For over 150 years, spectrally selective filters have been proposed to improve the vision of observers with color vision deficiencies. About 6% of males and <1% of females have anomalies in their gene arrays coded on the X chromosome that result in significantly decreased spectral separation between their middle- (M-) and long- (L-) wave sensitive cone photoreceptors. These shifts alter individuals’ color-matching and chromatic discrimination such that they are classified as anomalous trichromats. Broad-band spectrally selective filters proposed to improve the vision of color-deficient observers principally modify the illuminant and are largely ineffective in enhancing discrimination or perception because they do not sufficiently change the relative activity of M- and L-photoreceptors. Properly tailored notch filters, by contrast, might increase the difference of anomalous M- and L-cone signals. Here, we evaluated the effects of long-term usage of a commercial filter designed for thi...

Real-world Anomaly Detection in Surveillance Videos

Surveillance videos are able to capture a variety of realistic anomalies. In this paper, we propose to learn anomalies by exploiting both normal and anomalous videos. To avoid annotating the anomalous segments or clips in training videos, which is very time consuming, we propose to learn anomaly through the deep multiple instance ranking framework by leveraging weakly labeled training videos, i.e. the training labels (anomalous or normal) are at video-level instead of clip-level. In our approach, we consider normal and anomalous videos as bags and video segments as instances in multiple instance learning (MIL), and automatically learn a deep anomaly ranking model that predicts high anomaly scores for anomalous video segments. Furthermore, we introduce sparsity and temporal smoothness constraints in the ranking loss function to better localize anomaly during training. We also introduce a new large-scale first of its kind dataset of 128 hours of videos. It consists of 1900 long and untrimmed real-world surveillance videos, with 13 realistic anomalies such as fighting, road accident, burglary, robbery, etc. as well as normal activities. This dataset can be used for two tasks. First, general anomaly detection considering all anomalies in one group and all normal activities in another group. Second, for recognizing each of 13 anomalous activities. Our experimental results show that our MIL method for anomaly detection achieves significant improvement on anomaly detection performance as compared to the state-of-the-art approaches. We provide the results of several recent deep learning baselines on anomalous activity recognition. The low recognition performance of these baselines reveals that our dataset is very challenging and opens more opportunities for future work.

Problem & Motivation

One critical task in video surveillance is detecting anomalous events such as traffic accidents, crimes or illegal activities. Generally, anomalous events rarely occur as compared to normal activities. Therefore, to alleviate the waste of labor and time, developing intelligent computer vision algorithms for automatic video anomaly detection is a pressing need. The goal of a practical anomaly detection system is to timely signal an activity that deviates normal patterns and identify the time window of the occurring anomaly. Therefore, anomaly detection can be considered as coarse level video understanding, which filters out anomalies from normal patterns. Once an anomaly is detected, it can further be categorized into one of the specific activities using classification techniques. In this work, we propose an anomaly detection algorithm using weakly labeled training videos. That is we only know the video-level labels, i.e. a video is normal or contains anomaly somewhere, but we do not know where. This is intriguing because we can easily annotate a large number of videos by only assigning video-level labels. To formulate a weakly-supervised learning approach, we resort to multiple instance learning. Specifically, we propose to learn anomaly through a deep MIL framework by treating normal and anomalous surveillance videos as bags and short segments/clips of each video as instances in a bag. Based on training videos, we automatically learn an anomaly ranking model that predicts high anomaly scores for anomalous segments in a video. During testing, a longuntrimmed video is divided into segments and fed into our deep network which assigns anomaly score for each video segment such that an anomaly can be detected.

Method

Our proposed approach (summarized in Figure 1) begins with dividing surveillance videos into a fixed number of segments during training. These segments make instances in a bag. Using both positive (anomalous) and negative (normal) bags, we train the anomaly detection model using the proposed deep MIL ranking loss. https://www.crcv.ucf.edu/projects/real-world/method.png

UCF-Crime Dataset

We construct a new large-scale dataset, called UCF-Crime, to evaluate our method. It consists of long untrimmed surveillance videos which cover 13 realworld anomalies, including Abuse, Arrest, Arson, Assault, Road Accident, Burglary, Explosion, Fighting, Robbery, Shooting, Stealing, Shoplifting, and Vandalism. These anomalies are selected because they have a significant impact on public safety. We compare our dataset with previous anomaly detection datasets in Table 1. For more details about the UCF-Crime dataset, please refer to our paper. A short description of each anomalous event is given below. Abuse: This event contains videos which show bad, cruel or violent behavior against children, old people, animals, and women. Burglary: This event contains videos that show people (thieves) entering into a building or house with the intention to commit theft. It does not include use of force against people. Robbery: This event contains videos showing thieves taking money unlawfully by force or threat of force. These videos do not include shootings. Stealing: This event contains videos showing people taking property or money without permission. They do not include shoplifting. Shooting: This event contains videos showing act of shooting someone with a gun. Shoplifting: This event contains videos showing people stealing goods from a shop while posing as a shopper. Assault: This event contains videos showing a sudden or violent physical attack on someone. Note that in these videos the person who is assaulted does not fight back. Fighting: This event contains videos displaying two are more people attacking one another. Arson: This event contains videos showing people deliberately setting fire to property. Explosion: This event contains videos showing destructive event of something blowing apart. This event does not include videos where a person intentionally sets a fire or sets off an explosion. Arrest: This event contains videos showing police arresting individuals. Road Accident: This event contains videos showing traffic accidents involving vehicles, pedestrians or cyclists. Vandalism: This event contains videos showing action involving deliberate destruction of or damage to public or private property. The term includes property damage, such as graffiti and defacement directed towards any property without permission of the owner. Normal Event: This event contains videos where no crime occurred. These videos include both indoor (such as a shopping mall) and outdoor scenes as well as day and night-time scenes. https://www.crcv.ucf.edu/projects/real-world/dataset_table.png https://www.crcv.ucf.edu/projects/real-world/method.png

Gravity data measure small changes in gravity due to changes in the density of rocks beneath the Earth's surface. The data are collected on geophysical surveys conducted by Commonwealth, State & NT …Show full descriptionGravity data measure small changes in gravity due to changes in the density of rocks beneath the Earth's surface. The data are collected on geophysical surveys conducted by Commonwealth, State & NT Governments and the private sector.

This suite of CHLA and SST climatology and anomaly data products are derived from daily, 0.0125 degree x 0.0125 degree, MODIS Aqua CHLA and SST fields that cover the California Current System (22N - 51N, 155W - 105W) for the 11-year period July 2002 through June 2013. These daily fields, obtained from the NOAA CoastWatch West Coast Regional Node website, were processed using a successive 3x3, 5x5 and 7x7 grid cell hybrid median filtering technique. This technique was found to effectively reduce noise in the daily fields while maintaining features and detail in important regions such as capes and headlands. The resulting median filtered daily fields were then linearly interpolated to a 0.025 x 0.025 degree grid and averaged to create 132 monthly mean fields. The seasonal cycles at each 0.025 degree x 0.025 degree grid cell were obtained by fitting each multiyear time series of monthly means to a nine-parameter regression model consisting of a constant plus four harmonics (frequencies of N/(1-year), N-1,4; Risien and Chelton 2008, JPO). Even with the median filtering and the temporal averaging of the daily fields, the highly inhomogeneous nature of the MODIS fields still resulted in regression coefficients that were excessively noisy. We therefore applied the same successive 3x3, 5x5 and 7x7 hybrid median filtering technique, described above, to the regression coefficients before finally spatially smoothing the coefficients using a loess smoother (Schlax et al. 2001, JTECH) with filter cutoff wavelengths of 0.25 degree latitude by 0.25 degree longitude. The seasonal cycles were then calculated from the filtered regression coefficients for each 0.025 degree x 0.025 degree grid cell using the mean and all four harmonics. It is important to note that THIS SUITE OF DATA PRODUCTS IS HIGHLY EXPERIMENTAL and is strictly intended for scientific evaluation by experienced marine scientists.

We adaptively estimate both changepoints and local outlier processes in a Bayesian dynamic linear model with global-local shrinkage priors in a novel model we call Adaptive Bayesian Changepoints with Outliers (ABCO). We use a state-space approach to identify a dynamic signal in the presence of outliers and measurement error with stochastic volatility. We find that global state equation parameters are inadequate for most real applications and we include local parameters to track noise at each time-step. This setup provides a flexible framework to detect unspecified changepoints in complex series, such as those with large interruptions in local trends, with robustness to outliers and heteroscedastic noise. Finally, we compare our algorithm against several alternatives to demonstrate its efficacy in diverse simulation scenarios and two empirical examples on the U.S. economy.

Industrial sensor networks exhibit heterogeneous, federated, large-scale, and intelligent characteristics due to the increasing number of Internet of Things (IoT) devices and different types of sensors. Efficient and accurate anomaly detection of sensor data is essential for guaranteeing the system’s operational reliability and security. However, existing research on sensor data anomaly detection for industrial sensor networks still has several inherent limitations. First, most detection models usually consider centralized detection. Thus, all sensor data have to be uploaded to the control center for analysis, leading to a heavy traffic load. However, industrial sensor networks have high requirements for reliable and real-time communication. The heavy traffic load may cause communication delays or packets lost by corruption. Second, there are complex spatial and temporal features in industrial sensor data. The full extraction of such features plays a key role in improving detection performance. Nevertheless, the majority of existing methodologies face challenges in simultaneously and comprehensively analyzing both features. To solve the limitations above, this paper develops a cloud-edge collaborative data anomaly detection approach for industrial sensor networks that mainly consists of a sensor data detection model deployed at individual edges and a sensor data analysis model deployed in the cloud. The former is implemented using Gaussian and Bayesian algorithms, which effectively filter the substantial volume of sensor data generated during the normal operation of the industrial sensor network, thereby reducing traffic load. It only uploads all the sensor data to the sensor data analysis model for further analysis when the network is in an anomalous state. The latter based on GCRL is developed by inserting Long Short-Term Memory network (LSTM) into Graph Convolutional Network (GCN), which can effectively extract the spatial and temporal features of the sensor data for anomaly detection. The proposed approach is extensively assessed through experiments using two public industrial sensor network datasets compared with the baseline anomaly detection models. The numerical results demonstrate that the proposed approach outperforms the existing state-of-the-art models.

We develop a new global bathymetric model, named SYSU_FGGM, spanning from 80°S to 80°N with a grid resolution of 1′×1′. The model employs a filter combination method with a cutoff wavelength of 65 km, using the topo_25.1 model as the baseline topography for long-wavelength components (≥65 km). Short-wavelength components (

Gravity data measure small changes in gravity due to changes in the density of rocks beneath the Earth's surface. The data are collected on geophysical surveys conducted by Commonwealth, State & NT Governments and the private sector.

A digital magnetic anomaly database and map for the North American continent is the result of a joint effort by the Geological Survey of Canada (GSC), U. S. Geological Survey (USGS), and Consejo de Recursos Minerales of Mexico (CRM). The database and map represent a substantial upgrade from the previous compilation of magnetic anomaly data for North America, now over a decade old. This report presents three unique, gridded data sets used to make the magnetic anomaly map of North America. All three grids have 1-km spacing and are projected to the DNAG projection. These grids are provided in Geosoft binary grid format, with two files describing each of the grids (suffixes .grd and .gi). The first grids (NAmag_origmrg.grd and USmag_origmrg.grd) show the magnetic field at 1,000 m. above terrain. For the second grids (NAmag_hp500.grd and USmag_hp500.grd) we removed long-wavelength anomalies (500 km and greater) from the first grid. This grid was used for the published map. Although the North American merged grid represents a significant upgrade to older compilations, the existing patchwork of surveys is inherently unable to accurately represent anomalies with long (greater than roughly 150 km) wavelengths, particularly in the US and Canada (U.S. Magnetic- Anomaly Data Set Task Group, 1994). The lack of information about long wavelength anomalies is primarily related to datum shifts between merged surveys, caused by data acquisition at widely different times and by differences in merging procedures. Therefore, we removed anomalies with wavelengths greater than 500 km from the merged grid to reduce the effects caused by the spurious long wavelengths but still maintain the continuity of anomalies. The correction was accomplished by transforming the merged grid to the frequency domain, filtering the transformed data with a long-wavelength cutoff at 500 km, and subtracting the long-wavelength data grid from the merged grid. In addition to the 500-km high pass filter, an equivalent source method, based on long-wavelength characterization using satellite data (CHAMP satellite anomalies, Maus and others, 2002), was also used to correct for spurious shifts in the original magnetic anomaly grid (Ravat and others, 2002). These results are presented in the third grids (NAmag_CM.grd and USmag_CM.grd), in which the wavelengths longer than 500 km have been replaced by downward-continued satellite data.

🚗 Mel Spectrogram Dataset for Anomaly Detection in Road Traffic Sounds

📌 Overview

This dataset is designed for anomaly detection in road traffic sounds.
- Normal Data: Mel spectrograms of vehicle running sounds
- Anomalous Data: Mel spectrograms of non-vehicle sounds

📂 Dataset Structure

The dataset is organized into two main folders:

📷 Data Format

• File Format: PNG images
• Image Size: 224 × 251 pixels
• Sampling Rate: 16,000 Hz
• Mel Filter Bank Size: 224
• Time Axis: Horizontal direction
• Frequency Axis: Vertical direction
• Time Duration: 1 second per image

This suite of CHLA and SST climatology and anomaly data products are derived from daily, 0.0125 degree x 0.0125 degree, MODIS Aqua CHLA and SST fields that cover the California Current System (22N - 51N, 155W - 105W) for the 11-year period July 2002 through June 2013. These daily fields, obtained from the NOAA CoastWatch West Coast Regional Node website, were processed using a successive 3x3, 5x5 and 7x7 grid cell hybrid median filtering technique. This technique was found to effectively reduce noise in the daily fields while maintaining features and detail in important regions such as capes and headlands. The resulting median filtered daily fields were then linearly interpolated to a 0.025 x 0.025 degree grid and averaged to create 132 monthly mean fields. The seasonal cycles at each 0.025 degree x 0.025 degree grid cell were obtained by fitting each multiyear time series of monthly means to a nine-parameter regression model consisting of a constant plus four harmonics (frequencies of N/(1-year), N-1,4; Risien and Chelton 2008, JPO). Even with the median filtering and the temporal averaging of the daily fields, the highly inhomogeneous nature of the MODIS fields still resulted in regression coefficients that were excessively noisy. We therefore applied the same successive 3x3, 5x5 and 7x7 hybrid median filtering technique, described above, to the regression coefficients before finally spatially smoothing the coefficients using a loess smoother (Schlax et al. 2001, JTECH) with filter cutoff wavelengths of 0.25 degree latitude by 0.25 degree longitude. The seasonal cycles were then calculated from the filtered regression coefficients for each 0.025 degree x 0.025 degree grid cell using the mean and all four harmonics. It is important to note that THIS SUITE OF DATA PRODUCTS IS HIGHLY EXPERIMENTAL and is strictly intended for scientific evaluation by experienced marine scientists.This suite of CHLA and SST climatology and anomaly data products are derived from daily, 0.0125 degree x 0.0125 degree, MODIS Aqua CHLA and SST fields that cover the California Current System (22N - 51N, 155W - 105W) for the 11-year period July 2002 through June 2013. These daily fields, obtained from the NOAA CoastWatch West Coast Regional Node website, were processed using a successive 3x3, 5x5 and 7x7 grid cell hybrid median filtering technique. This technique was found to effectively reduce noise in the daily fields while maintaining features and detail in important regions such as capes and headlands. The resulting median filtered daily fields were then linearly interpolated to a 0.025 x 0.025 degree grid and averaged to create 132 monthly mean fields. The seasonal cycles at each 0.025 degree x 0.025 degree grid cell were obtained by fitting each multiyear time series of monthly means to a nine-parameter regression model consisting of a constant plus four harmonics (frequencies of N/(1-year), N-1,4; Risien and Chelton 2008, JPO). Even with the median filtering and the temporal averaging of the daily fields, the highly inhomogeneous nature of the MODIS fields still resulted in regression coefficients that were excessively noisy. We therefore applied the same successive 3x3, 5x5 and 7x7 hybrid median filtering technique, described above, to the regression coefficients before finally spatially smoothing the coefficients using a loess smoother (Schlax et al. 2001, JTECH) with filter cutoff wavelengths of 0.25 degree latitude by 0.25 degree longitude. The seasonal cycles were then calculated from the filtered regression coefficients for each 0.025 degree x 0.025 degree grid cell using the mean and all four harmonics. It is important to note that THIS SUITE OF DATA PRODUCTS IS HIGHLY EXPERIMENTAL and is strictly intended for scientific evaluation by experienced marine scientists.This suite of CHLA and SST climatology and anomaly data products are derived from daily, 0.0125 degree x 0.0125 degree, MODIS Aqua CHLA and SST fields that cover the California Current System (22N - 51N, 155W - 105W) for the 11-year period July 2002 through June 2013. These daily fields, obtained from the NOAA CoastWatch West Coast Regional Node website, were processed using a successive 3x3, 5x5 and 7x7 grid cell hybrid median filtering technique. This technique was found to effectively reduce noise in the daily fields while maintaining features and detail in important regions such as capes and headlands. The resulting median filtered daily fields were then linearly interpolated to a 0.025 x 0.025 degree grid and averaged to create 132 monthly mean fields. The seasonal cycles at each 0.025 degree x 0.025 degree grid cell were obtained by fitting each multiyear time series of monthly means to a nine-parameter regression model consisting of a constant plus four harmonics (frequencies of N/(1-year), N-1,4; Risien and Chelton 2008, JPO). Even with the median filtering and the temporal averaging of the daily fields, the highly inhomogeneous nature of the MODIS fields still resulted in regression coefficients that were excessively noisy. We therefore applied the same successive 3x3, 5x5 and 7x7 hybrid median filtering technique, described above, to the regression coefficients before finally spatially smoothing the coefficients using a loess smoother (Schlax et al. 2001, JTECH) with filter cutoff wavelengths of 0.25 degree latitude by 0.25 degree longitude. The seasonal cycles were then calculated from the filtered regression coefficients for each 0.025 degree x 0.025 degree grid cell using the mean and all four harmonics. It is important to note that THIS SUITE OF DATA PRODUCTS IS HIGHLY EXPERIMENTAL and is strictly intended for scientific evaluation by experienced marine scientists.

The increasingly high number of big data applications in seismology has made quality control tools to filter, discard, or rank data of extreme importance. In this framework, machine learning algorithms, already established in several seismic applications, are good candidates to perform the task flexibility and efficiently. sdaas (seismic data/metadata amplitude anomaly score) is a Python library and command line tool for detecting a wide range of amplitude anomalies on any seismic waveform segment such as recording artifacts (e.g., anomalous noise, peaks, gaps, spikes), sensor problems (e.g., digitizer noise), metadata field errors (e.g., wrong stage gain in StationXML). The underlying machine learning model, based on the isolation forest algorithm, has been trained and tested on a broad variety of seismic waveforms of different length, from local to teleseismic earthquakes to noise recordings from both broadband and accelerometers. For this reason, the software assures a high degree of flexibility and ease of use: from any given input (waveform in miniSEED format and its metadata as StationXML, either given as file path or FDSN URLs), the computed anomaly score is a probability-like numeric value in [0, 1] indicating the degree of belief that the analyzed waveform represents an anomaly (or outlier), where scores ≤0.5 indicate no distinct anomaly. sdaas can be employed for filtering malformed data in a pre-process routine, assign robustness weights, or be used as metadata checker by computing randomly selected segments from a given station/channel: in this case, a persistent sequence of high scores clearly indicates problems in the metadata

We present lists of anomaly-free charge assignments up to a maximum magnitude charge Qmax=10 for the chiral fermionic content of the MSSM plus 3 right-handed neutrinos.

Due to the large number of solutions, we compress the list into the file MSSMnuRcharges_Qmax10.gz. Please note that the unzipped file is approximately 130GB in size. We additionally include a smaller file, MSSMnuRcharges_Qmax4, containing the subset of anomaly-free charge assignments up to a maximum magnitude charge Qmax=4.

The files searchU1MSSM.cpp and searchU1MSSM.h contain C++ files (in the 2014 standard) to produce the solutions. runsearch.sh is a bash script that compiles the programs and then runs it for a sample set of inputs.

We provide Mathematica notebooks Analytic_solution_generator.nb and Analytic_Checks.nb which respectively provide the parametrisation of the analytic solution and checks thereof.

The files beginning 'filter' contain example programs that read in each line in the solution list, apply a filter and print only the solutions satisfying the conditions of that filter. runfilter.sh is a bash script that compiles the filters and then runs a single filter as an example.

These data and programs are based on this paper: https://arxiv.org/abs/2107.07926.

We present results from applying the SNAD anomaly detection pipeline to the third public data release of the Zwicky Transient Facility (ZTF DR3). The pipeline is composed of three stages: feature extraction, search of outliers with machine learning algorithms, and anomaly identification with followup by human experts. Our analysis concentrates in three ZTF fields, comprising more than 2.25 million objects. A set of four automatic learning algorithms was used to identify 277 outliers, which were subsequently scrutinized by an expert. From these, 188 (68 per cent) were found to be bogus light curves - including effects from the image subtraction pipeline as well as overlapping between a star and a known asteroid, 66 (24 per cent) were previously reported sources whereas 23 (8 per cent) correspond to non-catalogued objects, with the two latter cases of potential scientific interest (e.g. one spectroscopically confirmed RS Canum Venaticorum star, four supernovae candidates, one red dwarf flare). Moreover, using results from the expert analysis, we were able to identify a simple bi-dimensional relation that can be used to aid filtering potentially bogus light curves in future studies. We provide a complete list of objects with potential scientific application so they can be further scrutinised by the community. These results confirm the importance of combining automatic machine learning algorithms with domain knowledge in the construction of recommendation systems for astronomy. Our code is publicly available. Cone search capability for table J/MNRAS/502/5147/tabled1 (A complete list of anomaly candidates in the M 31, deep, and disk fields)

Filter matrix associated with the Tikhonov-regularized unfolding. The filter matrix $A$ encodes all biases coming from the unfolding process itself....

Link to the ScienceBase Item Summary page for the item described by this metadata record. Service Protocol: Link to the ScienceBase Item Summary page for the item described by this metadata record. Application Profile: Web Browser. Link Function: information

Sheet 2 - First Vertical Derivative of the Magnetic Field. Prepared by the USGS in cooperation with the Nevada Bureau of Mines and Geology Map 93B. To download this map PDF resource, please see the link provided.

Sheet 1 - Residual Total Magnetic Field Reduced to the North Magnetic Pole. Prepared by USGS in Cooperation with NBMG. Map 93B.To download this resource, please see the link provided.

The pediatric arterial filter market, while smaller than its adult counterpart, exhibits significant growth potential driven by rising prevalence of congenital heart defects and other cardiovascular anomalies in children requiring interventional procedures. The market's Compound Annual Growth Rate (CAGR) is estimated to be around 7% for the forecast period 2025-2033, reaching an estimated market value of $250 million by 2033, from a $150 million valuation in 2025. This growth is fueled by advancements in filter technology, leading to smaller, less invasive devices with improved biocompatibility and reduced risks of complications. Increasing awareness among healthcare professionals regarding the benefits of arterial filter placement in high-risk pediatric patients also contributes to market expansion. However, the market faces restraints including high procedural costs, stringent regulatory approvals, and the relatively limited number of specialized pediatric interventional cardiologists capable of performing the procedures. Key players like Medtronic, Terumo, LivaNova, EUROSETS, and Nipro are actively involved in research and development, driving innovation in filter design and materials. Regional variations in healthcare infrastructure and reimbursement policies influence market penetration, with North America and Europe currently holding the largest market shares. However, emerging economies in Asia-Pacific and Latin America present significant untapped potential, driven by rising healthcare expenditure and improving healthcare infrastructure. The forecast period will likely see increased competition and consolidation within the market, as companies strive to develop and market advanced filter technologies catering to the specific needs of the pediatric population.