Very High Resolution Research Articles

Characterizing extreme values of precipitation at very high resolution: An experiment over twenty European cities

The paper presents a new hourly high-resolution (i.e., at 0.02°, ≃2.2 km) precipitation dataset, labelled as ERA5@2km, obtained by dynamically downscaling ERA5 reanalysis at convection permitting scale (CPS) over 20 European cities for the recent past thirty years (1989–2018). The downscaling activity is performed within the framework of the Contract implemented by Fondazione CMCC to support Sectoral Information System about “Disaster Risk Reduction” (see https://climate.copernicus.eu/pluvial-flood-risk-assessment-urban-areas) of Copernicus Climate Change Service (C3S). Specifically, such an additional precipitation dataset is developed to provide precipitation data for estimating expected precipitations at fixed return periods used as input for pluvial flooding risk analysis where hazard (inundated areas, water levels) and risk (estimated damages) are assessed. The ambition is to support the Disaster Risk Reduction (DRR) community involved in pluvial flood risk assessment by providing a basis for impact analysis at city scale, in terms of extreme hourly precipitation, that matches with the expected spatial and temporal requirements. In this work, ERA5@2km precipitation dataset is introduced for the first time and its reliability and coherence are evaluated as for spatial patterns and trends as for extreme values against a set of available high-resolution observational datasets (comparable in terms of spatial and temporal resolution). Such an evaluation provides a clearer understanding about the added value of very high-resolution (VHR) dynamical downscaling reanalysis in terms of localization and magnitude of precipitation events at urban scale, confirming a general and relevant added value of this new configuration for the assessment of extreme atmospheric events (such as heavy precipitations).

Arterial health during early childhood following abnormal fetal growth

BackgroundAbnormal fetal growth is associated with increased cardiovascular risk in adulthood. We investigated the effect of fetal programming on arterial health and morphology during early childhood.MethodsWe examined 90 children (median age 5.81 years, interquartile range: 5.67; 5.95), born small for gestational age with fetal growth restriction, large or appropriate for gestational age (SGA, N = 23, LGA, N = 19, AGA N = 48). We measured body composition, anthropometrics, blood pressure, pulse wave velocity (PWV), lipids, glucose and inflammatory markers, and assessed carotid, brachial, radial and femoral arterial morphology and stiffness using very-high resolution ultrasound (46–71 MHz).ResultsLGA showed increased anthropometry, lean body mass and body mass index. SGA displayed decreased anthropometry and lean body mass. Blood pressure, PWV, carotid artery stiffness and blood work did not differ groupwise. Differences in lumen diameters, intima-media thicknesses (IMT) and adventitia thicknesses disappeared when adjusted for lean body mass and sex. In multiple regression models arterial dimensions were mainly predicted by lean body mass, with birth weight remaining associated only with carotid and brachial lumen dimensions, and not with IMTs. Carotid-femoral PWV was predicted by height and blood pressure only. No independent effect of adiposity was observed.ConclusionsArterial dimensions in childhood associate with current anthropometrics, especially lean body mass, and sex, explaining differences in arterial layer thickness. We found no signs of fetal programming of cardiovascular risk or arterial health in early childhood.

Satellite–Derived Topography and Morphometry for VHR Coastal Habitat Mapping: The Pleiades–1 Tri–Stereo Enhancement

The evolution of the coastal fringe is closely linked to the impact of climate change, specifically increases in sea level and storm intensity. The anthropic pressure that is inflicted on these fragile environments strengthens the risk. Therefore, numerous research projects look into the possibility of monitoring and understanding the coastal environment in order to better identify its dynamics and adaptation to the major changes that are currently taking place in the landscape. This new study aims to improve the habitat mapping/classification at Very High Resolution (VHR) using Pleiades–1–derived topography, its morphometric by–products, and Pleiades–1–derived imageries. A tri–stereo dataset was acquired and processed by image pairing to obtain nine digital surface models (DSM) that were 0.50 m pixel size using the free software RSP (RPC Stereo Processor) and that were calibrated and validated with the 2018–LiDAR dataset that was available for the study area: the Emerald Coast in Brittany (France). Four morphometric predictors that were derived from the best of the nine generated DSMs were calculated via a freely available software (SAGA GIS): slope, aspect, topographic position index (TPI), and TPI–based landform classification (TPILC). A maximum likelihood classification of the area was calculated using nine classes: the salt marsh, dune, rock, urban, field, forest, beach, road, and seawater classes. With an RMSE of 4 m, the DSM#2–3_1 (from images #2 and #3 with one ground control point) outperformed the other DSMs. The classification results that were computed from the DSM#2–3_1 demonstrate the importance of the contribution of the morphometric predictors that were added to the reference Red–Green–Blue (RGB, 76.37% in overall accuracy, OA). The best combination of TPILC that was added to the RGB + DSM provided a gain of 13% in the OA, reaching 89.37%. These findings will help scientists and managers who are tasked with coastal risks at VHR.

Open-Source Data-Driven Cross-Domain Road Detection From Very High Resolution Remote Sensing Imagery.

High-precision road detection from very high resolution (VHR) remote sensing images has broad application value. However, the most advanced deep learning based methods often fail to identify roads when there is a distribution discrepancy between the training samples and test samples, due to their limited generalization ability. In this paper, to address this problem, an open-source data-driven domain-specific representation (OSM-DOER) framework is proposed for cross-domain road detection. On the one hand, as the spatial structure information of the source and target domains is similar, but the texture information is different, the domain-specific representation (DOER) framework is proposed, which not only aligns the distributions of the spatial structure information, but also learns the domain-specific texture information. Furthermore, in order to enhance the representation of the target domain data distribution, open-source and freely available OpenStreetMap (OSM) road centerline data are utilized to generate target domain samples, which are then used in the network training as the supervised information for the target domain. Finally, to verify the superiority of the proposed OSM-DOER framework, we conducted extensive experiments with the public SpaceNet and DeepGlobe road datasets, and large-scale road datasets from Birmingham in the UK and Shanghai in China. The experimental results demonstrate that the proposed OSM-DOER framework shows obvious advantages over the mainstream road detection methods, and the use of OSM road centerline data has great potential for the road detection task.

Curriculum-Style Local-to-Global Adaptation for Cross-Domain Remote Sensing Image Segmentation

Although domain adaptation has been extensively studied in natural image-based segmentation tasks, the research on cross-domain segmentation for very-high-resolution (VHR) remote sensing images (RSIs) still remains underexplored. The VHR RSI-based cross-domain segmentation mainly faces two critical challenges: 1) large area land covers with many diverse object categories bring severe local patch-level data distribution deviations, thus yielding different adaptation difficulties for different local patches and 2) different VHR sensor types or dynamically changing modes cause the VHR images to go through intensive data distribution differences even for the same geographical location, resulting in different global feature-level domain gaps. To address these challenges, we propose a curriculum-style local-to-global cross-domain adaptation framework for the segmentation of VHR RSIs. The proposed curriculum-style adaptation performs the adaptation process in an easy-to-hard way according to the adaptation difficulties that can be obtained using an entropy-based score for each patch of the target domain and, thus, well aligns the local patches in a domain image. The proposed local-to-global adaptation performs the feature alignment process from the locally semantic to globally structural feature discrepancies and consists of a semantic-level domain classifier and an entropy-level domain classifier that can reduce the above cross-domain feature discrepancies. Extensive experiments have been conducted in various cross-domain scenarios, including geographic location variations and imaging mode variations, and the experimental results demonstrate that the proposed method can significantly boost the domain adaptability of segmentation networks for VHR RSIs.

Novel Automatic Approach for Land Cover Change Detection by Using VHR Remote Sensing Images

Many land cover change detection (LCCD) approaches applied on very high resolution (VHR) remote sensing images utilize spatial information by using a regular window or strict mathematical model. However, regular shape or strict models cannot fit the various shapes and sizes of the ground targets. In this article, a novel LCCD approach without the parameter is proposed to detect land cover change with VHR remote sensing images. First, an adaptive spatial-context extraction algorithm is applied to explore contextual information around a pixel. Second, the change magnitude between pairwise pixels is quantitatively measured by computing the band-to-band distance which is defined by the pairwise adaptive regions around the corresponding pixels. Finally, after the generation of a change magnitude image (CMI), a binary threshold method called double-window flexible pace search (DFPS) is adopted to divide CMI into a binary change detection map. The performance of the proposed approach is verified by comparing it with five state-of-the-art methods with three pairs of VHR images. The comparisons demonstrated that the proposed approach achieved the improved detected results comparing with state-of-the-art LCCD methods. The code of the proposed approach is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/TongfeiLiu/ASEA-CD</uri> .

Multisensor Fusion and Explicit Semantic Preserving-Based Deep Hashing for Cross-Modal Remote Sensing Image Retrieval

Cross-modal hashing is an important tool for retrieving useful information from very-high-resolution (VHR) optical images and synthetic aperture radar (SAR) images. Dealing with the intermodal discrepancies, including both spatial–spectral and visual semantic aspects, between VHR and SAR images is extremely vital to generate high-quality common hash codes in the Hamming space. However, existing cross-modal hashing methods ignore the spatial–spectral discrepancy when representing VHR and SAR images. Moreover, existing methods employ derived supervised signals, such as pairwise training images, to implicitly guide hashing learning, which fails to effectively deal with the visual semantic discrepancy, i.e., cannot adequately preserve the intraclass similarity and interclass discrimination between VHR and SAR images. To address these drawbacks, this article proposes a multisensor fusion and explicit semantic preserving-based deep Hashing method, termed as MsEspH, which can effectively deal with the discrepancies. Specifically, we design a novel cross-modal hashing network to eliminate the spatial–spectral discrepancies by fusing extra multispectral images (MSIs), which are generated in real time by a generative adversarial network. Then, we propose an explicit semantic preserving-based objective function by analyzing the connection between classification and hash learning. The objective function can preserve the intraclass similarity and interclass discrimination with class labels directly. Moreover, we theoretically verify that hash learning and classification can be unified into a learning framework under certain conditions. To evaluate our method, we construct and release a large-scale VHR-SAR image dataset. Extensive experiments on the dataset demonstrate that our method outperforms various state-of-the-art cross-modal hashing methods.

CGSANet: A Contour-Guided and Local Structure-Aware Encoder–Decoder Network for Accurate Building Extraction From Very High-Resolution Remote Sensing Imagery

Extracting buildings accurately from very high-resolution (VHR) remote sensing imagery is challenging due to diverse building appearances, spectral variability, and complex background in VHR remote sensing images. Recent studies mainly adopt a variant of the fully convolutional network (FCN) with an encoder–decoder architecture to extract buildings, which has shown promising improvement over conventional methods. However, FCN-based encoder–decoder models still fail to fully utilize the implicit characteristics of building shapes. This adversely affects the accurate localization of building boundaries, which is particularly relevant in building mapping. A contour-guided and local structure-aware encoder–decoder network (CGSANet) is proposed to extract buildings with more accurate boundaries. CGSANet is a multitask network composed of a contour-guided (CG) and a multiregion-guided (MRG) module. The CG module is supervised by a building contour that effectively learns building contour-related spatial features to retain the shape pattern of buildings. The MRG module is deeply supervised by four building regions that further capture multiscale and contextual features of buildings. In addition, a hybrid loss function was designed to improve the structure learning ability of CGSANet. These three improvements benefit each other synergistically to produce high-quality building extraction results. Experimental results on the WHU and NZ32km2 building datasets demonstrate that compared with the tested algorithms, CGSANet can produce more accurate building extraction results and achieve the best intersection over union value 91.55% and 90.02%, respectively. Experiments on the INRIA building dataset further demonstrate the ability for generalization of the proposed framework, indicating great practical potential.

Landslide Inventory Mapping on VHR Images via Adaptive Region Shape Similarity

Landslide inventory mapping (LIM) is an important application in remote sensing for assisting in the relief of landslide geohazards. However, while conducting LIM tasks performing change detection analysis using bi-temporal very high-resolution (VHR) remote sensing images, due to landslide usually occurred in a mountain area, the phenological difference and outcrop rock may bring pseudo-changes to LIM results. In this paper, a novel change detection approach based on Adaptive Region Shape Similarity (ARSS) is proposed for LIM with VHR remote sensing images to improve detection performance. First, an adaptive region around each pixel is extended to explore the contextual information. Then, direction lines within an adaptive region are defined to describe the shape of the adaptive region. Finally, the pixels located on each direction line are taken into account to build the corresponding histogram. The shape similarity between the pairwise histogram curves is measured by using the Discrete Frchet Distance (DFD). Once the bi-temporal images are processed by using the abovementioned steps, a change magnitude image (CMI) is generated, while a threshold is then used to obtain a final binary change map. The proposed approach is applied to three pairs of landslide sites images acquired with aerial plane and one land use change dataset acquired by Quick Bird Satellite. Compared with ten state-of-the-art methods, the proposed approach achieved LIMs and detection results with higher accuracies and better performance.

Deep Multimodal Fusion Network for Semantic Segmentation Using Remote Sensing Image and LiDAR Data

Extracting semantic information from very-high-resolution (VHR) aerial images is a prominent topic in the Earth observation research. An increasing number of different sensor platforms are appearing in remote sensing, each of which can provide corresponding multimodal supplemental or enhanced information, such as optical images, light detection and ranging (LiDAR) point clouds, infrared images, or inertial measurement unit (IMU) data. However, these current deep networks for LiDAR and VHR images have not fully utilized the complete potential of multimodal data. The stacked multimodal fusion network (MFNet) ignores the structural differences between the modalities and the manual statistical characteristics within the modalities. For multimodal remote sensing data and its corresponding carefully designed handcrafted features, we designed a novel deep MFNet that can use multimodal VHR aerial images and LiDAR data and the corresponding intramodal features, such as LiDAR-derived features [slope and normalized digital surface model (NDSM)] and imagery-derived features [infrared–red–green (IRRG), normalized difference vegetation index (NDVI), and difference of Gaussian (DoG)]. Technically, we introduce the attention mechanism and multimodal learning to adaptively fuse intermodal and intramodal features. Specifically, we designed a multimodal fusion mechanism, pyramid dilation blocks, and a multilevel feature fusion module. Through these modules, our network realized the adaptive fusion of multimodal features, improved the receptive field, and enhanced the global-to-local contextual fusion effect. Moreover, we used a multiscale supervision training scheme to optimize the network. Extensive experimental results and ablation studies on the ISPRS semantic dataset and IEEE GRSS DFC Zeebrugge dataset show the effectiveness of our proposed MFNet.

A Shallow-to-Deep Feature Fusion Network for VHR Remote Sensing Image Classification

With more detailed spatial information being represented in very-high-resolution (VHR) remote sensing images, stringent requirements are imposed on accurate image classification. Due to the diverse land-objects with intraclass variation and interclass similarity, efficient and fine classification of VHR images especially in complex scenes is challenging. Even for some popular deep learning (DL) frameworks, geometric details of land-object may be lost in deep feature levels, so it is difficult to maintain the highly-detailed spatial information (e.g., edges, small objects) only relying on the last high-level layer. Moreover, many of the newly developed DL methods require massive well-labeled samples, which inevitably deteriorates the model generalization ability under the few-shot learning. Therefore, in this paper, a lightweight shallow-to-deep feature fusion network (SDF2N) is proposed for VHR image classification, where the traditional machine learning (ML) and DL schemes are integrated to learn rich and representative information to improve the classification accuracy. In particular, the shallow spectral-spatial features are first extracted, and then a novel triple-stage fusion (TSF) module is designed to learn the saliency and discriminative information at different levels for classification. The TSF module includes three feature fusion stages, i.e., low-level spectral-spatial feature fusion, middle-level multi-scale feature fusion, and high-level multi-layer feature fusion. The proposed SDF2N takes advantages of the shallow-to-deep features, which can extract representative and complementary information of crossing layers. It is important to note that even with limited training samples, the SDF2N still can achieve satisfying classification performance. Experimental results obtained on three real VHR remote sensing data sets including two multispectral and one airborne hyperspectral images covering complex urban scenarios confirm the effectiveness of the proposed approach compared with the state-of-the-art methods.

Spatially and Semantically Enhanced Siamese Network for Semantic Change Detection in High-Resolution Remote Sensing Images

Given a pair of bitemporal very high resolution (VHR) remote sensing images, the semantic change detection task aims to locate land surface changes and identify their semantic classes. The existing algorithms use independent branches to locate and identify separately without considering the association between branches. In this article, we propose an end-to-end spatially and semantically enhanced Siamese network (SSESN) for semantic change detection. The SSESN aggregates the rich spatial and semantic information in the VHR image through a designed spatial and semantic feature aggregation module. Additionally, a change-aware module is proposed to decouple the aggregated features. Features in the binary branch are fused to the semantic branches as prior location information. This allows the spatially enhanced features to predict changed regions and the semantically enhanced features to refine the region categorizations. Experimental results show that our method provides comparable results with the state-of-the-art binary change detection and semantic change detection algorithms.

Feature-Selection High-Resolution Network With Hypersphere Embedding for Semantic Segmentation of VHR Remote Sensing Images

Very high resolution (VHR) remote sensing images contain various multi-scale objects, such as large-scale buildings and small-scale cars. However, these multi-scale objects cannot be considered simultaneously in the widely used backbones with large downsampling factor (e.g. VGG-like and ResNet-like), resulting in the appearance of various context aggregation approaches such as fusing low-level features and attention-based modules. To alleviate this problem caused by backbones with large downsampling factor, we propose a feature-selection high-resolution network (FSHRNet) based on an observation: if the features maintain high resolution throughout the network, a high precision segmentation result can be obtained by only using a 1&#x00D7;1 convolution layer with no need for complex context aggregation modules. Specifically, the backbone of FSHRNet is a multi-branch structure similar to HRNet where the high-resolution branch is the principal line. Then, a lightweight dynamic weight module, named feature-selection convolution layer (FSConv), is presented to fuse multi-resolution features, allowing adaptively feature selection based on the characteristic of objects. Finally, a specially designed 1&#x00D7;1 convolution layer derived from hypersphere embedding is used to produce the segmentation result. Experiments with other widely used methods show that the proposed FSHRNet obtains competitive performance on ISPRS Vaihingen dataset, ISPRS Potsdam dataset and iSAID dataset.

Building Outline Delineation From VHR Remote Sensing Images Using the Convolutional Recurrent Neural Network Embedded With Line Segment Information

Recently, several recurrent neural network (RNN)-based models have been proposed to delineate the outlines of buildings from very high resolution (VHR) remote sensing images. These models first use convolutional neural networks (CNNs) to recognize the boundary fragments by learning probability maps of both edges and corners and then feed them into RNN to find and link a set of sequent corners into external boundaries of buildings. However, caused by the category imbalance of edges and corners, the local ambiguity of edge detection is very serious, which significantly affects the accuracy of predicted outline corners. To tackle this challenge, this article introduces a convolutional RNN embedded with line segment information (LSI-RNN), a novel network that aims to directly detect line segment instead of edges. To achieve this, LSI-RNN utilizes an additional cotraining branch to generate an attraction field map (AFM) by neural discriminative dimensionality reduction (NDDR) layer. Consequently, the conventional classification problem of edges is converted to a regression problem of line segments, thus solving the aforementioned issues. Experimental results over three remote sensing datasets with different spatial resolutions show that the proposed method consistently outperforms other state-of-the-art methods.

Bilateral Semantic Fusion Siamese Network for Change Detection From Multitemporal Optical Remote Sensing Imagery

Change detection (CD) is an essential task in optical remote sensing, and it can be used to extract the valid information from sequential multitemporal images. However, since the character of long-term revisiting and very high resolution (VHR) development, the great differences of illumination, season, and interior textures between bitemporal images bring considerable challenges for pixel-wise CD. In this letter, focusing on accurate pixel-wise CD, a bilateral semantic fusion Siamese network (BSFNet) is proposed. First, to better map bitemporal images into semantic feature domain for comparison, a novel BSFNet is designed to effectively integrate shallow and deep semantic features, which can provide pixel-wise CD results with complete regions and clear boundary locations. Then, in order to facilitate the reasonable convergence of the proposed BSFNet, a scale-invariant sample balance (SISB) loss is designed for metric learning to avoid the problems of sample imbalance and scale variance. Finally, extensive experiments are carried out on two published CDD and LEVIR CD datasets, and results indicate that the proposed BSFNet can provide superior performance than the other state-of-the-art methods. Our work is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/ClarissaDHL/BSFNet</uri> .

Multitemporal UAV Photogrammetry For Sandbank Morphological Change Analysis: Evaluations of Camera Calibration Methods, Co-Registration Strategies, and the Reconstructed DSMs

Sandbank morphology is a longshore corridor with fast morphological changes. To analyze its changes, unmanned aerial vehicle (UAV) is the preferred effective and flexible data collection platform, which can collect very high resolution (VHR) images. UAV campaigns along corridor areas require well-distributed ground control points (GCP) and accurate geotagged image positions to avoid misalignment errors of generated digital surface models (DSMs). However, a sandy terrain can make the measurement of GCPs difficult or even impossible. Furthermore, UAVs are often equipped with consumer-grade devices (onboard GNSS and non-metric cameras), which, therefore, cannot provide accurate image positions and lead to geometric distortions of the image network due to inappropriate lens distortion corrections. This paper proposes a strategy to calibrate the used camera and co-register multi-temporal images without the need for accurate geotagged information on the whole datasets and well-distributed GCPs. The proposed strategy includes two improvements. First, we performed semi-on-the-job self-calibration (semi-OTJSC) using UAV images with favor orientations and flight altitudes to pre-calibrate interior orientation parameters (IOPs) and additional lens distortion parameters. This was then followed by another OTJSC of IOPs involving images with accurate geotagged positions. Second, a &#x201C;transferred aerial-triangulation (Trans-AT)&#x201D; strategy is proposed to co-register two consecutive UAV datasets through AT procedure similar to GNSS-supported AT within a strip image block (SIB). The experimental results revealed that the proposed method provides the best-fit camera parameters and generates high-accuracy co-registered DSMs. This strategy contributes significantly to multi-temporal camera calibration and co-registration procedures for determining morphological changes in corridor mapping.

Urban Surface Water Mapping from VHR Images Based on Superpixel Segmentation and Target Detection

Surface water is a fundamental resource in urban environments. Monitoring the spatio-temporal distribution of urban surface water from remotely sensed images is crucial for urban planning and management. Unfortunately, due to the limitation of spatial resolution, the method based on low&#x002F;medium resolution images is difficult to extract small water bodies accurately. Recently, very high resolution (VHR) images have shown considerable potential for urban compositions mapping. However, fewer spectral bands, shadows, and high spectral heterogeneity of VHR images hinder the application of traditional methods. In this study, we proposed an urban surface water mapping method called sparse superpixel-based water extraction (SSWE) from VHR images. The method includes three steps: (1) clustering water bodies into sparse targets at the object level by an improved scale-adaptive simple non-iterative clustering (SA-SNIC) superpixel segmentation; (2) generating new bands with additional spectral, spatial, and derived features, to increase the dimensions of original data and enhance the separability between water bodies and background covers; (3) constructing a positive-negative constrained energy minimization (PNCEM) multi-target sparse detector to highlight the water bodies while suppressing shadows. The proposed method was applied to GF-2 multispectral images of four cities in China. The results showed that SSWE achieved the highest accuracy compared with other methods, with an average OA of 98.91&#x0025; and an average kappa coefficient of 0.942. Furthermore, the separability analysis also indicated that SSWE could effectively distinguish urban water bodies from shadows and other land covers. And stable results can be acquired by the suggested parameters and thresholds of SSWE.

Transformer and CNN Hybrid Deep Neural Network for Semantic Segmentation of Very-High-Resolution Remote Sensing Imagery

This article presents a transformer and convolutional neural network (CNN) hybrid deep neural network for semantic segmentation of very high resolution (VHR) remote sensing imagery. The model follows an encoder–decoder structure. The encoder module uses a new universal backbone Swin transformer to extract features to achieve better long-range spatial dependencies modeling. The decoder module draws on some effective blocks and successful strategies of CNN-based models in remote sensing image segmentation. In the middle of the framework, an atrous spatial pyramid pooling block based on depthwise separable convolution (SASPP) is applied to obtain a multiscale context. A U-shaped decoder is used to gradually restore the size of the feature maps. Three skip connections are built between the encoder and decoder feature maps of the same size to maintain the transmission of local details and enhance the communication of multiscale features. A squeeze-and-excitation (SE) channel attention block is added before segmentation for feature augmentation. An auxiliary boundary detection branch is combined to provide edge constraints for semantic segmentation. Extensive ablation experiments were conducted on the International Society for Photogrammetry and Remote Sensing (ISPRS) Vaihingen and Potsdam benchmarks to test the effectiveness of multiple components of the network. At the same time, the proposed method is compared with the current state-of-the-art methods on the two benchmarks. The proposed hybrid network achieved the second highest overall accuracy (OA) on both the Potsdam and Vaihingen benchmarks (code and models are available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/zq7734509/mmsegmentation-</uri> multilayer).

HENet: Head-Level Ensemble Network for Very High Resolution Remote Sensing Images Semantic Segmentation

Semantic segmentation plays an important role in very high resolution (VHR) image understanding. Despite the potentials of the deep convolutional network in improving performance by end-to-end feature learning, each model has its limitations, and it is hard to discriminate complex features purely by a single model. Ensemble learning is promising for integrating the strengths of different models, however, the ensemble of deep models is challenging due to the huge amount of parameters and computation of the deep model itself as well as the difficulty in capturing complementarity between different models. To tackle these problems, a head-level ensemble network (HENet) is proposed in this letter, which reduces model complexity by sharing feature extraction networks and improves complementarity between models by novel cooperative learning (CL). Experiments on ISPRS 2-D semantic labeling benchmark demonstrate the effectiveness and advantage of the proposed method.

Image Defocus in an Airborne UWB VHR Microwave Photonic SAR: Analysis and Compensation

With the exploration of synthetic aperture radar (SAR), the requirements for its functionality, precision, and response time are inevitably increasing, in which the technical core is the generation, reception, and processing of wideband signals with low time cost. Owing to the excellent performance of modern photonics, such as the ultra-wide bandwidth (UWB), flat response, low transmission loss, fast analog signal processing, and microwave photonics (MWP) promises to be an appropriate solution to improve the capability of SAR in resolution, coverage, and efficiency. However, on the one hand, due to the high sensitivity of optical fiber, moisture, temperature, or physical vibration can produce an unknown extra propagation delay. On the other hand, the wavelength shift effect should be taken into consideration for the UWB system. Thereby, under very high resolution (VHR) circumstance, two-dimensional (2-D) defocus including migration through resolution cells (MTRC) and azimuth phase error (APE) becomes a challenge for MWP SAR imaging. Unfortunately, existing 2-D autofocus approaches concerning motion errors inherently fail for the commonly underlying assumption that the prominent nonsystematic residual range cell migration (RCM) is global in the azimuth time domain. In this article, we analyze the effects of the above-mentioned negative factors in the image processing of range migration algorithm (RMA) and reveal the structural characteristics of the 2-D phase errors. A novel two-step postprocessing compensation strategy is developed, and experiments on real data acquired by an airborne MWP SAR system demonstrate its effectiveness.