Improved Human Detection with a Fusion of Laser Scanner and Vision/Infrared Information for Mobile Applications
Source abstract
Abstract
This paper presents a method for human detection using a laser scanner with vision or infrared images. Mobile applications require reliable and efficient methods for human detection, especially as a part of driver assistance systems, including pedestrian collision systems. The authors propose an efficient method for multimodal human detection based on a combination of the features and context information. Strictly, the human is detected in the vision/infrared images using a combination of local binary patterns and histogram of oriented gradients features with a neural network in a cascade manner. Next, using coordinates of detected humans from the vision system, the moving trajectory is predicted until the scanner working distance is reached by the individual human. Then the segmentation of data from the laser scanner is further carried out with respect to the predicted trajectory. Finally, human detection in the laser scanner working distance is performed based on modelling of the human legs. The modelling is based on the adaptive breakpoint detection algorithm and proposed improved polylines definition and fitting algorithm. The authors conducted a set of experiments in predefined scenarios, discussed the identified weakness and advantages of the proposed method, and outlined detailed future work, especially for night-time and low-light conditions.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Video Surveillance and Tracking Methods · Hand Gesture Recognition Systems · Advanced Neural Network Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Institute of Automatic Control, Silesian University of Technology, Akademicka 16, 44-100 Gliwice, Poland
References (61)
Show source-provided references
1. Multi-spectral pedestrian detection10.1016/j.sigpro.2014.08.003 ↗
2. Context-aware obstacle detection for navigation by visually impaired10.1016/j.imavis.2017.06.002 ↗
3. Fusion of 3-D-LIDAR and camera data for scene parsing10.1016/j.jvcir.2013.06.008 ↗
4. Image-based on-road vehicle detection using cost-effective Histograms of Oriented Gradients10.1016/j.jvcir.2013.08.001 ↗
5. Moving object detection and segmentation in urban environments from a moving platform10.1016/j.imavis.2017.07.006 ↗
6. Based on cluster tree human action recognition algorithm for monocular video
7. A double-region learning algorithm for counting the number of pedestrians in subway surveillance videos10.1016/j.engappai.2017.06.017 ↗
8. Monocular pedestrian detection: Survey and experiments10.1109/TPAMI.2008.260 ↗
9. Scene-specific classifier for effective and efficient team sport players detection from a single calibrated camera10.1016/j.cviu.2017.01.001 ↗
10. Fast and reliable detection of hockey players10.1016/j.procs.2013.09.249 ↗
11. Using automated walking gait analysis for the identification of pedestrian attributes10.1016/j.trc.2014.08.004 ↗
12. Multiple human tracking in high-density crowds10.1016/j.imavis.2012.08.013 ↗
13. Efficient tracking of human poses using a manifold hierarchy10.1016/j.cviu.2014.10.005 ↗
14. Easy gesture recognition for Kinect10.1016/j.advengsoft.2014.07.005 ↗
15. Cosma, C., Brehar, R., and Nedevschi, S. (2013, January 5–7). Pedestrians detection using a cascade of LBP and HOG classifiers. Proceedings of the IEEE International Conference on ICCP, Cluj-Napoca, Romania.10.1109/ICCP.2013.6646084 ↗
16. Review of advanced driver assistance systems (ADAS)10.1063/1.5012394 ↗
17. Pedestrian protection systems: Issues, survey, and challenges10.1109/TITS.2007.903444 ↗
18. Survey on pedestrian detection for advanced driver assistance systems10.1109/TPAMI.2009.122 ↗
19. National Highway Traffic Safety Administration (2017, December 20). Traffic Safety Facts 2015 Data—Pedestrians, Available online: https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812375.
20. Centers for Disease Control and Prevention (2017, December 20). WISQARS (Web-Based Injury Statistics Query and Reporting System), Available online: http://www.cdc.gov/injury/wisqars.
21. Benenson, R., Matthias, M., Tuytrlaars, T., and Van Gool, L. (2013, January 25–27). Seeking the strongest rigid detector. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Portland, Oregon.10.1109/CVPR.2013.470 ↗
22. Luo, P., Tian, Y., Wang, X., and Tang, X. (2014, January 23–28). Switchable deep network for pedestrian detection. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.10.1109/CVPR.2014.120 ↗
23. Zhang, S., Bauckhage, C., and Cremers, A. (2014, January 23–28). Informed haar-like features improve pedestrian detection. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.10.1109/CVPR.2014.126 ↗
24. Viola, P., and Jones, M. (2001, January 8–14). Rapid object detection using a boosted cascade of simple features. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Kauai, HI, USA.
25. Vision-based pedestrian detection using Haar-like features
26. Dalal, N., and Triggs, B. (2005, January 20–25). Histograms of oriented gradients for human detection. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Diego, CA, USA.
27. Co–occurrence histograms of oriented gradients for pedestrian detection10.2197/ipsjtcva.2.39 ↗
28. Zhu, Q., Avidan, S., Yeh, M., and Cheng, K. (2006, January 17–22). Fast Human Detection Using a Cascade of Histograms of Oriented Gradients. Proceedings of the IEEE Conference Computer Vision and Pattern Recognition (CVPR), New York, NY, USA.
29. Park, W.-J., Kim, D.-H., Lyuh, C.-G., Roh, T.M., and Ko, S.-J. (October, January 30). Fast human detection using selective block-based HOG-LBP. Proceedings of the IEEE International Conference on Image Processing, Orlando, FL, USA.
30. Hussain, S., and Triggs, B. (September, January 29). Feature sets and dimensionality reduction for visual object detection. Proceedings of the British Machine Vision Conference, Dundee, UK.
31. Zheng, Y., Shen, C., and Huang, X. (2010, January 26–29). Pedestrian detection using center-symmetric local binary patterns. Proceedings of the IEEE International Conference on Image Processing, Hong Kong, China.
32. Nguyen, D.T., Zong, Z., Li, W., and Ogunbona, P. (2010, January 26–29). Object detection using non-redundant local binary patterns. Proceedings of the IEEE International Conference on Image Processing, Hong Kong, China.10.1109/ICIP.2010.5651633 ↗
33. Local Binary Patterns and Its Application to Facial Image Analysis: A Survey10.1109/TSMCC.2011.2118750 ↗
34. Ma, L., Yang, X., Xu, Y., and Zhu, J. (2011, January 11–14). Human Identification Using Body Prior and Generalized EMD. Proceedings of the 18th IEEE International Conference Image, Brussels, Belgium.10.1109/ICIP.2011.6115774 ↗
35. An efficient gait recognition system for human identification using modified ICA
36. Context aided pedestrian detection for danger estimation based on laser scanner and computer vision10.1016/j.eswa.2014.04.034 ↗
37. A real time human detection system based on far infrared vision10.1007/978-3-540-69905-7_9 ↗
38. Robust and fast pedestrian detection method for far–infrared automotive driving assistance systems10.1016/j.infrared.2013.06.003 ↗
39. Robust pedestrian detection in thermal infrared imagery using a shape distribution histogram feature and modified sparse representation classification10.1016/j.patcog.2014.12.013 ↗
40. Szarvas, M., Yoshizawa, A., Yamamoto, M., and Ogata, J. (2005, January 6–8). Pedestrian detection with convolutional neural networks. Proceedings of the IEEE Intelligent Vehicles Symposium, Las Vegas, NV, USA.10.1109/IVS.2005.1505106 ↗
41. Robust multiple cameras pedestrian detection with multi-view Bayesian network10.1016/j.patcog.2014.12.004 ↗
42. Laser-based detection and tracking of multiple people in crowds10.1016/j.cviu.2006.07.015 ↗
43. Combined lane and road attributes extraction by fusing data from digital map, laser scanner and camera10.1016/j.inffus.2010.01.005 ↗
44. Drage, T.H., Kalinowski, J., and Braunl, T. (2014, January 24–29). Development of an Autonomous Formula SAE Car with Laser Scanner and GPS. Proceedings of the IFAC, Cape Town, South Africa.10.3182/20140824-6-ZA-1003.01386 ↗
45. Laser-Based Pedestrian Tracking in Outdoor Environments by Multiple Mobile Robots10.3390/s121114489 ↗
46. Development of Human Following Mobile Robot System Using Laser Range Scanner10.1016/j.procs.2015.12.310 ↗
47. Premebida, C., Ludwig, O., Silva, M., and Nunes, U. (2010, January 19–22). A cascade classifier applied in pedestrian detection using laser and image-based features. Proceedings of the IEEE Intelligent Transportation Systems Conference on ITSC, Madeira Island, Portugal.10.1109/ITSC.2010.5625244 ↗
48. Spinello, L., and Siegwart, R. (2008, January 19–23). Human detection using multimodal and multidimensional features. Proceedings of the 2008 IEEE International Conference on Robotics and Automation, Pasadena, CA, USA.10.1109/ROBOT.2008.4543708 ↗
49. Camera calibration with distortion models and accuracy evaluation10.1109/34.159901 ↗
50. A Flexible New Technique for Camera Calibration10.1109/34.888718 ↗
51. Zhang, Q., and Pless, R. (October, January 28). Extrinsic Calibration of a Camera and Laser Range Finder (improves camera intrinsic calibration). Proceedings of the IEEE International Conference on Intelligent Robots and Systems, Sendai, Japan.
52. Scaramuzza, D., Harati, A., and Siegwart, R. (November, January 29). Extrinsic self-calibration of a camera and a 3-D laser range finder from natural scenes. Proceedings of the IEEE/RSJ International Intelligent Robots and Systems, San Diego, CA, USA.
53. Budzan, S. (2016, January 19–21). Human detection in low resolution thermal images based on combined HOG classifier. Computer vision and graphics. Proceedings of the International Conference, ICCVG, Lecture Notes in Computer Science, Warsaw, Poland.10.1007/978-3-319-46418-3_27 ↗
54. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns10.1109/TPAMI.2002.1017623 ↗
55. Luber, M., Stork, J., Tipaldi, G., and Arras, K. (2010, January 3–7). People Tracking with Human Motion Predictions from Social Forces. Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), Anchorage, AK, USA.10.1109/ROBOT.2010.5509779 ↗
56. Tracking pedestrians using local spatio-temporal motion patterns in extremely crowded scenes
57. Model-based pedestrian trajectory prediction using environmental sensor for mobile robots navigation
58. Recent advanced statistical background modeling for foreground detection: A systematic survey
59. Welch, G., and Bishop, G. (2001, January 12–17). An introduction to the Kalman filter. Proceedings of the ACM SIGGRAPH, Los Angeles, CA, USA.
60. Line extraction in 2-D range images for mobile robotics10.1023/B:JINT.0000038945.55712.65 ↗
61. Premebida, C., and Nunes, U. (2018, August 27). Laser and Image Pedestrian Detection Dataset—LIPD. Available online: http://www2.isr.uc.pt/~cpremebida/dataset.