第
45
卷第
2
期
2 0 1
3
年
2
月
哈尔滨工业大学学报
JOURNAL
OF
HARBIN
INSTITUTE
OF
TECHNOLOGY
Vol.
45
No.
2
Feb.
2013
水下图像边缘特征提取的
BEMD
自适应算法
刘波
1
,2
,林焰
1
,2
,王运龙
2
(
1.
大连理工大学工业装备结构分析国家重点实验室,
116024
辽宁大连;
2.
大连理工大学船舶工程学院,
116024
辽宁大连)
摘
要:针对应用二维经验模式分解算法进行水下图像边缘检测时需要人工设定检测阅值的问题,提出一种
BEMD
与
ROC
曲线分析相结合的自适应图像边缘检测新方法.首先通过
BEMD
算法将水下图像分解成多层内禀模式函数(
IMF)
分量图像,然后利用不同参数组合的
Canny
检测算子对
IMF
分量图像进行细化处理,生成各层
IMF
分量的二值化图像
集,最后利用
ROC
曲线分析技术求得
IMF
分量图像的最佳检测阀值,从而确定了理想的
BEMD
边缘特征提取图.实验结
果表明:该算法能够避免人工设置检测阅值带来的操作误差,可实现图像边缘特征提取检测阅值的自适应设定.水下图
像处理实例验证了所提方法的正确性和有效性
关键词:水下图像;二维经验模式分解;
ROC
曲线分析;边缘检测
中图分类号:
TP391
文献标志码:
A
文章编号:
0367
-6234(201~
)
02
-0117
-06
Bi-dimensional empirical mode decomposition algorithm for underwater
image edge detecting
LIU Bo1·2, LIN
Yan1'2,
WANG
Yunlong2
(
1.
State
Key
Laboratory
of
Structural
Analysis
for
Industrial Equipment,
Dalian
University
of
Technology,
116024
Dalian,
Liaoni
吨,
China;2.
School
of
Naval
Architecture
and
Ocean
Engineering,
Dalian
University
of
Technology,
116024
Dalian,Liaoni
吨,
China)
Abstract:
A novel
method
combining
BEMD
and
receiver
operating
characteristics
(
ROC)
curve
is
presented
in
this
paper
to solve
the
problem
that
the
threshold
is greatly affected by
personal
experience
when
underwater
image
edge
detection
is
performed
using
a
bi-dimensional
empirical
mode
decomposition
(
BEMD)
algorithm.
Firstly,
the
BEMD algorithm is
employed
to
decompose
an
underwater
image
into
several
intrinsic
mode
functions
(
IMFs)
and
a
residual.
Then
several
IMF
images
are
computed
using
combinations
of
the
Canny
detector
parameters,
and
the
image
binaryzation
results
are
generated
accordingly.
The
ideal
BEMD
edge
feature
extraction
maps
are
estimated
using
correspondence
threshold
which
is
optimized
by
ROC
analysis.
The
experimental
results
show
that
the
proposed
algorithm
is
able
to avoid
the
operation
e
町
or
caused
by
manual
setting
of
the
detection
threshold,
and
to
adaptively
set
the
image
feature
detection
threshold.
The
proposed
method
has
been
proved
to
be
accuracy
and
effectiveness
by
the
underwater
image
processing
examples.
Key
words:
underwater
image;
bi-dimensional
empirical
mode decomposition;
receiver
operating
characteristics
curve
;
edge
feature
detector
水下图像边缘检测技术在海洋探测、海洋平
台安全检测和水下目标识别等领域具有广泛的应
收稿日期:
2012
-05
-28.
基金项目:国家公益性行业科研专项(
201003024)
;
辽宁省教育厅科研项目(
LS20J
0046).
作者简介:刘波(
1977
一),男,博士研究生;
林焰(
1963
一),男
L
教授,博士生导师
通信作者:林焰,
linya1
向@
dlut.
edu.
en.
用.目前,随着水下机器视觉技术的发展,高分辨
率的图像技术给人们研究水下环境带来了方便.
然而在深海环境中,水下弱光条件限制了水下光
视觉图像技术的应用.在过去的十几年中,研究者
经过不懈的努力进行了大量的水下成像设计研
究,并相信终会解决这一困难[
I-3
].现在有一些学
者开始热衷于水下声纳图像[
4
5
]和水下激光图
第
45
卷第
2
期
2 0 1
3
年
2
月
哈尔滨工业大学学报
JOURNAL
OF
HARBIN
INSTITUTE
OF
TECHNOLOGY
Vol.
45
No.
2
Feb.
2013
水下图像边缘特征提取的
BEMD
自适应算法
刘波
1
,2
,林焰
1
,2
,王运龙
2
(
1.
大连理工大学工业装备结构分析国家重点实验室,
116024
辽宁大连;
2.
大连理工大学船舶工程学院,
116024
辽宁大连)
摘
要:针对应用二维经验模式分解算法进行水下图像边缘检测时需要人工设定检测阅值的问题,提出一种
BEMD
与
ROC
曲线分析相结合的自适应图像边缘检测新方法.首先通过
BEMD
算法将水下图像分解成多层内禀模式函数(
IMF)
分量图像,然后利用不同参数组合的
Canny
检测算子对
IMF
分量图像进行细化处理,生成各层
IMF
分量的二值化图像
集,最后利用
ROC
曲线分析技术求得
IMF
分量图像的最佳检测阀值,从而确定了理想的
BEMD
边缘特征提取图.实验结
果表明:该算法能够避免人工设置检测阅值带来的操作误差,可实现图像边缘特征提取检测阅值的自适应设定.水下图
像处理实例验证了所提方法的正确性和有效性
关键词:水下图像;二维经验模式分解;
ROC
曲线分析;边缘检测
中图分类号:
TP391
文献标志码:
A
文章编号:
0367
-6234(201~
)
02
-0117
-06
Bi-dimensional empirical mode decomposition algorithm for underwater
image edge detecting
LIU Bo1·2, LIN
Yan1'2,
WANG
Yunlong2
(
1.
State
Key
Laboratory
of
Structural
Analysis
for
Industrial Equipment,
Dalian
University
of
Technology,
116024
Dalian,
Liaoni
吨,
China;2.
School
of
Naval
Architecture
and
Ocean
Engineering,
Dalian
University
of
Technology,
116024
Dalian,Liaoni
吨,
China)
Abstract:
A novel
method
combining
BEMD
and
receiver
operating
characteristics
(
ROC)
curve
is
presented
in
this
paper
to solve
the
problem
that
the
threshold
is greatly affected by
personal
experience
when
underwater
image
edge
detection
is
performed
using
a
bi-dimensional
empirical
mode
decomposition
(
BEMD)
algorithm.
Firstly,
the
BEMD algorithm is
employed
to
decompose
an
underwater
image
into
several
intrinsic
mode
functions
(
IMFs)
and
a
residual.
Then
several
IMF
images
are
computed
using
combinations
of
the
Canny
detector
parameters,
and
the
image
binaryzation
results
are
generated
accordingly.
The
ideal
BEMD
edge
feature
extraction
maps
are
estimated
using
correspondence
threshold
which
is
optimized
by
ROC
analysis.
The
experimental
results
show
that
the
proposed
algorithm
is
able
to avoid
the
operation
e
町
or
caused
by
manual
setting
of
the
detection
threshold,
and
to
adaptively
set
the
image
feature
detection
threshold.
The
proposed
method
has
been
proved
to
be
accuracy
and
effectiveness
by
the
underwater
image
processing
examples.
Key
words:
underwater
image;
bi-dimensional
empirical
mode decomposition;
receiver
operating
characteristics
curve
;
edge
feature
detector
水下图像边缘检测技术在海洋探测、海洋平
台安全检测和水下目标识别等领域具有广泛的应
收稿日期:
2012
-05
-28.
基金项目:国家公益性行业科研专项(
201003024)
;
辽宁省教育厅科研项目(
LS20J
0046).
作者简介:刘波(
1977
一),男,博士研究生;
林焰(
1963
一),男
L
教授,博士生导师
通信作者:林焰,
linya1
向@
dlut.
edu.
en.
用.目前,随着水下机器视觉技术的发展,高分辨
率的图像技术给人们研究水下环境带来了方便.
然而在深海环境中,水下弱光条件限制了水下光
视觉图像技术的应用.在过去的十几年中,研究者
经过不懈的努力进行了大量的水下成像设计研
究,并相信终会解决这一困难[
I-3
].现在有一些学
者开始热衷于水下声纳图像[
4
5
]和水下激光图