Local Detectors Research Articles

Wind tunnel and flight testing of a lamb wave-based ice accretion sensor

Abstract Severe icing conditions are still a risk in aviation. Heavy ice accretion on aircraft can affect mass and aerodynamic performance in a way, that a catastrophic loss of control may occur. Detecting icing conditions and the extent of ice accretion is therefore essential to react instantly and to assess the severity. This eases the decision to either activate ice protection systems and continue the flight or to leave the icing conditions immediately. These functions can be provided by an icing sensor. Within the EU funded SENS4ICE project, several ice sensors have been developed. One of them is the so-called local ice layer detector (LILD). It uses lamb waves which are guided by the aircraft structure. If an ice accretion happens, the wave guide behavior is affected changing amplitude and lag time of the lamb wave pulses. This can be measured. If the sensor is mounted to a surface on the aircraft which is prone to icing, e.g. the leading edge of the wing, ice can be detected where it is relevant for the aerodynamic performance. Within the project, the sensor hard- and software have been developed to create and measure lamb wave pulses in a wide frequency range. To investigate the influence of ice accretion on the lamb waves, wind tunnel tests were undertaken. Finally, the sensor was investigated in a flight test. In this paper, an overview about the function of the sensor as well as the general signal behavior are given. The results of the wind tunnel test and some preliminary results from the flight tests will be shown. In both tests, the sensor was able to detect even very thin ice layers with minimum delay from the beginning of the ice accretion.

Enhanced Cooperative Compressive Spectrum Sensing in Cognitive Radio Networks

ABSTRACTThe demand for bandwidth in cognitive radio networks (CRNs) is growing as applications become increasingly data intensive. Techniques such as compressed spectrum sensing (CSS) and cooperative spectrum sensing (C‐SS) are employed to address this challenge. C‐SS enhances overall detection accuracy and reliability by enabling multiple nodes to share and combine their local sensing data. Conversely, CSS effectively reduces the required information for spectrum usage decision making, thereby improving bandwidth utilization. Integrating these two methods allows CRNs to utilize the spectrum reliably and efficiently, leading to increased spectral efficiency. To further improve reconstruction performance, we leverage the sparsity concept to transcend hardware constraints and merge restrictions from both real and synthesized channels. This approach involves the virtual synthesis of channels, which linearly enhances the signal‐to‐noise ratio (SNR) within the network's size range. Simulation results demonstrate that our proposed method offers significant advantages over single‐node recovery, as validated by simulations and software defined radio (SDR) Implementation. The integration of spectral estimations from various local CR detectors enhances spatial diversity gain and sensing quality, particularly in fading channels. Compared to traditional approaches, our method achieves superior performance, evidenced by an increase in (from 93.97% to 96.52%) with almost the same .

A high-efficiency local and global detector for diatom-based drowning diagnosis

Detection and classification of diatoms has been considered as the most effective method for the drowning diagnosis in forensic practice. However, challenges (such as limited samples, complex background interference, and detecting multiple objects) remain during detection and classification. To address these challenges, we propose a MDMS (multi-scale dynamic multi-head self-attention)-based deep learning network for the diatom-based drowning diagnosis. This framework uses TFFT (two-stage full fine-tuning training strategy) to solve the challenges of limited diatom samples and detecting multiple objects, and builds a multi-scale dynamic multi-head self-attention module to extract the local and global features of diatom images. In the meantime, we introduce an online hard example mining strategy to attenuate the complex background interference. The experimental results show that the proposed framework can effectively reduce the missing and false detection rates of diatom objects, with the mAP (mean Average Precision) reaching 92.434%, which is better than the result using the mainstream methods.

Particle detectors under chronological hazard

We analyze how the presence of closed timelike curves (CTCs) characterizing a time machine can be discerned by placing a local particle detector in a region of spacetime which is causally disconnected from the CTCs. Our study shows that not only can the detector tell if there are CTCs, but also that the detector can separate topological from geometrical information and distinguish periodic spacetimes without CTCs (like the Einstein cylinder), curvature, and spacetimes with topological identifications that enable time-machines.

Analysisof Beam Walk in Inter-Satellite Laser Link: Implications for Differential Wavefront Sensing in Gravitational Wave Detection

Achieving space-based gravitational wave detection requires the establishment of an interferometer constellation. It is necessary to establish and maintain stable laser interferometric links using the differential wavefront sensing (DWS) technnique. When the distant measurement beam experiences pointing jitter, it causes beam walk on the surface of the local detector. The reduced overlap between the local reference spot and the distant spot increases the nonlinear errors in the DWS technique, which need to be suppressed. Numerical analysis was conducted on the spatial beam interference signals of the DWS technique when the distant measurement beam experienced pointing jitter. An experimental measurement system was designed, and the beam walk was suppressed using a conjugate imaging system. The results show that within a range of 300 μrad, the optical path with the imaging system can reduce measurement errors by at least 83%. This way also helps to reduce pointing jitter noise in inter-satellite links, thereby improving laser pointing control accuracy.This method would provide a valuable reference for future DWS measurement systems.

Third sound detectors in accelerated motion

An accelerated observer moving through empty space sees particles appearing and disappearing, while an observer with a constant velocity does not register any particles. This phenomenon, generally known as the Unruh effect, relies on an initial vacuum state, thereby unifying the experience of all inertial observers. We propose an experiment to probe this observer-dependent detector response, using a laser beam in circular motion as a local detector of superfluid helium-4 surface modes or third sound waves. To assess experimental feasibility, we develop a theoretical framework to include a non-zero temperature initial state. We find that an acceleration-dependent signal persists, independent of the initial temperature. By introducing a signal-to-noise measure we show that observing this signal is within experimental reach.

Deep Anomaly Detection Framework Utilizing Federated Learning for Electricity Theft Zero-Day Cyberattacks.

Smart power grids suffer from electricity theft cyber-attacks, where malicious consumers compromise their smart meters (SMs) to downscale the reported electricity consumption readings. This problem costs electric utility companies worldwide considerable financial burdens and threatens power grid stability. Therefore, several machine learning (ML)-based solutions have been proposed to detect electricity theft; however, they have limitations. First, most existing works employ supervised learning that requires the availability of labeled datasets of benign and malicious electricity usage samples. Unfortunately, this approach is not practical due to the scarcity of real malicious electricity usage samples. Moreover, training a supervised detector on specific cyberattack scenarios results in a robust detector against those attacks, but it might fail to detect new attack scenarios. Second, although a few works investigated anomaly detectors for electricity theft, none of the existing works addressed consumers' privacy. To address these limitations, in this paper, we propose a comprehensive federated learning (FL)-based deep anomaly detection framework tailored for practical, reliable, and privacy-preserving energy theft detection. In our proposed framework, consumers train local deep autoencoder-based detectors on their private electricity usage data and only share their trained detectors' parameters with an EUC aggregation server to iteratively build a global anomaly detector. Our extensive experimental results not only demonstrate the superior performance of our anomaly detector compared to the supervised detectors but also the capability of our proposed FL-based anomaly detector to accurately detect zero-day attacks of electricity theft while preserving consumers' privacy.

STUDY OF LOCAL IMAGE FEATURES DETECTORS

The article is devoted to the optimization of local image feature detectors. They can be used in mobile augmented reality video information systems. A study was conducted of the effectiveness of using a brightness preliminary detector in conjunction with a Harris angle detector. The simulation results showed a reduction in the duration of detection of local features of the test image (on average 220 times), a reduction in the number of identified image features and a slight change in the threshold value of the angle response.

Micro LED defect detection with self-attention mechanism-based neural network

We propose a method utilizing a YOLO detector for the precise localization of defective chips and the identification of defect types within multi-scale multi-target images. To address the challenge of optimizing training costs and enhancing model generalization, we introduce an end-to-end deep neural network, CM-YOLOv5, specifically designed for chip detection. We incorporate a novel bottleneck layer, MA-CSP, in conjunction with Multi-Head Self-Attention mechanism (MHSA). Additionally, we propose a class-balanced loss function (CB-BCE Loss) to tackle the issue of uneven distribution of defective samples in the Micro LED dataset. To further enhance convergence speed and detection precision, we introduce the SIoU Loss combined with Meta-AconC. Our experimental results, conducted on the Micro LED dataset, demonstrate notable improvements with CM-YOLOv5 over the basic YOLOv5 algorithm. Specifically, CM-YOLOv5 exhibits a 3.8 % increase in mean average precision and a 3.7 % improvement in precision, surpassing current mainstream object detection algorithms, including YOLOR, YOLOX, and YOLOv6, etc., in terms of general evaluation metrics. Finally, upon deploying our proposed algorithm on the edge device NVIDIA Jetson Xavier NX, CM-YOLOv5 demonstrates commendable speed and detection performance in embedded scenarios.

A Tracking-Based Two-Stage Framework for Spatio-Temporal Action Detection

Spatio-temporal action detection (STAD) is a task receiving widespread attention and has numerous application scenarios, such as video surveillance and smart education. Current studies follow a localization-based two-stage detection paradigm, which exploits a person detector for action localization and a feature processing model with a classifier for action classification. However, many issues occur due to the imbalance between task settings and model complexity in STAD. Firstly, the model complexity of heavy offline person detectors adds to the inference overhead. Secondly, the frame-level actor proposals are incompatible with the video-level feature aggregation and Region-of-Interest feature pooling in action classification, which limits the detection performance under diverse action motions and results in low detection accuracy. In this paper, we propose a tracking-based two-stage spatio-temporal action detection framework called TrAD. The key idea of TrAD is to build video-level consistency and reduce model complexity in our STAD framework by generating action track proposals among multiple video frames instead of actor proposals in a single frame. In particular, we utilize tailored tracking to simulate the behavior of human cognitive actions and used the captured motion trajectories as video-level proposals. We then integrate a proposal scaling method and a feature aggregation module into action classification to enhance feature pooling for detected tracks. Evaluations in the AVA dataset demonstrate that TrAD achieves SOTA performance with 29.7 mAP, while also facilitating a 58% reduction in overall computation compared to SlowFast.

Development of PET Detector for Localization Using MLPE Based on Simulation Data

Development of PET Detector for Localization Using MLPE Based on Simulation Data

Semi-supervised object detection based on single-stage detector for thighbone fracture localization

Semi-supervised object detection based on single-stage detector for thighbone fracture localization

Series Arc Fault Identification in DC Distribution Based on Random Forest Predicted Probability

Series dc arc fault generates arcing noise cross-talk to adjacent electrical loads, making it challenging to identify the arc fault location, which creates a fire hazard that endangers the system’s safety. This study proposes a series arc fault identification method in a dc zonal electrical distribution using Random Forest based local detectors to monitor constant power loads, and output predicted nominal and arc fault probabilities. The predicted probabilities are then sent to a centralized master detector to obtain the final decision. With full communication capability among all local detectors, the master detector makes the final decision using Random Forest algorithms. If there is any disconnection in the communication links, the local detector can operate independently by using the predicted arc fault probability to output the flag signal autonomously, while the master detector continues to operate with other local detectors. The proposed fault identification method is experimentally verified with a zonal electrical distribution testbed that comprises three constant power loads.

Comparison of Tiling Artifact Removal Methods in Secondary Ion Mass Spectrometry Images.

Time-of-flight secondary ion mass spectrometry (ToF-SIMS) imaging is used across many fields for the atomic and molecular characterization of surfaces, with both high sensitivity and high spatial resolution. When large analysis areas are required, standard ToF-SIMS instruments allow for the acquisition of adjoining tiles, which are acquired by rastering the primary ion beam. For such large area scans, tiling artifacts are a ubiquitous challenge, manifesting as intensity gradients across each tile and/or sudden changes in intensity between tiles. Such artifacts are thought to be related to a combination of sample charging, local detector sensitivity issues, and misalignment of the primary ion gun, among other instrumental factors. In this work, we investigated six different computational tiling artifact removal methods: tensor decomposition, multiplicative linear correction, linear discriminant analysis, seamless stitching, simple averaging, and simple interpolating. To ensure robustness in the study, we applied these methods to three hyperspectral ToF-SIMS data sets and one OrbiTrapSIMS data set. Our study includes a carefully designed statistical analysis and a quantitative survey that subjectively assessed the quality of the various methods employed. Our results demonstrate that while certain methods are useful and preferred more often, no one particular approach can be considered universally acceptable and that the effectiveness of the artifact removal method is strongly dependent on the particulars of the data set analyzed. As examples, the multiplicative linear correction and seamless stitching methods tended to score more highly on the subjective survey; however, for some data sets, this led to the introduction of new artifacts. In contrast, simple averaging and interpolation methods scored subjectively poorly on the biological data set, but more highly on the microarray data sets. We discuss and explore these findings in depth and present general recommendations given our findings to conclude the work.

Magnon Confinement in an All-on-Chip YIG Cavity Resonator Using Hybrid YIG/Py Magnon Barriers.

Confining magnons in cavities can introduce new functionalities to magnonic devices, enabling future magnonic structures to emulate the established photonic and electronic components. As a proof-of-concept, we report magnon confinement in a lithographically defined all-on-chip YIG cavity created between two YIG/Permalloy bilayers. We take advantage of the modified magnetic properties of the covered/uncovered YIG film to define on-chip distinct regions with boundaries capable of confining magnons. We confirm this by measuring multiple spin-pumping voltage peaks in a 400 nm wide platinum strip placed along the center of the cavity. These peaks coincide with multiple spin-wave resonance modes calculated for a YIG slab with the corresponding geometry. The fabrication of micrometer-sized YIG cavities following this technique represents a new approach to control coherent magnons, while the spin-pumping voltage in a nanometer-sized Pt strip demonstrates to be a noninvasive local detector of the magnon resonance intensity.

Global and Local Context-Aware Ship Detector for High-Resolution SAR Images

Due to the large sizes of synthetic aperture radar (SAR) images, traditional deep learning-based ship detection methods usually utilize the sliding window preprocessing strategy to obtain the small-sized sub-images. However, there are amounts of background clutter areas without ships in SAR images. Thus, traditional sliding window-based methods may generate numerous sub-images without ships, which can bring high computation redundancy and numerous false alarms. To deal with the above problems, in this paper, a novel detection method called global and local context-aware ship detector (GLC-Det) for high-resolution SAR images is proposed. The proposed method mainly contains three parts: the global context-aware based sub-images selection (GCSS) module, the deep learning module, and the local context-aware based false alarms suppression (LCFS) module. The GCSS module employs the global context information to eliminate the sub-images without ships, which can enhance detection efficiency and avoid generating numerous false alarms. The deep learning module is used to further obtain the preliminary detection boxes. The LCFS module is a postprocessing step, which employs the local context information around the detection boxes to further eliminate the false alarms. The experimental results on the measured data of AIR-SARShip-1.0 and 2.0 high-resolution SAR images demonstrate that the proposed method has higher precision and efficiency than the original deep learning methods.

Intensity Tomography Method for Secure and High-Performance Quantum Key Distribution

Decoy-state method has greatly filled in loopholes from multi-photon pulses in the field of quantum key distribution (QKD) and it has been employed in almost all practical QKD systems in recent years. However, security and performance of the practical decoy-state QKDs are still limited by inaccurate modulations such as random intensity fluctuation and correlated intensity fluctuation. In this work, an intensity tomography method is proposed as a countermeasure against these fluctuations. It has an ability to tightly bound the unknown density matrices of photon-number state and accurately calibrate the fluctuations by measuring the decoy states with a local single-photon detector. We verified the tomography method by experiments and simulations. The results indicated that our method can efficiently mitigate the flaws from these fluctuations to improve the protocol performance and practical security. Noting that these fluctuations are inevitable in practical QKDs, our method would pave the avenue for QKD's practical applications.

Deep Corner

Recent studies have shown promising results on joint learning of local feature detectors and descriptors. To address the lack of ground-truth keypoint supervision, previous methods mainly inject appropriate knowledge about keypoint attributes into the network to facilitate model learning. In this paper, inspired by traditional corner detectors, we develop an end-to-end deep network, named Deep Corner, which adds a local similarity-based keypoint measure into a plain convolutional network. Deep Corner enables finding reliable keypoints and thus benefits the learning of the distinctive descriptors. Moreover, to improve keypoint localization, we first study previous multi-level keypoint detection strategies and then develop a multi-level U-Net architecture, where the similarity of features at multiple levels can be exploited effectively. Finally, to improve the invariance of descriptors, we propose a feature self-transformation operation, which transforms the learned features adaptively according to the specific local information. The experimental results on several tasks and comprehensive ablation studies demonstrate the effectiveness of our method and the involved components.

Cooperative Time-Frequency Localization for Wideband Spectrum Sensing With A Lightweight Detector

With the ability to solve the hidden terminal problem, cooperative spectrum sensing (CSS) in cognitive radio networks has attracted much attention. However, existing CSS mainly focuses on detecting the spectrum usage status of primary users (PUs) in time dimension under the premise of known center frequency and bandwidth of PU, which is limited in practical scenarios. For the emerging time-frequency localization based two-dimensional wideband spectrum sensing (SS), how to incorporate cooperation into secondary users (SUs) to enhance sensing performance is still a challenging problem. In this letter, considering the limited storage and computation capacity of SUs, we first propose a lightweight detector for time-frequency localization. Then, through analyzing the characteristics of the sensed spectrogram, we propose a non-maximum suppression based fusion algorithm. Finally, simulation results validate the advantages of the proposed lightweight detector compared with existing detectors, and show the recognition performance of CSS is superior to that of the individual SUs.

Traffic Sign Detection and Recognition Using Multi-Frame Embedding of Video-Log Images

The detection and recognition of traffic signs is an essential component of intelligent vehicle perception systems, which use on-board cameras to sense traffic sign information. Unfortunately, issues such as long-tailed distribution, occlusion, and deformation greatly decrease the detector’s performance. In this research, YOLOv5 is used as a single classification detector for traffic sign localization. Afterwards, we propose a hierarchical classification model (HCM) for the specific classification, which significantly reduces the degree of imbalance between classes without changing the sample size. To cope with the shortcomings of a single image, a training-free multi-frame information integration module (MIM) was constructed, which can extract the detection sequence of traffic signs based on the embedding generated by the HCM. The extracted temporal detection information is used for the redefinition of categories and confidence. At last, this research performed detection and recognition of the full class on two publicly available datasets, TT100K and ONCE. Experimental results show that the HCM-improved YOLOv5 has a mAP of 79.0 in full classes, which exceeds that of state-of-the-art methods, and achieves an inference speed of 22.7 FPS. In addition, MIM further improves model performance by integrating multi-frame information while only slightly increasing computational resource consumption.