Search Results (1,066)

Accurate as-built models of offshore platforms require component-level information from dense laser scans, but point-wise annotation is costly. We present Offshore Semi Supervised Contrastive Segmentation (OSCS), a framework for drilling platform point clouds when only a few complete scenes are annotated. OSCS constructs fixed-point-count 3D nearest-neighbor crops and uses a shared hierarchical attention encoder–decoder for supervised classification and contrastive representation learning. In unlabeled scenes, retained point identities establish positive correspondences between overlapping views. Predicted classes, confidence gating, and cross-batch class-wise feature queues organize contrastive samples without converting predictions into classification targets. Experiments used the Offshore Drilling Platform Tree Dataset (ODPT), with additional evaluation on Stanford Large-Scale 3D Indoor Spaces (S3DIS). At label budgets of approximately 10% and 20% on ODPT, OSCS achieved 88.03 ± 1.06% and 89.69 ± 0.97% mean intersection over union (mIoU) across three labeled-scene selections. The corresponding mean gains over matched supervised baselines were 4.57 and 4.42 percentage points. In the original fully labeled method comparison, OSCS achieved 92.27% mIoU, 2.67 percentage points above the strongest competing result under that protocol. Under the S3DIS Area 5 evaluation protocol, OSCS achieved 70.23% mIoU after retraining on the remaining areas. OSCS provides component-wise semantic point sets from partly annotated offshore scans, supporting the preparation of geometric information for retrofit planning and onshore prefabrication. Full article

Cucumber target spot disease, caused by Corynespora cassiicola, is a major constraint on greenhouse cucumber production. Accurate pixel-level segmentation and automated severity assessment remain challenging because early lesions are small, lesion boundaries are gradual, and field images often contain complex illumination and background interference. To address these issues, this study proposes a small-lesion and boundary-aware Mask2Former framework with a ResNet-50 (R50) backbone for cucumber target spot segmentation. Specifically, a Small Target Enhancement Module (ST) is designed by integrating atrous spatial pyramid pooling (ASPP), a high-resolution feature retention branch, and a small-target-weighted loss to improve sensitivity to early micro-lesions. In addition, a Boundary Segmentation Module (BS) is introduced to enhance boundary localization through boundary attention and explicit supervision with Dice and Focal losses. A field dataset containing 2559 cucumber leaf images was annotated at the pixel level and split into training, validation, and test sets at a ratio of 7:1:2. Disease severity was categorized into four grades based on the lesion-to-leaf-area ratio. On the test set, the proposed model achieved an mIoU of 85.05%, a Dice coefficient of 91.94%, and a precision of 92.30%, outperforming the Mask2Former baseline by 4.95, 4.72, and 4.58 percentage points, respectively. Moreover, severity assessment based on segmentation results reached an overall grading accuracy of 91.6% (Cohen’s Kappa = 0.89), with misclassifications predominantly confined to adjacent severity grades and no large cross-category errors observed. These results indicate that the proposed method has practical potential for automated disease monitoring and precision management in greenhouse cucumber production. Full article

Robust traffic sign detection is essential for reliable autonomous driving systems; however, detection performance can be substantially affected by adverse environmental conditions. Although simulation-based approaches are widely used to evaluate robustness, the extent to which simulated weather reproduces real-world environmental effects remains insufficiently understood. This study presents a controlled, comparative evaluation methodology for assessing a YOLOv8-based traffic sign detection model across four environmental conditions: clear (sunny), simulated rain, real-world rain, and simulated snow. The same road segment, camera configuration, and predefined set of 55 traffic-sign instances were maintained across the evaluated conditions, enabling interpretable comparisons while minimizing scene-level variability. Detection performance was assessed using representative detection confidence (RDC) at the traffic-sign-instance level, along with image-quality metrics such as sharpness, saturation, and intensity. Mean RDC was highest under clear conditions (0.825), followed descriptively by simulated snow (0.647), real-world rain (0.408), and simulated rain (0.395). However, simulated and real-world rain did not differ significantly in RDC (Holm-adjusted p = 0.770), while the Friedman test indicated a significant overall difference among conditions (χ²(3) = 98.767, p < 0.001; Kendall’s W = 0.599). Image-quality analysis further revealed substantial differences between rainfall conditions in successfully detected traffic-sign regions, particularly in Sharpness and Mean Saturation. Overall, the findings demonstrate that simulated weather can produce traffic-sign-level detector responses that are statistically comparable to those observed under independently recorded real-world rainfall, while producing substantially different image-level characteristics. The results support the use of controlled weather simulation as a complementary evaluation approach, alongside real-world validation, to investigate the environmental robustness of camera-based traffic-sign detection systems. Full article

Rapid urban population expansion in semi-arid environments poses considerable challenges for sustainable planning, particularly where demographic growth occurs under environmental constraints. This study examines the statistical association between climatic variability and population change in the Soran district center, Kurdistan Region of Iraq, and develops an exploratory framework for forecasting future demographic trends. It uses the official annual population series compiled by the Statistics Directorate of the Soran Independent Administration for 2010–2024 (15 observations) together with monthly meteorological records for 2015–2024 from the Soran agro-meteorological station. Two properties of the demographic record govern what can be inferred from it. The 2010–2023 values are inter-censal estimates reproduced to within 0.10% by a single declining-growth rule, so goodness-of-fit statistics obtained on that segment measure agreement with an interpolation rule rather than forecasting skill; and the 2024 value derives from the 2024 census round, representing a level shift of +18.4% against a 2.0% per year trend—178 residual standard deviations—rather than a year of growth. Detrended, lagged Spearman correlations across twelve climate–lag combinations yielded no association surviving Holm–Bonferroni correction; because the pre-2024 values are administratively smoothed, this is reported as an inability of the available data to resolve a climate–demography association rather than as evidence of independence. Eight specifications—naïve, drift and linear-trend benchmarks, a second-order polynomial, a log-linear trend, an AIC-selected ARIMA, and two multilayer perceptrons—were compared under identical rolling-origin cross-validation, producing seven one-step-ahead forecasts each. On the inter-censal segment a random walk with drift attained a mean absolute percentage error of 0.087%, better than every fitted specification; on the census fold every specification erred by at least 11,663 persons. The reported 2025–2030 projection is therefore anchored on the 2024 census level with growth extrapolated from its 2011–2023 trend, giving 93,342 residents in 2025 and 101,691 in 2030 (scenario range 100,395–104,064), reproduced to within 488 persons by an ARIMA(0,2,2) with a 2024 step term. Polynomial and log-linear trends fitted across the discontinuity project 2025 values are below the observed 2024 population and are reported as specification failures. The study demonstrates both the potential and the limitations of demographic forecasting in data-scarce semi-arid urban settings and shows that the provenance of an administrative population series materially constrains the claims such a study can make. Full article

Flushing planning requires utilities to prioritize pipes and service blocks using heterogeneous structural and hydraulic information. This study proposes a data envelopment analysis (DEA)-assisted framework that links pipe-level screening, service-block prioritization, and candidate flushing-segment planning. The framework separately evaluates sedimentation and detachment conditions using input-oriented Banker–Charnes–Cooper (BCC) models based on seven indicators, while avoiding the need to prescribe a common expert-defined weighting vector. The framework was applied to a real water distribution system with 481 pipes and 394 junctions, identifying 88 sedimentation-priority pipes and 12 detachment-priority pipes. Length-weighted aggregation within the existing block structure ranked Blocks 1, 7, 2, and 4 highest, whereas length-normalized analysis identified Block 2 as having the greatest concentration of priority pipes. The leading four-block group remained unchanged across the tested pipe-length exponents, and the top-12 detachment set was retained when the velocity-difference screening level was varied from 0.5 to 2.0 m/s. Additional model-form, preprocessing, indicator-omission, and equal-weight comparisons were used to characterize the sensitivity of pipe-level priorities. Finally, existing valve and hydrant locations were used to delineate five candidate flushing segments in a representative block. The framework provides a systematic basis for directing detailed hydraulic verification and field implementation. Full article

Background/Objectives: Supervised electrocardiography-based sleep apnea detection often depends on dense segment-level annotations, creating substantial expert burden. To reduce this dependence, we investigated whether adjacent temporal context could improve classification when ground-truth apnea labels are limited. Methods: We systematically evaluated five context lengths and seven apnea-label retention levels across 35 experimental configurations using participant-grouped five-fold cross-validation with three training repetitions. Results: The results demonstrated that using multi-window temporal context consistently improved classification performance across the evaluated apnea-label retention range. All multi-window configurations achieved higher mean F1-scores than the corresponding single-window baseline, with the best-performing multi-window setting at each retention level providing gains of 7.18–12.23 percentage points. Context length showed a clear overall effect on performance, whereas no systematic interaction was observed between context length and apnea-label retention. Targeted repeated-center and random-shuffle control experiments further supported a contribution from temporally adjacent ECG information rather than sequence-length expansion, repeated target presentation, or temporally distant same-recording context. Conclusions: Overall, leveraging adjacent temporal context provides a practical strategy for improving ECG-based apnea classification and may be particularly useful for developing screening models when dense expert-provided apnea annotations are limited. Full article

Background: Measurement of anterior chamber depth (ACD) is an important screening method for angle-closure glaucoma risk. We aimed to determine the precision and agreement of a smartphone-based method for measuring ACD when compared with anterior segment optical coherence tomography (AS-OCT) and Smith’s technique. Methods: A total of 43 participants (22 female, 21 male) were recruited. The mean age ± standard deviation of participants was 21.9 ± 2.1 years (range = 19–27 years). Three measurements were taken on each participant’s right eye in a random order for each technique. For this comparative study, differences in ACD, test–retest variability of a given method, and Bland–Altman level of agreement were computed to compare differences in performance between all three methods. Results: Smith’s technique over-estimated ACD, while smartphone photography under-estimated ACD, when compared to the reference AS-OCT. There was proportional bias between all three techniques. Smith’s technique had the largest degree of variability, whereas there was no statistically significant difference in test–retest variability between AS-OCT and smartphone photography. Conclusions: Zamir’s smartphone photography technique could be used for measuring ACD. Although there are some limitations compared to other established approaches, smartphone photography may have potential as a low-cost screening tool for estimating ACD in settings where specialised equipment is not available. Further studies are required to establish diagnostic performance for angle-closure screening in clinically relevant populations. Full article

Accurate point-cloud segmentation of cotton organs is essential for precise phenotypic characterization. However, reliable instance segmentation of cotton bolls in field-derived point clouds remains challenging because foliage occlusion, contact between adjacent bolls and incomplete reconstruction obscure instance boundaries. Here we present RQ-PointNeXt, an end-to-end framework that directly maps input point clouds to boll instance masks within a unified trainable network. Built on PointNeXt, it incorporates relative-elevation geometric channel attention in the shallow encoder to fuse global channel context with elevation and surface-normal cues. Its Query–mask branch integrates semantic guidance, center-seeded queries and a center-aware mask prior to suppress background responses, localize instances and constrain mask extent. Hungarian matching and multitask optimization establish one-to-one query–instance assignments, whereas query-based decoding produces instance masks without external geometric clustering. We evaluated the framework on 226 field-grown cotton plants containing 720 annotated boll instances reconstructed from UAV multi-view imagery using neural radiance fields. On the held-out test set, overall accuracy, mean class accuracy and mean intersection over union reached 0.8942, 0.8980 and 0.8076, respectively. AP25, AP50 and AP75 were 0.7359, 0.5585 and 0.2777, yielding an mAP25/50/75 of 0.5240. The framework provides instance-level outputs for boll counting and spatial analysis in high-throughput field phenotyping. Full article

Background/Objectives: To develop and evaluate an automated CT-based framework for the quantitative assessment of fibrotic interstitial lung disease (ILD), including idiopathic pulmonary fibrosis (IPF), using a standardised six-level anatomical protocol and deep-learning lung segmentation. Methods: The segmentation dataset comprised 3315 manually annotated development slices from 92 patients and a non-overlapping internal holdout of 845 slices from 5 patients. A separate 100-study localisation/scoring set yielded a 97-patient agreement cohort (84 IPF, 13 other ILD; 1164 per-level, per-lung observations) after three DICOM-conversion exclusions. YOLO11n-seg masks underwent vessel- and structure-removal fibrosis detection. The radial spatial score was compared with a non-blind expert-adjudicated reference; the per-level Fibrosis Index was an auxiliary read-out. Results: On the five-patient internal segmentation holdout, mean intersection over union (mIoU) was 0.926 ± 0.017; the in-sample development value was approximately 0.95. Model-only latency was 73.8 ± 9.7 ms/slice at batch size 1, and peak throughput was 1.48 ms/slice at batch size 512. In the separate 97-patient agreement cohort, the expert-adjudicated score was identical to the automated score for 967 of 1164 observations (83.1%) and differed for 197 (16.9%). In the modified-score subset, Pearson r was 0.918, mean absolute error was 3.07, and ICC(2,1) was 0.889 (patient-clustered 95% CI 0.828–0.923). The pooled ICC(2,1) was 0.988 (0.982–0.992), but this value was inflated because the 967 unchanged pairs were identical by construction. Sequential end-to-end processing, measured in seven study patients, took a mean of 22.5 s per patient (median 24.0 s, range 17.3–24.9 s); localisation accounted for 88.1% of this time. Conclusions: The framework combined lung segmentation, anatomically standardised sampling, and automated fibrosis scoring. The radial score showed preliminary analytical concordance under non-blind expert adjudication. The fibrosis detector remains a proof-of-concept implementation based on 8-bit windowed images and has not been compared with independently drawn pixel-level fibrosis masks. The radial partition is an exploratory scoring convention and was not compared with alternative partitions or validated against clinical outcomes. Larger external studies using native Hounsfield-unit data and independent blinded readers are required. Full article

Failure of axonal maintenance is proposed as a mechanism shared by Parkinson’s disease (PD) and amyotrophic lateral sclerosis (ALS). We reanalysed three public post-mortem resources under one rule set: bulk RNA-seq of 1242 samples from 319 donors (GSE153960) and midbrain single-nucleus RNA-seq (GSE157783, GSE178265). As a contributing project, a second batch variable tracked diagnosis in both cord segments and was completely separated from it in three cortical regions, which we removed. Adjusting for this removed a third of the naive differential expression. In ALS cord, the dominant depleted program was microtubule-based axonal transport (normalised enrichment score −2.40, FDR < 0.001); a regeneration-associated panel reached significance in none of ten fits. An ALS cord signature carried into the PD midbrain and scored highest on microglia in all 11 donors (+3.55 versus +0.73 next). Of 86 gene sets significant in both diseases, a translation block contained the GCN2 amino-acid-deficiency response. Both axes are compartment-level: within PD microglia, the complement panel is null (+0.21, p = 0.57). No GCN2 activity was measured, and the ALS cord is compared against PD midbrain. The axes that survive this control are glial and translational; the axonal question is not adjudicable in PD, where the panel score tracks dopaminergic content. Full article

Karst piedmont fault overflow springs are widely developed in structurally controlled, basin-margin settings, where basin-bounding faults obstruct regional groundwater flow. However, the coupled fault blocking, fault conduction, and caprock sealing mechanisms governing their genesis remain insufficiently quantified. This study investigates the Shentou Spring system in Shanxi Province, North China—a typical piedmont fault overflow spring—to develop a three-dimensional genetic model characterised by coupled blocking, conduction, and overflow processes. Integrating borehole datasets, multi-year groundwater level monitoring records, hydrochemical and isotopic measurements, and detailed structural mapping, this study identifies three key controlling mechanisms. First, spatial variations in the throw of the Mayi Fault control fault blocking efficiency, partitioning the fault zone into complete barrier and semi-permeable segments, which underpins the incomplete drainage behaviour of the spring system. Second, the Gengzhuang Fault intersects the high-permeability Qilihe and Yuanzihe groundwater flow zones, acting as the primary conduit that transports groundwater from distant recharge areas to the discharge zone. Third, the Quaternary caprock in the discharge area features a critical thickness threshold of approximately 30 m; confined karst groundwater breaches the overlying caprock and forms spring outlets where caprock thickness falls below this threshold. The proposed tripartite coupled model provides a semi-quantitative framework for interpreting the genesis of piedmont fault overflow springs. In practical terms, it supports the delineation of fault-conduit protection zones and the design of long-term water-quality monitoring networks along fault-controlled flow paths. Full article

Resource-intensive mobile edge computing (MEC) services are often provisioned on a per-request basis, resulting in repeated activation of equivalent service instances and redundant transmission of the same category-level state over overlapping inter-station links. Existing approaches rarely integrate demand aggregation, shared-instance provisioning, and reusable multi-target state distribution into a unified orchestration workflow. This paper proposes Coalition-based Shared Instance Provisioning and Routing (CoSIPR), a shared-service orchestration framework built around dynamic interest coalitions. CoSIPR predicts user requests and mobility, projects predicted locations onto the road network, filters unreliable or infeasible requests, and groups nearby users requesting the same service category. For each coalition, a marginal-gain-based candidate-reduction method and variable neighborhood search determine the serving stations, user assignments, and shared-instance counts. The selected stations then form the target set for a load-aware routing procedure that selects an existing state source and uses path-fusion reinforcement learning (PF-RL) to construct routes that reuse path segments across multiple targets. Experiments using real-world mobility and road-network data show that CoSIPR improves service-category matching, request satisfaction, and the average number of accepted requests per instance. It also reduces aggregate state-transfer cost and limits hotspot exposure while maintaining a controlled trade-off between end-to-end delay and state-transfer cost. These results demonstrate that dynamic interest coalitions and reusable multi-target paths can improve the efficiency of shared-service orchestration in MEC. Full article

Mapless navigation often removes global maps while retaining localization-derived goal vectors or bearings. We study a stricter setting in which a mobile robot observes only local LiDAR, scalar goal range, and short histories of executed actions; neither pose nor goal direction is provided to the policy. We introduce ACR-Nav, an action-conditioned range navigation framework that converts scalar-range evolution into closed-loop progress information. Its range–action history associates each distance change with the motion that produced it, while the sectorized LiDAR captures local geometry and short-term obstacle motion. A LiDAR-only safety filter provides immediate collision intervention, and a static-to-mixed curriculum stabilizes learning. A lightweight multilayer–perceptron is optimized with Proximal Policy Optimization (PPO), while the ACR-Nav formulation itself remains optimizer-agnostic. In corridor simulations, ACR-Nav achieved 93.2%, 80.4%, and 84.4% success in static, mixed, and dynamic environments. Removing the safety filter reduced success by 15.2, 14.6, and 16.0 percentage points in static, mixed, and dynamic environments, respectively, and random-goal tests yielded 91.2% and 81.4% success in static and mixed settings. Topology-shift experiments further quantified adaptation to an L-shaped corridor. The results show that action-conditioned scalar-range evolution can support goal-directed, segment-level navigation within locally straight corridor passages without exposing robot pose or target bearing to the policy. Full article

Disaster response often requires evidence from several sources, including satellite imagery, social media, news reports, audio and video recordings, and event metadata. Existing disaster datasets, however, usually focus on one source, a limited set of modalities or a single task. This separation makes it difficult to evaluate models that must combine regional observations with ground-level evidence for the same disaster context. We introduce DisasterScope, an event-centric multimodal dataset and benchmark for disaster-type recognition, severity assessment and evidence retrieval. DisasterScope organizes satellite observations, social images, synchronized audio–video segments, observation text, report passages, and provenance metadata around canonical events. Its primary benchmark contains 12,159 fixed event-context bundles from 24 events and nine disaster classes. The dynamic-media inventory includes 273 source videos, 628 curated audio–video segments, 9.02 h of material and 5473 temporal windows. Labels are harmonized through source-label inheritance, taxonomy mapping, teacher assistance and partition-specific human review. Validation and test annotations were reviewed in full, while training annotations were sampled for review and accepted through a threshold gate. The benchmark evaluates full-input and unavailable-view conditions on the same bundle identities, allowing direct measurement of how each view affects a model without changing the evaluated sample population. DisasterScope therefore provides a traceable setting for studying multimodal disaster assessment across different disasters, evidence sources and input-availability conditions. Full article

Background/Objectives: Accurate delineation of the optic disc and optic cup in retinal fundus photographs is a prerequisite for automated glaucoma screening. While encoder–decoder segmentation models have advanced considerably, the contribution of upstream preprocessing to segmentation accuracy, and the stability of that contribution across repeated training runs, remain insufficiently characterized. Methods: Five preprocessing pipelines, baseline, Contrast Limited Adaptive Histogram Equalization (CLAHE), Region of Interest (ROI) cropping, ROI+CLAHE, and CLAHE with heavy augmentation, were benchmarked under a fixed EfficientUNet++ model with an EfficientNet-B7 encoder on three publicly available fundus datasets (REFUGE, ORIGA, and Drishti-GS). Every configuration was retrained under three independent random seeds (42, 15, and 89) to assess run-to-run variability. Seed-level standard deviations accompany every reported mean and define the confidence limit on each ranking. Results: On REFUGE, CLAHE with augmentation (Config 5) achieved the strongest mean Dice (disc $0.9523\pm 0.0017$; cup $0.8348\pm 0.0018$). On ORIGA, all five configurations clustered within 0.0067 disc Dice; ROI+CLAHE (Config 4) was marginally ahead on disc ($0.9681\pm 0.0002$) and augmentation led on the cup ($0.8873\pm 0.0024$). On Drishti-GS, all five configurations converged successfully once optimizer and loss settings were corrected; the near-total failures seen in earlier single-run experiments reflected a configuration problem, not the small (81-image) training set. Conclusions: CLAHE applied to full-resolution images is the single most consistently beneficial preprocessing choice across all three datasets. ROI+CLAHE showed a small, initialization-stable advantage on ORIGA, but ROI crop centres were derived from ground-truth centroids, an oracle localization setting, and these results should not be interpreted as achievable by a fully automated pipeline. Data augmentation showed a consistent reduction in initialization sensitivity on small datasets and may be beneficial as a default strategy. Full article