Search Results (41)

This paper addresses the urgent scientific and applied problem of automatic tracking of highly maneuverable Unmanned Aerial Vehicles (UAVs) using systems based on platforms with limited computing power (Edge Computing). The paper analyzes the shortcomings of classical correlation trackers and detectors based on deep neural networks when tracking targets with non-linear trajectories. A hybrid tracking method is proposed, combining the speed of a Kernelized Correlation Filter (KCF) and the accuracy of a neural network detector (YOLO11s). A key feature of the method is the developed algorithm for adaptive Kalman Filter correction, which utilizes a dynamic, scale-invariant Prediction Error metric as a trigger for motion anomaly detection. This allows the system to distinguish between measurement noise and sharp target maneuvers, executing an adaptive state reset using finite differences only at critical moments. Experimental validation on edge hardware (Raspberry Pi 5) using highly dynamic video sequences from the UAV123 and VisDrone datasets demonstrated that the proposed approach maintains an average processing speed of 18.89 FPS. By limiting deep neural network invocations to merely 2.71% of total frames, the algorithm successfully curtails thermal throttling while achieving a global Mean Root Square Error (RMSE) of 259.10 pixels across highly erratic trajectories. The method ensures high tracking reliability without a critical increase in computational load, making it highly suitable for use in autonomous embedded systems. Full article

Edge detection is an essential problem in computer vision and is used in applications such as object recognition, scene analysis, and medical imaging. Conventional edge detectors based on integer-order derivatives are computationally efficient but sensitive to noise, whereas modern deep learning approaches generally achieve higher accuracy at the cost of increased model complexity. This paper presents GLF-ResFormer a lightweight hybrid CNN–Transformer architecture incorporating Grünwald–Letnikov (GL) fractional preprocessing. The discrete GL operator is approximated using a finite-difference convolution with a truncation level of $N=15$, where the fractional order $\alpha \in (0,1]$ controls the spatial memory of the operator. We establish an upper bound for the truncation error of the discrete GL approximation, $\mathcal{O}\left(h^{N-\alpha }\right)$ (Theorem 1), and present a gradient-sensitivity analysis (Lemma 1) that provides theoretical support for the proposed preprocessing strategy. Extensive experiments using 10 independent random seeds on the MNIST dataset show that, at the optimal fractional order of $\alpha =0.01$, GLF-ResFormer achieves a pixel-wise F1 score of $0.9967\pm 0.0002$ compared with $0.9891\pm 0.0033$ for a CNN baseline, while reducing the validation loss to $0.0043\pm 0.0003$. Additional experiments on the CIFAR-10 dataset and comparisons with the BSDS500 benchmark further demonstrate the effectiveness of the proposed framework across multiple edge detection evaluation settings while maintaining a lightweight architecture. Full article

This paper proposes a hybrid phase-correlation framework with a new multiscale detector, Fusion. Using a combination of FAST and Shi–Tomasi keypoints, followed by a probabilistic Hough transform and Canny edge detection, this detector improves repeatability. In addition, due to the limited ability of standard phase correlation to handle large geometric displacements, a complementary strategy is required to achieve sub-pixel matching precision. First, corners are extracted from both reference and sensed images using the Fusion detector. Corresponding points are then identified through coarse-to-fine phase correlation across a Gaussian pyramid. At each level, phase correlation yields an initial displacement, which is refined to sub-pixel accuracy using 1D parabolic fitting and propagated upward through the pyramid to obtain the final displacement. The proposed approach is evaluated using Pleiades and Sentinel-2 satellite images. Compared with the Scale-Invariant Feature Transform (SIFT)-based method and the detector-free Local Feature Transformer (LoFTR), the proposed framework achieves an RMSE below 0.2 and 0.4 pixels for Sentinel-2 and Pleiades imagery, respectively. Moreover, the results of the optimization analysis have revealed that shows that 2D paraboloid fitting combined achieves the lowest registration error of 0.010 pixels and the highest inlier ratio of 40.6%. The proposed approach achieves sub-pixel accuracy in the presence of noise and produces large numbers of correct matching points across different image resolutions. Full article

Low-altitude unmanned aerial vehicle (UAV)-based traffic object detection is challenged by substantial scale variations from aerial perspectives, the extremely small pixel proportions of distant traffic participants, complex road background interference, unstable illumination, and severe occlusion in dense traffic scenes. To address these problems, this paper proposes ACP2-YOLO, an improved YOLO11-based detection framework for low-altitude UAV traffic scenarios, with the goal of enhancing the detection of vehicles, pedestrians, and non-motorized traffic participants. The proposed framework introduces two key improvements. First, a lightweight hybrid ACmix module that integrates convolution and self-attention is embedded into the network, enabling the model to jointly capture local detailed features and global contextual dependencies and thereby strengthen feature representation under complex backgrounds. Second, a P2 small-object detection layer is added to the original three-scale detection structure of YOLO11 to construct a four-scale P2–P5 feature pyramid. By allowing shallow high-resolution features to directly participate in object prediction, this design effectively reduces spatial information loss caused by deep downsampling and improves small-object perception. Experiments on the VisDrone2019 dataset show that the improved model achieves 53.1% Precision, 41.1% Recall, 42.9% mAP@50, and 26.3% mAP@50–95, outperforming the baseline YOLO11 by 4.2, 4.2, 5.0, and 3.6 percentage points, respectively. Comparisons with mainstream YOLO-series detectors further demonstrate its superior overall accuracy, small-object detection capability, and adaptability to complex scenes, indicating its potential for UAV-based traffic monitoring, road safety inspection, and intelligent transportation perception. Full article

Pancreas segmentation in computed tomography (CT) images remains a challenging task due to the organ’s variable shape, size, and low contrast against surrounding tissues. In this study, we propose a two-stage pancreas segmentation framework that combines region localization using Faster R-CNN and pixel-level segmentation using a hybrid TransUNet architecture. To address issues related to class imbalance, we utilize the publicly available NIH pancreas CT dataset and apply downsampling techniques to construct a balanced training set. The segmentation stage incorporates a Transformer encoder into the U-Net framework and employs a novel DHD loss function, which combines the Dice Similarity Coefficient (DSC) and the Hausdorff Distance (HD) to enhance both region-level and boundary-level accuracy. Through the experiments on the open dataset, our proposed model achieved a mean DSC of 88.98%, precision of 91.0%, and recall of 94.4%, the highest among compared approaches on the NIH dataset, outperforming the baselines such as U-Net and standard-loss TransUNet; however, differences in preprocessing protocols should be considered when making direct comparisons with prior work. To assess generalizability beyond the training domain, we further evaluated the NIH-trained model on the BTCV Multi-Organ Segmentation dataset, a completely different institution’s CT dataset, using inference-time adaptation strategies without any fine-tuning, achieving a mean DSC of 62.96% in a zero-shot cross-dataset setting. A fully automatic end-to-end pipeline where a Faster R-CNN detector fine-tuned on BTCV training cases predicts bounding boxes used for cropping was evaluated using 5-fold cross-validation on all 30 BTCV cases and achieved a mean DSC of 66.50% ± 8.55%, with no manual annotation used at any stage. Full article

Bone densitometry in osteoporosis diagnosis via dual-energy X-ray absorptiometry (DXA) can benefit from advances in imaging detector technology. We devised a compact imaging scanner—DXA4A—using a photon-counting and energy-sensitive Timepix4 hybrid pixel detector (512 × 448 pixels, 55 µm pitch), for areal bone mineral density (aBMD) assessments in the distal radius and tibia in the clinic and for future in-flight astronauts’ bone health assessment. We present the design and Monte Carlo simulations of the scanner. A Timepix4 detector with a 1 mm thick CdTe sensor was tested in the laboratory with X-ray tube sources, acquiring first images of test samples. Monte Carlo simulations were implemented for scanner design and performance prediction, using 50 kVp unfiltered and 100 kVp Sm K-edge filtered spectra. With a digital twin of the scanner and patient wrist, we set up a virtual imaging study and determined the aBMD in the forearm of a patient (0.515 ± 0.048 g/cm²), in agreement with the clinical DXA value (0.571 g/cm² for the total forearm). This study highlights the feasibility of realizing a compact DXA scanner for the distal tibia and radius with spectral capabilities, exploiting Timepix4 hybrid detectors for its peculiar energy sensitivity and photon event timing properties for tissue identification. Full article

Forward-looking sonar (FLS) target detection is essential for autonomous underwater vehicles (AUVs), yet its effectiveness is severely hindered by complex acoustic distortions, environmental volatility and the scarcity of fine-annotated data, which limit the success of standard deep learning approaches. To address these challenges, this study proposes a novel quasi-static in situ learning paradigm for underwater acoustic target detection (UATD). The hybrid methodology integrates scene priors into a lightweight deep learning detector by incorporating explicit probability weighting based on echo-intensity statistics and acoustic attenuation compensation. By using these models for pixel-wise image enhancement and fusing statistical descriptors with deep learning predictions at the score level, the framework dynamically adapts to in situ environmental contexts during quasi-static operational tranches. Experimental evaluations on the UATD dataset demonstrate that this in situ adaptation significantly enhances overall detection performance, achieving an F1-score of 0.865 for our approach, an 8.1% improvement over the baseline YOLOv12n, with only a 2.1 G increase in FLOPs, while outperforming YOLOv12x (F1 = 0.844) with 95% fewer FLOPs. Ultimately, this paradigm overcomes the limitations of purely deep learning-based methods, offering a robust and interpretable solution tailored for practical AUV deployment. Full article

Balancing small object detection performance with model lightweighting remains a critical challenge in the remote sensing domain. To address the massive computational and parameter overhead of existing algorithms, we propose the hybrid scale dynamic detection transformer (HSD-DETR). This lightweight detector incorporates four core innovations to effectively enhance feature extraction for small objects. First, to reduce costs without compromising performance, we design a hybrid convolution and selective scanning fusion (HCSS-Fusion) module to reconstruct the backbone, combining local convolution with global linear scanning. Second, to preserve fine-grained information, we introduce a space-to-depth mixer (SPDMixer) to achieve pixel-level lossless downsampling. Third, to mitigate background interference and enhance small object representation, we develop a dynamic sparse adaptive intra-scale feature interaction (DSAIFI) module, employing a gating mechanism to dynamically select informative spatial tokens. Finally, to improve the localization precision for small objects, we propose the rational-focal minimum point distance intersection over union (RF-MPDIoU) loss, utilizing a non-linear mapping to dynamically modulate sample weights. Experimental results on public benchmarks confirmed that, compared to mainstream models, HSD-DETR achieves highly competitive accuracy while significantly reducing parameter scale and theoretical computational complexity. Ultimately, this research provides a lightweight and robust algorithmic solution for the field of remote sensing object detection. Full article

As synchrotron radiation sources (SRSs) expand to cover a broader energy range, the demand for hybrid detectors with improved spatial and energy resolution is increasing. This paper presents the design and characterization of a prototype pixel readout ASIC featuring a small pixel size and low noise, developed for low energy soft X-ray applications. This chip adopts the single photon-counting (SPC) approach and each pixel consists of a front-end amplifier, a discriminator, a charge injection circuitry and a pair of 15-bit counters with associated logic. Fabricated in a 130 nm CMOS process, the chip integrates a 2 × 16 pixel matrix with a 50 µm ×50 µm pixel size. Measurement results indicate the maximum pixel equivalent noise charge (ENC) across the matrix is 20 e⁻rms without sensor attached. The results validate that the chip design has the potential to deliver a low-energy resolution for soft X-ray applications. Full article

This study addresses the problem of robust video-based tracking of laboratory rats in open-field and Y-maze experiments under challenging acquisition conditions, including non-uniform illumination, low contrast, and heterogeneous recording setups. Existing approaches based on classical image processing or deep learning often fail to maintain stable localization under such conditions or require large, annotated datasets. We propose a hybrid tracking framework that combines an improved motion–appearance voting mechanism with consistency-constrained optimization for open-field experiments, together with a comparative deep learning-based detection strategy for Y-maze analysis. The proposed method introduces (i) adaptive dual-threshold motion extraction, (ii) directionally constrained temporal validation, and (iii) a robustness-driven fusion of motion and appearance cues. Experimental results demonstrate that the proposed approach achieves reliable tracking with a maximum localization error below 10 pixels under severe illumination variations. In the Y-maze scenario, a comparative evaluation of multiple detectors (YOLOv5, YOLOv9, YOLO12, Faster R-CNN) highlights the trade-off between accuracy and inference time, with YOLOv9 providing the best balance. The main contribution consists of enabling robust behavioral quantification in low-quality experimental conditions using limited training data, bridging the gap between classical tracking robustness and deep learning flexibility. Full article

Drones have proven effective for acquiring aerial imagery, and when equipped with onboard analysis tools, they can automatically identify objects of interest. Neural-network methods for image analysis typically require large training datasets and substantial computational resources. By contrast, algorithmic techniques can detect objects using simple features, such as pixel colors, thereby reducing the need for extensive training and computational resources. Once trained, both types of system can analyze images in a short time. In our experiments, each approach has distinct strengths. The YOLO-based detector is more accurate for complex-shaped objects, such as trees, whereas the pixel-color approach performs better on sparser objects. This paper proposes YOLO-C3, a hybrid system designed for onboard drone image processing. By leveraging the strengths of both YOLO-based and pixel-based approaches, YOLO-C3 balances detection accuracy with estimation confidence. Trained on Mediterranean imagery dataset, the system is optimized for identifying natural objects, including citrus groves and trees. To assess the robustness of the image classifier, a K-fold cross-validation is performed. Compared to existing models, YOLO-C3 detects a wider range of natural objects with high accuracy and minimal latency, achieving a processing speed of 0.01 s per image. By performing object detection locally, drones can adapt their trajectories to support emergency response, helping to map safe corridors and locate buildings where people may be awaiting rescue after a natural disaster. Full article

Early detection of agricultural fires with Unmanned Aerial Vehicles (UAVs) is important for environmental safety, yet it remains difficult because ignition cues are extremely small, smoke patterns vary widely, and farmland scenes often contain strong background interference such as specular reflections. Model development is further constrained by the scarcity of data from the early ignition stage. To address these challenges, we propose a joint data and model optimization framework. We first build a hybrid dataset through an ROI-guided synthesis pipeline, in which latent diffusion models are used to insert high-fidelity, carefully screened fire samples into real farmland backgrounds. We then introduce EF-YOLO, a detector designed for high sensitivity to small targets. The network uses SPD-Conv to reduce feature loss during spatial downsampling and includes a high-resolution P2 head to improve the detection of minute objects. To reduce background clutter, a Dual-Path Frequency–Spatial Enhancement (DP-FSE) module serves as a lightweight statistical surrogate that extracts global contextual cues and local salient features in parallel, thereby suppressing high-frequency noise. Experimental results show that EF-YOLO achieves an ${AP}_S$ of 40.2% on sub-pixel targets, exceeding the YOLOv8s baseline by 15.4 percentage points. With a recall of 88.7% and a real-time inference speed of 78 FPS, the proposed framework offers a strong balance between detection performance and efficiency, making it well suited for edge-deployed agricultural fire early-warning systems. Full article

To meet the requirements of quantitative elemental analysis in the ultraviolet (UV) spectrum, a UV single-photon imaging system was developed, integrating a digital micromirror device (DMD) and a single photon-counting imaging detector, enabling high sensitivity, high resolution, and a wide dynamic range. However, intrinsic geometric distortion poses a significant challenge to accurate spectral calibration. A hybrid correction framework is proposed, cascading polynomial coarse correction with multilayer perceptron (MLP) fine regression, improving calibration accuracy. The method utilizes a full-field dot-array mask projected by the DMD to acquire distortion-reference image pairs. The polynomial model rapidly captures the dominant high-order distortion, while a lightweight MLP performs non-parametric fine regression of residual displacements, achieving a mean error of 0.84 pixels. This approach reduces the root mean square (RMS) error to 1.01 pixels, outperforming traditional direct linear transformation (5.35 pixels) and pure polynomial models (1.33 pixels), while the nonlinearity index decreases from 0.35° to 0.05°. In addition, the method demonstrates stable performance across multi-scale checkerboard patterns ranging from 128 to 280 pixels, with RMS errors remaining around the 1-pixel level. These results validate the high-precision distortion suppression and robust cross-scale performance of the proposed framework. By leveraging DMD-generated patterns for self-calibration, this method eliminates the need for external targets, offering a scalable solution for high-end spectrometer calibration. Full article

This work presents a simulation study of a Timepix3 telescope composed of nine detectors for use as a Compton scatter polarimeter in the energy range of 35–100 keV. Four detectors carry 1 mm thick silicon (Si) sensors and five detectors carry 1 mm thick cadmium telluride (CdTe) sensors. The modulation factor for 100% linearly polarized X-ray beams was found to be ${\mu }_{100}>70\%$ in the energy range of 55–80 keV. The quality factor of the polarimeter has its maximum 12.8% at the energy 75 keV. The comparison of quality factors and the calculations of a hypothetical observation of the Crab nebula show that this multilayer Timepix3 approach is competitive with contemporary X-ray polarimeters. Full article

Precise recognition and localization of electronic components on printed circuit boards (PCBs) are crucial for industrial automation tasks, including robotic disassembly, high-precision assembly, and quality inspection. However, strong visual interference from silkscreen characters, copper traces, solder pads, and densely packed small components often degrades the accuracy of deep learning-based detectors, particularly under complex industrial imaging conditions. This paper presents a two-stage, coarse-to-fine PCB component localization framework based on an optimized YOLOv11 architecture and a sub-pixel geometric refinement module. The proposed method enhances the backbone with a Convolutional Block Attention Module (CBAM) to suppress background noise and strengthen discriminative features. It also integrates a tiny-object detection branch and a weighted Bi-directional Feature Pyramid Network (BiFPN) for more effective multi-scale feature fusion, and it employs a customized hybrid loss with vertex-offset supervision to enable pose-aware bounding box regression. In the second stage, the coarse predictions guide contour-based sub-pixel fitting using template geometry to achieve industrial-grade precision. Experiments show significant improvements over baseline YOLOv11, particularly for small and densely arranged components, indicating that the proposed approach meets the stringent requirements of industrial robotic disassembly. Full article