Click Detection Research Articles

Finger Click Detection Using a Depth Camera

Selection mechanisms are an essential tool for designating elements in a User Interface. The advances in Natural User Interfaces promoted the selection mechanisms from devices such as the mouse, to more natural methods such as hand gestures. Recently, depth sensors became widely available, and paved the way for new methods of hand gestures recognition. In this paper, we present Depth Click, a new approach of finger click detection using a depth camera. Our proposed method provides natural interactions in terms of selection mechanisms, and goes beyond that in allowing the clicks to be used for actions other than selection. Our technique proved to be fast, reliable and flexible. This paper explains the algorithm we used to detect the finger clicks in detail. We explain about some preliminary experiments for evaluating the accuracy of our approach, as well as its detection rate. Finally, we discuss the results of the evaluation.

Assessing segmentation processes by click detection: online measure of statistical learning, or simple interference?

Statistical learning can be used to extract the words from continuous speech. Gómez, Bion, and Mehler (Language and Cognitive Processes, 26, 212-223, 2011) proposed an online measure of statistical learning: They superimposed auditory clicks on a continuous artificial speech stream made up of a random succession of trisyllabic nonwords. Participants were instructed to detect these clicks, which could be located either within or between words. The results showed that, over the length of exposure, reaction times (RTs) increased more for within-word than for between-word clicks. This result has been accounted for by means of statistical learning of the between-word boundaries. However, even though statistical learning occurs without an intention to learn, it nevertheless requires attentional resources. Therefore, this process could be affected by a concurrent task such as click detection. In the present study, we evaluated the extent to which the click detection task indeed reflects successful statistical learning. Our results suggest that the emergence of RT differences between within- and between-word click detection is neither systematic nor related to the successful segmentation of the artificial language. Therefore, instead of being an online measure of learning, the click detection task seems to interfere with the extraction of statistical regularities.

Detection rates of wild harbour porpoises and bottlenose dolphins using static acoustic click loggers vary with depth

Acoustic echolocation click loggers are widely used to study the presence and behaviour of coastal cetaceans. Various studies have examined their performance and how animal behaviour may affect click detection. This study examined the effect of logger depth in the water column on click detection rates for harbour porpoises (Phocoena phocoena) and bottlenose dolphins (Tursiops truncatus) at a coastal nearshore site. There were significantly more harbour porpoise clicks logged at the click loggers moored in the water column, closer to the surface, compared with those near the seabed. No such difference was recorded for bottlenose dolphins. The results were consistent throughout the study period and across sites, apart from a site closest to a possible feeding site, indicating that cetacean behaviour is among the possible factors explaining these findings. The study emphasizes the need for prior understanding of the target species and its habitat use, and stresses the importance of pilot studies to achieve reliable and meaningful data from static acoustic monitoring programs.

Integrating passive acoustic and visual data to model spatial patterns of occurrence in coastal dolphins

AbstractFine-scale information on the occurrence of coastal cetaceans is required to support regulation of offshore energy developments and marine spatial planning. In particular, the EU Habitats Directive requires an understanding of the extent to which animals from Special Areas of Conservation (SAC) use adjacent waters, where survey effort is often sparse. Designing survey regimes that can be used to support these assessments is especially challenging because visual sightings are expected to be rare in peripheral parts of a population's range. Consequently, even intensive visual line-transect surveys can result in few encounters. Static passive acoustic monitoring (PAM) provides new opportunities to extend survey effort by using echolocation click detections to quantify levels of occurrence of coastal dolphins, but this does not provide information on species identity. In NE Scotland, assessments of proposed offshore energy developments required information on spatial patterns of occurrence of bottlenose dolphins in waters in and next to the Moray Firth SAC. Here, we illustrate how this can be achieved by integrating data from broad-scale PAM arrays with presence-only data from visual surveys. Generalized estimating equations were used with PAM data to model the occurrence of dolphins in relation to depth, distance to coast, slope, and sediment, and to predict the spatial variation in the cumulative occurrence of all dolphin species across a 4 × 4 km grid of the study area. Classification tree analysis was then applied to available visual sightings data to estimate the likely species identity of dolphins sighted in each grid cell in relation to local habitat. By multiplying these probabilities, it was possible to provide advice on spatial variation in the probability of encountering bottlenose dolphins from this protected population at a regional scale, complementing data from surveys that estimate average density or overall abundance within a region.

Counting odontocetes from click train detections using multiple independent autonomous acoustic sensors

Passive acoustic monitoring (PAM) is often suggested as an effective technology to mitigate impacts from anthropogenic activities; however, the ability to reliably and efficiently detect, locate, and count cetaceans using PAM is still in development. One particularly useful application of PAM is species density estimation, which requires an estimate of the number of individuals involved in a detection event. Efforts have been undertaken to develop methods to directly count the number of vocalizing animals during acoustic detection events. For odontocetes, discrete clicks are almost indistinguishable between individuals, making it more difficult to determine the number of vocalizing animals as the number increases. Using recordings from multiple closely spaced (≈200 m) GuardBuoy sensors deployed on the Canadian Scotian Shelf, cross-sensor correlograms were produced to estimate the number of individual sperm whales, and, as a more challenging case, the number of vocalizing delphinids. Using feature-based multipath reflection discrimination, the raw time series were reduced to a synthetic time series of binary click detections with the multipath arrivals removed. The synthesized click-detection time series were used for the cross-sensor correlograms to generate improved estimates of the number of vocalizing animals as compared with using the raw time series.

Research on Information Technology with Detecting the Fraudulent Clicks Using Classification Method

This paper aims to find an effective solution to detect fraudulent clicks from commissioners’ click logs. Given user’s click information of ads, we want to predict the user’s fraudulent label – malicious or not. Based on training set, we build our classification models for fraudulent click detection. We first create and extract features for the raw log data we show above. Next, we choose models by evaluating classifiers, and then build an ensemble model. Finally, we applied our model on the real dataset for human judge reference.

Joint classification of whistles and echolocation clicks from odontocetes

We propose to use acoustic features of both clicks and whistles to classify odontocete sounds to species. The species studied are Cuvier’s beaked whales (Ziphius cavirostris), bottlenose dolphin (Tursiops truncatus), melon-headed whale (Peponocephala electra), and short- and long-beaked common dolphin (Delphinus delphis and D. capensis). An energy-based detector is used for echolocation click detection, and Roch’s Silbido algorithm is used for whistle detection. Detected whistles are characterized by maximum and minimum frequencies, duration, slope, spectral maxima, spectral gaps, number and frequency of inflection points, number of “loop” repetitions, and other acoustic characteristics. Detected clicks are characterized by cepstral characteristics, as well as by a set of noise-resistant statistics. Clicks that occur within a certain time neighborhood of a whistle have the corresponding feature vectors merged to produce the input to the classification system. Random forest and Gaussian mixture model classifiers are tested on the resulting features and performance is characterized. [Funding from ONR.]

Tracking beaked whales with a passive acoustic profiler float

Acoustic methods are frequently used to monitor endangered marine mammal species. Advantages of acoustic methods over visual ones include the ability to detect submerged animals, to work at night, and to work in any weather conditions. A relatively inexpensive and easy-to-use acoustic float, the QUEphone, was developed by converting a commercially available profiler float to a mobile platform, adding acoustic capability, and installing the ERMA cetacean click detection algorithm of Klinck and Mellinger [(2011). J. Acoust. Soc. Am. 129(4), 1807-1812] running on a high-power DSP. The QUEphone was tested at detecting Blainville's beaked whales at the Atlantic Undersea Test and Evaluation Center (AUTEC), a Navy acoustic test range in the Bahamas, in June 2010. Beaked whale were present at AUTEC, and the performance of the QUEphone was compared with the Navy's Marine Mammal Monitoring on Navy Ranges (M3R) system. The field tests provided data useful to evaluate the QUEphone's operational capability as a tool to detect beaked whales and report their presence in near-real time. The range tests demonstrated that the QUEphone's beaked whale detections were comparable to that of M3R's, and that the float is effective at detecting beaked whales.

Population density of sperm whales in the Bahamas estimated using non-linked sensors

Estimates are presented of sperm whale click detection probability and sperm whale population density at the U.S. Navy’s Atlantic Undersea Test and Evaluation Center (AUTEC) in the Bahamas. The estimation of the probability of detecting whale echolocation clicks at multiple non-linked sensors uses estimates of sperm whale source level distribution, beam pattern of click emission, distribution of whale locations and orientations with respect to the sensors while clicking, acoustic transmission loss from source (whale) to receiver (bottom hydrophone), and noise levels at the receiver. These data are combined in a Monte Carlo model that propagates simulated clicks from whales at various random positions to each receiving hydrophone to estimate the signal-to-noise ratio at the receiver and the detection function, the probability of detecting clicks as a function of distance. The estimated detection function for each receiving hydrophone is then combined with information on the detector’s rate of missed calls and false detections as a function of signal-to-noise ratio, average sperm whale click rates, and the actual number of clicks detected in a given period of time in order to estimate population density. Results are compared to multi-sensor cases where detection functions were estimated analytically.

The energy ratio mapping algorithm: A tool to improve the energy-based detection of odontocete echolocation clicks

The energy ratio mapping algorithm (ERMA) was developed to improve the performance of energy-based detection of odontocete echolocation clicks, especially for application in environments with limited computational power and energy such as acoustic gliders. ERMA systematically evaluates many frequency bands for energy ratio-based detection of echolocation clicks produced by a target species in the presence of the species mix in a given geographic area. To evaluate the performance of ERMA, a Teager-Kaiser energy operator was applied to the series of energy ratios as derived by ERMA. A noise-adaptive threshold was then applied to the Teager-Kaiser function to identify clicks in data sets. The method was tested for detecting clicks of Blainville's beaked whales while rejecting echolocation clicks of Risso's dolphins and pilot whales. Results showed that the ERMA-based detector correctly identified 81.6% of the beaked whale clicks in an extended evaluation data set. Average false-positive detection rate was 6.3% (3.4% for Risso's dolphins and 2.9% for pilot whales).

The word segmentation process as revealed by click detection

The click detection paradigm was once conceived as a method to study online syntactic processing, but well-controlled empirical investigations casted many doubts on the early findings based on it. In this paper, we show that click methods can still prove valid and useful. We asked adult participants to listen to an artificial speech stream composed of statistically defined trisyllabic nonce words, while having to detect clicks superposed on the stream. The clicks were presented either within or between consecutive words. After 2 minutes of exposure to the stream, participants were slower to detect clicks located within words than clicks located between words. This result suggests that methods like click detection are sensitive to online statistical computations, opening new possibilities to obtain a richer picture of the segmentation process than what was hitherto possible.

Optimizing interoperability between video-oculographic and electromyographic systems

A new system is presented that enhances the interoperability between a video-oculographic (VOG) system for mouse movement control and an electromyographic (EMG) system for mouse click detection. The proposed VOG-EMG system combines gaze and muscle information to minimize the number of undesired clicks due to involuntary activations and environmental noise. We tested the system with 24 subjects, comparing three different configurations: one in which the VOG and EMG systems worked independently and two in which we used VOG gaze information to improve the EMG click detection. Results show that the number of false-positive click detections can be reduced when VOG and EMG information is combined. In addition, the third configuration, including extra processing, can reduce the activation delay produced because of the combined use of the VOG and EMG systems. The new VOG-EMG system is meant to be used in noisy environments in which the number of false clicks may impeach a reliable human-computer interaction.

Automatic detection of bioacoustics impulses based on kurtosis under weak signal to noise ratio

Passive acoustic monitoring (PAM) of marine mammal vocalizations has been efficiently used in a wide set of applications ranging from marine wildlife surveys to risk mitigation of military sonar emissions. The primary use of PAM is for detecting bioemissions, a good proportion of which are impulse sounds or clicks. A click detection algorithm based on kurtosis estimation is proposed as a general automatic click detector. The detector works under the assumption that click trains are embedded in stochastic but Gaussian noise. Under this assumption, kurtosis is used as a statistical test for detection. The algorithm explores acoustic sequences with the optimal frequency bandwidth for focusing on impulse sounds. The detector is successfully applied to field observations, and operates under weak signal to noise ratios and in presence of stochastic background noise. The algorithm adapts to varying click center frequency. Kurtosis appears as a promising approach to detect click trains, alone or in combination with other clicks detector, and to isolate individual clicks.

An efficient statistics-based method for the automated detection of sperm whale clicks

In this paper, we present a new method for the automated detection of sperm whale regular clicks and creaks based on statistical computations. In the first stage, a spectrogram is computed from the input waveform, followed by a noise normalisation process. A frequency domain filter is then applied, and the energy accumulated in each time frame is calculated. Two-second time-windows are then classified as containing either regular clicks, creaks, or noise based on statistical parameters using a neural network classifier. Finally, previously obtained statistical parameters are used to implement an energy-based detection criterion for the classified time-windows. Individual regular clicks and creaks are isolated by linking contiguous detected time frames. The proposed method was tested on five recordings of sperm whale sounds. Comparison of the detection performances to hand-labelled regular clicks and creaks revealed that this method outperforms two recently reported waveform-based methods when working with the same recordings files. An average percentage of detection of 86.97% was attained for the set of files. This method consumes also little computation time.

A real-time detection system for odontocete echolocation clicks in the low-energy processing environment of an acoustic glider.

This presentation will introduce a real-time system, the energy ratio mapping algorithm (ERMA), for detection of odontocete echolocation clicks. The system has been developed to run continuously over extended periods of time in low-energy processing environments such as acoustic gliders. The detector operates in the time domain and takes advantage of the species-specific differences in the power spectra and interclick interval of echolocation clicks to minimize the number of false positive detections. We will demonstrate the operation of ERMA using the clicks of beaked whales, family Ziphiidae, as example target species. To get information on the performance of ERMA, a comparison with more sophisticated and already established detection systems was conducted. Preliminary results indicate that the developed method is a promising tool to reliably detect echolocation clicks emitted by beaked whales in real time. On-going work is focused on implementation of the detector in the acoustic seaglider developed by the Applied Physics Laboratory of the University of Washington (APL-UW) and the use of this platform for autonomous passive acoustic monitoring and mitigation of beaked whales and other odontocete species.

Killer whale habitat use and prey fields from remote hydrophones and echosounders.

The Salish Sea Hydrophone Network has been providing live audio streams to listeners around the globe for the past three years. Five hydrophones around the core summer habitat of the endangered southern resident killer whales allow detection of killer whale presence and inference of pod identity based on call associations. Detection of calls, whistles, and clicks is accomplished automatically by spectral and time domain analysis and manually by a global network of trained human listeners. The hydrophone network is more effective than a sighting network for detecting killer whales during inclement weather and at night, and for generating real‐time notifications to alert researchers and managers. Additionally, the live streams and archived recordings have enriched educational activities of local museums and aquariums. In 2008, the hydrophone at Lime Kiln State Park was supplemented by hydroacoustic surveys conducted using a Biosonics echosounder (200 kHz). From August through December the echosounder was deployed at 15 m depth to conduct a continuous, fixed, horizontal survey. Remote display technology was used to monitor the echosounder in real time while also listening to the live audio stream. The resultant time series of fish density and acoustic behavior of killer whales is presented.

Stimulus-Response Profile during Single-Pulse Transcranial Magnetic Stimulation to the Primary Motor Cortex

We examined the stimulus-response profile during single-pulse transcranial magnetic stimulation (TMS) by measuring motor-evoked potentials (MEPs) with electromyographic monitoring and hemodynamic responses with functional magnetic resonance imaging (fMRI) at 3 Tesla. In 16 healthy subjects, single TMS pulses were irregularly delivered to the left primary motor cortex at a mean frequency of 0.15 Hz with a wide range of stimulus intensities. The measurement of MEP proved a typical relationship between stimulus intensity and MEP amplitude in the concurrent TMS-fMRI environment. In the population-level analysis of the suprathreshold stimulation conditions, significant increases in hemodynamic responses were detected in the motor/somatosensory network, reflecting both direct and remote effects of TMS, and also the auditory/cognitive areas, perhaps related to detection of clicks. The stimulus-response profile showed both linear and nonlinear components in the direct and remote motor/somatosensory network. A detailed analysis suggested that the nonlinear components of the motor/somatosensory network activity might be induced by nonlinear recruitment of neurons in addition to sensory afferents resulting from movement. These findings expand our basic knowledge of the quantitative relationship between TMS-induced neural activations and hemodynamic signals measured by neuroimaging techniques.

Patterns of coastal use by Hawaiian spinner dolphins (Stenella longirostris) observed using passive acoustic monitoring.

Spinner dolphins are the most commonly sighted cetacean in coastal Hawaiian waters and are an important higher‐level trophic component of the near‐shore ecosystem. Establishing their long‐term patterns of occurrence is important both for their conservation and for a better understanding of the relationships between dolphins, their prey, and other members of the ecosystem. Here we report on an effort to use ecological acoustic recorders (EARs) to monitor the presence of spinner dolphins along the leeward coast of the island of Oahu, Hawaii. The EAR is a bottom‐moored recorder with a bandwidth of 30 kHz, which allows the detection of both dolphin whistles and echolocation clicks. Eight EARs were deployed along the coast at depths ranging from 15 to 50 m. Five units were deployed in an array, providing information about the EAR’s detection range. The other three units were placed at sites along the coast commonly frequented by spinner dolphins. The results reveal distinct patterns of preference in habitat use, both for daytime resting behavior and night‐time foraging. The recordings also provide a measure of vessel traffic at locations important to spinner dolphins. Combined, these data demonstrate the value of passive acoustic methods for monitoring cetacean populations and their habitat over extended periods.

Beaked whale detection utilizing characteristics of foraging echolocation clicks.

A technique is presented for the automatic acoustic detection of beaked whale (family Ziphidae) foraging echolocation clicks. The technique currently employs a three‐stage process. The first stage is an acoustic detection screener that uses a submillisecond detection window. The second stage analyzes for click frequency modulation characteristic of two species of beaked whale (Mesoplodon densirostris and Ziphius cavirostris). The third stage performs temporal filtering to reduce false positives using information such as individual beaked whale interclick interval characteristics. Results from nine test files (90 min of data from US Navy instrumented test ranges AUTEC and SCORE) show good performance on eight of the nine test files. The ninth test file (short finned pilot whale) exhibited 930 (assumed) incorrect calls out of 9893 screener detections (9.4%). Over 70 000 min of data from the US Navy Pacific Missile Range Facility were also processed with third stage detections ranging from 3423 to 84 240 beaked whale click detections (depending on temporal processing utilized) from over 13×106 screener detections.

Detection and Localization of Magazine Insertion Clicks in Various Environmental Noises

The early detection and accurate localization of threatening sounds is vital for the survival of military personnel. The purpose of the present study was to investigate the detection, localization, and response time to an M-16 rifle magazine insertion click in three background conditions: jungle, pink, and ambient noise. The level of the click varied from -18 to 0 dB signal-to-noise ratio (SNR). Reported results indicate that detection rates of 50, 70, and 100% required SNRs of −6 to −12, −3, and 0 dB, respectively. Localization accuracy of 50 and 80% within a ±15° angle required SNRs of about −6 and 0 dB, respectively; 100% accuracy was never achieved. Response time was about 2.0–2.5 s for high SNRs but increased gradually at poorer SNRs. The military relevance of the reported findings is discussed.