结合自适应空间与CRF的遥感影像变化检测

提出了一种结合自适应空间与条件随机场（CRF）的新框架用于高分辨率遥感影像变化检测,解决了两幅影像配准误差造成的噪声和条件随机场产生的过平滑问题。通过加权自适应空间（WAS）获取差异图像消除了因配准误差造成的部分噪声,采用基于形态学重构的FCM聚类方法（MFCM）构建CRF一阶势减少了斑点噪声;引入带光谱—空间约束的模型改进CRF二阶势,进一步提高了算法的抗噪性能并可防止过平滑现象。与现有方法相比,该算法的检测精度、虚检率和漏检率都得到明显改进,同时较好地保留了边缘信息。     

Remote sensing image fusion based on Bayesian linear estimation

A new remote sensing image fusion method based on statistical parameter estimation is proposed in this paper. More specially, Bayesian linear estimation (BLE) is applied to observation models between remote sensing images with different spa- tial and spectral resolutions. The proposed method only estimates the mean vector and covariance matrix of the high-resolution multispectral (MS) images, instead of assuming the joint distribution between the panchromatic (PAN) image and low-resolution multispectral image. Furthermore, the proposed method can enhance the spatial resolution of several principal components of MS images, while the traditional Principal Component Analysis (PCA) method is limited to enhance only the first principal component. Experimental results with real MS images and PAN image of Landsat ETM demonstrate that the proposed method performs better than traditional methods based on statistical parameter estimation, PCA-based method and wavelet-based method.     

自适应PCNN和改进C-V结合的遥感图像变化检测

为获取保留图像信息较完好的差异图并得到更好的变化检测结果,提出一种基于自适应脉冲耦合神经网络（PC‐NN）和改进Chan‐Vese （C‐V）模型的非监督的不同时相遥感图像的变化检测算法。用差值法、比值法对两幅遥感图像进行差异图获取;用自适应PCNN图像融合算法对两幅差异图进行融合,获取保留图像信息较好的差异图;用基于改进C‐V模型的分割算法对融合后的差异图进行分割,得到变化检测结果图。实验结果表明,该算法具有很好的变化检测效果,总检测精度较高。     

基于贝叶斯网络的Android恶意行为检测方法

Android操作系统是市场占有率最高的移动操作系统,基于Android平台的恶意软件也呈现爆发式的增长,而目前仍然没有有效的手段进行Android恶意行为的检测,通过分析Android恶意行为的特点,采用基于贝叶斯网络的机器学习算法进行Android恶意行为的检测,通过静态分析的方法进行Android文件静态特征的提取,将Android恶意应用的静态分析与贝叶斯网络相结合,最后通过使用提出的方法构建贝叶斯网络模型,通过实验验证了提出的Android恶意行为检测模型的有效性。     

基于改进YOLOv5s的遥感目标检测

优化遥感目标检测对于军事民生方面都有着重要意义。由于遥感数据中成像模糊、目标较小、检测对象数量多，导致检测精度不高，现提出一种新网络：新网络在YOLOv5s（You Only Look Once v5s）原网络的基础上使用Mish激活函数代替SiLU（Sigmoid Weighted Liner Unit）激活函数；为解决遥感图像中小目标的问题，采用了对小目标、低分辨率友好的SPD-Conv（Space-to-depth-Conv）模块；考虑到使用耦合检测头会存在回归、分类两个任务之间的冲突问题，采用YOLOX (You Only Look Once X)中解耦的检测头，提高了模型查准率。实验结果表明，相比于原始YOLOv5s，新网络在mAP（mean Average Precision）平均精度均值方面提升了7%，查全率（recall）提升了10.9%，检测速度FPS提升了16.95%。改进后的网络模型相对于原始模型具有明显优越性，识别效果提升明显。     

基于FPGA的遥感数据采集与快视系统

提出了一种基于现场可编程门阵列（FPGA）的实时遥感卫星数据采集与快视系统的设计方案,该方案充分利用FPGA的硬件资源简化逻辑设计,并采用总线复用技术和内存映射文件的方法实现了采集和快视的实时并行处理以及海量数据的高速存取。     

Cover: Remote sensing based detection of defensive minefields

Crop surface temperature (CST) is an important parameter to monitor crop status. Satellite data in thermal region provide an opportunity to estimate CST over large regions at frequent intervals. In the present study, various split‐window algorithms are employed to estimate CST over rice areas in irrigation projects of Krishna basin, South India using multi‐resolution MODIS satellite data. NDVI is used to approximate the mean pixel emissivity, by taking known values for emissivity and NDVI for pure vegetation and bare soil pixels. Diurnal ground measurements are made to evaluate satellite‐derived CST. CST values obtained using the Sobrino method have been found to be closer to the ground‐measured values compared with other algorithms, as it takes into account view angle, atmospheric transmittance, and water vapour corrections. It has been observed that the error in estimating CST is comparatively lower for well‐grown crops.     

基于卡方检验的贝叶斯网络入侵检测的分析

朴素贝叶斯(NB)入侵检测没有考虑其入侵行为所涉及的数据属性间的差别.引入卡方检验改进传统的NB模型,利用它来对网络连接数据的属性进行特征选择,并删除一些冗余的属性,达到优化NB入侵检测模型的目的.实验结果表明,卡方检验对NB模型有一定的优化作用,相对神经网络模型有更高的检测率.     

Bayesian networks based rare event prediction with sensor data

A Bayesian network is a powerful graphical model. It is advantageous for real-world data analysis and finding relations among variables. Knowledge presentation and rule generation, based on a Bayesian approach, have been studied and reported in many research papers across various fields. Since a Bayesian network has both causal and probabilistic semantics, it is regarded as an ideal representation to combine background knowledge and real data. Rare event predictions have been performed using several methods, but remain a challenge. We design and implement a Bayesian network model to forecast daily ozone states. We evaluate the proposed Bayesian network model, comparing it to traditional decision tree models, to examine its utility.     

基于深度学习的遥感图像目标检测与识别

为解决目前的遥感图像目标检测算法存在的对小尺度目标检测精度低和检测速度慢等问题,提出了一种基于深度学习的遥感图像目标检测与识别算法。首先,构建一个含有不同尺度大小的遥感图像的数据集用于模型的训练和测试;其次,基于原始的多尺度单发射击（SSD）网络模型,融入了设计的浅层特征融合模块、浅层特征增强模块和深层特征增强模块;最后,在训练策略上引入聚焦分类损失函数,以解决训练过程中正负样本失衡的问题。在高分辨率遥感图像数据集上进行实验,结果表明所提算法的检测平均精度均值（mAP）达到77.95%,相较于SSD网络模型提高了3.99个百分点,同时检测速度为33.8 frame/s。此外,在拓展实验中,改进算法对高分辨率遥感图像中模糊目标的检测效果也优于原多尺度单发射击网络模型。实验结果说明,所提改进算法能够有效地提高遥感图像目标检测的精度。     

Multi-sensor remote sensing image change detection based on sorted histograms

Change detection using multi-sensor remote sensing images, such as synthetic aperture radar (SAR) and optical images, is poorly researched and thus remains a challenging task. In this study, we address this problem by proposing a novel automatic change detection method. Different sensors have completely different physical principles. Thus, the resulting multi-sensor images have completely different radiometric values. First, we introduce a sorted histogram concept that sorts the bins in descending order, noticing that multi-sensor images with absence of change have the same combination of objects, and each object in different images has the same proportions and a unique range of grey values. The sorted histogram discards the visual property correspondence between images and is capable of capturing the local internal image layout. Then, various histogram-based distances are employed to estimate the distance between each sorted histogram pair. After the whole image has been analysed, we obtain a divergence index map. The sorted histogram not only has the theoretical advantage of robustness in the intensity variations in multi-sensor images but also the practical advantage of low computational complexity compared with existing methods. Experiments on SAR and optical datasets with different resolutions show promising results in terms of detection capability and run time.     

基于深度学习的遥感图像目标检测与识别

为解决目前的遥感图像目标检测算法存在的对小尺度目标检测精度低和检测速度慢等问题，提出了一种基于深度学习的遥感图像目标检测与识别算法。首先，构建一个含有不同尺度大小的遥感图像的数据集用于模型的训练和测试；其次，基于原始的多尺度单发射击（SSD）网络模型，融入了设计的浅层特征融合模块、浅层特征增强模块和深层特征增强模块；最后，在训练策略上引入聚焦分类损失函数，以解决训练过程中正负样本失衡的问题。在高分辨率遥感图像数据集上进行实验，结果表明所提算法的检测平均精度均值（mAP）达到77.95%，相较于SSD网络模型提高了3.99个百分点，同时检测速度为33.8 frame/s。此外，在拓展实验中，改进算法对高分辨率遥感图像中模糊目标的检测效果也优于原多尺度单发射击网络模型。实验结果说明，所提改进算法能够有效地提高遥感图像目标检测的精度。     

基于改进YOLOv3的遥感图像小目标检测

YOLOv3算法被广泛地应用于目标检测任务。虽然在YOLOv3基础上改进的一些算法取得了一定的成果,但是仍存在表征能力不足且检测精度不高的问题,尤其对小目标的检测还不能满足需求。针对上述问题,提出了一种改进YOLOv3的遥感图像小目标检测算法。首先,使用K均值聚类变换（K-means-T）算法优化锚框的大小,从而提升先验框和真实框之间的匹配度;其次,优化置信度损失函数,以解决难易样本分布不均衡的问题;最后,引入注意力机制来提高算法对细节信息的感知能力。在RSOD数据集上进行实验的结果显示,与原始的YOLOv3算法、YOLOv4算法相比,所提算法在小目标“飞机（aircraft）”类上的平均精确率（AP）分别提高了7.3个百分点和5.9个百分点。这表明所提算法能够有效检测遥感图像小目标,具有更高的准确率。     

基于深度网络的遥感图像目标检测与识别

本文针对卫星遥感图像中的典型目标检测与识别问题,采用了基于深度神经网络的回归算法,利用深度网络所学习出来的特征结构针对多分类复杂场景具有很好的识别效果,同时将检测、分类与识别一同进行回归处理。可以有效缩短训练时间,并提升处理效率,保证目标检测的准确度,在实际应用中前景广泛。     

Bayesian change detection based on spatial sampling and Gaussian mixture model

A Gaussian mixture model (GMM) and Bayesian inferencing based unsupervised change detection algorithm is proposed to achieve change detection on the difference image computed from satellite images of the same scene acquired at different time instances. Each pixel of the difference image is represented by a feature vector constructed from the difference image values of the neighbouring pixels to consider the contextual information. The feature vectors of the difference image are modelled as a GMM. The conditional posterior probabilities of changed and unchanged pixel classes are automatically estimated by partitioning GMM into two distributions by minimizing an objective function. Bayesian inferencing is then employed to segment the difference image into changed and unchanged classes by using the conditional posterior probability of each class. Change detection results are shown on real datasets.     

基于旋转中心点估计的遥感目标精确检测算法

由于遥感图像背景复杂、目标密集分布以及目标尺度、形状差异巨大,给检测带来挑战.当前基于R-CNN的两阶段算法在水平框(HBB)检测上取得了良好效果,然而在定向框(OBB)检测上效果有限.基于点估计的HBB目标检测框架,提出用于定向遥感目标检测的旋转中心点估计网络(RCNet),大幅提升一阶段anchor-free算法在倾斜目标检测上的性能,同时保持较高的检测速度.RCNet通过添加一个用于方向预测的分支,实现旋转中心点估计.提出新的角度表示方式,解决回归角度参数loss不连续以及宽高交换导致训练过程不稳定的问题.所提方法在DOTA数据集上取得66.68 mAP的检测精度以及29.4 fps的检测速度,实现了最佳的速度和精度平衡.     

应用压缩感知的无线传感器网络数据处理综述

研究了压缩感知在无线传感器网络数据处理方面的应用。介绍了压缩感知技术和无线传感器网络的发展及研究现状,并从数据融合、信号采集、信号路由传输以及信号重构4个方面,对近年来基于压缩感知的无线传感器网络数据处理研究进行了详尽的分析,提出数据安全的重要性。总结并展望了压缩感知技术未来的研究方向。     

基于3MAD-PCA的软测量数据过失误差侦破

经典PCA是一种对软测量建模数据进行误差侦破的方法,但当数据中存在单变量大误差时,该方法不能准确确定主元(PC),从而影响了误差侦破效果.针对这一情况,结合单变量误差侦破技术提出了3MAD-PCA方法.该方法首先用3MAD对数据分别进行单变量误差侦破,再利用经典PCA进行多变量误差侦破,提高了经典PCA方法的稳定性,有效实现了数据的过失误差侦破.用该方法对丙烯浓度的软测量数据进行过失误差侦破,取得了良好的效果.