Home   >   CSC-OpenAccess Library   >    Manuscript Information
Performance Evaluation of CNN Based Pedestrian and Cyclist Detectors On Degraded Images
Tomáš Zemcík, Lukáš Kratochvíla, Šimon Bilík, Ondrej Boštík, Pavel Zemcík, Karel Horák
Pages - 1 - 13     |    Revised - 31-01-2021     |    Published - 28-02-2021
Volume - 15   Issue - 1    |    Publication Date - February 2021  Table of Contents
MORE INFORMATION
KEYWORDS
Object Detection, Image Degradation, Pedestrian Detection, Cyclist Detection, SSD, Faster R-CNN.
ABSTRACT
This paper evaluates the effects of input image degradation on performance of image object detectors. The purpose of the evaluation is to determine usability of the detectors trained on original images in adverse conditions. SSD and Faster R-CNN based pedestrian and cyclist detector performance with images degraded with motion blur, out-of-focus blur, and JPEG compression artefacts, most commonly occurring in mobile or static traffic systems. An experiment was designed to assess the effect of degradations on detection precision and cross class confusion. The paper describes the two datasets created for this evaluation, evaluation of a number of detectors on increasingly more degraded images, comparison of their performance, and assessment of their tolerance to different types of image degradation as well as a discussion of the results.
1 Google Scholar 
2 refSeek 
3 BibSonomy 
4 Doc Player 
5 J-Gate 
6 Scribd 
7 SlideShare 
Andreas Geiger, Philip Lenz, & Raquel Urtasun (2012). Are we ready for Autonomous Driving? The KITTI Vision Benchmark Suite. In Conference on Computer Vision and Pattern Recognition (CVPR).
European Commission (2019). EU Road Safety Policy Framework 2021-2030 - Next steps towards "Vision Zero". Luxembourg: Office for Official Publications of the European Communities.
Everingham, M., Van Gool, L., Williams, C., Winn, J., & Zisserman, A. (2010). The Pascal Visual Object Classes (VOC) challenge International Journal of Computer Vision, 88, 303-338.
Huang, J., Rathod, V., Sun, C., Zhu, M., Korattikara, A., Fathi, A., Fischer, I., Wojna, Z., Song, Y., Guadarrama, S., & Murphy, K. (2017, Apr.). Speed/Accuracy Trade-Offs for Modern Convolutional Object Detectors. In CVPR 2017 [Online] (pp. 3296-3297). Available: https://arxiv.org/abs/1611.10012. [Nov 15, 2020]
International Telecommunication Union (1992), T.81 – DIGITAL COMPRESSION AND CODING OF CONTINUOUS-TONE STILL IMAGES – REQUIREMENTS AND GUIDELINES. [Online]. Available: https://www.w3.org/Graphics/JPEG/itu-t81.pdf [Nov 12, 2020]
K. He, X. Zhang, S. Ren, & J. Sun (2016, Dec.). Deep Residual Learning for Image Recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [Online] (pp. 770-778). Available: https://arxiv.org/abs/1512.03385. [Dec 01, 2020]
Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., & Berg, A. (2016, Dec). SSD: Single Shot MultiBox Detector, Lecture Notes in Computer Science, [Online]. 21–37. Available: https://arxiv.org/abs/1512.02325. [Dec 15, 2020]
M. Hradiš, J. Kotera, P. Zemcík, & F. Šroubek (2015). Convolutional Neural Networks for Direct Text Deblurring. In Proceedings of BMVC 2015 (pp. 1–13). The British Machine Vision Association and Society for Pattern Recognition.
M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, & L. Chen (2018). MobileNetV2: Inverted Residuals and Linear Bottlenecks. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 4510-4520).
M. Tilgner. “Pedestrians Detection in Traffic Environment by Machine Learning” Master’s thesis, Brno University of Technology, 2019
N. Dalal, & B. Triggs (2005). Histograms of oriented gradients for human detection. In 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) (pp. 886-893 vol. 1).
O. Fard & S. Hamidreza. Efficient multi-class object detection with a hierarchy of classes. Aubičre, 2015, pp. 36-40.
P. F. Felzenszwalb, R. B. Girshick, D. McAllester, & D. Ramanan (2010). Object Detection with Discriminatively Trained Part-Based Models IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(9), 1627-1645.
P. Svoboda, M. Hradiš, D. Barina, & P. Zemcík (2016, May.). Compression Artifacts Removal Using Convolutional Neural Networks, Journal of WSCG, [Online]. 24(2), 63–72. Available: https://arxiv.org/abs/1605.00366. [Nov 21, 2020]
P. Svoboda, M. Hradiš, L. Maršík, & P. Zemcík (2016, Feb.). CNN for license plate motion deblurring. In IEEE International Conference on Image Processing (ICIP) [Online]. (pp. 3832–3836). IEEE Signal Processing Society. Available: https://arxiv.org/abs/1602.07873. [Dec 01, 2020]
R. Benenson, M. Omran, J. Hosang, & B. Schiele. (2014, Nov.). Ten Years of Pedestrian Detection, What Have We Learned?. [Online]. Available: https://arxiv.org/abs/1411.4304 [Nov 13, 2020]
R. Girshick (2015). Fast R-CNN. In 2015 IEEE International Conference on Computer Vision (ICCV) (pp. 1440-1448).
R. Girshick, J. Donahue, T. Darrell, & J. Malik (2014, Oct.). Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. In 2014 IEEE Conference on Computer Vision and Pattern Recognition [Online]. (pp. 580-587). Available: https://arxiv.org/abs/1311.2524. [Dec 01, 2020]
R. Padilla, S. L. Netto, & E. A. B. da Silva (2020, Jul.). A Survey on Performance Metrics for Object-Detection Algorithms. In 2020 International Conference on Systems, Signals and Image Processing (IWSSIP) [Online] (pp. 237-242). Available: https://www.researchgate.net/publication/343194514_A_Survey_on_Performance_Metrics_for_Object-Detection_Algorithms. [Nov 15, 2020
Redmon, J., Divvala, S., Girshick, R.B., & Farhadi, A. (2016). You Only Look Once: Unified, Real-Time Object Detection. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 779-788.
Roy, P., Ghosh, S., Bhattacharya, S., & Pal, U. (2018, Jul.). Effects of Degradations on Deep Neural Network Architectures. [Online]. Available: https://arxiv.org/abs/1807.10108. [Dec 01, 2020]
S. Kim, I. Wisanggeni, R. Ros & R. Hussein (2020, Jan.). Detecting Fatigue Driving Through PERCLOS: A Review, International Journal of Image Processing (IJIP), [Online] Volume (14) : Issue (1) : 2020. Available: https://www.cscjournals.org/manuscript/Journals/IJIP/Volume14/Issue1/IJIP-1194.pdf. [Jan 23, 2021]
S. Ren, K. He, R. Girshick, & J. Sun (2016, Jan). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks IEEE Transactions on Pattern Analysis and Machine Intelligence, [Online]. 39(6), 1137-1149. Available: https://arxiv.org/abs/1506.01497 [Dec 02, 2020]
V. Viswanathan & R. Hussein (2017, Apr.). Applications of Image Processing and Real-Time embedded Systems in Autonomous Cars: A Short Review, International Journal of Image Processing (IJIP), [Online] Volume (11) : Issue (2). Available: https://www.cscjournals.org/manuscript/Journals/IJIP/Volume11/Issue2/IJIP-1114.pdf. [Jan 23, 2021]
Viola, Jones, & Snow (2003, Oct.). Detecting pedestrians using patterns of motion and appearance. In Proceedings Ninth IEEE International Conference on Computer Vision [Online]. (pp. 734-741 vol.2). Available: https://ieeexplore.ieee.org/document/1238422 [Nov 15, 2020]
W. Tian, & M. Lauer (2015, Sep.). Fast Cyclist Detection by Cascaded Detector and Geometric Constraint. In 2015 IEEE 18th International Conference on Intelligent Transportation Systems [Online]. (pp. 1286-1291). Available: https://ieeexplore.ieee.org/document/7313303 [Dec 05, 2020]
Wojek C., Schiele B. (2008) A Performance Evaluation of Single and Multi-feature People Detection. In: Rigoll G. (eds) Pattern Recognition. DAGM 2008. Lecture Notes in Computer Science, vol 5096. Springer, Berlin, Heidelberg. [Online]. Available: https://doi.org/10.1007/978-3-540-69321-5_9 [Nov 23, 2020]
World Health Organization (2018). Global Status Report on Road Safety 2018 [Online]. Available:https://www.who.int/publications/i/item/9789241565684#:~:text=The%20Global%20status%20report%20on, people%20aged%205%2D29%20years. [Nov 15, 2020]
Y. Pei, Y. Huang, Q. Zou, X. Zhang, & S. Wang (2019, Nov.). Effects of Image Degradation and Degradation Removal to CNN-based Image Classification, IEEE Transactions on Pattern Analysis and Machine Intelligence, [Online]. 1-1. Available: https://ieeexplore.ieee.org/document/8889765. [Nov 18, 2020]
Z. Cheng, L. Jeng & K. Li (2018, Nov.) Behavioral Classification of Drivers for Driving Efficiency Related ADAS Using Artificial Neural Network, 2018 IEEE International Conference on Advanced Manufacturing (ICAM), [Online] pp. 173-176. Available: https://ieeexplore.ieee.org/document/8614836. [Nov 15, 2020]
Zhou Wang, A. C. Bovik, H. R. Sheikh, & E. P. Simoncelli (2004, Apr.). Image quality assessment: from error visibility to structural similarity, IEEE Transactions on Image Processing, [Online] 13(4), 600-612. Available: https://ieeexplore.ieee.org/document/1284395. [Dec 01,
Mr. Tomáš Zemcík
Faculty of Electrical Engineering and Communications, Dept. of Automation and Instrumentation, Brno University of Technology, Brno - Czech Republic
xzemci04@stud.feec.vutbr.cz
Mr. Lukáš Kratochvíla
Faculty of Electrical Engineering and Communications, Dept. of Automation and Instrumentation, Brno University of Technology, Brno - Czech Republic
Mr. Šimon Bilík
Faculty of Electrical Engineering and Communications, Dept. of Automation and Instrumentation, Brno University of Technology, Brno - Czech Republic
Mr. Ondrej Boštík
Faculty of Electrical Engineering and Communications, Dept. of Automation and Instrumentation, Brno University of Technology, Brno - Czech Republic
Professor Pavel Zemcík
Faculty of Information Technology, Dept. of Computer Graphics and Multimedia, Brno University of Technology, Brno - Czech Republic
Dr. Karel Horák
Faculty of Electrical Engineering and Communications, Dept. of Automation and Instrumentation, Brno University of Technology, Brno - Czech Republic