Transcript 下載/瀏覽
Detecting Moving Objects,
Ghosts, and Shadows in
Video Streams
Rita Cucchiara, Costantino Grana, Massimo Piccardi and Andrea Prati
Adviser : Chih-Hung Lin
Speaker : Kuan-Ju Chen
Date
: 2009/02/19
1
Author
Author
R. Cucchiara, C. Grana, and A. Prati are with the
Dipartimento di Ingegneria dell’Informazione,
Universita` di Modena e Reggio Emilia, Via Vignolese,
905/b, 41100 Modena, Italy.
. M. Piccardi is with Department of Computer Systems,
Faculty of IT, University of Technology, Sydney,
Broadway NSW 2007, Australia.
Accept
IEEE TRANSACTIONS ON PATTERN ANALYSIS
AND MACHINE INTELLIGENCE, VOL. 25, NO. 10,
OCTOBER 2003
revised 23 Sept. 2002; accepted 25 Feb.2003
2
Contents
1
INTRODUCTION
2
DETECTING MOVING OBJECTS
3
SHADOW DETECTION
4
RESULTS
5
CONCLUSIONS
3
INTRODUCTION
Detection of moving objects in video
streams is the first relevant step of
information extraction in many
computer vision applications
Video surveillance
People tracking
4
INTRODUCTION
In this work, the models of the target
objects and their motion are unknown
the most widely adopted approach for
moving object detection with fixed
camera is based on background
subtraction
5
INTRODUCTION
Background Subtraction
Two problems
• Accurately
• Reactivity
If the background model is neither
accurate nor reactive
Ghosts
Shadows
6
INTRODUCTION
Solution:
Statistical combination
• Combination frames to compute the background
model
Combine
• Combine the current frame and previous models
with recursive filtering to update the background
model
7
INTRODUCTION
In this paper, we propose a novel simple method
that exploits all these features, combining them
so as to efficiently provide detection of moving
objects, ghosts, and shadows.
The main contribution
the integration of knowledge of detected objects,
shadows, and ghosts in the segmentation
process
enhance object segmentation and background
update.
8
DETECTING MOVING OBJECTS
The aim
Detect real moving objects.
Avoiding detection of transient spurious
objects
Sakbot (Statistical And KnowledgeBased ObjecT detection)
Statistics and knowledge of the segmented
objects to improve both background modeling
and moving object detection
9
DETECTING MOVING OBJECTS
MVO (Moving visual object)
set of connected points belonging to object
characterized by nonnull motion
Ghost
a set of connected points detected as in motion by
means of background subtraction, but not
corresponding to any real moving object
Shadow
a set of connected background points modified by a
shadow cast over them by a moving object
10
DETECTING MOVING OBJECTS
Original
Image
Background
Foreground
Shadow
Object
MVO
Shadow
Ghost
Shadow
MVO
Ghost
Taxonomy
11
DETECTING MOVING OBJECTS
Frame t
background model
Foreground
12
DETECTING MOVING OBJECTS
Get a statistical information
Follow set of
S
t t
the statistical background model Bs p
Using the median function get a statistical information: Bkt t p
If p belong to MVO , p=background model
If p dosen`t belong to MVO, p= statistical background model
13
DETECTING MOVING OBJECTS
t t
used to update the background model to update background model B p
If p isn`t belong to known object , p= Bs p
If p is belong to known object , p= Bkt t p
t t
update background model
14
SHADOW DETECTION
Mean the process of classification of
foreground pixels as “shadow points”
based on their appearance
15
SHADOW DETECTION
Foreground
Ayalyze Hue-Saturation-Value(HSV) color space
Following three condition to mask shadow
identifying as shadows those points
define a darkening effect of shadows
I t p .V
t
t
p .S s D H H ; 0,1, 0,1
1
if
I
p
.
S
B
t
t
shadow mask:SP p
B p .V
0
otherwise
average image luminance
16
RESULTS
The reactivity of the background model.
17
RESULTS
Segmentation is provided via background subtraction including shadow detection.
(a) false positives and (b) false negatives.
18
CONCLUSIONS
This paper has presented Sakbot, a system
for moving object detection in image
sequences
This system has the unique characteristic of explicitly
addressing various troublesome situations such as
cast shadows and ghosts.
Cast shadows are detected and removed from the
background update function
Ghosts are also explicitly modeled and detected so as
to avoid a further cause of undesired background
modification
The method is highly computationally costeffective since it is not severe in
computational time
19
Add your company slogan
20
DETECTING MOVING OBJECTS
MVOs are validated by applying a set of
rules on area, saliency, and motion as
follows:
The MVO blob must be large enough
The MVO blob must be a “salient” foreground
blob
The MVO blob must have nonnegligible
motion
21
DETECTING MOVING OBJECTS
• Validation rules
22
Follow set of S
S I p , I
t
t - t
p ,..., I
t - nt
p b B p
t
the statistical background model:
Bst t p arg i 1,..., k min
Distance x , x
k
i
j 1
j
xi , x j S
knowledge-based background model:
t
t
t
B
p
if
p
MVO
MVO
SH
Bkt t p t t
t
t
B
p
if
p
G
G
SH
s
update the background model:
t t
p
B
t t
s
B p t t
Bk p
if O KOt : p O
otherwise
23