1. Mobile Model Car Navigation Based on Wireless Video Camera and Laser Sensors

2. Contents Team introduction Project Introduction Applicability
The whole picture
Discussing tracking algorithms
A project development
Results

3. Team introduction Supervisor: Professor Ehud Rivlin .
Instructor: Igor Katsman.
ISL Engineer: Yekutiel Pekar
Students : Alexander Sherman
Valentin Rozental

4. Introduction
Mobile navigation on unknown environment is an interesting and challenging task.
Consists of several computer vision based subtasks.
Each subtask is a project itself.

5. Where it could be used Space Exploring Medical applications
Military applications
Security applications
Civil Engineering
Science …

6. Subtasks Range-exploring algorithm for a mobile model. Tools
On-board wireless video camera
Laser sensors.
distance

7. Control interface between a mobile model and a PC.
Tools
RC CAR remote control software module

8. The most interesting task
Automatic tracking algorithm for a mobile model which would utilize an on-board wireless video camera to follow another leading mobile model.
Tools
on-board wireless video camera.

9. Lets focus on tracking algorithm
General Scheme
server
frame
target
Turn Left
Lets focus on tracking algorithm

10. Tracking algorithm features
Simple
Its should be real-time algorithm
Robust
It should focus on mobile features recognition
Correct
It should minimize misclassifications, “loosing” target, etc.

11. First steps Matlab environment acknowledgement Edge detection
Contour recognition
Noises

12. Known Algorithms discussion
Temporal Differencing
Background Subtraction
Motion Regions Detection
Motion Region Classification
Velocity Vector Computation

13. Advantages Simple Robust for static camera tracking
Background subtraction eliminates all static background’s pixels and moving objects could be easily detected.
No information is needed about a starting position
BUT …

14. Disadvantage A camera is in significant motion all the time.
NO STATIC BACKGROUND presents.

15. Solution Image stabilization algorithm
Drawbacks
It computationally expensive
Hard to get real-time speed

16. Conclusion Template Correlation matching algorithm may be used
Another question :
What is good template feature ???

17. Contour Advantages: Disadvantages: depends lightly on light conditions
good for the simple objects
Disadvantages:
Hard to detect if the noise level is high
Complex forms in different visual angles
A big variety of such forms

18. Color Advantages: Disadvantages: avoid the noise problem
more effective use of filters
Don’t depend on object form
Easier to detect the object location
center of mass
Disadvantages:
depends heavily on light conditions
Color is a feature that is widely distributed

19. Size Advantages : Disadvantages: Easy to detect if the object is big
Doesn’t depend on light conditions
Disadvantages:
Easy to misclassify the object if the object is small
Very sensitive to angle of sight
Depends heavily on the object distance

20. Template Advantages: Disadvantages: Easy to scan the image
Could be done by partial match
Less sensible to noises
Disadvantages:
A variety of templates must be provided and (sequent/parallel) diagnosed
Not a real-time method

21. But still there are good choices !
NO “SILVER BULLET” !
But still there are good choices !

22. A choice Color Simplicity is desirable Accuracy is mandatory
Real-time performance is crucial
Color

23. Color Histogram As flexible as memory-based methods
more compact representation
Estimation is trivial
Color histograms for an image are built from pixel values in one of color spaces.
Different weight to different color ranges
increases probability of quick and precise target detection.

24. Algorithm When new image becomes available A local search is performed
The best match is found
Velocity vector is calculated

25. Matching For each searched area Get a color histogram Normalize it.
Find an intersection with previously stored template histogram
minimize differences

26. Get a color histogram

27. Normalize and compare
If result < best result
best result = result

28. Making the tracking more robust and
efficient
Template updating
Template history management
Motion area prediction

29. Template updating
Solves disadvantage of great sensibility to light conditions
Ensures that current template accurately represents the new image of object
A method
merging the previous instance of template with current information we have

30. Template History Management
Solves partially or full occlusion of the object
Prevents from misclassification and “loosing” the target
A method
merging the previous instance of template with current one
A history parameter is estimated empirically

31. R(n) = a * M(n) + (1 - a) * R(n – 1)
For example:
R(n) = a * M(n) + (1 - a) * R(n – 1)
where
R(n) – is a template at n-th stage
M(n) – is a new motion detected
a – is a history parameter

32. Motion Area Prediction
Processing a whole picture is
computationally very expensive and is inapplicable to real-time applications
or requires specialized hardware to operate in the real-time domain.
To overcome this obstacle we use a motion region prediction that reduces search area.

33. A Method The 2D image velocity vector of the
target (u, v) (pixels/frame) is approximately
determined
Calculating the difference between centroid of the previous template R(n-1) and centroid of the new template R(n)

34. In the area of centroid exhaustive search is performed
In the area of centroid exhaustive search is performed. In case this search fails then area is enlarged in such way that ensures object detection.

35. First results
First we tried to track simple movements in slow motions

36. Back and Forward motions
Diagonal motions

37. Car come across
Parallel motion

38. Car is approaching
Car is moves away

39. Making things more complicated
Then we decided to add some obstacles to see how we deal with them

40. Car is partially occluded
Car is occluded by the obstacle of the same color range

41. Final Exam  freestyle 

42. A pleasant surprise
Although we concentrated on car tracking it still capable of tracking other objects
A HELICOPTER

43. AND finally some FUN
See
anybody
you
know?
  

44. Conclusions:
Using template correlation matching has three main advantages:
Continuous tracking despite occlusions and cessation of target motion.
Prevents template drifting onto background texture.
Provides robust tracking.

45. Although a module show good results it has several drawbacks
Although a module show good results it has several drawbacks. Some of them are:
Hard to detect if the noise level is high.
The probability to misclassifications of the object is great, since the color is a feature that is widely distributed.

46. Suggestions:
To overcome these drawbacks gradient-based algorithm might be implemented as secondary stage recognition. This will strongly decrease an probability for misclassification and “loosing” a target.

47. To be continued ...