Search Results (9,524)

High-resolution remote sensing object detection is a fundamental task in Earth observation. However, the storage, transmission, and downstream processing of high-resolution images impose substantial bandwidth, memory, and computational burdens. Compressed sensing (CS) samples and compresses the signals simultaneously, thereby reducing data transmission and storage overhead. Unfortunately, most existing CS-based pipelines require explicit image reconstruction before downstream inference, leading to heavy computational overheads and poor scalability for high-resolution remote sensing images (RSIs). This work focuses on post-acquisition image compression and explores algorithmically, rather than from a physical hardware implementation perspective, whether explicit image reconstruction is an indispensable intermediate step prior to object detection. To this end, we propose HR²SIOD-CL, an end-to-end compressed learning (CL) framework that performs object detection directly on CS measurements of high-resolution RSIs without explicit image reconstruction. HR²SIOD-CL integrates an entropy-driven content-aware adaptive sampling strategy and a measurement-domain detection backbone for multi-scale feature extraction. Although jointly optimized during training, the adaptive sampling and detection modules can be decoupled for flexible deployment. For fair comparison, an extra lightweight reconstruction network equipped with a single-step data-consistency correction is constructed as the baseline. Extensive experiments on the NWPU VHR-10 and DIOR datasets show that across various sampling ratios, HR²SIOD-CL surpasses the reconstruction-based detection method when integrated into two-stage detectors, and achieves comparable or superior detection performance to the reconstruction-based counterparts when integrated into single-stage detectors. Meanwhile, its computational overhead and GPU memory consumption are merely 6.97% and 35.74% of those of the reconstruction-based counterpart, respectively. These results indicate that CS measurements can function as an effective intermediate representation for object detection, and explicit image reconstruction is not a prerequisite when detection is the primary objective. Full article

Efficient parking space management in urban settings represents a growing challenge owing to the sustained increase in the vehicle fleet. This study presents a comparative evaluation of five object detection architectures —YOLOv8s, YOLOv11s, YOLOv12s, RT-DETR-L and Faster R-CNN—applied to real-time intelligent vehicle occupancy monitoring under variable lighting conditions. The models were trained via transfer learning on a custom dataset of 1463 source images (21,944 annotated instances; expanded to 3511 files and 52,664 instances through offline augmentation of the training subset; three classes: free, occupied and unavailable) captured on a university campus located in Juliaca (Puno region), Peru, at 3824 m a.s.l. under daytime and nighttime clear-sky conditions from a single fixed-camera viewpoint. Each architecture was evaluated in ten independent experiments. Six dataset partitioning schemes of increasing strictness—a random control (R0) plus five leakage-controlled partitions—were evaluated. Under the strictest scheme (D3), simultaneously disjoint in acquisition date and camera viewpoint and therefore the most rigorous generalization estimate obtained in this study, accuracy ranges from mAP@0.5:0.95 of 0.9325 for Faster R-CNN to 0.8763 for YOLOv11s. Under the random partitioning conventionally applied to fixed-camera datasets, the same five architectures fell within 0.0055 of one another, all above 0.985, and their ranking was essentially inverted (Spearman $\rho =-0.80$). The differences in computational efficiency across architectures were statistically significant ($H=47.06$, $p<0.001$). YOLOv8s was the fastest of the four non-dominated architectures under the disjoint partition and was selected in 73.3% of weightings, although it ranked fourth in accuracy; its recommendation therefore rests on computational efficiency under a real-time constraint, whereas deployments that prioritize accuracy are better served by Faster R-CNN. The integrated system YOLOv8s + ByteTrack + FastAPI + Next.js 14 achieved per-slot accuracies of 87.5% and 91.8% under daytime and nighttime clear-sky conditions, respectively, using 1395 observations collected in a single university parking lot. For YOLOv8s, the transition from random to disjoint partitioning costs 0.1085 in mAP@0.5:0.95 (0.9913 to 0.8828), indicating that the near-saturated performance obtained under random partitioning substantially reflects the memorization of a fixed spatial configuration rather than generalization. The results support the feasibility of single-stage CNN architectures for intelligent parking monitoring in high-altitude Andean university environments under the evaluated acquisition conditions. Full article

This study considers multi-class visual detection for processing tomato sorting under controlled laboratory conditions. In densely arranged images, occlusion and visual similarity can weaken the local boundary and texture cues needed to distinguish ripe tomatoes, unripe tomatoes, defective tomatoes, and soil clods. We propose SG-RTDETR, an RT-DETRv2 adaptation that combines detail-preserving downsampling, saliency-guided token encoding with residual spatial refill, context-aware feature organization, and adaptive cross-scale fusion. A four-class dataset was constructed from 1000 multi-object images and 400 single-object images used for supplementary representation learning. Across three independent training runs under controlled laboratory evaluation, SG-RTDETR achieved $87.8\pm 0.4$% mAP${}_{50:95}$, $92.4\pm 0.4$% mAP${}_{50}$, and $95.5\pm 0.3$% mAR${}_{50:95}$ (mean ± sample standard deviation). Relative to RT-DETRv2, the mean mAP${}_{50:95}$ increased by 3.4 percentage points, while total FLOPs remained at 61.17 G and forward-pass inference throughput decreased from 110.5 to 103.6 FPS. These results indicate an accuracy-oriented trade-off under the laboratory evaluation protocol; validation under realistic postharvest sorting conditions remains necessary. Full article

Manual fruit thinning is labor-intensive and inefficient, making the development of intelligent visual detection systems a crucial approach for improving the quality and production efficiency of the pear industry. However, in natural orchard environments, young pear fruits are small in size with slender fruit stalks, and their texture and color characteristics are highly similar to those of tender branches. Furthermore, variations in illumination and occlusions caused by branches and leaves make it difficult for existing detection models to simultaneously and accurately identify fruits and fruit stalks, limiting their application in automated thinning equipment. In this study, Yuluxiang pear was selected as the research object, and image data were collected under diverse field conditions, including forward-lighting, backlighting, close-range shooting, long-range shooting and fruit overlapping. A dedicated dataset containing 3057 images was established. Based on the RT-DETR-R18 network, a lightweight and high-precision fruit–stalk synchronous detection model was proposed. Specifically, the backbone network was reconstructed by integrating GCConv with C2f modules to enhance global feature extraction for slender fruit stalks. The bottleneck structure was optimized using GCConvC3 to reduce feature degradation under occlusion conditions, and an additional 4× down-sampling P2 detection head was introduced to improve the detection capability for small targets. To fully validate the model performance and stability, three types of experiments were conducted in this study: ablation experiments, repeated experiments with different random seeds, and comparative experiments. Ablation experiments verified the cumulative performance improvements brought by the introduced modules. Repeated experiments with different random seeds were performed to explore training randomness-induced performance fluctuations, and the results demonstrated that the proposed model maintains stable overall detection accuracy with minor metric fluctuations. Comparative experiments demonstrated that the proposed model achieved a compact parameter size of only 15.97 M, with a precision of 95.0% for young pear fruit detection and an mAP50 of 83.0% for fruit stalk detection, outperforming all comparative models in overall mAP50. The training convergence curves and Grad-CAM++ visualization results further confirmed the stable optimization process and enhanced feature attention capability of the proposed model. By achieving a favorable balance between detection accuracy and model lightweightness, this approach provides effective technical support for the development of intelligent fruit thinning equipment and vision-based systems for smart pear orchards. Full article

Oriented object detection (OOD) in remote sensing images (RSIs) suffers from insufficient feature representation caused by arbitrary rotation angles and small spatial resolutions. Existing spatial–frequency fusion paradigms merely implement single-granularity feature interaction, either global image-level frequency compensation or local instance-level feature refinement, and fail to simultaneously capture global scene semantic consistency and local object fine-grained discriminability. In this paper, we propose a unified dual-level spatial–frequency collaborative detector (DSCDet) for remote sensing OOD tasks. Different from previous decoupled designs, the proposed DSCDet constructs a complete spatial–frequency collaborative fusion paradigm that shares a generic wavelet-based frequency extraction mechanism and cross-feature fusion module, which is adaptively deployed at both image-level and instance-level granularities. Specifically, our method introduces Haar wavelet transform to extract multi-scale frequency mutation features. On this basis, a generic cross-domain attention fusion (GCDAF) is constructed with granularity-dependent positional encoding constraints. The core difference between dual granularity fusion lies in geometric positional encoding, where image-level fusion adopts global scene positional embedding to maintain overall semantic stability, and instance-level fusion leverages local pairwise instance positional embedding to optimize fine-grained target feature interaction. The unified dual-level fusion architecture comprehensively integrates global semantic integrity and local target specificity, forming a robust and universal spatial–frequency feature representation system. Extensive experiments on three public remote sensing datasets, including DOTA-v1.0, DOTA-v1.5 and DIOR-R, demonstrate that the proposed DSCDet achieves competitive and superior performance against state-of-the-art OOD detectors. Full article

UAV-based insulator detection is challenging because insulators often occupy small regions of aerial images and appear against complex backgrounds, while subtle local features may be lost during feature extraction and down-sampling. Moreover, practical UAV and edge-device applications require efficient models with limited computational and memory demands. This study proposes an optimized lightweight YOLOv11n model for high-voltage transmission-line insulator detection. The architecture integrates C3k2MBNV2 to reduce model complexity, SCDown to preserve spatial information during down-sampling, and C3k2WTDC to enhance multi-frequency feature representation. A diverse dataset containing 5750 insulator images acquired under different environmental conditions, viewing angles, and backgrounds was used for evaluation. Experimental results show that the proposed model reduces the parameter count from 6.20 M to 3.26 M and computational complexity from 20.5 to 12.7 GFLOPs, corresponding to reductions of 47.4% and 38.0%, respectively. Meanwhile, precision increases from 91.3% to 93.8%, recall from 73.4% to 75.2%, mAP50 from 71.2% to 73.9%, and mAP50–95 from 65.6% to 67.3%. These results demonstrate an improved accuracy–efficiency trade-off, supporting real-time insulator detection in resource-constrained UAV and edge-device applications. Full article

Background/Objectives: Computed tomography perfusion is widely used in acute ischemic stroke, but its performance is less reliable for lacunar and infratentorial infarcts. This technical review examines the acquisition and processing factors that influence their detectability. Methods: A targeted technical review of PubMed/MEDLINE, Scopus, and Web of Science was performed, focusing on acquisition, reconstruction, vascular input selection, deconvolution, filtering, spatial sampling, automated classification, posterior circulation stroke, and lacunar infarction. Results: Detectability depends on lesion size and contrast, posterior fossa artifacts, spatial and temporal resolution, vascular curve quality, mathematical stabilization, and automated thresholds. Larger cerebellar infarcts may remain visible, whereas brainstem, deep cerebellar, and perforator lesions are more vulnerable. Temporal maps and direct review of parametric images may reveal abnormalities absent from automated summaries. Conclusions: This technical review shows that computed tomography perfusion should be interpreted as a derived estimate rather than a direct representation of cerebral hemodynamics. Negative automated findings do not exclude lacunar or infratentorial infarction when clinical suspicion remains high. Full article

Objective: To address small bolt defect targets, complex background interference, and limited edge deployment in hilly and mountainous rail transport environments, a detection method balancing accuracy, lightweight design, and real-time performance was proposed. Methods: A track image dataset containing missing bolts, loose bolts, and missing nuts was constructed. Based on YOLOv11m, HFERBC3K2 was developed by replacing the standard bottleneck in C3K2 with a High-Frequency Enhancement Residual Block to strengthen edge, texture, and local structural feature extraction. A Spectral Enhanced Feed-Forward module was introduced into C2PSA to form SEFFNC2PSA, enhancing defect-related frequency components and suppressing background interference through adaptive frequency-domain modulation. The integrated model was named HFSE-YOLOv11. Channel-level structured pruning was then applied, and the model with a pruning ratio of 0.5 was named LHFSE-YOLOv11. Results: On the validation set, HFSE-YOLOv11 achieved 91.7% precision, 93.4% recall, 91.3% mAP@0.5, and 80.2% mAP@0.5:0.95, improving upon YOLOv11m by 3.6, 2.2, 1.2, and 3.1 percentage points, respectively. After pruning, LHFSE-YOLOv11 had 15.9 M parameters, 53.8 GFLOPs, and a 32.5 MB model size, representing reductions of 16.3%, 14.3%, and 11.7%, while mAP@0.5 and mAP@0.5:0.95 decreased by only 0.3 and 0.9 percentage points. On the independent test set, it achieved 91.7% precision, 93.4% recall, 91.0% mAP@0.5, 79.3% mAP@0.5:0.95, and 81.5 FPS, outperforming all compared models in the four detection metrics. Conclusion: LHFSE-YOLOv11 balances accuracy, efficiency, and model size, supporting deployment on vehicle-mounted inspection terminals and resource-constrained edge devices. Full article

In protected agriculture such as greenhouses, pest and disease monitoring via UAVs and fixed cameras suffers from extremely small object proportions owing to shooting altitude constraints, posing considerable detection challenges. Moreover, fine-grained annotation of numerous small-scale images incurs prohibitive costs. Targeting this bottleneck, this paper proposes a cross-scale unified semantic space learning framework and introduces an end-to-end DS-DETR detector based on DETR. Unlike existing methods relying on domain adaptation, multi-scale fusion, or super-resolution reconstruction, this work explicitly models instance-level cross-scale semantic correlation, transferring fine-grained semantics from large-scale close-up images to small-scale scene feature space. A Single-Point Dual-Shooting (SPDS) strategy is adopted to collect high-fidelity paired images via ordinary smartphones at low cost. A dual-stream encoder with cross-view attention and an instance-level contrastive loss align features of identical instances in a unified semantic space. A self-built CropScale-Det dataset covering three crop diseases is constructed in greenhouse scenarios. Experimental results show that DS-DETR achieves 42.5 ± 1.2% mAP@50 under limited annotations, outperforming YOLOv8-n by 11.2%, with small-target average precision reaching 26.8 ± 1.1%. Ablation experiments and feature visualization validate the effectiveness of the designed mechanism. This approach considerably reduces reliance on large-scale densely annotated data, establishing a data-efficient proof-of-concept for small-scale pest detection in protected agriculture. Full article

This paper focuses on the binary task of parking-slot occupancy detection (vacant vs. occupied) from UAV aerial imagery. Accurate parking-slot detection from UAV imagery is challenged by small target sizes, highly regular rectangular shapes, large-scale variations, complex backgrounds, and perspective distortions. Compared with fixed surveillance cameras, UAV-based detection offers flexible deployment, wide-area coverage, and no requirement for pre-installed infrastructure, making it especially suitable for large open-air parking lots and temporary parking scenarios. To address this, this study proposes a task-specific framework for UAV-based parking-slot detection, with improvements in backbone design, attention modeling, and bounding-box regression. A lightweight MnasNet-inspired backbone is used to improve multi-scale feature extraction at low computational cost. An enhanced EMA module with adaptive grouping, FFT-based frequency enhancement, and gated fusion is introduced to better model the structured patterns of parking lot scenes. In addition, a UIoU+ loss tailored to rectangular geometry is proposed to improve localization quality. Sensitivity analysis and repeated experiments show that the method is stable and statistically reliable. All main metrics are evaluated on an independent held-out test set to ensure generalization. Extensive experiments demonstrate that each component brings consistent performance gains. The proposed model achieves 99.44 ± 0.12% mAP@0.5, 90.31 ± 0.27% mAP@0.5:0.95, 99.27 ± 0.15% precision, and 99.00 ± 0.18% recall on the self-built UAV Parking Lot dataset. Its mAP@0.5:0.95 is 26.01 percentage points higher than the YOLOv11n baseline. Consistent performance improvements are also validated on two additional public benchmarks (CARPK and PKLot), confirming the generalization of the proposed method beyond the self-built dataset. Most importantly, the method supports real-time inference on embedded UAV platforms and achieves state-of-the-art performance among lightweight detectors, making it an ideal solution for practical intelligent parking management. Ablation studies further confirm the complementary synergy between the proposed backbone, attention module, and loss function. Full implementation code, pre-trained weights, and detailed reproduction guidelines are publicly available to ensure research reproducibility. Full article

Background: Early pressure injury detection is essential in acute care, yet diagnostic inequities persist for individuals with dark skin tones. Traditional visual assessment relies on colour changes that may not be visible in richly pigmented skin, contributing to delayed diagnosis and poorer outcomes. Objective: Our aim was to identify and map strategies reported in the literature for assessing skin and underlying tissue to support early pressure injury detection in adults with dark skin tones in acute care settings. Methods: A scoping review was conducted following Joanna Briggs Institute methodology and reported according to PRISMA-ScR. Searches were performed in CINAHL, PubMed, Scopus, Web of Science, and Google Scholar (April 2025; updated February 2026). Primary research published in English since 2017 was eligible. After screening and data extraction, descriptive thematic analysis was applied to identify recurrent strategies reported in the literature for assessment of skin and underlying tissues for detection of pressure injury. No appraisal of the included sources was conducted. Results: Ten studies met the inclusion criteria. Strategies identified for detecting pressure injuries were mainly for implementation across diverse skin tones, but all included aspects of dark-skin-tone participants. These approaches are clustered into four categories: thermographic imaging, subepidermal moisture sensors, image processing and colorimetry, and enhanced assessment and lighting techniques. The evidence highlights the growing value of objective, technology-supported methods for improving detection in people with dark skin tones. Routine bedside visual and tactile assessment was notably underrepresented, indicating a clear evidence gap. Overall, the review emphasises the need to integrate objective technologies into everyday practice to support more equitable pressure injury detection. Conclusions: This scoping review highlighted that evidence for routine visual and tactile assessment in dark skin tones remains limited. Technology-supported assessment approaches show promise for improving diagnostic equity in pressure injury detection. Identification of strategies used to perform skin assessment for pressure injury identification is critical to the wellbeing of persons with dark skin tones. This review demonstrates that while emerging technologies offer promising ways to improve pressure injury detection across diverse skin tones, evidence supporting everyday bedside visual and tactile assessment for individuals with dark skin tones remains notably limited. Strengthening this evidence base and integrating objective technologies into routine care are essential steps toward more accurate and equitable pressure injury detection. Full article

Background/Objectives: Contralateral prophylactic mastectomy (CPM) is increasingly performed despite negative preoperative imaging. We evaluated the prevalence and MRI detectability of occult pathology in the contralateral breast. Methods: In this single-center retrospective cohort, 82 patients with unilateral invasive breast cancer underwent simultaneous CPM after preoperative contralateral MRI. Occult findings were classified as occult malignancy, atypical/high-risk lesions, or other lesions of uncertain malignant potential (B3 lesions); the primary outcome was clinically significant occult pathology (malignancy or an atypical/high-risk lesion). Proportions are reported with exact 95% confidence intervals and associations with exact odds ratios and Benjamini–Hochberg correction. Results: Occult malignancy occurred in 1/82 (1.2%; a single DCIS), clinically significant occult pathology in 14/82 (17.1%) and any occult pathology in 21/82 (25.6%). Among 59 patients with a negative MRI (BI-RADS 1–2), clinically significant occult pathology occurred in 18.6% and atypical/high-risk lesions in 16.9% (whole cohort, 15.9%). The single occult malignancy arose in an MRI-negative breast. No factor remained significant after correction for multiple comparisons. Conclusions: After preoperative MRI, occult malignancy is rare, whereas atypical and high-risk lesions frequently remain occult; a negative contralateral MRI excluded neither. These findings support individualized, shared decision-making rather than the yield of occult pathology as the basis for CPM. Full article

Background/Objectives: The demand for reducing iodinated contrast agent (ICA) dose in cerebral CT angiography (CTA) for aneurysm detection has increased due to concerns over patient safety and image quality. The aim of this study is to systematically review the diagnostic accuracy, image quality, and dose efficiency of optimized ICA protocols in CTA for intracranial aneurysm detection. We also aimed to distinguish between direct diagnostic studies reporting aneurysm sensitivity/specificity and indirect technical studies, emphasizing patient-specific tailoring using cardiac output and lean body weight. Methods: Following PRISMA 2020 guidelines, PubMed, Scopus, and Web of Science were searched up to 2 June 2026. Two reviewers independently screened studies, extracted data, and assessed risk of bias using QUADAS-2 and the NIH tool. Due to clinical and technical heterogeneity, a narrative synthesis was performed. Results: Eleven studies comprising 716 patients were included. Low tube voltage (70–80 kVp) combined with reduced iodine load (8–30 mL, 3.2–9 gI) maintained diagnostic accuracy for aneurysms >3 mm compared to standard protocols. Iterative reconstruction and deep learning improved contrast-to-noise ratio by 23–45% at reduced dose. Technical studies showed feasibility of ultra-low-volume protocols but lacked aneurysm-specific DSA validation. Conclusions: Reduced-contrast and low-kVp CTA protocols may preserve diagnostic performance for intracranial aneurysm detection while lowering iodine exposure. However, current evidence is limited by small sample sizes and heterogeneous protocols. Prospective, aneurysm-specific validation is required before clinical implementation. Full article

Backgroundand Objectives: Retinal microcirculation may serve as a surrogate marker of systemic vascular status and can be non-invasively assessed using optical coherence tomography angiography (OCTA). Emerging evidence suggests that retinal and choroidal microvascular alterations may occur following coronavirus disease 2019 (COVID-19) and multisystem inflammatory syndrome in children (MIS-C), potentially reflecting systemic vascular and inflammatory processes. This review summarizes the current evidence regarding retinal and choroidal microvascular changes detected by OCTA in pediatric patients with COVID-19 and MIS-C. Materials and Methods: A structured literature search of the National Center for Biotechnology Information (NCBI) PubMed database was conducted from inception until April 2026. Original English-language studies evaluating retinal and choroidal microcirculation using OCTA in pediatric patients with COVID-19 or MIS-C were identified and reviewed. Results: Available studies suggest variable retinal and choroidal microvascular alterations following COVID-19 and MIS-C, including changes in vessel density, perfusion, and foveal avascular zone parameters. Several investigations reported reduced vessel density, particularly within the deep capillary plexus, whereas others demonstrated increased peripapillary vascular parameters or no significant differences during the observation period. Differences in patient characteristics, disease severity, timing of imaging, OCTA protocols, and study design likely contribute to the heterogeneity of the reported findings. Conclusions: OCTA represents a promising research tool for investigating retinal and choroidal microvascular alterations associated with pediatric COVID-19 and MIS-C. However, the available evidence remains limited by small observational studies, methodological heterogeneity and inconsistent findings. Larger prospective longitudinal studies using standardized imaging protocols are needed to determine the clinical significance and reproducibility of OCTA-derived parameters and to establish whether they may have potential as biomarkers of ocular or systemic vascular involvement. Full article

Osteonecrosis of the femoral head (ONFH) is a progressive and potentially disabling condition caused by compromised blood supply to the femoral head, leading to bone necrosis and collapse. Early diagnosis is essential to enable joint-preserving interventions and improve patient outcomes. Magnetic resonance imaging (MRI) is currently considered the most sensitive modality for early detection, whereas computed tomography (CT) provides superior assessment of subchondral bone integrity and structural collapse. In recent years, artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL) techniques, has emerged as a promising tool to enhance diagnostic accuracy, automate lesion segmentation, and predict disease progression. This review aims to provide an overview of current AI applications in MRI for ONFH, focusing on early diagnostic, disease staging and classification, volumetric assessment, differential diagnosis and prognostic prediction. A total of 61 articles were initially identified, of which 13 studies (2021–2025) met the inclusion criteria. Results indicate that DL models, particularly convolutional neural networks (CNNs), achieve excellent diagnostic performance, with reported accuracies up to 98.4% and area under the curve (AUC) values reaching 0.98 for early-stage detection. Several models demonstrated performance comparable to or exceeding that of experienced clinicians, particularly in differentiating ONFH from other hip pathologies and in early disease recognition. AI algorithms also showed high accuracy in staging and classification (AUC up to 99.7% in internal validation), as well as in automated segmentation and volumetric assessment (Dice coefficients up to 0.89), enabling objective quantification of necrotic lesions. Furthermore, prognostic models integrating radiomics and ML techniques demonstrated promising results in predicting femoral head collapse (AUC up to 0.85). From a clinical perspective, AI appears to function primarily as a supportive tool, improving diagnostic consistency, efficiency, and reproducibility, and acting as a “second reader” capable of reducing variability among less experienced clinicians. However, significant limitations remain, including dataset heterogeneity, predominance of retrospective and monocentric studies, and limited integration of clinical data. In conclusion, AI-based MRI analysis shows strong potential to enhance the diagnosis, staging, and management of ONFH. Future research should focus on multicenter prospective validation, integration of multimodal clinical data, and development of explainable and generalizable models to facilitate widespread clinical adoption. Full article