Search Results (31)

Urban traffic anomaly detection is essential for intelligent transportation systems, particularly in smart city environments where fast identification of abnormal events can improve road safety and traffic management. This work proposes a novel ULSTM-driven architecture that explicitly models temporal dependencies across consecutive traffic frames to achieve more stable and temporally coherent reconstructions. The proposed framework leverages sequential spatio-temporal representations to improve the distinction between normal traffic patterns and anomalous events. To further enhance reliability, we introduce a Hybrid Weighted Fusion strategy that synergistically combines structural, perceptual and pixel-wise metrics. The framework’s parameters are optimized using a Discrete Dirichlet Sampling approach, achieving a peak F1 Score of 70.28%. Evaluations were conducted on a manually curated traffic anomaly dataset with frame-level annotations. Experimental results demonstrate that the ULSTM framework significantly outperforms frame-independent generative models by suppressing high-frequency reconstruction noise, providing a robust solution for real-world smart city deployments. While highly effective in complex scenarios, the proposed framework is strictly applicable to highly dynamic traffic environments with active motion, as static background ensembles can degrade performance. Full article

Global Navigation Satellite System Reflectometry (GNSS-R) platforms, such as the CYGNSS constellation, provide unprecedented spatial coverage for monitoring ionospheric scintillation via the $S_4$ index. However, the operational utility of GNSS-R for space weather is substantially degraded by surface-induced signal contamination when sharp land–water boundaries (coastlines) trigger massive, false-positive $S_4$ pseudo-scintillations that imitate true ionospheric plasma irregularities. In this study, a robust machine learning (ML) methodology to autonomously distinguish surface-induced reflections from true atmospheric volumetric scattering was proposed. Using 1 Hz Level 1 continuous Signal-to-Noise Ratio (SNR) time-series data, morphologic features (e.g., maximum amplitude, peak prominence, and standard deviation) were extracted to train a Random Forest (RF) classifier. The model achieves 98% accuracy in differentiating coastal boundaries from ionospheric scintillation, evaluated on a global dataset of over ~450,000 anomalous events. Moreover, a multi-sensor case study of the historic May 2024 G5 geomagnetic storm is presented to validate the geophysical fidelity of the filtered data. The ML-isolated CYGNSS anomalies demonstrate strong spatial correlation with COSMIC-2 Radio Occultation (RO) F2-peak electron density (NmF2) variations and ground-based Rate of TEC Index (ROTI) maps. Furthermore, temporal cross-validation with 1 Hz localized ground magnetometer data in Northwest Mexico reveals positive synchronization between CYGNSS scattering events and localized electrodynamic disturbances. Finally, the results demonstrate that ML-de-aliased GNSS-R data can reliably link the oceanic observational gaps inherent to ground-based networks, offering a powerful new tool for global space weather monitoring. Full article

Seismic-while-tunneling signals acquired during coal-mine excavation are typically characterized by strong nonstationarity, intense mechanical noise, weak reflection responses, unstable inter-trace phases, and complex waveform fluctuations. These characteristics make conventional energy- or amplitude-based picking methods susceptible to false triggers and missed detections. To reveal the local complexity mutation of mine seismic signals under strong-noise backgrounds, this study proposes a multiscale fractal-dimension-constrained coherent phase processing method for signal enhancement, first-arrival picking, and fault prediction. First, the raw seismic-while-tunneling records are reorganized into shot gathers, windowed, and downsampled to preserve the effective early-arrival information. A damped multichannel singular spectrum analysis method is then used to extract coherent low-rank components and suppress incoherent random noise. Second, short-window and long-window box-counting fractal dimensions are calculated to characterize local and background waveform complexity, and a fractal-dimension mutation index is constructed to identify abrupt complexity transitions associated with effective seismic arrivals. On this basis, the fractal mutation index is incorporated into a coherent phase picking function that combines multichannel phase consistency and stacked amplitude, forming a fractal-dimension-constrained CCPP detection criterion. This criterion enhances true coherent arrivals while suppressing isolated noise spikes and unstable local amplitude disturbances. Finally, phase-weighted stacking is applied to further strengthen phase-consistent reflection responses and improve the interpretability of seismic-while-tunneling imaging profiles. Field application at the WII02040503 working face of Tunbao Coal Mine demonstrates that the proposed method can effectively improve the continuity of coherent events, stabilize automatic picking results, and enhance anomalous reflection bands under complex underground noise conditions. During the engineering trial, a total of 2558 m of ahead prospecting was completed, and 29 faults were predicted. The field-confirmation rates of the predicted faults with throws greater than 3 m, between 1 and 3 m, and less than 1 m were 100%, 87.50%, and 81.25%, respectively. Overall, 25 of the 29 predicted faults were confirmed by field exposure, corresponding to an overall field-confirmation rate of 86.21%. After velocity-synchronization time-difference correction, the average planar positioning deviation of the confirmed fault predictions decreased from 7.86 m to 5.08 m, corresponding to a 35.37% reduction in positioning error. These results indicate that the proposed fractal-dimension-constrained coherent processing framework provides an effective approach for complexity-aware signal enhancement and robust fault prediction in seismic-while-tunneling monitoring. Full article

In the realm of public safety, the automated identification of potential threats from voluminous surveillance streams is pivotal for developing intelligent security systems. Manual monitoring of such massive video feeds is highly inefficient, prone to human fatigue, and often leads to missed detections or false alarms. Leveraging deep learning for automatic anomaly detection is therefore essential to improve response efficiency and mitigate security risks. Weakly supervised video anomaly detection (WS-VAD) has emerged as a critical yet challenging task in this domain. In this study, we propose the Temporal-Enhanced and Visual-Text Adaptive Fusion (TE-VTAF) model for robust WS-VAD. Specifically, a Dynamic Local–Global Temporal Adaptive Module (DLG-TAM) is designed to capture multi-scale temporal dependencies and extract high-level video semantics. Concurrently, a Visual-Text Adaptive Fusion Module (VTAFM) is introduced to aggregate complementary cross-modal features, utilizing a competitive activation mechanism to suppress redundant information and enhance the discriminative power between normal and anomalous events. To further refine the learning process within the Multiple Instance Learning (MIL) framework, we incorporate a Top-K outer bag loss and a K-maxmin inner bag loss. These constraints effectively maximize the inter-class separability while suppressing label noise from normal instances within positive bags, thereby bolstering the detector’s robustness. Extensive experiments demonstrate that the proposed TE-VTAF consistently outperforms state-of-the-art methods on two large-scale benchmarks, achieving an AUC of 88.93% on UCF-Crime and an AP of 85.62% on XD-Violence. Full article

In automotive crash testing, trustworthy crash-test curves are essential for reliable crashworthiness assessment, yet automated anomaly detection is difficult due to limited labeled abnormal cases, event-level data scarcity, and distribution shifts across vehicle models and sensor configurations. This paper proposes MVCA-AD (Multi-View Context Augmentation for Anomaly Detection) for single-channel crash-test curves. MVCA-AD generates multiple context-rich views using deterministic time- and frequency-domain transformations to amplify subtle anomalous patterns under limited labeled supervision. A trend-aware modulation module and cross-view attention fuse these views to improve sensitivity to critical segments such as impact spikes and gradual transitions while remaining robust to noise. Experiments on three subsets derived from physical full-scale crash tests show that MVCA-AD improves Precision, Recall, F1-score, and area under the ROC curve (AUC) over strong baselines and achieves stable performance under event-level grouped evaluation across heterogeneous head and B-pillar crash-test signals. The proposed approach supports crash-test data quality control by automatically identifying abnormal curves for downstream crashworthiness assessment workflows. Full article

Since September 2021, numerous seismic events with spectral peaks below 1 Hz occurred on the island of Vulcano, Italy, 131 years after its last eruption. The local monitoring network recorded microseismicity mostly in the form of months-long swarms, concurrent with anomalous values of other geophysical and geochemical parameters. By applying a machine learning technique (Self-Organizing Maps, SOMs), we obtained an inventory of ~6600 seismic signals, identifying and separating exogenous signals (anthropic noise) from distinct families of events. These families were located below La Fossa Crater (where the last eruption of the volcano happened) from the surface to a depth of 2.2 km b.s.l. Based on the seismic signature and source location of these events, we hypothesize unsealed/sealed processes through a network of shallow fractures favored by fluid pressure. After the return to background values of geochemical and geophysical parameters in 2023, a resumption of microseismicity occurred between May and June 2024. A test application of the SOM to the new data confirmed the non-destructive source of the new recorded signals, which shared families, location, and depths with our previous inventory. This test showed that SOM can be an effective tool for supporting real-time monitoring and warning of future unrest at Vulcano. Full article

Although surface soil moisture (SSM) is particularly important in crop yield prediction, irrigation scheduling optimization, and runoff generation mechanisms, accurate monitoring of time-series SSM is still challenging for agricultural and hydrological research. This study presented an improved approach integrating Sentinel-1 C-band SAR and MODIS optical data (2019–2020) to estimate surface soil moisture. To address vegetation effects, we developed a piecewise function using fractional vegetation coverage (FVC) to correct soil moisture and backscatter extrema and established the normalized difference enhanced vegetation index (NDEVI) to characterize backscatter-vegetation relationships across various land covers. Furthermore, the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm identified anomalous surface changes, enabling segmentation of long-term series into invariant periods that satisfy the change detection method assumptions. Validation in the Shandian River Basin demonstrated significant improvement over traditional methods, achieving determination coefficients (R²) of 0.844 and root mean square errors (RMSE) of 0.030 m³/m³. The method effectively captured soil moisture dynamics from precipitation and irrigation events, providing reliable monitoring in heterogeneous landscapes. This integrated approach offers a robust technical framework for multi-source remote sensing of soil moisture in semi-arid areas, enhancing capability for agricultural water resource management. Full article

Traffic anomaly detection plays a crucial role in improving road safety and enabling timely responses to abnormal events. Recent research has explored generative and predictive models to enhance detection accuracy; however, the dynamic and complex nature of traffic scenes often introduces noise and uncertainty, reducing reliability. This work presents TGU-Net, a Temporal Generative U-Net framework designed for real-time traffic anomaly detection in urban environments. The proposed model integrates two key innovations: (1) a temporal modeling component that captures dependencies across consecutive frames, and (2) contextual scene enrichment that enhances the distinction between normal and anomalous behaviors. These additions mitigate reconstruction noise and improve detection robustness without compromising computational efficiency. Experimental evaluations on a synthetically generated CARLA-based dataset demonstrate that TGU-Net achieves strong performance in precision, recall, and early anomaly detection, confirming its potential as a scalable and reliable framework for real-world traffic monitoring systems. Full article

Background: Lumbar disc herniation (LDH) recurrence remains a pressing clinical challenge, with limited predictive tools available to support early identification and personalized intervention. Predicting recurrence after lumbar disc herniation (LDH) remains clinically important but algorithmically difficult due to extreme class imbalance and low signal-to-noise ratio. Objective: This study proposes a hybrid machine learning framework that integrates supervised classifiers, unsupervised anomaly detection, and decision threshold tuning to predict LDH recurrence using routine clinical data. Methods: A dataset of 977 patients from a Romanian neurosurgical center was used. We trained a deep neural network, random forest, and an autoencoder (trained only on non-recurrence cases) to model baseline and anomalous patterns. Their outputs were stacked into a meta-classifier and optimized via sensitivity-focused threshold tuning. Evaluation was performed via stratified cross-validation and external holdout testing. Results: Baseline models achieved high accuracy but failed to recall recurrence cases (0% sensitivity). The proposed ensemble reached 100% recall internally with a threshold of 0.05. Key predictors included hospital stay duration, L4–L5 herniation, obesity, and hypertension. However, external holdout performance dropped to 0% recall, revealing poor generalization. Conclusions: The ensemble approach enhances detection of rare recurrence cases under internal validation but exhibits poor external performance, emphasizing the challenge of rare-event modeling in clinical datasets. Future work should prioritize external validation, longitudinal modeling, and interpretability to ensure clinical adoption. Full article

This study reports a method for the detection of mechanical signaling anomalies in cardiac tissue through the use of deep learning and the design of two anomaly detectors. In contrast to anomaly classifiers, anomaly detectors allow accurate identification of the time position of the anomaly. The first detector used a recurrent neural network (RNN) of long short-term memory (LSTM) type, while the second used an autoencoder. Mechanical contraction data present several challanges, including high presence of noise due to the biological variability in the contraction response, noise introduced by the data acquisition chain and a wide variety of anomalies. Therefore, we present a robust deep-learning-based anomaly detection framework that addresses these main issues, which are difficult to address with standard unsupervised learning techniques. For the time series recording, an experimental model was designed in which signals of cardiac mechanical contraction (right and left atria) of a CD-1 mouse could be acquired in an automatic organ bath, reproducing the physiological conditions. In order to train the anomaly detection models and validate their performance, a database of synthetic signals was designed (n = 800 signals), including a wide range of anomalous events observed in the experimental recordings. The detector based on the LSTM neural network was the most accurate. The performance of this detector was assessed by means of experimental mechanical recordings of cardiac tissue of the right and left atria. Full article

Nowadays, there has been a growing interest in understanding earthquake forerunners, i.e., anomalous variations that are possibly associated with the complex process of earthquake evolution. In this context, the Robust Satellite Technique was coupled with 10 years (2012–2021) of daily night-time MODIS-Land Surface Temperature remote sensing data to detect thermal anomalies likely related to the 27 September 2021, strong onshore earthquake of magnitude Mw6.0 occurring near the Arkalochori village in Central Crete, Greece. Eight intense (signal-to-noise ratio > 3) and infrequent, quite extensive, and temporally persistent thermal signal transients were detected and characterized as pre-seismic anomalies, while one thermal signal transient was identified as a co-seismic effect on the day of the main tectonic event. The thermal anomalies dataset was combined with tectonic parameters of Central Crete, such as active faults and fault density, seismogenic zones and ground displacement maps produced using Sentinel-1 satellite imagery and the Interferometric Synthetic Aperture Radar technique. Regarding the thermal anomaly of 27 September, its greatest portion was observed over the footwall part of the fault where a significant subsidence up to 20 cm exists. We suggest that the thermal anomalies are possibly connected with gas release which happens due to stress changes and is controlled by the existence of tectonic lines and the density of the faults, even if alternative explanations could not be excluded. Full article

An ionospheric anomaly is the irregular change of the ionosphere. It may result in potential threats for the ground-based augmentation system (GBAS) supporting the high-level precision approach. To counter the hazardous anomalies caused by the steep gradient in ionospheric delays, customized monitors are equipped in GBAS architectures. A major challenge is to rapidly detect the ionospheric gradient anomaly from environmental noise to meet the safety-critical requirements. A one-class support vector machine (OCSVM)-based monitor is developed to clearly detect ionospheric anomalies and to improve the robust detection speed. An offline-online framework based on the OCSVM is proposed to extract useful information related to anomalous characteristics in the presence of noise. To validate the effectiveness of the proposed framework, the influence of noise is fully considered and analyzed based on synthetic, semi-simulated, and real data from a typical ionospheric anomaly event. Synthetic results show that the OCSVM-based monitor can identify the anomaly that cannot be detected by other commonly-used monitors, such as the CCD-1OF, CCD-2OF and KLD-1OF. Semi-simulation results show that compared with other monitors, the newly proposed monitor can improve the average detection speed by more than 40% and decrease the minimum detectable gradient change rate to 0.002 m/s. Furthermore, in the real ionospheric anomaly event experiment, compared with other monitors, the OCSVM-based monitor can improve the detection speed by 16%. The result indicates that the proposed monitor has encouraging potential to ensure integrity of the GBAS. Full article

Noise annoyance depends not only on sound energy, but also on other features, such as those in its spectrum (e.g., low frequency and/or tonal components), and, over time, amplitude fluctuations, such as those observed in road, rail, or aircraft noise passages. The larger these fluctuations, the more annoying a sound is generally perceived. Many algorithms have been implemented to quantify these fluctuations and identify noise events, either by looking at transients in the sound level time history, such as exceedances above a fixed or time adaptive threshold, or focusing on the hearing perception process of such events. In this paper, four criteria to detect sound were applied to the acoustic monitoring data collected in two urban areas, namely Andorra la Vella, Principality of Andorra, and Milan, Italy. At each site, the 1 s A-weighted short L_Aeq,1s time history, 10 min long, was available for each hour from 8:00 a.m. to 7:00 p.m. The resulting 92-time histories cover a reasonable range of urban environmental noise time patterns. The considered criteria to detect noise events are based on: (i) noise levels exceeding by +3 dB the continuous equivalent level L_AeqT referred to the measurement time (T), criteria used in the definition of the Intermittency Ratio (IR) to detect noise events; (ii) noise levels exceeding by +3 dB the running continuous equivalent noise level; (iii) noise levels exceeding by +10 dB the 50th noise level percentile; (iv) progressive positive increments of noise levels greater than 10 dB from the event start time. Algorithms (iii) and (iv) appear suitable for notice-event detection; that is, those that (for their features) are clearly perceived and potentially annoy exposed people. The noise events detected by the above four algorithms were also evaluated by the available anomalous noise event detection (ANED) procedure to classify them as produced by road traffic noise or something else. Moreover, the assessment of the sonic environment by the Harmonica index was correlated with the single event level (SEL) of each event detected by the four algorithms. The threshold value of 8 for the Harmonica index, separating the “noisy” from the “very noisy” environments, corresponds to lower SEL levels for notice-events as identified by (iii) and (iv) algorithms (about 88–89 dB(A)) against those identified by (i) and (ii) criteria (92 dB(A)). Full article

Anomaly detection has gained considerable attention in the past couple of years. Emerging technologies, such as the Internet of Things (IoT), are known to be among the most critical sources of data streams that produce massive amounts of data continuously from numerous applications. Examining these collected data to detect suspicious events can reduce functional threats and avoid unseen issues that cause downtime in the applications. Due to the dynamic nature of the data stream characteristics, many unresolved problems persist. In the existing literature, methods have been designed and developed to evaluate certain anomalous behaviors in IoT data stream sources. However, there is a lack of comprehensive studies that discuss all the aspects of IoT data processing. Thus, this paper attempts to fill this gap by providing a complete image of various state-of-the-art techniques on the major problems and core challenges in IoT data. The nature of data, anomaly types, learning mode, window model, datasets, and evaluation criteria are also presented. Research challenges related to data evolving, feature-evolving, windowing, ensemble approaches, nature of input data, data complexity and noise, parameters selection, data visualizations, heterogeneity of data, accuracy, and large-scale and high-dimensional data are investigated. Finally, the challenges that require substantial research efforts and future directions are summarized. Full article

Recent advances in technology have enabled the development of affordable low-cost acoustic monitoring systems, as a response of several fields of application that require a close acoustic analysis in real-time: road traffic noise in crowded cities, biodiversity conservation in natural parks, behavioural tracking in the elderly living alone and even surveillance in public places for safety reasons. This paper presents a low-cost wireless acoustic sensor network developed to gather acoustic data to build a 24/7 real-time soundmap. Each node of the network comprises an omnidirectional microphone and a computation unit, which processes acoustic information locally to obtain nonsensitive data (i.e., equivalent continuous loudness levels or acoustic event labels) that are sent to a cloud server. Moreover, it has also been studied the placement of the acoustic sensors in a real scenario, following acoustics criteria. The ultimate goal of the deployed system is to enable the following functions: (i) to measure the Leq in real-time in a predefined window, (ii) to identify changing patterns in the previous measurements so that anomalous situations can be detected and (iii) to prevent and attend potential irregular situations. The proposed network aims to encourage the use of real-time non-invasive devices to obtain behavioural and environmental information, in order to take decisions in real-time. Full article