基于生成对抗网络的半监督图像语义分割

提出了一种基于生成对抗网络的语义分割模型,包括一个全卷积语义分割网络以及一个判别网络,其中语义分割网络负责生成与输入图像对应的语义分割图,判别网络负责检测分割图与真实标签的区别,以促使分割网络改进分割效果。为了更好的提取全局结构信息,语义分割网络中采用了金字塔池化模块,对不同规模的空间区域进行池化操作。另外,为了应对语义分割训练数据集人工标注成本过高的问题,利用判别网络生成伪标签协助语义分割网络进行训练,从而实现了半监督训练效果。模型在PASCAL VOC2012数据集中进行了测试,结果表明该模型在全监督和半监督条件下均优于已有方法。     

模具结构对连续纤维增强热塑性复合材料性能的影响

采用自行设计的两种不同结构的熔融浸渍模具制备了连续玻璃纤维增强聚丙烯预浸带，测试了模具结构对预浸带的孔隙率、纤维断裂率、界面形貌、纤维分散均匀度和拉伸强度的影响，建立了纤维浸渍模型和纤维断裂模型，并通过理论模型对预浸带的孔隙率和断裂率进行理论预测。结果表明，本文建立的数学模型能够有效预测预浸带的浸渍程度和纤维断裂率，可用于浸渍模具结构的优化设计；在本文范围内，与波浪形模具相比，斜齿形模具的多楔形区结构可以有效地降低预浸带孔隙率和提升纤维分散程度；波浪形模具的流道圆角半径较大，楔形区个数较少，与斜齿形模具相比，可有效降低纤维断裂率并提升拉伸性能。     

次酉极因子的扰动界

矩阵扰动问题不仅对矩阵论,而且对控制论、力学、线性系统以及工程都有着重要的意义.主要利用矩阵特征值与奇异值的性质,对广义极分解中次酉极因子的扰动界进行研究,得到F范数下新的扰动界,并利用最新的换子不等式,对李仁仓的研究结论重新证明,该证明更加简短有趣.     

知识辅助的机载MIMO雷达STAP非均匀样本检测方法

When the covariance matrix is estimated with training samples contaminated by target like signals, the performance of target detection in multiple input multiple output (MIMO) radar space time adaptive processing (STAP) decreases. Aiming at this deficiency, a knowledge aided (KA) generalized inner product (GIP) method for non homogeneous samples detection is proposed. Firstly the clutter subspace knowledge estimated by prolate spheroidal wave functions is utilized to construct the clutter covariance matrix offline. Then the GIP non homogeneity detector (GIP NHD) is integrated to realize the effective selection of training samples, which eliminates the effect of the target like signals in training samples on target detection. The simulation results show that compared with the conventional GIP method, the KA GIP method can screen out contaminated training samples more effectively and the target detection performance of MIMO radar STAP can be improved significantly. 〖JP2〗Thus the proposed KA GIP method is more valuable for practical engineering application.     

迹为整数的3×3阶正交矩阵的谱

讨论了3×3阶正交矩阵的特征值和迹的关系,证明了迹为整数的3×3阶正交矩阵的谱可由迹确定,为应用广泛的3×3阶正交矩阵的谱的计算提供了简单实用的方法.     

Hierarchical Community Detection Based on Partial Matrix Convergence Using Random Walks

Random walks are a standard tool for modeling the spreading process in social and biological systems.But in the face of large-scale networks,to achieve convergence,iterative calculation of the transition matrix in random walk methods consumes a lot of time.In this paper,we propose a three-stage hierarchical community detection algorithm based on Partial Matrix Approximation Convergence(PMAC) using random walks.First,this algorithm identifies the initial core nodes in a network by classical measurement and then utilizes the error function of the partial transition matrix convergence of the core nodes to determine the number of random walks steps.As such,the PMAC of the core nodes replaces the final convergence of all the nodes in the whole matrix.Finally,based on the approximation convergence transition matrix,we cluster the communities around core nodes and use a closeness index to merge two communities.By recursively repeating the process,a dendrogram of the communities is eventually constructed.We validated the performance of the PMAC by comparing its results with those of two representative methods for three real-world networks with different scales.     

基于热中子探测评价裂缝的示踪剂特性研究

针对裂缝热中子示踪探测方法,从中子双组扩散理论入手结合蒙特卡洛数值模拟方法,分析压裂前后地层热中子分布影响因素,并模拟不同类型示踪剂及含量的热中子裂缝响应规律,筛选最佳示踪剂并给出其在支撑剂中含量的上、下限。结果表明:压裂前后热中子计数变化量主要受裂缝宽度和标记支撑剂中示踪剂含量的影响;钆对裂缝宽度变化反应最灵敏,钆元素是作为标记支撑剂的最佳示踪元素;随支撑剂中Gd 2O 3含量的增加,热中子计数率降低,热中子计数变化量D增加;示踪剂中氧化钆的上、下限值随裂缝宽度呈指数降低,且当裂缝宽度为1.0 cm时,Gd 2O 3的质量分数上限为0.68%,下限为0.03%。     

测绘工程专业CAD课程教学改革与实践

结合测绘工程专业本身的特点，该文首先介绍了CAD课程的作用，分析了该课程的教学现状，总结了在现实教学中存在的问题，进一步引出教学改革的必要性。接下来结合实际教学经验，分别从课程的教学内容、教材的选择、教学方法与教学手段、考核方式、师资队伍建设等方面探讨了AutoCAD教学的改革措施，对提高CAD课程教学质量，培养学生实际动手能力，更好地将CAD应用到专业中，适应实际行业工作需求，具有十分重要的意义。     

heory, controls parameter and application of the packed-bed electroosmotic pump

Theory and controls parameters affecting the output pressure and flow rate of the packed-bed electroosmotic pump (p-EOP), including zeta potential ζ, electric intensity E, length of the column L, cross-section area of the column A, dielectric constant ε and viscosity coefficient η of the liquid being pumped, are discussed concisely. And also, the fabrication and application of the p-EOP are introduced.