改进U-Net网络及在遥感影像道路提取中的应用

高分辨率遥感图像分割在军事、民用等领域具有良好的应用前景,但由于复杂的背景条件以及干扰物的遮挡,导致现有算法无法较好地从遥感影像中提取道路细节信息。研究基于改进U Net网络模型,提出了MDAU-Net（multi dimension attention U-Net）网络结构模型,通过对U-Net网络结构加深至七层结构来提升精细分割道路的能力;并提出了一种多维注意力模块MD-MECA（multi dimension modified efficient channel attention）,将其添加至编码部分的特征传递步骤中,以达到对编码部分的特征传递进行优化的目的;其中利用DropBlock与Batch Normalization解决网络训练过程中出现的过拟合。试验结果表明:改进后算法可以有效提升道路的提取效果,在测试集上的准确率达到了97.04%。     

Effects of coarse riblets on air flow structures over a slender delta wing using particle image velocimetry

This research investigates the aerodynamic performance and flow characteristics of a delta wing with 65° sweep angle and with coarse axial riblets,and then compares with that of a smooth-surface delta wing.Particle Image Velocimetry(PIV)were utilized to visualize the flow over the wing at 6 cross-sections upright to the wing surface and parallel to the wing span,as well as 3 longitudinal sections on the leading edge,symmetry plane,and a plane between them at Angles of Attack(AOA)=20°and 30°and Re=1.2×10~5,2.4×10~5,and 3.6×10~5.The effects of the riblets were studied on the vortices diameter,vortex breakdown location,vortices distance from the wing surface,flow lines pattern nearby the wing,circulation distribution,and separation.The results show that the textured model has a positive effect on some of the parameters related to drag reduction and lift increase.The riblets increase the flow momentum near the wing’s upper surface except near the apex.They also increase the flow momentum behind the wing.     

典型二次曲面实时成象的改进算法

在旋转的空间几何模型的Ｐｈｏｎｇ光照模型分析基础上，基于以往的二次曲面等灰度理论，提出了一种新的镜面反射模型和投影等灰度线生成算法，该算法很好地解决了典型二次曲面在三维空间旋转任意角度之后实时地生成Ｐｈｏｎｇ模型光照图象，为二次曲面图元接拼形体奠定了良好的基础。     

基于NSCT变换和PCA的图像融合算法

针对小波变换与Contourlet变换的不足,提出了一种将非下采样Contourlet 变换与主成分分析相结合的图像融合算法。对已配准好的融合源图像作NSCT分解,采用PCA加权融合规则对低频系数处理,采用边缘强度最大的规则对各高频子带系数作融合处理,分别得到融合图像的变换系数,对该系数经NSCT逆变换重构得到融合图像。实验结果表明,算法有效地融合了源图像中的重要信息,融合图像边缘、纹理、细节清晰,得到较好视觉效果和较优的评价指标。     

Evaluation of Characteristics of Non-Metallic Inclusions in P/M Ni-Base Superalloy by Automatic Image Analysis

Non-metallic inclusions,especially the large ones,within P/M Ni-base superalloy have a major influence on fatigue characteristics,but are not directly measurable by routine inspection.In this paper,a method,automatic image analysis,is proposed for estimation of the content,size and amount of non-metallic inclusions in superalloy.The methodology for the practical application of this method is described and the factors affecting the precision of the estimation are discussed.In the experiment,the characteristics of the non-metallic inclusions in Ni-base P/M superalloy are analyzed.     

多分辨率分析及其在图像分解和编码中的应用

多分辨率分析是信号处理特别是图像处理的一种新的有效方法，可以在不同尺度上给出信号的信息。在分析其理论的基础上，将其应用在图像的分析的编码中，最后给出了处理结果。     

土地类型分类的人工神经网络方法

:研究了用人工神经网络对航空影像进行土地类型分类的方法 ,用选定窗口的平均值 ,标准差和傅利叶频谱平均值作为神经网络的输入 ,土地类型的标准值为神经网络的输出。将计算出来的未知土地类型标准值与已知土地类型的标准值相比较 ,可以确定出未知的土地类型。     

Combining Locations from Working Memory and Long-Term Memory into a Common Spatial Image

Abstract

This research uses a novel integration paradigm to investigate whether target locations read in from long-term memory (LTM) differ from perceptually encoded inputs in spatial working-memory (SWM) with respect to systematic spatial error and/or noise, and whether SWM can simultaneously encompass both of these sources. Our results provide evidence for a composite representation of space in SWM derived from both perception and LTM, albeit with a loss in spatial precision of locations retrieved from LTM. More generally, the data support the concept of a spatial image in working memory and extend its potential sources to representations retrieved from LTM.     

卫星转动条件下APS星敏感器星像目标中心提取精度分析

 目前对星敏感器星像定位的研究多限于静态情况,而卫星转动过程中,在曝光期间星像在像平面不断移动,从而影响星像定位的精度,重点分析动态情况下APS星敏感器星像目标中心的提取精度。首先分析采用质心法计算星像目标中心的误差源,提出动态精度的估计方法,并推导相应的计算公式;进而以给定的APS星敏感器参数为基础,研究了计算窗口、曝光时间等精度影响因素的选择方法;最后通过仿真进行了验证。