This study proposes a method for the three-dimensional (3D) inversion of Interferometric Synthetic Aperture Radar (InSAR) time series measurements, focusing on land subsidence in urban contexts characterized by slow, ...
详细信息
ISBN:
(数字)9798350360325
ISBN:
(纸本)9798350360332
This study proposes a method for the three-dimensional (3D) inversion of Interferometric Synthetic Aperture Radar (InSAR) time series measurements, focusing on land subsidence in urban contexts characterized by slow, long-term, and small deformation magnitudes. Tailored for situations primarily dependent on single-track SAR satellite data, it integrates a physical constraint model between horizontal and vertical deformation gradients. Given the localized nature of urban deformations, this method avoids using a uniform subsidence model for an entire Region of Interest (ROI). Instead, it opts for a pixel-by-pixel estimation of constraint parameters, based on a detailed analysis of subsidence in affected areas. Experiments were conducted in representative scenarios, including the subsidence observed in ocean reclaimed zones and in residential areas impacted by tunneling activities for metro line construction. The results, validated through comparison with leveling measurements, suggest that the proposed method facilitates a precise 3D inversion, capturing the complex dynamics of urban surface deformation.
Water contamination is a major concern worldwide due to its impact on the global ecology. Accurate forecasting of water quality parameters is crucial for effective water contamination control. However, existing models...
详细信息
Due to the significant differences between the source and target domains, semantic segmentation models for remote sensing images trained on the source domain often struggle to generalize effectively to new target doma...
详细信息
Water contamination is a major concern worldwide due to its impact on the global ecology. Accurate forecasting of water quality parameters is crucial for effective water contamination control. However, existing models...
Water contamination is a major concern worldwide due to its impact on the global ecology. Accurate forecasting of water quality parameters is crucial for effective water contamination control. However, existing models face challenges in accurately predicting water quality series due to their nonlinearity and non-stationarity. Additionally, further investigation is needed to find out the performance of the Bidirectional Long Short-Term Memory network (BiLSTM) when combined with optimization of signal processing and adjustment of hyperparameters. In this article, we introduce a novel hybrid approach named CEEMDAN-QPSO-BiLSTM, which combines three techniques - Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Quantum behaved Particle Swarm optimization (QPSO), and BiLSTM - for predicting water quality parameters. To evaluate the effectiveness of this hybrid model, we developed four benchmark models, namely LSTM, BiLSTM, PSO-BiLSTM, and QPSO-BiLSTM, and compared their predictive accuracy with that of our hybrid approach. The training and testing of these models utilized actual water quality data from the Chinese National Marine environmental Monitoring Center’s official system, specifically focusing on Dissolved Oxygen (DO), pH value, and Chemical Oxygen Demand (COD) parameters. The experimental results show that the CEEMDAN-QPSO-BiLSTM surpasses the baseline models, providing better error performance and prediction accuracy.
Remote sensing(RS)technologies are extensively exploited by scientists and a vast audience of local authorities,urban managers,and city *** regions,geohazard-prone areas,and highly populated cities represent natural l...
详细信息
Remote sensing(RS)technologies are extensively exploited by scientists and a vast audience of local authorities,urban managers,and city *** regions,geohazard-prone areas,and highly populated cities represent natural laboratories to apply RS technologies and test new *** the last decades,many efforts have been spent on improving Earth's surface monitoring,including intensifying Earth Observation(EO)operations by the major national space *** oversee to plan and make operational constellations of satellite sensors providing the scientific community with extensive research and development oppor-tunities in the geoscience *** instance,within this framework,the European Space Agency(ESA)and the Ministry of Science and Technology of China(MOST)have sponsored,since the early 2000s,the DRAGON initiative jointly carried out by the European and Chinese RS scientific *** manuscript aims to provide a synthetic overview of some research activities and new methods recently designed and applied and trace the route for further *** main findings are related to ⅰ)the analysis of flood risk in China,ⅱ)the potential of new methods for the estimation and removal of ground displacement biases in small-baseline oriented interferometric Synthetic Aperture Radar(SAR)methods,ⅲ)the analy-sis of the inundation risk in low-lying regions using coherent and incoherent SAR methods;andⅳ)the use of SAR-based technologies for marine applications.
As the function of the decomposition of fungi has been clearly researched in the global carbon cycle,it is obviously of value to explore the decomposition rate of fungal *** study analyzed the relationship between env...
详细信息
As the function of the decomposition of fungi has been clearly researched in the global carbon cycle,it is obviously of value to explore the decomposition rate of fungal *** study analyzed the relationship between environmental factors and biodiversity step by *** order to explore the interaction between the fungi and the relationship between the decomposition rate of fungi with time,the model based on the Logistic model was built and the Lotka-Volterra model was employed in the condition of two kinds of fungi existing in an environment with limited *** changing trend of population number and decomposition rate of several fungi under different environmental conditions can be predicted through the *** illustrate the applicability of the model,Laetiporus conifericola and Hyphoderma setigerum were applied as *** results showed that the higher the degree of population diversity,the greater the decomposition rate,and the higher the decomposition efficiency of the *** rich species diversity is conducive to accelerating the decomposition of litter,lignocellulose,and the circulation of the entire *** on the above model and using the data from measuring the mycelial elongation rate of each isolate at 10℃,16℃,and 22℃ under standardized laboratory conditions,the growth patterns of the five fungi combinations were *** results revealed a general increase in growth rate with increasing temperature,which verifies the accuracy of the ***,it also revealed that the total decomposition rate after fungal incorporation was negatively correlated with the decomposition rate of a fungal single *** on the above model,predictions can be made for fungal growth in different environments,and suitable environments for fungal growth can be *** the future,the model can be further optimized,and lignin and cellulose decomposition factors can be added to fit the decomposition of *** application
DARTS has emerged as a popular method for neural architecture search (NAS) owing to its efficiency and simplicity. It employs gradient-based bi-level optimization to iteratively optimize the upper-level architecture p...
详细信息
Parameter-efficient fine-tuning (PEFT) techniques have emerged to address issues of overfitting and high computational costs associated with fully fine-tuning in the paradigm of self-supervised learning. Mainstream me...
详细信息
China is experiencing accelerated urbanisation,with a large number of people moving from rural to urban areas[1].It has resulted in large losses in the net primary production(NPP),biodiversity and carbon stocks and an...
详细信息
China is experiencing accelerated urbanisation,with a large number of people moving from rural to urban areas[1].It has resulted in large losses in the net primary production(NPP),biodiversity and carbon stocks and an increase in environmental pollution and CO_(2)emissions[2–4].In 2015,196 countries signed the Paris Agreement and committed to setting long-term goals to jointly manage climate change and reduce their individual emissions,aiming to control the increase in global average temperature from the pre-industrial level to below 2℃and to curtail the temperature rise within 1.5℃till the end of the 21st century[5].China is bolstering its efforts to achieve the climate change mitigation goals and has announced a plan for achieving carbon neutrality by 2060[6].The carbon neutrality goal poses a challenge to the current policies promoting rapid urbanisation across China.
Histopathology and transcriptomics are fundamental modalities in cancer diagnostics, encapsulating the morphological and molecular characteristics of the disease. Multi-modal self-supervised learning has demonstrated ...
详细信息
Histopathology and transcriptomics are fundamental modalities in cancer diagnostics, encapsulating the morphological and molecular characteristics of the disease. Multi-modal self-supervised learning has demonstrated remarkable potential in learning pathological representations by integrating diverse data sources. Conventional multi-modal integration methods primarily emphasize modality alignment, while paying insufficient attention to retaining the modality-specific intrinsic structures. However, unlike conventional scenarios where multi-modal inputs often share highly overlapping features, histopathology and transcriptomics exhibit pronounced heterogeneity, offering orthogonal yet complementary insights. Histopathology data provides morphological and spatial context, elucidating tissue architecture and cellular topology, whereas transcriptomics data delineates molecular signatures through quantifying gene expression patterns. This inherent disparity introduces a major challenge in aligning these modalities while maintaining modality-specific fidelity. To address these challenges, we present MIRROR, a novel multi-modal representation learning framework designed to foster both modality alignment and retention. MIRROR employs dedicated encoders to extract comprehensive feature representations for each modality, which is further complemented by a modality alignment module to achieve seamless integration between phenotype patterns and molecular profiles. Furthermore, a modality retention module safeguards unique attributes from each modality, while a style clustering module mitigates redundancy and enhances disease-relevant information by modeling and aligning consistent pathological signatures within a clustering space. Extensive evaluations on The Cancer Genome Atlas (TCGA) cohorts for cancer subtyping and survival analysis highlight MIRROR’s superior performance, demonstrating its effectiveness in constructing comprehensive oncological feature representations and be
暂无评论