Search Results (2,444)

Multimodal remote sensing segmentation commonly integrates optical imagery with Digital Surface Models (DSMs) to improve land-cover understanding. However, existing methods often overlook a critical issue, namely the inconsistency between spatial-domain structural details and frequency-domain semantics, which can lead to misaligned features and degraded performance in complex scenes. To address this problem, we formulate spatial–frequency representation conflict as a unified multimodal learning problem and propose the Conflict-Aware Fusion Network (CAF-Net). Specifically, Cross-Modal Structure Guidance (CMSG) extracts DSM-derived high-pass structural cues and conditionally modulates optical features to improve boundary consistency. The Adaptive Cross-Frequency Module (ACFM) separates DCT coefficients using a fixed radial mask, adaptively reweights low- and high-frequency components, and performs cross-modal alignment at the highest encoder stage. Uncertainty-Aware Fusion (UAF) predicts pixel-wise relative reliability scores and normalizes them across modalities to suppress low-confidence responses. This coordinated design links spatial refinement, frequency alignment, and reliability-guided fusion instead of treating them as independent feature-enhancement operations. Experiments on the ISPRS Vaihingen and Potsdam datasets yield mIoU scores of 84.35% and 86.86%, respectively. The results indicate that explicitly modeling spatial–frequency discrepancies can improve multimodal segmentation accuracy and representation consistency. Full article

The accuracy of optical tracking tools is crucial for surgical navigation. While commercial tools are reliable, their proprietary design knowledge limits accessibility and adaptability for specialized clinical and research applications. This study introduces an open, reproducible optimization framework based on point-based rigid registration theory, defining a unified pose estimation deviation metric and deriving its analytical expression for both expectation and variance. The approach incorporates constraints for intra-group uniqueness and inter-group compatibility, using exhaustive configuration generation and geometric evaluation to rank designs by predicted accuracy. Numerical simulations confirmed the derived formula, with under 5% average prediction error for the expectation and strong agreement for the variance. Optimized four-fiducial tools were compared to commercial references via tip calibration, distance measurement, and registration tests. Most optimized tools (75%) achieved accuracy comparable to or modestly better than commercial tools (e.g., tip calibration 0.22 mm vs. 0.28 mm, distance measurement 0.18 mm vs. 0.20 mm, FRE 0.13 mm vs. 0.16 mm, TRE 0.48 mm vs. 0.51 mm). All four experiments showed a strong positive correlation between the theoretical metric and measured error (Pearson’s $r>0.98$, all $p<2.2\times {10}^{-7}$; exact Spearman $p\le 2.8\times {10}^{-6}$). Beyond these numerical results, the primary contribution is a systematic, open design methodology that formalizes knowledge historically proprietary to commercial vendors, enabling researchers and engineers to generate high-precision custom tracking tools for diverse surgical navigation scenarios. Full article

Accurate maize distribution information is critical for crop-area statistics, food-security assessment, and agricultural monitoring, but large-scale maize-mapping remains difficult in regions with limited reference samples, heterogeneous crop calendars, and frequent optical data gaps. This study proposes a phenology-adaptive maize mapping framework based on Sentinel-2 time-series imagery and an Enhanced Red-edge NDVI (ENDVIre). ENDVIre was constructed from the Sentinel-2 red-edge 4 and red-edge 2 bands to enhance the spectral response of maize during the silking-to-grain-filling stage, when maize develops a dense canopy and high chlorophyll content but is often confused with soybean. The framework first reconstructed the NDVI time series using an upper-envelope-constrained Whittaker smoother to identify key phenological stages, including sowing–emergence, vigorous growth, and maturity–harvest. NDVI, ENDVIre, and LSWI were then integrated into an interpretable decision-tree model with phenology-aligned time windows to distinguish maize from soybean, rice, wheat, and other non-maize backgrounds. The method was evaluated in six representative maize-growing regions across the United States, Brazil, China, Kenya, and Ukraine, covering different crop calendars, field sizes, and agricultural systems. The mean overall accuracy, F1-score, and Kappa coefficient across the six regions reached 93.27%, 93.14%, and 0.8652, respectively. Cross-year experiments in a winter-wheat–summer-maize rotation region from 2020 to 2024 achieved overall accuracies of 89.80–96.80%, while spatial-transfer experiments in six independent regions achieved overall accuracies of 87.40–95.40%. A comparison with existing high-resolution maize products in the Huang-Huai-Hai Plain further showed that the proposed method better balanced omission and commission errors. These results indicate that ENDVIre-based phenology rules provide an interpretable and transferable solution for maize mapping under limited-sample conditions, although persistent cloud contamination and fragmented smallholder landscapes remain important challenges. Full article

This study proposes TPA-ConvNeXt, a ConvNeXt-Small-based deep learning architecture for retinal image classification using a phase-based trigonometric attention mechanism. The proposed Trigonometric Phase Attention (TPA) block recalibrates feature maps through learnable phase and amplitude modulation derived from channel-wise and spatial context. In addition, a stable fine-tuning strategy combining layer-wise learning rate decay (LLRD), linear warmup, and cosine annealing is employed to adapt pretrained backbones to medical image data. The method was evaluated on both fundus and optical coherence tomography (OCT) datasets. On the HYAMD fundus dataset, 5-fold cross-validation yielded a mean Macro F1 of 0.8940 ± 0.0403 and an ROC-AUC of 0.9449. External fundus validation further demonstrated cross-dataset robustness, achieving 0.9312 Macro F1 and 0.9803 ROC-AUC when trained on HYAMD and tested on AMDNet23. In OCT experiments, the model achieved Macro F1 scores of 0.9790 on MAK1_OCT, 0.9387 on OCTDL, and 0.9989 on the cleaned OCT2017 benchmark test set after MD5 duplicate removal. Ablation results showed that warmup contributed most strongly to stable optimization, while the proposed TPA block provided a smaller but consistent performance gain. Additional statistical analysis across the five HYAMD folds showed that TPA-ConvNeXt achieved comparable performance to the ConvNeXt-Small baseline, with a small numerical accuracy difference that did not reach statistical significance. Grad-CAM visualizations indicated that the model focused on clinically relevant retinal regions, and complexity analysis showed that the full model required 51.82 M parameters and 17.41 GFLOPs, adding only 2.36 M parameters and 0.02 GFLOPs over the ConvNeXt-Small baseline. These findings suggest that TPA-ConvNeXt provides a robust and generalizable framework for retinal image classification across both fundus and OCT modalities. Full article

This article presents the design and implementation of a control system for an adaptive light luminaire with variable light distribution. The developed solution enables dynamic shaping of the light distribution characteristics by simultaneously controlling the geometry of the optical system and the spatial distribution of the emitted light flux. The system utilizes two cooperating control mechanisms. The first is implemented by four independently controlled reflectors with adjustable angles of inclination. The second is based on the independent control of eight sections of LED light sources. The coordination of both systems enables the implementation of various operating scenarios, including symmetric, asymmetric, and adaptive configurations, with variants of narrow and wide beam distribution. The central unit of the system is an ESP32 microcontroller that performs control functions, generates PWM signals, and coordinates the operation of the actuators. The system was implemented as a dedicated embedded system. The main contribution of this work is the implementation and experimental validation of an embedded control platform integrating mechanical beam shaping and segmented LED control within a single adaptive lighting system. As part of the work, predefined control scenarios for lighting system configuration were developed and experimentally tested. The developed solution increases the functionality of adaptive lighting systems and may contribute to reducing energy consumption by directing light only where required. However, the quantitative evaluation of the energy savings was beyond the scope of the present study. Full article

In optical remote sensing 3D reconstruction, high-resolution satellite stereo matching is a critical task, yet it is challenged by extreme imaging geometries, texture-less and repetitive patterns, occlusions, and scene variations caused by spatio-temporal heterogeneity. To address these issues, we propose IFMA-Stereo, an innovative binocular disparity estimation method that leverages a monocular depth foundation model. Our approach constructs a multi-scale spatial information pyramid to jointly integrate the foundation model with a disparity extraction network. At the feature level, an attention interaction mechanism captures multi-dimensional contextual dependencies and transforms general scene understanding priors into long-range associative features suitable for stereo cost volume construction. At the pixel level, a cyclic iterative refinement module embeds depth information from the foundation model throughout the iteration process and performs joint optimization, enhancing the model’s adaptability in geometrically complex regions. Experiments on the US3D and GaoFen-7 datasets demonstrate that IFMA-Stereo achieves superior performance in challenging areas (texture-less regions, disparity discontinuities, repetitive patterns) and effectively mitigates prediction errors caused by spatio-temporal heterogeneity, albeit at the cost of increased inference time compared to baseline methods. Quantitatively, the method achieves an end-point error (EPE) of 1.347 and a D1 error of 7.26% on the US3D dataset, and an EPE of 1.585 and a D1 error of 13.41% on the GaoFen-7 dataset. Notably, the method also yields precise predictions for unseen urban areas, indicating strong generalization. These results confirm that IFMA-Stereo achieves state-of-the-art accuracy in remote sensing disparity estimation. Full article

Among myrmecophilous insects, beetles represent the most specialised and diverse group. Myrmecophily is a complex evolutionary strategy encompassing a wide spectrum of interactions with ants, ranging from occasional to obligate relationships, and from mutualistic associations (e.g., trophobionts) to fully parasitic symbioses (social parasites). One of the most remarkable examples of an obligate ant parasite is Paussus favieri Fairmaire,1851 (Carabidae, Paussinae, Paussini), a West-Mediterranean ant-nest beetle. This species spends most of its life inside the nests of Pheidole pallidula (Nylander, 1849) (Hymenoptera, Formicidae), where it exploits the colony’s most valuable resources (ant larvae, pupae, and tenerals) through a suite of sophisticated chemical and structural adaptations that allow it to evade detection and integrate seamlessly into the host colony. For these reasons, P. favieri has recently emerged as a key model organism for studying host–parasite interactions in eusocial systems. In this study, we investigated possible correlations between the nervous system of P. favieri and its remarkable morphological and behavioural adaptations, shedding light on how an extreme environment such as the ant nest may have shaped the beetle’s brain. Our results, although requiring more in-depth analysis, reveal an exceptional development of the central body and the antennal lobes, which rank among the largest recorded across all insect species studied to date. We also report two previously undescribed morphological asymmetries affecting the optic lobes and mushroom bodies. Together, these findings provide new insights into the neuroanatomy of carabid beetles and, more broadly, into the biology of a unique model of ant parasitism, advancing our understanding of the evolutionary adaptations that characterise the highly specialised Paussinae subfamily, laying down the basis for further analysis. Full article

Three-dimensional surface reconstruction is essential for accurately acquiring the external quality parameters of watermelons, such as size, volume, and defect area. Binocular stereo vision provides a low-cost and easily deployable solution for the single-view 3D reconstruction of watermelons. However, watermelons present highly similar surface textures, and as typical spheroid-like objects, the excessive angle between surface normals of edge regions and the camera optical axis leads to insufficient feature representation. Consequently, directly applying existing stereo matching algorithms often introduces matching ambiguities, and lightweight networks struggle to balance real-time performance with matching accuracy. This study focuses on the high-precision single-view point cloud generation of Kirin watermelons. To address these issues, we first construct a cross-modal, high-precision Kirin watermelon stereo matching dataset. Building upon the Fast-ACVNet+ architecture, we then propose MI-ACVNet, a lightweight stereo matching network tailored for high-precision watermelon point cloud acquisition. In the feature extraction stage, a Multi-Scale Stereo Feature Extraction (MSFE) module is adapted. By incorporating the re-parameterized network MobileOne and Epipolar-Enhanced Coordinate Attention (E2CA), MSFE improves the discriminative capability for weak and similar textures without compromising inference speed. For cost computation, a Coarse-to-Fine Cascaded Residual Correction (C2F-CRC) strategy is incorporated to construct a fine-grained cost volume via sub-pixel interpolation, enhancing the network’s ability to capture subtle surface fluctuations. Furthermore, a Semantics-Guided Region-Aware Loss (SGRA-Loss) is formulated, leveraging semantic masks to apply differentiated supervision weights across edge, center, and background regions to significantly improve edge matching accuracy. Ablation studies validate the effectiveness of the MSFE, C2F-CRC, and SGRA-Loss components. Compared to the baseline model, the full MI-ACVNet reduces the End-Point Error (EPE) by 19.5% and the Bad-0.5 error rate by 34.5% in the watermelon region. Furthermore, when compared against five mainstream algorithms (StereoNet, AANet, HSMNet, LightStereo-L, and NMRF-swint), MI-ACVNet achieves state-of-the-art performance: EPE and Bad-0.5 are reduced to 0.091 pixels and 1.159%, respectively, with a single-frame inference time of only 46 ms. The average depth error of the reconstructed point clouds is merely 0.26 mm. By ensuring both real-time efficiency and high-precision depth estimation, this method demonstrates promising potential for deployment in industrial Kirin watermelon sorting lines, driving sorting equipment toward higher precision and intelligence. Full article

Microrobots navigating the human body require biocompatible actuators capable of functioning in physiological fluids. The microalga Euglena gracilis offers precise phototactic control; however, its operational stability in simulated physiological environments remains unproven. Here, we report that a stepwise acclimation process preserves the robotic functionality of E. gracilis in 100% phosphate-buffered saline (PBS), 100% fetal bovine serum (FBS), and a NaCl solution at a concentration equivalent to PBS (137 mM). We compared direct transfer against a graduated adaptation protocol, evaluating morphology, swimming speed, motility rate, and phototaxis. Direct transfer to each medium caused near-total immobilization, whereas stepwise acclimation retained motility. Acclimated cells exhibited size reduction (miniaturization) while maintaining their characteristic eccentricity. These adapted cells sustained a negative phototactic response among the remaining motile population, supporting optical controllability despite reduced swimming speed. These results indicate that stepwise acclimation allows E. gracilis to retain partial motility and phototactic controllability under simulated physiological saline conditions, and that the observed miniaturization and preserved photo-responsiveness may be useful features for future biohybrid microrobotics. Full article

Underwater optical sensor networks are essential for fish monitoring, yet imagery is often affected by illumination variability, low contrast, and complex backgrounds. Attention mechanisms are vital for feature representation in deep networks, yet existing approaches often struggle with spatial information loss and limited multi-scale interaction under such challenging conditions. This paper introduces Convolution to Interactive Capture and Recalibration Enhancement (C2ICARE), a lightweight attention module designed to overcome these challenges. The principal contribution of C2ICARE is the adaptation of memory interaction principles into an edge-oriented attention framework that enhances feature discrimination while maintaining computational efficiency. The architecture employs three core innovations: a 1:3 memory-feature split to preserve context while reducing cost, parallel multi-scale depthwise convolutions (3 × 3 and 7 × 7) for fine-grained and broad feature extraction, and a cross-branch interaction mechanism coupled with a ConvNeXt-style feed-forward network that avoids dimensionality reduction. Experimental results on an underwater fish dataset demonstrate that YOLO26n with C2ICARE achieves a mean average precision (mAP@0.5:0.95) of 0.7033, outperforming Coordinate Attention (+3.8%), FasterBlock (+1.7%), and CBAM (+0.4%) while adding only 0.05M parameters and 0.16 GFLOPs. Multi-objective Pareto Frontier analysis confirms that C2ICARE provides an effective balance between accuracy, efficiency, and generalization for resource-constrained deployment. EigenCAM visualizations further validate that the model focuses on biological morphology rather than background noise. Its lightweight design enables seamless integration with underwater sensor networks and fog platforms for real-time fish detection in aquaculture, commercial fisheries, and scientific research. Future work will investigate broader marine applications and cross-platform deployment scenarios. The code is available on GitHub. Full article

Accurate change detection in high-resolution remote sensing imagery is essential for urban planning, land-use monitoring, and disaster response. This study introduces CTA-Net, a Cross-Temporal Attention Network for binary change detection in bi-temporal optical imagery, designed to improve robustness against pseudo-changes caused by illumination variation, seasonal effects, and sensor noise. The proposed method employs a shared Siamese encoder with multi-scale Cross-Temporal Attention modules that derive spatial and channel attention from L2 feature differences, along with a lightweight confidence estimation head for per-pixel uncertainty modelling. A hybrid loss function combining confidence-weighted binary cross-entropy and focal loss is used to address class imbalance. Experiments on the LEVIR-CD dataset demonstrate that CTA-Net achieves an overall accuracy of 98.99%, an F1-score of 87.68%, an Intersection over Union of 78.06%, a Cohen’s kappa of 0.8715, and a Matthews Correlation Coefficient of 0.8721, with stable convergence and minimal overfitting. Qualitative and calibration analyses further indicate that the model produces interpretable attention maps and reliable probabilistic outputs. To evaluate cross-domain generalization, we conduct a transfer learning case study on multispectral Sentinel-2 agricultural imagery. The model is adapted to 11-channel input and fine-tuned on automatically generated change masks derived from NDVI-delta thresholding. Under this supervision protocol, CTA-Net achieves an F1-score of 95.18% and an IoU of 90.81% on a held-out test region, with balanced precision and recall. While these results demonstrate effective adaptation across sensor modality, spatial resolution, and semantic domain, the evaluation reflects agreement with the mask generation procedure rather than independently annotated ground truth. While CTA-Net shows strong performance and reasonable interpretability, its cross-domain evaluation is limited by the use of automatically generated labels. As a result, the reported transferability should be interpreted cautiously until validated on human-annotated datasets. Full article

Within the domain of Earth observation, tracking terrestrial transitions via high-resolution optical data plays a fundamental role. Nevertheless, current methods face critical challenges, including the difficulty in collaborative modeling of local details and global features and the singularity of bi-temporal difference representation, along with insufficient cross-scale feature communication, thereby constraining both the precision and resilience of models when applied to complicated environments. To solve these problems, we propose LADENet (Locally Adaptive Mamba and Multi-scale Feature Enhancement Network), an innovative framework that synergizes CNN, Transformer, and Mamba paradigms. By leveraging customized local contextual refinement alongside sophisticated hierarchical fusion, this integration delivers highly precise and resilient detection performance. LADENet adopts a weight-sharing multi-level Transformer encoder combined with a sequence reduction mechanism to generate multi-scale global features, achieving precise alignment of bi-temporal features and global context modeling while reducing computational complexity. To realize accurate localization and local enhancement of changed regions, we design a dual spatiotemporal adaptive local feature marking module based on State-Space Scanning (SSS). This module screens high-saliency changed regions through an adaptive scanning strategy, realizes pixel-aligned spatiotemporal feature fusion via cross-temporal state-space scanning, and introduces a sliding window boundary calibration mechanism to alleviate boundary information loss caused by window segmentation. To strengthen the feature representation of changed regions, a dual-branch difference enhancement module is constructed, which collaboratively captures global change trends and fine-grained local features through an attention-enhanced difference branch and a multi-scale convolution concatenation branch, effectively suppressing background interference. To address the semantic gap between cross-scale features, a global cross-scale spatial feature fusion decoder is proposed, which balances local detail preservation and global context perception through the synergy of spatial attention and two-dimensional selective scanning, completing refined multi-scale feature fusion and spatial resolution recovery. To rigorously validate the proposed LADENet, comprehensive experiments were conducted across four widely adopted bi-temporal benchmarks: LEVIR-CD, WHU-CD, CLCD-CD, and GVLM-CD. The presented architecture establishes substantial superiority over existing cutting-edge methodologies across primary evaluation criteria. Specifically, it yields an F1-measure of 91.06% alongside an IoU of 85.28% in the LEVIR-CD tests, while registering 90.51% (F1) and 82.45% (IoU) for WHU-CD. Similarly, robust outcomes are delivered on CLCD-CD (82.15% F1, 72.83% IoU) as well as GVLM-CD (89.12% F1, 77.78% IoU). These results demonstrate that LADENet possesses excellent detection accuracy, boundary delineation capability and generalization performance in diverse and intricate bi-temporal observation environments. Full article

Synthetic microbial community (SynCom) biosensors are emerging from the convergence of whole-cell biosensing, synthetic ecology, and computational design. Conventional whole-cell biosensors (WCBs) use a single microbial chassis to convert analyte recognition into optical, electrochemical, gaseous, or growth-linked outputs. This compact architecture supports low-cost and field-oriented detection, but it can be limited by cellular burden, narrow dynamic range, environmental interference, and difficulty in interpreting multicomponent signals. Natural microbial consortia provide an ecological template in which sensing, transformation, stress tolerance, and response are distributed across interacting populations. SynCom biosensors seek to translate this logic into engineered platforms with defined members, assigned functional roles, designed communication, and interpretable readouts. This review traces the transition from WCBs to natural consortia and engineered multicellular biosensors, emphasizing functional partitioning, signal routing, community control, and artificial intelligence (AI)-assisted design. AI is discussed as a practical tool for narrowing design space, predicting interactions, decoding complex biosignals, and supporting adaptive operation. Key challenges remain in community stability, orthogonal communication, data quality, biosafety, standardization, and real-sample validation. Future progress will depend on parsimonious community design, reliable containment, quantitative validation, and computational workflows that connect community composition with sensing performance. Full article

Multimodal fusion unlocks the potential of point cloud semantic segmentation, thereby driving advancements in surface observation and visual perception tasks. Although light detection and ranging (LiDAR) systems capture precise 3D structural geometry and optical images provide rich semantic and textural information, existing fusion methods struggle with limited cross-modal perception and insufficient information complementarity. To address these limitations, we propose a multi-stage LiDAR-image collaborative perception fusion network (MCPFNet) for point cloud semantic segmentation of urban scenes. At the middle fusion stage, the network incorporates an elevation-guided geometric-aware fusion module and a semantic-aware cross-attention fusion module to enable bidirectional feature injection between LiDAR and image modalities. In the late fusion stage, a bidirectional adaptive fusion module further refines semantic representations through gated weighting and bidirectional cross-attention mechanisms. Extensive experiments on three multimodal datasets with different resolutions, i.e., ISPRS Vaihingen, N3C-California, and UAVScenes, demonstrate that MCPFNet outperforms existing fusion methods, achieving mIoUs of 74.51%, 95.15%, and 62.76%, respectively. Hence, our multi-stage fusion and bidirectional interaction strategy is more reliable and accurate than existing methods in performing segmentation across diverse and complex urban scenes. Full article

This work presents an automated vision-based measurement system designed for the precise dimensional characterization of high-aspect-ratio micro-holes, achieving a relative dimensional error of less than 1% for characterizing high-aspect-ratio damage geometries. The system integrates coaxial microscopic imaging with a precision motorized scanning stage. To ensure high-fidelity measurements in early-stage warning applications, depth is determined using a focus variation method driven by a robust data fusion strategy. By capturing a sequence of images along the Z-axis, the focal planes of the defect’s surface orifice and internal base are automatically identified using a data fusion algorithm based on a consensus evaluation of three parallel sharpness metrics (Tenengrad, Laplacian, and Brenner variants). The Z-axis scanning module, featuring encoder feedback and bi-directional compensation, achieves a repeated positioning error of ±0.5 µm. For lateral damage assessment, the system’s high magnification provides an effective sampling resolution of 0.09 µm. The equivalent diameter of the focused orifice image is calculated through a robust pipeline involving adaptive thresholding, morphological filtering, and sub-pixel ellipse fitting, which serves as a highly sensitive indicator for early-stage structural deformation. The entire process can be completed within five minutes, demonstrating a rapid, highly accurate, and localized optical inspection solution that generates high-precision dimensional data crucial for quality inspection in aerospace and precision engineering. Full article