Search Results (46)

Accurate rooftop segmentation from aerial imagery is essential for large-scale urban analysis, including applications such as solar potential assessment and urban monitoring. However, it remains constrained by the high cost of dense annotation and the limited generalisation of supervised models across heterogeneous urban morphologies. This study investigates binary rooftop segmentation for fine-tuning large image-editing foundation models using parameter-efficient Low-Rank Adaptation (LoRA). Using parts of Brisbane metropolitan dataset (split 80/20 into 97 training and 24 testing tiles), three paradigms were evaluated under a unified protocol: zero-shot image-editing models (including Gemini 3 Pro), a segmentation-first baseline (Segment Anything Model 3, SAM3), and LoRA-adapted diffusion models (FLUX.1 Kontext, FLUX.2, and Qwen Image Edit 2509) fine-tuned each 250 steps up to 5000 steps. Evaluated under zero-shot conditions, the generative models demonstrated varying levels of boundary fidelity. The Gemini model achieved a strong zero-shot baseline with [IoU, Dice] scores of [85%, 91%], followed by the SAM3 baseline, which also achieved a stable [84%, 91%] but exhibited increased false negatives in visually complex scenes. The tested diffusion models (FLUX.1 Kontext, FLUX.2, and Qwen) showed more limited initial spatial overlap, scoring [45%, 55%], [67%, 78%], and [33%, 46%], respectively. Following LoRA adaptation, the FLUX and Qwen models showed substantial improvements, with their respective [IoU, Dice] metrics increasing to [89%, 94%], [82%, 90%], and [87%, 93%]. FLUX.1 Kontext achieved the strongest overall performance at step 4250, yielding a mean IoU of 89% (SD = 3.16%) and a pixel accuracy exceeding 96%. These results demonstrate that parameter-efficient fine-tuning, combined with rigorous evaluation under class-imbalanced conditions, can transform general-purpose generative models into competitive, scalable spatial analysis tools that match or exceed both dedicated segmentation baselines and strong zero-shot multimodal models. Full article

Ion-adsorption rare-earth mining in southern China often leaves small, fragmented disturbances in rugged, forested terrain, making UAV-based enforcement challenging due to confusion with bare ground, canopy gaps, and shadows. We propose YOLO11-MSCAM, an enhanced YOLO11vm detector in which the original SPPF at the backbone–neck junction is replaced by a Multi-Scale Convolution–Attention Module that cascades channel attention, spatial attention, and multi-scale residual convolutions to enhance context aggregation and suppress background clutter. We build a field-acquired UAV dataset, SIMA (0.05 m GSD; September–November 2023), generating 1630 non-overlapping 640 × 640 orthomosaic tiles split into 1320/147/163 for training/validation/testing; five-lens raw images (nadir + oblique) are additionally used as auxiliary training samples and for post-detection verification. On the test set, YOLO11-MSCAM achieves mAP@0.5 = 83.24%, mAP@0.5:0.95 = 58.29%, and F1 = 79.92%, outperforming YOLOv11m and other detectors (YOLOv5m/6m/8m/9m/10m and Faster R-CNN with ResNet-50). With 19.67 M parameters, 67.34 GFLOPs@640, and 45.86 FPS, it supports tile-based batch screening to prioritize suspicious sites for field checks and evidence collection. Full article

This study presents a geometry-based post-processing framework developed to refine deep-learning detections of tornado-damaged trees. The YOLO11-based instance segmentation framework served as the baseline, but its predictions often included multiple masks for a single tree or incomplete fragments of the same trunk, particularly in dense canopy areas or within tiled orthomosaics. Overlapping masks led to duplicated predictions of the same tree, while fragmentation broke a single fallen trunk into disconnected parts. Both issues reduced the accuracy of tree-count estimates and weakened orientation analysis, two factors that are critical for treefall methods. To resolve these problems, a Shape-Aware Non-Maximum Suppression (SA-NMS) procedure was introduced. The method evaluated each mask’s collinearity and, based on its geometric condition, decided whether segments should be merged, separated, or suppressed. A spatial assessment then aggregated prediction vectors within a defined Region of Interest (ROI), reconnecting trunks that were divided by obstacles or tile boundaries. The proposed method, applied to high-resolution orthomosaics from the December 2021 Land Between the Lakes tornado, achieved 76.4% and 77.1% instance-level orientation agreement accuracy in two validation zones. Full article

Accurate delineation of farmland cultivation zones (FCZs) is crucial for advancing precision agriculture. However, identifying FCZs in landscapes where standardized and non-standard (fragmented) farmlands coexist remains a pressing challenge, primarily due to the lack of high-quality datasets covering such mixed patterns. To address this, we propose a novel tiling-based dataset generation method that integrates boundary probes and minimum-overlap Poisson-disk sampling (BP-MOPS). Using this strategy, we constructed a multi-temporal unmanned aerial vehicle (UAV) imagery dataset of FCZs—the multi-period farmland cultivation zones (MPFCZ) dataset—which encompasses three critical phenological stages: the dormant period (DP), the intermediate growing period (IGP), and the vigorous growing period (VGP). The source imagery was acquired over Zhouhu Village in China. The MPFCZ dataset comprises 6467 image patches (1024 × 1024 pixels), containing both standardized fields and fragmented cultivation zones typically missed by conventional methods. Both Transformer- and CNN-based models trained on MPFCZ surpassed those trained on the dataset generated by conventional segmentation strategy. The best-performing model achieved remarkable temporal change detection accuracy (mIoU > 0.82 across three phenological stages) and demonstrated strong cross-region generalization capability (0.8817 precision under zero-shot transfer). MPFCZ thus provides essential support for precise farmland identification in complex agricultural landscapes with standard and nonstandard fields mixed. Full article

This work addresses computational inefficiency in ultra-wide-area remote sensing image (RSI) object detection. Traditional homogeneous tiling strategies enforce computational symmetry by processing all image regions uniformly, ignoring the intrinsic spatial asymmetry of target distribution where target-dense coexist with vast target-sparse areas (e.g., deserts, farmlands), thereby wasting computational resources. To overcome symmetry mismatch, we propose a heat-guided adaptive blocking and dual-model collaboration (HAB-DMC) framework. First, a lightweight EfficientNetV2 classifies initial 1024 × 1024 tiles into semantic scenes (e.g., airports, forests). A target-scene relevance metric converts scene probabilities into a heatmap, identifying high-attention regions (HARs, e.g., airports) and low-attention regions (LARs, e.g., forests). HARs undergo fine-grained tiling (640 × 640 with 20% overlap) to preserve small targets, while LARs use coarse tiling (1024 × 1024) to minimize processing. Crucially, a dual-model strategy deploys: (1) a high-precision LSK-RTDETR-base detector (with Large Selective Kernel backbone) for HARs to capture multi-scale features, and (2) a streamlined LSK-RTDETR-lite detector for LARs to accelerate inference. Experiments show 23.9% faster inference on 30k-pixel images and reduction in invalid computations by 72.8% (from 50% to 13.6%) versus traditional methods, while maintaining competitive mAP (74.2%). The key innovation lies in repurposing heatmaps from localization tools to dynamic computation schedulers, enabling system-level efficiency for Ultra-Wide-Area RSIs. Full article

As a new type of bridge deck skin material, tiled laminates (TLs) are often subjected to bending actions in actual working conditions. This study employs a 3D progressive damage model (3D PDM) based on the Hashin damage criterion to investigate the influence of overlap configuration and inclination angle on the bending performance of tiled laminates in both elastic and non-linear stages through three-point bending (3PB) numerical simulations. The results indicate that, in the elastic stage, overlapped TLs (TLOs) exhibit a more uniform stress distribution due to their more rational geometric structure, and their bending stiffness is significantly less sensitive to the inclination angle compared to the non-overlapped TLs (TLNs). In the non-linear stage, damage in both configurations begins at the reduced section, and the ultimate midspan bending moment decreases with increasing inclination angles. Notably, cracks in the TLO configuration extend internally, enabling the structure to maintain a partial bending resistance up to failure, whereas cracks in the TLN configuration propagate externally, resulting in a rapid complete loss of structural bending performance. Furthermore, regardless of the geometric configuration and inclination degree, the final failure of the TL under bending is dominated by tensile failure. This research provides comprehensive insights into the bending mechanical behaviour of tiled laminates, offering scientific foundations for their optimised engineering design. Full article

Anti-tank landmines endanger post-conflict recovery. Detecting camouflaged TM-62 landmines in low-altitude unmanned aerial vehicle (UAV) imagery is challenging because targets occupy few pixels and are low-contrast and often occluded. We introduce a single-class anti-tank dataset and a YOLOv5 pipeline augmented with a SAHI-based small-object stage and Weighted Boxes Fusion. The evaluation combines COCO metrics with an operational operating point (score = 0.25; IoU = 0.50) and stratifies by object size and occlusion. On a held-out test partition representative of UAV acquisition, the baseline YOLOv5 attains mAP@0.50:0.95 = 0.553 and AP@0.50 = 0.851. With tuned SAHI (768 px tiles, 40% overlap) plus fusion, performance rises to mAP@0.50:0.95 = 0.685 and AP@0.50 = 0.935—ΔmAP = +0.132 (+23.9% rel.) and ΔAP@0.50 = +0.084 (+9.9% rel.). At the operating point, precision = 0.94 and recall = 0.89 (F1 = 0.914), implying a 58.4% reduction in missed detections versus a non-optimized SAHI baseline and a +14.3 AP@0.50 gain on the small/occluded subset. Ablations attribute gains to tile size, overlap, and fusion, which boost recall on low-pixel, occluded landmines without inflating false positives. The pipeline sustains real-time UAV throughput and supports actionable triage for humanitarian demining, as well as motivating RGB–thermal fusion and cross-season/-domain adaptation. Full article

Instance segmentation in high-resolution images is essential for applications such as remote sensing, medical imaging, and precision agriculture, yet remains challenging due to factors such as small object sizes, irregular shapes, and occlusions. Tiling-based approaches, such as Slicing-Aided Hyper Inference (SAHI), alleviate some of these challenges by processing smaller patches but introduce border artifacts and increased computational cost. Overlapping tiles can mitigate certain boundary effects but often result in duplicate detections and boundary inconsistencies, particularly along patch edges. Conventional deduplication techniques, including Non-Maximum Suppression (NMS) and Non-Mask Merging (NMM), rely on Intersection over Union (IoU) thresholds and frequently fail to merge fragmented or adjacent masks with low mutual IoU that nonetheless correspond to the same object. To address deduplication and mask fragmentation, Spatial Mask Merging (SMM) is proposed as a graph clustering approach that integrates pixel-level overlap and boundary distance metrics while using R-tree indexing for efficient candidate retrieval. SMM was evaluated on the iSAID benchmark using standard segmentation metrics, with tile overlap configurations systematically examined to determine the optimal setting for segmentation accuracy. The method achieved a nearly 7% increase in precision, with consistent gains in F1 score and Panoptic Quality over existing approaches. The integration of R-tree indexing facilitated faster candidate retrieval, enabling computational performance improvements over standard merging algorithms alongside the observed accuracy gains. Full article

Accurate segmentation of kidney microstructures in whole slide images (WSIs) is essential for the diagnosis and monitoring of renal diseases. In this study, an end-to-end instance segmentation pipeline was developed for the detection of glomeruli and blood vessels in hematoxylin and eosin (H&E) stained kidney tissue. A tiling-based strategy was employed using Slicing Aided Hyper Inference (SAHI) to manage the resolution and scale of WSIs and the performance of two segmentation models, YOLOv11 and YOLOv12, was comparatively evaluated. The influence of tile overlap ratios on segmentation quality and inference efficiency was assessed, with configurations identified that balance object continuity and computational cost. To address object fragmentation at tile boundaries, an Enhanced Syncretic Mask Merging algorithm was introduced, incorporating morphological and spatial constraints. The algorithm’s hyperparameters were optimized using Particle Swarm Optimization (PSO), with vessel and glomerulus-specific performance targets. The optimization process revealed key parameters affecting segmentation quality, particularly for vessel structures with fine, elongated morphology. When compared with a baseline without postprocessing, improvements in segmentation precision were observed, notably a 48% average increase for glomeruli and up to 17% for blood vessels. The proposed framework demonstrates a balance between accuracy and efficiency, supporting scalable histopathology analysis and contributing to the Vasculature Common Coordinate Framework (VCCF) and Human Reference Atlas (HRA). Full article

Aging port infrastructure demands frequent and reliable inspections, yet the existing automated systems often require expensive industrial drones, posing significant adoption barriers for local governments with limited resources. To address this challenge, this study develops a low-cost, automated crack detection system for port quay walls utilizing orthophotos generated from a small general-purpose drone. The system employs the YOLOR (You Only Learn One Representation) object detection algorithm, enhanced by two novel image processing techniques—overlapping tiling and pseudo-altitude slicing—to overcome the resolution limitations of low-cost cameras. While official guidelines for port facilities designate 3 mm as an inspection threshold, our system is specifically designed to achieve a higher-resolution detection capability for cracks as narrow as 1 mm. This approach ensures reliable detection with a sufficient safety margin and enables the proactive monitoring of crack progression for preventive maintenance. The effectiveness of the proposed image processing techniques was validated, with an F₁ score-based analysis revealing key trade-offs between maximizing detection recall and achieving a balanced performance depending on the chosen simulated altitude. Furthermore, evaluation using real-world inspection data demonstrated that the proposed system achieves a detection performance comparable to that of a well-established commercial system, confirming its practical applicability. Crucially, by mapping the detected cracks to real-world coordinates on georeferenced orthophotos, the system provides a foundation for advanced, data-driven asset management, allowing for the quantitative tracking of deterioration over time. These results confirm that the proposed workflow is a practical and sustainable solution for infrastructure monitoring. Full article

Convolution plays a significant role in many scientific and technological computations, such as artificial intelligence and signal processing. Convolutional computations consist of many dot-product operations (multiplication–accumulation, or MAC), for which the Winograd algorithm is currently the most widely used method to reduce the number of MACs. The Karatsuba algorithm, since its introduction in the 1960s, has been traditionally used as a fast arithmetic method to perform multiplication between large-bit-width operands. It had not been exploited to accelerate 2D convolution computations before. In this paper, we revisited the Karatsuba algorithm and exploited it to reduce the number of MACs in 2D convolutions. The matrices are first segmented into tiles in a divide-and-conquer method, and the resulting submatrices are overlapped to construct the final output matrix. Our analysis and benchmarks have shown that for convolution operations of the same dimensions, the Karatsuba algorithm requires the same number of multiplications but fewer additions as compared with the Winograd algorithm. A pseudocode implementation is also provided to demonstrate the complexity reduction in Karatsuba-based convolution. FPGA implementation of Karatsuba-based convolution also achieves 33.6% LUTs (Look -up Tables) reduction compared with Winograd-based implementation. Full article

Stimulated Raman Scattering (SRS) nanoscopy imaging offers unprecedented insights into tissue molecular architecture but often requires stitching multiple high-resolution tiles to capture large fields of view. This process is time-consuming and frequently introduces vignetting artifacts—grid-like intensity fluctuations that degrade image quality and hinder downstream quantitative analyses and processing such as super-resolution deconvolution. We present VIgnetted Stitched-Tile Adjustment using Morphological Adaptive Processing (VISTAmap), a simple tool that corrects these shading artifacts directly on the final stitched image. VISTAmap automatically detects the tile grid configuration by analyzing intensity frequency variations and then applies sequential morphological operations to homogenize the image. In contrast to conventional approaches that require increased tile overlap or pre-acquisition background sampling, VISTAmap offers a pragmatic, post-processing solution without the need for separate individual tile images. This work addresses pressing concerns by delivering a robust, efficient strategy for enhancing mosaic image uniformity in modern nanoscopy, where the smallest details make tremendous impacts. Full article

Per- and polyfluoroalkyl substances (PFAS) are a group of environmental contaminants associated with various health risks; however, their relationship with all-cause mortality in individuals with diabetes remains unclear. A total of 1256 participants from the National Health and Nutrition Examination Survey (NHANES) were included to explore the association between seven PFAS compounds and all-cause mortality in diabetic patients. Preliminary logistic regression identified three PFAS compounds (perfluorooctanoic acid [PFOA], perfluorooctane sulfonic acid [PFOS], and 2-(N-methyl-PFOSA) acetate acid [MPAH]) as significantly associated with mortality in the diabetic population. The optimal cut-off values for PFOS, PFOA, and MPAH were determined using the X-tile algorithm, and participants were categorized into high- and low-exposure groups. Kaplan–Meier survival curves and multivariable Cox proportional hazards regression models were used to assess the relationship between PFAS levels and mortality risk. The results showed that high levels of PFOS were significantly associated with increased all-cause mortality risk in diabetic patients (hazard ratio [HR]: 1.55, 95% confidence interval [CI]: 1.06–2.29), while PFOA and MPAH showed no significant associations. To explore mechanisms underlying the PFOS–mortality link, toxicogenomic analysis identified 95 overlapping genes associated with PFOS exposure and diabetes-related mortality using the Comparative Toxicogenomics Database (CTD) and GeneCards. Functional enrichment analysis revealed key biological processes, such as glucose homeostasis and response to peptide hormone, with pathways including the longevity regulating pathway, apoptosis, and p53 signaling pathway. Protein–protein interaction network analysis identified 10 hub genes, and PFOS was found to upregulate or downregulate their mRNA expression, protein activity, or protein expression, with notable effects on mRNA levels. These findings suggest that PFOS exposure contributes to increased mortality risk in diabetic patients through pathways related to glucose metabolism, apoptosis, and cellular signaling. Our study provides new insights into the association between PFAS and all-cause mortality in diabetes, highlighting the need for large-scale cohort studies and further in vivo and in vitro experiments to validate these findings. Full article

This paper analyzes the advantages and the drawbacks of using the Synthetic Aperture Radar (SAR) azimuth multichannel technique known as Multi-Aperture Processing Scheme (MAPS), in a set of relevant application cases that are far from the canonical ones. In the scientific literature on this topic, equally distributed azimuth channels with the quasi-monostatic deployment are assumed. With this research, we aim at extending the models from the current literature to (i) a generic bistatic acquisition geometry, (ii) a set of cases where the number of receiving tiles is not the same for each channel, or (iii) the tiles are shared between adjacent channels thus creating an overlapping configuration. The paper introduces the mathematical models for the listed non-conventional MAPS cases. Dealing with the bistatic MAPS, we first solve the problem by interpreting multichannel acquisition as a bank of Linear Time Invariant (LTI) filters. Then, a more physical approach, based on discrimination of the direction of arrivals (DoAs) is pursued. The effectiveness of the two methods and the advantages of the second approach on the first are proved by using a simplified 1D end-to-end simulation. Even limiting to the monostatic configuration, the azimuth antenna tiles have always been supposed equally partitioned among the RX channels. Overcoming this limit has two advantages: (i) more MAPS possible solutions in case few azimuth tiles are available, as in the ROSE-L mission; (ii) the number of channels can be designed independently of the number of tiles, also allowing asymmetric solutions, useful for a phase array antenna with an odd number of tiles such as in the SAOCOM-1 mission. Conversely, sharing one or more receiving tiles in different receiving channels makes the input noise partially correlated. The drawback is an increase in the noise level. A trade-off is determined for the different solutions obtained using simulations with real mission parameters. The theoretical performance and the end-to-end simulations are compared. Full article

Studies addressing the supervised extraction of geospatial elements from aerial imagery with semantic segmentation operations (including road surface areas) commonly feature tile sizes varying from 256 × 256 pixels to 1024 × 1024 pixels with no overlap. Relevant geo-computing works in the field often comment on prediction errors that could be attributed to the effect of tile size (number of pixels or the amount of information in the processed image) or to the overlap levels between adjacent image tiles (caused by the absence of continuity information near the borders). This study provides further insights into the impact of tile overlaps and tile sizes on the performance of deep learning (DL) models trained for road extraction. In this work, three semantic segmentation architectures were trained on data from the SROADEX dataset (orthoimages and their binary road masks) that contains approximately 700 million pixels of the positive “Road” class for the road surface area extraction task. First, a statistical analysis is conducted on the performance metrics achieved on unseen testing data featuring around 18 million pixels of the positive class. The goal of this analysis was to study the difference in mean performance and the main and interaction effects of the fixed factors on the dependent variables. The statistical tests proved that the impact on performance was significant for the main effects and for the two-way interaction between tile size and tile overlap and between tile size and DL architecture, at a level of significance of 0.05. We provide further insights and trends in the predictions of the extensive qualitative analysis carried out with the predictions of the best models at each tile size. The results indicate that training the DL models on larger tile sizes with a small percentage of overlap delivers better road representations and that testing different combinations of model and tile sizes can help achieve a better extraction performance. Full article