software-defined and fpga computational...
TRANSCRIPT
May 8, 2017 Sam Siewert, GPU Tech 2017 –
Drone Net Concept
“Drone Net”
Using Tegra for Multi-Spectral Detection and Tracking in
Shared Air Space
Significance
2
Motivation – Large Numbers of sUAS – Droneii, FAA, Sandia, ASSURE – Counter UAS Challenge – senseFly Catalog of Uses
Problem – Default solution – Part 107 for sUAS and beyond – ADS-B for sUAS insufficient, infeasible – RADAR/LIDAR feasibility
Drone Net hypothesis – Networked, multi-modal
(passive/active), information and sensor data fusion
– EO/IR + acoustic, spectral fusion, machine learning
– Compare to and validate with LIDAR/RADAR, ADS-B
The sUAS Market is Vigorous [Droneii]
Proposed and State of Art
4
Counter UAS [DARPA, ONR, SRC, Mitre] – RADAR/LIDAR, IFF – Neutralization, Jamming, Interception
Security [FAA, DHS, FBI, Sandia] – EO/IR, RADAR/LIDAR, Geo-fencing – Detection Field Testing [FBI at JFK]
Safety and Compliance [DOT FAA] – Part 107 - sUAS Registration, Pilot
Certification, Classification – ADS-B NAS, See/Sense-and-avoid – PrecisionHawk LATAS (Low-Altitude Traffic
and Airspace Safety) – NASA / Verizon UTM [UAS Traffic
Management]
Goals and Objectives
5
EO/IR sUAS Detection Feasibility – Baseline – UAS, GA, Wildlife, Insects, … – DNN, DBN, SVM Machine Vision and
Learning Classification, Identification
Passive, Passive + Active – Performance Evaluation – ROC, PR, F-measure, Confusion Matrices – Data, Image, Information Fusion – Acoustic Camera, LIDAR [Next Steps]
Complimentary Spatial, Temporal, Spectral Resolution – flightradar24.com – Enhanced aggregation – Network of Sensor Fusion Nodes – Long baseline [optical and acoustic
localization] – Campus Drone Net – Geo-Net [Florida, Alaska, Arizona,
Colorado]
Tegra K1, X1,X2
Method – Information Fusion
6
EO/IR Multi-spectral imaging – Visible, NIR to LWIR – Pixel Level and Feature Level Registration – Lower Cost than common bore-sight – Test sUAS with “Flash Pop” Stimulators
Link Drone Net Nodes [wireless, wired] ADS-B aggregation
– Drone Net receivers - Garmin, Appareo, etc. – Test UAS transceivers– e.g. uAvionix ping1090,
ping2020
Acoustic cameras and cues [improve classification] Active LIDAR [RADAR] GP-GPU Real-time Processing
Method – Machine Vision & Learning
7
Machine Vision using SoC Linux – Salient Object Detection
Shape, Behavior and Contrast/Color/Texture in Multiple Bands Performance [ROC, PR, F-measure, confusion matrices]
– Real-Time Detection, Segmentation, Tracking, Classification, Identification
Machine Learning (Traditional, ANN) – Expert systems – Bayesian inference, Dempster-Shafer
reasoning [DBN] – PCA [Principal Component Analysis] – SVM [Support Vector Machines] – Clustering [e.g. K-means] – GPU Accelerated DNN (cuDNN) – Supervised, Unsupervised learning
Leverage Open Source: ROS, OpenCV, PyBrain, PyML, MLpack, cuDNN, Caffe
Conceptual Configuration
Sam Siewert 8
Panchromatic, NIR, RGB
Jetson Tegra X1 With GP-GPU Co-Processing
Cloud Analytics and
Machine Learning
Flash SD Card (local
database) LWIR
Saliency & Behavioral Assessment Thermal Fusion Assessment
Many multi-spectral focal planes …
2D/3D Spatial Assessment
Experimental System Block Diagram 2 Watts at Idle, Plus 1.5 Watts per Camera = 6.5W E.g. Sobel, 30Hz, 20 Mega Pixels/Sec/Watt, 7.3W Peak – SPIE Sensor Tech + Apps
Sam Siewert 9
1) Sync’d Capture
2) Resolution Match
3) Image Registration
4) Detection 5) Classification 6) Identification
Needs Debugging – Literally! Many Insects Detected in Visible to LWIR Opportunity to work on Bird / Aviation Interaction Testing
Sam Siewert 10
2015/16 – ADAC & ERAU Sponsored UAA – ADAC, SmartCam ERAU (Undergraduate Research Team) – Dr. Sam Siewert, PI, Assistant Professor – Demi Matthew Vis – AE/SE Student – Ryan Claus – SE Student – Nicholas DiPinto – SE Student – Arctic Power Team – Power Team Poster
CU Boulder – Embedded Systems Engineering Graduate Program – Ram Krishnamurthy – MS EE – Surjith Singh – MS, ESE – Akshay Singh – ME, ESE – Shivasankar Gunasekaran – ME,ESE – Swaminath Badrinath – ME, ESE
Industry Advising/Collaboration Participants – Randall Myers, Mentor Graphics
Sam Siewert 11
This material is based upon work supported by the U.S. Department of Homeland Security under Grant Award Number, DHS-14-ST-061-COE-001A-02. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the U.S. Department of Homeland Security.
2016/17 Team – ERAU Sponsored ERAU – Drone Net – Dr. Sam Siewert, PI, Assistant Prof. – Dr. Iacopo Gentilini, Co-I – Dr. Stephen Bruder, Co-I – Dr. Mehran Andalibi, Co-I – Demi Matthew Vis – AE/SE Student – Ryan Claus – SE Student
CU Boulder – Embedded Systems Graduate – Ram Krishnamurthy – MS EE – Surjith Singh – MS, ESE – Akshay Singh – ME, ESE – Shivasankar Gunasekaran – ME,ESE – Omkar Seelam – ME, ESE
Industry Advising/Collaboration Participants – Randall Myers, Mentor Graphics (PCB, CAD, Systems) – Joe Butler, Intel Corporation (IoT)
Sam Siewert 12
Detection Experiments for Aircraft and UAS
Preliminary Roof-top Field Trials at ERAU Prescott
Sam Siewert
13
AIAA – Drone Net Feasibility Results
Sam Siewert
14
S. Siewert, M. Vis, R. Claus, R. Krishnamurthy, S. B. Singh, A. K. Singh, S. Gunasekaran, “Image and Information Fusion Experiments with a Software-Defined Multi-Spectral Imaging System for Aviation and Marine Sensor Networks”, AIAA SciTech 2017, Grapevine, Texas, January 2017.
Open Reference SDMSI Configuration 2 Basler Pulse Visible Cameras 1 FLIR Vue LWIR Camera with ZnSe Window Jetson TK1, Panda Wireless, USB3 Hub, Power, NEMA Enclosure
Sam Siewert 15
Smart Camera Deployment - Aerial UAV Systems - ERAU ICARUS Group Experimental Aviation and Small Aircraft - ERAU Kite Aerial Photography, Balloon Missions (ERAU, UAA, CU Boulder)
Sam Siewert 16
Sam Siewert – ERAU ICARUS Group
Actual - Roof Mount Experiment Starting point – evolve to aircraft, buoy and UAS later Embry Riddle flight line provides lots of light aircraft traffic Simple UAS testing in Campus (semi-Urban) environment Wildlife – insects, bats, birds, etc. Sam Siewert 17
Information Fusion Concepts Integration and System of Systems Between ADS-B and S-AIS for Vessel / Aircraft / UAS Awareness Smart Cameras Can Monitor and Plan Uplink Opportunity as Well as Wake up and Uplink
Sam Siewert 18
System Fusion For Uplink
Baseline Motion Trigger Detection
Difference Images over Time (adjustable) Threshold - Statistically Significant Pixel Change Filters (Atmospheric, Cloud, Constant Background Motion) – Dispersion of Changes Detection Performance – ROC, PR-Curve, F-measure [TP, FP, FN, TN analysis] Classification/Identification - Confusion Matrix
Sam Siewert 19
https://en.wikipedia.org/wiki/Precision_and_recall
PR best for Image Retrieval E.g. https://images.google.com/ ROC best for Target Detection
Frame by Frame Analysis TP – Determined by Human Review Frame by Frame Alternative is by Physical Experiment Design “Autoit” Program to Analyze Sam Siewert 20
Aircraft Detection Performance - Baseline Video Links – Aircraft, Bugs, FP, TP+FP, [TN], [Full]
Sam Siewert 21
UAS Detection Performance – Baseline Video Link – UAS+Aircraft, Bugs, FP, TP+FP, [TN], [Full]
Sam Siewert 22
Candidate SOD (BinWang14) - Aircraft Modified to Run BinWang14 SOD => MD Baseline Video Links – TP+FP, [TN], [Full]
Sam Siewert 23
Candidate SOD (BinWang14) - UAS Modified to Run BinWang14 SOD => MD Baseline Video Links – TP+FP, [TN], [Full]
Sam Siewert 24
AIAA – Drone Detection Conclusions
Drone Net Likely Requires Custom Detection – SOD Classification Based on Shape, Behavior and Contrast/Color/Texture in Multiple Bands (RGB, NIR, LWIR) Considering Acoustic Cue Fusion Cross Check with ADS-B, RADAR/LIDAR Data Produce Improved flightradar24.com Meta-data Find Ghost UAS and Aircraft [Non-compliant], Log Others
Sam Siewert 25
SPIE – Benchmark Results
Sam Siewert
26
S. Siewert, V. Angoth, R. Krishnamurthy, K. Mani, K. Mock, S. B. Singh, S. Srivistava, C. Wagner, R. Claus, M. Vis, “Software Defined Multi-Spectral Imaging for Arctic Sensor Networks”, SPIE Proceedings, Volume 9840, Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XXII, Baltimore, Maryland, April 2016.
FPGA Results - Sobel ALUTs: 10187 Registers: 13,561 Logic utilization: 7,427 / 32,070 ( 23 % )
Sam Siewert 27
Resolution Transform (Watts) (Pixel/sec) per Watt
Saturation FPS Bus transfer rate (MB/sec)
320x240 5.655 2,050,716 151 11.06
640x480 5.700 2,107,284 39.1 11.46
1280x960 5.704 2,143,506 9.95 11.66
2560x1920 5.696 2,157,303 2.50 11.72
Table 2. Sobel Continuous Transform Power Consumption by Cyclone V FPGA
FPGA Results – Pyramidal ALUTs: 24456 Registers: 34,062 Logic utilization: 17,721 / 32,070 ( 55 % ) ( 55 % )
Sam Siewert 28
Table 3. Pyramidal Laplacian Resolution Up-Conversion Continuous Transform Power
Resolution Transform (Watts) (Pixel/sec) per Watt
Saturation FPS Bus transfer rate (MB/sec)
320x240 6.009 889,546 69.6 5.10
640x480 6.013 904,281 17.7 5.19
1280x960 6.038 905,624 4.45 5.21
2560x1920 6.192 889,054 1.12 5.25
Table 4. Pyramidal Gaussian Resolution Down-Conversion Continuous Transform Power Resolution Continuous Transform
Power (Watts) (Pixel/sec) / Watt Saturation
FPS Bus transfer rate (MB/sec)
320x240 5.968 2,445,040 190 13.92 640x480 6.018 2,399,202 47.0 13.77 1280x960 6.023 2,427,813 11.9 13.95 2560x1920 6.109 2,309,154 2.87 13.45
GP-GPU Results - Sobel
Sam Siewert 29
Table 5. Sobel Continuous Transform Power Resolution Continuous
Power at 1Hz (Watts)
Continuous Power at 30Hz (Watts)
(pixels/sec) per Watt @ 1Hz
(pixels/sec) per Watt @ 30Hz
Saturation FPS
320x240 4.241 4.932 18,109 467,153 1624 640x480 4.256 4.984 72,180 1,849,117 840 1280x960 4.266 5.142 288,045 7,169,195 237 2560x1920 4.325 7.326 1,136,462 20,127,764 55
GP-GPU Results - Pyramidal
Sam Siewert 30
Table 6. Pyramidal Up and Down Conversion Continuous Transform Power Resolution Continuous
Power at 1Hz (Watts)
Continuous Power at 20Hz (Watts)
(pixels/sec) / Watt @ 1Hz
(pixels/sec) / Watt @ 20Hz
Saturation FPS
320x240 4.104 4.824 18,713 477,612 1120 640x480 4.116 5.460 74,636 1,687,912 325 1280x960 4.152 6.864 295,954 5,370,629 82 2560x1920 4.224 13.44 1,163,636 10,971,429 20
SPIE On-going Work We Have Completed Hough Lines Continuous Transform Test, Available on GitHub Hough Power Curves Not Yet Produced – In Progress Goal to Identify all Continuous Transform Primitives Used in Infrared + Visible Fusion and 3D Mapping Pixel Level Emphasis, But Also Plan to Review Feature Level – Camera Extrinsic and Intrinsic Transformations – Registration – Resolution and AR Matching – Methods of Pixel Level Fusion in Review [10] [11], [12], [14]
Sam Siewert 31
SPIE Benchmark Conclusions Please Download our Benchmarks – https://github.com/siewertserau/fusion_coproc_benchmarks – MIT License
Test on NVIDIA GP-GPU or FPGA SoCs (Altera, Xilinx) Share Results Back Please Help Us Add Benchmarks Critical to Continuous 3D Mapping and Infrared + Visible Fusion (Suite of Primitives) Open Source Hardware, Firmware, Software for Multispectral Smart Camera Applications
Sam Siewert 32
Research References
Sam Siewert
33
References
Sam Siewert 34
1S. Siewert, V. Angoth, R. Krishnamurthy, K. Mani, K. Mock, S. B. Singh, S. Srivistava, C. Wagner, R. Claus, M. Demi Vis, “Software Defined Multi-Spectral Imaging for Arctic Sensor Networks”, SPIE Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XXII, Baltimore, Maryland, April 2016. 2 S. Siewert, J. Shihadeh, Randall Myers, Jay Khandhar, Vitaly Ivanov, “Low Cost, High Performance and Efficiency Computational Photometer Design”, SPIE Sensing Technology and Applications, SPIE Proceedings, Volume 9121, Baltimore, Maryland, May 2014. 3Piella, G. (2003). A general framework for multiresolution image fusion: from pixels to regions. Information fusion, 4(4), 259-280. 4Blum, R. S., & Liu, Z. (Eds.). (2005). Multi-sensor image fusion and its applications. CRC press. 5Liu, Z., Blasch, E., Xue, Z., Zhao, J., Laganiere, R., & Wu, W. (2012). Objective assessment of multiresolution image fusion algorithms for context enhancement in night vision: a comparative study. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 34(1), 94-109. 6Simone, G., Farina, A., Morabito, F. C., Serpico, S. B., & Bruzzone, L. (2002). Image fusion techniques for remote sensing applications. Information fusion, 3(1), 3-15. 7Mitchell, H. B. (2010). Image fusion: theories, techniques and applications. Springer Science & Business Media. 8Szeliski, R. (2010). Computer vision: algorithms and applications. Springer Science & Business Media. 9Sharma, G., Jurie, F., & Schmid, C. (2012, June). Discriminative spatial saliency for image classification. In Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on (pp. 3506-3513). IEEE. 10Toet, A. (2011). Computational versus psychophysical bottom-up image saliency: A comparative evaluation study. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 33(11), 2131-2146.
References
Sam Siewert 35
11Valenti, R., Sebe, N., & Gevers, T. (2009, September). Image saliency by isocentric curvedness and color. In Computer Vision, 2009 IEEE 12th International Conference on (pp. 2185-2192). IEEE. 12Wang, M., Konrad, J., Ishwar, P., Jing, K., & Rowley, H. (2011, June). Image saliency: From intrinsic to extrinsic context. In Computer Vision and Pattern Recognition (CVPR), 2011 IEEE Conference on (pp. 417-424). IEEE. 13Liu, F., & Gleicher, M. (2006, July). Region enhanced scale-invariant saliency detection. In Multimedia and Expo, 2006 IEEE International Conference on (pp. 1477-1480). IEEE. 14Cheng, M. M., Mitra, N. J., Huang, X., & Hu, S. M. (2014). Salientshape: Group saliency in image collections. The Visual Computer, 30(4), 443-453. 15http://global.digitalglobe.com/sites/default/files/DG_WorldView2_DS_PROD.pdf 16http://www.spaceimagingme.com/downloads/sensors/datasheets/DG_WorldView3_DS_2014.pdf 17Richards, Mark A., James A. Scheer, and William A. Holm. Principles of modern radar. SciTech Pub., 2010. 18Brown, Christopher D., and Herbert T. Davis. "Receiver operating characteristics curves and related decision measures: A tutorial." Chemometrics and Intelligent Laboratory Systems 80.1 (2006): 24-38. 19Wang, Bin, and Piotr Dudek. "A fast self-tuning background subtraction algorithm." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2014. 20Panagiotakis, Costas, et al. "Segmentation and sampling of moving object trajectories based on representativeness." IEEE Transactions on Knowledge and Data Engineering 24.7 (2012): 1328-1343.
References
Sam Siewert 36
21Public SDMSI shared data web site for video sequences captured and used in two experiments presented in this paper - http://mercury.pr.erau.edu/~siewerts/extra/papers/AIAA-SDMSI-data-2017/ 22Perazzi, Federico, et al. "Saliency filters: Contrast based filtering for salient region detection." Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on. IEEE, 2012. 23Achanta, Radhakrishna, et al. "SLIC superpixels compared to state-of-the-art superpixel methods." IEEE transactions on pattern analysis and machine intelligence 34.11 (2012): 2274-2282. 24Hou, Xiaodi, and Liqing Zhang. "Saliency detection: A spectral residual approach." 2007 IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 2007. 25Global Contrast based Salient Region Detection. Ming-Ming Cheng, Niloy J. Mitra, Xiaolei Huang, Philip H. S. Torr, Shi-Min Hu. IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI), 37(3), 569-582, 2015. 26flightradar24.com, ADS-B, primary/secondary RADAR flight localization and aggregation services. 27Birch, Gabriel Carisle, John Clark Griffin, and Matthew Kelly Erdman. UAS Detection Classification and Neutralization: Market Survey 2015. No. SAND2015-6365. Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States), 2015.
Drone Detection and Neutralization Companies
37
Leading Drone Detection Companies
• https://www.blacksagetech.com/ • https://www.droneshield.com/ • http://www.dedrone.com/en/
List of Drone Detection Companies and Experiments
• Test at JFK by FBI - https://www.faa.gov/news/updates/?newsId=85546
• https://www.x20.org/uav-drone-detection-anti-drone-defense-systems/
• http://industrialcamera.net/security-surveillance-system-products-services/uav-detection/
• https://www.hgh-infrared.com/es/Aplicaciones/Seguridad/UAV-Detection
• http://www.cerbair.com/en/ • http://www.israeldefense.co.il/en/content/rafael-unveils-drone-dome-
drone-detection-and-neutralization-system