采用自适应字典学习的InSAR降噪方法

We consider the phase noise filtering problem for interferometric synthetic aperture radar (InSAR) based on the dictionary learning technique. Due to the non-convexity of the optimization problem is difficult to solve. By using the splitting technique and employing the augmented Lagrangian framework, we obtain a relaxed nonlinear constraint optimization problem with l₁-norm regularization which can be solved efficiently by the alternating direction method of multipliers (ADMM). Specifically, we firstly train dictionaries from the InSAR complex phase data, and then reconstruct the desired complex phase image from the sparse representation. Simulation results based on simulated and measured data show that this new InSAR phase noise reduction method has a much better performance than several classical phase filtering methods in terms of residual count, mean square error (MSE) and preservation of the fringe completeness.     

群策式智能多联机末端在大学教学楼的设计应用与分析

建筑能耗占全国能耗比重较大,中央空调在公共建筑中使用广泛,滥用无管理使得公共建筑夏季空调能耗很高。本文针对大学中多联机中央空调系统以末端智能调控系统控制策略方向为目的进行基于群策控制智能末端控制改造,达到节能的目的。通过单片机对空调控制面板进行电信号控制,数据判断及传输通过总服务器进行计算。控制方式采用自动控制与群策相结合,由微信公众号进行查询及发送命令,通过网络传输和数据库相连。数据库录入空调系统使用教室信息、工作及学习时段表等相关数据。从调研、设计、模拟来看,节能效果显著,最理想节能量可达84%,在调研过程中大部分学生对此可接受。     

A customised approach to determining the geomorphic effectiveness of small flood events in a regulated river

The construction of dams significantly alters flow and sediment regimes with subsequent deleterious effects on the morphological and ecological character of rivers. Effective experimental floods can ameliorate the downstream geomorphic impacts of dams. The traditional view is that large floods are required to perform effective geomorphic work, and the geomorphic outcomes of small floods are often overlooked. Many river restoration frameworks do not consider small floods. Yet, there is evidence that the hydrological characteristics that ameliorate specific geomorphic impacts in a river are unique to each river, and a customised approach to setting the right mix of floods (including small experimental floods) is needed. In this study, we modify an existing flood effectiveness model developed for large floods, for determining the geomorphic effectiveness of small floods in a highly regulated Australian river. Two flood classes were added to the model (medium peak stream power and moderate total energy expenditure), and the flood power characteristics were rescaled to reflect the relative difference in the magnitude of the small floods and the magnitude of the geomorphic work performed. Using a step‐wise approach, this customised model determined the geomorphic effectiveness of small floods. The best flood for ameliorating the geomorphic impacts of flow regulation had medium to long duration (10 to 51 days), high peak unit stream power (77 to 123 Wm^?2) and moderate to large total energy expenditure (78,600 to 342,320 × 10³ J). This approach to determining flood effectiveness for small floods is applicable to other geomorphically impacted river channels downstream of dams and can be used to inform experimental flood releases for geomorphic outcomes.     

基于GAED-MADDPG多智能体强化学习的协作策略研究

目前多智能体强化学习算法多采用集中学习,分散行动的框架。该框架存在算法收敛时间过长和可能无法收敛的问题。为了加快多智能体的集体学习时间,提出多智能体分组学习策略。通过使用循环神经网络预测出多智能体的分组矩阵,通过在分组内部共享智能体之间经验的机制,提高了多智能体的团队学习效率;同时,为了弥补分组带来的智能体无法共享信息的问题,提出了信息微量的概念在所有智能体之间传递部分全局信息;为了加强分组内部优秀经验的留存,提出了推迟组内优秀智能体死亡时间的生灭过程。最后,在迷宫实验中,训练时间比MADDPG减少12%;夺旗实验中,训练时间比MADDPG减少17%。     

基于多层注意力机制—柔性AC算法的机器人路径规划

针对行动者—评论家（AC）算法存在的经验学习样本维度高、策略梯度模型鲁棒性低等问题,依据多代理系统的信息协作优势,构建注意力机制网络并作为代理体,引入多层并行注意力机制网络模型对AC算法进行改进,提出一种基于多层并行注意力机制的柔性AC算法。将其用于解决动态未知环境下的机器人路径规划问题,可增强行动者的策略梯度鲁棒性并降低评论家的回归误差,实现机器人路径规划最优方案的快速收敛。实验结果表明,该算法有效克服机器人路径规划的局部最优,具有计算速度快、稳定收敛的优点。     

融合特征关联性的深度哈希图像表示方法

针对深度描述子无法提供图像特征之间关联性的问题进行了研究,提出了一种融合特征关联性的深度哈希图像表示方法,这种方法将深度描述子之间的关系融入到图像内容的描述中,用于提高图像检索性能。首先,通过预训练网络生成图像的特征映射,并在此基础上提取出深度特征描述子。然后,将深度特征描述子映射为深度视觉词,从而用于深度视觉词的频繁项集发现。接下来将离散值的深度视觉词图像表示和哈希值的频繁项集图像表示连接生成图像表示。最后,算法通过图像类内、类间的相似性关系构造优化,得到最优的阈值,用于将图像表示变为哈希值。实验中,将提出的方法与一些优秀的图像表示方法在holiday、Oxford和Paris图像集的图像检索任务中进行了性能比对,用于证明此方法的有效性。     

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。     

基于ResNeXt和WGAN网络的单图像超分辨率重建

为解决现有基于生成对抗网络的单图像超分辨率重建模型SRGAN网络训练不稳定、学习速率慢等问题,提出了一种基于ResNeXt和WGAN的单图像超分辨率重建模型Res_SRGAN。该模型参考ResNeXt网络结构构造生成器,降低了生成器的复杂度,仅为SRGAN的1/8;通过WGAN来构造判别器解决了SRGAN模型不稳定的问题;实验结果表明,在四个公开数据集上本文所提模型相较于现有的单图像超分辨率重建模型在主客观评价中均取得了更加优越的性能。     

基于卷积神经网络和语义相关的协同显著性检测

针对目前协同显著性检测方法中存在的语义特征类相差悬殊的物体被误检测为协同对象等问题,提出了一种基于卷积神经网络和语义相关的协同显著性检测算法CSCCD。首先,采用引导超像素滤波方法对SLIC分割出的超像素区域和DSS生成的显著性区域进行处理,清晰地显示了目标边界轮廓;然后使用Mask R-CNN提取语义特征,给出了图像语义特征和语义一致性的定义,并针对提取语义特征过程中出现的同一语义类别的物体在不同形态下被检测为不同语义类别的问题,提出了图像组语义相关类的概念,在此概念的基础上定义了图像组语义关联类,解决了多幅图像的语义关联问题;最后融合显著性检测区域和图像组语义一致性区域得到协同显著性检测结果。在公开基准数据集上的实验结果表明,该算法能够有效凸显目标整体及轮廓,在客观量化方面的综合性能有明显提升。     

Theoretical analysis of a new segmented anchoring style in weakly cemented soft surrounding rock

According to the tensile failure of rock bolt in weakly cemented soft rock, this paper presents a new segmented anchoring style in order to weaken the cumulative effect of anchoring force associated with the large deformation. Firstly, a segmented mechanical model was established in which free and anchoring section of rock bolt were respectively arranged in different deformation zones. Then, stress and displacement in elastic non-anchoring zone, elastic anchoring zone, elastic sticking zone, softening sticking zone and broken zone were derived respectively based on neural theory and tri-linear strain softening constitutive model of soft rock. Results show that the anchoring effect can be characterized by a supporting parameter b. With its increase, the peak value of tangential stress gradually moves to the roadway wall, and the radial stress significantly increases, which means the decrease of equivalent plastic zone and improvement of confining effect provided by anchorage body. When b increases to 0.72, the equivalent plastic zone disappears, and stresses tend to be the elastic solutions. In addition, the anchoring effect on the displacement of surrounding rock can be quantified by a normalization factor δ.