Remote Sensing Image Interpretation Research Articles

Quantitative Study on Morphological Characteristics of Barchan Dunes in Yamarak Desert, China

Although scholars have conducted many studies on barchan dunes in deserts and sandy areas around the world, few studies have been conducted on the morphological characteristics and changes in movement patterns of barchan dunes in the Yamarak Desert, China. To assess the changes in and movement patterns of these dunes, As well as the impact on the wind-sand hazards and geohazards around the study area, we selected several typical barchan dunes of different sizes in the Yamarak Desert and measured and calculated their morphologies using a combination of a 3-D laser scanner (RIEGL VZ-2000) and remote sensing image interpretation. The results show that the average coverage of the barchan dunes in the Yamarak Desert was 37.19%. Moreover, 62.81% of this area was covered by interdune sites. The movement velocity of the barchan dunes in this study area was 0–20 m/a, with an average movement velocity of 8.45 m/a. Of these dunes, 44.18% were fast-dominated dunes, 37.20% were particularly fast-dominated dunes, and 18.62% were moderately fast-dominated dunes. The barchan dunes moved in the 95–130° direction, which was basically consistent with the main wind direction of the area. The determination and function fitting of the barchan dune morphology shows that the correlation coefficients between all of the morphological parameters are greater than 0.8, except for the degree of spread and the symmetry of the two wings. We suggest adding the influences of other factors affecting barchan dunes and dune chains in the Yamarak Desert in future studies and paying attention to new wind-sand hazards and geohazards in the downwind corridor.

Extraction of rock and alteration geons by FODPSO segmentation and GP regression on the HyMap imagery: A case study of SW Birjand, Eastern Iran

In lithological and mineralogical mapping geons are generated by transforming the solid spatial/spectral information of imagery objects into individual rocks and minerals through complex algorithms including interpolation, segmentation, generalization, and regionalization. In geological applications, the foundation of most remote sensing image interpretation is based on mapping and classification. The most common classification techniques are established for multispectral data analysis and are pixel-based. As a consequence, such analysis show disadvantages for classification of the hyperspectral images that contain abundant detailed spatial, spectral, and textural information. The major objective of this research was to extract the mineral and rock geons of HyMap mosaicked imagery scenes from Birjand, eastern Iran. The approach combines object based segmentation and machine learning classification algorithms. The utilized imagery was segmented by FODPSO algorithm that creates higher between-class variances while the output image was classified by Gaussian process (GP) algorithm using 272 and 127 geons as training and testing areas. The key alteration minerals including alunite, jarosite, kaolinite, muscovite, and montmorillonite, and the dominant rock types consisting of granodiorite, tuff breccia, tuff, ignimbrite, dacite, and andesite were mapped accordingly. All of the geons were verified through field observations and the final output map showed an overall accuracy of 89.66 %. This investigation showed that integration of FODPSO and GP algorithms could be introduced as a new approach to create geons of rocks and minerals, and for lithological and mineralogical mapping in various geological settings.

Assessment on Changes of Ecosystem Carbon Storage in Reservoir Area due to Hydroproject.

Hydropower offers significant value for global carbon peak and carbon neutrality. However, the construction of hydropower stations leads to significant changes in land use and cover structure in reservoir areas, which affect ecosystem services including carbon balance. Furthermore, the development and operation of hydropower project require vast investment. However, the reservoir ecosystem's carbon storage and carbon emission reduction caused by hydropower could offer economic benefits when the official carbon market trading in China was launched in 2021. Therefore, it is necessary to assess comprehensively the changes in carbon storage and its value to the ecosystem in reservoir areas. The evaluation is of great importance for carbon loss reduction, land management, and hydropower development. This study provides a comprehensive and effective framework for evaluating changes in carbon storage and has its value to the reservoir ecosystem. It combines land utilization classification data obtained from remote sensing image interpretation and the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) carbon storage model. Based on the case study of the Xiluodu reservoir area, they were evaluated from two aspects: physical quantity and value quantity. The results show that the carbon storage in the Xiluodu reservoir area increased by 8,504.42 Mg from 2000 to 2018. The spatial distribution of the carbon storage shows a trend of high in the north and west, but low in the south and east. The construction of hydropower stations and the rise of reservoir water level covered a large amount of land, which led to the loss of carbon storage in reservoir areas. By implementing soil and water conservation and vegetation protection policies, parts of the cultivated land and grassland were converted into forestland, which was the main source for increasing the ecosystem's carbon storage. Moreover, carbon emission reduction was achieved by hydropower. In terms of the monetary value, the carbon storage value of the reservoir ecosystem increased to 19 million RMB during the construction period (2005–2015). The carbon storage value of the reservoir ecosystem increased to 611 million RMB during the operation period (2015–2018). The latter was greater than the maintenance cost of the hydropower station and exceeded the amortized cost of hydropower development, indicating the feasibility and economic benefits of hydropower development. These findings provide guidance for future hydropower development decisions in Jinsha River Basin and also others.

A Lightweight Network for Building Extraction From Remote Sensing Images

Building extraction is a fundamental research topic in remote sensing image interpretation. Convolutional neural network (CNN)-based building extraction algorithms have achieved high accuracy but require a large account of parameters and calculations, which hinders the practical application of these algorithms. To address the challenge, we propose a lightweight network (RSR-Net) for building extraction from remote sensing images. The network consists of three basic units with only a few parameters, and uses the idea of the fusion of shallow features and deep features, which is proposed by U-Net. Before features fusion, the squeeze-and-excitation (SE) module in RSR-Net assigned channel weights to these deep and shallow features. This operation can effectively reduce the influence of noise caused by shallow features in feature fusion, so as to improve the performance of the model. We estimated our network on datasets and achieved 88.32%, 71.58%, and 77.07% intersection-over-union (IoU) on datasets of aerial image and satellite image in Wuhan University (WHU) dataset, and the self-made building dataset of Guangzhou University Town, with only 2.81 M parameters and 6.91 G floating point operations (FLOPs). In addition, we propose a strategy combining target and background prediction, which makes RSR-Net achieve 0.37% improvement in IoU on WHU aerial image dataset. The effectiveness of RSR-Net is high. It showed that the proposed network is light and fast for the application of convolution neural network algorithm in practice.

Dual Learning-Based Graph Neural Network for Remote Sensing Image Super-Resolution

High-resolution (HR) remote sensing imagery plays a critical role in remote sensing image interpretation, and single image super-resolution (SISR) reconstruction technology is becoming increasingly valuable and significant. The state-of-the-art deep-learning-based SISR methods have demonstrated remarkable advantages, while reconstructing complex texture details still remains a big challenge. Besides, as a typical ill-posed inverse problem, how to determine the optimal solution is another important topic. To address these problems, in this work, a dual learning-based graph neural network (DLGNN) is proposed, in which the GNN is utilized to consider the self-similarity patches in remote sensing imagery by aggregating cross-scale neighboring feature patches, and dual learning strategy is adopted to refine the reconstruction results by constraining the mapping process in terms of the loss function, transferring the typical ill-posed problem to a well-posed one. Abundant experiments on 3K VEHICLE_SR datasets and Massachusetts Roads demonstrate the validity and outstanding performance for remote sensing image super-resolution tasks compared with other state-of-the-art super-resolution construction methods. Code is available at https://github.com/CUG-RS/DLGNN.

Remote Sensing Image Interpretation With Semantic Graph-Based Methods: A Survey

With the significant improvements in Earth observation (EO) technologies, remote sensing (RS) data exhibit the typical characteristics of Big Data. Propelled by the powerful feature extraction capabilities of intelligent algorithms, remote sensing image interpretation has drawn remarkable attention and achieved progress. However, the semantic relationship and domain knowledge hidden in massive RS images have not been fully exploited. To the best of our knowledge, a comprehensive review of recent achievements regarding semantic graph-based methods for comprehension and interpretation of remote sensing images is still lacking. Specifically, this article discusses the main challenges of remote sensing image interpretation and presents a systematic survey of typical semantic graph-based methodologies for RS knowledge representation and understanding, including the Ontology Model, Geo-Information Tupu, and Semantic Knowledge Graph. Furthermore, we categorize and summarize how the existing technologies address different challenges in RS image interpretation based on semantic graph-based methods, which indicates that the semantic information about potential relationships and prior knowledge of variant RS targets are central to the solution. In addition, a case study of RS geological interpretation based on the semantic knowledge graph is demonstrated to show the promising capability of intelligent RS image interpretation. Finally, the future directions are discussed for further research.

Multiple Embeddings Contrastive Pretraining for Remote Sensing Image Classification

This letter focuses on remote sensing image interpretation and aims to promote the use of contrastive self-supervised learning in varied applications of remote sensing image classification. The proposed method is a contrastive self-supervised pre-training framework that encourages the network to learn image representations by comparing image embeddings extracted by different encoders and predictors. Experiments were carried out on a variety of remote sensing image datasets to determine the efficacy of the proposed method for classification tasks. Results show that the proposed framework exploits the capabilities of encoders and outperforms the supervised learning method in terms of classification accuracy. Besides, it takes a few pre-training epochs to find a suboptimal initialization of network weights, and the pre-trained encoders use a little training data to get outstanding classification results, which shows the time and data efficiency of the proposed framework. Code is available at https://github.com/yinxu98/MECo.

Anchor-Free Oriented Proposal Generator for Object Detection

Oriented object detection is a practical and challenging task in remote sensing image interpretation. Nowadays, oriented detectors mostly use horizontal boxes as intermedium to derive oriented boxes from them. However, the horizontal boxes are inclined to get small Intersection-over-Unions (IoUs) with ground truths, which may have some undesirable effects, such as introducing redundant noise, mismatching with ground truths, detracting from the robustness of detectors, etc. In this paper, we propose a novel Anchor-free Oriented Proposal Generator (AOPG) that abandons horizontal box-related operations from the network architecture. AOPG first produces coarse oriented boxes by a Coarse Location Module (CLM) in an anchor-free manner and then refines them into high-quality oriented proposals. After AOPG, we apply a Fast R-CNN head to produce the final detection results. Furthermore, the shortage of large-scale datasets is also a hindrance to the development of oriented object detection. To alleviate the data insufficiency, we release a new dataset on the basis of our DIOR dataset and name it DIOR-R. Massive experiments demonstrate the effectiveness of AOPG. Particularly, without bells and whistles, we achieve the accuracy of 64.41%, 75.24% and 96.22% mAP on the DIOR-R, DOTA and HRSC2016 datasets respectively. Code and models are available at https://github.com/jbwang1997/AOPG.

基于原型思维的城市边缘区景观格局演变研究——以北京市大兴区为例

PDF HTML阅读 XML下载 导出引用 引用提醒 基于原型思维的城市边缘区景观格局演变研究——以北京市大兴区为例 DOI: 10.5846/stxb202103170716 作者: 作者单位: 作者简介: 通讯作者: 中图分类号: 基金项目: 北京林业大学建设世界一流学科和特色发展引导专项（2019XKJS0317）；北京市共建项目专项（D171100000217002） Evolution of landscape pattern in urban fringe based on prototype thinking: A case study of Daxing District of Beijing Author: Affiliation: Fund Project: Supported by: Beijing Forestry University (2019XKJS0317) and Beijing Municipal Co-construction Project (D171100000217002) 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:城市边缘区是城市扩张过程中空间结构变化最大、土地利用转变最快的区域。以北京市大兴区为例，从原型思维出发，以2000、2010、2020年遥感影像解译数据、自然和社会经济等数据为基础，充分考虑地物地情及政策背景等因素，借助ArcGIS 10.5、Fragstats 4.2、SPSS等工具，对不同规划尺度和历时性维度下的景观格局指数进行分析、对比和验证，通过总体分析，场地切片和原型筛选三个步骤，筛选出大兴区6个空间原型组成框架，审视不同发展导向下的空间原型发展变化，探讨近20年间大兴区城市景观格局演变规律。旨在提出一种原型理论指导下的研究思路，尝试为城市转型、用地结构优化、区域规划等工作提供新的视角和一定的借鉴意义。 Abstract:In the process of urban expansion, urban fringe is the area with the biggest change in spatial structure and the fastest change in land use. This paper takes Daxing District of Beijing as an example, starting from the prototype thinking, based on natural, social, economic data and remote sensing image interpretation in 2000, 2010 and 2020, fully considering the factors of land conditions and policy background. With the help of ArcGIS 10.5,FRAGSTATS 4.2 and SPSS, we analyze, compare and verify the landscape pattern index under different planning scales and diachronic dimensions. Through steps of overall analysis, site slicing and prototype screening, six spatial prototypes of Daxing District were screened out to examine the development and changes of spatial prototypes under different development orientations, thus to explore the evolution law of urban landscape pattern in Daxing District in last 20 years. This paper aims to put forward a research idea under the guidance of prototype theory, and try to provide a new perspective and certain reference for urban transformation, land use structure optimization and regional planning. 参考文献 相似文献 引证文献

LaST: Label-Free Self-Distillation Contrastive Learning With Transformer Architecture for Remote Sensing Image Scene Classification

The increase in self-supervised learning (SSL), especially contrastive learning, has enabled one to train deep neural network models with unlabeled data for remote sensing image (RSI) scene classification. Nevertheless, it still suffers from the following issues. 1. The performance of the contrastive learning method is significantly impacted by the hard negative sample (HNS) issue, since the RSI scenario is complex in semantics and rich in surface features. 2. The multiscale characteristic of RSI is missed in the existing contrastive learning methods. 3. As the backbone of a deep learning model, especially in the case of limited annotation, a CNN does not include the adequate receptive field of convolutional kernels to capture the broad contextual information of RSI. In this regard, we propose label-free self-distillation contrastive learning with a transformer architecture (LaST). We introduce the self-distillation contrastive learning mechanism to address the HNS issue. Specifically, the LaST architecture comprises two modules: scale alignment with a multicrop module and a long-range dependency capture backbone module. In the former, we present global-local crop and scale alignment to encourage local-to-global correspondence and acquire multiscale relations. Then, the distorted views are fed into a transformer as a backbone, which is good at capturing the long-range-dependent contextual information of the RSI while maintaining the spatial smoothness of the learned features. Experiments on public datasets show that in the downstream scene classification task, LaST improves the performance of the self-supervised trained model by a maximum of 2.18% compared to the HNS-impacted contrastive learning approaches, and only 1.5% of labeled data can achieve the performance of supervised training CNNs with 10% labeled data. Moreover, this letter supports the integration of a transformer architecture and self-supervised paradigms in RSI interpretation.

Multimodal Semantic Consistency-Based Fusion Architecture Search for Land Cover Classification

Multimodal Land Cover Classification (MLCC) using the optical and Synthetic Aperture Radar (SAR) modalities has resulted in outstanding performances over using only unimodal data due to their complementary information on land properties. Previous multimodal deep learning (MDL) methods have relied on handcrafted multi-branch convolutional neural networks (CNN) to extract the features of different modalities and merged them for land cover classification. However, natural images-oriented handcrafted CNN models may not the optimal strategies to handle Remote Sensing (RS) image interpretation problems, due to the huge difference in terms of imaging angles and imaging ways. Furthermore, few MDL methods have analyzed optimal combinations of hierarchical features from different modalities. In this article, we propose an efficient multimodal architecture search framework, namely Multimodal Semantic Consistency-Based Fusion Architecture Search (M²SC-FAS) in continuous search space with the gradient-based optimization method, which can not only discover optimal optical- and SAR-specific architectures according to the different characteristics of the optical and SAR images, respectively, but also realizes the search of optimal multimodal dense fusion architecture. Specifically, the semantic-consistency constraint is introduced to guarantee dense fusion between hierarchical optical and SAR features with high semantic consistency and then capture the complementary performance on land properties. Finally, the basis of curriculum learning strategy is adopted on the M²SC-FAS. Extensive experiments show superior performances of our work on three broad co-registered optical and SAR datasets.

DFAF-Net: A Dual-Frequency PolSAR Image Classification Network Based on Frequency-Aware Attention and Adaptive Feature Fusion

Multifrequency (MF) polarization synthetic aperture radar (PolSAR) systems can obtain more abundant and continuous earth resource information than single-frequency ones and have been widely used in the remote sensing community. However, there are relatively a few researches for the fine MF PolSAR image classification, which is an important part of remote sensing image interpretation. The main focus currently is on the single-frequency part. Therefore, for dual-frequency PolSAR image classification, this article proposes the dual-frequency attention fusion network (DFAF-Net). It is based on frequency-aware attention and adaptive feature fusion to improve classification performance. First, the dual-frequency PolSAR data is input into the joint feature extraction (JFE) module to obtain the joint feature representation. Meanwhile, two frequency-aware attention block (FAB) modules with the same structure are constructed, which, respectively, generate frequency-specific attention masks based on the guidance of different frequency data. Subsequently, these masks are used to weigh the joint features to highlight the description of different frequency-aware importance. The activated frequency-aware features can fully mine and utilize the complementary information provided by different frequencies, thereby enhancing the discrimination of similar landcover categories. Finally, the adaptive feature fusion block (AFFB) module is utilized to adaptively aggregate different frequency-aware features multiple times, which can effectively eliminate information differences. The obtained fusion features are more compact within classes and separable between classes, thereby effectively improving the classification performance. Experiments on three measured spaceborne and airborne dual-frequency PolSAR datasets verify that DFAF-Net can better perceive frequency characteristics and fully mine the complementary. Therefore, the classification accuracy is effectively enhanced, and the inaccuracy of single-frequency classification can be eliminated. Meanwhile, due to the introduction of the attention module, the classification performance of the proposed DFAF-Net is more competitive than the related deep learning networks. Quantitatively, the overall accuracy of DFAF-Net on the three datasets is respectively 98.30%, 97.42%, and 99.45%, which is better than other methods.

A Statistical-Spatial Feature Learning Network for PolSAR Image Classification

Polarimetric synthetic aperture radar (PolSAR) image classification plays an important role in the development of remote sensing image interpretation for the rich polarization information. Generative methods learn the statistical distribution characteristic of the scattered echoes data with heavy noise. However, it is tedious to design and solve different generative likelihood functions. Discriminative methods learn image features and classifier in an end-to-end framework whose classification performance is also limited without polarization statistical characteristics. In order to make full use of the statistical characteristic of PolSAR echoes and spatial feature of PolSAR image, a novel real-value hybrid generative/discriminative (HGD) deep network is proposed to learn statistical-spatial feature for PolSAR image classification with the data expressing ability of the generative term and the end-to-end learning ability of the discriminative term. First, a derived replacement form is obtained from the typical Wishart distribution of PolSAR data with an eigenvalue generative term. In this way, the complex-value form of PolSAR data converts into real-value expression, which makes the PolSAR image classification easy to understand and implement in the deep network. Finally, these two terms are integrated from the variational Bayesian theory with an alternative optimization equation derived for PolSAR image classification. It provides a normal framework for PolSAR multifeature learning. Experiments are tested on different PolSAR datasets compared with several state-of-the-art individual generative or discriminative methods. All visual experimental results and accuracy values demonstrate the superiority of the proposed method.

When CNNs Meet Vision Transformer: A Joint Framework for Remote Sensing Scene Classification

Scene classification is an indispensable part of remote sensing image interpretation, and various convolutional neural network (CNN)-based methods have been explored to improve classification accuracy. Although they have shown good classification performance on high-resolution remote sensing (HRRS) images, discriminative ability of extracted features is still limited. In this letter, a high-performance joint framework combined CNNs and vision transformer (ViT) (CTNet) is proposed to further boost the discriminative ability of features for HRRS scene classification. The CTNet method contains two modules, including the stream of ViT (T-stream) and the stream of CNNs (C-stream). For the T-stream, flattened image patches are sent into pretrained ViT model to mine semantic features in HRRS images. To complement with T-stream, pretrained CNN is transferred to extract local structural features in the C-stream. Then, semantic features and structural features are concatenated to predict labels of unknown samples. Finally, a joint loss function is developed to optimize the joint model and increase the intraclass aggregation. The highest accuracies on the aerial image dataset (AID) and Northwestern Polytechnical University (NWPU)-RESISC45 datasets obtained by the CTNet method are 97.70% and 95.49%, respectively. The classification results reveal that the proposed method achieves high classification performance compared with other state-of-the-art (SOTA) methods.

A 3-D-Swin Transformer-Based Hierarchical Contrastive Learning Method for Hyperspectral Image Classification

Deep convolutional neural networks have been dominating in the field of hyperspectral image (HSI) classification. However, single convolutional kernels can limit the receptive field and fail to capture the sequential properties of data. Self-Attention-based Transformer can build global sequence information, among which, the Swin Transformer (SwinT) integrates sequence modeling capability and priori information of the visual signals (e.g., locality and translation invariance). Based on SwinT, we propose a 3D Swin Transformer (3DSwinT) to accommodate the 3D properties of HSI and capture the rich spatial-spectral information of HSI. Currently, supervised learning is still the most commonly used method for remote sensing image interpretation. However, pixel-by-pixel HSI classification demands a large number of high-quality labeled samples, which are time-consuming and costly to collect. As an unsupervised learning, self-supervised learning (SSL), especially contrastive learning, can learn semantic representations from unlabeled data, and hence, is becoming a potential alternative to supervised learning. On the other hand, current contrastive learning methods are all single-level or single-scale, which do not consider complex and variable multi-scale features of objects. Therefore, this paper proposes a novel 3DSwinT-based hierarchical contrastive learning method (3DSwinT-HCL), which can fully exploit multi-scale semantic representations of images. Besides, we propose a multi-scale local contrastive learning (MS-LCL) module to mine the pixel-level representations in order to adapt to downstream dense prediction tasks. A series of experiments verify the great potential and superiority of 3DSwinT-HCL.

Remote Sensing Scene Classification via Multi-Branch Local Attention Network.

Remote sensing scene classification (RSSC) is a hotspot and play very important role in the field of remote sensing image interpretation in recent years. With the recent development of the convolutional neural networks, a significant breakthrough has been made in the classification of remote sensing scenes. Many objects form complex and diverse scenes through spatial combination and association, which makes it difficult to classify remote sensing image scenes. The problem of insufficient differentiation of feature representations extracted by Convolutional Neural Networks (CNNs) still exists, which is mainly due to the characteristics of similarity for inter-class images and diversity for intra-class images. In this paper, we propose a remote sensing image scene classification method via Multi-Branch Local Attention Network (MBLANet), where Convolutional Local Attention Module (CLAM) is embedded into all down-sampling blocks and residual blocks of ResNet backbone. CLAM contains two submodules, Convolutional Channel Attention Module (CCAM) and Local Spatial Attention Module (LSAM). The two submodules are placed in parallel to obtain both channel and spatial attentions, which helps to emphasize the main target in the complex background and improve the ability of feature representation. Extensive experiments on three benchmark datasets show that our method is better than state-of-the-art methods.

DO-Net: Dual-Output Network for Land Cover Classification From Optical Remote Sensing Images

Land cover classification is the basic task of remote sensing image interpretation. Related methods have developed rapidly, especially the branch based on deep learning (DL). For high-resolution remote sensing images, the smaller inter-class difference and greater intra-class difference are two obstacles to improving the classification accuracy. For the former, the DL models generally use a deeper encoder to extract more powerful classification features. Considering that the scale of different land cover categories varies greatly, multi-scale feature extraction modules are also used to improve the classification accuracy. While the latter is always overlooked, and thus we propose a dual-output model, which uses a dense spatial pyramid pooling (DSPP) module to generate both the pixel-level and region-level predictions, to reduce the influence of intra-class differences. To further increase the classification accuracy, we investigate the band selection technique to apply the pre-trained encoder from the natural red green blue (RGB) dataset to multi-spectral remote sensing images. Extensive experiments on two datasets demonstrate the effectiveness of our model.

DMML-Net: Deep Metametric Learning for Few-Shot Geographic Object Segmentation in Remote Sensing Imagery

Geographic object segmentation is a fundamental yet challenging problem for remote sensing image interpretation. The prevalent paradigm to solve this problem is to train deep neural networks on massive labeled samples. Although remarkable achievements have been attained, these methods suffer from the severe dependence on the large-scale dataset and require a long training process with high computation burden. To address these issues, a deep metametric learning framework, named DMML-Net, consisting of the metametric learner and the base-metric learner, is proposed for few-shot geographic object segmentation. First, DMML-Net formulates the segmentation as the metric-based pixel classification and develops a deep feature pyramid comparison network as the architecture of the metric learner for multiscale metric learning. Benefiting from this design, the segmentation can be efficiently solved, as well as being robust to deal with the scale variations of geographic objects. Second, an affinity-based fusion mechanism is introduced to adaptively reweight and fuse the semantic information across samples, effectively calibrating the deviation of prototypes induced by the intraclass variations. Third, considering the impact of the large interclass distribution divergences, DMML-Net presents a metametric training paradigm to provide the metric model with flexible scalability for fast adaptation to novel tasks. After metatraining, DMML-Net can be applied for the few-shot segmentation tasks of novel geographic objects with only a few gradient steps on the small training set. Experimental results on two benchmark remote sensing datasets demonstrate the validity and the superiority of our method in low-shot conditions where there are only one to ten labeled samples.

HMCNet: Hybrid Efficient Remote Sensing Images Change Detection Network Based on Cross-Axis Attention MLP and CNN

As an important task in the field of remote sensing image interpretation, change detection (CD) has been extensively studied by scholars in recent years. Affected by the illumination and the environment during bitemporal images’ acquisition, there will be many pseudochanges, and the pseudochanges will seriously affect the effect of CD. Based on this, we propose a CD model named HMCNet, which introduces multilayer perceptron (MLP) into a convolutional neural network (CNN)-based CD model to form an MLP-CNN hybrid model. HMCNet has both the good feature extraction of CNN and the long-term dependence modeling ability of MLP, which can effectively overcome the interference of pseudochanges. In addition, the proposed cross-axis attention MLP can induce window attention of local features through shifted windows and, at the same time, form global attention to features through the interaction between information flows on the cross-axis, which effectively improves the comprehensive performance of MLP block. Extensive experiments on three public benchmark datasets show that HMCNet can achieve better performance with fewer parameters and Flops, and still maintain good generalization ability with fewer train data.

Remote-Sensing Image Tile-Level Annotation Based on Discriminative Features and Expressive Visual Word Descriptors

Remote-sensing images play a crucial role in a wide range of applications and have been receiving significant attention. In recent years, great efforts have been made in developing various methods for intelligent interpretation of remote-sensing images. Generally speaking, machine learning-based methods of remote-sensing image interpretation require a large number of labeled samples and there are still not enough annotated datasets in the field of remote sensing. However, manual annotation of remote-sensing images is usually labor-intensive and requires expert knowledge and the accuracy of annotation results is relatively low. The goal of this paper is to propose a novel tile-level annotation method of remote-sensing images to obtain remote-sensing datasets which are well-labeled and contain accurate semantic concepts. Firstly, we use a set of images with defined semantic concepts to represent the training set and divide them into several nonoverlapping regions. Secondly, the color features, texture features, and spatial features of each region are extracted, and discriminative features are obtained by the weight optimization feature fusion method. Then, the features are quantized into visual words by applying a density-based clustering center selection method and an isolated feature point elimination method. And the remote-sensing images can be represented by a series of visual words. Finally, the LDA model is used to calculate the probabilities of semantic categories for each region. The experiments are conducted on remote-sensing images which demonstrate that our proposed method can achieve good performance on remote-sensing image tile-level annotation. The implications of our research can obtain annotated datasets with accurate semantic concepts for intelligent interpretation of remote-sensing images.