Micro-expression recognition from local facial regions

MiE is a facial involuntary reaction that reflects the real emotion and thoughts of a human being. It is very difficult for a normal human to detect a Micro-Expression (MiE), since it is a very fast and local face reaction with low intensity. As a consequence, it is a challenging task for researchers to build an automatic system for MiE recognition. Previous works for MiE recognition have attempted to use the whole face, yet a facial MiE appears in a small region of the face, which makes the extraction of relevant features a hard task. In this paper, we propose a novel deep learning approach that leverages the locality aspect of MiEs by learning spatio-temporal features from local facial regions using a composite architecture of Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM). The proposed solution succeeds to extract relevant local features for MiEs recognition. Experimental results on benchmark datasets demonstrate the highest recognition accuracy of our solution with respect to state-of-the-art methods.     

四旋翼长续航飞行模式的设计与实现

更长的飞行时间是四旋翼无人机领域研究热点方向之一;在对实际飞行中瞬时消耗电流和电池电压数据的研究中发现,过大姿态角下电池电量消耗显著提升;为了延长飞行时间和提升电池电量使用效率,提出一种长续航飞行模式;在该模式下,基于现有的角速度串级PID姿态控制器,将飞行加速度的控制算法改为飞行速度控制,限制过大姿态角的操作;在无风、微风和强风环境下的飞行实验表明,长续航飞行模式比传统飞行方式飞行时间增加8%~20%;长续航飞行模式可广泛应用于多种无需快速变换飞行路径,但需要更长飞行时间的的应用场景中。     

电气化铁路既有线深路堑扩堑石方控制爆破

介绍电气化铁路既有线无隔墙台阶爆破扩堑方法。沿既有线方向设置低台阶 ,边界处布置光爆孔 ,采用“同列同段和列间微差”的起爆网路。采用自制的“炮被”和架设“钢管排架”阻挡飞石、滚石和滑石 ,确保了既有线的安全。文中还概述了“炮被”和“钢管排架”的制做以及作者的认识和体会。     

Bayesian Inference for ARFIMA Models

This article develops practical methods for Bayesian inference in the autoregressive fractionally integrated moving average (ARFIMA) model using the exact likelihood function, any proper prior distribution, and time series that may have thousands of observations. These methods utilize sequentially adaptive Bayesian learning, a sequential Monte Carlo algorithm that can exploit massively parallel desktop computing with graphics processing units (GPUs). The article identifies and solves several problems in the computation of the likelihood function that apparently have not been addressed in the literature. Four applications illustrate the utility of the approach. The most ambitious is an ARFIMA(2,d,2) model for the Campito tree ring time series (length 5405), for which the methods developed in the article provide an essentially uncorrelated sample of size 16,384 from the exact posterior distribution in under four hours. Less ambitious applications take as little as 4 minutes without exploiting GPUs.     

Microstructure and Mechanical Properties of Mg–RE–TM Cast Alloys Containing Long Period Stacking Ordered Phases: A Review

Casting magnesium alloys hold the greatest share of magnesium application products due to their short processing period, low cost and near net shape forming. Compared with conventional commercial magnesium alloys or other Mg–RE-based alloys, the novel Mg–RE–TM cast alloys with long period stacking ordered(LPSO) phases usually possess a higher strength and are promising candidates for aluminum alloy applications. Up to now, two ways: alloying design and casting process control(including subsequent heat treatments), have been predominantly employed to further improve the mechanical properties of these alloys. Alloying with other elements or ceramic particles could alter the solidifi cation pattern of alloys, change the morphology of LPSO phases and refi ne the microstructures. Diff erent casting techniques(conventional casting, rapidly solidifi cation, directional solidifi cation, etc.) introduce various microstructure characteristics, such as dendritic structure, nanocrystalline, metastable phase, anisotropy. Further heat treatments could activate the transformation of various LPSO structures and precipitation of diverse precipitates. All these evolutions exert great impacts on the mechanical properties of the LPSO-containing alloys. However, the underlying mechanisms still remain a subject of debate. Therefore, this review mainly provides the state of the art of the casting magnesium alloys research and the accompanying challenges and summarizes some topics that merit future investigation for developing high-performance Mg–RE–TM cast alloys.     

Evolution mechanism of active groups and thermal effects of Chinese lignite in low-temperature oxidation

Abstract

The evolution of active groups at low temperature was examined using Chinese lignite by infrared technology and X-ray photoelectron spectroscopy (XPS). The results showed that the hydroxyl, aliphatic ether, methylene, and methyl groups played important roles in the low-temperature oxidation of lignite below 200?°C. Carbonyl and carboxyl groups were important intermediates. Thus, a multi-step evolution mechanism involving the hydroxyl, aliphatic ether groups, and alkane was reasoned to describe the low-temperature oxidation of lignite. In addition, according to the oxidation kinetics experiment and the evolution laws of the active groups, the ratios of the reaction lines were determined considering the accuracy of thermal effects. The thermal effects and the heat release intensities of each temperature interval were obtained based on the evolution mechanism and the reaction ratios. The shortest spontaneous combustion period of lignite was calculated and compared with the experimental value, which proved that the reasoned evolution mechanism of the active groups and the calculations of the thermal effects were reliable.     

Operational time-series data modeling via LSTM network integrating principal component analysis based on human experience

Today’s information technologies involve increasingly intelligent systems, which come at the cost of increasingly complex equipment. Modern monitoring systems collect multi-measuring-point and long-term data which make equipment health prediction a “big data” problem. It is difficult to extract information from such condition monitoring data to accurately estimate or predict health statuses. Deep learning is a powerful tool for big data processing that is widely utilized in image and speech recognition applications, and can also provide effective predictions in industrial processes. This paper proposes the Long Short-term Memory Integrating Principal Component Analysis based on Human Experience (HEPCA-LSTM), which uses operational time-series data for equipment health prognostics. Principal component analysis based on human experience is first conducted to extract condition parameters from the condition monitoring system. The long short-term memory (LSTM) framework is then constructed to predict the target status. Finally, a dynamic update of the prediction model with incoming data is performed at a certain interval to prevent any model misalignment caused by the drifting of relevant variables. The proposed model is validated on a practical case and found to outperform other prediction methods. It utilizes a powerful deep learning analysis method, the LSTM, to fully process big condition monitoring series data; it effectively extracts the features involved with human experience and takes dynamic updates into consideration.     

LTE系统多小区上行联合功控算法

针对传统的小区内开环功率控制算法通常以提升本小区的吞吐量性能为目标，忽略了当前小区用户对邻小区用户同频干扰的问题，为提升边缘用户性能的同时兼顾系统整体性能，提出了一种LTE系统小区间上行联合功率控制（UJPC）算法。该算法采用单基站三扇区为系统模型，以最优化系统吞吐量比例公平函数为目标，首先根据最小信干噪比（SINR）约束值和用户最大发射功率这两个约束条件得到相应的数学优化模型，然后采用连续凸近似的方法求解优化问题得出各个基站所管辖的小区内所有用户的最优发射功率。仿真结果表明，与基准的开环功控方案相比，联合功控方案在保证系统平均频谱利用率的情况下能够较大幅度地提高小区边缘频谱利用率，其最佳性能增益能达到50%。     

基于空间合作关系的基站流量预测模型

针对传统的自回归积分移动平均（ARIMA）模型和长短时记忆（LSTM）单元在基站流量预测中没有利用基站（BS）间合作关系的问题，提出一种利用由用户群体在不同基站下访问产生的基站合作关系的流量预测（TPBC）算法。首先，通过基站之间的合作关系构建基站合作网络，并对此合作网络进行社区划分得到基站社区；然后，通过格兰杰因果关系检验方法寻找与目标基站同一社区且关系最紧密的若干基站，作为目标基站的合作基站；最后，使用LSTM和词嵌入层（Embedding）搭建混合神经网络，并根据目标基站和合作基站的流量信息进行流量预测。实验结果表明，TPBC在基站流量预测上的均方根误差（RMSE）相比ARIMA和LSTM分别减小了29.19%和27.47%。TPBC能有效提高基站流量预测准确率，在流量卸载和绿色节能等领域具有重要意义。