Detecção de Faixas de Trânsito em Tempo Real usando um Sistema Embarcados de Baixo Custo e Implementação em CUDA

Detecção de Faixas de Trânsito em Tempo Real usando um Sistema Embarcados de Baixo Custo e Implementação em CUDA

Autores

DOI:

https://doi.org/10.5433/1679-0375.2023.v44.48268

Palavras-chave:

detecção de faixas de trânsito, aplicação em tempo real, CUDA, computação henerogênea

Resumo

Este trabalho avalia a eficácia da computação heterogênea, com base em uma implementação CUDA, para detecção de faixas de sinalização de trânsito em tempo real um computador embarcado de baixo custo típico. O trabalho propõe e analisa um algoritmo com otimizações CUDA usando uma abordagem heterogênea baseada na extração de características de uma imagem em perspectiva aérea. O método incorpora algoritmos conhecidos otimizados para obter uma solução muito eficiente com altas taxas de detecção, além de combinar técnicas para melhorar as marcações e remover ruídos. A solução baseada em CUDA é comparada a uma biblioteca OpenCV e a uma implementação sequencial em CPU. O método é avaliado por um experimento prático usando conjuntos de dados de imagens do banco de dados TuSimple em um computador embarcado NVIDIA Jetson Nano. O algoritmo detecta até 97,9% das faixas de sinalização com uma precisão de 99,0% no melhor cenário avaliado. Além disso, o algoritmo com otimizações em CUDA resulta em taxas superiores a 300 fps, acelerando 25 vezes e 140 vezes a implementação do OpenCV e da CPU, respectivamente, todas avaliadas no computador embarcado NVIDIA Jetson Nano. Esses resultados mostram que algoritmos e soluções mais complexos podem ser empregados para obter melhores taxas de detecção, mantendo os requisitos em tempo real em um computador embarcado de baixa potência típico usando uma implementação CUDA.

Downloads

Não há dados estatísticos.

Biografia do Autor

Guilherme Brandão da Silva, Universidade Estadual de Londrina - UEL

Mestre, Programa de Mestrado em Engenharia Elétrica, UEL, Londrina, PR, Brasil

Daniel Strufaldi Batista, Universidade Estadual de Londrina - UEL

Prof. Dr., Departamento de Engenharia Elétrica, UEL, Londrina, PR, Brasil

Décio Luiz Gazzoni Filho, Universidade Estadual de Londrina - UEL

Prof. Dr., Departamento de Engenharia Elétrica, UEL, Londrina, PR, Brasil

Marcelo Carvalho Tosin, Universidade Estadual de Londrina - UEL

Prof. Dr., Departamento de Engenharia Elétrica, UEL, Londrina, PR, Brasil

Leonimer Flávio Melo, Universidade Estadual de Londrina - UEL

Prof. Dr., Departamento de Engenharia Elétrica, UEL, Londrina, PR, Brasil

Referências

Afif, M., Said, Y., & Atri, M. (2020). Computer vision algorithms acceleration using graphic processors NVIDIA CUDA. Cluster Computing, 23(4), 3335–3347. DOI: https://doi.org/10.1007/s10586-020-03090-6

Borkar, A., Hayes, M., & Smith, M. T. (2012). A novel lane detection system with efficient ground truth generation. IEEE Transactions on Intelligent Transportation Systems, 13(1), 365–374. DOI: https://doi.org/10.1109/TITS.2011.2173196

Borkar, A., Hayes, M., & Smith, M. T. (2009). Robust lane detection and tracking with Ransac and Kalman filter. In Institute of Electrical and Electronics Engineers, Conferences [Proceedings]. 16th IEEE International Conference on Image Processing (ICIP), Cairo, Egypt. DOI: https://doi.org/10.1109/ICIP.2009.5413980

Bradski, G. (2000). The OpenCV Library. Dr. Dobb’s Journal of Software Tools, 120, 122–125.

Cao, J., Song, S., Xiao, W., & Peng, Z. (2019). Lane detection algorithm for intelligent vehicles in complex road conditions and dynamic environments. Sensors, 19, 3166. DOI: https://doi.org/10.3390/s19143166

Gansbeke, W. V., Brabandere, B. D., Neven, D., Proesmans, M., & Gool, L. V. (2019). End-to-end Lane Detection through Differentiable Least-Squares Fitting. ArXiv:1902.00293v3 [cs.CV].

Hartley, R., & Zisserman, A. (2003). Multiple View Geometry in Computer Vision (2nd ed.). Cambridge University Press. DOI: https://doi.org/10.1017/CBO9780511811685

He, B., Ai, R., Yan, Y., & Lang, X. (2016). Accurate and robust lane detection based on Dual-View Convolutional Neutral Network. In Institute of Electrical and Electronics Engineers, Conferences [Proceedings]. IEEE 4th Intelligent Vehicles Symposium, Gothenburg, Sweden.

Hernández, D. C., Filonenko, A., Shahbaz, A., & Jo, K.-H. (2017). Lane marking detection using image features and line fitting model. In Institute of Electrical and Electronics Engineers, Conferences [Proceedings]. 10th International Conference on Human System Interactions (HSI), Ulsan, Korea.

Hillel, A. B., Lerner, R., Levi, D., & Raz, G. (2014). Recent progress in road and lane detection: A survey. Machine Vision and Applications, 25, 727–74. DOI: https://doi.org/10.1007/s00138-011-0404-2

Huang, Y., Li, Y., Hu, X., & Ci, W. (2018). Lane Detection Based on Inverse Perspective Transformation and Kalman Filter. KSII Transactions on Internet and Information Systems. Korean Society for Internet Information (KSII), 12(2), 643–661. DOI: https://doi.org/10.3837/tiis.2018.02.006

Jaiswal, D., & Kumar, P. (2020). Real-time implementation of moving object detection in UAV videos using GPUs. Journal of Real-Time Image Processing, 17(5), 1301–1317. DOI: https://doi.org/10.1007/s11554-019-00888-5

Kim, H.-S., Beak, S.-H., & Park, S.-Y. (2016). Parallel Hough Space Image Generation Method for RealTime Lane Detection. In J. Blanc-Talon, C. Distante, W. Philips, D. Popescu, & P. Scheunders (Eds.), Advanced Concepts for Intelligent Vision Systems (pp. 81–91). Springer International Publishing. DOI: https://doi.org/10.1007/978-3-319-48680-2_8

Kim, J., Kim, J., Jang, G.-J., & Lee, M. (2017). Fast learning method for convolutional neural networks using extreme learning machine and its application to lane detection. Neural Networks, 87, 109–121. DOI: https://doi.org/10.1016/j.neunet.2016.12.002

Kirk, D. B., & Hwu, W.-M. W. (2016). Programming massively parallel processors: A hands-on approach (3 th ed.). Elsevier.

Küçükmanisa, A., Tarim, G., & Urhan, O. (2019). Realtime illumination and shadow invariant lane detection on mobile platform. Journal of Real-Time Image Processing, 16(5), 1781–1794. DOI: https://doi.org/10.1007/s11554-017-0687-2

Kühnl, T., Kummert, F., & Fritsch, J. (2012). Spatial ray features for real-time ego-lane extraction. In Institute of Electrical and Electronics Engineers, Conferences [Proceedings]. 15th International IEEE Conference on Intelligent Transportation Systems, Anchorage, USA. DOI: https://doi.org/10.1109/ITSC.2012.6338740

Lee, C., & Moon, J.-H. (2018). Robust Lane Detection and Tracking for Real-Time Applications. IEEE Transactions on Intelligent Transportation Systems, 19(12), 4043–4048. DOI: https://doi.org/10.1109/TITS.2018.2791572

Li, J., Deng, G., Zhang, W., Zhang, C., Wang, F., & Liu, Y. (2020). Realization of CUDA-based real-time multi-camera visual SLAM in embedded systems. Journal of Real-Time Image Processing, 17(3), 713–727. DOI: https://doi.org/10.1007/s11554-019-00924-4

Li, W., Qu, F., Wang, Y., Wang, L., & Chen, Y. (2019). A robust lane detection method based on hyperbolic model. Soft Computing, 23(19), 9161–9174. DOI: https://doi.org/10.1007/s00500-018-3607-x

Li, X., Fang, X., Ci, W., & Zhang, W. (2014). Lane Detection and Tracking Using a Parallel-snake Approach. Journal of Intelligent and Robotic Systems, 77(3), 597–609. DOI: https://doi.org/10.1007/s10846-014-0075-0

Mammeri, A., Boukerche, A., & Tang, Z. (2016). A realtime lane marking localization, tracking and communication system. Computer Communications, 73, 132–143. DOI: https://doi.org/10.1016/j.comcom.2015.08.010

Muthalagu, R., Bolimera, A., & Kalaichelvi, V. (2020). Lane detection technique based on perspective transformation and histogram analysis for selfdriving cars. Computers & Electrical Engineering, 85, 106653. DOI: https://doi.org/10.1016/j.compeleceng.2020.106653

Narote, S. P., Bhujbal, P. N., Narote, A. S., & Dhane, D. M. (2018). A review of recent advances in lane detection and departure warning system. Pattern Recognition, 73, 216–234. DOI: https://doi.org/10.1016/j.patcog.2017.08.014

Nguyen, V., Kim, H., Jun, S., & Boo, K. (2018). A study on real-time detection method of lane and vehicle for lane change assistant system using vision system on highway. Engineering Science and Technology, an International Journal, 21(5), 822–833. DOI: https://doi.org/10.1016/j.jestch.2018.06.006

Nvidia. (2011). NVIDIA CUDA Programming Guide. NVIDIA Corporation.

Nvidia. (2014). GeForce GTX 980 - Featuring Maxwell, The Most Advanced GPU Ever Made. NVIDIA Corporation.

Paula, M. B., & Jung, C. R. (2015). Automatic Detection and Classification of Road Lane Markings Using Onboard Vehicular Cameras. IEEE Transactions on Intelligent Transportation Systems, 16(6), 3160–3169. DOI: https://doi.org/10.1109/TITS.2015.2438714

Philion, J. (2019). FastDraw: Addressing the Long Tail of Lane Detection by Adapting a Sequential Prediction Network. In Institute of Electrical and Electronics Engineers, Conferences [Proceedings]. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach, USA, 11574–11583. DOI: https://doi.org/10.1109/CVPR.2019.01185

Reichenbach, M., Liebischer, L., Vaas, S., & Fey, D. (2018). Comparison of Lane Detection Algorithms for ADAS Using Embedded Hardware Architectures. In Institute of Electrical and Electronics Engineers, Conferences [Proceedings]. Conference on Design and Architectures for Signal and Image Processing (DASIP). Porto, Portugal, 48–53. DOI: https://doi.org/10.1109/DASIP.2018.8596994

Sanders, J., & Kandrot, E. (2010). CUDA by Example: An Introduction to General-Purpose GPU Programming. Addison-Wesley Professional.

Selim, E., Alci, M., & Uğur, A. (2022). Design and implementation of a real-time LDWS with parameter space filtering for embedded platforms. Journal of Real-Time Image Processing, 19(3), 663–673. DOI: https://doi.org/10.1007/s11554-022-01213-3

Silva, G., Batista, D., Tosin, M., & Melo, L. (2020). Estratégia de detecção de faixas de trânsito baseada em câmera monocular para sistemas embarcados. Sociedade Brasileira de Automática, Anais do 23º Congresso Brasileiro de Automática [Anais]. Congresso Brasileiro de Automática, Campinas, Brasil.

Silva, G. B. (2021a). Cuda-based real-time ego-lane detection in embedded system-tusimple #0601. https: //youtu.be/1q%5C_Jy8MJjFw

Silva, G. B. (2021b). Método heterogêneo de detecção de faixas de trânsito em tempo real para sistemas embarcados. [Dissertation Master’s thesis, State University of Londrina]. Digital Library. http:// www.bibliotecadigital.uel.br/document/?code=vtls 000236022.

Singh, S. (2015). Critical reasons for crashes investigated in the national motor vehicle crash causation survey (Traffic Safety Facts Crash Stats Report DOT HS 812 115). National Highway Traffic Safety Administration.

Sivaraman, S., & Trivedi, M. M. (2013). Looking at vehicles on the road: A survey of vision-based vehicle detection, tracking, and behavior analysis. IEEE Transactions on Intelligent Transportation Systems, 14(4), 1773–1795. DOI: https://doi.org/10.1109/TITS.2013.2266661

Son, J., Yoo, H., Kim, S., & Sohn, K. (2015). Real-time illumination invariant lane detection for lane departure warning system. Expert Systems with Applications, 42(4), 1816–1824. DOI: https://doi.org/10.1016/j.eswa.2014.10.024

Son, Y., Lee, E., & Kum, D. (2019). Robust multi-lane detection and tracking using adaptive threshold and lane classification. Machine Vision and Applications, 30, 111–124. DOI: https://doi.org/10.1007/s00138-018-0977-0

Wang, Y., Teoh, E., & Shen, D. (2004). Lane detection and tracking using b-snake. Image and Vision Computing, 22, 269–280. DOI: https://doi.org/10.1016/j.imavis.2003.10.003

Wu, C.-B., Wang, L.-H., & Wang, K.-C. (2019). Ultra-low complexity block-based lane detection and departure warning system. IEEE Trans on Circuits and Systems for Video Technology, 29(2), 582–593. DOI: https://doi.org/10.1109/TCSVT.2018.2805704

Yenikaya, S., Yenikaya, G., & Düven, E. (2013). Keeping the vehicle on the road: A survey on on-road lane detection systems. ACM Computing Surveys, 46(1). DOI: https://doi.org/10.1145/2522968.2522970

Yonglong, Z., Kuizhi, M., Xiang, J., & Peixiang, D. (2013). Parallelization and Optimization of SIFT on GPU Using CUDA. In Institute of Electrical and Electronics Engineers, Conferences [Proceedings]. 10th International Conference on High Performance Computing and Communications & 2013 IEEE International Conference on Embedded and Ubiquitous Computing. Zhangjiajie, China. DOI: https://doi.org/10.1109/HPCC.and.EUC.2013.192

Yu, Y., & Jo, K.-H. (2018). Lane detection based on color probability model and fuzzy clustering. In H. Yu & J. Dong (Eds.), Ninth International Conference on Graphic and Image Processing (ICGIP 2017). [Proceedings]. Washington, USA. DOI: https://doi.org/10.1117/12.2302941

Zeng, H., Peng, N., Yu, Z., Gu, Z., Liu, H., & Zhang, K. (2015). Visual tracking using multi-channel correlation filters. In Institute of Electrical and Electronics Engineers, Conferences [Proceedings]. IEEE International Conference on Digital Signal Processing (DSP), Singapore. DOI: https://doi.org/10.1109/ICDSP.2015.7251861

Zhang, Z., & Ma, X. (2019). Lane Recognition Algorithm Using the Hough Transform Based on Complicated Conditions. Journal of Computer and Communications, 7(11), 65–75. DOI: https://doi.org/10.4236/jcc.2019.711005

Zheng, F., Luo, S., Song, K., Yan, C.-W., & Wang, M.-C. (2018). Improved lane line detection algorithm based on hough transform. Pattern Recognition and Image Analysis, 28(2), 254–260. DOI: https://doi.org/10.1134/S1054661818020049

Zhi, X., Yan, J., Hang, Y., & Wang, S. (2019). Realization of CUDA-based real-time registration and target localization for high-resolution video images. Journal of Real-Time Image Processing, 16(4), 1025–1036. DOI: https://doi.org/10.1007/s11554-016-0594-y

Zou, Q., Jiang, H., Dai, Q., Yue, Y., Chen, L., & Wang, Q. (2020). Robust lane detection from continuous driving scenes using deep neural networks. IEEE Trans on Vehicular Technology, 69(1), 41–54. DOI: https://doi.org/10.1109/TVT.2019.2949603

Publicado

2023-09-11

Como Citar

da Silva, G. B., Batista, D. S., Gazzoni Filho, D. L., Tosin, M. C., & Melo, L. F. (2023). Detecção de Faixas de Trânsito em Tempo Real usando um Sistema Embarcados de Baixo Custo e Implementação em CUDA. Semina: Ciências Exatas E Tecnológicas, 44, e48268. https://doi.org/10.5433/1679-0375.2023.v44.48268

Edição

Seção

Engenharias
Loading...