A deep learning approach to multi-track location and orientation in gaseous drift chambers-猫眼课题宝

气体漂移室中多轨道定位和定向的深度学习方法

DOI：

10.1016/j.nima.2020.164640

发表时间：

2020-05

期刊：

Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Ai Pengcheng;Wang Dong;Sun Xiangming;Huang Guangming;Li Zili

在光束监测系统中精确测量单个粒子的位置和方向是多学科研究人员特别感兴趣的问题。在可行的方法中，采用混合像素传感器的气体漂移室具有实现长期稳定、高精度测量的巨大潜力。在本文中，我们引入深度学习来分析光束投影图像中的模式，以促进粒子轨迹的三维重建。我们提出了一种基于分割和拟合的端到端神经网络用于特征提取和回归。两个分割分支分别称为二值分割和语义分割，分别进行初始航迹确定和像素航迹关联。然后将像素分配到多个轨道上，并利用全反向传播实现加权最小二乘拟合。此外，我们还引入了一种中心角测量方法，结合两个独立的因素来判断定位和定向的精度。初始位置分辨率单道为8.8 μ m， 1-3道（1-5道）为11.4 μ m (15.2 μ m)，角度分辨率分别为0.15°和0.21°（0.29°）。这些结果表明，与传统方法相比，在精度和多道兼容性方面有了显著提高。

Accurate measuring the location and orientation of individual particles in a beam monitoring system is of particular interest to researchers in multiple disciplines. Among feasible methods, gaseous drift chambers with hybrid pixel sensors have the great potential to realize long-term stable measurement with considerable precision. In this paper, we introduce deep learning to analyze patterns in the beam projection image to facilitate three-dimensional reconstruction of particle tracks. We propose an end-to-end neural network based on segmentation and fitting for feature extraction and regression. Two segmentation branches, named binary segmentation and semantic segmentation, perform initial track determination and pixel-track association. Then pixels are assigned to multiple tracks, and a weighted least squares fitting is implemented with full back-propagation. Besides, we introduce a center-angle measure to judge the precision of location and orientation by combining two separate factors. The initial position resolution achieves 8.8 μ m for the single track and 11.4 μ m (15.2 μ m) for the 1–3 tracks (1–5 tracks), and the angle resolution achieves 0.15° and 0.21°(0.29°) respectively. These results show a significant improvement in accuracy and multi-track compatibility compared to traditional methods.