Appl Opt. 2021 May 10;60(14):4074-4083. doi: 10.1364/AO.419158.
An algorithm of laser curve segmentation for a train wheelset based on an encoder- decoder network is proposed. Aiming at the rich local features and simple semantic features of the train wheelset laser curve image, a neural network with shallow depth, high resolution, and good detail performance was designed. The proposed neural network makes full use of the dense connection mechanism and the upsampling module to enhance feature reuse and feature propagation. It can extract context semantic information at multiple scales with fewer parameters. Experimental results show that the encoder-decoder network has better performance than other neural networks in laser curve extraction of train wheelset. Based on the encoder-decoder neural network, mIOU, Recall, Accuracy, and F1_score of the laser curve dataset of the train wheelset, the score index reached 86.5%, 89.2%, 99.9%, and 85.0%, which can accurately extract the laser stripe of the train wheelset. Additionally, the encoder-decoder network can diminish the influence of noise on the extraction of laser fringes of a train wheelset to a certain extent. Therefore, it has good application in railway safety.
提出了一种基于编解码器网络的火车轮对激光曲线分割算法。针对火车轮对激光曲线图像丰富的局部特征和简单的语义特征，设计了一种深度浅、分辨率高、细节性能好的神经网络。所提出的神经网络充分利用密集连接机制和上采样模块，增强特征重用和特征传播，可以用较少的参数提取多个尺度的上下文语义信息。实验结果表明，在火车轮对激光曲线提取方面，编解码器网络的性能优于其他神经网络。基于编解码器神经网络，火车轮对激光曲线数据集的 mIOU、Recall、Accuracy 和 F1_score 得分指标分别达到 86.5%、89.2%、99.9%和 85.0%，可以准确提取火车轮对的激光条纹。此外，编解码器网络可以在一定程度上减少噪声对火车轮对激光条纹提取的影响。因此，它在铁路安全中具有很好的应用前景。
