基于改进Mask-RCNN的遥感影像建筑物提取

由于遥感影像中建筑物种类繁多且与周围环境信息混淆,传统方法难以实现建筑物的准确高效提取.本文提出了一种基于改进Mask-RCNN的建筑物自动提取方法,利用PyTorch深度学习框架搭建改进Mask-RCNN网络模型架构,在网络的设计中添加了路径聚合网络和特征增强功能,通过监督和迁移学习的方式对Inria航空影像标签数据集进行多线程迭代训练与模型优化学习,实现了建筑物的自动精确分割和提取.基于不同开源数据集,分别与SVM、FCN、U-net和Mask-RCNN等建筑物提取算法进行对比,实验表明,本文方法可以高效准确、高效地提取建筑物,对于同一个数据集,提取结果的mAP、mRecall、mPrecision和F1分数这4个评价指标均优于对比算法.     

结合OTSU与迭代三角网的机载LiDAR建筑物点云提取

针对机载LiDAR建筑物点云提取过程中受地形影响参数设置困难,建筑物、树木难以区分等问题,提出一种结合最大类间方差法与迭代三角网相结合的机载LiDAR建筑物点云提取算法。在已有滤波结果的基础上,首先采用最大类间方差法对滤波得到的非地面点进行预处理,提取初始建筑物点;然后运用改进的迭代三角网方法对初始建筑物点云进行精确提取,得到最终的建筑物点云。实验选取国际摄影测量与遥感协会提供的三组LiDAR点云数据进行建筑物点云提取。结果表明,该算法可以较好地实现建筑物点云的高精度自动提取,且对不同屋顶类型以及地形具有良好的自适应性,验证了算法的可靠性。     

结合OTSU与迭代三角网的机载LiDAR建筑物点云提取

针对机载LiDAR建筑物点云提取过程中受地形影响参数设置困难,建筑物、树木难以区分等问题,提出一种结合最大类间方差法与迭代三角网相结合的机载LiDAR建筑物点云提取算法。在已有滤波结果的基础上,首先采用最大类间方差法对滤波得到的非地面点进行预处理,提取初始建筑物点;然后运用改进的迭代三角网方法对初始建筑物点云进行精确提取,得到最终的建筑物点云。实验选取国际摄影测量与遥感协会提供的三组LiDAR点云数据进行建筑物点云提取。结果表明,该算法可以较好地实现建筑物点云的高精度自动提取,且对不同屋顶类型以及地形具有良好的自适应性,验证了算法的可靠性。     

利用格式塔的高分辨率遥感影像建筑物提取

目的 格式塔心理学的理论基础为通过对事物的部分感知,实现对事物整体的认识。本文将该思想应用到建筑物提取中,提出一种兼顾目标细节及整体几何特征的高分辨率遥感影像建筑物提取方法。方法 首先,利用SIFT算法提取特征点作为候选边缘点;然后定义格式塔序列连续性原则判别边缘点,从而得到边缘点点集;并由边缘点点集拟合边缘,实现遥感影像建筑物提取。结果 利用提出算法,对WorldView-2遥感影像进行建筑物提取实验。通过与基于多尺度分割和区域合并的建筑物提取算法对比可以看出,提出算法能够更加准确、完整地提取出建筑物。采用分支因子、遗漏因子、检测率和完整性4个定量化指标对实验结果的定量评价,本文算法的检测率和完整性均大于对比算法,且本文算法的检测率均在95%以上,验证了提出基于格式塔理论的高分辨率遥感影像建筑物提取算法的有效性和准确性。结论 基于格式塔的高分辨率遥感影像建筑物提取算法能够准确刻画建筑物细节特征,同时兼顾建筑物整体几何轮廓,准确提取高分辨率遥感影像中的建筑物。本文算法针对高分辨率遥感影像,适用于提取边缘具有直线特征的建筑物。使用本文算法进行遥感影像建筑物提取时,提取精度会随分辨率降低而降低,建议实验影像分辨率在5 m以上。     

改进KAZE算法的移动目标检测

针对KAZE算法特征提取速度慢以及移动目标检测效果差等问题,提出一种改进KAZE算法的移动目标检测算法。在原有KAZE算法基础上,对求解线性非尺度空间的迭代步长进行了重新修正;在特征匹配阶段,对特征点在相邻帧间位移进行求解后,用于背景估计。实验结果表明:改进后的算法加速了非线性尺度空间的求解,特征点提取速度提升了9. 3%,特征匹配率提升了1. 6%;用于无人机移动目标检测中,也取得了良好效果。     

基于边缘特征的多源高分辨率影像配准

针对多源高分辨率影像之间较大的非线性辐射差异和局部几何变形造成较差配准精度的问题,提出一种基于边缘特征的多源高分辨率影像配准方法。该方法首先通过各向异性扩散滤波构造输入影像的非线性尺度空间,在此基础上计算每一尺度的扩展相位一致性最大矩以获取丰富的边缘特征,并利用基于分块策略的FAST检测器提取稳定的特征点;其次利用多尺度多方向Log-Gabor滤波生成主方向索引图（Main Orientated Index Map,MOIM）,并结合高斯加权构建一种稳健的特征描述子;最后采用巴氏距离和快速采样一致（Fast Sample Consensus,FSC）方法获取同名点。选择多组多源高分辨率影像进行实验,结果表明：该方法能够有效克服多源高分辨率影像间非线性辐射差异和局部几何变形,配准效果好于其他相关方法,并且平均配准精度优于1个像素。     

基于边缘保持滤波的高光谱影像光谱-空间联合分类

针对高光谱遥感影像分类过程中,高维数据引起的"维数灾难"以及空间邻域一致性信息没有得到充分利用的问题,提出一种基于边缘保持滤波（Edge-preserving filtering,EPF）的高光谱影像光谱-空间联合分类算法.该算法首先进行波段子集划分和主成分提取,构造新的低维特征集,在保存影像结构信息的前提下降低数据维度;其次利用支持向量机（Support vector machine,SVM）获得低维特征集的初始分类概率图;然后利用原始影像主成分对初始分类概率图进行边缘保持滤波,融合光谱信息和空间信息;最后根据滤波后分类概率图对应像素点值的大小确定每个像素的类别.在Indian Pines和Pavia University两组高光谱数据上进行仿真实验,相同实验条件下,本文算法都获得最高分类精度和最少的时间消耗.仿真结果表明本文算法在高光谱遥感影像分类任务中具有明显的优势.     

基于非线性扩散滤波的指纹增强算法

根据指纹图像的特性提出了一种利用非线性扩散滤波增强指纹图像以提取细节点的 方法.该算法利用指纹图像的结构张量构造扩散张量.使得滤波沿着指纹脊线方向扩散.在 求解非线性扩散滤波中最重要是怎样选取迭代时间和迭代步长.为了解决这两个问题,根据 扩散张量在大小为(2n+1)×(2n+1)窗口上的离散形式,分解扩散滤波为4n个方向的和, 然后利用托马斯-高斯消元法在每个方向快速求解非线性滤波后的指纹增强图像.该算法的 优点是可以有效地连接断裂脊线,而不会改变指纹脊线的连续性和奇异性.在NIST27数据 库及FVC2002数据库上的实验结果表明,基于非线性扩散滤波的指纹增强算法可以有效地 提高提取细节点的性能和指纹匹配率.     

高分辨率SAR影像形态学层级分析的建筑物检测

目的 现有基于结构分析的高分辨率SAR影像建筑物检测方法,只考虑了直线和L形结构建筑物,并且依赖建筑物高亮线条处阴影区作为建筑物识别的主要特征;当处于复杂场景时,阴影区受制于背景较暗或建筑物密集而无法准确得到,导致建筑物检测误差大、检测率低。针对上述问题,提出一种基于形态学层级分析的高分辨率SAR影像无监督建筑物检测算法。方法 该方法基于单幅单极化高分辨率SAR影像,首先利用改进的形态学交替滤波算子有效抑制其固有的斑点噪声,大大剔除了同质区背景噪声的干扰;然后利用层级分析形态学差分属性断面算法来实现对SAR影像建筑物的几何结构特征的提取;最后结合特征融合和属性阈值分割等后处理步骤得到复杂场景下建筑物提取信息。结果 将上述方法在建筑物密集的城区SAR影像中实验,通过与其他方法对比分析,具有检测率高、误差小的特点,准确率和召回率分别为95.38%、86.31%,并对降低虚警率方面有明显的优势。结论 将形态学交替滤波与形态学属性滤波的改进与结合,在对不同走向、尺寸和形状的高密度建筑物检测中具有较好的适应性。     

一种粒子势概率假设密度滤波纯方位多目标跟踪算法

针对基于势概率假设密度算法(CPHD)的纯方位多目标跟踪,提出一种新型的多传感器粒子CPHD滤波算法.该算法通过分析混合线性/非线性状态模型的结构信息,结合粒子滤波(PF)与卡尔曼滤波(KF)对各个目标的状态进行预测与估计,运用Mean-Shift方法提取概率假设密度的峰值作为目标状态估计值,并对算法复杂度进行了分析.仿真结果表明,算法可改善目标跟踪效果.     

Localized image watermarking based on feature points of scale-space representation

This paper proposes a novel method for content-based watermarking based on feature points of an image. At each feature point, the watermark is embedded after scale normalization according to the local characteristic scale. Characteristic scale is the maximum scale of the scale-space representation of an image at the feature point. By binding watermarking with the local characteristics of an image, resilience against affine transformations can be obtained easily. Experimental results show that the proposed method is robust against various image processing steps including affine transformations, cropping, filtering and JPEG compression.     

Fingerprint classification using fast Fourier transform and nonlinear discriminant analysis

In this paper, we present a new approach for fingerprint classification based on discrete Fourier transform (DFT) and nonlinear discriminant analysis. Utilizing the DFT and directional filters, a reliable and efficient directional image is constructed from each fingerprint image, and then nonlinear discriminant analysis is applied to the constructed directional images, reducing the dimension dramatically and extracting the discriminant features. The proposed method explores the capability of DFT and directional filtering in dealing with low-quality images and the effectiveness of nonlinear feature extraction method in fingerprint classification. Experimental results demonstrates competitive performance compared with other published results.     

抑制光照不均匀影响的图像特征点提取算法

为解决在光照不均匀情况下图像特征点提取算法表现效果不佳的问题,提出了一种改进的尺度不变特征转换（Scale Invariant Feature Transform,SIFT）算法抑制光照不均的影响。该方法在尺度空间构造中对输入的图像进行频域上的高斯高通滤波处理来滤除光照成分,并结合Top-hat变换弱化高斯滤波器参数选取难度,利用高斯卷积构建基于光照滤除与参数弱化的高斯差分金字塔,融合SIFT算法生成具有良好光照不变性的GT-SIFT描述子,进行特征点提取与匹配。实验结果表明,与传统算法相比改进算法在光照不均匀条件下具有更好的鲁棒性,图像特征点提取与匹配效果更好。     

A New Digital Image Watermarking Algorithm Resilient to Desynchronization Attacks

Synchronization is crucial to design a robust image watermarking scheme. In this paper, a novel feature-based image watermarking scheme against desynchronization attacks is proposed. The robust feature points, which can survive various signal-processing and affine transformation, are extracted by using the Harris-Laplace detector. A local characteristic region (LCR) construction method based on the scale-space representation of an image is considered for watermarking. At each LCR, the digital watermark is repeatedly embedded by modulating the magnitudes of discrete Fourier transform coefficients. In watermark detection, the digital watermark can be recovered by maximum membership criterion. Simulation results show that the proposed scheme is invisible and robust against common signal processing, such as median filtering, sharpening, noise adding, JPEG compression, etc., and desynchronization attacks, such as rotation, scaling, translation, row or column removal, cropping, and random bend attack, etc.     

一种基于Gabor小波的局部特征尺度提取方法

图像的局部特征尺度在进行特征提取和构造尺度不变量时非常重要。提出了一种基于Gabor小波的局部特征尺度提取方法,该方法利用视皮层简单细胞的2维Gabor函数模型, 构造了一个Gabor尺度空间核函数,利用该核函数计算图像的Gabor尺度空间分解,并在尺度空间中搜索局部极大值作为特征点的固有尺度。实验结果表明,该方法可在不同对比度 条件下有效地提取各类特征的局部尺度,并且相比高斯拉普拉斯(LoG)方法有更好的适应性和可靠性。     

Fast implementation of scale-space by interpolatory subdivisionscheme

While the scale-space approach has been widely used in computer vision, there has been a great interest in fast implementation of scale-space filtering. We introduce an interpolatory subdivision scheme (ISS) for this purpose. In order to extract the geometric features in a scale-space representation, discrete derivative approximations are usually needed. Hence, a general procedure is also introduced to derive exact formulae for numerical differentiation with respect to this ISS. Then, from ISS, an algorithm is derived for fast approximation of scale-space filtering. Moreover, the relationship between the ISS and the Whittaker-Shannon sampling theorem and the commonly used spline technique is discussed. As an example of the application of ISS technique, we present some examples on fast implementation of λτ-spaces as introduced by Gokmen and Jain (1997), which encompasses various famous edge detection filters. It is shown that the ISS technique demonstrates high performance in fast implementation of the scale-space filtering and feature extraction     

A novel white blood cell segmentation scheme based on feature space clustering

This paper presents a novel white blood cell (WBC) segmentation scheme based on two feature space clustering techniques: scale-space filtering and watershed clustering. In this scheme, nucleus and cytoplasm, the two components of WBC, are extracted, respectively, using different methods. First, a sub image containing WBC is separated from the original cell image. Then, scale-space filtering is used to extract nucleus region from sub image. Later, a watershed clustering in 3-D HSV histogram is processed to extract cytoplasm region. Finally, morphological operations are performed to obtain the entire connective WBC region. Through feature space clustering techniques, this scheme successfully avoids the variety and complexity in image space, and can effectively extract WBC regions from various cell images of peripheral blood smear. Experiments demonstrate that the proposed scheme performs much better than former methods.     

Automatic building height extraction by volumetric shadow analysis of monoscopic imagery

This article presents a new approach to automatic extraction of building heights from monoscopic urban scenes. A volumetric shadow analysis (VSA) method was proposed previously for extraction of 3D building information (height, shape, and footprint location) and for handling occluded building footprints or shadows. It determined building heights by adjusting building height manually until the projected shadows generated for an assumed height and actual shadows in the image matched. In this article, we propose an intelligent scheme based on the VSA for automatic building height extraction. We achieve this by checking the location change of projected shadow lines with respect to the actual shadow regions while building heights are increased incrementally. In this article, the performance of the proposed automatic height extraction was compared to that of manual extraction. The method was first applied to IKONOS, KOMPSAT-2, QuickBird, and Worldview-1 images with manually extracted building roofs. The root mean square error (RMSE) of building heights was under 3 m by automatic height extraction and 2 m by manual extraction. The RMSE of building footprint location was close to twice that of image ground sample distance (GSD) by automatic height extraction and under twice that of image GSD by manual extraction. These results support the capability of the proposed method in automatic height extraction from a single image efficiently and accurately, and in handling occluded building footprints and shadows. Second, the method was combined with an existing roof extraction method and tested for automated building roof extraction. The results showed that the proposed method can also provide a powerful cue for automatic building roof extraction from a single image.     

Exploring and exploiting the structure of saddle points in Gaussian scale space

When an image is filtered with a Gaussian of width σ and σ is considered as an extra dimension, the image is extended to a Gaussian scale-space (GSS) image. In earlier work it was shown that the GSS-image contains an intensity-based hierarchical structure that can be represented as a binary ordered rooted tree. Key elements in the construction of the tree are iso-intensity manifolds and scale-space saddles.A scale-space saddle is a critical point in scale space. When it connects two different parts of an iso-intensity manifold, it is called “dividing”, otherwise it is called “void”. Each dividing scale-space saddle is connected to an extremum in the original image via a curve in scale space containing critical points. Using the nesting of the iso-intensity manifolds in the GSS-image and the dividing scale-space saddles, each extremum is connected to another extremum. In the tree structure, the dividing scale-space saddles form the connecting elements in the hierarchy: they are the nodes of the tree. The extrema of the image form the leaves, while the critical curves are represented as the edges.To identify the dividing scale-space saddles, a global investigation of the scale-space saddles and the iso-intensity manifolds through them is needed.In this paper an overview of the situations that can occur is given. In each case it is shown how to distinguish between void and dividing scale-space saddles. Furthermore, examples are given, and the difference between selecting the dividing and the void scale-space saddles is shown. Also relevant geometric properties of GSS images are discussed, as well as their implications for algorithms used for the tree extraction.As main result, it is not necessary to search through the whole GSS image to find regions related to each relevant scale-space saddle. This yields a considerable reduction in complexity and computation time, as shown in two examples.