Automatic%20Target%20Recognition%20with%20Support%20Vector%20Machines

1
Automatic Target Recognition with Support Vector
Machines

• Qun Zhao, Jose Principe

Computational Neuro-Engineering
Laboratory Department of Electrical and Computer
Engineering University of Florida
December 4, 1998
2
Overview

• Introduction to SAR ATR

• 4 Classifiers

• Experiment results

• Conclusions

3
1. Introduction
Recognition of vehicles in synthetic aperture
radar (SAR) is a difficult problem due to the low
resolution of the sensor (1 meter) and the
speckle (noise) intrinsic to the image formation.
Another difficulty is due to the operating
conditions. Vehicles can be placed in high
clutter backgrounds, partial occluded, and NEW
vehicles may be found that were not used in the
training set. Training data is always limited.
We use here the MSTAR I and II database (Veda).
4
1. Data Examples
BMP2 BTR72 T72 DS1 D7
5
2. Four Classifiers
1). Perceptron with hard limiter (perceptron
training) 2). Perceptron with sigmoids
(delta rule)
6
2. Four Classifiers

• 3). Optimal Separating Hyperplane

7
2. Four Classifiers

• 4). Support vector machine
• Training kernel-Adatron (FrieB, T., Cristianini,
  N., and Campbell, C. 1998).
• Use Gaussian Kernel.

8
3. Experiments

• 3 Target classes
• T72, BTR70, and BMP2
• Pairwise classification
• Image sizes 80 x 80. Aspect 0 180 degrees.
• Training 17 degree depression
• Number of Training samples 406
• Testing 15 degree depression
• Number of Testing samples 724

9
3. Experiments

• 1. Classification

10
3. Experiments - Recognition

• Added two more vehicles to test set. They are
  called confusers.
• Confusers 2S1 and D7
• Number of confuser images 275
• This becomes a recognition problem. The point
  PD0.9 of the receiver operating characteristics
  (ROC) is chosen for the comparison. Output of
  classifiers are thresholded to achieve PD0.9.
• Now performance is measured by error rate and
  false alarms.

11
3. Experiments - Recognition
12
4. Conclusion

• Classification and recognition are different
  problems, and the latter is more realistic (and
  hard).
• SVMs with the Gaussian kernel perform better for
  recognition. The local shape of the Gaussian
  kernel is very useful and should be utilized
  (samples that are far away from the class centers
  tend to have small feature values).
• In our problem (large input space) the optimal
  separating hyperplane performs better for
  classification.
• Kernel-Adatron easy and fast training