Volume 3, Special Issue 1, ICSTSD 2016
113
An approach for moving object tracking in
image processing methods
Miss. N.S.Bharti Dr.S.N.Kale Dr. V. M. Thakare
SGBAU, Amravati SGBAU, Amravati SGBAU, Amravati
India. India. India.
[email protected]
[email protected]
ABSTRACT: Image processing is processing of images using mathematical operations by using any form of signal processing for which the input is an image, a series of images, or a video ,such as photograph or video frame, the output of image processing may be either an image or a set of characteristics or parameters related to the image. This paper focus on moving object tracking method, this method is used for describe a moving object in a tracking algorithm. Color, motion information, edge and texture are common features to represent moving objects. Moving object detection method, this method is used to extract moving object in video image sequences. Detecting the moving and sounding objects via utilization of canonical correlation analysis (CCA) method, it is utilized to identify the moving objects which are most correlated to the audio signal. The search & detection of moving objects method is used for addresses the detection of moving objects that could interfere with driver behavior, either through distraction or by posing an actual danger. Digital image correlation (DIC) method using fisheye lens, this method is used to inspect the interior wall displacement and strain of hollow cylinder, the fisheye lens provides a long depth of field and wide angle of view that make it suitable for this work. The proposed method can give better performance in result.
Keywords:Moving object tracking,canonical correlation analysis, Digital image correlation, Moving object
detection, Driver behavior, fisheye lens.
I. INTRODUCTION
Moving objects tracking is one of the important tasks
in computer vision. It is used widely in visual
surveillance, intelligent transport systems, industrial
vision etc. it is one of the key techniques in intelligent
video surveillance. It is an important aspect to
describe a moving object in a tracking algorithm.
Moving object detection method, this method is used
to extract moving object in video image sequences.it
is one of the basic step of target tracking, target
classification and behavior understanding. The search
& detection of moving objects in an image sequences
using a moving camera, this method addresses the
detection of moving objects that could interfere with
driver behavior, either through distraction or by
posing an actual danger. The scene is recorded as
seen by the driver, i.e., with a moving camera. Digital
image correlation (DIC) method using fisheye lens,
this method is used to inspect the interior wall
displacement and strain of hollow cylinder, the
fisheye lens provides a long depth of field and wide
angle of view that make it suitable for this work.
II. BACKGROUND
The study on image Processing discusses the most
relevant moving object tracking techniques developed
in recent years. Moving object tracking is one of the
important task in computer vision, it is used in visual
surveillance, intelligent transport systems, industrial
vision etc. this method is used for describe a moving
object in a tracking algorithm. Color, motion
information, edge and texture are common features to
represent moving objects[1].Moving object detection
Volume 3, Special Issue 1, ICSTSD 2016
114 in video image sequences.it is one of the basic step of
target tracking, target classification and behavior
understanding [2].Detecting the moving and sounding
objects via utilization of canonical correlation
analysis (CCA) method, it is utilized to identify the
moving objects which are most correlated to the
audio signal. In addition to moving-sounding object
identification, the same framework is also exploited
to solve the problem of audio-video synchronization,
and is used to aid interactive segmentation [3].The
search & detection of moving objects in an image
sequence using a moving camera method, this
method interfere with driver behavior, either through
distraction or by posing an actual danger. The scene is
recorded as seen by the driver, i.e., with a moving
camera the camera is installed inside a conventional
vehicle driving on public roads. The moving camera
and the variable natural lighting pose a serious
challenge for calculating optical flow [4].Digital
image correlation method using fisheye lens method;
full-field optical measuring is increasingly being
popular measurement tools, such as digital image
correlation (DIC) method. DIC method is an optical
metrology that utilizes sub-pixel registration
algorithms for accurate measurement of full-field
deformation [5].
This paper introduces five methods for image
processing i.e. these are organizes as follows. Section
I Introduction. Section II discusses Background.
Section III discusses previous work. Section IV
discusses existing methodologies. Section V
discusses attributes and parameters and how these are
affected on images. Section VI proposed method and
outcome result possible. Finally section VII
Conclude this review paper.
III. PREVIOUS WORK DONE
In research literature, to improved moving object
detection, increase efficiency using recent techniques
[1][2][3][4][5].Moving object tracking method,
Moving objects tracking is often one of the important
tasks in computer vision. It is used widely in visual
surveillance, intelligent transport systems, industrial
vision etc.Color features are most widely used. But
when an object and its corresponding background
have similar color or the illumination varies rapidly,
the tracking accuracy is not ideal, Guo-wu YUAN
and Yun GAO, et al (2011) [1]. Moving objects
detection method is used to extract moving objects in
video image sequences. It's a basic step of target
tracking, target classification and behavior
understanding. Many researchers have been studying
about this issue and many algorithms of moving
objects detection are proposed, such as Background
Subtraction [1], Frame difference [2] and so on.
Either Background subtraction or Frame difference
method is hindered by the illumination change in the
scene, Chunsheng GUO and Feng XUAN,et
al(2011)[2]. Detecting the moving & sounding object
via utilization of canonical correlation analysis
(CCA) method, the canonical correlation analysis
(CCA) is utilized to identify the moving objects
which are most correlated to the audio signal, Hamid
Izadinia and Imran Saleemi,et al(2013)[3].The search
& detection of moving objects in an image sequence
using a moving camera method, the detection of
moving objects that interfere with driver behavior,
either through distraction or by posing an actual
danger. The scene is recorded as seen by the driver,
i.e., with a moving camera.The camera is installed
inside a conventional vehicle driving on public roads.
The moving camera and the variable natural lighting
pose a serious challenge for calculating optical flow,
Antonio Garcia-Dopico and José Luis Pedraza,et al
(2014)[4]. Digital image correlation method using
fisheye lens, this method is used to inspect the
interior wall displacement and strain of hollow
cylinder by digital image correlation (DIC) method
Volume 3, Special Issue 1, ICSTSD 2016
115 depth of field and wide angle of view that make it
suitable for work,Wei-Chung Wang and Chi-Hung
Hwang,et al(2014)[5].
IV. EXISTING METHODOLOGY
Many image processing methods has been
implemented over the last several decades. There are
different methodologies that are implemented for
image processing, i.e. moving object tracking method.
Moving object detection method. Moving object
tracking method [1] this method is one of the
important task in computer vision. It is an important
aspect to describe a moving object in a tracking
algorithm.it is used in visual surveillance, intelligent
transport systems, industrial vision etc. this method is
used for describe a moving object in a tracking
algorithm. Moving object detection method [2] in
video moving objects detection, the same
illumination and perspective, will lead to that moving
objects and background is nonlinearly mixed. In
which Kernel Independent Component Analysis
(KICA) algorithm is proposed to detect the video
moving objects. Detecting the moving & sounding
object via utilization of canonical correlation analysis
(CCA) method [3] This method exploits correlation
between audio-visual dynamics of a video to segment
and localize objects that are the dominant source of
audio. The canonical correlation analysis (CCA) is
utilized to identify the moving objects which are most
correlated to the audio signal .The search & detection
of moving objects in an image sequence using a
moving camera method [4] this method addresses the
detection of moving objects that could interfere with
driver behavior, either through distraction or by
posing an actual danger. The scene is recorded as
seen by the driver, i.e., with a moving camera. The
camera is installed inside a conventional vehicle
driving on public roads.Digital image correlation
method using fisheye lens method [5] this method is
used to inspect the interior wall displacement and
strain of hollow cylinder by digital image correlation
(DIC) method using fisheye lens.
V. ANALYSIS AND DISCUSSION
Moving objects tracking is often one of the important
tasks in computer vision. It is used widely in visual
surveillance, intelligent transport systems, industrial
vision etc. How to describe moving objects is a key
issue in a tracking algorithm. Color, motion
information, edge and texture are common features to
represent moving objects. Color features are most
widely used [1]. Moving objects detection method is
used to extract moving objects in video image
sequences. It's a basic step of target tracking, target
classification and behavior understanding
[2].Detecting the moving & sounding object via
utilization of canonical correlation analysis (CCA)
method, the canonical correlation analysis (CCA) is
utilized to identify the moving objects which are most
correlated to the audio signal [3]. The search &
detection of moving objects in an image sequence
using a moving camera method, this method interfere
with driver behavior, either through distraction or by
posing an actual danger. The scene is recorded as
seen by the driver, i.e., with a moving camera [4].
Digital image correlation method using fisheye lens
method, full-field optical measuring is increasingly
being popular measurement tools, such as digital
image correlation (DIC) method[5].
Moving object
detection
Techniques
Volume 3, Special Issue 1, ICSTSD 2016
116 Moving object
tracking
method.
In the region with similar color between moving objects and background, LBP texture can often achieve certain effects; and in the region of lacking texture.
The shadow region, which is widespread in a video, has lower brightness than the
corresponding background region, so it will also affect the tracking accuracy.
Moving object
detection
method.
In addition to using prior information & image object information but also make full use of the complementary information of different images. This method cannot satisfy the multi-frame image sequence super resolution processing requirements, single image blind restoration has become an effective means. Detecting the moving & sounding object via utilization of canonical correlation This segmentation can be used in higher level recognition and perception systems as it can determine motion of
This method does not actually perform video segmentation; assumes
availability of actor name and script.
analysis
(CCA).
interest in the scene.
The search &
detection of
moving objects
in an image
sequence using
a moving
camera.
The trade-offs between
efficiency and accuracy in optical flow algorithms is highlighted.
The main
drawback of most of the algorithms is their high computational and memory costs.
Digital image
correlation
method using
fisheye lens.
In this method, due to the features of fisheye lens, the area can be analyzed by DIC method was significantly increased. The omnidirectional image captured by fisheye lens cannot be used by DIC method directly.
Table : comparison between Moving object tracking method., Moving object detection method.,Detecting the moving & sounding object via utilization of canonical correlation analysis (CCA),The search & detection of moving objects in an image
sequence using a moving camera,Digital image correlation method using fisheye lens.
VI. PROPOSED METHODOLOGY
Many image processing strategies have been used in
this paper propose a new method for object tracking,
Volume 3, Special Issue 1, ICSTSD 2016
117 model and optic flow. This method uses Gaussian
mixture model (GMM) and optical flow approach for
object tracking. There are two important steps to
establish the background for model, and background
updates which separate the foreground and
background. The GMM approach consists of three
different Gaussian distributions, the average, standard
deviation and weight respectively. The advantages of
Optical Flow are quick calculations and the
disadvantage is a lack of complete object tracking.
The advantage of GMM is complete results of the
operation the disadvantage is not a complete object
tracking, GMM result of the operation complete but
disadvantages include computing for a long time with
more noise. These two methods can complement each
other and image filtering results in the successful
tracking of objects. It has variety of uses such as
video communication and compression, traffic
control, medical imaging and video editing. Optical
flow method can detect the moving object even when
the camera moves, but it needs more time for its
computational complexity, and it is very sensitive to
the noise. The motion area usually appears quite
noisy in real images and optical flow estimation
involves only local computation. So the Optical Flow
method can't detect the exact contour of the moving
object, so it can conclude that there are some
shortcomings in the traditional moving object
detection methods:
- Frame difference cannot detect the exact contour of
the moving object.
- Optical Flow method is sensitive to the noise.
GMM can be used in the context of a complex
environment while Optical Flow can be used for
quick calculation with simple background. GMM is
not a complete object tracking while Object Flow
provides complete computation tracking. The
Gaussian mixture model is a single extension of the
Gaussian probability density function. As the GMM
can approximate any smooth shape of the density
distribution, so often used in image processing in
recent years for good results. Optical flow or optic
flow is the pattern of apparent motion of objects,
surfaces, and edges in a visual scene caused by the
relative motion between an observer and the scene.
Combine the GMM with the Optical Flow method, it
can obtained the results of moving object tracking.
Combine the advantages of GMM and Optical Flow.
One of the key tasks in a tracking system is to update
the object model. In most of the tracking scenarios,
the underlying image data, the object, and the scene,
evolve over time in a temporal sequence. In such
scenarios, the assumption of a constant object or
background model over the entire sequence will lead
to an impoverished tracker which cannot handle
photometric differences and occlusions. Hence it is
essential to learn the object model and adapt
accordingly.
- Input: captured with a fixed camera containing one
or more moving objects of interest
- Processing goals: determine the image regions
where significant motion has occurred
- Combine GMM and Optical Flow algorithm.
- Foreground extraction.
- Output: an outline of the motion within the image
sequence.
- Output: an outline of the motion within the image
sequence.
Input image
Backgro und model Backgro
und update
GMM Optical
flow
Backgrou nd based detection method
Noises remove
Time gradient
image
Foreground extract
Foreground extract Math
morpho logy median
filter
Object segmentati
on
The purposed
Volume 3, Special Issue 1, ICSTSD 2016
118 Fig: flow diagram for object tracking using
Gaussian mixture model and optical flow.
VII. OUTCOME& POSSIBLE RESULT
Combine the GMM with the Optical Flow method,
then obtained the results of moving object tracking.
Combine the advantages of GMM and Optical Flow.
One of the key tasks in a tracking system is to update
the object model.
VIII. CONCLUSION
This paper focused on the study of different moving
object detection techniques i.e. Moving object
tracking method., Moving object detection
method,Detecting the moving & sounding object via
utilization of canonical correlation analysis
(CCA),The search & detection of moving objects in
an image sequence using a moving camera,Digital
image correlation method using fisheye lens. This
paper proposed the GMM and Optical Flow method
successfully applied in a continuous image.it uses the
GMM approach as the main tracking algorithm, with
morphological and median filtering to remove noise
and also it uses the optical flow method to subtract
successive images, also using morphological and
median filters to remove noise.
IX. FUTURE SCOPE:
In future this method can be modified to differentiate
different class objects in real time video. Later
characteristics are extracted and applied to a Neural
Network so that segmented objects are classified as
vehicles and non-vehicles and, in the case of vehicles,
they will be classified according to the size of the
vehicle as follows: large size, intermediate size, small
size.
REFERENCES
[1] Guo-wu YUAN and Yun GAO, "A Moving Objects Tracking Method Based On a Combination of Local Binary Pattern Texture and Hue ,"sciencedirect, PP. 3964-3968, 2011.
[2] ChunshengGUO and FengXUAN,"Moving object detection based on kernel independent component analysis,"sciencedirect,PP.1046-1050,2011.
[3] Hamid Izadinia and Imran Saleemi,"Multimodal Analysis for
Identification andSegmentation of Moving-Sounding Objects,"IEEE TRANSACTIONS ON MULTIMEDIA, Vol: 15, No. 2, PP.378-390, FEBRUARY 2013
[4] Antonio Garcia-Dopicoand José Luis Pedraza,"Locating moving objects in car-drivingsequences,"Journal,PP.1-23, January 2014.
[5] Wei-Chung Wang and Chi-Hung Hwang,"Displacement Measurement of Interior Wall of Hollow Cylinder by Digital Image Correlation Method Using Fisheye Lens ,"sciencedirect, PP.437-446,2014. Output