Search Results (9)

Background: Accurate white blood cell (WBC) segmentation is important for quantitative microscopic image analysis, but conventional supervised approaches depend on labor-intensive pixel-level annotations and may exhibit reduced robustness across datasets. This study proposes a detector-guided, region of interest (ROI)-constrained framework that combines automatic localization, pseudo-mask-based weak supervision, and diffusion-assisted segmentation. Methods: You Only Look Once version 13 Nano (YOLOv13-N) was used to localize WBCs and define ROIs, within which segmentation was performed using the proposed diffusion-assisted model. Automatically generated pseudo-masks served as segmentation-training targets, while expert masks were retained for reference evaluation. Experiments used a verified Dicle cohort of 14,721 records, partitioned into 10,305 training, 2208 validation, and 2208 held-out test records. Performance was evaluated separately at ROI-conditional and end-to-end levels. Generalization was assessed on 11,200 Raabin-WBC records using the frozen framework without external tuning. Results: The automatically generated pseudo-masks achieved a mean Dice score of approximately 0.702, demonstrating usable but imperfect weak supervision. On the held-out Dicle test set, the proposed framework achieved a Dice score of 0.7188, intersection over union (IoU) of 0.5796, precision of 0.8966, and recall of 0.6422 under ROI-conditional evaluation. End-to-end performance was 0.6705 Dice, 0.5408 IoU, 0.8378 precision, and 0.5986 recall, demonstrating the influence of localization on overall performance. Comparison with reference segmentation architectures showed that the proposed framework did not maximize internal Dice, while achieving the highest reported ROI-conditional precision. Frozen external evaluation on Raabin-WBC revealed further degradation under dataset shift, with failure analysis identifying detector/ROI transfer as an important end-to-end bottleneck. Conclusions: The proposed framework demonstrates the feasibility of WBC segmentation using detector-guided ROI processing and pseudo-mask-based weak supervision, reducing reliance on expert pixel-level segmentation targets. The findings further show that robust cross-dataset localization is critical to end-to-end generalization and provide a clear direction for improving weakly supervised WBC segmentation across heterogeneous microscopy datasets. Full article

UAV-based ship segmentation is important for berthing and unberthing monitoring, ship–berth distance estimation, and situational awareness in port waters. However, direct video mask propagation with Segment Anything Model 2 (SAM 2) remains susceptible to local contour degradation in high-resolution UAV videos containing weak berth-side boundaries, adjacent tugboats, quay-side structures, water-surface reflections, and target-scale variations. To address this problem, a constrained local boundary refinement method is proposed for target-ship segmentation. The method follows a training-free, first-frame-mask-initialized semi-supervised video object segmentation setting, with all SAM 2 parameters remaining frozen. ROI Boundary Re-Inference first enhances weak contours within local target neighborhoods. Prompt-Consensus Refinement then retains boundary candidates consistently supported by multiple structured prompt variants. Finally, Boundary-Constrained Non-Erosive Fusion restricts supplementation to a narrow neighborhood of the propagated boundary and incorporates reliable candidates without deleting the original foreground. Experiments on a self-built UAV berthing and unberthing video dataset show that the proposed method improves Boundary F@5 px from 89.68% to 94.03% and J&F from 94.08% to 96.39%. These results demonstrate that the proposed method improves target-ship boundary delineation without model training or fine-tuning. Full article

Accurate instance segmentation of oranges in complex orchard environments is crucial for obtaining clean regions of interest (ROIs). Coarse region extraction may include non-target pixels from leaves, shadows, background, and adjacent fruits, thereby increasing boundary pixel mixing in subsequent hyperspectral-assisted observation. This study proposes an improved lightweight YOLO11n-Seg method as an RGB-based visual front-end for cleaner single-fruit ROI extraction. Its contribution lies in the task-oriented integration of three complementary components: a Local Deformable Convolution Backbone (LDC-Backbone) for representing irregular and occluded fruit contours, a Boundary-Guided GSConv (BG-GSConv) module for efficiently fusing shallow boundary details with deep semantic features, and an ROI-Purity-Oriented Dice Boundary Loss for constraining mask integrity and boundary adherence. Evaluated on a complex orchard dataset, the improved model achieved a Mask mAP@0.5 of 0.962, a Mask mAP@0.5:0.95 of 0.692, a Box mAP@0.5 of 0.942, and an inference speed of 101 FPS with 3.20 M parameters. Background leakage analysis further showed that the proposed model reduced the inclusion of non-fruit pixels in extracted ROIs, supporting cleaner mask-based single-fruit region extraction. Preliminary ROI-based reflectance observation indicated that the reflectance curves obtained from the improved-model ROIs were closer to those of manually referenced pure ROIs than those obtained from the baseline extraction. These results suggest that the proposed method can serve as a real-time RGB-based front-end for cleaner single-fruit ROI extraction and later hyperspectral-assisted sampling. Complete closed-loop spectral quality modeling with paired RGB–HSI data remains a direction for future work. Full article

This work introduces the MassSeg-Framework, a fully automatic two-stage pipeline for breast mass analysis in mammography that integrates YOLOv11-based detection with Chan–Vese ACM refinement to achieve accurate mass localization and segmentation with a lightweight computational footprint. The framework was trained and evaluated on two publicly available datasets using consistent experimental protocols. In the detection stage, YOLOv11-nano was the most effective architecture, with a confidence threshold of 0.4, achieving statistically significant mAP50 values of 0.862 and 0.709 on the dINbreast and dCBIS datasets, respectively. These results confirm that a moderate threshold preserves clinically relevant true-positive candidates, which is particularly important for screening-oriented settings where missed lesions are costly. In the segmentation stage, the proposed framework achieved mean DICE scores of 0.721 and 0.700 on the test sets of the same datasets, demonstrating consistent overlap with expert annotations. Compared with state-of-the-art approaches that commonly assume lesion-centered ROIs or rely on heavier backbones, the proposed pipeline addresses a more realistic scenario by performing automatic detection followed by segmentation while maintaining substantially lower computational requirements. This balance between performance and efficiency makes the MassSeg-Framework a promising tool for scalable mammography analysis, particularly in resource-constrained environments or high-throughput screening workflows that require rapid processing. Full article

Recently, deep learning methods have seen major advancements and are preferred for medical image analysis. Clinically, deep learning techniques for cancer image analysis are among the main applications for early diagnosis, detection, and treatment. Consequently, segmentation of breast histology images is a key step towards diagnosing breast cancer. However, the use of deep learning methods for image analysis is constrained by challenging features in the histology images. These challenges include poor image quality, complex microscopic tissue structures, topological intricacies, and boundary/edge inhomogeneity. Furthermore, this leads to a limited number of images required for analysis. The U-Net model was introduced and gained significant traction for its ability to produce high-accuracy results with very few input images. Many modifications of the U-Net architecture exist. Therefore, this study proposes the watershed encoder–decoder neural network (WEDN) to segment cancerous lesions in supervised breast histology images. Pre-processing of supervised breast histology images via augmentation is introduced to increase the dataset size. The augmented dataset is further enhanced and segmented into the region of interest. Data enhancement methods such as thresholding, opening, dilation, and distance transform are used to highlight foreground and background pixels while removing unwanted parts from the image. Consequently, further segmentation via the connected component analysis method is used to combine image pixel components with similar intensity values and assign them their respective labeled binary masks. The watershed filling method is then applied to these labeled binary mask components to separate and identify the edges/boundaries of the regions of interest (cancerous lesions). This resultant image information is sent to the WEDN model network for feature extraction and learning via training and testing. Residual convolutional block layers of the WEDN model are the learnable layers that extract the region of interest (ROI), which is the cancerous lesion. The method was evaluated on 3000 images–watershed masks, an augmented dataset. The model was trained on 2400 training set images and tested on 600 testing set images. This proposed method produced significant results of $98.53\%$ validation accuracy, $96.98\%$ validation dice coefficient, and $97.84\%$ validation intersection over unit (IoU) metric scores. Full article

Accurate segmentation of coronary artery calcifications (CAC) from cardiac CT is challenged by class imbalance, small lesion size, and anatomical ambiguity. We present an anatomically guided, cascaded framework that couples heart and vessel priors with a heterogeneous U-Net ensemble for robust, vessel-aware CAC segmentation. First, a ResU-Net trained on MM-WHS isolates the heart region of interest (ROI). Second, a ResU-Net trained on ASOCA—using Frangi vesselness enhancement—segments the coronary arteries, yielding vessel masks that constrain downstream lesion detection. Third, calcifications are segmented within the vessel-constrained ROI using an ensemble of U-Net variants (baseline U-Net, Residual U-Net, Attention U-Net, UNet++). At inference, a rank-based selective fusion strategy prioritizes predictions with strong morphological consistency and vessel conformity, suppressing false positives. On the Stanford COCA gated dataset, the proposed ensemble outperforms individual models (Dice 84.25%, sensitivity 87.10%, specificity 98.00%), with ablations demonstrating additional gains when vessel priors are integrated into selective fusion (Dice 85.50%, sensitivity 88.53%). Results confirm that combining dataset-specific anatomical priors with selective ensembling improves boundary sharpness, small-lesion detectability, and anatomical plausibility, supporting reliable CAC segmentation in clinical imaging workflows. Full article

The safety and reliability of modern power systems are increasingly challenged by adverse environmental conditions. (1) Background: Ice accumulation on power transmission lines is recognized as a severe threat to grid stability, as tower collapse, conductor breakage, and large-scale outages may be caused, thereby making accurate monitoring essential. (2) Methods: A rule-driven and interpretable stereo vision framework is proposed for three-dimensional (3D) detection and quantitative measurement of transmission line icing. The framework consists of three stages. First, adaptive preprocessing and segmentation are applied using multiscale Retinex with nonlinear color restoration, graph-based segmentation with structural constraints, and hybrid edge detection. Second, stereo feature extraction and matching are performed through entropy-based adaptive cropping, self-adaptive keypoint thresholding with circular descriptor analysis, and multi-level geometric validation. Third, 3D reconstruction is realized by fusing segmentation and stereo correspondences through triangulation with shape-constrained refinement, reaching millimeter-level accuracy. (3) Result: An accuracy of 98.35%, sensitivity of 91.63%, specificity of 99.42%, and precision of 96.03% were achieved in contour extraction, while a precision of 90%, recall of 82%, and an F1-score of 0.8594 with real-time efficiency (0.014–0.037 s) were obtained in stereo matching. Millimeter-level accuracy (Mean Absolute Error: 1.26 mm, Root Mean Square Error: 1.53 mm, Coefficient of Determination = 0.99) was further achieved in 3D reconstruction. (4) Conclusions: Superior accuracy, efficiency, and interpretability are demonstrated compared with two existing rule-based stereo vision methods (Method A: ROI Tracking and Geometric Validation Method and Method B: Rule-Based Segmentation with Adaptive Thresholding) that perform line icing identification and 3D reconstruction, highlighting the framework’s advantages under limited data conditions. The interpretability of the framework is ensured through rule-based operations and stepwise visual outputs, allowing each processing result, from segmentation to three-dimensional reconstruction, to be directly understood and verified by operators and engineers. This transparency facilitates practical deployment and informed decision making in real world grid monitoring systems. Full article

In the surveillance of interested regions by unmanned aerial vehicle (UAV), system performance relies greatly on the motion control strategy of the UAV and the operation characteristics of the onboard sensors. This paper investigates the 2D path planning problem for the lightweight UAV synthetic aperture radar (SAR) system in an environment of multiple regions of interest (ROIs), the sizes of which are comparable to the radar swath width. Taking into account the special requirements of the SAR system on the motion of the platform, we model path planning for UAV SAR as a constrained multiobjective optimization problem (MOP). Based on the fact that the UAV route can be designed in the map image, an image-based path planner is proposed in this paper. First, the neighboring ROIs are merged by the morphological operation. Then, the parts of routes for data collection of the ROIs can be located according to the geometric features of the ROIs and the observation geometry of UAV SAR. Lastly, the route segments for ROIs surveillance are connected by a path planning algorithm named the sampling-based sparse A* search (SSAS) algorithm. Simulation experiments in real scenarios demonstrate that the proposed sensor-oriented path planner can improve the reconnaissance performance of lightweight UAV SAR greatly compared with the conventional zigzag path planner. Full article

Monitoring of mechanical structures is a Big Data challenge and includes Structural Health Monitoring (SHM) and Non-destructive Testing (NDT). The sensor data produced by common measuring techniques, e.g., guided wave propagation analysis, is characterized by a high dimensionality in the temporal and spatial domain. There are off- and on-line methods applied at maintenance- or run-time, respectively. On-line methods (SHM) usually are constrained by low-resource processing platforms, sensor noise, unreliability, and real-time operation requiring advanced and efficient sensor data processing. Commonly, structural monitoring is a task that maps high-dimensional input data on low-dimensional output data (information, which is feature extraction), e.g., in the simplest case a Boolean output variable “Damaged”. Machine Learning (ML), e.g., supervised learning, can be used to derive such a mapping function. But ML quality and performance depends strongly on the input data size. Therefore, adaptive and reliable input data reduction (that is feature selection) is required at the first layer of an automatic structural monitoring system. Assuming some kind of two-dimensional sensor data (or n-dimensional data in general), image segmentation can be used to identify Regions of Interest (ROI), e.g., of wave propagation fields. Wave propagation in materials underlie reflections that must be distinguished, especially in hybrid materials (e.g., combining metal and fibre-plastic composites) there are complex wave propagation fields. The image segmentation is one of the most crucial parts of image processing. Major difficulties in image segmentation are noise and the differing homogeneity (fuzziness and signal gradients) of regions, complicating the definition of suitable threshold conditions for the edge detection or region splitting/clustering. Many traditional image segmentation algorithms are constrained by this issue. Artificial Intelligence can aid to overcome this limitation by using autonomous agents as an adaptive and self-organizing software architecture, presented in this work. Using a collection of co-operating agents decomposes a large and complex problem in smaller and simpler problems with a Divide-and-Conquer approach. Related to the image segmentation scenario, agents are working mostly autonomous (de-coupled) on dynamically bounded data from different regions of a signal or an image (i.e., distributed with simulated mobility), adapted to the locality, being reliable and less sensitive to noisy sensor data. In this work, self-organizing agents perform segmentation. They are evaluated with measured high-dimensional data from piezo-electric acusto-ultrasonic sensors recording the wave propagation in plate-like structures. Commonly, SHM deploys only a small set of sensors and actuators at static positions delivering only a few temporal resolved sensor signals (1D), whereas NDT methods additionally can use spatial scanning to create images of wave signals (2D). Both one-dimensional temporal and two-dimensional spatial segmentation are considered to find characteristic ROI. Full article