An object-based and heterogeneous segment filter convolutional neural network for high-resolution remote sensing image classification

In recent years, object-based segmentation methods and shallow-model classification algorithms have been widely integrated for remote sensing image supervised classification. However, as the image resolution increases, remote sensing images contain increasingly complex characteristics, leading to higher intraclass heterogeneity and interclass homogeneity and thus posing substantial challenges for the application of segmentation methods and shallow-model classification algorithms. As important methods of deep learning technology, convolutional neural networks (CNNs) can hierarchically extract higher-level spatial features from images, providing CNNs with a more powerful recognition ability for target detection and scene classification in high-resolution remote sensing images. However, the input of the traditional CNN is an image patch, the shape of which is scarcely consistent with a given segment. This inconsistency may lead to errors when directly using CNNs in object-based remote sensing classification: jagged errors may appear along the land cover boundaries, and some land cover areas may overexpand or shrink, leading to many obvious classification errors in the resulting image. To address the above problem, this paper proposes an object-based and heterogeneous segment filter convolutional neural network (OHSF-CNN) for high-resolution remote sensing image classi?cation. Before the CNN processes an image patch, the OHSF-CNN includes a heterogeneous segment filter (HSF) to process the input image. For the segments in the image patch that are obviously different from the segment to be classified, the HSF can differentiate them and reduce their negative influence on the CNN training and decision-making processes. Experimental results show that the OHSF-CNN not only can take full advantage of the recognition capabilities of deep learning methods but also can effectively avoid the jagged errors along land cover boundaries and the expansion/shrinkage of land cover areas originating from traditional CNN structures. Moreover, compared with the traditional methods, the proposed OHSF-CNN can achieve higher classification accuracy. Furthermore, the OHSF-CNN algorithm can serve as a bridge between deep learning technology and object-based segmentation algorithms thereby enabling the application of object-based segmentation methods to more complex high-resolution remote sensing images.     

噪声鲁棒的轻量级深度遥感场景图像分类检索

目的 基于深度神经网络的遥感图像处理方法在训练过程中往往需要大量准确标注的数据,一旦标注数据中存在标签噪声,将导致深度神经网络性能显著降低。为了解决噪声造成的性能下降问题,提出了一种噪声鲁棒的轻量级深度遥感场景图像分类检索方法,能够同时完成分类和哈希检索任务,有效提高深度神经网络在有标签噪声遥感数据上的分类和哈希检索性能。方法 选取轻量级神经网络作为骨干网,而后设计能够同时完成分类和哈希检索任务的双分支结构,最后通过设置损失基准的正则化方法,有效减轻模型对噪声的过拟合,得到噪声鲁棒的分类检索模型。结果 本文在两个公开遥感场景数据集上进行分类测试,并与8种方法进行比较。本文方法在AID（aerial image datasets）数据集上,所有噪声比例下的分类精度比次优方法平均高出7.8%,在NWPU-RESISC45（benchmark created by Northwestern Polytechnical University for remote sensing image scene classification covering 45 scene classes）数据集上,分类精度比次优方法平均高出8.1%。在效率方面,本文方法的推理速度比CLEOT（classification loss with entropic optimal transport）方法提升了2.8倍,而计算量和参数量均不超过CLEOT方法的5%。在遥感图像哈希检索任务中,在AID数据集上,本文方法的平均精度均值（mean average precision,mAP）在3种不同哈希比特下比MiLaN（metric-learning based deep hashing network）方法平均提高了5.9%。结论 本文方法可以同时完成遥感图像分类和哈希检索任务,在保持模型轻量高效的情况下,有效提升了深度神经网络在有标签噪声遥感数据上的鲁棒性。     

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。     

深度学习在高分遥感图像场景分类中的应用

高分遥感图像场景分类致力于自动辨别土地利用或覆盖的类别,在军事和国土资源勘探等领域具有重要的应用价值。基于深度学习的高分遥感图像场景分类方法取得了比传统方法更好的效果,也是当前研究的热点,对此类方法进行归纳总结和综合评估。按照监督方式的不同,对基于深度学习的流行方法进行了逐类分析。对不同监督方式下的流行方法在三个公开数据集上进行了定量实验评估。总结了基于不同监督方式方法的特点,并对下一步发展趋势进行了展望。     

基于U-Net的高分辨率遥感图像土地利用信息提取

随着现代遥感技术的迅速发展,遥感图像的质量和数量得到了显著的提升,新技术带来的高分辨率遥感图像所蕴含的信息也更加丰富,如何利用人工智能手段辅助挖掘这些丰富的信息也成为了遥感图像分析与理解的重要内容.与此同时,以深度卷积神经网络为代表的人工智能技术在图像处理领域大放异彩.得益于类人眼的分层卷积池化模型,深度卷积神经网络可...     

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。     

U-Net Neural Networks and Its Application in High Resolution Satellite Image Classification

High-resolution remote sensing images have precise geometric structure and spatial layout, but the spectral information is limited, which increases the difficulty of classifying similar features of spectral features. Aiming at the problem of high resolution remote sensing image classification, a U-Net convolutional neural network classification method based on deep learning is proposed. Based on the remote sensing image of the Ejina Oasis GF-2 in the lower reaches of the Heihe River, the U-Net model was used to extract the five types of land cover types of Populus euphratica, Tamarix chinensis, cultivated land, grassland and bare land. The overall classification accuracy and Kappa coefficient were 85.024% and 0.795 6 respectively. Compared with the traditional Support Vector Machine(SVM) and object-oriented method, the results show that compared with SVM and object-oriented method, the U-Net model is used to classify the high-resolution remote sensing, and the classification effect is better. The ground extracts the essential features of the features to meet the accuracy requirements.     

深度学习U-Net方法及其在高分辨卫星影像分类中的应用

高分辨率遥感影像有精确的几何结构和空间布局,但是光谱信息有限,增大了对光谱特征相似地物的分类难度。针对高分辨率遥感影像分类的问题,采用深度学习U-Net模型分类方法。基于黑河下游额济纳绿洲高分二号遥感影像,通过U-Net模型提取胡杨、柽柳、耕地、草地和裸地五种地物覆被类型,分类总体精度和Kappa系数分别为85.024%和0.795 6,并与传统的支持向量机（SVM, Support Vector Machine）和面向对象的分类方法比较,结果表明：相对于SVM和面向对象,基于U-Net模型的高分辨率卫星影像地物覆被分类,能够更好地对地物本质特征进行提取,分类效果较好,满足精度要求。     

基于多特征融合与CELM的场景分类算法

场景分类对于场景图像的语义解译至关重要,是遥感领域近期的主要研究课题之一。针对大部分图像分类方法中提取的特征结构单一,依赖于大量人工标记的数据以及分类器的训练过程缓慢等问题,提出了一种基于多特征融合与约束极限学习机（constrained extreme learning machines,CELM）的场景图像分类方法。该方法采用三种不同结构的预训练卷积神经网络,利用特定数据集对其进行微调,将微调后网络提取到的三种特征进行融合并送入CELM分类器进行分类,最终得到图像的类别标签。以SIRI-WHU、WHU-RS19与UC-Merced数据集作为实验数据集,在预训练卷积神经网络、单一特征和传统分类器上进行的对比实验表明,基于多特征融合与CELM相结合的方法产生了较好的分类效果,三种数据集上的总分类精度分别高达99.25%、98.26%与97.70%。     

基于有效通道注意力的遥感图像场景分类

针对基于人工设计特征的方法不能提取高层次遥感图像信息以及以往利用VGGNet、ResNet等卷积神经网络（CNN）无法关注到遥感图像中显著分类特征的问题,提出了一种基于有效通道注意力（ECA）机制的遥感图像场景分类新模型——ECA-ResNeXt-8-SVM。为了建立高效模型,一方面,设计了嵌入ECA模块的深度特征提取网络ECA-ResNeXt-8,通过端到端的学习使网络更关注分类特征明显的通道;另一方面,利用支持向量机（SVM）代替全连接层作为已提取到的深度特征的分类器,从而进一步提高模型的分类准确率与泛化能力。该模型在实验数据集UC Merced Land-Use上的分类准确率达到95.81%,相较于使用SE-ResNeXt50与ResNeXt50网络,分别提高了6%与18%,且在分类准确率为75%时所提模型的训练时间比上述两个网络分别减少了82%与81%。实验结果表明,所提模型能够有效地减少模型的收敛时间并提升遥感图像场景分类的准确率。     

Using random walker for knowledge transfer in classifying multi-temporal VHR images

Acquiring land cover types from very high resolution (VHR) images is of great significance to many applications and has been intensively studied for many years. The difficulties in image classification and the high frequencies of remote sensing image acquisition make it urgent to develop efficient knowledge transfer approaches for understanding multi-temporal VHR images. This letter proposed a knowledge transfer approach that uses the label information of the existing VHR images to classify multi-temporal images. The approach was implemented in three steps: object-based change detection, knowledge transfer of label information, and random walker (RW) classification. The proposed approach was tested by two datasets with each having two temporal images acquired on the same geographical areas. The experimental results showed that the proposed approach outperformed the support vector machine (SVM) algorithm in classifying multi-temporal images and can reduce the influence of spectral confusions on image classification.     

基于模糊高斯基函数神经网络的遥感图像分类

针对遥感图像分类的特点,提出了一种基于模糊高斯基函数神经网络的遥感图像分类器。该分类器将模糊技术与神经网络相结合,采用神经网络来实现模糊推理,利用神经网络的学习能力来达到调整模糊隶属函数和模型规则的目的,从而使系统具备了自适应的特性,实验结果表明,这种基于模糊高斯基孙数神经网络的分类器经过训练后,可应用于遥感图像的分类,其分类精度明显高于传统的最大似然分类法。     

Growing random forest on deep convolutional neural networks for scene categorization

Breakthrough performances have been achieved in computer vision by utilizing deep neural networks. In this paper we propose to use random forest to classify image representations obtained by concatenating multiple layers of learned features of deep convolutional neural networks for scene classification. Specifically, we first use deep convolutional neural networks pre-trained on the large-scale image database Places to extract features from scene images. Then, we concatenate multiple layers of features of the deep neural networks as image representations. After that, we use random forest as the classifier for scene classification. Moreover, to reduce feature redundancy in image representations we derived a novel feature selection method for selecting features that are suitable for random forest classification. Extensive experiments are conducted on two benchmark datasets, i.e. MIT-Indoor and UIUC-Sports. Obtained results demonstrated the effectiveness of the proposed method. The contributions of the paper are as follows. First, by extracting multiple layers of deep neural networks, we can explore more information of image contents for determining their categories. Second, we proposed a novel feature selection method that can be used to reduce redundancy in features obtained by deep neural networks for classification based on random forest. In particular, since deep learning methods can be used to augment expert systems by having the systems essentially training themselves, and the proposed framework is general, which can be easily extended to other intelligent systems that utilize deep learning methods, the proposed method provide a potential way for improving performances of other expert and intelligent systems.     

Multiscale attention for few-shot image classification

In recent years, the application of traditional deep learning methods in the agricultural field using remote sensing techniques, such as crop area and growth monitoring, crop classification, and agricultural disaster monitoring, has been greatly facilitated by advancements in deep learning. The accuracy of image classification plays a crucial role in these applications. Although traditional deep learning methods have achieved significant success in remote sensing image classification, they often involve convolutional neural networks with a large number of parameters that require extensive optimization using numerous remote sensing images for training purposes. To address these challenges, we propose a novel approach called multiscale attention network (MAN) for sample-based remote sensing image classification. This method consists primarily of feature extractors and attention modules to effectively utilize different scale features through multiscale feature training during the training phase. We evaluate our proposed method on three datasets comprising agricultural remote sensing images and observe superior performance compared to existing approaches. Furthermore, we validate its generalizability by testing it on an oil well indicator diagram specifically designed for classification tasks.     

A Hierarchical Classification and Iterative Model based Methodfor Remote Sensing Classification of Land Cover

Land cover classification based on remote sensing is an important means to analyze the change and spatial pattern of land use.In order to further improve the classification accuracy,this paper proposed a hierarchical classification and iterative CART model based method for remote sensing classification of landcover.Firstly,the extraction order of land cover classes was determined based on the class separability evaluation,which was water,vegetation,bare soil and built-up land.Secondly,we selected the optimal image segmentation parameters and a set of sensitive features for each class during the hierarchical classification process.Finally,object-based training samples were selected to be fed into the iterative CART algorithm for the successive extraction of the first three classes,with the remaining unclassified objects being directly assigned to the last class.Results demonstrated that the proposed method can significantly reduce the mixture between bare soil and built-up land,and is capable of achieving landcover classification with much higher accuracy.The proposed method achieved an overall accuracy of 85.76% and a Kappa efficient of 0.72,with the performance improvements ranging from 10.67% to 16.5% and 0.15 to 0.21 as compared SVM and CART single classification methods.The classification accuracy of a specific class can be flexibly adjusted using this method,giving different purposes of classification.This method can also be easily extended to other districts and disciplines involving remote sensing image classification.     

对抗学习遥感图像场景识别

目的 在高分辨率遥感图像场景识别问题中,经典的监督机器学习算法大多需要充足的标记样本训练模型,而获取遥感图像的标注费时费力。为解决遥感图像场景识别中标记样本缺乏且不同数据集无法共享标记样本问题,提出一种结合对抗学习与变分自动编码机的迁移学习网络。方法 利用变分自动编码机（variational auto-encoders,VAE）在源域数据集上进行训练,分别获得编码器和分类器网络参数,并用源域编码器网络参数初始化目标域编码器。采用对抗学习的思想,引入判别网络,交替训练并更新目标域编码器与判别网络参数,使目标域与源域编码器提取的特征尽量相似,从而实现遥感图像源域到目标域的特征迁移。结果 利用两个遥感场景识别数据集进行实验,验证特征迁移算法的有效性,同时尝试利用SUN397自然场景数据集与遥感场景间的迁移识别,采用相关性对齐以及均衡分布适应两种迁移学习方法作为对比。两组遥感场景数据集间的实验中,相比于仅利用源域样本训练的网络,经过迁移学习后的网络场景识别精度提升约10%,利用少量目标域标记样本后提升更为明显;与对照实验结果相比,利用少量目标域标记样本时提出方法的识别精度提升均在3%之上,仅利用源域标记样本时提出方法场景识别精度提升了10%~40%;利用自然场景数据集时,方法仍能在一定程度上提升场景识别精度。结论 本文提出的对抗迁移学习网络可以在目标域样本缺乏的条件下,充分利用其他数据集中的样本信息,实现不同场景图像数据集间的特征迁移及场景识别,有效提升遥感图像的场景识别精度。     

基于多尺度特征多对抗网络的雾天图像识别

目的 当前的大型数据集,例如ImageNet,以及一些主流的网络模型,如ResNet等能直接高效地应用于正常场景的分类,但在雾天场景下则会出现较大的精度损失。雾天场景复杂多样,大量标注雾天数据成本过高,在现有条件下,高效地利用大量已有场景的标注数据和网络模型完成雾天场景下的分类识别任务至关重要。方法 本文使用了一种低成本的数据增强方法,有效减小图像在像素域上的差异。基于特征多样性和特征对抗的思想,提出多尺度特征多对抗网络,通过提取数据的多尺度特征,增强特征在特征域分布的代表性,利用对抗机制,在多个特征上减少特征域上的分布差异。通过缩小像素域和特征域分布差异,进一步减小领域偏移,提升雾天场景的分类识别精度。结果 在真实的多样性雾天场景数据上,通过消融实验,使用像素域数据增强方法后,带有标签的清晰图像数据在风格上更趋向于带雾图像,总的分类精度提升了8.2%,相比其他的数据增强方法,至少提升了6.3%,同时在特征域上使用多尺度特征多对抗网络,相比其他的网络,准确率至少提升了8.0%。结论 像素域数据增强以及多尺度特征多对抗网络结合的雾天图像识别方法,综合考虑了像素域和特征域的领域分布差异,结合了多尺度的丰富特征信息,同时使用多对抗来缩小雾天数据的领域偏移,在真实多样性雾天数据集上获得了更好的图像分类识别效果。     

基于图卷积网络的多标签遥感图像分类

由于遥感图像包含物体类别多样,单个语义类别标签无法全面地描述图像内容,而多标签图像分类任务更加具有挑战性.通过探索深度图卷积网络(GCN),解决了多标签遥感图像分类缺乏对标签语义信息相关性利用的问题,提出了一种新的基于图卷积的多标签遥感图像分类网络,它包含图像特征学习模块、基于图卷积网络的分类器学习模块和图像特征差异化模块三个部分.在公开多标签遥感数据集Planet和UCM上与相关模型进行对比,在多标签遥感图像分类任务上可以得到了较好的分类结果.该方法使用图卷积等模块将多标签图像分类方法应用到遥感领域,提高了模型分类能力,缩短了模型训练时间.     

联合集成学习与EfficientNet的光学遥感图像场景分类

深度学习能够提高光学遥感图像场景分类的准确率和效率,但光学遥感图像语义丰富,部分场景仍存在易误分类的情况,同时由网络模型规模扩大带来的硬件要求过高、时间成本消耗过大等问题制约着深度学习网络模型的推广应用。为此,提出一种基于轻量化网络模型的光学遥感图像场景分类方法。通过EfficientNet网络提取图像特征,对图像特征进行复合提取以生成语义信息更丰富的新特征,利用多个子分类器构建集成学习模块解析新特征得到预分类结果,集成加权预分类结果以获得最终的分类结果。在AID和NWPU-RESISC45数据集上的实验结果表明,即使只训练20%的数据样本,该方法也能分别达到94.32%和93.36%的准确率,相对D-CNNs、CNN-CapsNet等方法,所提方法对易误分类场景有更好的分类效果,且参数量和浮点运算量大幅减少。