High Detection Probability Research Articles

Classified Early Warning and Forecast of Severe Convective Weather Based on LightGBM Algorithm

Severe convective weather can lead to a variety of disasters, but they are still difficult to be pre-warned and forecasted in the meteorological operation. This study generates a model based on the light gradient boosting machine (LightGBM) algorithm using C-band radar echo products and ground observations, to identify and classify three major types of severe convective weather (i.e., hail, short-term heavy rain (STHR), convective gust (CG)). The model evaluations show the LightGBM model performs well in the training set (2011-2017) and the testing set (2018) with the overall false identification ratio (FIR) of only 4.9% and 7.0%, respectively. Furthermore, the average probability of detection (POD), critical success index (CSI) and false alarm ratio (FAR) for the three types of severe convective weather in two sample sets are over 85%, 65% and lower than 30%, respectively. The LightGBM model and the storm cell identification and tracking (SCIT) product are then used to forecast the severe convective weather 15 - 60 minutes in advance. The average POD, CSI and FAR for the forecasts of the three types of severe convective weather are 57.4%, 54.7% and 38.4%, respectively, which are significantly higher than those of the manual work. Among the three types of severe convective weather, the STHR has the highest POD and CSI and the lowest FAR, while the skill scores for the hail and CG are similar. Therefore, the LightGBM model constructed in this paper is able to identify, classify and forecast the three major types of severe convective weather automatically with relatively high accuracy, and has a broad application prospect in the future automatic meteorological operation.

Dim and Small Target Detection in Multi-Frame Sequence Using Bi-Conv-LSTM and 3D-Conv Structure

Infrared dim and small target detection is widely used in military and civil fields. Traditional methods in that application rely on the local contrast between the target and background for single-frame detection. On the other hand, those algorithms depend on the motion model with fixed parameters for multi-frame association. For the great similarity of gray value and the dynamic changes of motion model parameters in the condition of low SNR and strong clutter, those methods possess weak robustness, low detection probability, and high false alarm rate. In this paper, an infrared video sequences encoding and decoding model based on Bidirectional Convolutional Long Short-Term Memory structure (Bi-Conv-LSTM) and 3D Convolutional structure (3D-Conv) is proposed, addressing the problem of high similarity and dynamic changes of parameters. For solving the problem of dynamic change in parameters, Bi-Conv-LSTM structure is used to learn the motion model of targets. And for the problem of low local contrast, 3D-Conv structure is adopted to extend receptive field in the time dimension. In order to improve the precision of detection, the Decoding part is divided into two different full connections with distinctive active function. Simulation results show that the trajectory detection accuracy of the proposed model is more than 90% under the condition of low SNR and maneuvering motion, which is better than traditional method of 80% in DB-TBD 20% in others. Real data experiment to illustrate that that our proposed method can detect small infrared targets of a low false alarm rate and high detection probability.

On Trajectory Design for Intruder Detection in Wireless Mobile Sensor Networks

We address the problem of detecting the invasion of an intruder into a region of interest (ROI) which is monitored by a distributed bandwidth-constrained wireless mobile sensor network (WMSN). We design periodic trajectories for the mobile sensor nodes (MSNs) such that high detection probabilities are obtained while maintaining the MSNs' energy consumption low. To reduce the transmission and processing burden on the MSNs, we propose an operation algorithm based on two modes, surveying mode and confirmation mode. In the former, to efficiently detect the intruder while using little mechanical energy, we optimize the surveying path such that the sensed area is maximized. During this mode, each MSN performs local detection and switches to the confirmation mode if and only if the intruder is suspected to be present. In the confirmation mode, each MSN collects further measurements over a predefined duration to reduce the detection uncertainly. A binary local hypothesis testing is performed at each MSN and only positive test statistics are transmitted to the FC where the ultimate decision is taken. Simulations results show the merits of the proposed two-mode operation algorithm in terms of detection performance and energy efficiency.

Space-Time Adaptive Detection at Low Sample Support

An important problem in space-time adaptive detection is the estimation of the large p-by-p interference covariance matrix from training signals. When the number of training signals n is greater than 2p, existing estimators are generally considered to be adequate, as demonstrated by fixed-dimensional asymptotics. But in the low-sample-support regime (n < 2p or even n < p) fixed-dimensional asymptotics are no longer applicable. The remedy undertaken in this paper is to consider the "large dimensional limit" in which n and p go to infinity together. In this asymptotic regime, a new type of estimator is defined (Definition 2), shown to exist (Theorem 1), and shown to be detection-theoretically ideal (Theorem 2). Further, asymptotic conditional detection and false-alarm rates of filters formed from this type of estimator are characterized (Theorems 3 and 4) and shown to depend only on data that is given, even for non-Gaussian interference statistics. The paper concludes with several Monte Carlo simulations that compare the performance of the estimator in Theorem 1 to the predictions of Theorems 2-4, showing in particular higher detection probability than Steiner and Gerlach's Fast Maximum Likelihood estimator.

A Ship Detection and Tracking Algorithm for an Airborne Passive Interferometric Microwave Sensor (PIMS)

Ship detection by an airborne passive interferometric microwave sensor has been proposed as a powerful complementary tool for ship detection, especially for large ships in low visibility conditions and high sea clutter. In this article, the brightness temperature (TB) models of the sea background and the ship on the sea are established. Given the TB models, a ship detection and tracking algorithm with four steps is proposed, which is devoted to ship tracking. In addition, the impacts of the noises in the passive interferometric microwave images (PIMI) and the algorithm parameters are analyzed and discussed. Numerical simulations and airborne experiments are performed to assess the practicability of the algorithm. Both numerical and experimental results have demonstrated the practicability of the proposed algorithm with dealing with the additional systematic errors and estimating the ship heading, and the results also exhibit that the additional noises in the PIMI caused by the stochastic fluctuations of the aircraft and the algorithm parameters have a non-negligible impact on the algorithm detection performance. As a result, the “ship-free” observed TB maps should be carefully considered as the sea background TB map to improve the algorithm detection performance by reducing the noises caused by the stochastic fluctuations of the aircraft. In addition, the algorithm parameter should also be carefully considered to get a high ship detection probability with a low ship false-alarm rate.

Adaptive δ-Generalized Labeled Multi-Bernoulli Filter for Multi-Object Detection and Tracking

The δ-generalized labeled multi-Bernoulli (δ-GLMB) filter is an efficient approach for multiobject tracking in case of high clutter density and low detection probability. However, the formulation of the original δ-GLMB filter requires that the birth δ-GLMB filtering density is known a priori. It is inapplicable for the birth object appearing from unknown positions. To address this problem, an adaptive δ-GLMB filter is proposed to detect and track the birth objects with unknown position information. This adaptive filter establishes the birth δ-GLMB filtering density by using measurements at previous three successive times. Simulation results indicate that the proposed adaptive δ-GLMB filter may efficiently detect and track the multiple objects with unknown positions. Simulation results also demonstrate that the proposed adaptive δ-GLMB filter performs better than the other existing adaptive filters.

Ground Validation and Error Sources Identification for GPM IMERG Product over the Southeast Coastal Regions of China

The Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement mission (IMERG) has been widely evaluated. However, most of these studies focus on the ultimate merged satellite-gauge precipitation estimate and neglect the valuable intermediate estimates which directly guide the improvement of the IMERG product. This research aims to identify the error sources of the latest IMERG version 6 by evaluating the intermediate and ultimate precipitation estimates, and further examine the influences of regional topography and surface type on these errors. Results show that among six passive microwave (PMW) sensors, the Microwave Humidity Sounder (MHS) has outstanding comprehensive behavior, and Special Sensor Microwave Imager/Sounder (SSMIS) operates advanced at precipitation detection, while the Sounder for Atmospheric Profiling of Humidity in the Intertropics by Radiometry (SAPHIR) has the worst performance. More precipitation events are detected with larger quantitative uncertainty in low-lying places than in highlands, in urban and water body areas than in other places, and more in coastal areas than in inland regions. Infrared (IR) estimate has worse performance than PMW, and the precipitation detectability of IR is more sensitive to the factors of elevation and the distance to the coast, as larger critical successful index (CSI) over lowlands and coastal areas. PMW morphing and the mixing of PMW and IR algorithms partly reverse the conservative feature of the precipitation detection of PMW and IR estimates, resulting in higher probability of detection (POD) and false alert ratio (FAR). Finally, monthly gauge calibration improves most of the statistical indicators and reduces the influence of elevation and surface type factor on these errors.

Environmental DNA Assessment Reveals Restoration Success for Mudpuppies (Necturus maculosus)

Mudpuppies (Necturus maculosus) are secretive, fully aquatic salamanders with a range that spans much of the eastern United States and Canada including the Great Lakes region. Although this species was once abundant, there have been widespread declines due to habitat loss and modification, pollution, lampricide use, and overcollection. We compared environmental DNA (eDNA) and trapping surveys conducted from 2014 to 2016 to determine Mudpuppy occupancy along the St. Clair-Detroit River System, where this indicator species could be a gauge for success of ongoing restoration. Mudpuppy eDNA was detected at all sites with positive trapping records, as well as one site where individuals have not been trapped previously. Sites with shoreline restoration had the highest occupancy estimates, whereas deep-water restoration did not affect Mudpuppy occupancy. Additionally, eDNA surveys resulted in higher detection probability than setline and minnow trap survey methods, illustrating the benefit of using eDNA to detect secretive species. This study demonstrates the success of restoration efforts in increasing the occupancy of an indicator species and can be used as a template for other restoration initiatives.

Generalized self‐tuning system for adaptive threshold estimators in cognitive radio systems using swarm and evolutionary‐based approaches

AbstractParameter‐based adaptive threshold estimators (ATEs) are widely used for signal detection in cognitive radio (CR) systems. However, their performance deteriorates under dynamic spectra conditions owing to a lack of valid methods to accurately self‐tune the different parameters of such ATEs. In this article, we address this limitation by proposing a generalized system for self‐tuning the parameters of any ATE based only on the input signal measured per time. We adopt swarm and evolutionary‐based metaheuristic optimization techniques to effectively search for the optimal parameter values of any ATE. Our system controls the search process by applying the between‐class variance function adapted from Otsu's algorithm as the objective function. We tested the system using five different metaheuristic optimization algorithms (MOAs) to self‐tune two different ATEs, namely the recursive one‐sided hypothesis testing (ROHT) technique and the histogram partitioning algorithm under Rayleigh and Rician fading channels, as well as under different modulation schemes, including the 4‐quadrature amplitude modulation and 4‐phase shift keying schemes. Our findings suggest that our proposed system yields generally a small error rate irrespective of the MOA used. In addition, the ROHT‐cuckoo search optimization configuration yielded a reasonably high and low probability of detection and probability of false alarm, respectively, as a function of the signal‐to‐noise‐ratio of the input signal at a fast average processing time of 0.0699 seconds. We concluded that our system presents an effective mechanism that can be used to automatically tune the parameters of any ATE for useful signal detection in CR.

Estimating the Audibility of Industrial Noise to Denning Polar Bears

ABSTRACTOil and gas activities on Alaska's North Slope overlap spatially with polar bear (Ursus maritimus) maternal denning habitat and temporally with the peri‐partum and emergence periods. Noise associated with these activities can be substantial and concerns regarding the effects on polar bears have been acknowledged. But the secluded and ephemeral nature of subnivean maternal dens renders the measurement of behavioral and physiological responses of bears to noise exposure challenging, except for rare cases when disturbance‐prompted den abandonment has been documented. These limitations, coupled with the uncertainty associated with the synergistic effects of anthropogenic activities on bears in a rapidly changing Arctic, prompt the need to develop predictive models of disturbance to ensure management guidelines effectively mitigate disturbance. Towards this end, we characterized noise propagation from 9 sources (2 aircraft, 2 over‐tundra tracked vehicles, 4 wheeled on‐road vehicles, and humans walking) used to support industrial activities around artificial snow dens near Milne Point, Alaska, USA, in March–April 2010. We built dens in 4 configurations to mimic variability in den roof thickness and to evaluate differences in noise propagation from when the den was closed compared to being open, similar to den emergence. We integrated these data with an existing polar bear audiogram and developed models to predict auditory‐detection probabilities as a function of distance from the den. Within a closed den, aircraft had high probabilities (detection probability ≥75%) of being detected by polar bears at distances ≤1.6 km and ground‐based sources had high probabilities of detection at distances ≤0.8 km. On average, closed dens reduced noise levels by 15 decibels (dB) relative to open dens. Our findings indicate that although polar bear snow dens effectively attenuate acoustic sound pressure levels, noise from some industrial support vehicles was likely to be detected farther from dens than previously documented. These results reinforce the importance of maintaining buffer zones around polar bear dens to minimize the potential for den disturbance. © 2020 The Wildlife Society.

Development of algorithm for identification of maligant growth in cancer using artificial neural network

The precise identification and characterization of small pulmonary nodules at low-dose CT is a necessary requirement for the completion of valuable lung cancer screening. It is compulsory to develop some automated tool, in order to detect pulmonary nodules at low dose ct at the beginning stage itself. The numerous algorithms had been proposed earlier by many researchers in the past, but, the accuracy of prediction is always a challenging task. In this work, an artificial neural network based methodology is proposed to find the irregular growth of lung tissues. Higher probability of detection is taken as a goal to get an automated tool, with great accuracy. The finest feature sets derived from Haralick Gray level co occurrence Matrix and used as the dimension reduction way for feeding neural network. In this work, a binary Binary classifier neural network has been proposed to identify the normal images out of all the images. The capability of the proposed neural network has been quantitatively computed using confusion matrix and found in terms of classification accuracy.

Analysis of energy detector with improved ED and variable threshold ED suitable for digital TV IEEE 802.22 WRAN

Now a days demand of wireless devices are increasing very rapidly by different users, while available spectrum resource is very limited. To overcome the need of different users, spectrum should be utilized efficiently. Cognitive Radio is introduced to access the spectrum dynamically by taking the permission of the licensed users for specific time interval when their spectrum is freely available. For this task of CR, many spectrum sensing techniques are introduced to sense the spectrum. Here in this paper, we have focused on energy detector(ED) spectrum sensing technique and overcome the problems with the ED at low SNR. Improved Energy Detector and Variable Threshold Energy Detector (VTED) give better results than normal ED technique which has been demonstrated through different receiver operating characteristics. Under Cooperative Spectrum Sensing (CSS) environment, it has been shown that OR logic gives probability of detection greater than 0.9 rather than the AND logic for optimized number of cooperative users 10. Results have been shown that VTED technique overcome the low SNR region problem of normal ED technique with higher probability of detection i.e. greater than 0.9 for cognitive radio networks.

Energy Efficient and Interference-aware Spectrum Sensing Technique for Improving the Throughput in Cognitive Radio Networks

Cognitive Radio network (CRN) depends on opportunistic spectrum access and spectrum sensing for improving the wireless networks’ spectrum efficiency. Since throughput maximization can result in high energy consumption, the spectrum sensing technique should address the energy-throughput trade off. The spectrum sensing time has to be determined by the considering the residual battery energies of each secondary user (SU). The Primary User (PU) interference degrades the throughput of the entire network. Hence the transmit power level should be determined considering the PU interference and SU battery energy. This paper proposes Energy Efficient and Interference-aware Spectrum Sensing (EEISS) technique for improving the throughput in CRN. In this work, the sensing time is dynamically estimated based on the battery energy levels of SUs and the transmit power is determined depending on PU signal and battery energy levels of SUs. A Game theory model is formulated to maximize the throughput based on these constraints. Experimental results show that EEISS attains improved throughput, higher probability of detection and higher residual energy.

Detectability and Abundance of Snowy Plovers at Salt Plains National Wildlife Refuge, Oklahoma

AbstractIn the past two decades, Salt Plains National Wildlife Refuge has been increasingly recognized as important habitat for both breeding and migratory shorebirds. North American snowy plovers Charadrius nivosus in particular rely on the nearly 5,000-ha salt flat at Salt Plains National Wildlife Refuge, which thousands use as breeding and stopover habitat. Elsewhere on the Southern Great Plains, decadal declines up to 75% within snowy plover subpopulations have been documented and attributed to vegetation encroachment, increased rates of nest predation, and decreased availability of fresh surface water. Despite many attempts to estimate this species' abundance across the continent, to date, no known attempt at distance sampling of snowy plovers has occurred. To address this paucity of data, we assessed feasibility of distance sampling methods to accurately estimate snowy plover abundance and detectability. Distance sampling surveys (2017–2018) indicated high detection probability (P = 0.80) and the population abundance estimate across the salt flat extrapolated to 3,307 individuals. The distance-sampling population abundance estimate is lower than population abundance estimates determined by two previous studies within the past decade but far greater than 2,105 estimated for a study in 2006. Overall, distance sampling snowy plovers at Salt Plains National Wildlife Refuge proved to be an effective addition to pre-established survey protocols but further investigation is needed to compare accuracy and precision of methods used in this study, annual surveys conducted by Salt Plains National Wildlife Refuge, and other potential snowy plover surveys.

Effect of Stream Permanence on Predation Risk of Lotic Crayfish by Riparian Predators

Given the importance of crayfish in stream ecosystems, gaining insight into the role of stream permanence in maintaining predator–prey interactions is critical. Our objectives were to determine the influence of stream permanence and season on crayfish predation and assess the role of stream permanence and crayfish density on the presence of predators, while accounting for imperfect detection. We conducted surveys of crayfish density, mammalian scat, and environmental variables within 10 intermittent and 10 permanent streams in the Ozark Highlands. We used occupancy modeling to assess the relationship between predator presence, crayfish density, and environmental variables. Stream permanence did not play a role in determining relative frequency of occurrence or percent volume of crayfish prey in mammalian diets. However, percent volume and relative frequency of crayfish prey found in scats differed by season, with both highest in spring and summer. The relative frequency and percent volume of fish prey showed a significant interaction of season by stream permanence, which may be the first instance of this observation. Procyon lotor (Raccoon) had the highest detection probability (p = 0.39), whereas Neovison vison (American Mink; p = 0.15) and Lontra canadensis (River Otter; p = 0.03) had low detection probabilities. Further study into predator–prey interactions in the context of hydrology, particularly when related to imperiled groups like freshwater crayfishes, is needed since climate change is expected to alter hydrologic patterns.

Cost-effectiveness of longitudinal surveillance for Piscirickettsia salmonis using qPCR in Atlantic salmon farms (Salmo salar) in Chile.

Costs of diagnostic testing including sample collection, sampling frequency and sample size are an important consideration in the evaluation of the economic feasibility of alternative surveillance strategies for detection of infectious diseases in aquatic animals. In Chile, Piscirickettsia salmonis is the primary reason for antibiotic treatments in farmed Atlantic salmon. In 2012, a surveillance and control programme for piscirickettsiosis was established with an overall goal of reducing antibiotic use. The present study estimated the cost-effectiveness of different sampling frequencies and sample sizes to achieve at least 95% confidence of early detection of P.salmonis at the netpen and farm levels using a validated qPCR test. We developed a stochastic model that incorporated variability in test accuracy, within-pen prevalence and sampling costs. Our findings indicated that the current piscirickettsiosis surveillance programme based on risk-based sampling of five moribund or dead fish from 2 to 3 netpens is cost-effective and gives a high probability of detection of P.salmonis in Atlantic salmon farms in Chile at both the netpen and farm levels. Results from this study should incentivize salmon farmers to establish cost-effective strategies for early detection of P.salmonis infection and the application of this approach to other highly infectious diseases.

Object Detection Using Stacked YOLOv3

Object detection is a stimulating task in the applications of computer vision. It is gaining a lot of attention in many real-time applications such as detection of number plates of suspect cars, identifying trespassers under surveillance areas, detecting unmasked faces in security gates during the COVID-19 period, etc. Region-based Convolution Neural Networks(R-CNN), You only Look once (YOLO) based CNNs, etc., comes under Deep Learning approaches. In this proposed work, an improved stacked Yolov3 model is designed for the detection of objects by bounding boxes. Hyperparameters are tuned to get optimum performance. The proposed model evaluated using the COCO dataset, and the performance is better than other existing object detection models. Anchor boxes are used for overlapping objects. After removing all the predicted bounding boxes that have a low detection probability, bounding boxes with the highest detection probability are selected and eliminated all the bounding boxes whose Intersection Over Union value is higher than 0.4. Non-Maximal Suppression (NMS) is used to only keep the best bounding box. In this experimentation, we have tried with various range of values, but finally got better result at threshold 0.5.

Combining disciplines: Dealing with observed and cryptic animal residencies in passive telemetry data by applying econometric decision-making models

Migratory species do not necessarily behave migratory continuously. An important aspect of studying migratory species is therefore to distinguish between movement and resident behavior. Telemetry is a rapidly evolving technique to study animal movement, but the number of data processing techniques to account for resident behavior remains limited. In this study we describe how models that were initially developed to predict human customer behavior, i.e. two-part and three-part models, provide new insights in the movement of migrating eel by accounting for resident behavior apparent from telemetry data sets. In econometrics, two-part models take into account that the decision of a customer to purchase an item and the decision of the customer on the purchase quantity of the concerning product, might be affected by different factors. Similarly, the factors that affect the decision of a fish to migrate or to stay resident might be different from the factors that affect the swimming speed of the fish. Telemetry data of eel movement in the Permanent Belgian Acoustic Receiver Network (PBARN) of the Scheldt Estuary was used. This network with high detection probabilities allowed residencies to be recognized, defined, and introduced as zero values in a movement-residency data set. Two-part models, which consider movement decision, i.e. residency or movement, and movement intensity, i.e. swimming speed, as two different processes or parts of one larger model, outperformed one-part models that do not make that distinction. This underlines the complex migration behavior eels exhibit. These two-part models in turn were outperformed by three-part models that also accounted for cryptic (i.e. unobserved) residencies. While the one-part model identified the tides and the distance from the most upstream gate as most important for movement, the three-part models identified the tides as most important for the movement decision and the distance from the most upstream gate as most important for the movement intensity. Considering movement decisions, cryptic residencies and movement intensity in modeling efforts increased model performance by 9.8%, underlining the importance of acknowledging the potentially complex behavior animals exhibit.

Blind Wireless Network Topology Inference

This work proposes a framework to discover the topology of a non-collaborative packet-based wireless network using radio-frequency (RF) sensors. The methodology developed is blind, allowing topology sensing of a network whose key features (i.e., number of nodes, physical layer signals, and medium access control (MAC) and routing protocols) are unknown. Because of the wireless medium, over-the-air signals captured by the sensors are mixed; therefore, blind source separation (BSS) and measurement association are used to separate traffic patterns. Then, to infer the topology, we detect directed data flows among nodes by identifying causal relationships between the separated transmitted patterns. We propose causal inference methods such as Granger causality (GC), transfer entropy (TE), and conditional transfer entropy (CTE) that use the times series of traffic profiles, and a solution based on a neural network (NN) that exploits distilled time-based features. The framework is validated on an ad-hoc wireless network accounting for MAC protocol, packet collisions, nodes mobility, the spatial density of sensors, and channel impairments, such as path-loss, shadowing, and noise. Numerical results reveal that the proposed approach reaches a high probability of link detection and a moderate false alarm rate in mild shadowing regimes and low to moderate network nodes mobility.

Comparison of Two Sampling Methods to Estimate the Abundance of Lucanus cervus with Application of n-Mixture Models

Monitoring programs should be based on the measurement of two main pillars for evaluating the conservation status of a species: population size and geographical distribution. To date, the only way reported in the literature to obtain detailed information on L. cervus population size is to use the capture-mark-recapture method. This is an expensive and time-consuming technique that implies physical capture and handling of individuals, which could affect their survival. Therefore, in this study we tested and compared two non-invasive sampling approaches, namely evening walk transects and diurnal tree trunk surveys, to derive accurate abundance estimates by means of N-mixture models in a Bayesian framework. In our study, both methods showed relatively high detection probability (≥56%). However, tree surveys performed better than walk transects (≈80%), especially with the progression of the sampling season. Tree surveys proved to be more effective than walk transects in providing data for an accurate population density estimate (much smaller 95% Bayesian Confidence Intervals). In light of a cost and benefit assessment, the tree survey is undoubtedly more convenient, as well as more effective, as it is more time consuming but less expensive than a walk transect (one operator for 2–3 h vs. two operators for 30 min each). Moreover, it needs fewer expert operators because of the greater proximity to the species, increasing the probability of correctly identifying it, i.e., reducing type I error (false positive or overestimation of counts). For the first time, we applied N-mixture models for estimating population abundance of L. cervus. Overcoming all the limits imposed by the use of the capture-mark-recapture method, in this study we performed a further step forward in the planning of monitoring aimed at the conservation of L. cervus and the evaluation of its demographic trend.