Search

The traditional method for retrieving cloud optical thickness (COT) is carried out through a Look-Up Table (LUT). Researchers must make a series of idealized assumptions and conduct extensive observations and record features in this scenario, consuming considerable resources. The emergence of deep learning effectively addresses the shortcomings of the traditional approach. In this paper, we first propose a daytime (SOZA < 70°) COT retrieval algorithm based on FY-4A AGRI. We establish and train a Convolutional Neural Network (CNN) model for COT retrieval, CM4CR, with the CALIPSO's COT product spatially and temporally synchronized as the ground truth. Then, a deep learning method extended from video prediction models is adopted to predict COT values based on the retrieval results obtained from CM4CR. The COT prediction model (CPM) consists of an encoder, a predictor, and a decoder. On this basis, we further incorporated a time embedding module to enhance the model's ability to learn from irregular time intervals in the input COT sequence. During the training phase, we employed Charbonnier Loss and Edge Loss to enhance the model's capability to represent COT details. Experiments indicate that our CM4CR outperforms existing COT retrieval methods, with predictions showing better performance across several metrics than other benchmark prediction models. Additionally, this paper also investigates the impact of different lengths of COT input sequences and the time intervals between adjacent frames of COT on prediction performance. [ABSTRACT FROM AUTHOR]

Sea surface temperature (SST) is a key parameter in ocean hydrology. Currently, existing SST prediction methods fail to fully utilize the potential spatial correlation between variables. To address this challenge, we propose a spatiotenporal UNet (ST-UNet) model based on the UNet model. In particular, in the encoding phase of ST-UNet, we use parallel convolution with different kernel sizes to efficiently extract spatial features, and use ConvLSTM to capture temporal features based on the utilization of spatial features. Atrous Spatial Pyramid Pooling (ASPP) module is placed at the bottleneck of the network to further incorporate the multi-scale features, allowing the spatial features to be fully utilized. The final prediction is then generated in the decoding stage using parallel convolution with different kernel sizes similar to the encoding stage. We conducted a series of experiments on the Bohai Sea and Yellow Sea SST data set, as well as the South China Sea SST data set, using SST data from the past 35 days to predict SST data for 1, 3, and 7 days in the future. The model was trained using data spanning from 2010 to 2021, with data from 2022 being utilized to assess the model's predictive performance. The experimental results show that the model proposed in this research paper achieves excellent results at different prediction scales in both sea areas, and the model consistently outperforms other methods. Specifically, in the Bohai Sea and Yellow Sea sea areas, when the prediction scales are 1, 3, and 7 days, the MAE of ST-UNet outperforms the best results of the other three compared models by 17%, 12%, and 2%, and the MSE by 16%, 18%, and 9%, respectively. In the South China Sea, when the prediction ranges are 1, 3, and 7 days, the MAE of ST-UNet is 27%, 18%, and 3% higher than the best of the other three compared models, and the MSE is 46%, 39%, and 16% higher, respectively. Our results highlight the effectiveness of the ST-UNet model in capturing spatial correlations and accurately predicting SST. The proposed model is expected to improve marine hydrographic studies. [ABSTRACT FROM AUTHOR]

Ship detection aims to automatically identify whether there are ships in the images, precisely classifies and localizes them. Regardless of whether utilizing early manually designed methods or deep learning technology, ship detection is dedicated to exploring the inherent characteristics of ships to enhance recall. Nowadays, high-precision ship detection plays a crucial role in civilian and military applications. In order to provide a comprehensive review of ship detection in optical remote-sensing images (SDORSIs), this paper summarizes the challenges as a guide. These challenges include complex marine environments, insufficient discriminative features, large scale variations, dense and rotated distributions, large aspect ratios, and imbalances between positive and negative samples. We meticulously review the improvement methods and conduct a detailed analysis of the strengths and weaknesses of these methods. We compile ship information from common optical remote sensing image datasets and compare algorithm performance. Simultaneously, we compare and analyze the feature extraction capabilities of backbones based on CNNs and Transformer, seeking new directions for the development in SDORSIs. Promising prospects are provided to facilitate further research in the future. [ABSTRACT FROM AUTHOR]

Despite notable advancements achieved on Hyperspectral (HS) pansharpening tasks through deep learning techniques, previous methods are inherently constrained by convolution or self-attention intrinsic defects, leading to limited performance. In this paper, we proposed an Attention-Interactive Dual-Branch Convolutional Neural Network (AIDB-Net) for HS pansharpening. Our model purely consists of convolutional layers and simultaneously inherits the strengths of both convolution and self-attention, especially the modeling of short- and long-range dependencies. Specially, we first extract, tokenize, and align the hyperspectral image (HSI) and panchromatic image (PAN) by Overlapping Patch Embedding Blocks. Then, we specialize a novel Spectral-Spatial Interactive Attention which is able to globally interact and fuse the cross-modality features. The resultant token-global similarity scores can guide the refinement and renewal of the textural details and spectral characteristics within HSI features. By deeply combined these two paradigms, our AIDB-Net significantly improve the pansharpening performance. Moreover, with the acceleration by the convolution inductive bias, our interactive attention can be trained without large scale dataset and achieves competitive time cost with its counterparts. Compared with the state-of-the-art methods, our AIDB-Net makes 5.2%, 3.1%, and 2.2% improvement on PSNR metric on three public datasets, respectively. Comprehensive experiments quantitatively and qualitatively demonstrate the effectiveness and superiority of our AIDB-Net. [ABSTRACT FROM AUTHOR]

Several deep learning and transformer models have been recommended in previous research to deal with the classification of hyperspectral images (HSIs). Among them, one of the most innovative is the bidirectional encoder representation from transformers (BERT), which applies a distance-independent approach to capture the global dependency among all pixels in a selected region. However, this model does not consider the local spatial-spectral and spectral sequential relations. In this paper, a dual-dimensional (i.e., spatial and spectral) BERT (the so-called D2BERT) is proposed, which improves the existing BERT model by capturing more global and local dependencies between sequential spectral bands regardless of distance. In the proposed model, two BERT branches work in parallel to investigate relations among pixels and spectral bands, respectively. In addition, the layer intermediate information is used for supervision during the training phase to enhance the performance. We used two widely employed datasets for our experimental analysis. The proposed D2BERT shows superior classification accuracy and computational efficiency with respect to some state-of-the-art neural networks and the previously developed BERT model for this task. [ABSTRACT FROM AUTHOR]

The new generation of available (i.e., PRISMA, ENMAP, DESIS) and future (i.e., ESA-CHIME, NASA-SBG) spaceborne hyperspectral missions provide unprecedented data for environmental and agricultural monitoring, such as crop trait assessment. This paper focuses on retrieving two crop traits, specifically Chlorophyll and Nitrogen content at the canopy level (CCC and CNC), starting from hyperspectral images acquired during the CHIME-RCS project, exploiting a self-supervised learning (SSL) technique. SSL is a machine learning paradigm that leverages unlabeled data to generate valuable representations for downstream tasks, bridging the gap between unsupervised and supervised learning. The proposed method comprises pre-training and fine-tuning procedures: in the first stage, a de-noising Convolutional Autoencoder is trained using pairs of noisy and clean CHIME-like images; the pre-trained Encoder network is utilized as-is or fine-tuned in the second stage. The paper demonstrates the applicability of this technique in hybrid approach methods that combine Radiative Transfer Modelling (RTM) and Machine Learning Regression Algorithm (MLRA) to set up a retrieval schema able to estimate crop traits from new generation space-born hyperspectral data. The results showcase excellent prediction accuracy for estimating CCC (R2 = 0.8318; RMSE = 0.2490) and CNC (R2 = 0.9186; RMSE = 0.7908) for maize crops from CHIME-like images without requiring further ground data calibration. [ABSTRACT FROM AUTHOR]

Earthquake Disaster Assessment (EDA) plays a critical role in earthquake disaster prevention, evacuation, and rescue efforts. Deep learning (DL), which boasts advantages in image processing, signal recognition, and object detection, has facilitated scientific research in EDA. This paper analyses 204 articles through a systematic literature review to investigate the status quo, development, and challenges of DL for EDA. The paper first examines the distribution characteristics and trends of the two categories of EDA assessment objects, including earthquakes and secondary disasters as disaster objects, buildings, infrastructure, and areas as physical objects. Next, this study analyses the application distribution, advantages, and disadvantages of the three types of data (remote sensing data, seismic data, and social media data) mainly involved in these studies. Furthermore, the review identifies the characteristics and application of six commonly used DL models in EDA, including convolutional neural network (CNN), multi-layer perceptron (MLP), recurrent neural network (RNN), generative adversarial network (GAN), transfer learning (TL), and hybrid models. The paper also systematically details the application of DL for EDA at different times (i.e., pre-earthquake stage, during-earthquake stage, post-earthquake stage, and multi-stage). We find that the most extensive research in this field involves using CNNs for image classification to detect and assess building damage resulting from earthquakes. Finally, the paper discusses challenges related to training data and DL models, and identifies opportunities in new data sources, multimodal DL, and new concepts. This review provides valuable references for scholars and practitioners in related fields. [ABSTRACT FROM AUTHOR]

The despeckling of synthetic aperture radar images using two different convolutional neural network architectures is presented in this paper. The first method presents a novel Siamese convolutional neural network with a dilated convolutional network in each branch. Recently, attention mechanisms have been introduced to convolutional networks to better model and recognize features. Therefore, we propose a novel design for a convolutional neural network using an attention mechanism for an encoder–decoder-type network. The framework consists of a multiscale spatial attention network to improve the modeling of semantic information at different spatial levels and an additional attention mechanism to optimize feature propagation. Both proposed methods are different in design but they provide comparable despeckling results in subjective and objective measurements in terms of correlated speckle noise. The experimental results are evaluated on both synthetically generated speckled images and real SAR images. The methods proposed in this paper are able to despeckle SAR images and preserve SAR features. [ABSTRACT FROM AUTHOR]

This paper presents the design of a high-voltage pulse-based radar and a supervised data processing method for soil moisture estimation. The goal of this research was to design a pulse-based radar to detect changes in soil moisture using a cross-hole approach. The pulse-based radar with three transmitting antennas was placed into a 12 m deep hole, and a receiver with three receive antennas was placed into a different hole separated by 100 m from the transmitter. The pulse generator was based on a Marx generator with an LC filter, and for the receiver, the high-frequency data acquisition card was used, which can acquire signals using 3 Gigabytes per second. Used borehole antennas were designed to operate in the wide frequency band to ensure signal propagation through the soil. A deep regression convolutional network is proposed in this paper to estimate volumetric soil moisture using time-sampled signals. A regression convolutional network is extended to three dimensions to model changes in wave propagation between the transmitted and received signals. The training dataset was acquired during the period of 73 days of acquisition between two boreholes separated by 100 m. The soil moisture measurements were acquired at three points 25 m apart to provide ground truth data. Additionally, water was poured into several specially prepared boreholes between transmitter and receiver antennas to acquire additional dataset for training, validation, and testing of convolutional neural networks. Experimental results showed that the proposed system is able to detect changes in the volumetric soil moisture using Tx and Rx antennas. [ABSTRACT FROM AUTHOR]

Due to the advantages of Synthetic Aperture Radar (SAR), the study of Automatic Target Recognition (ATR) has become a hot topic. Deep learning, especially in the case of a Convolutional Neural Network (CNN), works in an end-to-end way and has powerful feature-extracting abilities. Thus, researchers in SAR ATR also seek solutions from deep learning. We review the related algorithms with regard to SAR ATR in this paper. We firstly introduce the commonly used datasets and the evaluation metrics. Then, we introduce the algorithms before deep learning. They are template-matching-, machine-learning- and model-based methods. After that, we introduce mainly the SAR ATR methods in the deep-learning era (after 2017); those methods are the core of the paper. The non-CNNs and CNNs, that is, those used in SAR ATR, are summarized at the beginning. We found that researchers tend to design specialized CNN for SAR ATR. Then, the methods to solve the problem raised by limited samples are reviewed. They are data augmentation, Generative Adversarial Networks (GAN), electromagnetic simulation, transfer learning, few-shot learning, semi-supervised learning, metric leaning and domain knowledge. After that, the imbalance problem, real-time recognition, polarimetric SAR, complex data and adversarial attack are also reviewed. The principles and problems of them are also introduced. Finally, the future directions are conducted. In this part, we point out that the dataset, CNN architecture designing, knowledge-driven, real-time recognition, explainable and adversarial attack should be considered in the future. This paper gives readers a quick overview of the current state of the field. [ABSTRACT FROM AUTHOR]

Seismic impedance inversion is a vital way of geological interpretation and reservoir investigation from a geophysical perspective. However, it is inevitably an ill-posed problem due to the noise or the band-limited characteristic of seismic data. Artificial neural network have been used to solve nonlinear inverse problems in recent years. This research obtained an acoustic impedance profile by feeding seismic profile and background impedance into a well-trained self-attention U-Net. The U-Net got convergence by appropriate iteration, and the output predicted the impedance profiles in the test. To value the quality of predicted profiles from different perspectives, e.g., correlation, regression, and similarity, we used four kinds of indexes. At the same time, our results were predicted by conventional methods (e.g., deconvolution with recursive inversion, and TV regularization) and a 1D neural network was calculated in contrast. Self-attention U-Net showed to be robust to noise and does not require prior knowledge. Furthermore, spatial continuity is also better than deconvolution, regularization, and 1D deep learning methods in contrast. The U-Net in this paper is a type of full convolutional neural network, so there are no limits to the shape of the input. Based on this, a large impedance profile can be predicted by U-Net, which is trained by a patchy training dataset. In addition, this paper applied the proposed method to the field data obtained by the Ceduna survey without any label. The predictions prove that this well-trained network could be generalized from synthetic data to field data. [ABSTRACT FROM AUTHOR]

With the continuous progress of remote sensing image object detection tasks in recent years, researchers in this field have gradually shifted the focus of their research from horizontal object detection to the study of object detection in arbitrary directions. It is worth noting that some properties are different from horizontal object detection during oriented object detection that researchers have yet to notice much. This article presents the design of a straightforward and efficient arbitrary-oriented detection system, leveraging the inherent properties of the orientation task, including the rotation angle and box aspect ratio. In the detection of low aspect ratio objects, the angle is of little importance to the orientation bounding box, and it is even difficult to define the angle information in extreme categories. Conversely, in the detection of objects with high aspect ratios, the angle information plays a crucial role and can have a decisive impact on the quality of the detection results. By exploiting the aspect ratio of different targets, this letter proposes a ratio-balanced angle loss that allows the model to make a better trade-off between low-aspect ratio objects and high-aspect ratio objects. The rotation angle of each oriented object, which we naturally embed into a two-dimensional Euclidean space for regression, thus avoids an overly redundant design and preserving the topological properties of the circular space. The performance of the UCAS-AOD, HRSC2016, and DLR-3K datasets show that the proposed model in this paper achieves a leading level in terms of both accuracy and speed. [ABSTRACT FROM AUTHOR]

In recent years, the rapid development of deep learning technology has brought new opportunities for specific emitter identification and has greatly improved the performance of radar emitter identification. The most specific emitter identification methods, based on deep learning, have focused more on studying network structures and data preprocessing. However, the data selection and utilization have a significant impact on the emitter recognition efficiency, and the method to adaptively determine the two parameters by a specific recognition model has yet to be studied. This paper proposes a knowledge graph-driven convolutional neural network (KG-1D-CNN) to solve this problem. The relationship network between radar data is modeled via the knowledge graph and uses 1D-CNN as the metric kernel to measure these relationships in the knowledge graph construction process. In the recognition process, a precise dataset is constructed based on the knowledge graph according to the task requirement. The network is designed to recognize target emitter individuals from easy to difficult by the precise dataset. In the experiments, most algorithms achieved good recognition results in the high SNR case (10–15 dB), while only the proposed method could achieve more than a 90% recognition rate in the low SNR case (0–5 dB). The experimental results demonstrate the efficacy of the proposed method. [ABSTRACT FROM AUTHOR]

The deep-learning-based image super-resolution opens a new direction for the remote sensing field to reconstruct further information and details from captured images. However, most current SR works try to improve the performance by increasing the complexity of the model, which results in significant computational costs and memory consumption. In this paper, we propose a lightweight model named pixel-wise attention residual network for optical remote sensor images, which can effectively solve the super-resolution task of multi-satellite images. The proposed method consists of three modules: the feature extraction module, feature fusion module, and feature mapping module. First, the feature extraction module is responsible for extracting the deep features from the input spatial bands with different spatial resolutions. Second, the feature fusion module with the pixel-wise attention mechanism generates weight coefficients for each pixel on the feature map and fully fuses the deep feature information. Third, the feature mapping module is aimed to maintain the fidelity of the spectrum by adding the fused residual feature map directly to the up-sampled low-resolution images. Compared with existing deep-learning-based methods, the major advantage of our method is that for the first time, the pixel-wise attention mechanism is incorporated in the task of super-resolution fusion of remote sensing images, which effectively improved the performance of the fusion network. The accuracy assessment results show that our method achieved superior performance of the root mean square error, signal-to–reconstruction ratio error, universal image quality index, and peak signal noise ratio compared to competing approaches. The improvements in the signal-to-reconstruction ratio error and peak signal noise ratio are significant, with a respective increase of 0.15 and 0.629 dB for Sentinel-2 data, and 0.196 and 1 dB for Landsat data. [ABSTRACT FROM AUTHOR]

The precise classification of crop types using hyperspectral remote sensing imaging is an essential application in the field of agriculture, and is of significance for crop yield estimation and growth monitoring. Among the deep learning methods, Convolutional Neural Networks (CNNs) are the premier model for hyperspectral image (HSI) classification for their outstanding locally contextual modeling capability, which facilitates spatial and spectral feature extraction. Nevertheless, the existing CNNs have a fixed shape and are limited to observing restricted receptive fields, constituting a simulation difficulty for modeling long-range dependencies. To tackle this challenge, this paper proposed two novel classification frameworks which are both built from multilayer perceptrons (MLPs). Firstly, we put forward a dilation-based MLP (DMLP) model, in which the dilated convolutional layer replaced the ordinary convolution of MLP, enlarging the receptive field without losing resolution and keeping the relative spatial position of pixels unchanged. Secondly, the paper proposes multi-branch residual blocks and DMLP concerning performance feature fusion after principal component analysis (PCA), called DMLPFFN, which makes full use of the multi-level feature information of the HSI. The proposed approaches are carried out on two widely used hyperspectral datasets: Salinas and KSC; and two practical crop hyperspectral datasets: WHU-Hi-LongKou and WHU-Hi-HanChuan. Experimental results show that the proposed methods outshine several state-of-the-art methods, outperforming CNN by 6.81%, 12.45%, 4.38% and 8.84%, and outperforming ResNet by 4.48%, 7.74%, 3.53% and 6.39% on the Salinas, KSC, WHU-Hi-LongKou and WHU-Hi-HanChuan datasets, respectively. As a result of this study, it was confirmed that the proposed methods offer remarkable performances for hyperspectral precise crop classification. [ABSTRACT FROM AUTHOR]

Advances in deep learning techniques for remote sensing as well as the increased availability of high-resolution data enable the extraction of more detailed information from aerial images. One promising task is the semantic segmentation of roof segments and their orientation. However, the lack of annotated data is a major barrier for deploying respective models on a large scale. Previous research demonstrated the viability of the deep learning approach for the task, but currently, published datasets are small-scale, manually labeled, and rare. Therefore, this paper extends the state of the art by presenting a novel method for the automated generation of large-scale datasets based on semantic 3D city models. Furthermore, we train a model on a dataset 50 times larger than existing datasets and achieve superior performance while applying it to a wider variety of buildings. We evaluate the approach by comparing networks trained on four dataset configurations, including an existing dataset and our novel large-scale dataset. The results show that the network performance measured as intersection over union can be increased from 0.60 for the existing dataset to 0.70 when the large-scale model is applied on the same region. The large-scale model performs superiorly even when applied to more diverse test samples, achieving 0.635. The novel approach contributes to solving the dataset bottleneck and consequently to improving semantic segmentation of roof segments. The resulting remotely sensed information is crucial for applications such as solar potential analysis or urban planning. [ABSTRACT FROM AUTHOR]

Deep reflection seismic data are usually accompanied by large-offset data, and the accurate and rapid identification of the first arrivals of seismic records plays an important role in eliminating the effects of topography and other factors that increase with the increasing offsets. In this paper, we propose a method based on convolutional neural networks (CNNs) that can accurately identify the first arrivals of large-offset seismic data. A time window for linear dynamic correction was established to convert the raw seismic data into rectangular images so as to reduce the amount of invalid sample data and improve the training efficiency. In order to enhance the prediction effect of the far-offset first arrivals, we propose the strategy of adjusting the weight of the far-offset data to increase the weight of the far-offset data in the training dataset and, thus, to improve the first arrival accuracy. The manually picked first arrivals are used as labels and the input to the CNNs for training, and the full-offset first arrivals are the output. The travel time tomography velocity is modeled and compared based on the first arrivals obtained through manual picking, industrial software automatic picking, and CNN prediction. The results show that the application of CNNs to large-offset seismic datasets can help researchers to obtain the first arrivals at different offsets, while the inclusion of far-offset weights can effectively improve the modeling depth of the tomography inversion, and the accuracy of the results is high. [ABSTRACT FROM AUTHOR]

Global solar photovoltaic capacity has consistently doubled every 18 months over the last two decades, going from 0.3 GW in 2000 to 643 GW in 2019, and is forecast to reach 4240 GW by 2040. However, these numbers are uncertain, and virtually all reporting on deployments lacks a unified source of either information or validation. In this paper, we propose, optimize, and validate a deep learning framework to detect and map solar farms using a state-of-the-art semantic segmentation convolutional neural network applied to satellite imagery. As a final step in the pipeline, we propose a model to estimate the energy generation capacity of the detected solar energy facilities. Objectively, the deep learning model achieved highly competitive performance indicators, including a mean accuracy of 96.87%, and a Jaccard Index (intersection over union of classified pixels) score of 95.5%. Subjectively, it was found to detect spaces between panels producing a segmentation output at a sub-farm level that was better than human labeling. Finally, the detected areas and predicted generation capacities were validated against publicly available data to within an average error of 4.5% Deep learning applied specifically for the detection and mapping of solar farms is an active area of research, and this deep learning capacity evaluation pipeline is one of the first of its kind. We also share an original dataset of overhead solar farm satellite imagery comprising 23,000 images (256 × 256 pixels each), and the corresponding labels upon which the machine learning model was trained. [ABSTRACT FROM AUTHOR]

Change detection (CD) is a crucial task in remote sensing (RS) to distinguish surface changes from bitemporal images. Recently, deep learning (DL) based methods have achieved remarkable success for CD. However, the existing methods lack robustness to various kinds of changes in RS images, which suffered from problems of feature misalignment and inefficient supervision. In this paper, a deeply supervised attentive high-resolution network (DSAHRNet) is proposed for remote sensing image change detection. First, we design a spatial-channel attention module to decode change information from bitemporal features. The attention module is able to model spatial-wise and channel-wise contexts. Second, to reduce feature misalignment, the extracted features are refined by stacked convolutional blocks in parallel. Finally, a novel deeply supervised module is introduced to generate more discriminative features. Extensive experimental results on three challenging benchmark datasets demonstrate that the proposed DSAHRNet outperforms other state-of-the-art methods, and achieves a great trade-off between performance and complexity. [ABSTRACT FROM AUTHOR]

Hyperspectral sharpening has been considered an important topic in many earth observation applications. Many studies have been performed to solve the Visible-Near-Infrared (Vis-NIR) hyperpectral sharpening problem, but there is little research related to hyperspectral sharpening including short-wave infrared (SWIR) bands despite many hyperspectral imaging systems capturing this wavelength range. In this paper, we introduce a novel method to achieve full-spectrum hyperspectral sharpening by fusing the high-resolution (HR) Vis-NIR multispectral image (MSI) and the Vis-NIR-SWIR low-resolution (LR) hyperspectral image (HSI). The novelty of the proposed approach lies in three points. Firstly, our model is designed for sharpening the full-spectrum HSI with high radiometric accuracy. Secondly, unlike most of the big-dataset-driven deep learning models, we only need one LR-HSI and HR-MSI pair for training. Lastly, per-pixel classification is implemented to test the spectral accuracy of the results. [ABSTRACT FROM AUTHOR]

In recent years, deep-learning-based hyperspectral image (HSI) classification networks have become one of the most dominant implementations in HSI classification tasks. Among these networks, convolutional neural networks (CNNs) and attention-based networks have prevailed over other HSI classification networks. While convolutional neural networks with perceptual fields can effectively extract local features in the spatial dimension of HSI, they are poor at capturing the global and sequential features of spectral–spatial information; networks based on attention mechanisms, for example, Transformer, usually have better ability to capture global features, but are relatively weak in discriminating local features. This paper proposes a fusion network of convolution and Transformer for HSI classification, known as FusionNet, in which convolution and Transformer are fused in both serial and parallel mechanisms to achieve the full utilization of HSI features. Experimental results demonstrate that the proposed network has superior classification results compared to previous similar networks, and performs relatively well even on a small amount of training data. [ABSTRACT FROM AUTHOR]

In recent years, hyperspectral image (HSI) classification (HSIC) methods that use deep learning have proved to be effective. In particular, the utilization of convolutional neural networks (CNNs) has proved to be highly effective. However, some key issues need to be addressed when classifying hyperspectral images (HSIs), such as small samples, which can influence the generalization ability of the CNNs and the HSIC results. To address this problem, we present a new network that integrates hybrid pyramid feature fusion and coordinate attention for enhancing small sample HSI classification results. The innovative nature of this paper lies in three main areas. Firstly, a baseline network is designed. This is a simple hybrid 3D-2D CNN. Using this baseline network, more robust spectral-spatial feature information can be obtained from the HSI. Secondly, a hybrid pyramid feature fusion mechanism is used, meaning that the feature maps of different levels and scales can be effectively fused to enhance the feature extracted by the model. Finally, coordinate attention mechanisms are utilized in the network, which can not only adaptively capture the information of the spectral dimension, but also include the direction-aware and position sensitive information. By doing this, the proposed CNN structure can extract more useful HSI features and effectively be generalized to test samples. The proposed method was shown to obtain better results than several existing methods by experimenting on three public HSI datasets. [ABSTRACT FROM AUTHOR]

Three-dimensional (3D) point cloud classification methods based on deep learning have good classification performance; however, they adapt poorly to diverse datasets and their classification accuracy must be improved. Therefore, FSDCNet, a neural network model based on the fusion of static and dynamic convolution, is proposed and applied for multiview 3D point cloud classification in this paper. FSDCNet devises a view selection method with fixed and random viewpoints, which effectively avoids the overfitting caused by the traditional fixed viewpoint. A local feature extraction operator of dynamic and static convolution adaptive weight fusion was designed to improve the model's adaptability to different types of datasets. To address the problems of large parameters and high computational complexity associated with the current methods of dynamic convolution, a lightweight and adaptive dynamic convolution operator was developed. In addition, FSDCNet builds a global attention pooling, integrating the most crucial information on different view features to the greatest extent. Due to these characteristics, FSDCNet is more adaptable, can extract more fine-grained detailed information, and can improve the classification accuracy of point cloud data. The proposed method was applied to the ModelNet40 and Sydney Urban Objects datasets. In these experiments, FSDCNet outperformed its counterparts, achieving state-of-the-art point cloud classification accuracy. For the ModelNet40 dataset, the overall accuracy (OA) and average accuracy (AA) of FSDCNet in a single view reached 93.8% and 91.2%, respectively, which were superior to those values for many other methods using 6 and 12 views. FSDCNet obtained the best results for 6 and 12 views, achieving 94.6%, 93.3%, 95.3%, and 93.6% in OA and AA metrics, respectively. For the Sydney Urban Objects dataset, FSDCNet achieved an OA and F1 score of 81.2% and 80.1% in a single view, respectively, which were higher than most of the compared methods. In 6 and 12 views, FSDCNet reached an OA of 85.3% and 83.6% and an F1 score of 85.5% and 83.7%, respectively. [ABSTRACT FROM AUTHOR]

In recent years, due to its powerful feature extraction ability, the deep learning method has been widely used in hyperspectral image classification tasks. However, the features extracted by classical deep learning methods have limited discrimination ability, resulting in unsatisfactory classification performance. In addition, due to the limited data samples of hyperspectral images (HSIs), how to achieve high classification performance under limited samples is also a research hotspot. In order to solve the above problems, this paper proposes a deep learning network framework named the three-dimensional coordination attention mechanism network (3DCAMNet). In this paper, a three-dimensional coordination attention mechanism (3DCAM) is designed. This attention mechanism can not only obtain the long-distance dependence of the spatial position of HSIs in the vertical and horizontal directions, but also obtain the difference of importance between different spectral bands. In order to extract the spectral and spatial information of HSIs more fully, a convolution module based on convolutional neural network (CNN) is adopted in this paper. In addition, the linear module is introduced after the convolution module, which can extract more fine advanced features. In order to verify the effectiveness of 3DCAMNet, a series of experiments were carried out on five datasets, namely, Indian Pines (IP), Pavia University (UP), Kennedy Space Center (KSC), Salinas Valley (SV), and University of Houston (HT). The OAs obtained by the proposed method on the five datasets were 95.81%, 97.01%, 99.01%, 97.48%, and 97.69% respectively, 3.71%, 9.56%, 0.67%, 2.89% and 0.11% higher than the most advanced A2S2K-ResNet. Experimental results show that, compared with some state-of-the-art methods, 3DCAMNet not only has higher classification performance, but also has stronger robustness. [ABSTRACT FROM AUTHOR]

Recently, with the extensive application of deep learning techniques in the hyperspectral image (HSI) field, particularly convolutional neural network (CNN), the research of HSI classification has stepped into a new stage. To avoid the problem that the receptive field of naive convolution is small, the dilated convolution is introduced into the field of HSI classification. However, the dilated convolution usually generates blind spots in the receptive field, resulting in discontinuous spatial information obtained. In order to solve the above problem, a densely connected pyramidal dilated convolutional network (PDCNet) is proposed in this paper. Firstly, a pyramidal dilated convolutional (PDC) layer integrates different numbers of sub-dilated convolutional layers is proposed, where the dilated factor of the sub-dilated convolution increases exponentially, achieving multi-sacle receptive fields. Secondly, the number of sub-dilated convolutional layers increases in a pyramidal pattern with the depth of the network, thereby capturing more comprehensive hyperspectral information in the receptive field. Furthermore, a feature fusion mechanism combining pixel-by-pixel addition and channel stacking is adopted to extract more abstract spectral–spatial features. Finally, in order to reuse the features of the previous layers more effectively, dense connections are applied in densely pyramidal dilated convolutional (DPDC) blocks. Experiments on three well-known HSI datasets indicate that PDCNet proposed in this paper has good classification performance compared with other popular models. [ABSTRACT FROM AUTHOR]

In high-resolution earth observation systems, object detection in high spatial resolution remote sensing images (HSRIs) is the key technology for automatic extraction, analysis and understanding of image information. With respect to the multi-angle features of object orientation in HSRIs object detection, this paper presents a novel HSRIs object detection method based on convolutional neural networks (CNN) with adaptive object orientation features. First, an adaptive object orientation regression method is proposed to obtain object regions in any direction. In the adaptive object orientation regression method, five coordinate parameters are used to regress the object region with any direction. Then, a CNN framework for object detection of HSRIs is designed using the adaptive object orientation regression method. Using multiple object detection datasets, the proposed method is compared with some state-of-the-art object detection methods. The experimental results show that the proposed method can more accurately detect objects with large aspect ratios and densely distributed objects than some state-of-the-art object detection methods using a horizontal bounding box, and obtain better object detection results for HSRIs. [ABSTRACT FROM AUTHOR]

With the increasing availability of SAR imagery in recent years, more research is being conducted using deep learning (DL) for the classification of ice and open water; however, ice and open water classification using conventional DL methods such as convolutional neural networks (CNNs) is not yet accurate enough to replace manual analysis for operational ice chart mapping. Understanding the uncertainties associated with CNN model predictions can help to quantify errors and, therefore, guide efforts on potential enhancements using more–advanced DL models and/or synergistic approaches. This paper evaluates an approach for estimating the aleatoric uncertainty [a measure used to identify the noise inherent in data] of CNN probabilities to map ice and open water with a custom loss function applied to RADARSAT–2 HH and HV observations. The images were acquired during the 2014 ice season of Lake Erie and Lake Ontario, two of the five Laurentian Great Lakes of North America. Operational image analysis charts from the Canadian Ice Service (CIS), which are based on visual interpretation of SAR imagery, are used to provide training and testing labels for the CNN model and to evaluate the accuracy of the model predictions. Bathymetry, as a variable that has an impact on the ice regime of lakes, was also incorporated during model training in supplementary experiments. Adding aleatoric loss and bathymetry information improved the accuracy of mapping water and ice. Results are evaluated quantitatively (accuracy metrics) and qualitatively (visual comparisons). Ice and open water scores were improved in some sections of the lakes by using aleatoric loss and including bathymetry. In Lake Erie, the ice score was improved by ∼2 on average in the shallow near–shore zone as a result of better mapping of dark ice (low backscatter) in the western basin. As for Lake Ontario, the open water score was improved by ∼6 on average in the deepest profundal off–shore zone. [ABSTRACT FROM AUTHOR]

Geographic object-based image analysis (GEOBIA) is a remote sensing image analysis paradigm that defines and examines image-objects: groups of neighboring pixels that represent real-world geographic objects. Recent reviews have examined methodological considerations and highlighted how GEOBIA improves upon the 30+ year pixel-based approach, particularly for H-resolution imagery. However, the literature also exposes an opportunity to improve guidance on the application of GEOBIA for novice practitioners. In this paper, we describe the theoretical foundations of GEOBIA and provide a comprehensive overview of the methodological workflow, including: (i) software-specific approaches (open-source and commercial); (ii) best practices informed by research; and (iii) the current status of methodological research. Building on this foundation, we then review recent research on the convergence of GEOBIA with deep convolutional neural networks, which we suggest is a new form of GEOBIA. Specifically, we discuss general integrative approaches and offer recommendations for future research. Overall, this paper describes the past, present, and anticipated future of GEOBIA in a novice-accessible format, while providing innovation and depth to experienced practitioners. [ABSTRACT FROM AUTHOR]

Wildfires are one of the most destructive natural disasters that can affect our environment, with significant effects also on wildlife. Recently, climate change and human activities have resulted in higher frequencies of wildfires throughout the world. Timely and accurate detection of the burned areas can help to make decisions for their management. Remote sensing satellite imagery can have a key role in mapping burned areas due to its wide coverage, high-resolution data collection, and low capture times. However, although many studies have reported on burned area mapping based on remote sensing imagery in recent decades, accurate burned area mapping remains a major challenge due to the complexity of the background and the diversity of the burned areas. This paper presents a novel framework for burned area mapping based on Deep Siamese Morphological Neural Network (DSMNN-Net) and heterogeneous datasets. The DSMNN-Net framework is based on change detection through proposing a pre/post-fire method that is compatible with heterogeneous remote sensing datasets. The proposed network combines multiscale convolution layers and morphological layers (erosion and dilation) to generate deep features. To evaluate the performance of the method proposed here, two case study areas in Australian forests were selected. The framework used can better detect burned areas compared to other state-of-the-art burned area mapping procedures, with a performance of >98% for overall accuracy index, and a kappa coefficient of >0.9, using multispectral Sentinel-2 and hyperspectral PRISMA image datasets. The analyses of the two datasets illustrate that the DSMNN-Net is sufficiently valid and robust for burned area mapping, and especially for complex areas. [ABSTRACT FROM AUTHOR]

Road extraction is important for road network renewal, intelligent transportation systems and smart cities. This paper proposes an effective method to improve road extraction accuracy and reconstruct the broken road lines caused by ground occlusion. Firstly, an attention mechanism-based convolution neural network is established to enhance feature extraction capability. By highlighting key areas and restraining interference features, the road extraction accuracy is improved. Secondly, for the common broken road problem in the extraction results, a heuristic method based on connected domain analysis is proposed to reconstruct the road. An experiment is carried out on a benchmark dataset to prove the effectiveness of this method, and the result is compared with that of several famous deep learning models including FCN8s, SegNet, U-Net and D-Linknet. The comparison shows that this model increases the IOU value and the F1 score by 3.35–12.8% and 2.41–9.8%, respectively. Additionally, the result proves the proposed method is effective at extracting roads from occluded areas. [ABSTRACT FROM AUTHOR]

Mapping land surface water automatically and accurately is closely related to human activity, biological reproduction, and the ecological environment. High spatial resolution remote sensing image (HSRRSI) data provide extensive details for land surface water and gives reliable data support for the accurate extraction of land surface water information. The convolutional neural network (CNN), widely applied in semantic segmentation, provides an automatic extraction method in land surface water information. This paper proposes a new lightweight CNN named Lightweight Multi-Scale Land Surface Water Extraction Network (LMSWENet) to extract the land surface water information based on GaoFen-1D satellite data of Wuhan, Hubei Province, China. To verify the superiority of LMSWENet, we compared the efficiency and water extraction accuracy with four mainstream CNNs (DeeplabV3+, FCN, PSPNet, and UNet) using quantitative comparison and visual comparison. Furthermore, we used LMSWENet to extract land surface water information of Wuhan on a large scale and produced the land surface water map of Wuhan for 2020 (LSWMWH-2020) with 2m spatial resolution. Random and equidistant validation points verified the mapping accuracy of LSWMWH-2020. The results are summarized as follows: (1) Compared with the other four CNNs, LMSWENet has a lightweight structure, significantly reducing the algorithm complexity and training time. (2) LMSWENet has a good performance in extracting various types of water bodies and suppressing noises because it introduces channel and spatial attention mechanisms and combines features from multiple scales. The result of land surface water extraction demonstrates that the performance of LMSWENet exceeds that of the other four CNNs. (3) LMSWENet can meet the requirement of high-precision mapping on a large scale. LSWMWH-2020 can clearly show the significant lakes, river networks, and small ponds in Wuhan with high mapping accuracy. [ABSTRACT FROM AUTHOR]

The interest in change detection in the field of remote sensing has increased in the last few years. Searching for changes in satellite images has many useful applications, ranging from land cover and land use analysis to anomaly detection. In particular, urban change detection provides an efficient tool to study urban spread and growth through several years of observation. At the same time, change detection is often a computationally challenging and time-consuming task; therefore, a standard approach with manual detection of the elements of interest by experts in the domain of Earth Observation needs to be replaced by innovative methods that can guarantee optimal results with unquestionable value and within reasonable time. In this paper, we present two different approaches to change detection (semantic segmentation and classification) that both exploit convolutional neural networks to address these particular needs, which can be further refined and used in post-processing workflows for a large variety of applications. [ABSTRACT FROM AUTHOR]

Super-resolution (SR) technology has emerged as an effective tool for image analysis and interpretation. However, single hyperspectral (HS) image SR remains challenging, due to the high spectral dimensionality and lack of available high-resolution information of auxiliary sources. To fully exploit the spectral and spatial characteristics, in this paper, a novel single HS image SR approach is proposed based on a spatial correlation-regularized unmixing convolutional neural network (CNN). The proposed approach takes advantage of a CNN to explore the collaborative spatial and spectral information of an HS image and infer the high-resolution abundance maps, thereby reconstructing the anticipated high-resolution HS image via the linear spectral mixture model. Moreover, a dual-branch architecture network and spatial spread transform function are employed to characterize the spatial correlation between the high- and low-resolution HS images, aiming at promoting the fidelity of the super-resolved image. Experiments on three public remote sensing HS images demonstrate the feasibility and superiority in terms of spectral fidelity, compared with some state-of-the-art HS image super-resolution methods. [ABSTRACT FROM AUTHOR]

The registration of multi-temporal remote sensing images with abundant information and complex changes is an important preprocessing step for subsequent applications. This paper presents a novel two-stage deep learning registration method based on sub-image matching. Unlike the conventional registration framework, the proposed network learns the mapping between matched sub-images and the geometric transformation parameters directly. In the first stage, the matching of sub-images (MSI), sub-images cropped from the images are matched through the corresponding heatmaps, which are made of the predicted similarity of each sub-image pairs. The second stage, the estimation of transformation parameters (ETP), a network with weight structure and position embedding estimates the global transformation parameters from the matched pairs. The network can deal with an uncertain number of matched sub-image inputs and reduce the impact of outliers. Furthermore, the sample sharing training strategy and the augmentation based on the bounding rectangle are introduced. We evaluated our method by comparing the conventional and deep learning methods qualitatively and quantitatively on Google Earth, ISPRS, and WHU Building Datasets. The experiments showed that our method obtained the probability of correct keypoints (PCK) of over 99% at α   = 0.05 (α : the normalized distance threshold) and achieved a maximum increase of 16.8% at α   = 0.01, compared with the latest method. The results demonstrated that our method has good robustness and improved the precision in the registration of optical remote sensing images with great variation. [ABSTRACT FROM AUTHOR]

Recently, deep learning has been successfully and widely used in hyperspectral image (HSI) classification. Considering the difficulty of acquiring HSIs, there are usually a small number of pixels used as the training instances. Therefore, it is hard to fully use the advantages of deep learning networks; for example, the very deep layers with a large number of parameters lead to overfitting. This paper proposed a dynamic wide and deep neural network (DWDNN) for HSI classification, which includes multiple efficient wide sliding window and subsampling (EWSWS) networks and can grow dynamically according to the complexity of the problems. The EWSWS network in the DWDNN was designed both in the wide and deep direction with transform kernels as hidden units. These multiple layers of kernels can extract features from the low to high level, and because they are extended in the wide direction, they can learn features more steadily and smoothly. The sliding windows with the stride and subsampling were designed to reduce the dimension of the features for each layer; therefore, the computational load was reduced. Finally, all the weights were only from the fully connected layer, and the iterative least squares method was used to compute them easily. The proposed DWDNN was tested with several HSI data including the Botswana, Pavia University, and Salinas remote sensing datasets with different numbers of instances (from small to big). The experimental results showed that the proposed method had the highest test accuracies compared to both the typical machine learning methods such as support vector machine (SVM), multilayer perceptron (MLP), radial basis function (RBF), and the recently proposed deep learning methods including the 2D convolutional neural network (CNN) and the 3D CNN designed for HSI classification. [ABSTRACT FROM AUTHOR]

The aboveground fresh weight of weeds is an important indicator that reflects their biomass and physiological activity and directly affects the criteria for determining the amount of herbicides to apply. In precision agriculture, the development of models that can accurately locate weeds and predict their fresh weight can provide visual support for accurate, variable herbicide application in real time. In this work, we develop a two-stream dense feature fusion convolutional network model based on RGB-D data for the real-time prediction of the fresh weight of weeds. A data collection method is developed for the compilation and production of RGB-D data sets. The acquired images undergo data enhancement, and a depth transformation data enhancement method suitable for depth data is proposed. The main idea behind the approach in this study is to use the YOLO-V4 model to locate weeds and use the two-stream dense feature fusion network to predict their aboveground fresh weight. In the two-stream dense feature fusion network, DenseNet and NiN methods are used to construct a Dense-NiN-Block structure for deep feature extraction and fusion. The Dense-NiN-Block module was embedded in five convolutional neural networks for comparison, and the best results were achieved with DenseNet201. The test results show that the predictive ability of the convolutional network using RGB-D as the input is better than that of the network using RGB as the input without the Dense-NiN-Block module. The mAP of the proposed network is 75.34% (IoU value of 0.5), the IoU is 86.36%, the detection speed of the fastest model with a RTX2080Ti NVIDIA graphics card is 17.8 fps, and the average relative error is approximately 4%. The model proposed in this paper can provide visual technical support for precise, variable herbicide application. The model can also provide a reference method for the non-destructive prediction of crop fresh weight in the field and can contribute to crop breeding and genetic improvement. [ABSTRACT FROM AUTHOR]

Satellite images have drawn increasing interest from a wide variety of users, including business and government, ever since their increased usage in important fields ranging from weather, forestry and agriculture to surface changes and biodiversity monitoring. Recent updates in the field have also introduced various deep learning (DL) architectures to satellite imagery as a means of extracting useful information. However, this new approach comes with its own issues, including the fact that many users utilize ready-made cloud services (both public and private) in order to take advantage of built-in DL algorithms and thus avoid the complexity of developing their own DL architectures. However, this presents new challenges to protecting data against unauthorized access, mining and usage of sensitive information extracted from that data. Therefore, new privacy concerns regarding sensitive data in satellite images have arisen. This research proposes an efficient approach that takes advantage of privacy-preserving deep learning (PPDL)-based techniques to address privacy concerns regarding data from satellite images when applying public DL models. In this paper, we proposed a partially homomorphic encryption scheme (a Paillier scheme), which enables processing of confidential information without exposure of the underlying data. Our method achieves robust results when applied to a custom convolutional neural network (CNN) as well as to existing transfer learning methods. The proposed encryption scheme also allows for training CNN models on encrypted data directly, which requires lower computational overhead. Our experiments have been performed on a real-world dataset covering several regions across Saudi Arabia. The results demonstrate that our CNN-based models were able to retain data utility while maintaining data privacy. Security parameters such as correlation coefficient (−0.004), entropy (7.95), energy (0.01), contrast (10.57), number of pixel change rate (4.86), unified average change intensity (33.66), and more are in favor of our proposed encryption scheme. To the best of our knowledge, this research is also one of the first studies that applies PPDL-based techniques to satellite image data in any capacity. [ABSTRACT FROM AUTHOR]

Wetlands are important ecosystems that are linked to climate change mitigation. As 25% of global wetlands are located in Canada, accurate and up-to-date wetland classification is of high importance, nationally and internationally. The advent of deep learning techniques has revolutionized the current use of machine learning algorithms to classify complex environments, specifically in remote sensing. In this paper, we explore the potential and possible limitations to be overcome regarding the use of ensemble deep learning techniques for complex wetland classification and discusses the potential and limitation of various solo convolutional neural networks (CNNs), including DenseNet, GoogLeNet, ShuffleNet, MobileNet, Xception, Inception-ResNet, ResNet18, and ResNet101 in three different study areas located in Newfoundland and Labrador, Canada (i.e., Avalon, Gros Morne, and Grand Falls). Moreover, to improve the classification accuracies of wetland classes of bog, fen, marsh, swamp, and shallow water, the results of the three best CNNs in each study area is fused using three supervised classifiers of random forest (RF), bagged tree (BTree), Bayesian optimized tree (BOT), and one unsupervised majority voting classifier. The results suggest that the ensemble models, in particular BTree, have a valuable role to play in the classification of wetland classes of bog, fen, marsh, swamp, and shallow water. The ensemble CNNs show an improvement of 9.63–19.04% in terms of mean producer's accuracy compared to the solo CNNs, to recognize wetland classes in three different study areas. This research indicates a promising potential for integrating ensemble-based learning and deep learning for operational large area land cover, particularly complex wetland type classification. [ABSTRACT FROM AUTHOR]

Although hyperspectral images capture very detailed information about the scanned objects, their efficient analysis, transfer, and storage are still important practical challenges due to their large volume. Classifying and segmenting such imagery are the pivotal steps in virtually all applications, hence developing new techniques for these tasks is a vital research area. Here, deep learning has established the current state of the art. However, deploying large-capacity deep models on-board an Earth observation satellite poses additional technological challenges concerned with their memory footprints, energy consumption requirements, and robustness against varying-quality image data, with the last problem being under-researched. In this paper, we tackle this issue, and propose a set of simulation scenarios that reflect a range of atmospheric conditions and noise contamination that may ultimately happen on-board an imaging satellite. We verify their impact on the generalization capabilities of spectral and spectral-spatial convolutional neural networks for hyperspectral image segmentation. Our experimental analysis, coupled with various visualizations, sheds more light on the robustness of the deep models and indicate that specific noise distributions can significantly deteriorate their performance. Additionally, we show that simulating atmospheric conditions is key to obtaining the learners that generalize well over image data acquired in different imaging settings. [ABSTRACT FROM AUTHOR]

Deep-learning methods, especially convolutional neural networks (CNN), have become the first choice for hyperspectral image (HSI) classification to date. It is a common procedure that small cubes are cropped from hyperspectral images and then fed into CNNs. However, standard CNNs find it difficult to extract discriminative spectral–spatial features. How to obtain finer spectral–spatial features to improve the classification performance is now a hot topic of research. In this regard, the attention mechanism, which has achieved excellent performance in other computer vision, holds the exciting prospect. In this paper, we propose a double-branch network consisting of a novel convolution named pyramidal convolution (PyConv) and an iterative attention mechanism. Each branch concentrates on exploiting spectral or spatial features with different PyConvs, supplemented by the attention module for refining the feature map. Experimental results demonstrate that our model can yield competitive performance compared to other state-of-the-art models. [ABSTRACT FROM AUTHOR]

In recent years, benefiting from the rapid development of deep learning technology in the field of computer vision, the study of hyperspectral image (HSI) classification has also made great progress. However, compared with ordinary RGB images, HSIs are more like 3D cubes; therefore, it is necessary and beneficial to explore classification methods suitable for the very special data structure of HSIs. In this paper, we propose Multiple Spectral Resolution 3D Convolutional Neural Network (MSR-3DCNN) for HSI classification tasks. In MSR-3DCNN, we expand the idea of multi-scale feature fusion and dilated convolution from the spatial dimension to the spectral dimension, and combine 3D convolution and residual connection; therefore, it can better adapt to the 3D cubic form of hyperspectral data and make efficient use of spectral information in different bands. Experimental results on four benchmark datasets show the effectiveness of the proposed approach and its superiority as compared with some state-of-the-art (SOTA) HSI classification methods. [ABSTRACT FROM AUTHOR]

Supraglacial meltwater accumulation on ice sheets can be a main driver for accelerated ice discharge, mass loss, and global sea-level-rise. With further increasing surface air temperatures, meltwater-induced hydrofracturing, basal sliding, or surface thinning will cumulate and most likely trigger unprecedented ice mass loss on the Greenland and Antarctic ice sheets. While the Greenland surface hydrological network as well as its impacts on ice dynamics and mass balance has been studied in much detail, Antarctic supraglacial lakes remain understudied with a circum-Antarctic record of their spatio-temporal development entirely lacking. This study provides the first automated supraglacial lake extent mapping method using Sentinel-1 synthetic aperture radar (SAR) imagery over Antarctica and complements the developed optical Sentinel-2 supraglacial lake detection algorithm presented in our companion paper. In detail, we propose the use of a modified U-Net for semantic segmentation of supraglacial lakes in single-polarized Sentinel-1 imagery. The convolutional neural network (CNN) is implemented with residual connections for optimized performance as well as an Atrous Spatial Pyramid Pooling (ASPP) module for multiscale feature extraction. The algorithm is trained on 21,200 Sentinel-1 image patches and evaluated in ten spatially or temporally independent test acquisitions. In addition, George VI Ice Shelf is analyzed for intra-annual lake dynamics throughout austral summer 2019/2020 and a decision-level fused Sentinel-1 and Sentinel-2 maximum lake extent mapping product is presented for January 2020 revealing a more complete supraglacial lake coverage (~770 km2) than the individual single-sensor products. Classification results confirm the reliability of the proposed workflow with an average Kappa coefficient of 0.925 and a F 1 -score of 93.0% for the supraglacial water class across all test regions. Furthermore, the algorithm is applied in an additional test region covering supraglacial lakes on the Greenland ice sheet which further highlights the potential for spatio-temporal transferability. Future work involves the integration of more training data as well as intra-annual analyses of supraglacial lake occurrence across the whole continent and with focus on supraglacial lake development throughout a summer melt season and into Antarctic winter. [ABSTRACT FROM AUTHOR]

Recently, deep learning has been reported to be an effective method for improving hyperspectral image classification and convolutional neural networks (CNNs) are, in particular, gaining more and more attention in this field. CNNs provide automatic approaches that can learn more abstract features of hyperspectral images from spectral, spatial, or spectral-spatial domains. However, CNN applications are focused on learning features directly from image data—while the intrinsic relations between original features, which may provide more information for classification, are not fully considered. In order to make full use of the relations between hyperspectral features and to explore more objective features for improving classification accuracy, we proposed feature relations map learning (FRML) in this paper. FRML can automatically enhance the separability of different objects in an image, using a segmented feature relations map (SFRM) that reflects the relations between spectral features through a normalized difference index (NDI), and it can then learn new features from SFRM using a CNN-based feature extractor. Finally, based on these features, a classifier was designed for the classification. With FRML, our experimental results from four popular hyperspectral datasets indicate that the proposed method can achieve more representative and objective features to improve classification accuracy, outperforming classifications using the comparative methods. [ABSTRACT FROM AUTHOR]

Recently, deep learning methods based on three-dimensional (3-D) convolution have been widely used in the hyperspectral image (HSI) classification tasks and shown good classification performance. However, affected by the irregular distribution of various classes in HSI datasets, most previous 3-D convolutional neural network (CNN)-based models require more training samples to obtain better classification accuracies. In addition, as the network deepens, which leads to the spatial resolution of feature maps gradually decreasing, much useful information may be lost during the training process. Therefore, how to ensure efficient network training is key to the HSI classification tasks. To address the issue mentioned above, in this paper, we proposed a 3-DCNN-based residual group channel and space attention network (RGCSA) for HSI classification. Firstly, the proposed bottom-up top-down attention structure with the residual connection can improve network training efficiency by optimizing channel-wise and spatial-wise features throughout the whole training process. Secondly, the proposed residual group channel-wise attention module can reduce the possibility of losing useful information, and the novel spatial-wise attention module can extract context information to strengthen the spatial features. Furthermore, our proposed RGCSA network only needs few training samples to achieve higher classification accuracies than previous 3-D-CNN-based networks. The experimental results on three commonly used HSI datasets demonstrate the superiority of our proposed network based on the attention mechanism and the effectiveness of the proposed channel-wise and spatial-wise attention modules for HSI classification. The code and configurations are released at Github.com. [ABSTRACT FROM AUTHOR]

Accurate hyperspectral image classification has been an important yet challenging task for years. With the recent success of deep learning in various tasks, 2-dimensional (2D)/3-dimensional (3D) convolutional neural networks (CNNs) have been exploited to capture spectral or spatial information in hyperspectral images. On the other hand, few approaches make use of both spectral and spatial information simultaneously, which is critical to accurate hyperspectral image classification. This paper presents a novel Synergistic Convolutional Neural Network (SyCNN) for accurate hyperspectral image classification. The SyCNN consists of a hybrid module that combines 2D and 3D CNNs in feature learning and a data interaction module that fuses spectral and spatial hyperspectral information. Additionally, it introduces a 3D attention mechanism before the fully-connected layer which helps filter out interfering features and information effectively. Extensive experiments over three public benchmarking datasets show that our proposed SyCNNs clearly outperform state-of-the-art techniques that use 2D/3D CNNs. [ABSTRACT FROM AUTHOR]

Many wildlife species inhabit inaccessible environments, limiting researchers ability to conduct essential population surveys. Recently, very high resolution (sub-metre) satellite imagery has enabled remote monitoring of certain species directly from space; however, manual analysis of the imagery is time-consuming, expensive and subjective. State-of-the-art deep learning approaches can automate this process; however, often image datasets are small, and uncertainty in ground truth labels can affect supervised training schemes and the interpretation of errors. In this paper, we investigate these challenges by conducting both manual and automated counts of nesting Wandering Albatrosses on four separate islands, captured by the 31 cm resolution WorldView-3 sensor. We collect counts from six observers, and train a convolutional neural network (U-Net) using leave-one-island-out cross-validation and different combinations of ground truth labels. We show that (1) interobserver variation in manual counts is significant and differs between the four islands, (2) the small dataset can limit the networks ability to generalise to unseen imagery and (3) the choice of ground truth labels can have a significant impact on our assessment of network performance. Our final results show the network detects albatrosses as accurately as human observers for two of the islands, while in the other two misclassifications are largely caused by the presence of noise, cloud cover and habitat, which was not present in the training dataset. While the results show promise, we stress the importance of considering these factors for any study where data is limited and observer confidence is variable. [ABSTRACT FROM AUTHOR]

In recent years, deep learning has dramatically improved the cognitive ability of the network by extracting depth features, and has been successfully applied in the field of feature extraction and classification of hyperspectral images. However, it is facing great difficulties for target detection due to extremely limited available labeled samples that are insufficient to train deep networks. In this paper, a novel target detection framework for deep learning is proposed, denoted as HTD-Net. To overcome the few-training-sample issue, the proposed framework utilizes an improved autoencoder (AE) to generate target signatures, and then finds background samples which differ significantly from target samples based on a linear prediction (LP) strategy. Then, the obtained target and background samples are used to enlarge the training set by generating pixel-pairs, which is viewed as the input of a pre-designed network architecture to learn discriminative similarity. During testing, pixel-pairs of a pixel to be labeled are constructed with both available target samples and background samples. Spectral difference between these pixel-pairs is classified by the well-trained network with results of similarity measurement. The outputs from a two-branch averaged similarity scores are combined to generate the final detection. Experimental results with several real hyperspectral data demonstrate the superiority of the proposed algorithm compared to some traditional target detectors. [ABSTRACT FROM AUTHOR]

High-Performance Computing (HPC) has recently been attracting more attention in remote sensing applications due to the challenges posed by the increased amount of open data that are produced daily by Earth Observation (EO) programs. The unique parallel computing environments and programming techniques that are integrated in HPC systems are able to solve large-scale problems such as the training of classification algorithms with large amounts of Remote Sensing (RS) data. This paper shows that the training of state-of-the-art deep Convolutional Neural Networks (CNNs) can be efficiently performed in distributed fashion using parallel implementation techniques on HPC machines containing a large number of Graphics Processing Units (GPUs). The experimental results confirm that distributed training can drastically reduce the amount of time needed to perform full training, resulting in near linear scaling without loss of test accuracy. [ABSTRACT FROM AUTHOR]

Remote sensing image dehazing is an extremely complex issue due to the irregular and non-uniform distribution of haze. In this paper, a prior-based dense attentive dehazing network (DADN) is proposed for single remote sensing image haze removal. The proposed network, which is constructed based on dense blocks and attention blocks, contains an encoder-decoder architecture, which enables it to directly learn the mapping between the input images and the corresponding haze-free image, without being dependent on the traditional atmospheric scattering model (ASM). To better handle non-uniform hazy remote sensing images, we propose to combine a haze density prior with deep learning, where an initial haze density map (HDM) is firstly extracted from the original hazy image, and is subsequently utilized as the input of the network, together with the original hazy image. Meanwhile, a large-scale hazy remote sensing dataset is created for training and testing of the proposed method, which contains both uniform and non-uniform, synthetic and real hazy remote sensing images. Experimental results on the created dataset illustrate that the developed dehazing method obtains significant progresses over the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Due to technical and budget limitations, there are inevitably some trade-offs in the design of remote sensing instruments, making it difficult to acquire high spatiotemporal resolution remote sensing images simultaneously. To address this problem, this paper proposes a new data fusion model named the deep convolutional spatiotemporal fusion network (DCSTFN), which makes full use of a convolutional neural network (CNN) to derive high spatiotemporal resolution images from remotely sensed images with high temporal but low spatial resolution (HTLS) and low temporal but high spatial resolution (LTHS). The DCSTFN model is composed of three major parts: the expansion of the HTLS images, the extraction of high frequency components from LTHS images, and the fusion of extracted features. The inputs of the proposed network include a pair of HTLS and LTHS reference images from a single day and another HTLS image on the prediction date. Convolution is used to extract key features from inputs, and deconvolution is employed to expand the size of HTLS images. The features extracted from HTLS and LTHS images are then fused with the aid of an equation that accounts for temporal ground coverage changes. The output image on the prediction day has the spatial resolution of LTHS and temporal resolution of HTLS. Overall, the DCSTFN model establishes a complex but direct non-linear mapping between the inputs and the output. Experiments with MODerate Resolution Imaging Spectroradiometer (MODIS) and Landsat Operational Land Imager (OLI) images show that the proposed CNN-based approach not only achieves state-of-the-art accuracy, but is also more robust than conventional spatiotemporal fusion algorithms. In addition, DCSTFN is a faster and less time-consuming method to perform the data fusion with the trained network, and can potentially be applied to the bulk processing of archived data. [ABSTRACT FROM AUTHOR]