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.     

用于电弧检测的电力专用SoC设计

针对传统的电弧电路故障检测结果不准确的问题,设计用于电弧检测的SoC系统,并且在55nm工艺下进行流片验证。采用包含两种结构的模数转换器的片上电压源,设计了锁相环以及复位电路,精度最高可达8.67 bit。验证结果表明,本设计可提高电弧检测的准确性。     

Machine learning algorithm selection for windage alteration fault diagnosis of mine ventilation system

Machine learning algorithms have been widely used in mine fault diagnosis. The correct selection of the suitable algorithms is the key factor that affects the fault diagnosis. However, the impact of machine learning algorithms on the prediction performance of mine fault diagnosis models has not been fully evaluated. In this study, the windage alteration faults (WAFs) diagnosis models, which are based on K-nearest neighbor algorithm (KNN), multi-layer perceptron (MLP), support vector machine (SVM), and decision tree (DT), are constructed. Furthermore, the applicability of these four algorithms in the WAFs diagnosis is explored by a T-type ventilation network simulation experiment and the field empirical application research of Jinchuan No. 2 mine. The accuracy of the fault location diagnosis for the four models in both networks was 100%. In the simulation experiment, the mean absolute percentage error (MAPE) between the predicted values and the real values of the fault volume of the four models was 0.59%, 97.26%, 123.61%, and 8.78%, respectively. The MAPE for the field empirical application was 3.94%, 52.40%, 25.25%, and 7.15%, respectively. The results of the comprehensive evaluation of the fault location and fault volume diagnosis tests showed that the KNN model is the most suitable algorithm for the WAFs diagnosis, whereas the prediction performance of the DT model was the second-best. This study realizes the intelligent diagnosis of WAFs, and provides technical support for the realization of intelligent ventilation.     

Synthesis,characterization and magnetic properties of polythiophene/SrFe12-x(ZrZn)0.5xO19 nanocomposite as a microwave absorber

Ferrites are an important group of magnetic materials which are used as absorbers. The incorporation of ferrite and conducting polymer achieves great enhancement in microwave absorption properties. The nanocomposites of hexagonal ferrites embedded by conducting polymers such as polypyrrole, polyaniline and polythiophene (PTH) have been paid much attention. In the present study, strontium hexagonal ferrite doped by Zr and Zn with the final formula of SrFe_12-x(ZrZn)_0.5xO19 considering x = 0.9 and embedded by PTH was produced to achieve a nanocomposite with the highest microwave absorbing ability. In this study, after synthesis of SrFe₁₂O₁₉(ZrZn)_0.5xO19 and PTH, the nanocomposite was prepared by in situ polymerization. Wrapping the ferrite particles and PTH chains could form nanocomposite properly, and therefore acceptable interactions were observable between SrFe_12-x(ZrZn)_0.5xO19ferrite particles and PTH polymer chains in the composites. Assessing the X-ray diffraction (XRD) patterns of SrFe_12-x(ZrZn)_0.5xO19, PTH, and PTH/SrFe_12-x(ZrZn)_0.5xO19 nanocomposite indicated that the PTH characteristic peak shifts slightly and its peak intensity reduces, which may be attribute to the coating of PTH polymer chains onto SrFe_12-x(ZrZn)_0.5xO19 particles. We revealed also lower magnetic properties in the obtained nanocomposite. The morphological assessment also suggested that PTH could effectively coat the SrFe_12-x(ZrZn)_0.5xO19 particles. The synergistic effect of SrFe_12-x(ZrZn)_0.5xO19 particle plus PTH leads to microwave absorption percentage higher than 95% by PTH/SrFe_12-x(ZrZn)_0.5xO19 nanocomposite. Overall, nanocomposite creating by coupling interaction between SrFe_12-x(ZrZn)_0.5xO19 particles (x = 0.9) and PTH can effectively lead to achieve the highest rate of absorption of electromagnetic waves.     

毫米波三维异质异构集成技术的进展与应用

三维异质异构集成技术是实现电子信息系统向着微型化、高效能、高整合、低功耗及低成本方向发展的最重要方法,也是决定信息化平台中微电子和微纳系统领域未来发展的一项核心高技术。文章详细介绍了毫米波频段三维异质异构集成技术的优势、近年来的发展趋势以及面临的挑战。利用硅基MEMS 光敏复合薄膜多层布线工艺可实现异质芯片的低损耗互连,同时三维集成高性能封装滤波器、高辐射效率封装天线等无源元件,还能很好地处理布线间的电磁兼容和芯片间的屏蔽问题。最后介绍了一款新型毫米波三维异质异构集成雷达及其在远距离生命体征探测方面的应用。     

基于PBFT区块链技术的电网企业仓储系统

提出了一种基于实用拜占庭容错(PBFT)算法的区块链技术,首先对传统的实用拜占庭容错算法原理进行了阐述,该传统算法包含前期、需求、预准备、准备、确认、答复6个阶段,但传统算法具有实时性差、缺乏惩罚机制、带宽高的缺点。针对出现的这些问题,又对传统算法进行了改进,具体涉及记账节点、共识过程以及视图切换过程。通过测试进一步证明了该改进算法的实用性,并将该算法应用于电网企业中,构建的虚拟仓库实现了联储联备,降低了库存资金的耗费,并且提高了电网企业库存管理的效率。     

Performance investigation on a novel 3D wave flow channel design for PEMFC

Flow field structure can largely determine the output performance of Polymer electrolyte membrane fuel cell. Excellent channel configuration accelerates electrochemical reactions in the catalytic layer, effectively avoiding flooding on the cathode side. In present study, a three-dimensional, multi-phase model of PEMFC with a 3D wave flow channel is established. CFD method is applied to optimize the geometry constructions of three-dimensional wave flow channels. The results reveal that 3D wave flow channel is overall better than straight channel in promoting reactant gases transport, removing liquid water accumulated in microporous layer and avoiding thermal stress concentration in the membrane. Moreover, results show the optimal flow channel minimum depth and wave length of the 3D wave flow channel are 0.45 mm and 2 mm, respectively. Due to the periodic geometric characteristics of the wave channel, the convective mass transfer is introduced, improving gas flow rate in through-plane direction. Furthermore, when the cell output voltage is 0.4 V, the current density in the novel channel is 23.8% higher than that of conventional channel.     

A precise sensor fault detection technique using statistical techniques for wireless body area networks

One of the major challenges in wireless body area networks (WBANs) is sensor fault detection. This paper reports a method for the precise identification of faulty sensors, which should help users identify true medical conditions and reduce the rate of false alarms, thereby improving the quality of services offered by WBANs. The proposed sensor fault detection (SFD) algorithm is based on Pearson correlation coefficients and simple statistical methods. The proposed method identifies strongly correlated parameters using Pearson correlation coefficients, and the proposed SFD algorithm detects faulty sensors. We validated the proposed SFD algorithm using two datasets from the Multiparameter Intelligent Monitoring in Intensive Care database and compared the results to those of existing methods. The time complexity of the proposed algorithm was also compared to that of existing methods. The proposed algorithm achieved high detection rates and low false alarm rates with accuracies of 97.23% and 93.99% for Dataset 1 and Dataset 2, respectively.     

Polyaniline-based networks combined with Fe3O4 hollow spheres and carbon balls for excellent electromagnetic wave absorption

Polyaniline (PANI)-based networks combined with Fe₃O₄ hollow spheres and carbon balls (FCP) for improved electromagnetic wave (EMW) absorption were investigated using an easy-to-industrialize solvothermal and physical method. Hollow structure Fe₃O₄ spheres with a lower density than that of the common solid sphere were prepared. As a thin and light magnetic material, Fe₃O₄ hollow spheres generate magnetic loss, carbon balls and PANI networks generate dielectric loss. The magnetic and conductive parts play appropriate roles in achieving complementarity in the EMW absorption. The relatively high specific surface area introduced by PANI networks promotes interfacial polarization and further supports dielectric loss. In conclusion, the above reasons provide multiple attenuation mechanisms. Samples FCP1 (?65.109 dB, at 12.800 GHz, 1.966 mm, from 5.6 to 18.0 GHz) and FCP2 (?61.033 dB, at 8.480 GHz, 3.328 mm, from 4.3 to 18.0 GHz) demonstrated a wide bandwidth, a small thickness, a minimum reflection loss (R_L), and a low loading ratio (25%) in paraffin-based composites. Specifically, their loading ration of 25% is much lower than the loading ratio of conventional materials (usually 50% and above). In addition, the bandwidth is excessively wide, above 12 GHz, possessing good absorption performance in continuous intervals with different thicknesses. Such excellent characteristics have rarely been reported in literature.