Search Results (84)

Background: Automatic vehicle recognition has recently become an area of great interest, providing substantial support for multiple use cases, including law enforcement and surveillance applications. In real traffic conditions, where for various reasons license plate recognition is impossible or license plates are forged, alternative solutions are required to support human personnel in identifying vehicles used for illegal activities. In such cases, appearance-based approaches relying on vehicle make and model recognition (VMMR) and vehicle color recognition (VCR) can successfully complement license plate recognition. Methods: This research addresses appearance-based vehicle identification, in which VMMR and VCR rely on inherent visual cues such as body contours, stylistic details, and exterior color. In the first stage, vehicles passing through an intersection are detected, and essential visual characteristics are extracted for the two recognition tasks. The proposed system employs deep learning with semantic segmentation and data augmentation for color recognition, while histogram of oriented gradients (HOG) feature extraction combined with a support vector machine (SVM) classifier is used for make-model recognition. For the VCR task, five different neural network architectures are evaluated to identify the most effective solution. Results: The proposed system achieves an overall accuracy of 94.89% for vehicle make and model recognition. For vehicle color recognition, the best-performing models obtain a Top-1 accuracy of 94.17% and a Top-2 accuracy of 98.41%, demonstrating strong robustness under real-world traffic conditions. Conclusions: The experimental results show that the proposed automatic vehicle recognition system provides an efficient and reliable solution for appearance-based vehicle identification. By combining region-tailored data, segmentation-guided processing, and complementary recognition strategies, the system effectively supports real-world surveillance and law-enforcement scenarios where license plate recognition alone is insufficient. Full article

As industrialized construction and smart building continue to advance, rebar-tying robots place higher demands on the real-time and accurate recognition of rebar intersections and their tying status. Existing deep learning-based detection methods generally rely on heavy backbone networks and complex feature-fusion structures, making it difficult to deploy them efficiently on resource-constrained mobile robots and edge devices, and there is also a lack of dedicated datasets for rebar intersections. In this study, 12,000 rebar mesh images were collected and annotated from two indoor scenes and one outdoor scene to construct a rebar-intersection dataset that supports both object detection and instance segmentation, enabling simultaneous learning of intersection locations and tying status. On this basis, a lightweight improved YOLOv8-based method for rebar intersection detection and segmentation is proposed. The original backbone is replaced with ShuffleNetV2, and a C2f_Dual residual module is introduced in the neck; the same improvements are further transferred to YOLOv8-seg to form a unified lightweight detection–segmentation framework for joint prediction of intersection locations and tying status. Experimental results show that, compared with the original YOLOv8L and several mainstream detectors, the proposed model achieves comparable or superior performance in terms of mAP@50, precision and recall, while reducing model size and computational cost by 51.2% and 58.1%, respectively, and significantly improving inference speed. The improved YOLOv8-seg also achieves satisfactory contour alignment and regional consistency for rebar regions and intersection masks. Owing to its combination of high accuracy and low resource consumption, the proposed method is well suited for deployment on edge-computing devices used in rebar-tying robots and construction quality inspection, providing an effective visual perception solution for intelligent construction. Full article

The quality of scanned topographic maps—including parameters such as image compression, scanning resolution, and bit depth—may strongly influence the performance of deep learning models for contour vectorization. In this study, we investigate this dependence by training eight U-Net models on the same map data but under varying input quality conditions. Each model is trained to segment contour lines from the raster input, followed by a postprocessing pipeline that converts segmented output into vector contours. We systematically compare the models with respect to topological error metrics (such as contour intersections and dangling ends) in the resulting vector output and overlay metrics of matched contour segments within given tolerance. Our experiments demonstrate that while the input data quality indeed matters, moderate lowering of quality parameters doesn’t introduce significant practical tradeoff, while storage and computational requirements remain low. We discuss implications for the preparation of archival map scans and propose guidelines for choosing scanning settings when the downstream goal is automated vectorization. Our results highlight that deep learning methods, though resilient against reasonable compression, remain measurably sensitive to degradation in input fidelity. Full article

Substation equipment operating in harsh environments is highly susceptible to corrosion, yet conventional image segmentation methods often fail to achieve precise delineation of corroded regions. Here, we propose an enhanced TransU-Net-based approach for corrosion segmentation. Deformable convolution is incorporated into the encoder to strengthen the model’s capacity to represent irregular corrosion morphologies. A composite color–texture fusion module is developed to jointly exploit color information from HSV and Lab spaces together with multi-scale texture features. In addition, a Shape-IoU loss function is introduced to refine boundary fitting and improve contour accuracy. Experimental evaluations demonstrate that the proposed method consistently outperforms state-of-the-art models across multiple metrics, achieving an Intersection over Union (IoU) of 75.42% and a Recall (PA) of 83.14%. These results confirm that the model substantially enhances corrosion recognition accuracy and edge integrity under complex background conditions, offering a promising strategy for intelligent maintenance of substation infrastructure. Full article

Digital surface models (DSMs) derived from high-resolution satellite imagery often contain mismatches, voids, and coarse building geometry, limiting their suitability for accurate and standardized 3D reconstruction. The scarcity of finely annotated samples further constrains generalization to complex structures. To address these challenges, an automated building reconstruction method based on two-stage polygon decomposition and adaptive roof fitting is proposed. Building polygons are first extracted and standardized to preserve primary contours while improving geometric regularity. A two-stage decomposition is then applied. In the first stage, polygons are coarsely decomposed, and redundant rectangles are removed by analyzing containment relationships. In the second stage, non-flat regions are identified and further decomposed to accommodate complex building connections. For 3D model fitting, flat-roof buildings are reconstructed by integrating structural analysis of DSM elevation distributions with adaptive rooftop partitioning, which enables accurate modeling of complex flat structures with auxiliary components. For non-flat roofs, a representative parameter space is defined and explored through systematic search and optimization to obtain precise fits. Finally, intersecting primitives are normalized and optimally merged to ensure structural coherence and standardized representation. Experiments on the US3D, MVS3D, and Beijing-3 datasets demonstrate that the proposed method achieves higher geometric accuracy and more standardized models, with an average IOU3 of 91.26%, RMSE of 0.78 m, and MHE of 0.22 m. Full article

Discontinuity trace provides critical geological data for engineering design and construction optimization. However, current extraction methods relying on discontinuity intersection fitting are highly sensitive to the segmentation accuracy of individual discontinuity, while trace segment connectivity remains suboptimal. To address these challenges, we propose an ARCG (Adaptive Region Contour Growing) method using 3D point clouds. By dynamically adjusting parameter thresholds, our approach simultaneously extracts both discontinuities and their boundaries. We then evaluate the fitting performance of different discontinuity models using area ratios, identifying the parallelogram as the most suitable representation. The method then detects intersection lines between paired discontinuities through spatial intersection analysis, with dynamic partitioning preserving original geometric properties. Finally, a bidirectional weighted graph-based growth algorithm connects intersection lines belonging to the same discontinuity, generating the final trace results. The proposed method was validated using slope data from two case studies. Results demonstrate that, compared to existing methods and point cloud processing software, our approach achieves robust extraction of complex traces while maintaining high connectivity. Moreover, it improves computational efficiency by 48.8% without compromising trace accuracy. Thus, this method offers a novel solution for the digital characterization of rock mass discontinuity parameters. Full article

In this study, we utilize Gaofen-2 satellite remote sensing images to optimize and enhance the extraction of feature information from rural compounds, addressing key challenges in high-resolution remote sensing analysis: traditional methods struggle to effectively capture long-distance spatial dependencies for scattered rural compounds. To this end, we implement the RS³Mamba+ deep learning model, which introduces the Mamba state space model (SSM) into its auxiliary branching—leveraging Mamba’s sequence modeling advantage to efficiently capture long-range spatial correlations of rural compounds, a critical capability for analyzing sparse rural buildings. This Mamba-assisted branch, combined with multi-directional selective scanning (SS2D) and the enhanced STEM network framework (replacing single 7 × 7 convolution with two-stage 3 × 3 convolutions to reduce information loss), works synergistically with a ResNet-based main branch for local feature extraction. We further introduce a multiscale attention feature fusion mechanism that optimizes feature extraction and fusion, enhances edge contour extraction accuracy in courtyards, and improves the recognition and differentiation of courtyards from regions with complex textures. The feature information of courtyard utilization status is finally extracted using empirical methods. A typical rural area in Weifang City, Shandong Province, is selected as the experimental sample area. Results show that the extraction accuracy reaches an average intersection over union (mIoU) of 79.64% and a Kappa coefficient of 0.7889, improving the F1 score by at least 8.12% and mIoU by 4.83% compared with models such as DeepLabv3+ and Transformer. The algorithm’s efficacy in mitigating false alarms triggered by shadows and intricate textures is particularly salient, underscoring its potential as a potent instrument for the extraction of rural vacancy rates. Full article

The extraction of architectural structural line features can simplify the 3D spatial representation of built environments, reduce the storage and processing burden of large-scale point clouds, and provide essential geometric primitives for downstream modeling tasks. However, existing 3D line extraction methods suffer from incomplete and fragmented contours, with missing or misaligned intersections. To overcome these limitations, this study proposes a patch-level framework for 3D line extraction and structural optimization from building point clouds. The proposed method first partitions point clouds into planar patches and establishes local image planes for each patch, enabling a structured 2D representation of unstructured 3D data. Then, graph-cut segmentation is proposed to extract compact boundary contours, which are vectorized into closed lines and back-projected into 3D space to form the initial line segments. To improve geometric consistency, regularized geometric constraints, including adjacency, collinearity, and orthogonality constraints, are further designed to merge homogeneous segments, refine topology, and strengthen structural outlines. Finally, we evaluated the approach on three indoor building environments and four outdoor scenes, and experimental results show that it reduces noise and redundancy while significantly improving the completeness, closure, and alignment of 3D line features in various complex architectural structures. Full article

Background/Objectives: Accurate segmentation of brain tumors in Magnetic Resonance Imaging (MRI) scans is critical for diagnosis and treatment planning due to their life-threatening nature. This study aims to develop a robust and automated method capable of precisely delineating heterogeneous tumor regions while improving segmentation accuracy and generalization. Methods: We propose Attention Res-UNet (ARU-Net), a novel Deep Learning (DL) architecture integrating residual connections, Adaptive Channel Attention (ACA), and Dimensional-space Triplet Attention (DTA) modules. The encoding module efficiently extracts and refines relevant feature information by applying ACA to the lower layers of convolutional and residual blocks. The DTA is fixed to the upper layers of the decoding module, decoupling channel weights to better extract and fuse multi-scale features, enhancing both performance and efficiency. Input MRI images are pre-processed using Contrast Limited Adaptive Histogram Equalization (CLAHE) for contrast enhancement, denoising filters, and Linear Kuwahara filtering to preserve edges while smoothing homogeneous regions. The network is trained using categorical cross-entropy loss with the Adam optimizer on the BTMRII dataset, and comparative experiments are conducted against baseline U-Net, DenseNet121, and Xception models. Performance is evaluated using accuracy, precision, recall, F1-score, Dice Similarity Coefficient (DSC), and Intersection over Union (IoU) metrics. Results: Baseline U-Net showed significant performance gains after adding residual connections and ACA modules, with DSC improving by approximately 3.3%, accuracy by 3.2%, IoU by 7.7%, and F1-score by 3.3%. ARU-Net further enhanced segmentation performance, achieving 98.3% accuracy, 98.1% DSC, 96.3% IoU, and a superior F1-score, representing additional improvements of 1.1–2.0% over the U-Net + Residual + ACA variant. Visualizations confirmed smoother boundaries and more precise tumor contours across all six tumor classes, highlighting ARU-Net’s ability to capture heterogeneous tumor structures and fine structural details more effectively than both baseline U-Net and other conventional DL models. Conclusions: ARU-Net, combined with an effective pre-processing strategy, provides a highly reliable and precise solution for automated brain tumor segmentation. Its improvements across multiple evaluation metrics over U-Net and other conventional models highlight its potential for clinical application and contribute novel insights to medical image analysis research. Full article

Traditional orebody modeling methods struggle to efficiently integrate new geological data. Therefore, we propose a novel framework for dynamically updating 3D geological models by directly interpolating geological logging data. The core innovation lies in the innovative interpolation of raw interpreted cross polylines into an implicit scalar field representation without intermediate explicit surface extraction or manual remodeling. To obtain reliable vectorized polylines, we developed image recognition and digitization techniques that are based on the pattern recognition of geological sketches. Moreover, different from existing implicit techniques, we present an improved approach to interpolate complex cross polylines that are dynamically based on the improved principal component analysis. The method allows specifying a priori constraints to adjust the erroneous estimated normal to improve the reliability of the normal estimation results of cross-contour polylines. The a priori information can be combined into the normal estimation algorithm to update the normals of the corresponding adjacent contour polylines in the process of normal estimation at the intersection points and in the process of normal propagation. By leveraging the radial basis functions-based spatial interpolators, the method continuously assimilates incremental geological observations into the interpolation constraints to update the implicit model. Case studies demonstrate a reduction in the modeling cycle time compared to conventional explicit methods while maintaining geologically coherent boundaries. The framework significantly enhances decision agility in resource estimation and mine planning workflows by bridging geological interpretation and dynamic model iteration. Full article

This study proposes an improved YOLO11n-seg instance segmentation model to address the limitations of existing models in accurately identifying mature blueberries in complex greenhouse environments. Current methods often lack sufficient accuracy when dealing with complex scenarios, such as fruit occlusion, lighting variations, and target overlap. To overcome these challenges, we developed a novel approach that integrates a Spatial–Channel Adaptive (SCA) attention mechanism and a Dual Attention Balancing (DAB) module. The SCA mechanism dynamically adjusts the receptive field through deformable convolutions and fuses multi-scale color features. This enhances the model’s ability to recognize occluded targets and improves its adaptability to variations in lighting. The DAB module combines channel–spatial attention and structural reparameterization techniques. This optimizes the YOLO11n structure and effectively suppresses background interference. Consequently, the model’s accuracy in recognizing fruit contours improves. Additionally, we introduce Normalized Wasserstein Distance (NWD) to replace the traditional intersection over union (IoU) metric and address bias issues that arise in dense small object matching. Experimental results demonstrate that the improved model significantly improves target detection accuracy, recall rate, and mAP@0.5, achieving increases of 1.8%, 1.5%, and 0.5%, respectively, over the baseline model. On our self-built greenhouse blueberry dataset, the mask segmentation accuracy, recall rate, and mAP@0.5 increased by 0.8%, 1.2%, and 0.1%, respectively. In tests across six complex scenarios, the improved model demonstrated greater robustness than mainstream models such as YOLOv8n-seg, YOLOv8n-seg-p6, and YOLOv9c-seg, especially in scenes with dense occlusions. The improvement in mAP@0.5 and F1 scores validates the effectiveness of combining attention mechanisms and multiple metric optimizations, for instance, segmentation tasks in complex agricultural scenes. Full article

Automatic segmentation of nuclei on breast cancer histology images is a basic and important step for diagnosis in a computer-aided diagnostic approach and helps pathologists discover cancer early. Nuclei segmentation remains a challenging problem due to cancer biology and the variability of tissue characteristics; thus, their detection in an image is a very tedious and time-consuming task. In this context, overlapping nuclei objects present difficulties in separating them by conventional segmentation methods; thus, active contours can be employed in image segmentation. A major limitation of the active contours method is its inability to resolve image boundaries/edges of intersecting objects and segment multiple overlapping objects as a single object. Therefore, we present a hybrid active contour (connected component + active contours) method to segment cancerous lesions in unsupervised human breast histology images. Initially, this approach prepares and pre-processes data through various augmentation methods to increase the dataset size. Then, a stain normalization technique is applied to these augmented images to isolate nuclei features from tissue structures. Secondly, morphology operation techniques, namely erosion, dilation, opening, and distance transform, are used to highlight foreground and background pixels while removing overlapping regions from the highlighted nuclei objects on the image. Consequently, the connected components method groups these highlighted pixel components with similar intensity values and assigns them to their relevant labeled component to form a binary mask. Once all binary-masked groups have been determined, a deep-learning recurrent neural network (RNN) model from the Keras architecture uses this information to automatically segment nuclei objects having cancerous lesions on the image via the active contours method. This approach, therefore, uses the capabilities of connected components analysis to solve the limitations of the active contour method. This segmentation method is evaluated on an unsupervised, augmented human breast cancer histology dataset of 15,179 images. This proposed method produced a significant evaluation result of $98.71\%$ accuracy score. Full article

In the production process of cross-cutting boards, real-time measurement of dimensions online has been a long-standing technical problem in the production field. Currently, the detection of board dimensions in the production field relies on manual observation based on workers’ operational experience or stopping the machine for measurement. This paper proposes a machine vision-based real-time online measurement system for dimensional measurements of cross-cutting units. A certain angle measurement model is established by using a face-array industrial camera, and a more accurate edge contour extraction is realized by deep learning. A novel edge intersection extraction algorithm based on line fitting and least squares method was proposed to accurately measure the length, width, diagonal lines of cross-cutting boards using four intersection coordinates. The measurement of 100 cross-cutting boards in the industrial production site shows that the proposed online measurement system for cross-cut board dimensions in this article has high accuracy, with a length perception error of ±50 mm, width of ±2 mm, and diagonal difference of ±5 mm, meeting the production requirements in industrial settings. The on-site shutdown measurement work was reduced, thereby doubling the production efficiency and saving two staff members. Full article

Traditional offline programming has limitations for large parts with significant machining or assembly deviations. This study proposes a 3D scanning-assisted method that generates accurate STereoLithography (STL) models and enables multi-layer multi-bead welding trajectory planning for large intersecting line workpieces. The proposed framework implements a robust STL model processing pipeline incorporating Random Sample Consensus (RANSAC)-based cylindrical approximation, cross-sectional slicing, and automated feature detection to achieve high-precision groove feature recognition. For asymmetric variable-section grooves, a multi-layer and multi-pass path-planning algorithm based on template affine projection transformation is developed to ensure accurate deposition of welds along complex geometric contours. Experimental validation demonstrates sub-millimeter trajectory accuracy (positional errors < 1.0 mm), meeting stringent arc welding specifications and substantially expanding the applicability of offline programming systems. Full article

Plant height (PH), leaf width (LW), and leaf angle (LA) are critical phenotypic parameters in maize that reliably indicate plant growth status, lodging resistance, and yield potential. While various lidar-based methods have been developed for acquiring these parameters, existing approaches face limitations, including low automation, prolonged measurement duration, and weak environmental interference resistance. This study proposes a novel estimation method for maize PH, LW, and LA based on point cloud projection. The methodology comprises four key stages. First, 3D point cloud data of maize plants are acquired during middle–late growth stages using lidar sensors. Second, a Gaussian mixture model (GMM) is employed for point cloud registration to enhance plant morphological features, resulting in spliced maize point clouds. Third, filtering techniques remove background noise and weeds, followed by a combined point cloud projection and Euclidean clustering approach for stem–leaf segmentation. Finally, PH is determined by calculating vertical distance from plant apex to base, LW is measured through linear fitting of leaf midveins with perpendicular line intersections on projected contours, and LA is derived from plant skeleton diagrams constructed via linear fitting to identify stem apex, stem–leaf junctions, and midrib points. Field validation demonstrated that the method achieves 99%, 86%, and 97% accuracy for PH, LW, and LA estimation, respectively, enabling rapid automated measurement during critical growth phases and providing an efficient solution for maize cultivation automation. Full article