Traffic violation is one of the leading causes traffic crashes. In context global aging, it important to study violations by elderly drivers for improving safety in preparation a worldwide aging population. this study, hybrid approach Latent Class Analysis (LCA) and XGBoost based SHAP proposed identify hidden clusters understand key contributing factors on severity drivers, police-reported dataset Beijing (China). First, LCA applied segment into several latent homogeneous clusters, then...

Water vapor, the most significant greenhouse gas, plays a critical role in Earth's climate system, influencing hydrological cycle, energy distribution, and atmospheric dynamics. Integrated Vapor (IWV) serves as vital parameter for understanding these processes. While traditional methods, including ground-based instruments satellite observations, provide IWV measurements, they are often limited by spatial resolution, coverage, accuracy. Advances numerical weather models (NWM) remote sensing...

Abstract Presurgical investigations for categorizing focal patterns are crucial, leading to localization and surgical removal of the epileptic focus. This paper presents a machine learning approach using information theoretic features extracted from high-frequency subbands detect focus interictal intracranial electroencephalogram (iEEG). It is known that (>80 Hz) include important biomarkers such as oscillations (HFOs) identifying commonly referred seizure onset zone (SOZ). In this...

Relief logistics is vital to disaster relief management. Herein, a risk-averse two-stage distributionally robust programming model proposed provide decision support for planning logistics. It distinct from the conventional problem in that (i) facility location-inventory and multi-commodity network flow formulation are integrated; (ii) probability distribution information of supply, demand, road link capacity partially known, (iii) optimisation (DRO) method based on worst-case...

The acquisition of road information from remote sensing images is significant value with regard to intelligent transportation research. This study focuses on enhancing the contour-learning ability mitigate phenomenon fragmented segments and missing connections in extraction. A novel Deep Feature-Review (FR) Transmit Network (TransNet) proposed review facilitate flow contour features into an encoder network. Meanwhile, multiscale are linked via a bridge between decoder. Compared...

The main goal of this study is to investigate the unobserved heterogeneity in VRU-MV crash data and determine relatively important contributing factors injury severity. For end, a latent class analysis (LCA) coupled with random parameters logit model (LCA-RPL) developed segment crashes into homogeneous clusters explore differences among clusters. random-forest-based SHapley Additive exPlanation (RF-SHAP) approach used relative importance for severity each cluster. results show that,...

Classifying point clouds obtained from mobile laser scanning of road environments is a fundamental yet challenging problem for asset management and unmanned vehicle navigation. Deep learning networks need no prior knowledge to classify multiple objects, but often generate certain amount false predictions. However, traditional clustering methods involve leveraging priori knowledge, may lack generalisability compared deep networks. This paper presents classification method that coarsely...

Background People usually spend most of their time indoors, so indoor fine particulate matter (PM 2.5 ) concentrations are crucial for refining individual PM exposure evaluation. The development concentration prediction models is essential the health risk assessment in epidemiological studies involving large populations. Methods In this study, based on monitoring data multiple types places, classical linear regression (MLR) method and random forest (RFR) algorithm machine learning were used...

Distributed scatter (DS) interferometric synthetic aperture radar is a powerful technology for analyzing displacements of the earth's surface. Unfortunately, preparatory step DS pre-processing enormously time consuming. The present research puts forward deep learning-based approach called Scatterers Prediction Net (DSPN), that can reduce computational load considerably. DSPN convolutional neural network, which generates candidate masks based on nine input layers. Masked pixels with low...

Mobile laser scanning (MLS) technology directly collects dense 3D coordinate information at expressway and satisfied with the demands of transportation-related road surveying. This paper aims to ex...

Abstract. Full-waveform LiDAR is an active technology of photogrammetry and remote sensing. It provides more detailed information about objects along the path a laser pulse than discrete-return topographic LiDAR. The point cloud waveform with high quality can be obtained by decomposition, which could make contributions to accurate filtering. surface fitting filtering method proposed present such advantage. Firstly, discrete parameters are resolved global convergent Levenberg Marquardt...

Traffic congestion is a critical problem in urban areas, and understanding the relationship between land use source crucial for traffic management planning. This study investigates land-use characteristics pattern features of parcels Second Ring Road Xi’an, China. The combines cell-phone data, POI data empirical analysis, uses spatial clustering approach to identify congested road sections trace them back parcels. correlations building factors patterns are explored using XGBoost algorithm....

Abstract The integrated design of traffic signal control (TSC) and reversible lane (RL) is an effective way to solve the problem tidal congestion with uncertainty at isolated intersections, because its advantage in making full use temporal‐spatial transportation facilities. Considering contradiction between dynamic TSC scheme fixed RL one period, a two‐stage optimization method based on improved mean‐standard deviation (MSD) model for intersections historical real‐time uncertain flow...

Multi-temporal high-density terrestrial laser scanning (TLS) datasets are processed to delineating possible movements from terrain surfaces and trees. Terrain surface estimated with the help of segmentation random sample consensus (RANSAC) algorithm. Tree interpreted by iterative closest point (ICP) solved translations rotations tree clouds. The capabilities proposed methodology were tested using a case study Slumgullion landslide, where trees without clear trunks cover surfaces....

In recent years, geodesy based on spaceborne microwave remote sensing has gained significant advances. However, whether the observations from Global Navigation Satellite System (GNSS) or Interferometric Synthetic Aperture Radar (InSAR), results are inevitably influenced by atmospheric tropospheric delay. Although zenith total delay (ZTD) can be estimated through gridded meteorological data products and empirical models provided ERA5 reanalysis product, its accuracy is still insufficient to...

Abstract In recent years, the evaluation of impact meteorological marine environment on combat effectiveness surface ships has been an important content and hot topic in military oceanography research naval battlefield support. this paper, analytic hierarchy process is adopted to establish hierarchical structure model, determine weights each index, quantitatively evaluate proportion influence wind, sea wave, rain, visibility, atmospheric waveguide, other factors ships. Combined with...

Abstract Special vehicles (SVs) are which conduct tasks such as the maintenance of urban roads and typically characterized by travelling at a lower speed constant rate within same lane. In order to reduce influence SVs, guidance zone is designed provides traffic suggestions (TGS) for human‐driven (HVs) helping drivers better decision between car‐following (CF) lane‐changing (LC). To verify effectiveness TGS, an improved Dogit‐agent‐based model established simulate captive not choice CF LC...