A simplification method for capacitance models in AlGaN/GaN high electron mobility transistors under large drain voltage using channel analysis

This article analyzes the bias dependence of gate‐drain capacitance (C_gd) and gate‐source capacitance (C_gs) in the AlGaN/GaN high electron mobility transistors under a high drain‐to‐source voltage (V_ds) from the perspective of channel shape variation, and further simplifies C_gd and C_gs to be gate‐to‐source voltage (V_gs) dependent only at high V_ds. This method can significantly reduce the number of parameters to be fitted in C_gd and C_gs and therefore lower the difficulty of model development. The Angelov capacitance models are chosen for verifying the effectiveness of simplification. Good agreement between simulated and measured small‐signal S‐parameters, large‐signal power sweep, and power contours comprehensively proves the accuracy of this simplification method.     

Hyperspectral image super-resolution through clustering-based sparse representation

Promoting the spatial resolution of hyperspectral sensors is expected to improve computer vision tasks. However, due to the physical limitations of imaging sensors, the hyperspectral image is often of low spatial resolution. In this paper, we propose a new hyperspectral image super-resolution method from a low-resolution (LR) hyperspectral image and a high resolution (HR) multispectral image of the same scene. The reconstruction of HR hyperspectral image is formulated as a joint estimation of the hyperspectral dictionary and the sparse codes based on the spatial-spectral sparsity of the hyperspectral image. The hyperspectral dictionary is learned from the LR hyperspectral image. The sparse codes with respect to the learned dictionary are estimated from LR hyperspectral image and the corresponding HR multispectral image. To improve the accuracy, both spectral dictionary learning and sparse coefficients estimation exploit the spatial correlation of the HR hyperspectral image. Experiments show that the proposed method outperforms several state-of-art hyperspectral image super-resolution methods in objective quality metrics and visual performance.

     

基于空间注意力下边缘图融合的草图图像检索

基于草图的图像检索任务根据用户提供的手绘草图,从图像数据库中检索得到与该草图对应的自然图像.与传统基于内容的图像检索不同,草图和自然图像间存在明显的域差异,这使得二者的特征难以直接进行比较.针对自然图像边缘图和草图的相似性,提出了空间注意力下的边缘图融合模型,将自然图像和对应的边缘图分别编码到各自的特征空间,再通过空间注意力掩膜进行加权融合,进而用于草图图像检索.所提模型可以更有效地编码物体边缘轮廓的特征,分别在Sketchy和Flickr15K数据集的草图图像检索任务上取得了比前人方法更高的Recall@1和MeanAP指标.     

基于有源定标器的HY-2B卫星雷达高度计时钟频率偏差在轨测试

雷达高度计的主要功能是测量全球平均海平面高度（SSH：Sea Surface Height）,它通过准确测量收发脉冲之间的时间间隔来实现精准的星地距离测量,时间测量的准确度取决于高度计时钟的准确性。卫星在轨运行期间,雷达高度计时钟频率会产生缓慢漂移,时钟频率漂移会影响星地距离的测量准确度,造成对海面高程的测量偏差。在基于重建型有源定标器对HY-2A卫星高度计时钟偏差在轨测试方法的基础上,提出了从星地斜距函数曲线时间偏移量中提取高度计时钟偏差的估计方法,应用于HY-2B卫星雷达高度计在轨定标测试中,测量了雷达高度计时钟频率漂移量,精度优于0.001 Hz。结果表明：HY-2B高度计原子钟性能稳定,时钟频率偏差造成的测距偏差为毫米量级,平均测距漂移率为2.95×10^-7 m/d。     

基于深度强化学习的舰载机动态避障方法

针对高度异构、动态的航母甲板作业场景中的舰载机避障问题,提出一种结合预测算法和深度强化学习的避障方法.该方法包含场景建模、奖励模型和轨迹预测模型等模块.首先基于智能体状态和动作空间对航母甲板场景进行建模;然后利用最小二乘法对场景中动态障碍物的位置进行实时轨迹预测,并构造了包含路径预测模块的深度强化学习方法——环境预测深度Q网络(PDQN);最后利用该方法实现航母甲板作业场景中的舰载机动态避障.利用Python绘图集Matplotlib进行仿真实验,实验数据结果表明,相比于Q-learning,SARSA等方法,所提方法的准确率提升了15％～25％,路径长度短9％～39％,平均奖励值高30％～100％,收敛速度快1～2倍且训练平稳后准确率的标准差小2％～50％.     

Hermite-Shannon-Cosine区间小波及其在图形自适应分划插值中的应用

动态曲线图是描述客观现象变换规律的常用工具,为实现动态曲线特征点的动态捕捉及精确表达,构造了同时具有插值性、光滑性、紧支撑性和对称性的Shannon-Cosine小波函数.首先利用Shannon小波函数的波动性和连续性,根据积分中值定理,设计了一种参数化的窗函数,通过参数调整,可满足Shannon-Cosine小波对支撑区间和光滑度的自适应控制的要求;其次,分析确定了对曲线进行小波变换时边界效应归因于曲线边界的不连续,因此,采用2点3次Hermite插值函数构建了区间小波;最后,采用多尺度Shannon-Cosine小波对冲击波传播曲线和反射Burger曲线进行多尺度自适应细分和逼近,自动捕捉曲线的特征点进行重构.实例结果表明,与其他方法相比,Hermite Shannon-Cosine区间小波逼近曲线具有较高的数值精度和较低的算法复杂度.     

改进U-Net型网络的遥感图像道路提取

目的 遥感图像道路提取在城市规划、交通管理、车辆导航和地图更新等领域中发挥了重要作用,但遥感图像受光照、噪声和遮挡等因素以及识别过程中大量相似的非道路目标干扰,导致提取高质量的遥感图像道路有很大难度。为此,提出一种结合上下文信息和注意力机制的U-Net型道路分割网络。方法 使用Resnet-34预训练网络作为编码器实现特征提取,通过上下文信息提取模块对图像的上下文信息进行整合,确保对道路的几何拓扑结构特征的提取;使用注意力机制对跳跃连接传递的特征进行权重调整,提升网络对于道路边缘区域的分割效果。结果 在公共数据集Deep Globe道路提取数据集上对模型进行测试,召回率和交并比指标分别达到0.847 2和0.691 5。与主流方法U-Net和CE-Net（context encoder network）等进行比较,实验结果表明本文方法在性能上表现良好,能有效提高道路分割的精确度。结论 本文针对遥感图像道路提取中道路结构不完整和道路边缘区域不清晰问题,提出一种结合上下文信息和注意力机制的遥感道路提取模型。实验结果表明该网络在遥感图像道路提取上达到良好效果,具有较高的研究和应用价值。     

自适应多模态特征融合胶质瘤分级网络

目的 胶质瘤的准确分级是辅助制定个性化治疗方案的主要手段,但现有研究大多数集中在基于肿瘤区域的分级预测上,需要事先勾画感兴趣区域,无法满足临床智能辅助诊断的实时性需求。因此,本文提出一种自适应多模态特征融合网络（adaptive multi-modal fusion net,AMMFNet）,在不需要勾画肿瘤区域的情况下,实现原始采集图像到胶质瘤级别的端到端准确预测。方法 AMMFNet方法采用4个同构异义网络分支提取不同模态的多尺度图像特征;利用自适应多模态特征融合模块和降维模块进行特征融合;结合交叉熵分类损失和特征嵌入损失提高胶质瘤的分类精度。为了验证模型性能,本文采用MICCAI （Medical Image Computing and Computer Assisted Intervention Society）2018公开数据集进行训练和测试,与前沿深度学习模型和最新的胶质瘤分类模型进行对比,并采用精度以及受试者曲线下面积（area under curve,AUC）等指标进行定量分析。结果 在无需勾画肿瘤区域的情况下,本文模型预测胶质瘤分级的AUC为0.965;在使用肿瘤区域时,其AUC高达0.997,精度为0.982,比目前最好的胶质瘤分类模型——多任务卷积神经网络同比提高1.2%。结论 本文提出的自适应多模态特征融合网络,通过结合多模态、多语义级别特征,可以在未勾画肿瘤区域的前提下,准确地实现胶质瘤分级预测。     

A multi-view attention-based deep learning system for online deviant content detection

With the exponential growth of user-generated content, policies and guidelines are not always enforced in social media, resulting in the prevalence of deviant content violating policies and guidelines. The adverse effects of deviant content are devastating and far-reaching. However, the detection of deviant content from sparse and imbalanced textual data is challenging, as a large number of stakeholders are involved with different stands and the subtle linguistic cues are highly dependent on complex context. To address this problem, we propose a multi-view attention-based deep learning system, which combines random subspace and binary particle swarm optimization (RS-BPSO) to distill content of interest (candidates) from imbalanced data, and applies the context and view attention mechanisms in convolutional neural network (dubbed as SSCNN) for the extraction of structural and semantic features. We evaluate the proposed approach on a large-scale dataset collected from Facebook, and find that RS-BPSO is able to detect whether the content is associated with marijuana with an accuracy of 87.55%, and SSCNN outperforms baselines with an accuracy of 94.50%.

     

Guaranteeing performance for uncertain nonlinear systems with bounded state constraint and mismatching condition

The control design problem for the uncertain nonlinear system with bounded state constraint and mismatching condition is considered in this paper. The uncertainty in the system, which may be due to unknown system parameters and external disturbance, is nonlinear and time‐varying. The state of the system is constrained to be bounded. The system does not satisfy the (global) matching condition. A creative one‐to‐one state transformation is proposed by converting the bounded states into the unbounded ones. A step‐by‐step state transformation is proposed to convert the mismatched system into a matched system. The robust control is then proposed based on the transformed system. The control is demonstrated to be able to guarantee the uniform boundedness and uniform ultimate boundedness of the system in the presence of uncertainty, while the state constraint can be always guaranteed.