Automatic Target Recognition in high resolution Optical Aerial Images - PowerPoint PPT Presentation

Loading...

PPT – Automatic Target Recognition in high resolution Optical Aerial Images PowerPoint presentation | free to download - id: b6645-ZDc1Z



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Automatic Target Recognition in high resolution Optical Aerial Images

Description:

... and Scene Understanding for Meter Resolution Image 29/03/07 - Oberpfaffenhofen ... Automatic Target Recognition in high resolution Optical Aerial Images ... – PowerPoint PPT presentation

Number of Views:68
Avg rating:3.0/5.0
Slides: 17
Provided by: xavierpe
Learn more at: http://www.dlr.de
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Automatic Target Recognition in high resolution Optical Aerial Images


1
Automatic Target Recognition in high resolution
Optical Aerial Images
  • 7th CNES/DLR Workshop on Information Extraction
    and Scene Understanding for Meter Resolution Image

Xavier PERROTTON Marc STURZEL Michel ROUX
Xavier.perrotton_at_eads.com marc.sturzel_at_eads.com
michel.roux_at_enst.fr Image Signal
Processing Laboratory Telecom Paris
2
Why ATR for EADS?
  • Objective
  • make a breakthrough on ATR in visible images
  • Context
  • Observation systems (satellites, UAVs, aircraft)
  • Huge volume of data sent back by current and
    future systems
  • Limited number of operators
  • Pressure to shorten the loops
  • Autonomous systems (missiles, UAVs)
  • More intelligence onboard
  • Strong need in the future for
  • Fully automatic processing
  • Autonomous systems
  • ATR still unsolved for operational use

3
The problem
  • Challenging problems
  • Lighting, occlusion and background
  • Difficult segmentation
  • Targets size
  • Local descriptors approach
  • Learning appearance characteristics
  • Focusing on discriminative parts of the target

4
Questions
  • Can we efficiently use local descriptors?
  • How to extend application domain by statistical
    learning?

5
Local descriptors method
6
Local descriptors method
  • Descriptors
  • GLOH (Gradient Location orientation Histogram)
    1
  • How to match Keypoints?
  • Looking for the best match on each pixel
  • Associating a limited number of matched points
    for each learned keypoint
  • How to define an hypothesis ?
  • Choosing three points among the best matches
  • Evaluating the affine transform
  • How to propagate an hypothesis?
  • Checking for agreement between each candidate
    point and the geometric model

1 K. Mikolajczykand C. Schmid. A performance
evaluation of local descriptors. In Proc. IEEE
CVPR, June 2003
7
Local descriptors tests on real images (1)
Learned target
Matched targets
8
Local descriptors tests on real images (2)
Learned target
Matched targets
9
Local descriptors tests on real images (3)
X
  • Aerial Images difficulties
  • Few points
  • Not robust to background
  • We must find a way to learn the variability of
    appearance characteristics

10
AdaBoost a powerful learning concept
  • Principle
  • Iterative learning algorithm introduced by Freund
    and Schapire 2
  • Constructing a strong classifier in combining
    weak classifiers
  • Selecting a weak classifier at each iteration
  • Used for face detection by Viola and Jones 3
  • Advantages
  • often outperforms most monolithic strong
    classifiers such as Neural Networks
  • Few parameters to tune

2 Y. Freund and R. E. Schapire. A
decision-theoretic generalization of on-line
learning and an application to boosting. 97 3
Paul Viola and Michael J. Jones. Rapid Object
Detection using a Boosted Cascade of Simple
Features.IEEE CVPR, 2001
11
AdaBoost algorithm
  • Adaboost starts with a uniform distribution of
    weights over training samples
  • We obtain a weak classifier from the weak
    learning algorithm, hj(x) at each round
  • We compute ?j that measures the confidence
    assigned to hj(x)
  • We increase the weights on the training samples
    that were misclassified
  • Repeat
  • At the end, make a weighted linear combination of
    the weak classifiers obtained at all iterations

12
Weak classifier
Database Positive, negative samples
Feature output database
X
X
X
X
X
X
Feature Threshold weak classifier
X
X
X
Feature
X
X
X
X
  • A weak classifier is only required to be better
    than chance
  • Very simple and computationally inexpensive

X
X
X
X
  • Haar like features
  • Gabor filters
  • Steerable filters
  • orientation estimation features

X
13
Database
  • Generation of a representative database with
    positive and negative samples
  • The classifier is learned on images of fixed size
  • Detection is done through a sliding search window
  • Angle variations -5 to 5

14
Tests on real images
Learned different appearance characteristics
successfully
15
Descriptors
  • Challenge
  • Finding descriptors less sensitive to background
    and target texture
  • Haar like features learn only difference of
    contrasts
  • Not enough to discriminate complex textures
  • But can be very efficient on shadow
  • Gabor filters, steerable filters, orientation
    estimation features
  • More robust to background and target texture

16
Conclusion
  • Local descriptors enable to define an efficient
    ATR algorithm
  • Targets can be modelled as a collection of
    regions
  • Geometric constraints are efficient to eliminate
    false alarms
  • Statistical learning enables to extend the
    application domain
  • Selecting the discriminating features
  • Learning the variability of appearance
    characteristics
  • Descriptors
  • To detect particular oriented edges
  • To detect different regions
About PowerShow.com