Multi-Scale PIIFD for Registration of Multi-Source Remote Sensing Images

This paper aims at providing multi-source remote sensing images registered in geometric space for image fusion.Focusing on the characteristics and differences of multi-source remote sensing images,a feature-based registration algorithm is implemented.The key technologies include image scale-space for implementing multi-scale properties,Harris corner detection for keypoints extraction,and partial intensity invariant feature descriptor(PIIFD)for keypoints description.Eventually,a multi-scale Harris-PIIFD image registration algorithm framework is proposed.The experimental results of fifteen sets of representative real data show that the algorithm has excellent,stable performance in multi-source remote sensing image registration,and can achieve accurate spatial alignment,which has strong practical application value and certain generalization ability.

Accuracy Analysis on the Automatic Registration of Multi-Source Remote Sensing Images Based on the Software of ERDAS Imagine

The automatic registration of multi-source remote sensing images (RSI) is a research hotspot of remote sensing image preprocessing currently. A special automatic image registration module named the Image Autosync has been embedded into the ERDAS IMAGINE software of version 9.0 and above. The registration accuracies of the module verified for the remote sensing images obtained from different platforms or their different spatial resolution. Four tested registration experiments are discussed in this article to analyze the accuracy differences based on the remote sensing data which have different spatial resolution. The impact factors inducing the differences of registration accuracy are also analyzed.

Smart Photogrammetric and Remote Sensing Image Processing for Very High Resolution Optical Images——Examples from the CRC-AGIP Lab at UNB

This paper introduces some of the image processing techniques developed in the Canada Research Chair in Advanced Geomatics Image Processing Laboratory (CRC-AGIP Lab) and in the Department of Geodesy and Geomatics Engineering (GGE) at the University of New Brunswick (UNB), Canada. The techniques were developed by innovatively/“smartly” utilizing the characteristics of the available very high resolution optical remote sensing images to solve important problems or create new applications in photogrammetry and remote sensing. The techniques to be introduced are: automated image fusion (UNB-PanSharp), satellite image online mapping, street view technology, moving vehicle detection using single set satellite imagery, supervised image segmentation, image matching in smooth areas, and change detection using images from different viewing angles. Because of their broad application potential, some of the techniques have made a global impact, and some have demonstrated the potential for a global impact.

Robust multi-sensor image matching based on normalized self-similarity region descriptor

Multi-modal image matching is crucial in aerospace applications because it can fully exploit the complementary and valuable information contained in the amount and diversity of remote sensing images.However,it remains a challenging task due to significant non-linear radiometric,geometric differences,and noise across different sensors.To improve the performance of heterologous image matching,this paper proposes a normalized self-similarity region descriptor to extract consistent structural information.We first construct the pointwise self-similarity region descriptor based on the Euclidean distance between adjacent image blocks to reflect the structural properties of multi-modal images.Then,a linear normalization approach is used to form Modality Independent Region Descriptor(MIRD),which can effectively distinguish structural features such as points,lines,corners,and flat between multi-modal images.To further improve the matching accuracy,the included angle cosine similarity metric is adopted to exploit the directional vector information of multi-dimensional feature descriptors.The experimental results show that the proposed MIRD has better matching accuracy and robustness for various multi-modal image matching than the state-of-the-art methods.MIRD can effectively extract consistent geometric structure features and suppress the influence of SAR speckle noise using non-local neighboring image blocks operation,effectively applied to various multi-modal image matching.

Multi-source Remote Sensing Image Registration Based on Contourlet Transform and Multiple Feature Fusion

Image registration is an indispensable component in multi-source remote sensing image processing. In this paper, we put forward a remote sensing image registration method by including an improved multi-scale and multi-direction Harris algorithm and a novel compound feature. Multi-scale circle Gaussian combined invariant moments and multi-direction gray level co-occurrence matrix are extracted as features for image matching. The proposed algorithm is evaluated on numerous multi-source remote sensor images with noise and illumination changes. Extensive experimental studies prove that our proposed method is capable of receiving stable and even distribution of key points as well as obtaining robust and accurate correspondence matches. It is a promising scheme in multi-source remote sensing image registration.

面向无人机绝对定位的遥感影像快速检索方法

针对在复杂环境下无人机景象匹配导航中的视觉绝对定位问题,提出了一种聚合深度学习特征的实时影像快速检索方法。首先,引入可训练软分配深度学习框架—NetVLAD,结合VGG16网络提取并聚合生成影像稳定的全局特征表达向量;其次,在初始检索阶段,使用KD树结构对影像全局特征向量构建检索索引,在不损失检索精度的前提下提高检索速度;最后,使用皮尔逊积矩相关系数对初始检索结果进行快速预判断,自动过滤初始检索结果,对于需要重排序的影像则采用特征学习匹配算法——图神经网络SuperGlue进行匹配重排序。所提方法在公开的夏季和冬季遥感影像数据集分组进行实验,实验结果表明:未重排序条件下,初始检索结果第一张影像平均准确率达到了58.27%,部分特征较好地区准确率达到了85%,对不同时相遥感影像也有很好的适应性,平均检索一张影像耗时3.7 s,可为无人机景象匹配导航的初始定位提供参考。

基于改进ORB算法的遥感图像匹配

针对遥感图像匹配中存在的误匹配率高、匹配速度慢等问题,结合相似图像局部结构一致的思想,提出一种将ORB特征提取算子与BRISK特征描述算子相结合的遥感图像匹配算法。首先使用ORB算法的特征点提取算子对特征点进行检测,再使用BRISK算法的特征点描述算子对检测到的特征点建立特征描述符,然后基于汉明距离使用暴力匹配完成特征点匹配,最后使用LPM算法剔除错误匹配点。实验结果表明:本文算法提取的特征点数量、匹配精确度和运行速度都满足遥感图像匹配的要求;相较于ORB、BRISK算法,做到了速度和精度的较好平衡,展现出了良好的匹配性能。

基于注意力机制和软匹配的多标签遥感图像检索方法

针对卷积神经网络对于多标签遥感图像特征提取能力弱、不能准确反映遥感图像多标签复杂性的问题,提出基于注意力机制和软匹配的多标签遥感图像检索方法。在特征提取阶段,以密集卷积神经网络模型为基础,在每个密集块(Dense Block)后添加CBAM(Convolutional Block Attention Module)层,实现对多标签图像区域特征提取。在模型训练时,利用区分硬匹配与软匹配的联合损失函数,学习图像的哈希编码表示。通过评估遥感图像哈希编码间的汉明距离,实现相似图像的检索。实验结果表明,所提方法在数据集NUS-WIDE和多标签遥感图像数据集DLRSD上与其他基于全局特征的深度哈希方法相比,明显提升了检索准确率。

面向海岛海岸带区域的高分遥感影像智能化色彩增强方法

原始高空间分辨率海岛海岸带遥感影像往往存在影像灰暗、偏色、地物信息较难辨识的现象。为及时获取清晰、信息丰富、反差适中、亮度均匀的海岛礁遥感影像,满足日益强烈的海岛海岸带地理信息保障需求,针对海岛海岸带高空间分辨率遥感影像,该文提出一种深度学习结合改进直方图匹配的智能化调色方法。首先,进行数据重采样与自适应分块获取抽稀影像;其次,应用MBLLEN网络对抽稀影像进行真彩色增强;最后,采用改进直方图匹配的方法对原始影像进行色彩映射,最终得到符合人眼视觉、色彩一致、细节丰富的遥感影像。采用主客观相结合的方式综合评价调色效果,结果表明:相较于Retinex,HE和MASK等常用调色方法,该文算法结果更符合人眼视觉、色彩一致、细节丰富,可有效改善海岛海岸带高空间分辨率遥感影像视觉效果,较好地保留原始地物的细节信息,大幅提升调色效率。

结合注意力机制的卫星遥感影像立体匹配

采用立体匹配技术对多视卫星遥感影像进行三维场景重建一直是摄影测量与遥感领域的核心问题。基于卷积神经网络的深度学习方法极大地促进了立体匹配技术的发展,然而其中涉及匹配困难和误匹配问题的相关研究仍然不足。为了提升卫星遥感影像不适定区域中视差估计的精度,本研究提出了一种结合注意力机制的立体匹配深度学习网络,在特征提取模块中加入注意力机制,分别从通道和空间两个维度捕获全局信息,对特征进行优化;在代价体的构建模块中构建新的代价体积,并重新设置视差的回归范围。为了验证本文方法的有效性,在US3D、WHU-Stereo两个数据集上分别与已有方法Stereo-Net、PSM-Net进行了比较分析。结果表明,本文方法在EPE(endpointerror)和D_1两个指标上均能达到最优,取得了较好的性能,提高了立体匹配的精度,尤其在无纹理、重复纹理、遮挡及视差不连续区域表现出良好的鲁棒性。

多模态遥感图像模板匹配Log-Gabor滤波方法

针对多模态遥感图像匹配难的问题,本文提出了一种基于Log-Gabor滤波的高精度匹配方法。该方法采用由粗到细的多层级密集匹配框架,无须进行特征点检测,避开了多模态图像特征点检测重复率低的问题,能够提取大量高精度匹配点对。本文方法主要分为两步:首先,利用多尺度多角度Log-Gabor滤波器构建对图像间非线性辐射差异稳健的特征金字塔;然后,利用粗尺度的底层特征图进行密集模板匹配,提取大量粗粒度的特征匹配点对,在此基础上再利用特征金字塔,实现粗匹配点自下而上的逐层优化,完成高精度特征匹配点对的提取。同时,针对模板匹配滑窗运算效率不高的问题,提出了一种密集模板匹配的快速实现方式,有效减少了密集模板匹配的运算时间。本文使用多组不同模态的遥感图像进行试验,结果表明,本文方法能够克服图像间非线性辐射差异的影响,在正确匹配数目、匹配准确率与匹配精度上均优于现有多模态图像特征匹配方法。

一种结合局部与半全局几何保持的影像匹配算法

遥感影像匹配是众多遥感应用中数据处理的关键前置步骤,但高程差导致的影像局部畸变和影像匹配的复杂性严重限制了高分辨率影像的匹配精度。提出一种适用于局部畸变和高外点比例的鲁棒匹配算法,首先利用Delaunay剖分算法在假定匹配点集上施加几何约束,得到特征点局部邻接关系;然后基于邻接信息进行预过滤;采用多尺度的策略建立局部邻接关系一致约束模型;最后定义三角形相似度函数实现匹配恢复。利用3组高分辨率影像开展对比实验,实验结果表明该算法的平均精度比RANSAC提高7.69%,在外点率高于90%时仍旧稳健。

基于红外影像层次旋转匹配的飞行器定位方法

基于图像匹配的飞行器自主视觉定位技术是飞行器导航制导、态势感知和自主决策的关键技术之一。针对红外遥感影像在大角度旋转下飞行器匹配定位失效的问题,提出了一种基于层次结构强化的特征点旋转匹配定位方法。该方法通过融合深度特征点提取与层次结构强化的旋转匹配定位技术,有效实现了飞行器匹配定位。首先设计了一个集成残差连接编码器的RBN-SuperPoint深度特征点提取模型,用于检测和描述待匹配图像中的深度特征点。其次构建了基于线性注意力和置信度分类器的L-LightGlue自适应匹配算法,利用L-LightGlue进行特征点粗匹配,生成单应性变换矩阵。随后采用层次结构强化的旋转匹配策略,根据粗匹配得到的单应性变换矩阵对图像进行旋转处理,消除图像间的角度差异,并进行精确匹配。再通过将结果映射至原图像,得到旋转校正后的特征点匹配结果和对应的单应性变换矩阵。最后利用图像间变换关系确定飞行器在图像中的位置,完成视觉定位。实验结果表明:RBN-SuperPoint可以高效提取大量均匀分布的特征点,所提基于L-LightGlue的匹配定位算法的匹配准确率最高可达98.57%,平均定位误差仅为4.08 pixel。

融合卷积和上下文变压器的遥感图像配准

针对卷积神经网络在遥感图像配准上鲁棒性弱、精度低的问题,提出一种结合残差网络和变压器并融入四重注意力的配准算法。提出一种卷积和遥感上下文变压器混合的网络结构,替换残差网络残差块并与其它预训练卷积层结合,以特征提取。将四重注意力机制融入特征提取网络,提高遥感图像区分性特征表示。设计双向匹配网络,采用皮尔逊互相关算法建立遥感图像之间的对应关系。实验结果表明,该模型在遥感图像配准多个评估指标下均优于其它模型。

A novel dense descriptor based on structure tensor voting for multi-modal image matching

Automatic and robust matching of multi-modal images can be challenging owing to the nonlinear intensity differences caused by radiometric variations among these images.To address this problem,a novel dense feature descriptor and improved similarity measure are proposed for enhancing the matching performance.The proposed descriptor is built on a voting scheme of structure tensor that can effectively capture the geometric structural properties of images.It is not only illumination and contrast invariant but also robust against the degradation caused by significant noise.Further,the similarity measure is improved to adapt to the reversal of orientation caused by the intensity inversion between multi-modal images.The proposed dense feature descriptor and improved similarity measure enable the development of a robust and practical templatematching algorithm for multi-modal images.We verify the proposed algorithm with a broad range of multi-modal images including optical,infrared,Synthetic Aperture Radar(SAR),digital surface model,and map data.The experimental results confirm its superiority to the state-of-the-art methods.

Generation of daily snow depth from multi-source satellite images and in situ observations

Snow depth （SD） is a key parameter for research into global climate changes and land surface processes. A method was developed to obtain daily SD images at a higher 4 km spatial resolution and higher precision with SD measurements from in situ observations and passive microwave remote sensing of Advanced Microwave Scanning Radiometer-EOS （AMSR-E） and snow cover measurements of the Interactive Multisensor Snow and Ice Mapping System （IMS）. AMSR-E SD at 25 km spatial resolution was retrieved from AMSR-E products of snow density and snow water equivalent and then corrected using the SD from in situ observations and IMS snow cover. Corrected AMSR-E SD images were then resampled to act as ＂virtual＂ in situ observations to combine with the real in situ observations to interpolate at 4 km spatial resolution SD using the Cressman method. Finally, daily SD data generation for several regions of China demonstrated that the method is well suited to the generation of higher spatial resolution SD data in regions with a lower Digital Elevation Model （DEM） but not so well suited to regions at high altitude and with an undulating terrain, such as the Tibetan Plateau. Analysis of the longer time period SD data generation for January between 2003 and 2010 in northern Xinjiang also demonstrated the feasibility of the method.

基于机器学习的激光遥感图像多目标检测方法

目前方法检测多目标激光遥感图像时,其未处理遥感图像的辐射量失真情况,导致检测方法的平均识别率、平均精度均值低和检测效率低。提出基于机器学习的激光遥感图像多目标检测方法,该方法处理激光遥感图像辐射量失真情况,通过校正辐射和辐射匹配获取到更清晰和真实的遥感图像,结合改进的原始Mask R-CNN网络,引入分级跳连法和K均值聚类算法,解决Mask R-CNN网络不适用于激光遥感图像检测的问题,采用改进的机器学习,实现激光遥感图像多目标检测。实验结果表明,所提方法检测出激光遥感图像中的全部目标,平均识别率为95.6%,目标检测时间为0.02 s,平均精度均值达到77.5%,因此,该方法有效提高了平均识别率、平均精度均值和检测效率。

基于改进SIFT算法的无人机遥感影像快速拼接

近几十年来,无人机遥感在地球观测领域发展十分迅速,然而,无人机影像快速拼接是阻碍其应用的难题.针对无人机遥感影像的特点与SIFT(scale invariant feature transform)拼接算法的缺点,该文提出了一种基于随机抽样一致性算法RANSAC(random sample consensus)和最小二乘匹配改进的SIFT影像拼接算法.首先采用随机采样法RANSAC进行粗略匹配数据的提纯,剔除伪特征点对,以减少特征点数量,降低时耗;然后再以最小二乘匹配进行更加精确的匹配,最终实现了无人机影像的自动拼接.实验结果表明:基于RANSAC和最小二乘匹配改进的SIFT拼接算法的平均正确匹配率为92.8%,拼接精度由1个像元提高到0.1个像元,同时拼接运算效率也得到了较大的提升.经改进的SIFT拼接算法在海量特征数据库中可以进行快速、准确的匹配、甚至可以达到实时的要求,具有更强的鲁棒性,可以满足低空遥感影像的相对定向高度自动化的需要,应用前景广阔.

星载光学遥感图像特征误匹配剔除方法研究

误匹配剔除是提高特征匹配精度的重要手段,星载光学遥感图像由于数据量大、纹理重复、光照强度变化等特点,导致现有误匹配剔除方法的性能下降。针对此问题,提出一种基于局部和全局几何约束的误匹配剔除方法。在初始匹配集的基础上,首先利用特征局部一致性对误匹配进行初步过滤;然后根据图像间的变换关系构建特征拓扑结构,并提取其几何属性描述结构相似度,基于此建立特征全局结构一致性约束模型,通过推导模型的最优解剔除残留误匹配;采用引导式匹配策略,选取局部一致性高的匹配点组成高内点率匹配集,以此作为特征全局邻域,提高全局约束的鲁棒性和效率。实验结果表明,与现有方法相比,所提方法对星载光学遥感图像的匹配性能更优,平均精确率、召回率分别为0.9和0.89;在不同内点率的初始匹配集上表现鲁棒,平均F分数为0.86。

一种改进的航天遥感影像数据清洗方法

针对航天遥感影像数据量大,存在重复和低质量影像,导致重复处理和存储,进而浪费计算、存储资源的问题,提出一种改进的航天遥感影像数据清洗方法。采用一种改进的DBSCAN聚类算法,基于遥感影像编目数据,对相似重复影像进行检测,形成相似重复记录集;对上一步确定的潜在相似重复影像执行影像匹配,根据匹配结果确判相似重复影像;最后,对相似重复影像进行质量评价,根据评价结果确定清洗策略,依据策略对数据进行清洗,剔除重复和低质量影像。可对海量原始遥感影像数据中重复、低质量数据进行准确、快速、有效清洗。实验结果表明,可有效识别相似重复遥感影像,并对影像进行准确分级,最终去除重复和低质量影像。