没有合适的资源?快使用搜索试试~ 我知道了~
关于视频监控领域步态识别方法的一篇综述(英文)
4星 · 超过85%的资源 需积分: 9 55 下载量 43 浏览量
2010-03-19
09:44:11
上传
评论 14
收藏 948KB PDF 举报
温馨提示
试读
13页
步态识别是视频监控中远距离识别人的方法之一。本文是这方面非常详尽优秀的综述。内容涉及了步态识别的一整套方法。
资源推荐
资源详情
资源评论
D
r. Walker arrives at the high-security research
facility where he works, eager to see the results of
his latest experiments. To access his office, he has
to undergo an authentication process. The main
entrance is at the end of a well-lit corridor,
20 m in length, equipped with several cameras. Dr. Walker
walks steadily towards the entrance. As he gets close, his gait is
recognized, the door opens automatically, and the intelligent
system that manages the building welcomes him with a friend-
ly, albeit synthesized, voice.
Although today there is no practical system that can support the
above authentication scenario, the latest research on gait-based
identification—identification by observation of a person’s walking
style—provides evidence that such a system is realistic and is likely
to be developed and used in the years to come. The study of human
gait, as well as its deployment as a biometric for identification pur-
poses, is currently an active research area. This article outlines the
application of gait technologies for security and other purposes.
IDENTIFICATION METHODS
Despite the imperative need for efficient security architectures in
airports, border crossings, and other public access areas, most cur-
rently deployed identification methods were developed and estab-
lished several years ago. It is now clear that these methods cannot
cover contemporary security needs. Moreover, in some cases, such
as in metropolitan public transport stations, authentication or
IEEE SIGNAL PROCESSING MAGAZINE [78] NOVEMBER 2005 1053-5888/05/$20.00©2005IEEE
© PHOTO DISC
[
Nikolaos V. Boulgouris, Dimitrios Hatzinakos,
and Konstantinos N. Plataniotis
]
Gait
Recognition:
A challening signal
processing
technology for
biometric
identification
Gait
Recognition:
A challening signal
processing
technology for
biometric
identification
Authorized licensed use limited to: Nanjing University. Downloaded on April 12, 2009 at 03:17 from IEEE Xplore. Restrictions apply.
verification using conventional technologies is practically infeasi-
ble. For the above reasons, the development and deployment of
biometric authentication methods (fingerprint, hand geometry,
iris, face, and gait identification) has recently attracted the atten-
tion of government agencies and other institutions. Gait analysis
and recognition can form the basis of unobtrusive technologies for
the detection of individuals who represent a security threat or
behave suspiciously. The concerns that arise from the capturing
and analysis of gait in surveillance applications are outlined in the
“Privacy and Other Concerns” sidebar.
GAIT VERSUS OTHER BIOMETRIC TRAITS
Compared to other biometrics, gait has some unique characteris-
tics. The most attractive feature of gait as a biometric trait is its
unobtrusiveness, i.e., the fact that, unlike other biometrics, it can
be captured at a distance and without requiring the prior consent
of the observed subject. Most other biometrics such as finger-
prints [1], face [2], hand geometry [3], iris [4], voice [5], and sig-
nature [6] can be captured only by physical contact or at a close
distance from the recording probe. Gait also has the advantage of
being difficult to hide, steal, or fake.
Although the study of kinesiological parameters that define
human gait can form a basis for identification, there are appar-
ent limitations in gait capturing that make it extremely difficult
to identify and record all parameters that affect gait. Instead,
gait recognition has to rely on a video sequence taken in con-
trolled or uncontrolled environments. Even if the accuracy
with which we are able to measure certain gait parameters
improves, we still do not know if the knowledge of these param-
eters provides adequate discrimination power to enable large-
scale deployment of gait recognition technologies. Moreover,
studies report both that gait changes over time and that it is
affected by clothes, footwear, walking surface, walking speed,
and emotional condition [7]. The above facts impose limitations
on the inherent accuracy of gait and question its deployment as
a discriminative biometric.
The ambiguity regarding the efficiency of gait-assisted identifica-
tion differentiates gait from other biometrics whose uniqueness and
invariability, and therefore appropriateness for use in identification
applications, can be more conclusively determined by the study of
the similarities and differences between biometrics captured from
several subjects under varying conditions. This is why, at present,
gait is not generally expected to be used as a sole means of identifica-
tion of individuals in large databases; instead, it is seen as a poten-
tially valuable component in a multimodal biometric system.
GAIT AS MULTIBIOMETRIC COMPONENT
Research conducted thus far in the area of gait recognition has
shown that gait can be reliable in combination with other bio-
metrics. If we assume that palm, fingerprint, and iris methods
belong to a different (obtrusive) class of biometrics, additional
biometrics that could be used in conjunction with gait in a
multibiometric system would be face and foot pressure [8] (the
latter requiring some specialized equipment for measuring the
ground reaction force).
IEEE SIGNAL PROCESSING MAGAZINE [79] NOVEMBER 2005
PRIVACY AND OTHER CONCERNS
Although contemporary security needs, as outlined previ-
ously, largely necessitate the use of biometrics as a means of
identification, several concerns have been raised regarding
the wide deployment of biometric systems. The most impor-
tant of these concern the fact that surveillance infrastruc-
tures might violate a citizen’s right to anonymity and invade
his/her privacy. In the particular case of gait recognition,
these concerns are even more pronounced due to the unob-
trusiveness of the gait capturing process, which could allow
continuous monitoring and recording of all traffic in public
places. Apart from this, there is a growing concern that bio-
metrics might be used for purposes beyond the original
scope or that unauthorized persons may gain access to bio-
metric information and use it for unlawful purposes. Others
are concerned about parameters that affect gait, such as
fatigue, injuries, and psychological condition; this may make
the gait of a lawful person resemble the gait of a suspicious
person. In such cases, a lawful person may be held, ques-
tioned, and possibly banned from boarding a flight or
accessing a public place. Other concerns will also be raised if
radiation is used to get an accurate picture of the human
body. Such technologies, which are currently tested in a few
airports for the purpose of detecting dangerous objects hid-
den under passengers’ clothes, are likely to be met with
objections by passengers who fear that the process is not
safe for their health or who feel uncomfortable with having
pictures of their body seen by airport officers. Finally, since
the study of gait patterns might possibly reveal medical con-
ditions, the compilation of gait databases will be regarded
as a threat to the medical privacy of individuals.
Some of the above concerns sound reasonable and
should generally be expected. The major concern about
continuous monitoring is largely unfounded. Gait-based
technologies could be used in verification applications in
controlled environments or in other monitoring applica-
tions, but at present it is technically infeasible to record all
gait parameters of a person walking in a public place and
identify him/her by searching in a database of thousands of
subjects. A conceivable application could be the discrimina-
tion between classes of people (gender, age) for statistical
purposes. But the implementation of a large-scale surveil-
lance system that continuously identifies and monitors all
individuals in public places is not possible, at least in the
foreseeable future.
The other concerns are largely unfounded as well. It is
unlikely that dangerous rays will ever be used for gait cap-
turing since this would represent a greater threat to the
population than the one the technology tries to deter.
Moreover, regarding medical privacy, gait is not more
revealing than other biometrics. Iris, palm, and fingerprints
can also disclose medical conditions. In the case of gait,
however, the large volume of information that exists in a
gait sequence prohibits the storage of all the information.
Instead, the most practical approach would be that only
information that is useful for recognition will be retained,
rendering the possibility of gait being used for diagnostic
purposes even more remote.
Authorized licensed use limited to: Nanjing University. Downloaded on April 12, 2009 at 03:17 from IEEE Xplore. Restrictions apply.
In a multibiometric system, gait and foot pressure could be
used to narrow down the database of subjects. Subsequently,
face recognition could be used for identification of a test subject
among the reduced set of candidate subjects. Otherwise, the
three biometrics could be combined altogether, e.g., using the
techniques described in [9].
A system fusing gait and ground reaction force was presented
in [10]. The combination of gait with face recognition was
examined in [11] and [12]. In [12], it was shown that gait is
more efficiently utilized in a multimodal framework when it is
combined directly with the facial features rather than preceding
the face recognition module as a filter. In both works, concrete
recognition performance gains were reported compared to using
face or gait alone. The above results indicate that there is much
value in combining gait with other biometrics.
TERMINOLOGY
Despite the differences among walking styles, the process of walk-
ing is similar for all humans. A typical sequence of stances in a gait
cycle is shown in Figure 1. A detailed analysis of gait phases can be
found in [13]. For simplicity, we consider the following four main
walking stances [14]: right double support (both legs touch the
ground, right leg in front), right midstance (legs are closest togeth-
er, right leg touches the ground), left double support, and left mid-
stance. Although some other definitions would also be appropriate,
in this article we define a gait cycle as the interval between two con-
secutive left/right midstances. The interval between any two con-
secutive midstances is termed half cycle. The time interval in which
a gait cycle is carried out is called the gait period, whereas the walk-
ing frequency is termed the fundamental gait frequency.
A GENERIC GAIT RECOGNITION SYSTEM
Gait recognition is a multistage process (see Figure 2). It is
important that gait capturing is performed in environments
where the background is as uniform as possible. Moreover, since
gait recognition algorithms are not, in general, invariant to the
capturing viewpoint, care must be taken to conduct capturing
from an appropriate viewpoint. Preferably, the walking subject
should be walking in a direction perpendicular to the optical axis
of the capturing device since the side view of walking individuals
discloses the most information about their gait. Once a walking
sequence is captured, the walking subject is separated from its
background using a process called background subtraction. A
critical step in gait recognition is feature extraction, i.e., the
extraction, from video sequences depicting walking persons, of
signals that can be used for recognition. This step is very impor-
tant since there are numerous conceivable ways to extract signals
from a gait video sequence, e.g., spatial, temporal, spatiotempo-
ral, and frequency-domain feature extraction. Therefore, one
must ensure that the feature extraction process compacts as
much discriminatory information as possible. Finally, there is a
recognition step, which aims to compare the extracted gait sig-
nals with gait signals that are stored in a database. Apart from
the apparent usefulness of gait analysis in biometric applications,
gait has several important nonbiometric applications that are
summarized in the “Nonbiometric Applications of Gait” sidebar.
PREVIOUS WORK
The study of gait as a discriminating trait was first attempted a few
decades ago from a medical/behavioral viewpoint [16], [17]. Later,
several attempts were made to investigate the gait recognition
problem from the perspective of captur-
ing and analyzing gait signals [18]–[22].
Most recent work investigating the
appropriateness of gait as a biometric for
human identification has taken place in
the context of the HumanID project
sponsored by the U.S. Defense Advanced
Research Project Agency (DARPA). Each
of the participating institutions has
established its own database of
sequences depicting humans walking.
IEEE SIGNAL PROCESSING MAGAZINE [80] NOVEMBER 2005
[FIG1] Several stances during a gait cycle. The silhouettes are from CMU’s MoBo database [15].
Midstance Midstance MidstanceDouble Support Double Support
[FIG2] General block diagram of a gait recognition/authentication system.
Background
Subtraction
Feature
Extraction
Recognition
Database
Authorized licensed use limited to: Nanjing University. Downloaded on April 12, 2009 at 03:17 from IEEE Xplore. Restrictions apply.
剩余12页未读,继续阅读
资源评论
- tp612_ingodwt2013-10-15对步态识别的发展做了比较清晰的概述 不错 适合初学者看看
- qq_154201352015-01-19不错 适合初学者看看
- myw1983512015-01-16不错,听详细的
sunryuan
- 粉丝: 0
- 资源: 1
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功