Mangrove forests, as important ecological and economic resources, have suffered a loss in the area due to natural human activities. Monitoring distribution of obtaining accurate information on mangrove species is necessary for ameliorating damage protecting restoring forests. In this study, we compared performance UAV Rikola hyperspectral images, WorldView-2 (WV-2) satellite-based multispectral fusion data from both classification species. We first used recursive feature elimination‒random...

Compositing an image usually inevitably suffers from inharmony problem that is mainly caused by incompatibility of foreground and background two different images with distinct surfaces lights, corresponding to material-dependent light-dependent characteristics, namely, reflectance illumination intrinsic images, respectively. Therefore, we seek solve harmonization via separable illumination, i.e., harmonization. Our method based on autoencoder disentangles composite into for further separate...

Abstract The underutilization of active sites limits the performance enhancement 2D transition metal boride (MBene) in electrocatalytic nitrogen reduction reaction (NRR). Herein, a highly efficient NRR electrocatalyst with S atoms bridging Fe and Mo on surface MBene is successfully constructed by using an site electron optimization strategy, which increases charge density around enhances activation ability catalyst to N 2 molecules. It noteworthy that FeS ‐MBene demonstrates low intrinsic...

Background The association between body mass index (BMI) and kidney stone formation may vary across different age groups follow nonlinear patterns. Methods This study analyzed data from NHANES 2009–2018, including 14,880 participants aged ≥20 years, to evaluate the BMI risk of stones. was categorized as normal weight (<25.0 kg/m 2 ), overweight (25.0–29.9 obesity (≥30.0 ). Weighted logistic regression models were employed adjust for multiple confounders, sex, age, race/ethnicity,...

The extensive expansion of coastal land-based aquaculture ponds (CLAP) has resulted in substantial wetland loss. To effectively monitor the dynamics CLAP on a large scale, critical spatial information derived from continuous datasets is essential. Here, we generate inaugural map depicting changes pond coverage across regions both China and Southeast Asia 1990 to 2020 by utilizing long time-series Landsat archive with an object-oriented hierarchical classification method. We extracted water...

An iterative two-stage structural damage identification approach that combines the modal strain energy decomposition (MSED) method with multiobjective particle swarm optimization algorithm is presented. The proposed scheme inspired by general noniterative approach, which attempts to locate in first stage and estimate severity of second stage; however, present differs constructing an MSED indicator perform localization estimation a cyclic manner. Consequently, false positive indications are...

An efficient meta-heuristic algorithm, named beetle swarm optimization (BSO), is proposed to localize and quantify structural damage using limited vibration measurement data. The antennae search (BAS) algorithm that imitats a random walking mechanism in nature was recently developed solve the problem. However, ratio of convergence this significantly relys on direction deviation for high-dimensional problems. To overcome shortcoming, BSO inspired by intelligence strategy proposed. In...

Scientific studies on fishes have considerable importance to the fishery industry and marine ecosystem. Fish images are typically collected by scuba divers or autonomous underwater vehicles then annotated manually biologists. In recent years, with rise of deep learning, field image classification has been greatly improved. But, fish which can be considered as fine-grained problem is more challenging than common classification. Usually, low-quality belong small-scale data, while classic CNN...

Abstract An adaptive control system with good following characteristics is designed for the electro-hydraulic servo which are strongly coupled and seriously nonlinear. The model of position on air rudder discussed uncertain equivalent to one parametric variable double input single output establish a Popov-MRAC (Model Reference Adaptive Control) system. simulation results show that Popov MRAC has characteristics, precision robustness.

A novel multistage approach is developed for structural model updating based on sensitivity ranking of the selected parameters. Modal energy-based sensitivities are formulated, and maximum-normalized indices designed ranking. Based strategy, a proposed, where these parameters to be corrected with similar levels updated simultaneously at same stage, complete procedure continues sequentially several stages, from large small, according predefined At every single previously cross mode (CMCM)...

The purpose of this study is to develop an effective iterative two-stage method (ITSM) for structural damage identification offshore platform structures. In each iteration, a new index, Modal Energy-Based Damage Index (MEBI), proposed help effectively locate the potential elements in first stage. Then, second stage, beetle antenna search (BAS) algorithm used estimate severity these elements. Compared with well-known particle swarm optimization (PSO) and genetic (GA), has lower computational...

The complex and changeable underwater environment light conditions lead to quality degradation problems of image in different degrees. Therefore, enhancement is an important challenging task computer vision. Relying on a powerful visual system, human beings have the ability restore real scene from degraded image, estimate object material situation image. Inspired by this, we believe that separating lighting effect produce enhanced result, effective method for enhancement. We utilize fitting...

Abstract Corrosion-induced defects, often extensive and unavoidable in marine structures, significantly compromise structural performance, posing a substantial risk to safety. To investigate the residual axial compression strength of corroded circular hollow section (CHS) steel tubes evaluate reparative effects composite patches, feasibility analysis is conducted through verifying performance tube with uniformly distributed corrosion pits perfect strengthened by Aramid fiber-reinforced...

Medical imaging data is inherently heterogeneous across different modalities and clinical centers, posing unique challenges for developing generalizable foundation models. Conventional entails training distinct models per dataset or using a shared encoder with modality-specific decoders. However, these approaches incur heavy computational overheads suffer from poor scalability. To address limitations, we propose the Multimodal Mixture of Experts (M$^4$oE) framework, leveraging SwinUNet...