Active Obstacle Detection System Based on Video Recognition and Lidar Information Fusion

Authors

  • Jiang Yaodong

DOI:

https://doi.org/10.37819/nm.001.01.0073

Keywords:

Obstacle detection, Convolutional neural network, Autonomous driving, Rail transit

Abstract

In terms of the requirements for obstacle detection in the rail transit application field, an architecture and implementation method for active obstacle detection system based on the fusion of video recognition and lidar information is proposed. The studies on the video recognition algorithms based on deep learning neural network and lidar for orbit area recognition, pedestrian vehicle recognition, and small foreign object recognition are analyzed, and the necessity of the fusion of video recognition and lidar data and the related key technical points are discussed. Through the tests on domestic metro and tram lines, the feasibility of the scheme is verified, and the technical parameters are optimized, which can effectively reduce the probability of accidents caused by foreign object intrusion.

References

Gowtham, R. (2016). Obstacle Detection and Avoidance Methods for Autonomous Mobile Robot. International Journal of Science & Research, 5(1), 1245-1249.

He, K., Gkioxari G, Dollar P, et al. (2020). Mask R-CNN. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(2), 386-397.

Hua, C. (2019). Tensorflow and convolutional neural network: from algorithm introduction to project practice. Beijing: Publishing House of Electronics Industry.

Kim, J., & Cho, J. (2019). YOLO-based real-time object detection scheme combining RGB image with LiDAR point cloud. Journal of Korean Institute of Information Technology, 17(8), 93-105.

Lin, G., Liu, F., Milan, A., Shen, C., & Reid, I. (2019). Refinenet: Multi-path refinement networks for dense prediction. IEEE transactions on pattern analysis and machine intelligence, 42(5), 1228-1242. https://doi.org/10.1109/TPAMI.2019.2893630

Ople, J. J. M., Yeh, P. Y., Sun, S. W., Tsai, I. T., & Hua, K. L. (2020). Multi-Scale Neural Network with Dilated Convolutions for Image Deblurring. IEEE Access, 8, 53942-53952. https://doi.org/10.1109/ACCESS.2020.2980996

Ren, S., He, K., Girshick, R., & Sun, J. (2017). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE transactions on pattern analysis and machine intelligence, 39(6), 1137-1149. https://doi.org/10.1109/tpami.2016.2577031

Downloads

Published

2020-12-02

How to Cite

Yaodong, J. . (2020). Active Obstacle Detection System Based on Video Recognition and Lidar Information Fusion. New Metro, 1(1), 11-21. https://doi.org/10.37819/nm.001.01.0073

Issue

Section

Intelligent Metro