基于快速均值漂移和水平集方法的图像分割技术

通过加强边缘检测函数的约束作用,对L i提出的无需重新初始化的水平集方法进行了改进,用于提高水平集方法对弱边缘与低对比度图像的分割处理性能。同时,利用快速均值漂移法对图像进行预分割,将得到的轮廓作为水平集方法的初始轮廓,从而降低水平集函数的迭代次数。脑部M R图像的分割实验结果表明,该算法能够准确地提取相应的脑组织,同时大大地提高水平集函数的演化速度。     

基于极点与极性的车辆分割及阴影处理方法

视频交通参数检测中的车辆分割需要准确地检测与车辆连在一起的阴影和车灯产生的亮斑。一般地,路面与车辆的图像在灰度结构上存在显著差异。本文推导了图像在光照变化情况下的一种灰度结构——极点与极性分布图,并提出了一个基于该分布图的车辆阴影与亮斑检测算法。该算法能够精确地检测车辆阴影,车灯照射产生的路面亮斑和因其他原因被误为车辆的路面。该方法使用对光照变化不敏感的灰度结构,阴影检测准确而稳定,计算量小。     

一种深度学习分割遥感影像道路的损失函数

传统的根据光谱特征或形态学算法来分割道路,存在精度低、阈值难确定等缺点,而深度学习中已有的方法并未考虑道路的特性,只是利用通用方法来分割道路.针对上述不足,提出了一种针对道路特有形态的深度学习损失函数——形态损失函数.首先使用连通性算法将预测结果划分为若干个相互分离的连通区域,分别计算这些区域的面积与外接圆面积的比值,...     

Flow field generated by a dielectric barrier discharge plasma actuator in quiescent air at initiation stage

A single Dielectric Barrier Discharge (DBD) plasma actuator driven by Alternating Current (AC) power, capable of inducing a starting vortex and a wall jet in quiescent air, is suited for low-Reynolds-number flow control. However, the starting vortex and the wall jet are usually observed after the plasma actuator has been operated for dozens of and hundreds of cycles of the voltage, respectively. The detail of the induced flow field at the initiation stage of the plasma actuator has rarely been addressed. At the initiation stage, a thin jet that provides the impetus for the entrainment of the induced flow at the beginning of the plasma actuation is first observed by using a high-accuracy phase-lock Schlieren technique and a high-speed Particle Image Velocimetry (PIV) system. This is the initial form of the momentum transfer from the plasma to the fluid. Then, an arched type jet is created by the plasma actuator. In addition, the whole development process of the induced flow field from the starting point of the thin jet to the quasi-steady stage of wall jet is presented for providing a comprehensive understanding of the plasma actuator and proposing a relevant enhancement of the numerical simulation model.     

一种适用于遥感图像的可逆信息隐藏算法

可逆信息隐藏技术，可用于实现遥感图像重要目标区域的无损传输。通过对现有可逆信息隐藏算法进行研究，发现现有算法的单次嵌入容量不足，且需要额外的无损二值压缩处理附加信息，无法满足遥感图像处理的有效性和实时性要求。借助给定门限对载体图像不重叠子块的平滑性进行划分，通过编码平滑子块内基准像素与其余像素之差，并结合比特替换，实现秘密信息的可逆隐藏。对于512×512像素的灰度遥感图像，单次嵌入容量可达5×105 bit以上。试验结果表明，该算法可实现单次高嵌入容量的可逆隐藏，并不需要任何额外无损二值压缩来处理辅助信息。     

基于数字图像相关的预腐蚀2024-T4铝合金疲劳开裂实验研究

通过三维数字图像相关(3D-DIC)方法研究预腐蚀2024-T4铝合金在三种不同最大应力水平和应力比下的疲劳开裂行为。通过应变场演化直观地显示裂纹萌生和扩展的时空特征,并通过扫描电镜观测关键损伤区域的断裂微观形貌。结果表明:试样边缘的局部腐蚀促进了疲劳裂纹萌生,影响了裂纹的形核位置,并引起材料氢脆现象;疲劳裂纹扩展方向与加载方向呈60°~68°角,表明疲劳裂纹扩展可以用KⅠ/KⅡ混合模式来描述;预腐蚀铝合金疲劳失效存在单裂纹断裂、多裂纹联合、多裂纹竞争和多裂纹平行扩展4种典型的失效模式。     

基于Otsu和FMM方法的风洞试验图像修复

各种光学测试技术在风洞试验中应用越来越广泛，图像中的破损或无关信息均会影响数据的处理和分析。以存在缺陷的风洞纹影图像为例，对数字图像修复技术进行了研究。根据缺陷离散分布、大小及形状各异，核心区域的灰度值较低以及边缘处灰度梯度变化较大等特点，将图像修复流程设计为确定待处理区域、缺陷识别和缺陷修复等3个步骤。前2个步骤使用Otsu算法进行图像的分割处理，然后使用快速行进方法（FMM）完成缺陷修复。经处理，纹影图像的缺陷被自动识别，并得到了减少或削弱，而且修复的信息客观合理。该方法可推广至其他光学测试技术的图像修复。     

Evaluating the statistical performance of less applied algorithms in classification of worldview-3 imagery data in an urbanized landscape

In recent decade, analyzing the remotely sensed imagery is considered as one of the most common and widely used procedures in the environmental studies. In this case, supervised image classification techniques play a central role. Hence, taking a high resolution Worldview-3 over a mixed urbanized landscape in Iran, three less applied image classification methods including Bagged CART, Stochastic gradient boosting model and Neural network with feature extraction were tested and compared with two prevalent methods: random forest and support vector machine with linear kernel. To do so, each method was run ten time and three validation techniques was used to estimate the accuracy statistics consist of cross validation, independent validation and validation with total of train data. Moreover, using ANOVA and Tukey test, statistical difference significance between the classification methods was significantly surveyed. In general, the results showed that random forest with marginal difference compared to Bagged CART and stochastic gradient boosting model is the best performing method whilst based on independent validation there was no significant difference between the performances of classification methods. It should be finally noted that neural network with feature extraction and linear support vector machine had better processing speed than other.     

地基光学图像中本体-帆板结构航天器姿态估计方法

提出一种利用地基光学图像估计空间目标姿态的方法。针对本体-帆板结构航天器外形结构特点，定义了一种不易遮挡的两轴结构特征，设计了地基光学图像中该特征的提取方法；根据航天器成像过程建立了特征点二三维几何投影方程，实现了航天器三维姿态的解算。利用单站序列图像，可获得航天器姿态变化角速度信息。在轨航天器的试验结果表明，本方法能有效地从地基光学图像估计出航天器姿态。     

基于颜色、形状和纹理的多特征融合图像检索

基于单一特征的图像检索的许多方法都已经相当成熟,如何有效利用这些特征是基于内容图像检索的一个关键问题。利用颜色、纹理、形状等特征,在多特征融合检索时用动态生成权值的方法进行图像检索。实验结果表明,算法能够准确、高效地检索出目标图像,相对于单一特征的检索,有效提高了图像的检索精度。