is configured to provide the compressed signal to an output device such that the compressed signal has a dynamic range has been compressed to within the dynamic range available at the output unit.

### Claim 21 (depends on 1)

A non-transitory computer readable medium including a computer program comprising:
computer program instructions configured to implement the method of claim 1 when executed by a processor.
