Biologically adaptive robust mean shift algorithm with Cauchy predator-prey BBO and space variant resolution for unmanned helicopter formation

Visual tracking technology can provide measurement information for unmanned helicopter formation and thus, more attention is being paid to this research area. We propose a novel mean shift （MS） algorithm that is both adaptive and robust for unmanned helicopter formation and apply it to the leading unmanned helicopter tracking. The movement of an unmanned helicopter is very flexible and changeable, which makes the tracking there of more difficulty than for common targets. In creating an algorithm that can adapt to the acceleration of the unmanned helicopter and estimates both the scale and orientation of the movement changes, we combine the traditional MS with the bio-inspired Cauchy predator-prey biogeography-based optimization （CPPBBO） evolutionary algorithm, and also the space variant resolution （SVR） mechanism of the human visual system （MS-CPPBBO-SVR）. To demonstrate the effectiveness and robustness of the proposed method and justify the importance of the CPPBBO algorithm and SVR mechanism at the same time, a series of comparative exper- iments were carried out. The experimental results of the proposed MS-CPPBBO-SVR method are compared with other competitive tracking methods, such as MS, MS with SVR （MS-SVR）, MS-SVR with several other optimization algorithms, and the robust particle filter algorithm. The experimental results demonstrate that our proposed tracking approach, MS-CPPBBO-SVR, is more adaptive, robust and efficient in target tracking than the other methods.     

民航领域突发事件的实体链接方法

实体链接的相关研究主要集中于医疗、生物和新闻领域,但在民航领域的研究较少。因此针对民航领域实体链接任务进行了研究,发现在实体链接中存在实体变体多、歧义少等问题。为解决上述问题,提出了一种基于语义推断的实体链接框架以及一种用于增强框架鲁棒性的负采样策略。在民航领域数据集上进行了对比实验,结果表明所提框架链接效果优于现有基准框架,并通过消融实验,验证了负采样策略的有效性。在负采样策略的作用下,该实体链接框架的Acc@top1高达0.875。     

基于深度网络的出租车Pick-up需求预测

在智慧城市建设中,区域的出租车需求预测是一个十分重要的模块。为了预测指定区域未来时刻的出租车需求,本文通过拓展已有序列模型,提出一个多时间分辨率的基于层次注意力机制的循环高速网络（Multi-Time Resolution Hierarchical Attention-Based Recurrent Highway Networks, MTR-HRHN）。MTR-HRHN将对外生数据时空特征的提取和目标数据的时空建模整合到单个框架中,并且通过多分辨率（例如每个小时或者每天）对序列数据不同的时间属性建模,从而捕获更全面的时间模式。最后,在纽约市出租车数据集上评估MTR-HRHN的预测性能。实验结果表明,与其他经典时间序列预测方法相比,MTR-HRHN在多个高需求区域的短期需求预测上表现出更好的预测性能。     

基于机器学习合成高分辨率布料褶皱

为了精化布料网格得到逼真的布料模拟效果,提出了一种基于机器学习的方法合成高分辨率布料褶皱。首先模拟真实的布料运动,获取布料运动的帧数据信息,将布料运动的帧数据信息转换为图像信息的形式进行存储。然后将图像信息作为输入,输入到卷积神经网络(CNN)中进行训练,通过将卷积神经网络与缩小网络相结合,最终得到高分辨率布料图像。最后将高分辨率布料图像转换为高分辨率布料网格,对布料进行模拟。实验结果表明,与初始的低分辨率布料网格对比,合成的高分辨率布料网格模拟出的布料有着大量且细微的褶皱,并且能够模拟出真实的布料效果,与真实场景中的布料模拟效果相似。该方法在不同的场景中都可以模拟出高质量的布料动画效果,而且减少了仿真速度,验证了该方法的有效性。     

基于二次虚拟扩展的高分辨率波达方向估计方法

为进一步提高天线阵波达方向估计的分辨率,在四阶量多重信号分类方法的基础上,提出一种高分辨率的波达方向估计方法.利用阵列接收数据的四阶矩量进行虚拟阵列扩展,再利用阵列接收数据的共轭进行虚拟阵列扩展,实现二次虚拟扩展;将扩展后的阵列导向矢量和协方差矩阵用于波达方向估计,与原阵列导向矢量和协方差矩阵相比,相当于构造了更多的虚拟阵元,并扩展了阵列的孔径.仿真结果表明:与四阶量多重信号分类等波达方向估计方法相比,所提出的方法在波达方向估计中成功概率更高,均方误差更低,具有更高的分辨率.所提方法通过二次虚拟扩展,构造了更多的虚拟阵元,有效地提高了天线阵波达方向估计的分辨率.     

High precision frequency measurement based on the characteristic of the fuzzy area

The phase difference change is random in the discrete fuzzy area by considering the limit of device resolution, which the device can not stably capture, and the precision of the measured value only reaches the level of ns, but the phase difference change in the concentrated fuzzy area is step by step, and the device can stably capture the change of ps. By using the fuzzy area conversion technology, a high precision frequency measurement is designed. Through computer control and frequency synthesis,the discrete fuzzy area is converted to the concentrated fuzzy area, the ps stability properties of phase difference change on the border of the concentrated fuzzy area is used, the precision can be improved two or three orders of magnitude, and the high precision measurement in a wide range is achieved.     

Single Radar Measurement Method for Antiaircraft Projectile Miss Distance

To measure miss distance for antiaircraft projectile,a radial velocity identification and positioning method with a single radar is proposed.By analyzing the spatial resolution of multi-frequency ranging radar,the discrimination and testing model of this radar for multi-targets (projectile and target) is established to analyze the systematic error of antiaircraft miss distance.Then through the aerial target flight test and contrast test with optical test equipment,the validity of the measurement method is verified.This new method has the potential to be used in the measurement of antiaircraft projectile miss distance.     

基于WRF的城市热岛效应高分辨率评估方法

为解决城市总规尺度气候评估工具缺乏的问题,分析了不同气候评估方法的特点和应用范围,研究了气象学工具WRF在城市热岛效应评估中的应用方法及有效性。建立了针对城市总规尺度的高精度城市规划信息数据库、气象数据库生成WRF模拟边界的方法;明确了高精度城市热岛效应模拟所需的WRF物理方案。以沿海城市大连夏季典型晴朗高温天气为例,利用WRF模型进行了300 m的高分辨率模拟。结果表明,WRF可以较好地模拟出城市热岛效应的强度和时空变化;将模拟结果与气象站和现场观测值进行了对比,两者变化趋势一致,偏差在可接受的范围之内。     

学习小波超分辨率系数的人脸超分算法

目前基于卷积神经网络的超分方法虽然在峰值信噪比和结构相似性评价指标上能取得优异的结果,但是得到的超分图像视觉质量较差,会丢失人脸五官区域的细节信息。针对这一现象,设计了一种新的深度神经网络来预测超分小波系数以获得信息丰富的超分辨率人脸图像,首先利用人脸图像的先验知识手动地给予五官区域更多的关注,然后在网络中引入线性低秩卷积运算,最后利用长距离依赖的思想补充超分图像的细节。实验验证该算法可以在获得较高的峰值信噪比和结构相似性的同时,使超分人脸图像五官区域更加清晰、视觉质量更优。     

地球同步轨道SAR曲线轨迹二维空间分辨率分析

针对地球同步轨道合成孔径雷达(GEO SAR)中,卫星轨道是曲线轨迹,且由于地球自转等因素的影响,卫星相对于点目标的运动轨迹没有显式表达式,且航迹速度在不同的卫星轨道位置差异很大的问题,提出了一种基于相位梯度分析的高轨SAR二维空间分辨率计算方法.文中提出了一种猜想,并采用介值定理证明:在合成孔径时间内,存在1个等效航迹速度,使得理论方位向分辨率位于1个范围之内,并计算得到了该方位向分辨率范围.最后,后向投影算法仿真实验验证了该方法的有效性.