Deep Learning Method on Target Echo Signal Recognition for Obscurant Penetrating Lidar Detection in Degraded Visual Environments
Abstract
:1. Introduction
- (1)
- We propose a novel multi-distance measurement method that acquires multiple echo signals in the process of approaching the target. Unlike the fixed-distance measurement, this method acquires multiple echo signals during the movement of the Lidar sensor.
- (2)
- We propose a novel target echo signal recognition method based on 2-D spectrogram images. The 2-D spectrogram images are constructed by using the frequency-distance relation derived from the 1-D echo signals of the Lidar sensor individual cell in the course of approaching target.
- (3)
- We propose a customized deep learning algorithm based on Faster Region Convolutional Neural Network (R-CNN) to recognize the target echo signal and to predict the target distance.
2. The Obscurant Penetrating Lidar Detection Method
2.1. Method Overview
2.2. The Multi-Distance Measurement
2.3. The 2-D Spectrogram Images
2.4. The Target Echo Signal Recognition Algorithm
3. Simulation and Experimental Results
3.1. The Simulation Results
3.1.1. The Simulation and Training Settings
3.1.2. Detection Results of the Simulated Signals
3.2. Experimental Results
3.2.1. Experimental and Training Settings
3.2.2. Experimental Results
4. Conclusions
Author Contributions
Funding
Conflicts of Interest
References
- Zhang, Y.; Wang, J.; Wang, X.; Dolan, J.M. Road-segmentation-based curb detection method for self-driving via a 3D-LiDAR sensor. IEEE Trans. Intell. Transp. Syst. 2018, 19, 3981–3991. [Google Scholar] [CrossRef]
- Heinzler, R.; Piewak, F.; Schindler, P.; Stork, W. CNN-Based Lidar Point Cloud De-Noising in Adverse Weather. IEEE Robot. Autom. Lett. 2020, 5, 2514–2521. [Google Scholar] [CrossRef] [Green Version]
- Hecht, J. Lidar for self-driving cars. Opt. Photo. News 2018, 29, 26–33. [Google Scholar] [CrossRef]
- Wallace, L.; Lucieer, A.; Watson, C.S. Evaluating tree detection and segmentation routines on very high resolution UAV LiDAR data. IEEE Trans. Geosci. Remote Sens. 2014, 52, 7619–7628. [Google Scholar] [CrossRef]
- Starr, J.W.; Lattimer, B.Y. A comparison of IR stereo vision and LIDAR for use in fire environments. Sensors 2012, 8, 1–4. [Google Scholar]
- Ma, R. DEM generation and building detection from lidar data. Photogramm. Eng. Remote Sens. 2005, 71, 847–854. [Google Scholar] [CrossRef]
- Jiang, Z.; Song, C. The Simulation of FMCW Laser Transmission in Smoke Environment. In Proceedings of the 2019 4th International Conference on Control and Robotics Engineering (ICCRE), Nanjing, China, 20–23 April 2019; pp. 175–178. [Google Scholar]
- Fritsche, P.; Zeise, B.; Hemme, P.; Wagner, B. Fusion of radar, LiDAR and thermal information for hazard detection in low visibility environments. In Proceedings of the 2017 IEEE International Symposium on Safety, Security and Rescue Robotics (SSRR), Shanghai, China, 11–13 October 2017; pp. 96–101. [Google Scholar]
- Donoho, D.L. De-noising by soft-thresholding. IEEE Trans. Inf. Theory 1995, 41, 613–627. [Google Scholar] [CrossRef] [Green Version]
- Huang, N.E.; Long, S.R.; Zheng, S. The Mechanism for Frequency Downshift in Nonlinear Wave Evolution. Adv. Appl. Mech. 1996, 32, 59–117. [Google Scholar]
- Kopsinis, Y.; McLaughlin, S. Improved EMD using doubly-iterative sifting and high order spline interpolation. EURASIP J. Adv. Signal Process. 2008, 1, 128293. [Google Scholar] [CrossRef]
- Kopsinis, Y.; McLaughlin, S. Empirical mode decomposition based soft-thresholding. In Proceedings of the 2008 16th European Signal Processing Conference, Lausanne, Switzerland, 6 April 2008; pp. 1–5. [Google Scholar]
- Kopsinis, Y.; McLaughlin, S. Development of EMD-Based Denoising Methods Inspired by Wavelet Thresholding. IEEE T. Signal Proces. 2009, 57, 1351–1362. [Google Scholar] [CrossRef]
- Shamsudin, A.U.; Ohno, K.; Westfechtel, T.; Takahiro, S.; Okada, Y.; Tadokoro, S. Fog removal using laser beam penetration, laser intensity, and geometrical features for 3D measurements in fog-filled room. Adv. Robot. 2016, 30, 729–743. [Google Scholar] [CrossRef]
- Charron, N.; Phillips, S.; Waslander, S.L. De-noising of Lidar point clouds corrupted by snowfall. In Proceedings of the 2018 15th Conference on Computer and Robot Vision (CRV), Toronto, ON, Canada, 9–11 May 2018; pp. 254–261. [Google Scholar]
- Illig, D.W.; Rumbaugh, L.K.; Banavar, M.K.; Bollt, E.M.; Jemison, W.D. Backscatter suppression via blind signal separation for a 532 nm underwater chaotic lidar rangefinder. In Proceedings of the 2015 SPIE, Washington, DC, USA, 19–22 October 2015; pp. 1–6. [Google Scholar]
- O’connor, S.; Lee, R.; Mullen, L.; Cochenour, B. In Waveform design considerations for modulated pulse lidar. In Proceedings of the Ocean Sensing and Monitoring VI, Maryland, MD, USA, 5 May 2014; p. 91110P. [Google Scholar]
- Djuricic, A.; Jutzi, B. Supporting Uavs in Low Visibility Conditions by Multiple-Pulse Laser Scanning Devices. Isprs Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 2013, XL-1/W1, 93–98. [Google Scholar] [CrossRef] [Green Version]
- Trickey, E.; Church, P.; Cao, X. Characterization of the OPAL obscurant penetrating LiDAR in various degraded visual environments. In Proceedings of the Degraded Visual Environments: Enhanced, Synthetic, and External Vision Solutions, International Society for Optics and Photonics, Maryland, MD, USA, 29 April 2013; p. 87370E. [Google Scholar]
- Starr, J.W.; Lattimer, B.Y. Evidential Sensor Fusion of Long-Wavelength Infrared Stereo Vision and 3D-LIDAR for Rangefinding in Fire Environments. Fire Technol. 2017, 1961–1983. [Google Scholar] [CrossRef]
- Bosch, T.; Lescure, M.; Myllyla, R.; Rioux, M.; Amann, M.C. Laser ranging: A critical review of usual techniques for distance measurement. Opt. Eng. 2001, 40, 10–12. [Google Scholar] [CrossRef]
- Journet, B.; Bazin, G. A low-cost laser range finder based on an FMCW-like method. IEEE Trans. Instrum. Meas. 2000, 49, 840–843. [Google Scholar] [CrossRef]
- Liu, B.; Song, C.; Li, Y.; Duan, Y. A frequency-modulated-continuous-wave laser detection system based on the four-quadrant photodetector. Appl. Phys. B 2018, 124, 186. [Google Scholar] [CrossRef]
- Batet, O.; Dios, F.; Comeron, A.; Agishev, R. Intensity-modulated linear-frequency-modulated continuous-wave lidar for distributed media: Fundamentals of technique. Appl. Opt. 2010, 49, 3369–3379. [Google Scholar] [CrossRef]
- Stann, B.L.; Ruff, W.C.; Sztankay, Z.G. Intensity-modulated diode laser radar using frequency-modulation/continuous-wave ranging techniques. Opt. Eng. 1996, 35, 3270–3278. [Google Scholar] [CrossRef]
- Ren, S.; He, K.; Girshick, R.; Sun, J. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Trans. Pattern Anal. Mach. Intell. 2017, 39, 1137–1149. [Google Scholar] [CrossRef] [Green Version]
- Bruscaglioni, P.; Ismaelli, A.; Zaccanti, G. Monte-Carlo calculations of LIDAR returns: Procedure and results. Appl. Phys. B 1995, 60, 325–329. [Google Scholar] [CrossRef]
- Liu, B.; Song, C.; Duan, Y. The characteristics simulation of FMCW laser backscattering signals. Opt. Rev. 2018, 25, 197–204. [Google Scholar] [CrossRef]
- Lee, H.; Wang, Q.; Shan, B.; Li, X.; Gong, Y.; Zhao, J.; Peng, Z. A New Monte Carlo Simulation Model for Laser Transmission in Smokescreen. In MATLAB; SPIE: Washington, DC, USA, 2016. [Google Scholar]
- Ryde, J.; Hillier, N. Performance of laser and radar ranging devices in adverse environmental conditions. J. Field Robot. 2009, 8, 712–727. [Google Scholar] [CrossRef]
- Cao, X.; Peng, W.; Cai, M.; Bai, X.; Gong, G.; Liu, M.; Qi, J. Region Based CNN for Foreign Object Debris Detection on Airfield Pavement. Sensors 2018, 18, 737. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Uijlings, J.R.R.; Sande, K.E.A.V.D. Selective Search for Object Recognition. Int. J. Comput. Vis. 2013, 104, 154–171. [Google Scholar] [CrossRef] [Green Version]
Visibility/m | Measurement Distance | Measurement Times | Annotation Situation |
---|---|---|---|
1−3 | 13 m to 3 m | 11 | No |
4−6 | Yes | ||
7−15 | 16 m to 6 m |
Methods | Resolution | AP (%) | ARE (m) | Time(s) |
---|---|---|---|---|
YOLO | 227 × 227 | 0 | 0.02 | |
500 × 500 | 0 | 0.08 | ||
R-CNN | 227 × 227 | 57.0 | 2.75 | 5.63 |
500 × 500 | 75.1 | 1.32 | 53.34 | |
Fast R-CNN | 227 × 227 | 75.9 | 2.53 | 1.24 |
500 × 500 | 86.3 | 1.34 | 4.42 | |
Faster R-CNN | 227 × 227 | 85.0 | 1.91 | 0.49 |
500 × 500 | 91.4 | 0.24 | 0.86 |
Visibility/m | Measurement Distance | Measurement Times | Annotation Situation |
---|---|---|---|
1−2 | 6 m to 2 m | 5 | No |
3−15 | Yes |
Methods | Wavelet | EMD-DT | EMD-IT | EMD-Correlation | Our Method |
---|---|---|---|---|---|
ARE(m) | 0.32 | 0.48 | 0.42 | 0.45 | 0.14 |
Time(s) | 0.82 | 5.27 | 4.86 | 4.42 | 0.75 |
Algorithms | Resolution | AP (%) | ARE (m) | Time(s) |
---|---|---|---|---|
YOLO | 227 × 227 | 0 | 0.04 | |
500 × 500 | 0 | 0.11 | ||
R-CNN | 227 × 227 | 62.9 | 1.46 | 6.72 |
500 × 500 | 80.6 | 0.81 | 46.65 | |
Fast R-CNN | 227 × 227 | 78.3 | 1.08 | 0.95 |
500 × 500 | 89.1 | 0.34 | 4.48 | |
Faster R-CNN | 227 × 227 | 82.5 | 0.84 | 0.34 |
500 × 500 | 92.3 | 0.14 | 0.75 |
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Liang, X.; Huang, Z.; Lu, L.; Tao, Z.; Yang, B.; Li, Y. Deep Learning Method on Target Echo Signal Recognition for Obscurant Penetrating Lidar Detection in Degraded Visual Environments. Sensors 2020, 20, 3424. https://doi.org/10.3390/s20123424
Liang X, Huang Z, Lu L, Tao Z, Yang B, Li Y. Deep Learning Method on Target Echo Signal Recognition for Obscurant Penetrating Lidar Detection in Degraded Visual Environments. Sensors. 2020; 20(12):3424. https://doi.org/10.3390/s20123424
Chicago/Turabian StyleLiang, Xujia, Zhonghua Huang, Liping Lu, Zhigang Tao, Bing Yang, and Yinlin Li. 2020. "Deep Learning Method on Target Echo Signal Recognition for Obscurant Penetrating Lidar Detection in Degraded Visual Environments" Sensors 20, no. 12: 3424. https://doi.org/10.3390/s20123424
APA StyleLiang, X., Huang, Z., Lu, L., Tao, Z., Yang, B., & Li, Y. (2020). Deep Learning Method on Target Echo Signal Recognition for Obscurant Penetrating Lidar Detection in Degraded Visual Environments. Sensors, 20(12), 3424. https://doi.org/10.3390/s20123424