YOLSO: You Only Look Small Object

Small object detection is challenging and far from satisfactory. Most general object detectors suffer from two critical issues with small objects: (1) Feature extractor based on classification network cannot express the characteristics of small objects reasonably due to insufficient appearance information of targets and a large amount of background interference around them. (2) The detector requires a much higher location accuracy for small objects than for general objects. This paper proposes an effective and efficient small object detector YOLSO to address the above problems. For feature representation, we analyze the drawbacks in previous backbones and present a Half-Space Shortcut(HSSC) module to build a background-aware backbone. Furthermore, a coarse-to-fine Feature Pyramid Enhancement(FPE) module is introduced for layer-wise aggregation at a granular level to enhance the semantic discriminability. For loss function, we propose an exponential L1 loss to promote the convergence of regression, and a focal IOU loss to focus on prime samples with high classification confidence and high IOU. Both of them significantly improves the location accuracy of small objects. The proposed YOLSO sets state-of-the-art results on two typical small object datasets, MOCOD and VeDAI, at a speed of over 200 FPS. In the meantime, it also outperforms the baseline YOLOv3 by a wide margin on the common COCO dataset.     

Experimental study on the propagation characteristics of hydrogen/methane/air premixed flames in a narrow channel

This work is focused on the explosion characteristics of premixed gas containing different volume fractions of hydrogen in a narrow channel (1000 mm × 50 mm × 10 mm) under the circumstance of stoichiometric ratio. The ignition positions were set in the closed end and the middle of the pipeline respectively. The results showed that when the gas was ignited at the pipeline closed end, the propagating flame was tulip structure for different premixed gas. When the hydrogen volume fraction was less than 40%, the flame propagation speed increased significantly with the rise of hydrogen volume fraction, and the overpressure peak also appeared obviously in advance. However, when the volume fraction of hydrogen was more than 40%, the increase of flame propagation speed and the overpressure peak occurrence time varied slightly. Furthermore, when the ignition position was placed in the middle of the pipeline, the flame propagation speed propagating to the opening end was much faster than that propagating to the closing end, and there was no tulip shape when the flame propagates to the opening end. The flame propagating to the closed end appeared tulip shape under the influence of airflow, and high-frequency flame oscillation occurred during the propagation. This work shows that the hydrogen volume fraction and ignition position significantly affected the flame structure, flame front speed, and explosion overpressure.     

Feature selection methods for root-cause analysis among top-level product attributes

Manufacturing companies not only strive to deliver flawless products but also monitor product failures in the field to identify potential quality issues. When product failures occur, quality engineers must identify the root cause to improve any affected product and process. This root-cause analysis can be supported by feature selection methods that identify relevant product attributes, such as manufacturing dates with an increased number of product failures. In this paper, we present different methods for feature selection and evaluate their ability to identify relevant product attributes in a root-cause analysis. First, we compile a list of feature selection methods. Then, we summarize the properties of product attributes in warranty case data and discuss these properties regarding the challenges they pose for machine learning algorithms. Next, we simulate datasets of warranty cases, which emulate these product properties. Finally, we compare the feature selection methods based on these simulated datasets. In the end, the univariate filter information gain is determined to be a suitable method for a wide range of applications. The comparison based on simulated data provides a more general result than other publications, which only focus on a single use case. Due to the generic nature of the simulated datasets, the results can be applied to various root-cause analysis processes in different quality management applications and provide a guideline for readers who wish to explore machine learning methods for their analysis of quality data.     

基准平面垂直断面法在爆破漏斗试验中的应用

系统阐述了基准平面垂直断面法在爆破漏斗试验中测量爆破漏斗体积的基本原理,并将隧道激光断面仪应用于金厂河矿1 750 m水平15^#采场底部切割巷道爆破漏斗试验爆破漏斗体积测量中。通过与传统体重法等计算法所得漏斗体积分析比较,结果表明基于隧道激光断面仪与3D Mine软件分析的基准平面垂直断面法实用性强、操作方便、结果直观可靠,达到试验预期目的。     

Understanding the asymmetric perceptions of smartphone security from security feature perspective: A comparative study

Smartphones are being used and relied on by people more than ever before. The open connectivity brings with it great convenience and leads to a variety of risks that cannot be overlooked. Smartphone vendors, security policy designers, and security application providers have put a variety of practical efforts to secure smartphones, and researchers have conducted extensive research on threat sources, security techniques, and user security behaviors. Regrettably, smartphone users do not pay enough attention to mobile security, making many efforts futile. This study identifies this gap between technology affordance and user requirements, and attempts to investigate the asymmetric perceptions toward security features between developers and users, between users and users, as well as between different security features. These asymmetric perceptions include perceptions of quality, perceptions of importance, and perceptions of satisfaction. After scoping the range of smartphone security features, this study conducts an improved Kano-based method and exhaustively analyzes the 245 collected samples using correspondence analysis and importance satisfaction analysis. The 14 security features of the smartphone are divided into four Kano quality types and the perceived quality differences between developers and users are compared. Correspondence analysis is utilized to capture the relationship between the perceived importance of security features across different groups of respondents, and results of importance-satisfaction analysis provide the basis for the developmental path and resource reallocation strategy of security features. This article offers new insights for researchers as well as practitioners of smartphone security.     

急性炎性疾病患者乳酸林格氏液液体动力学

目的:探索炎性疾病患者的乳酸林格氏液（Ringer's lactate,RL）液体动力学特征以及炎性生物标记物是否可以作为协变量影响RL分布和排泄。方法:本研究为前瞻性队列研究。选择40例美国麻醉医师分级（ASA）I-II级，腹腔镜下择期胆囊切除术（胆囊炎组，n=20）或者腹腔镜下急诊阑尾切除术（阑尾炎组，n=20）。所有患者麻醉诱导前开始输注RL，按15 mL/kg，35 min内输毕。采用酶联免疫（enzyme-linked immunosorbent assay,ELISA）方法测定血浆炎症（TNF-α,IL-10和CRP）或者内皮损伤生物标记物（syndecan-1,SDC-1）；利用血红蛋白（Hb）稀释-时间曲线和尿量，使用Phoenix软件，采用非线性混合效应模型分析计算RL液体动力学参数和协变量的影响。结果:与胆囊炎组相比，阑尾炎组RL从组织间隙到血浆的转运速率常数（k21）显著降低（14×10-3min-1 versus 35×10-3min-1;P=0.012）。阑尾炎组C反应蛋白（CRP）升高［中位数38.1(1.8-143.6) μg/mL versus 1.3(0.1-159.0) μg/mL;P<0.001］；与清醒状态相比，麻醉期间（输液开始后30～45 min），液体从中央室中到外周室的转运速率常数（k12）显著增加（57×10-3min-1 versus 32×10-3min-1；P<0.01）。清除速率常数（k10）降低90%（0.6×10-3min-1 versus 5.3×10-3min-1;P<0.001）。无论在清醒状态还是麻醉状态下低血压均能降低液体清除；炎症或者内膜损伤的生物标记物不能作为显著影响RL液体动力学参数的协变量。结论:阑尾炎或者胆囊炎患者术前输入液体后“炎症反应的生物标记物”不是RL的液体动力学的协变量，但是两组患者中，全身麻醉期间输入液体的清除率下降。     

基于梯度提升决策树的车轮轮缘厚度磨耗预测

轮对在列车走行过程中起着导向、承受以及传递载荷的作用，其踏面及轮缘磨耗对地铁列车运行安全性和钢轨的寿命都将产生重要影响。根据地铁列车车轮磨耗机理，分析车轮尺寸数据特点，针对轮缘厚度这一型面参数，基于梯度提升决策树算法构建轮缘厚度磨耗预测模型。在该模型的基础上，任意选取某轮对数据进行验证分析，结果表明：基于梯度提升决策树的轮对磨耗预测模型具有较好的预测精度，可预测出1~6个月的轮缘厚度变化趋势范围，预测时间范围较长，可为地铁维保部门对轮对的维修方式由状态修转为预防修提供指导性建议。     

基于深度学习的秀丽隐杆线虫显微图像分割方法

利用计算机实现自动、准确的秀丽隐杆线虫(C.elegans)的各项形态学参数分析,至关重要的是从显微图像上分割出线虫体态,但由于显微镜下的图像噪声较多,线虫边缘像素与周围环境相似,而且线虫的体态具有鞭毛和其他附着物需要分离,多方面因素导致设计一个鲁棒性的C.elegans分割算法仍然面临着挑战。针对这些问题,提出了一种基于深度学习的线虫分割方法,通过训练掩模区域卷积神经网络(Mask R-CNN)学习线虫形态特征实现自动分割。首先,通过改进多级特征池化将高级语义特征与低级边缘特征融合,结合大幅度软最大损失(LMSL)损失算法改进损失计算;然后,改进非极大值抑制;最后,引入全连接融合分支等方法对分割结果进行进一步优化。实验结果表明,相比原始的Mask R-CNN,该方法平均精确率(AP)提升了4.3个百分点,平均交并比(mIOU)提升了4个百分点。表明所提出的深度学习分割方法能够有效提高分割准确率,在显微图像中更加精确地分割出线虫体。     

Design and experimental investigation of the high efficiency 1.5 W/4.2 K pneumatic-drive G-M cryocooler

The development of a high cooling power and high efficiency 4.2 K two stage G-M cryocooler is critically important given its broad applications in low temperature superconductors, MRI, infrared detector and cryogenic electronics. A high efficiency 1.5 W/4.2 K pneumatic-drive G-M cryocooler has recently been designed and developed by ARS. The effect of expansion volume rate and operation conditions on the cooling performance has been experimentally investigated. A typical cooling performance of 1.5 W/4.2 K has been achieved, and the minimum temperature of the second stage is 2.46 K. The steady input power of the compressor at 60 Hz is 6.8 kW, while the operation speed of the rotary valve is 30 rpm. A maximum cooling power of 1.75 W/4.2 K has been obtained in test runs.     

基于RFID标签阵列的睡眠期间呼吸量连续监测系统

睡眠期间连续且准确的呼吸量监测有助于推断用户的睡眠阶段以及提供一些慢性疾病的线索。现有工作主要针对呼吸频率进行感知和监测,缺乏对呼吸量进行连续监测的手段。针对上述问题提出了一种基于商用无线射频识别(RFID)标签的无线感知用户睡眠期间呼吸量的系统——RF-SLEEP。RF-SLEEP通过阅读器连续收集附着在胸部表面的标签阵列返回的相位值及时间戳数据,计算出呼吸引起的胸部不同点的位移量,基于广义回归神经网络(GRNN)构建胸部不同点的位移量与呼吸量之间的关系模型,从而实现对用户睡眠期间呼吸量的评估。RF-SLEEP通过在用户肩膀处附着双参考标签,消除用户睡眠期间翻转身体对胸部位移计算造成的误差。实验结果表明,RFSLEEP对不同用户睡眠期间的呼吸量连续监测的平均精确度为92.49%。