Zhang Haibo, Yang Jinfeng, Tao Wei, Zhao Hui
Department of Instrument Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China.
Appl Opt. 2011 Jul 10;50(20):3658-65. doi: 10.1364/AO.50.003658.
The increasing train speed on railways generates an urgent need for more powerful automatic inspection of railway tracks, including real-time fastening component inspection. To obtain better high-speed performance with lower cost, this paper has proposed a novel structured light method based on motion image (SLMMI) for moving object inspection. The motion images in the proposed method are insensitive to motion, abundant with information, and easy to process, resulting in a low performance requirement of the hardware. Compared to the conventional unstructured light method and structured light method, the proposed method inherits the virtues of both thus offering a fresh perspective when inspecting missing fastening components on high-speed railways. By using the SLMMI and the recognition method based on a neural network, the experimental results yield good performance in terms of speed and accuracy. Furthermore, the robustness of the proposed method is also discussed and simulated by adding typical interferences, such as ambient light, vibration, and obstacles.
铁路列车速度的不断提高，迫切需要对铁轨进行更强大的自动检测，包括实时紧固部件检测。为了以更低的成本获得更好的高速性能，本文提出了一种基于运动图像的新型结构光方法（SLMMI）用于移动物体检测。该方法中的运动图像对运动不敏感，信息丰富且易于处理，从而对硬件的性能要求较低。与传统的非结构光方法和结构光方法相比，该方法兼具两者优点，为高速铁路上缺失紧固部件的检测提供了新视角。通过使用SLMMI和基于神经网络的识别方法，实验结果在速度和准确性方面表现良好。此外，还通过添加典型干扰（如环境光、振动和障碍物）对该方法的鲁棒性进行了讨论和模拟。
