Segmentation Tracking Research Articles

A Railway Track Extraction Method Based on Improved DeepLabV3+

Extracting railway tracks is crucial for creating electronic railway maps. Traditional methods require significant manual labor and resources while existing neural networks have limitations in efficiency and precision. To address these challenges, a railway track extraction method using an improved DeepLabV3+ model is proposed, which incorporates several key enhancements. Firstly, the encoder part of the method utilizes the lightweight network MobileNetV3 as the backbone extraction network for DeepLabV3+. Secondly, the decoder part adopts the lightweight, universal upsampling operator CARAFE for upsampling. Lastly, to address any potential extraction errors, morphological algorithms are applied to optimize the extraction results. A dedicated railway track segmentation dataset is also created to train and evaluate the proposed method. The experimental results demonstrate that the model achieves impressive performance on the railway track segmentation dataset and DeepGlobe dataset. The MIoU scores are 88.93% and 84.72%, with Recall values of 89.02% and 86.96%. Moreover, the overall accuracy stands at 97.69% and 94.84%. The algorithm’s operation time is about 5% lower than the original network. Furthermore, the morphological algorithm effectively eliminates errors like holes and spots. These findings indicate the model’s accuracy, efficiency, and enhancement brought by the morphological algorithm in error elimination.

Performance Evaluation of a Combined Transition System in Slab-Ballasted Railway Track Using a Vehicle-Track-Substructure Interaction Model

Abrupt stiffness variations along the railway track may increase the geometrical and mechanical defects of railway lines. The conjunction points of a railway track with concrete and ballast pavements, which are called slab-ballasted track transitions, are one of the main areas where vertical track stiffness changes sharply. Therefore, the potential benefits of a combined transition system along the slab-ballasted transition, made of an approach slab and additional rails, are studied in this paper. For this purpose, a vehicle-track-substructure interaction model, which included three main segments of the railway track (slab track, transition zone, and ballasted track) was programmed based on the finite element method. A test line with the mentioned combined transition system was built to measure the railway track responses through field study. Then, the three-dimensional (3D) numerical model was validated using the results obtained from the experimental tests. Afterwards, a number of parametric studies were performed to analyze the dynamic responses of the combined transition zone. The results indicated that this type of transition system promotes a smoother stiffness transition between the slab track segment and the ballasted track segment by making the transition in three gradual steps. The track displacements in the analyzed case-study gradually increased by about 22%, 28%, and 34% along the combined transition zone in the junction points of the slab and ballasted tracks.

A Fully Automated Post-Surgical Brain Tumor Segmentation Model for Radiation Treatment Planning and Longitudinal Tracking.

Glioblastoma (GBM) has a poor survival rate even with aggressive surgery, concomitant radiation therapy (RT), and adjuvant chemotherapy. Standard-of-care RT involves irradiating a lower dose to the hyperintense lesion in T2-weighted fluid-attenuated inversion recovery MRI (T2w/FLAIR) and a higher dose to the enhancing tumor on contrast-enhanced, T1-weighted MRI (CE-T1w). While there have been several attempts to segment pre-surgical brain tumors, there have been minimal efforts to segment post-surgical tumors, which are complicated by a resection cavity and postoperative blood products, and tools are needed to assist physicians in generating treatment contours and assessing treated patients on follow up. This report is one of the first to train and test multiple deep learning models for the purpose of post-surgical brain tumor segmentation for RT planning and longitudinal tracking. Post-surgical FLAIR and CE-T1w MRIs, as well as their corresponding RT targets (GTV1 and GTV2, respectively) from 225 GBM patients treated with standard RT were trained on multiple deep learning models including: Unet, ResUnet, Swin-Unet, 3D Unet, and Swin-UNETR. These models were tested on an independent dataset of 30 GBM patients with the Dice metric used to evaluate segmentation accuracy. Finally, the best-performing segmentation model was integrated into our longitudinal tracking web application to assign automated structured reporting scores using change in percent cutoffs of lesion volume. The 3D Unet was our best-performing model with mean Dice scores of 0.72 for GTV1 and 0.73 for GTV2 with a standard deviation of 0.17 for both in the test dataset. We have successfully developed a lightweight post-surgical segmentation model for RT planning and longitudinal tracking.

The influence of the melt pool structure on the mechanical properties of laser powder bed fusion Nickel Superalloy 625

Qualification of additive manufacturing (AM) parts requires exploring the process-structure-property relationships for a large number of process variables and unique macro/microstructure features associated with AM processes. This investigation explores the unique melt pool structure influences on a Nickel superalloy 625. Samples were built with different laser parameters and build orientations. They were then systematically characterized, including the macrostructure down to the microstructure sub-micrometer dendrites. Mechanical characterization was performed through tensile testing, microhardness, strain hardening behavior, and fracture analysis. The significance of the melt pool structure was shown to directly affect properties. The elongation of the parts was dependent on the continuity of the track segments, with easier crack propagation paths appearing in samples with more whole segment traces in the fracture (especially in Z samples). Interactions between built layers and tensile direction were also observed. The investigation elucidates properties and features that are process-related (independent of the material system), which can carry over to other material systems fabricated with additive manufacturing. The investigation also highlights ways that process-related and material-related properties and features interact.

Migrating eagles stopping in the Sahara Desert: Aberrant behaviour or foraging strategy?

AbstractThe Sahara Desert is one of the major ecological barriers of the Palearctic‐Africa bird migration system. We analysed the behaviour of GPS‐tagged booted eagles (Aquila pennata) breeding in Spain and wintering in Tropical Africa, focusing on the Sahara crossing. We searched for anomalous track segments recorded during the desert crossing, finding that non‐directional and/or slow movements occur in 35% of the migratory journeys and last up to 31 days (average: 5.9). The analyses show that this behaviour is not triggered by adverse weather conditions, and is shown especially by females during spring. The length of the stopping periods affects the arrival date to the breeding grounds only in males, which means that females buffer the potential delay with an earlier migration start, suggesting that this behaviour is followed deliberately. These wandering movements occur in hyper‐arid areas without vegetation cover. A similar behaviour has not been previously recorded in any other trans‐Saharan raptor and is probably related to the ability of booted eagles to take advantage of food resources unexploited by other predators, such as migratory songbirds stopping in the desert during the day, and migratory locusts. If confirmed by further research involving higher resolution tracking and field observations, our findings may show a new way of combining foraging and migration, without a well‐defined directed trajectory, nor stopping in a restricted spot, but instead wandering over wide areas.

An animal movement track segmentation framework for forecasting range adaptation under global change

The methods used for predicting space use and geographic distribution adaptations of animals in response to global change have relied on fitting statistical and machine learning models to environmentally-contextualized movement and spatial distribution data. These predictions, however, are made at particular spatiotemporal scales (from home range to species distribution), but no comprehensive methods have been proposed for predicting how changes to subdiel segments of individual movement tracks may lead to emergent changes in the lifetime tracks of individuals, and hence in the redistribution of species under global change. In this article, we discuss in terms of a hierarchical movement track segmentation framework that, anchored by diel activity routines (DARs), how adaptions in the canonical activity modes (CAMs) of movement can be used to assess space use adaptations to landscape and climate change at scales ranging from subdiel movement segments to the lifetime tracks (LiTs) of individuals.

Retracted: Study of Clustering Algorithms in Object Tracking and Image Segmentation

Retracted: Study of Clustering Algorithms in Object Tracking and Image Segmentation

Identifying emissions hotspots and strategies to reduce real-world fuel use and emissions for passenger rail: A spatially resolved approach

The objectives of this work are to model spatially resolved passenger locomotive fuel use and emission rates, locate emissions hotspots, and identify strategies to reduce trip train fuel use and emissions. Train fuel use and emission rates, speed, acceleration, track grade, and track curvature were quantified based on over-the-rail measurements, using portable emission measurement systems, for diesel and biodiesel passenger rail service on the Amtrak-operated Piedmont route. Measurements included 66 one-way trips and 12 combinations of locomotives, consists, and fuels. A locomotive power demand (LPD) based emissions model was developed based on the physics of resistive forces opposing train motion, taking into account factors such as speed, acceleration, track grade, and curvature. The model was applied to locate spatially-resolved locomotive emissions hotspots on a passenger rail route, and also identify train speed trajectories with low trip fuel use and emissions. Results show that acceleration, grade, and drag are the major resistive forces affecting LPD. Hotspot track segments have 3 to 10 times higher emission rates than non-hotspot segments. Real-world trajectories are identified that reduce trip fuel use and emissions by 13 % to 49 % compared to the average. Strategies for reducing trip fuel use and emissions include dispatching energy-efficient and low-emitting locomotives, using a 20 % blend of biodiesel, and operating on low-LPD trajectories. Implementing these strategies will not only decrease trip fuel use and emissions but reduce the number and intensity of hotspots and, thus, lowering the potential for exposure to train-generated pollution near railroad tracks. This work provides insights on reducing railroad energy use and emissions, which would lead to a more sustainable and environmental-friendly rail transportation system.

Research on the association algorithm of space-target track segments

In order to improve the tracking continuity and stability of space targets, the track segments association algorithms are studied. The extrapolation prediction based on the C-W motion model in the main spacecraft orbital coordinates is carried out. The model of space target is used to forecast and extrapolate the trajectory of the interrupted target. Based on the extrapolation of the terminated track, the track segments are associated by using the extrapolation results and the state update values at the first few moments of each track segment after the interruption. In this way, the association and fusion of the orbit of the space target interruption is completed. The results can improve the tracking accuracy of space targets and the trajectory association.

The New Small Wheel electronics

The increase in luminosity, and consequent higher backgrounds, of the LHC upgrades require improved rejection of fake tracks in the forward region of the ATLAS Muon Spectrometer. The New Small Wheel upgrade of the Muon Spectrometer aims to reduce the large background of fake triggers from track segments that don't originate from the interaction point. The New Small Wheel employs two detector technologies, the resistive strip Micromegas detectors and the “small” Thin Gap Chambers, with a total of 2.45 million electrodes to be sensed. The two technologies require the design of a complex electronics system given that it consists of two different detector technologies and is required to provide both precision readout and a fast trigger. It will operate in a high background radiation region up to about 20 kHz/cm2 at the expected HL-LHC luminosity of ℒ = 7.5 × 1034 cm-2 s-1. The architecture of the system is strongly defined by the GBTx data aggregation ASIC, the newly-introduced FELIX data router and the software based data handler of the ATLAS detector. The electronics complex of this new detector was designed and developed in the last ten years and consists of multiple radiation tolerant Application Specific Integrated Circuits, multiple front-end boards, dense boards with FPGA's and purpose-built Trigger Processor boards within the ATCA standard. The New Small Wheel has been installed in 2021 and is undergoing integration within ATLAS for LHC Run 3. It should operate through the end of Run 4 (December 2032). In this manuscript, the overall design of the New Small Wheel electronics is presented.

Disamenities of living close to transit tracks: Evidence from Singapore's MRT system

This study empirically tests how disamenities associated with Mass Rapid Transit (MRT) tracks affect public housing prices in Singapore. Residents living near aboveground rail tracks are more affected by MRT train noise than those living near underground rail tracks. Prices of public housing flats near the aboveground tracks are higher than those near the underground tracks by 3%–4% for every 1 km away from the MRT rail tracks. We use the erection of noise barriers on the selected segments of MRT tracks to mitigate noise effects as the shock in the models. The results show that housing prices increased by 2% and 3% for every 1 km reduction in distance to the segments of MRT tracks with new noise barriers after the announcement and completion of the projects, respectively. The welfare gains of the noise barrier project accrued to public housing residents near the noise-mitigating segments of rail tracks are estimated to be more than 700 million Singapore dollars.

Modeling of scavenging systems in water radiolysis with Geant4-DNA

PurposeThis paper presents the capabilities of the Geant4-DNA Monte Carlo toolkit to simulate water radiolysis with scavengers using the step-by-step (SBS) or the independent reaction times (IRT) methods. It features two examples of application areas: (1) computing the escape yield of H2O2 following a 60Co γ-irradiation and (2) computing the oxygen depletion in water irradiated with 1 MeV electrons. MethodsTo ease the implementation of the chemical stage in Geant4-DNA, we developed a user interface that helps define the chemical reactions and set the concentration of scavengers. The first application area example required two computational steps to perform water radiolysis using NO2- and NO3- as scavengers and a 60Co irradiation. The oxygen depletion computation technique for the second application area example consisted of simulating track segments of 1 MeV electrons and determining the radio-induced loss and gain of oxygen molecules. ResultsThe production of H2O2 under variable scavenging levels is consistent with the literature; the mean relative difference between the SBS and IRT methods is 7.2 % ± 0.5 %. For the oxygen depletion 1 µs post-irradiation, the mean relative difference between both methods is equal to 9.8 % ± 0.3 %. The results in the microsecond scale depend on the initial partial pressure of oxygen in water. In addition, the computed oxygen depletions agree well with the literature. ConclusionsThe Geant4-DNA toolkit makes it possible to simulate water radiolysis in the presence of scavengers. This feature offers perspectives in radiobiology, with the possibility of simulating cell-relevant scavenging mechanisms.

Influence of wheel tread wear on Rolling Contact Fatigue and on the dynamics of railway vehicles

This paper evaluates the influence of railway tread wear on the occurrence of Rolling Contact Fatigue (RCF) as well as other dynamic parameters of the railway vehicle in order to propose a tread maintenance limit. Around 160 unworn and worn AAR-1/20wheel profiles were measured and simulated inside the commercial software UM® - Universal Mechanism on a representative track segment of a Brazilian railway. With the simulated dataset, RCF was evaluated using the Shakedown Diagram, the Energy Dissipation Model, and the Wheel Gauge Side False Flange Gradient. Vehicle dynamics were assessed using the Nadal Criterion, contact pressure, and Rolling Resistance. The results show when the tread wear depth exceeds 2.0 mm, RCF is very likely to occur due to larger contact pressures. However, no clear relationship between tread wear, safety (Nadal Criterion), and fuel expenditure were observed. Therefore, this study suggests keeping wheelsets in service until a tread wear depth of 2.0 mm occurs in order to avoid RCF.

Multitarget Detection and Tracking by Mitigating Spot Jammer Attack in 77-GHz mm-Wave Radars: An Experimental Evaluation

Small form factor radar sensors at millimeter wavelengths find numerous applications in the industrial and automotive sectors. These radar sensors provide improved range resolution, good angular resolution, and enhanced Doppler resolution for short range and ultrashort ranges. However, it is challenging to detect and track the targets accurately when a radar is interfered by another radar. This article proposes an experimental evaluation of a 77-GHz IWR1642 radar sensor in the presence of a second 77-GHz AWR1642 radar sensor acting as a spot jammer. A real-time experiment is carried out by considering five different targets of various cross sections, such as a car, a larger size motorcycle, a smaller size motorcycle, a cyclist, and a pedestrian. The collected real-time data are processed by four different constant false alarm rate detectors, cell averaging (CA)-CFAR, ordered statistics (OS)-CFAR, greatest of CA (GOCA)-CFAR, and smallest of CA (SOCA)-CFAR. Following that, data from these detectors are fed into two different clustering algorithms (density-based spatial clustering of applications with noise (DBSCAN) and K-means), followed by the extended Kalman filter (EKF)-based tracker with global nearest neighbor (GNN) data association, which provide tracks of various targets with and without the presence of a jammer. Furthermore, four different metrics [tracks reported (TR), track segments (TSs), false tracks (FTs), and track loss (TL)] are used to evaluate the performance of various tracks generated for two clustering algorithms with four detection schemes. The experimental results show that the DBSCAN clustering algorithm outperforms the K-means clustering algorithm for many cases.

SiamMask: A Framework for Fast Online Object Tracking and Segmentation.

In this article, we introduce SiamMask, a framework to perform both visual object tracking and video object segmentation, in real-time, with the same simple method. We improve the offline training procedure of popular fully-convolutional Siamese approaches by augmenting their losses with a binary segmentation task. Once the offline training is completed, SiamMask only requires a single bounding box for initialization and can simultaneously carry out visual object tracking and segmentation at high frame-rates. Moreover, we show that it is possible to extend the framework to handle multiple object tracking and segmentation by simply re-using the multi-task model in a cascaded fashion. Experimental results show that our approach has high processing efficiency, at around 55 frames per second. It yields real-time state-of-the art results on visual-object tracking benchmarks, while at the same time demonstrating competitive performance at a high speed for video object segmentation benchmarks.

Laser powder bed fusion of oxidized microscale SiC-particle-reinforced AlSi10Mg matrix composites: Microstructure, porosity, and mechanical properties

Laser powder bed fusion (LPBF)-fabricated Al matrix composites reinforced with SiC particles are strongly influenced by SiC particle features. In this study, the microstructural features, mechanical properties, and porosity of moderate volume fraction (10 vol%) microscale SiC-particle-reinforced AlSi10Mg composites fabricated by LPBF under different oxidation temperatures of SiC particles were investigated. The results demonstrate that LPBF-deposited SiCp/AlSi10Mg composites mainly exhibited laser track segments in macroscale and fine cellular-like dendrite grains and (Al + Si) eutectic phases within the melt pool on a microscale. The transition layer of MgAl2O4 was present around the SiC particles in the oxidized-SiCp reinforced AlSi10Mg composites. In addition, the pore spatial distribution of the deposited composites using the microcomputed tomography test and the microstructural formation mechanism using Thermal Calc software were also studied. Using the finite element to model the deposited composites, it was found that the stress and strain distribution presented higher values than those without long-stripe precipitation phases in the deposited parts with long-stripe precipitation phases. The LPBF-fabricated oxidized-10 vol% SiCp/AlSi10Mg composite part with a relatively large ultimate tensile strength (∼405.3 MPa), relative density (∼98.89%), compressive strength (∼1046 MPa) and friction coefficient (0.32) were obtained. In particular, the ultimate tensile strength was larger than that of the other reported SiCp/AlSi10Mg composites with similar volume fractions and particle sizes.

Dealing with disruptions in railway track inspection using risk-based machine learning

Unplanned track inspections can be a direct consequence of any disruption to the operation of on-board track geometry monitoring activities. A novel response strategy to enhance the value of the information for supplementary track measurements is thus established to construct a data generation model. In this model, artificial (synthetic) data is assigned on each measurement point along the affected track segment over a short period of time. To effectively generate artificial track measurement data, this study proposes a NARX (nonlinear autoregressive with exogenous variables) model, which incorporates short-range memory dependencies in the dependent variable and integrates interdependent effects from external factors. Nonlinearities in the proposed model have been determined using an artificial neural network that allowed fast computation of a mapping function in line with the needs of effective disruption management. The risk of over fitting the data generation model, which reflected its generalisation ability, has been effectively managed through risk aversion concept. For the model evaluation, the deviation of track longitudinal level has been taken as a case study, predicted using its degradation rate and track alignment and gauge as exogenous variables. Simulation results on two datasets that are statistically different showed that the data generation model for disrupted track measurements is reliable, accurate, and easy-to-use. This novel model is an essential breakthrough in railway track integrity prediction and resilient operation management.

Smoothing and compressing algorithm of toolpath with complex contour in NC machining

The continuous short line segment processing method in CNC machining has the defect of frequent fluctuation of feedrate, which affects the quality and efficiency. For this problem, the local and global methods are common solutions. The local method is to insert transition curves at the corners to obtain a smooth toolpath, but this method does not essentially reduce the number of accelerations and decelerations, and it is difficult to insert transition curves between tiny straight sections. Although the global method can obtain smooth tool trajectories, the parametric interpolation method is not suitable for middle- and low-end CNC machine tools. The article proposes an adaptive tool path generation method with the original part contour as the research object and discretizes the NURBS curves into smooth trajectories consisting of large segments of straight lines and circular arcs. The article mimics the crawling characteristics of snakes and designs a double-headed snake algorithm based on the least squares method to implement this process. First, given the start and end points of the curve, two snakeheads search for the maximum linear and circular segments that satisfy the error constraint in the curve direction. The winner will then be selected as the candidate track segment through a competition mechanism. Finally, all trajectory segments are connected to obtain a smooth tool path. Experiments show that the toolpath data of the method in this paper are reduced by 75.56% compared with commercial CAM software when the contour error threshold is the same, and the contour error is reduced by more than one order of magnitude when the number of toolpath segments is equal. In addition, the method in this paper can obtain a smoother machining surface and more stable cutting force, thus achieving a win‒win situation of quality and efficiency.

An ATCA processor for Level-1 trigger primitive generation and readout of the CMS barrel muon detectors

An ATCA processor was designed to instrument the first layer of the CMS Barrel Muon Trigger. The processor receives and processes DT and RPC data and produces muon track segments. Furthermore, it provides readout for the DT detector. The ATCA processor is based on a Xilinx XCVU13P FPGA, receives data via 10 Gbps optical links and transmits track segments via 25 Gbps optical links. The processor is instrumented with a Zynq Ultrascale+ SoM connected with an SSD which provides the necessary resources for enhanced monitoring and control information. The design of the board as well as results on its performance are presented.

Data-driven design of the Belle II track segment finder

The Belle II experiment relies on a level-1 trigger system to reduce noise background and preselect events of interest for particle physics. The Central Drift Chamber is the main track detector which makes its trigger system important for online track reconstruction. To improve its hit efficiency, an extension of the track segment finder for low angle tracks is proposed. By combining hardware and software development flows, an automated data-driven pipeline is created and three different-sized hardware concepts are implemented. The operation point is adjustable to balance hit efficiency against hit purity in the trigger system.