Download presentation
Presentation is loading. Please wait.
Published byHorace Hart Modified over 9 years ago
1
Sukhum Sattaratnamai Advisor: Dr.Nattee Niparnan 1
2
Outline Introduction Objective Calibration Process Our Work Improving Laser Data Automate Data Collection Conclusion 2
3
LRF-Camera System 3 α
4
4 α
5
LRF-Camera Calibration Problem Definition Find the transformation [R |t ] of the camera w.r.t. LRF 5
6
Objective Related Work LRF-Camera Calibration Calibration of a multi-sensor system laser rangefinder/camera, 1995 More Accurate Result Extrinsic calibration of a camera and laser range finder (improves camera calibration), 2004 Easier Process An algorithm for extrinsic parameters calibration of a camera and a laser range finder using line features, 2007 6 Proposal Improving Laser Data Filtering Laser Data
7
Objective 7 Proposal Improving Laser Data Filtering Laser Data Thesis Improving Laser Data On Improving Laser Data for Extrinsic LRF/Camera Calibration, 2011 Automated Process Automated Calibration Data Collection in LRF/Camera Calibration with Online Feedback, 2012 Related Work LRF-Camera Calibration Calibration of a multi-sensor system laser rangefinder/camera, 1995 More Accurate Result Extrinsic calibration of a camera and laser range finder (improves camera calibration), 2004 Easier Process An algorithm for extrinsic parameters calibration of a camera and a laser range finder using line features, 2007
8
Objective Related Work LRF-Camera Calibration Calibration of a multi-sensor system laser rangefinder/camera, 1995 More Accurate Result Extrinsic calibration of a camera and laser range finder (improves camera calibration), 2004 Easier Process An algorithm for extrinsic parameters calibration of a camera and a laser range finder using line features, 2007 8 Proposal Improving Laser Data Filtering Laser Data Thesis Improving Laser Data On Improving Laser Data for Extrinsic LRF/Camera Calibration, 2011 Automated Process Automated Calibration Data Collection in LRF/Camera Calibration with Online Feedback, 2012
9
Calibration Process 9 Data Collection Optimization Check Result End Start Feature Detection
10
Calibration Process Data Collection 10
11
Calibration Process Feature Detection 11
12
Calibration Process Projection Error 12
13
Calibration Process Optimization Simulated Annealing : Find global minimum Levenberg-Marquardt : Find local minimum 13
14
Calibration Process Result Project laser data onto an image 14
15
Our Work Improving Laser Data Automatic Data Collection 15
16
Improving Laser Data Angular Error => 16
17
Simulation Angular Error => 17
18
Simulation Laser Data Improvement 18 Target Methodimp plai n ratioimp plai n ratio average RMS 0.3 2 0.5 4 54.8 % 0.9 2 2.0 4 48.0 % S.D. of RMS 0.0 07 0.0 32 22.1 % 0.0 03 0.0 19 16.5 %
19
Experiment Laser Range Finder Camera 19
20
Experiment Laser Data Improvement 20 TargetStingrayLegria Methodimp plai n ratioimp plai n ratio average RMS 1.6 6 2.9 0 57.3 % 6.1 8 11.3 8 54.3 % S.D. of RMS 0.0 2 0.0 4 40.7 % 0.1 9 0.4 8 38.8 %
21
Experiment Number of Data 21
22
Improving Laser Data Lower bound 22
23
Simulation Lower Bound 23 การ ทดลอง RMS Rati o 1 27 0 0.5 0.5 9 0.5592.9 2 54 0 0.5 1.1 8 1.0992.9 3 27 0 1.0 1.1 8 1.0992.3 454 0 1.02.3 6 2.1892.3
24
Automate Data Collection 24 Data Collection Optimization Check Result End Start Feature Detection 5 นาที 30 นาที 1 วินาที 5 นาที 30 นาที 2 นาที
25
Automate Data Collection Feature Detection 25
26
Automate Data Collection False Detection => Tracking 26
27
Experiment Data Distribution 27 บริเว ณ ปรับแก้ทดสอบ ค่าเฉลี่ยค่า เบี่ยงเบ น ค่าเฉลี่ยค่า เบี่ยงเบ น ซ้าย 0.440.0021.030.063 กลาง 0.390.0030.780.037 ขวา 0.500.0031.160.078 ทั้งหม ด 0.440.0050.690.011
28
Automate Data Collection Working Space Covering Data Bin (x, y, angle) 28
29
Automate Data Collection Moving Calibration Object => Velocity Metric 29
30
Experiment Velocity & Accuracy 30
31
Experiment Accuracy & Time 31
32
Experiment 32
33
Automate Data Collection User Interface Data Quality Metric Tracking, Velocity Data Distribution Data Bins, Current Bin, Target Bin 33
34
Automate Data Collection User Interface Result Laser Data Projection Acknowledge & Warning Sound Data Acquire, Tracking Lost 34
35
Conclusion Improved calibration method Reduce projection error to 50 percent Automatic data collection process Faster and easier for all user 35
36
36
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.