Real-Time Detection of Multiple Moving Objects in Complex Image Sequences

1
Real-Time Detection of Multiple Moving Objects in
Complex Image Sequences

• 
  International journal of imaging systems and
  technology
• 10,305-317,1999

Speaker M. Q. Jing
2
Outline

• Introduction
• Change Detection Algorithms
• Simple difference method
• Derivative model
• Shading model
• LIG model
• Binary Statistical Morphology
• Multiple object detection
• Result Conclusion

3
Introduction(1)

• Change detection (CD) is an process for many
  machine vision systes
• Track moving object
• Analyze the traffic flow
• Guide autonomous vehicles
• Task finding significant differences between
  images
• Difference may be caused by
• Motion of the camera
• Entrance or exit of an object from the scene
• Change in illuminations Bad environmental
  conditions

4
Introduction(2)

• CD is performed at pixel, edge or higher feature
  levels.
• Feature levels require more computational effort
• CD take two digital images as input. The output
  is a binary image

5
Change Detection Algorithms -- Simple
Difference(SD) method

• Input Two N x N digitized images It(x,y),
  It-1(x,y)
• Output Binary image Bt(x,y)
• Dt(x,y) It(x,y), It-1(x,y)

Fig 2 a,b
6
Change Detection Algorithms -- Derivative Model
(DM)
Fig 2 c,d
7
Change Detection Algorithms -- Shading Model (SM)

• The intensity Ip at point (x,y)
• Ip Ii Sp , where Ii illumination
• Sp shading coefficient
• Calcuate the varance of the intensity Ip/Ip-1
• 0 gt no changed
• Fig2 (e,f)

8
Change Detection Algorithms -- Local
intensity gradient(LIG) method

• Pixels at the location having a high gray-level
  gradient form a part of an object
• Nearly pixles having similar gray levels will be
  also part of the same object.
• Large negative gradients at pixels belonging to
  object boundaries

9
Change Detection Algorithms -- Local
intensity gradient(LIG) method

• 1 I -gt G(I)
• 2 G(I) -gt m x m blocks
• Limit the effects of illumination changing
• 3 for each block, calcuating the mean variance
• 4 regional means and variances are smoothed
  using the neighboring regions
• 5. then interpolated to fill an m x m again
• 6 compare with background and foregound

Fig 2 I,l
10
Binary statistical morphology

• Mathematical morphology(MM) describe images as
  sets and image-porcessing operators.
• The main drawback of such operators is the high
  sensitivity to noise
• Statistical morphology provides a noise-robust
  probabilistic generalization of MM.

11
Binary statistical morphology

• SM interprets erosion and dilation as
  probabilistic lta and wta operator
• Statistical erosion statistical dilation
  minimum variance estimators of the ouput
  distribution, P(HI)

12
Binary statistical morphology

• SD computed at an image location m as
• Binary Statistical Dilation(BSD) is obtained by
  threshold SD at value

13
(No Transcript)
14
(No Transcript)
15

• In similar way, Binary Statistical Erosion(BSE)

16
Multiple object detection
It(x,y)
Simple Difference
BCKt(x,y)
Dt(x,y)
First Level
Binary Statistical Erosion with TSE
Ki bata
Fig 4 a,b
Bt1(x,y)
17
(No Transcript)
18
Multiple object detection -- The
first step

• Temporal structuring element (TSE) 3x3 mask
• A low ki value should be selected to maximize the
  probability of detecting moving object points.
• The first level is represent the pixel where are
  object or background.

19
Multiple object detection -- The Second step
(noise reduction)
Bt1(x,y)
Binary Statistical Erosion with TSE1
Zt(x,y)
Set filtering
Yt(x,y)
Binary Statistical dilation with TSE2
Bt2(x,y)
20
Multiple object detection -- The Second step
(noise reduction)

• (a) elimnate isolated points
• (b) (c) taking into account the compactness
  constraint by favoring dense agglomerate of
  changed pixels.

Fig 4. c,d
21
Multiple object detection -- The Third step
(blob tracking)

• Blob tracking is a difficult task
• Many blobs may appear or vanish in successive
  frames
• Blobs may be occluded by overlapping of other
  blobs
• Problems are caused by the high dimensionality of
  the set of possible combinations
• See Fig 4. e

22
Multiple object detection -- The Third step
(blob tracking)

• Step1 a blob is not expected to be too far from
  where it was in the previous frame.
• Choose a mask,(2ucosa1 ) in x direction and
  2usina1 in y direction
• The mask is centerred in Is barycenter
• The blobs in I1 frame whose barycenter fall
  within this mask are selected.

23
Multiple object detection -- The Third step
(blob tracking)

• Step2
• Let CBt1(bi) set of blobs on the frame t1
  which are candidate to match with the blob bi
  detected on the frame t
• gt ???blob bi ??????frame,?? bi????blob ??

24
The cost function
? blob ?????,cost function???
25
Result-- The detection performance

• Tests on Outdoor Scenes
• Fig 5.
• The detection performance
• P(correctly identified ) / (the method detected
  )
• ???????detected?blob?,???????.
• R(correct detected )/ (original image )
• G(correctly identified ) -(false detected )/
• (original image )

Fig 7.
26
Result-- The Illumination conditions
Fig 8
27
Result-- Overflow error gap error

• r(i)1, if i belong to the
• detected blob
• rgr(i) 1, if i belong to
• the manually extracted
• blobs

Fig 9
28
Result-- Test on Noisy Images

• A random noise with uniform distrubution was
  added to some frames of the image sequence.
• Fig 10(a)
• SD and SM method could not be applied to very
  noisy images.
• Fig 10(b) is proposed method result.
• Fig 11(a) is different noise level result

29
Result-- Time performance

• The time performance of the different CD methods
  were compared.
• SD and proposed method are 10 times fast than the
  SM and DM
• 15 times faster than the LIG method

30
Conclusion

• If the system works outdoors and in badly
  illuminated enviroments,
• gt LIG , the proposed method
• If real-time is required,
• gt the proposed method

31
Fig1 Original
32
SD method
DM method
Fig 2
33
Fig 2 SM result
34
Fig 2 LIG Result
35
Fig4 the proposed result
36
Fig5 more complex
37
Fig 6 a The proposed
38
Fig6b SD result
39
Fig 6 c SM result
40
Fig 6 d LIG result
41
Fig 7 a,b
42
Fig 7 c
43
Fig 8 illum result
44
Fig 8 illum result
45
Fig 9 Overflow err
46
Fig 10 Noise