图像欧氏距离在人脸识别中的应用研究

图像欧氏距离可以嵌入到许多传统的图像分类识别算法中,该嵌入是通过对原始图像的线性变换来实现的,给出了一种基于数据场的图像线性变换方法,将其应用到图像欧氏距离中.实验结果表明,基于数据场的线性变换方法是一种可行的图像线性变换方法,该方法可以完成大尺度图像的线性变换,方便地将图像欧氏距离嵌入到传统人脸识别算法中.     

利用数据场和欧氏距离的图像边缘提取

图像边缘是图像分析和识别的基础,图像边缘信息的准确性和完整性对后续图像分析和识别有重要影响。为实现图像边缘有效提取,提出一种利用数据场和图像欧氏距离的图像边缘提取方法。首先,该方法利用数据场理论构建图像数据场,实现图像灰度值特征空间到数据场势值空间的转换。然后,在对图像数据场的势值计算时引入图像欧氏距离,利用图像区域欧氏距离扩大像素差异,抑制微小细节和噪声,得到"背景"和"目标"相对分离的势值图。最后,用改进Canny算法对势值图进行边缘提取。实验表明,用本文方法可以有效提高边缘提取的准确性,减少伪边缘,抑制冗余细节和噪声。     

HJ-1卫星在港口海冰监测中的应用

以黄骅港煤炭航道海冰监测为例,采用国产环境一号卫星CCD数据,研究环境一号卫星监测海冰原理,实验环境一号卫星在煤炭航道中海冰的监测方法和流程,通过对航道、港池和锚地海冰反射率的研究,确定海冰类型,估算海冰厚度;利用经几何精纠正后的影像,统计不同海冰类型在航道、港池和锚地中的位置、范围和密集度;根据环境一号卫星日或双日的过境周期,实施了2010/2011年冬季黄骅港海冰业务化监测,从而直接为港口调度航道破冰、装船、北煤南运提供遥感技术支持。     

卫星遥感技术在海冰减灾中的应用

海冰是一种严重的海洋灾害。60年代由于缺乏有效的监测技术,曾造成人民生命财产的重大损失。气象卫星具有观测周期短、观测面积大、观测内容丰富和同步性强的特点,是海冰动态监测的有效手段。自80年代以来,国内相继开展了     

基于欧氏距离和精英交叉的免疫算法研究

针对免疫算法(AIA)的运行速度和收敛速度慢的缺点,提出一种精英交叉策略,并将精英交叉策略与基于欧氏距离的免疫算法相结合,得到一种基于欧氏距离和精英交叉的免疫算法(DKBAIA).提出了抗体相似矩阵的概念和一种改进的措施,对DKBAIA进行改进,从而得到改进的DKBAIA算法(MDKBAIA).仿真结果表明,精英交叉策略对免疫算法的收敛性能有显著的改善;MDKBAIA的运行速度也得到极大的提高,已接近GA的运行速度.     

高维空间中用计算街区和棋盘距离的线性组合代替计算欧氏距离

在高维空间中点的超球范围查找问题是 :已知一个高维数据点集 ,输入一个点和半径数值 ,询问所确定超球范围内包含有给出点集中哪些点 .考查了解决这个问题时利用计算街区和棋盘距离的线性组合代替计算欧氏距离的方法 .这一方法由于减少了乘法计算而明显地可以提高效率 .为提高计算精度 ,对如何选择构造线性组合时的系数进行了深入分析 ,提出了使选择系数达到上、下确界或最优值的计算方法 .为提出的方法设计了实现算法并进行了运行实验 .结果表明方法是有效的 ,可以应用到有关高维空间中距离计算的广泛问题中     

卫星遥感海冰监测技术在河北省近海海域的应用

卫星遥感技术已成为海冰监测的重要技术手段之一。介绍了应用EOS系列卫星、HY-1B卫星和HJ-1A/B卫星等遥感数据提取河北省近海海域海冰分布、海冰类型、海冰厚度等海冰信息的方法。重点研究了应用HJ-1A/B卫星遥感数据进行海冰分类,及根据海冰类型与厚度的对应关系估算海冰厚度的方法,并以2013年1月17日的卫星遥感数据为例,提取海冰信息,制作相关海冰专题产品。结果表明:利用多种卫星遥感数据进行海冰监测,既可相互补充又能提高监测效率和准确率,制作更精细化的海冰监测产品,为河北海洋管理部门制定防灾减灾措施提供重要依据。     

一种图像距离定义及其在置乱效果评价中的应用

针对传统欧氏距离不能真实刻画图像间差异程度的不足,考虑到图像中任意位置灰度信息与其周围邻域灰度存在紧密关系,提出一种考虑邻域灰度信息的图像距离新定义,并用其构造一种描述图像置乱效果好坏的评价函数。实验结果表明,该评价方法能够较好地刻画图像的置乱程度,反映了加密次数与置乱程度之间的关系,与人的视觉基本相符。而且对于不同的图像,该评价方法能在一定程度上反映所用的置乱变换在各置乱阶段的效果。     

基于马氏距离的文本聚类算法在自动阅卷系统中的应用

基于欧氏距离的传统模糊划分聚类算法较适用于球型结构的聚类。将其应用于维度较高的文本聚类时,准确率和效率均有所下降。为解决这一问题,提出一种基于马氏距离的文本聚类算法。该算法可发现非球形结构的类簇,在不需要先验知识的情况下,仅通过数学迭代即可得到聚类结果。鉴于当前无纸化考试系统的广泛应用,将该算法应用于主观题的自动阅卷系统中。通过对多种主观题的仿真实验,表明了该算法与C均值和FCM算法相比,不仅能获得较高的准确率,算法收敛速度也较快。     

Snakes模型在卫星图片道路提取中的应用

航空影像中自动提取道路已经成为了一个研究热点,并提出了大量的提取算法,然而道路网的自动提取仍然是件具有挑战性的任务。文中介绍的Snakes模型（基于主动轮廓模型）对于卫星影像中道路提取具有很好的效果。该方法分为道路初始条件的获取阶段和利用初始条件进行道路网跟踪处理阶段,而文中着重分析后面一阶段,且道路网的跟踪处理是在初始条件的基础上完成的。最后通过实验证明,该方法能正确清晰地提取出主干道,并且对于道路弯曲轮廓的提取也非常有效。     

Studies of the Antarctic sea ice edge and ice extent from satellite and ship observations

Passive microwave-derived ice edge locations in the Antarctic are assessed against in situ observations from ships between 1989 and 2000. During the growth season (March-October), the ship data agrees with satellite data very well, with r² values of 0.99 and 0.97 for the Bootstrap and Team algorithms, respectively. During the melt season (November-February), the agreement is not so good with the passive microwave ice edge typically 1-2° of latitude south of the observations. This is due to the low concentration and saturated nature of the ice, and the r² values for this period are 0.92 and 0.80 for the Bootstrap and Team algorithms, respectively. Sensitivity studies show that such an offset in the summer ice edge location can cause significant errors in trend studies of the extent of sea ice cover in the Southern Ocean. The passive microwave ice concentration at the ice edge observed by ship varies greatly, averaging 14% for the Team algorithm and 19% for the Bootstrap. Comparisons between passive microwave data and SAR, Landsat and OLS data during the ice growth season show that while small-scale details in ice edge location are lost, the passive microwave data generally provide good and consistent representation of the higher resolution imagery.     

恶意模型下计算欧几里德距离的协议

安全多方计算（MPC）是一个允许多个参与方在保持各自输入隐私的前提下联合计算一个函数。Yao和Goldreich等人（STOC’87）开创性的工作表明,存在陷门置换的前提下,任何一个函数都存在安全多方计算协议,并给出了安全多方计算的一个通用解决方案,但是该方案由于效率问题而不实用。因此,Goldreich同时指出需要针对特定问题提出特定的安全多方计算协议。提出了一个新的基于分布式EI Gamal加密的计算两个向量欧几里德距离的安全协议,并在混合模型下给出了协议的安全性证明。与原来的方案比较,该协议的计算和通信复杂度都较低,适用于计算和通信能力都有限的应用环境。     

Detection of small-scale roughness and refractive index of sea ice in passive satellite microwave remote sensing

Polar ice masses and sheets are sensitive indicators of climate change. Small-scale surface roughness significantly impacts the microwave emission of the sea ice/snow surface; however, published results of surface roughness measurements of sea ice are rare. Knowing the refractive index is important to discriminate between objects. In this study, the small-scale roughness and refractive index over sea ice are estimated with AMSR-E observations and a unique method. Consequently, the small-scale surface roughness of 0.25 cm to 0.5 cm at AMSR-E 6.9 GHz shows reasonable agreement with the results of known observations, ranging from 0.2 cm to 0.6 cm for the sea ice in the Antarctic and Arctic regions. The refractive indexes are retrieved from 1.6 to 1.8 for winter, from 1.2 to 1.4 for summer in the Arctic and the Antarctic, which are similar to those of the sea ice and results from previous studies. This research shows the physical characteristics of the sea ice edges and melting process. Accordingly, this investigation provides an effective procedure for retrieving the small-scale roughness and refractive index of sea ice and snow. Another advantage of this study is the ability to distinguish sea ice from the sea surface by their relative small-scale roughness.     

利用欧氏距离变换Snake模型分割脊椎CT图像

在计算机虚拟脊椎矫正系统中,为模拟钢钉打入椎骨后相关脊椎骨产生位移和旋转,需要建立患者脊柱的三维模型。由于脊椎结构复杂,空间位置相互交错,造成了在靠近上下椎骨连接区域的CT图像中既有当前椎骨的图像,还有相邻其它椎骨的部分骨组织图像的状态,用传统的分割方法很难从脊柱的CT序列图像中分割出单个的椎骨。在传统Snake模型算法基础上,通过引进一种新的能量表达方法,改善了传统算法对具有复杂凹形轮廓物体分割效果差的缺点,在脊椎CT图像分割中取得了良好的效果。     

基于欧氏距离变换的肺部粘连肿瘤分割算法

以肺部CT图像为研究对象,针对肺部粘连肿瘤图片本身的特点,提出了一种基于边缘跟踪的二维欧氏距离变换算法.从目标区域的最外层边界开始,自外向内对目标区域进行边缘跟踪、腐蚀,直到肿瘤区粘连部分与肺部边界分离.算法能够计算精确的欧氏距离.通过实验分割出的肿瘤和放射科医生手工勾画的肿瘤轮廓对比,5幅病例图像重叠率达到了75%左右,实验结果表明该方法对于中晚期肺部粘连肿瘤的分割有一定的效果.     

Constant-time algorithm for computing the Euclidean distance maps of binary images on 2D meshes with reconfigurable buses

The computation of Euclidean distance maps (EDM), also called Euclidean distance transform, is a basic operation in computer vision, pattern recognition, and robotics. Fast computation of the EDM is needed since most of the applications using the EDM require real-time computation. It is shown in L. Chen and H.Y.H. Chuang [Information Processing Letters, 51, pp. 25–29 (1994)] that a lower bound Ω(n²) is required for any sequential EDM algorithm due to the fact that in any EDM algorithm each of the n² pixels has to be scanned at least once. Recently, many parallel EDM algorithms have been proposed to speedup its computation. Chen and Chuang proposed an algorithm for computing the EDM on an n×n mesh in O(n) time [L. Chen and H.Y.H. Chuang Parallel Computing, 21, pp. 841–852 (1995)]. Clearly, the VLSI complexities of both the sequential and the mesh algorithm described in L. Chen and H.Y.H. Chuang [Parallel Computing, 21, pp. 841–852 (1995)] are AT²=O(n⁴), where A is the VLSI layout area of the design and T is the computation time using area A when implemented in VLSI. In this paper, we propose a new and faster parallel algorithm for computing the EDM problem on the reconfigurable VLSI mesh model. For the same problem, our algorithm runs in O(1) time on a two-dimensional n²×n² reconfigurable mesh. We show that the VLSI complexity of our algorithm is the same as those of the above sequential algorithm and the mesh algorithm, while it uses much less time. To our best knowledge, this is the first constant-time EDM algorithm on any parallel computational model.     

Mapping sediment-laden sea ice in the Arctic using AVHRR remote-sensing data: Atmospheric correction and determination of reflectances as a function of ice type and sediment load

Exploiting the fact that the spectral characteristics of light backscattered from sediment-laden ice differ substantially from those of clean ice and that sediment tends to accumulate at the ice surface during the first melt season, remote-sensing techniques provide a valuable tool for mapping the extent of particle-laden ice in the Arctic basin and assessing its particulate loading. This study considers two fundamental problems that still need to be addressed in order to make full use of satellite observations for this type of assessment: (i) the effects of the atmosphere on surface reflectances derived from radiances measured by the satellite sensor need to be quantified and ultimately corrected for, and (ii) the spectral reflectance of the ice surface as a function of particle loading and sub-pixel distribution needs to be determined in order to derive quantitative estimates from the at-sensor satellite signal. Here, spectral albedos have been computed for different ice surfaces of variable sediment load with a radiative transfer model for sea ice coupled with an optical model for particulates included in sea ice. In a second step, the role of the atmosphere in modulating the surface reflectance signal is assessed with the aid of an atmospheric radiative transfer model applied to a “standard” Arctic atmosphere and surface boundary conditions as prescribed by the sea ice radiative transfer model. A series of sensitivity studies helps assess differences between top-of-the-atmosphere and true surface reflectance and has been utilized to derive a look-up table for atmospheric correction of Advanced Very High Resolution Radiometer (AVHRR) data over sediment-laden sea ice surfaces. In particular, the effects of solar elevation, viewing geometry, and atmospheric properties are considered. The atmospheric corrections are necessary for certain geometries and surface types. Large discrepancies between raw and corrected data are particularly evident in the derived coverage of clean ice and ice with small sediment loading.     

Classification of normal and epileptic seizure EEG signals using wavelet transform,phase-space reconstruction,and Euclidean distance

This paper proposes new combined methods to classify normal and epileptic seizure EEG signals using wavelet transform (WT), phase-space reconstruction (PSR), and Euclidean distance (ED) based on a neural network with weighted fuzzy membership functions (NEWFM). WT, PSR, ED, and statistical methods that include frequency distributions and variation, were implemented to extract 24 initial features to use as inputs. Of the 24 initial features, 4 minimum features with the highest accuracy were selected using a non-overlap area distribution measurement method supported by the NEWFM. These 4 minimum features were used as inputs for the NEWFM and this resulted in performance sensitivity, specificity, and accuracy of 96.33%, 100%, and 98.17%, respectively. In addition, the area under Receiver Operating Characteristic (ROC) curve was used to measure the performances of NEWFM both without and with feature selections.     

Comparison of Envisat radar and airborne laser altimeter measurements over Arctic sea ice

Sea ice thickness is a crucial, but very undersampled cryospheric parameter of fundamental importance for climate modeling. Advances in satellite altimetry have enabled the measurement of sea ice freeboard using satellite microwave altimeters. Unfortunately, validation of these new techniques has suffered from a lack of ground truth measurements. Therefore, an airborne campaign was carried out in March 2006 using laser altimetry and photo imagery to validate sea ice elevation measurements derived from the Envisat/RA-2 microwave altimeter.We present a comparative analysis of Envisat/RA-2 sea ice elevation processing with collocated airborne measurements collected north of the Canadian Archipelago. Consistent overall relationships between block-averaged airborne laser and Envisat elevations are found, over both leads and floes, along the full 1300 km aircraft track. The fine resolution of the airborne laser altimeter data is exploited to evaluate elevation variability within the RA-2 ground footprint. Our analysis shows good agreement between RA-2 derived sea ice elevations and those measured by airborne laser altimetry, particularly over refrozen leads where the overall mean difference is about 1 cm. Notwithstanding this small 1 cm mean difference, we identify a larger elevation uncertainty (of order 10 cm) associated with the uncertain location of dominant radar targets within the particular RA-2 footprint. Sources of measurement uncertainty or ambiguity are identified, and include snow accumulation, tracking noise, and the limited coverage of airborne measurements.