基于分层特征关联条件随机场的遥感图像分类

针对高分辨率遥感图像分类中空间上下文信息表达的难题,提出了一种新的多尺度条件随机场(CRF)模型。首先将图像内容表示成从细到粗三个超像素层:区域层、对象层、场景层,并将超像素特征逐层关联形成特征向量;再利用支持向量机(SVM)定义CRF关联势函数,利用相邻超像素特征对比度加权的Potts模型定义CRF交互势函数,最后形成一个分层特征关联的多尺度SVM-CRF模型。以Quickbird遥感图像中两个复杂场景为测试数据对该模型的分类有效性进行了验证,结果表明:该模型比基于上述三个超像素层的单尺度SVM-CRF模型分类精度分别平均提高了2.68%、1.66%、3.75%,而且分类时耗时较少。     

一种基于CRF-MR的自顶向下显著性目标检测方法*

针对自顶向下显著性目标检测边界模糊及准确率低的问题,提出一种结合条件随机场（conditional random field,CRF）和流行排序（manifold ranking,MR）的自顶向下显著性目标检测方法。首先对图像进行超像素分割,以超像素块特征为节点建立无向图;然后输入具有目标先验的CRF中得到节点的显著值并通过边缘背景先验MR修改显著值;最后扩展初步显著性目标得到最终显著性图。实验结果表明,在行人、汽车和自行车类目标检测中目标边界明确,与基于CRF的方法相比,本文方法在保证运算效率的同时具有更好的鲁棒性。     

基于目标识别与显著性检测的图像场景多对象分割

提出了一种基于目标识别与显著性检测的图像场景多对象分割方法。该方法的步骤包括:在图像训练集上训练语义对象的检测器,用来检测输入图像中对象的位置,标定对象的包围盒;对输入的图像进行过分割处理,得到超像素集合,根据包围盒的位置和超像素的语义概率值计算兴趣区域;在3种稠密尺度上进行场景显著性检测,得到输入图像的显著图;在兴趣区域内计算超像素的邻接关系,形成邻接矩阵,构建条件随机场模型,将多对象分割问题转化成多类别标记问题,每一个对象是一种类别;以每个超像素作为场模型的节点,超像素的邻接关系对应场模型中节点之间的连接关系,将显著性和图像特征转化为节点和边的权重值;利用图割算法,在条件随机场模型上进行优化,迭代终止时得到像素的对象标记结果,从而实现对多个对象的分割。实验结果表明该方法效果较好。     

超像素词包模型与SVM分类的图像标注

为了改善基于词包模型与支持向量机(SVM)分类一幅图对应一个标签的单标签分类问题,提出了一种基于超像素词包模型与SVM分类的图像标注算法.将超像素分割结果作为词包模型的基本单元,用词包模型生成的视觉词汇表示超像素区域特征,保留了图像中的同质区域,很好地利用了图像的区域特征.仿真结果表明,该方法能有效改善基于词包模型与SVM分类的单标签分类问题,且分类的准确性有所提高.     

基于多图流形排序的图像显著性检测

针对现有图像显著性检测算法中显著目标检测不完整和显著目标内部不均匀的问题,本文提出了一种基于多图流形排序的图像显著性检测算法.该算法以超像素为节点构造KNN图（K nearest neighbor graph）模型和K正则图（K regular graph）模型,分别在两种图模型上利用流形排序算法计算超像素节点的显著性值,并将每个图模型中超像素节点的显著值加权融合得到最终的显著图.在公开的MSRA-10K、SED2和ECSSD三个数据集上,将本文提出的算法与当前流行的14种算法进行对比,实验结果显示本文算法能够完整地检测出显著目标,并且显著目标内部均匀光滑.     

基于超像素的Grabcut彩色图像分割

针对以像素为节点建立图模型进行图像分割耗时的特点,文中提出了一种基于超像素的Grabcut彩色图像分割方法.首先用户在目标所在区域手动标定一个矩形框;然后用两次分水岭算法将图像过分割成区域内颜色相似的小区域(超像素),用分割得到的超像素作为图的结点构建图模型;以每个超像素的颜色均值代表所在分块的全部像素点估计GMM(高斯混合模型)参数;最后用最小割算法求得吉布斯能量的最小值达到最优分割.实验结果表明,该算法以极少数超像素代替海量像素,在得到较好分割结果的同时,极大地缩短了运行时间,加快了分割速度,提高了效率.     

基于超像素的人工神经网络图像分类

基于人工神经网络对图像标签分类,为简化后续数据处理,先用Normalized Cut将图像分割为超像素,提取特征向量,通过输入训练样本集,对网络进行训练,在最小均方误差意义下得到网络参数,最后在Matlab的仿真实验中基于不同隐藏层节点,使用BP神经网络模型对图像超像素进行分类。     

基于仿射不变离散哈希和条件随机场的遥感图像目标检测

建立了一种结合仿射不变离散哈希（Affined-invariant discrete hashing, AIDH）和条件随机场（Confidential random field, CRF）的模型,实现遥感图像的目标检测。对遥感图像进行超像素分割,构建适用于CRF的以超像素块为顶点的无向图结构。以超像素块作为测试样本,使用AIDH学习方法作为CRF一元势函数,生成初始类别标签。采用Potts模型构建CRF的二元势函数进行标签的再学习,平滑目标邻域信息,解决目标检测中的漏判问题。最后,使用基于凸壳边界的方法生成最小外接目标框作为目标检测结果。实验表明,本文方法在目标检测的精度和效率上取得了较好的平衡。     

自动标注技术概述

图像标签的自动标注技术是当前信息检索领域的热点问题。图像标注本质上是一个机器学习问题,该文对基于超像素的CRF图像分类标注和基于纠错输出编码的图像分类标注两类不同标签标注方法进行对比,阐述其优缺点,指出了实现低复杂度的图像分类标签标注方法的途径。     

基于SLIC的改进GrabCut彩色图像快速分割

GrabCut算法用户交互量少且分割精度高,但它迭代使用GraphCuts的求解模式使得在处理高分辨率图像时,耗时巨大。提出了一种快速GrabCut算法,在高斯混合模型参数估计过程中,通过SLIC算法构建精简的GraphCuts模型以实现加速。通过SLIC算法将原始图像快速地预分割成具有确定边界且区域内相似度高的超像素图,并以此构建精简的网络图。以块内的RGB均值描述超像素特征进行高斯混合模型参数估计。为了提高分割精度,使用得到的GMM参数对原始图像进行分割。实验结果证明了该算法在时效和精度上都有很好的性能。     

Superpixel-based active contour model for unsupervised change detection from satellite images

This study proposes a superpixel-based active contour model (SACM) for unsupervised change detection from satellite images. The accuracy of change detection produced by the traditional active contour model suffers from the trade-off parameter. The SACM is designed to address this limitation through the incorporation of the spatial and statistical information of superpixels. The proposed method mainly consists of three steps. First, the difference image is created with change vector analysis method from two temporal satellite images. Second, statistical region merging method is applied on the difference image to produce a superpixel map. Finally, SACM is designed based on the superpixel map to detect changes from the difference image. The SACM incorporates spatial and statistical information and retains the accurate shapes and outlines of superpixels. Experiments were conducted on two data sets, namely Landsat-7 Enhanced Thematic Mapper Plus and SPOT 5, to validate the proposed method. Experimental results show that SACM reduces the effects of the trade-off parameter. The proposed method also increases the robustness of the traditional active contour model for input parameters and improves its effectiveness. In summary, SACM often outperforms some existing methods and provides an effective unsupervised change detection method.     

结合深度学习与条件随机场的遥感图像分类

目的 为进一步提高遥感影像的分类精度,将卷积神经网络（CNN）与条件随机场（CRF）两个模型结合,提出一种新的分类方法。方法 首先采用CNN对遥感图像进行预分类,并将其类成员概率定义为CRF模型的一阶势函数;然后利用高斯核函数的线性组合定义CRF模型的二阶势函数,用全连接的邻域结构代替常见的4邻域或8邻域;接着加入区域约束,使用Mean-shift分割方法得到超像素,通过计算超像素的后验概率均值修正各像素的分类结果,鼓励连通区域结果的一致性;最后采用平均场近似算法实现整个模型的推断。结果 选用3组高分辨率遥感图像进行地物分类实验。本文方法不仅能抑制更多的分类噪声,同时还可以改善过平滑现象,保护各类地物的边缘信息。实验采用类精度、总体分类精度OA、平均分类精度AA,以及Kappa系数4个指标进行定量分析,与支持向量机（SVM）、CNN和全连接CRF相比,最终获得的各项精度均得到显著提升,其中,AA提高3.28个百分点,OA提高3.22个百分点,Kappa提高5.07个百分点。结论 将CNN与CRF两种模型融合,不仅可以获得像元本质化的特征,而且同时还考虑了图像的空间上下文信息,使分类更加准确,后加入的约束条件还能进一步保留地物目标的局部信息。本文方法适用于遥感图像分类领域,是一种精确有效的分类方法。     

BoVSG: bag of visual SubGraphs for remote sensing scene classification

ABSTRACT

Remote sensing scene classification is gaining much more interest in the recent few years for many strategic fields such as security, land cover and land use monitoring. Several methods have been proposed in the literature and they can be divided into three main classes based on the features used: handcrafted features, features obtained by unsupervised learning and those obtained from deep learning. Handcrafted features are generally time consuming and suboptimal. Unsupervised learning based features which have been proposed later gave better results but their performances are still limited because they mainly rely on shallow networks and are not able to extract powerful features. Deep learning based features are recently investigated and gave interesting results. But, they cannot be usually used because of the scarcity of labelled remote sensing images and are also computationally expensive. Most importantly, whatever kind of feature is used, the neighbourhood information of them is ignored. In this paper, we propose a novel remote sensing scene representation and classification approach called Bag of Visual SubGraphs (BoVSG). First, each image is segmented into superpixels in order to summarize the image content while retaining relevant information. Then, the superpixels from all images are clustered according to their colour and texture features and a random label is assigned to each cluster that probably corresponds to some material or land cover type. Thus superpixels belonging to the same cluster have the same label. Afterwards, each image is modelled with a graph where nodes correspond to labelled superpixels and edges model spatial neighbourhoods. Finally, each image is represented by a histogram of the most frequent subgraphs corresponding to land cover adjacency patterns. This way, local spatial relations between the nodes are also taken into account. Resultant feature vectors are classified using standard classification algorithms. The proposed approach is tested on three popular datasets and its performance outperforms state-of-the-art methods, including deep learning methods. Besides its accuracy, the proposed approach is computationally much less expensive than deep learning methods.     

Salient object detection method using random graph

In this paper, a bottom-up salient object detection method is proposed by modeling image as a random graph. The proposed method starts with portioning input image into superpixels and extracting color and spatial features for each superpixel. Then, a complete graph is constructed by employing superpixels as nodes. A high edge weight is assigned into a pair of superpixels if they have high similarity. Next, a random walk prior on nodes is assumed to generate the probability distribution on edges. On the other hand, a complete directed graph is created that each edge weight represents the probability for transmitting random walker from current node to next node. By considering a threshold and eliminating edges with higher probability than the threshold, a random graph is created to model input image. The inbound degree vector of a random graph is computed to determine the most salient nodes (regions). Finally, a propagation technique is used to form saliency map. Experimental results on two challenging datasets: MSRA10K and SED2 demonstrate the efficiency of the proposed unsupervised RG method in comparison with the state-of-the-art unsupervised methods.     

基于区域协方差的图像超像素生成

图像分割是指将图像分割成一些互不重叠的区域,各区域内部具有相同或相近的某些特定属性,而不同区域之间的属性则相差明显。在图像处理的许多应用中,由于像素级处理的方法因图像数据量庞大、运算规模较大而需要耗费大量的运行时间,因此对图像进行超像素分割预处理是很有必要的一个步骤。基于区域协方差分析,提出了一种新的像素块相似度度量方法;基于像素块相似度度量提出了一种图像超像素生成的鲁棒方法。该方法首先利用K-means算法对输入图像 进行初始聚类分割成若干小区域,对每个小区域利用区域协方差矩阵描述其特征信息;然后利用小区域块之间的区域协方差距离来构造相似度矩阵,结合Graph-based与K-means方法对区域块聚类生成图像超像素。与其它方法相比,该方法在生成较紧凑超像素的同时能更好地保持图像边缘特征信息,改善了图像欠分割错误,减少了不必要的过分割现象。将图像超像素生成方法应用于图像风格化中可以快速生成油画风格的风格化图像。     

Spatial-spectral neighbour graph for dimensionality reduction of hyperspectral image classification

Recently, graph embedding-based methods have drawn increasing attention for dimensionality reduction (DR) of hyperspectral image (HSI) classification. Graph construction is a critical step for those DR methods. Pairwise similarity graph is generally employed to reflect the geometric structure in the original data. However, it ignores the similarity of neighbouring pixels. In order to further improve the classification performance, both spectral and spatial-contextual information should be taken into account in HSI classification. In this paper, a novel spatial-spectral neighbour graph (SSNG) is proposed for DR of HSI classification, which consists of the following four steps. First, a superpixel-based segmentation algorithm is adopted to divide HSI into many superpixels. Second, a novel distance metric is utilized to reflect the similarity of two spectral pixels in each superpixel. In the third step, a spatial-spectral neighbour graph is constructed according to the above distance metric. At last, support vector machine with a composite kernel (SVM-CK) is adopted to classify the dimensionality-reduced HSI. Experimental results on three real hyperspectral datasets demonstrate that our method can achieve higher classification accuracy with relatively less consumed time than other graph embedding-based methods.     

基于完全联系的条件随机场的图像标注

传统的图像标注模型通常存在两个问题：只能够对短距离的像素上下文信息进行建模和复杂的模型推理过程。为了提高图像标注的精度、简化图像标注的模型推理过程,采用完全联系的条件随机场模型进行图像标注,提出利用基于高斯kd树的平均场估计方法实现该模型的高效推理。为了更好地验证算法的有效性,实验的图片数据库不仅包含标准的图片库--剑桥大学微软研究图片库（MSRC-9）,还包含作者制作的机械零件图片库（MyDataset_1）和办公桌图片库（MyDataset_2）。新算法在三个图片库上的平均标注精度分别可以达到77.96%、97.15%和95.35%,每幅图的平均运行时间为2s。实验结果表明,基于完全联系的条件随机场的图像标注能够更充分地考虑不同的像素上下文信息来提高标注精度,而基于高斯kd树的模型推理能够提高模型推理的效率。     

基于SLIC和主动学习的高光谱遥感图像分类方法

在主动学习的基础上,提出一种基于SLIC的高光谱遥感图像主动分类方法。首先提取图像纹理特征并与光谱特征融合,使用PCA对新数据进行降维,取前三个主成分构成假彩色图像,然后使用SLIC处理该图像获得超像素;接着随机抽取定量超像素作为初始训练样本,样本光谱信息为超像素样本中所有像素点的光谱信息均值,样本标签为超像素中出现次数最多的类别;然后通过主动学习得到SVM分类器;最后使用分类器对超像素分类得到其类别,并将超像素类别赋予其包含的像素点,从而达到高光谱遥感图像分类的目的。实验表明：该方法明显降低了主动学习过程的时间消耗,有效地提高了分类效果,其OA,AA和Kappa值显著优于未使用SLIC的主动学习方法。     

背景吸收的马尔可夫显著性目标检测

目的 现有的基于马尔可夫链的显著目标检测方法是使用简单线性迭代聚类（SLIC）分割获得超像素块构造图的节点,再将四边界进行复制获得吸收节点,但是SLIC分割结果的好坏直接影响到后续结果,另外,很大一部分图像的显著目标会占据12个边界,特别是对于人像、雕塑等,如果直接使用四边界作为待复制的节点,必然影响最终效果。针对以上存在的缺陷,本文提出一种背景吸收的马尔可夫显著目标检测方法。方法 首先通过差异化筛选去除差异较大的边界,选择剩余3条边界上的节点进行复制作为马尔可夫链的吸收节点,通过计算转移节点的吸收时间获得初始的显著图,从初始显著图中选择背景可能性较大的节点进行复制作为吸收节点,再进行一次重吸收计算获得显著图,并对多层显著图进行融合获得最终的显著图。 结果 在ASD、DUT-OMRON和SED 3个公开数据库上,对比实验验证本文方法,与目前12种主流算法相比,在PR曲线、F值和直观上均有明显的提高,3个数据库计算出的F值分别为0.903、0.544 7、0.775 6,验证了算法的有效性。结论 本文针对使用图像边界的超像素块复制作为吸收节点和SLIC分割技术的缺陷,提出了一种基于背景吸收马尔可夫显著目标检测模型,实验表明,本文的方法适用于自底向上的图像显著目标检测,特别是适用于存在人像或雕塑等目标的图像,并且可以应用于图像检索、目标识别、图像分割和图像压缩等多个领域。