Vision transformers are widely used in hyperspectral image classification, with their core feature extractor being self-attention. Self-attention has a wider receptive field than convolution. However, existing vision transformers for the classification of hyperspectral images (HSIs) with a large number of bands generally suffer from high computational complexity and a large number of parameter requirements. In this paper, we propose an Ultra-lightweight Feature-compressed Multi-head Self-attention Learning Network (UFMS-LN), which mainly consists of a novel Compressed Feature Multi-Head Self-Attention (CF-MHSA), a Spatial Feature Enhancement- Enhancing Transformation Reduction (SFE-ETR) and a Spatial-spectral Hybridization-Receptive Field Attention Convolutional operation (SH-RFAConv). By effectively compressing feature maps in spatial-spectral dimensions, CF-MHSA achieves the same feature extraction capabilities as state-of-the-art self-attention mechanisms, and its floating-point operations (FLOPs) and parameters are two orders of magnitude lower than state-of-the-art self-attention mechanisms. SH-RFAConv is designed to emphasize local features, which have the ability to extract both spatial-spectral features simultaneously and have a wider receptive field than traditional convolutional operations. Furthermore, SFE-ETR is a preprocessing module for UFMS-LN that combines global spatial feature enhancement methods with Enhancing Transformation Reduction (ETR). Extensive experiments conducted on four benchmark HSI datasets have shown that this method achieves superior results compared to existing state-of-the-art HSI classification networks.

The shipping industry is pivotal in transporting approximately 90% of the world’s goods, and it is characterized by evolving trends in vessel sizes and energy-efficient designs. Continuous advancements in technology for ship management have focused on detecting and analyzing anomalous and illicit vessels. In this study, we introduce ShipGeoNet, a model designed to extract geometric features from ships captured in Sentinel-1 synthetic aperture radar (SAR) images. ShipGeoNet employs a combination of convolutional neural networks (CNNs) and nonlinear regression techniques to extract various geometric features of ships from SAR imagery. The model follows a two-step approach. First, it utilizes a modified Mask R-CNN architecture and the ViTDet model to accurately detect ships, generating high-quality object masks for precise localization. In the subsequent step, a regression model utilizes the detected ship masks to extract key geometric attributes, including length, width, and orientation. The proposed nonlinear regression techniques are specifically crafted to address the complex nonlinear deformations inherent in SAR images. Through extensive experiments on a large-scale SAR dataset, ShipGeoNet demonstrates its efficiency and accuracy in ship size extraction and matching, outperforming existing methods. Developing the ShipGeoNet model opens up possibilities for future applications in maritime surveillance, navigation, and environmental monitoring.

Through the oil spill simulation experiment of designing outdoor small scenes, this paper finds that: (1) the thermal infrared image obtained by the two sensors shows that the diurnal variation trend of the thermal infrared brightness temperature (BT) value of different OFT is the same, but the values will be different depending on the sensitivity of the sensor and the shooting angle. (2) the larger the oil film thickness (OFT), the larger the brightness temperature difference (BTD) between oil and water in the daytime, while the relationship between OFT and BTD is not monotonous in the nighttime. When the OFT is less than 317μm, the larger the OFT, the smaller the BTD, and when OFT is greater than 317μm, the conclusion is reversed. (3) there is a strong correlation between OFT and BTD. The correlation is greatest around noon, which is most conducive to the detection of OFT.

Object identification is one of the fundamental challenges in computer vision. Dealing with tiny, fuzzy, and widely dispersed objects is challenging despite tremendous improvements. This article introduces the detector in our model as a tool for detecting tiny objects in SAR (Synthetic Aperture Radar) photos. To overcome the aforementioned issues, the suggested model makes use of shallow layer information, spatial attention, and contextual information. To avoid missing the tiny object features, a shallow semantic information extraction model that incorporates low-level semantic data into the backbone has been designed. The original Neck has been replaced by an MSCFP (Multi-scale Context Feature Pyramid) in order to enhance the exploitation of lower levels of information and provide context information. The recognition of attention locations of various sizes is made possible by the introduction of a spatial attention system. Tests using open-source SSDD (SAR Ship Detection Dataset) datasets show that our model has effective identifying capabilities.

Oil film thickness (OFT) is an important indicator for estimating the amount of oil spill, and accurately quantifying the OFT is of great significance for loss assessment. In this paper, hyperspectral images (HSIs) of different OFTs (0.01-3.04 mm) through a ground experiment were obtained, and the spectral characteristics were analyzed. To address the issue of poor spectral separability for different OFTs, the 1DConvolutional Neural Network_Gate Recurrent Unit (1DCNN_GRU) model was developed for the quantitative inversion of OFT. It was validated through experiments on airborne Cubert-S185 HSI. The experimental results indicated that: (1) The proposed 1DCNN_GRU model effectively addressed the issue of reduced quantitative inversion accuracy resulting from poor spectral separability. The inversion results of it outperformed those of the SVR, CNN, and GRU models. Moreover, the optimal time for hyperspectral sensor to monitor OFT was at noon. (2) The proposed model using airborne hyperspectral data exhibited excellent inversion performance for OFT greater than 0.07 mm, especially with the best performance in 0.60-0.90mm. (3) The accuracy of HSI based OFT inversion assisted by brightness temperature (BT) data was superior to that of OFT inversion using single-source data. In particular, the proposed model had advantages in the feature level and decision level inversion of OFT in the ranges of 0.01-0.30mm and 1.00-3.04mm, respectively. This research provides technical support for the detection of OFT.

Autoencoders (AEs) are commonly utilized for acquiring low-dimensional data representations and performing data reconstruction, which makes them suitable for hyperspectral unmixing. However, AE networks trained pixel by pixel and those employing localized convolutional filters disregard the global material distribution and distant interdependencies, resulting in the loss of necessary spatial feature information essential for the unmixing process. To overcome this limitation, we propose an innovative deep neural network model named U-shaped transformer network using shifted windows (UST-Net). UST-Net prioritizes spatial information in the scene that is more discriminative and significant by using multi-head self-attention blocks based on shifted windows. Unlike patch-based unmixing networks, UST-Net operates on the complete image, eliminating inconsistencies associated with patches. Moreover, the downsampling and upsampling stages are used to extract HSI feature maps at different scales. This process generates a context-rich and spatially accurate abundance map without losing local details. The experimental results of one synthetic dataset and three real datasets demonstrate that UST-Net significantly outperforms both traditional and several other advanced neural network methods. Our code is publicly available at https://github.com/UPCGIT/UST-Net.

In order to research the multi-year spatial-temporal distribution of Harmful Algal Bloom (HAB) in the South Yellow Sea of China, NDVI method was used to extract HAB from MODIS images after Cloud removal by a comprehensive threshold method explored in this paper. Then, the growth law and spatial-temporal distribution of HAB are analyzed by the method of standard deviation ellipse and superposition analysis. It shows that the periphery of the radial sand ridge area off the coast of Jiangsu Province is the main birthplace of the HAB. Entering the breeding period, the northward drift speed as well as the diffusion speed of HAB accelerates. Until late June, it invades the coast of Shandong Province. In the early stage of HAB growth, preventing the spreading along the northeast-southwest direction will effectively restrain the HAB.

Drought is one of the agricultural natural disasters, which seriously threatens the natural ecological environment and world's food security. The FY-3B satellite has the ability of drought monitoring, but its spatial resolution is low, which isn't suitable for small and medium scale drought monitoring. In view of this, this paper calculates the High resolution Soil Moisture Drought Index (HSMDI) in the study area through the effective fusion of MODIS optical data which is small and medium scale and FY-3B soil moisture data of large scale. The results were verified by the measured meteorological data and simulated soil moisture data, which showed that HSMDI had a good consistency with precipitation and soil moisture (P < 0.05). The spatial and temporal characteristics of drought described by HSMDI were consistent with the actual drought situation in the study area, indicating that the method can effectively achieve small and medium scale drought monitoring.

Cloud interference often occurs in Enteromorpha prolifera (EP) extraction from MODIS images, with the purpose of solving this problem, a pseudo-invariant feature-based relative radiometric correction algorithm was proposed in this paper for cloud detection, and named PIF-RAC. The pseudo- invariant feature pixels were carried out to find the linear relationship of reflectance between target image and reference image in this algorithm. Then, the cloud detection threshold of the target image was corrected by the above established linear relationship and manual cloud detection threshold of the reference image. The experimental results show the automatic cloud detection effect of the proposed algorithm is close to that of the artificial threshold algorithm, which enables to effectively eliminate different kind of cloud interference for EP information from MODIS images. The PIF-RAC is an unsupervised algorithm with a high level of automation, which can be applied on EP disasters remote sensing operational monitoring.

Yellow Sea and Bohai Sea are the most frequent sea fog regions in China. The research on sea fog detection methods is of great significance to sea safety and human production activities. In this paper, the MODIS images are used for sea fog detection experiment from 2016 to 2018. The sea fog detection threshold algorithm is established based on MODIS 1, 2, 3, 5, 17, 26 and 32 bands. The temporal and spatial characteristics of sea fog are analyzed in the Yellow Sea and Bohai Sea.

In this paper, the spectral and texture features of Sargassum first are analyzed through calculating four measures of GLCM and sampling spectrum from typical pixels of Sargassum blooms with high-resolution satellite data. The four-dimensional spectral bands of the image, the first principal component and NVDI are used as the spectral features of the image, and four measures of GLCM are used as the texture features of the image. And then the Sargassum is extracted using SVM by constructing spectral-texture eigenvectors. The experiment achieves superior results compared with the conventional NDVI threshold method.

In this paper the quality of Jason-3 data was evaluated based on Jason-2 GDR data in the tandem stages. The percentage of data loss and data edition and the daily mean changes of the main physical parameters are calculated for each cycle. The sea surface height discrepancy at the intersection point and the sea level anomaly along the track are analyzed, and the system deviation between Jason-2 and Jason-3 are calibrated. Jason-3 and Jason-2 data have uniform change trend and spatial distribution in significant wave heights(SWH), mean backscattering coefficient, ionosphere delay correction mean values and mean wet tropospheric delay differences between microwave radiometer observations and ECMWF model, and there are systematic deviations between Jason-3 and Jason-2. The standard deviations of the sea surface height bias at the self-intersection of the Jason-3 and Jason-2 are 4.98cm and 4.94cm, respectively. It can be indicated that the accuracy of Jason-3 is comparable to that of Jason-2 with the systematic deviation of 2.93cm.

The sea level rise rate of the China Sea and its vicinity is 4.16 mm/a based on the altimeter data of TOPEX/Poseidon, Jason-1, Jason-2 and Jason-3 satellites. Then the sea level time series are denoised by Morlet wavelet transform. And the multi-time scale characteristics of sea level time series are analyzed by wavelet transform. The results show that the strength and distribution of various periods and the breaking point can be given by the wavelet analysis. The sea level change in China Sea and its vicinity is dominated by annual signals with signal periods of 1.5a and 4a.

Ocean oil spill is an emergency with great harm. Optical remote sensing is an important means to monitor oil spill on the sea surface. Due to the influence of cloud and weather and the limitation of satellite revisit period, only limited sample data can be obtained. In the case of limited samples, the ability of learning sample features using a single supervised classifier is limited, which can not meet the needs of accurately monitor oil spill. This paper takes GF-1 WFV oil spill image as data source, and uses four classical supervised classification algorithms to extract oil spill information. From the point of view of target recognition information fusion, the advantages of multiple supervised classification algorithms are integrated. Decision fusion algorithm is used to realize multi-source oil spill information fusion, so as to improve the detection accuracy of remote sensing oil spill.

The ocean gravity field has always been an important part of marine scientific research. Combining ship survey data and satellite data to obtain high-resolution and high-precision gravity data attracts many researchers. In this paper, we present two fusion methods of gravity anomaly data, which are data fusion based on the analysis of error trend surface and the data fusion method based on the combination of net function and fractal interpolation. The experimental results show that the data fusion based on the analysis of error trend surface can effectively improve the effect of data fusion. The data fusion method based on the combination of net function and fractal interpolation is not obvious for accuracy improvement, but it shows great potential in terms of detail.