{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

3630-08-lec14-features

3630-08-lec14-features - Introduction Detection/Estimation...

Info iconThis preview shows pages 1–12. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Introduction Detection/Estimation Sonar Features Line Features Summary References Feature Extraction Henrik I Christensen Robotics & Intelligent Machines @ GT Georgia Institute of Technology, Atlanta, GA 30332-0760 [email protected] Henrik I Christensen ([email protected]) Features 1 / 38 Introduction Detection/Estimation Sonar Features Line Features Summary References Outline 1 Introduction 2 Detection/Estimation 3 Sonar Features 4 Line Features 5 Summary Henrik I Christensen ([email protected]) Features 2 / 38 Introduction Detection/Estimation Sonar Features Line Features Summary References System Context Henrik I Christensen ([email protected]) Features 3 / 38 Introduction Detection/Estimation Sonar Features Line Features Summary References Introduction Compression of sensor data to key “features” How do we detect these “features”? How do we estimate the parameters for the “features”? A few examples Henrik I Christensen ([email protected]) Features 4 / 38 Introduction Detection/Estimation Sonar Features Line Features Summary References Outline 1 Introduction 2 Detection/Estimation 3 Sonar Features 4 Line Features 5 Summary Henrik I Christensen ([email protected]) Features 5 / 38 Introduction Detection/Estimation Sonar Features Line Features Summary References Point Estimation Many sensor generate range estimates (Sonar, GPS, Laser) Triangulation is a well known technique for estimation of points Fusion of multiple range readings into an estimate Theory is well known from phased array radar Henrik I Christensen ([email protected]) Features 6 / 38 Introduction Detection/Estimation Sonar Features Line Features Summary References Point Estimation r1 r1 r2 r1 r2 Henrik I Christensen ([email protected]) Features 7 / 38 Introduction Detection/Estimation Sonar Features Line Features Summary References Point Estimation Generalizes to many sensors hits x T = x s 1 + 1 d 2 s ( dxs d 2 r ± | dys | q r 2 2 d 2 s- d 4 r ) y T = y s 1 + 1 d 2 s ( dys d 2 r ± | dxs | q r 2 2 d 2 s- d 4 r ) The intersection “point” has no general solution so it becomes a minimization problem. Henrik I Christensen ([email protected]) Features 8 / 38 Introduction Detection/Estimation Sonar Features Line Features Summary References Line Estimation Lines are a predominant feature in engineered environments There is an abundance of methods for line estimation LSQ, Split-Merge, Hough, EM-estimation, RANSAC is frequently used (Fischler & Bolles, 1981) Henrik I Christensen ([email protected]) Features 9 / 38 Introduction Detection/Estimation Sonar Features Line Features Summary References Voting based estimators Voting provides a simple estimator for detection Voting requires: 1 A Voting Space 2 A voting function (structure function) 3 A decision function (often local extrema) Hough (1962) is one of the most widely used. Can also be used for lines and other shapes (Ballard, 1981) Henrik I Christensen ([email protected]) Features 10 / 38 Introduction Detection/Estimation Sonar Features Line Features Summary References The Hough Transform ρ θ x y Line model: ρ = x * cos( θ...
View Full Document

{[ snackBarMessage ]}

Page1 / 41

3630-08-lec14-features - Introduction Detection/Estimation...

This preview shows document pages 1 - 12. Sign up to view the full document.

View Full Document Right Arrow Icon bookmark
Ask a homework question - tutors are online