Detailed and accurate information on the spatial variation of land cover use is a critical component local ecology environmental research. For these tasks, high resolution images are required. Considering trade-off between temporal in remote sensing images, many learning-based models (e.g., Convolutional neural network, sparse coding, Bayesian network) have been established to improve coarse both computer vision fields. However, data for training testing methods usually limited certain...

The thermal band of a satellite platform enables the measurement land surface temperature (LST), which captures spatial-temporal distribution energy exchange between Earth and atmosphere. LST plays critical role in simulation models, enhancing our understanding physical biochemical processes nature. However, limitations swath width orbit altitude prevent single sensor from providing data with both high spatial temporal resolution. To tackle this challenge, unmixing-based spatiotemporal...

Sub-Pixel Offset Tracking (sPOT) is applied to derive high-resolution centimetre-level landslide rates in the Three Gorges Region of China using TerraSAR-X Hi-resolution Spotlight (TSX HS) space-borne SAR images. These results contrast sharply with previous use conventional differential Interferometric Synthetic Aperture Radar (DInSAR) techniques areas steep slopes, dense vegetation and large variability water vapour which indicated around 12% phase coherent coverage. By contrast, sPOT...

Timely and accurate crop type mapping is a critical prerequisite for the estimation of water availability environmental carrying capacity. This research proposed method to integrate time series Sentinel-1 (S1) Sentinel-2 (S2) data over oasis agricultural areas through case study in Northwest China. Previous studies using synthetic aperture radar (SAR) alone often yield quite limited accuracy identification due speckles. To improve quality SAR features, we adopted statistically homogeneous...

A new method to identify short-rotation eucalyptus plantations by exploring both the changing pattern of vegetation indices due tree crop rotation and spectral characteristics in red-edge region is presented. It can be adopted produce maps high spatial resolution (30 m) at large scales, with use open remote sensing images from Landsat 8 Operational Land Imager (OLI), MODerate Imaging Spectroradiometer (MODIS), Sentinel-2 MultiSpectral Instrument (MSI), as well a free cloud computing...

In emergency responses to natural disasters, actionable information provided by remote sensing images is crucial help managers become aware of the situation and assess magnitude damage. Without accurate prediction time consumption, choosing an algorithm for land use/land cover (LULC) classification under these circumstances could be blind subjective. Here, we proposed a full parameter complexity (FPTC) analysis corresponding coefficient ω estimate actual running LULC without actually code....

Accurate and efficient extraction of cultivated land data is great significance for agricultural resource monitoring national food security. Deep-learning-based classification remote-sensing images overcomes the two difficulties traditional learning methods (e.g., support vector machine (SVM), K-nearest neighbors (KNN), random forest (RF)) when extracting land: (1) limited performance same land-cover type with high intra-class spectral variation, such as both vegetation non-vegetation cover,...

Hyperspectral remote sensing image classification has been widely employed for numerous applications, such as environmental monitoring, agriculture, and mineralogy. During classification, the number of training samples in each class often varies significantly. This imbalance dataset is not identified because most classifiers are designed under a balanced assumption, which can distort minority classes or even treat them noise. may lead to biased inaccurate results. issue be alleviated by...

Sub-pixel offset tracking has been used in various applications, including measurements of glacier movement, earthquakes, landslides, etc., as a complementary method to time series InSAR. In this work, we explore the use small baseline subset (SBAS) Offset Tracking approach monitor very slow landslides with centimetre-level annual displacement rate, and challenging areas characterized by high humidity, dense vegetation cover, steep slopes. This approach, herein referred SBAS Tracking, is...

The unmixing-based spatiotemporal fusion model is one of the effective ways to solve limitations in temporal and spatial resolution tradeoffs a single satellite sensor. By using data from different platforms, high both domains can be produced. However, due ill-posed characteristic unmixing function, performance may vary setups. key factors affecting stability most how set up strategy for downscaling remain unknown. In this study, we use multisource land surface temperature as case focus on...

The spatial fragmentation of high-resolution remote sensing images makes the segmentation algorithm put forward a strong demand for noise immunity. However, stronger immunity, more serious loss detailed information, which easily leads to neglect effective characteristics. In view difficulty balancing immunity and characteristic retention, an adaptive distance-weighted Voronoi tessellation technology is proposed image segmentation. distance between pixels seed points in established by...

Tropospheric correction is a crucial step for interferometric synthetic aperture radar (InSAR) monitoring of small deformation magnitude. However, most the corrections are implemented without rigorous evaluation their influences on InSAR measurements. In this paper, we present three statistical metrics to evaluate performance. Firstly, propose time series decomposition method estimate tropospheric noise and mitigate bias caused by ground displacement. On basis, calculate root-mean-square...

The timely and accurate mapping of sugarcane cultivation is significant to ensure the sustainability industry, including production, rural society, sugar futures, crop insurance. Synthetic aperture radar (SAR), due its all-weather all-time imaging capability, plays an important role in cloudy areas. However, inherent speckle noise SAR data worsens “salt pepper” effect map. Therefore, previous studies, additional land cover map or optical image was still required. This study proposes a new...

Mapping rubber plantations in a large area is still challenging high-cloud-cover and complex-vegetation landscapes. Existing studies were often confined to the discrimination of trees from natural forests rarely concerned other tropical tree species. The Sentinel-2 constellation, with improved spatial, spectral, temporal resolution, offers new opportunities improve previous efforts. In this paper, four Hainan composites generated based on detailed phenological stages delineation trees....

The amplitude and coherence attributes of three-dimensional (3D) seismic data are used to confirm gas hydrate occurrence delineate its distribution in the Zhongjiannan basin, South China Sea. High anomalies (HAAs) distributed above or below regional base stability zone (BGHSZ), which intersect with bottom simulating reflectors (BSRs) interrupted by different types pockmarks. maximum attribute extracted along T1 (5.5 Ma) horizon is controlled widely faults. layer-bound polygonal faults (PFs)...

摘要: 寻找与识别高饱和度天然气水合物富集层是当前海域天然气水合物商业性开采的重点.通过对南海中建盆地局部三维地震资料开展宽频重处理和波阻抗反演,结合全区三维地震的精细解释、属性分析和稳定带计算,剖析中建盆地BSR(bottom simulating reflector)反射特征和天然气水合物富集程度.研究发现:(1)研究区BSR特征与南海北部钻探区略微不同,主要分布在半深海细粒沉积物,分布广且连续,当断层到达水合物稳定带处时,BSR振幅强且连续;(2)天然气水合物层饱和度约30%~40%,局部较高,富集程度为中等饱和度;(3)受沉积与构造活动影响,水合物富集层位于BSR上部不同地层深度,断层控制着水合物分布. 关键词: 天然气水合物 / 富集 宽频处理 中建盆地 构造 沉积学

The deep-learning-network performance depends on the accuracy of training samples. samples are commonly labeled by human visual investigation or inherited from historical land-cover land-use maps, which usually contain label noise, depending subjective knowledge and time map. Helping network to distinguish noisy labels during process is a prerequisite for applying model across locations. This study proposes an antinoise framework, Weight Loss Network (WLN), achieve this goal. WLN contains...

Deep-learning-based object detectors have substantially improved state-of-the-art detection in remote sensing images terms of precision and degree automation. Nevertheless, the large variation scales makes it difficult to achieve high-quality across multiresolution images, where quality is defined by Intersection over Union (IoU) threshold used training. In addition, imbalance between positive negative samples worsens precision. Recently, was found that a Cascade region-based convolutional...

Aiming at the problem of misdetection caused by traditional texture characteristic extraction model, which does not describe correlation among multiple bands, an object-oriented remote sensing image change detection method based on a color co-occurrence matrix is proposed. First, divided into multi-scale objects graph-based superpixel segmentation, and optimal scale determined overall goodness F-measure (OGF). Then, except for spectral features, multi-channel features (CCM) are extracted to...

Abstract The complex contacts between gas hydrate and free have been confirmed by pilot production in the Pearl River Mouth Basin, South China Sea. laminated clay silt reservoirs high amplitude reflections which may be filled with different saturations of hydrates, coexistence gas. To distinguish variations caused properties below bottom simulating reflection, we combine forward modelling, pre‐stack gathers incident angles well log analysis to conduct seismic responses. A number geological...

Superpixel segmentation is an essential step of object-oriented remote sensing image classification, the accuracy superpixel boundary will directly affect classification result. Most traditional algorithms rely on spectral similarity and spatial connectivity to construct superpixels. They can not find accurate in complex scenes, such as distribution ground features being relatively broken, large differences size shape, especially long-thin shape circular shape. Aiming at this problem, a...