A multi-factor knuckle and nail bed verification tool for forensic imagery analysis.

BACKGROUND: The grooming process involves sexually explicit images or videos sent by the offender to the minor. Although offenders may try to conceal their identity, these sexts often include hand, knuckle, and nail bed imagery. OBJECTIVE: We present a novel biometric hand verification tool designed to identify online child sexual exploitation offenders from images or videos based on biometric/forensic features extracted from hand regions. The system can match and authenticate hand component imagery against a constrained custody suite reference of a known subject by employing advanced image processing and machine learning techniques. DATA: We conducted experiments on two hand datasets: Purdue University and Hong Kong. In particular, the Purdue dataset collected for this study allowed us to evaluate the system performance on various parameters, with specific emphasis on camera distance and orientation. METHODS: To explore the performance and reliability of the biometric verification models, we considered several parameters, including hand orientation, distance from the camera, single or multiple fingers, architecture of the models, and performance loss functions. RESULTS: Results showed the best performance for pictures sampled from the same database and with the same image capture conditions. CONCLUSION: The authors conclude the biometric hand verification tool offers a robust solution that will operationally impact law enforcement by allowing agencies to investigate and identify online child sexual exploitation offenders more effectively. We highlight the strength of the system and the current limitations.

Weighted pressure matching with windowed targets for personal sound zones.

Personal sound zones (PSZ) systems use an array of loudspeakers to render independent audio signals to multiple listeners within a room. The performance of a PSZ system, designed using weighted pressure matching, depends on the selected target responses for the bright zone. In reverberant environments, the target responses are generally chosen to be the room impulse responses from one of the loudspeakers to the control points in the selected bright zone. This approach synthesizes the direct propagation component and all the reverberant components in the bright zone, while minimizing the energy in the dark zone. We present a theoretical analysis to show that high energy differences cannot be achieved for the diffuse reverberant components in the bright and dark zones, and so trying to synthesize these components in the bright zone does not lead to the best performance. It is then shown that the performance can be improved by using windowed versions of these measured impulse responses as target signals, in order to control which reverberant components are synthesized in the bright zone and which are not. This observation is supported by experimental measurements in two scenarios with different levels of reverberation.

Superposition of the uncertainties in acoustic responses and the robust design of active control systems.

The use of virtual sensing allows the frequency range of a local active noise control system located close to a listener's ears to be extended beyond what is possible when only controlling from remote physical sensors, particularly if head tracking is also used to determine the position of the virtual sensors. As the frequency range is extended, however, the uncertainties in the acoustic responses become more significant, and the design of multichannel adaptive controllers that are robustly stable to these uncertainties becomes more important. In order to fully characterise the uncertainties, a very large number of measurements would, in principle, need to be taken, due to the combination of all the possible changes in the acoustic environment. For uncertainties due to the simultaneous change in position of several objects within the acoustic environment, however, it is shown that the uncertainties can be accurately predicted by the superposition of these uncertainties, due to the change in position of the objects individually. This allows the uncertainty that is due to the change in position of a number of objects to be rapidly evaluated from a limited number of experiments and considerably simplifies the controller design process, which is illustrated here for an active headrest system using two different virtual sensing techniques.

Cochlea-inspired design of an acoustic rainbow sensor with a smoothly varying frequency response.

A number of physical arrangements for acoustic rainbow sensors have been suggested, where the aim is to separate different frequency components into different physical locations along the sensor. Although such spatial discrimination has been achieved with several designs of sensor, the resulting frequency responses at a given position along the sensor are generally not smoothly varying. In contrast, the cochlea provides an interesting natural example of a rainbow sensor, which has an exponential frequency distribution and whose response does vary smoothly with frequency. The design of a rainbow sensor is presented that has a number of discrete resonators and an exponential frequency distribution. We discuss the conditions for a smoothly varying frequency response in such a sensor, as part of a broader design strategy. It is shown that the damping within the resonators determines the trade-off between the frequency resolution and the number of elements required to achieve a smooth response. The connection is explained between this design and that of an effective acoustic absorber. The finite number of hair cells means that the cochlea itself can be thought of as being composed of discrete units and the conditions derived above are compared with those that are observed in the cochlea.

Active control of broadband sound through the open aperture of a full-sized domestic window.

Shutting the window is usually the last resort in mitigating environmental noise, at the expense of natural ventilation. We describe an active sound control system fitted onto the opening of the domestic window that attenuates the incident sound, achieving a global reduction in the room interior while maintaining natural ventilation. The incident sound is actively attenuated by an array of control modules (a small loudspeaker) distributed optimally across the aperture. A single reference microphone provides advance information for the controller to compute the anti-noise signal input to the loudspeakers in real-time. A numerical analysis revealed that the maximum active attenuation potential outperforms the perfect acoustic insulation provided by a fully shut single-glazed window in ideal conditions. To determine the real-world performance of such an active control system, an experimental system is realized in the aperture of a full-sized window installed on a mockup room. Up to 10-dB reduction in energy-averaged sound pressure level was achieved by the active control system in the presence of a recorded real-world broadband noise. However, attenuation in the low-frequency range and its maximum power output is limited by the size of the loudspeakers.

Active control of the sound power scattered by a locally-reacting sphere.

Active control of the sound power scattered by a sphere is theoretically investigated using spherical harmonic expansions of the primary and secondary fields. The sphere has a surface impedance that is uniform, real, and locally reacting while being subjected to an incident monochromatic plane wave. The scattered power is controlled with a number of monopole sources, initially on the surface of the sphere, and is expressed as the sum of squared amplitudes of the spherical harmonics due to both the scattered and control fields. This quadratic function is minimized to identify the optimal strengths for different numbers of control sources. At low frequencies, the scattered field is dominated by the first few spherical harmonic terms, and its power can be significantly reduced with a single controlling monopole for a soft or absorbent sphere and with two monopoles for a hard sphere. The number of secondary sources required to significantly attenuate the scattered field at higher frequencies is found to be proportional to the square of the frequency, and the attenuation also falls off rapidly if the secondary sources are moved away from the surface of the sphere, no matter what its surface impedance.

Estimation of the pressure at a listener's ears in an active headrest system using the remote microphone technique.

The remote microphone technique is considered in this paper as a way of estimating the error signals at a listener's ears in an active headrest system using remotely installed monitoring microphones. A least-squares formulation for the optimal observation filter is presented, including a regularization factor that is chosen to satisfy both the estimation accuracy and robustness to uncertainties. The accuracy of the nearfield estimation is first investigated for a diffuse field via simulations. Additionally, simulations of a free field are also used to investigate the effect of the spatial directivity of the primary field. Finally, experiments in an anechoic chamber are conducted with 24 monitoring microphones around a dummy head positioned in an active headrest system. When six loudspeakers driven by uncorrelated random disturbances are used to generate the primary field, the best arrangement of monitoring microphones is considered, taking into account both accuracy and robustness.

Application of the remote microphone method to active noise control in a mobile phone.

Mobile phones are used in a variety of situations where environmental noise may interfere with the ability of the near-end user to communicate with the far-end user. To overcome this problem, it might be possible to use active noise control technology to reduce the noise experienced by the near-end user. This paper initially demonstrates that when an active noise control system is used in a practical mobile phone configuration to minimise the noise measured by an error microphone mounted on the mobile phone, the attenuation achieved at the user's ear depends strongly on the position of the source generating the acoustic interference. To help overcome this problem, a remote microphone processing strategy is investigated that estimates the pressure at the user's ear from the pressure measured by the microphone on the mobile phone. Through an experimental implementation, it is demonstrated that this arrangement achieves a significant improvement in the attenuation measured at the ear of the user, compared to the standard active control strategy. The robustness of the active control system to changes in both the interfering sound field and the position of the mobile device relative to the ear of the user is also investigated experimentally.

Head tracking extends local active control of broadband sound to higher frequencies.

Local active sound control systems provide useful reductions in noise within a zone of quiet which only extends to about one tenth of an acoustic wavelength. If active control is required above a few hundred hertz, this generally limits the movement of a listener to unrealistically small changes in head position. We describe a local active sound control system using a fixed array of monitoring microphones, in which the pressures at the ear positions are estimated from these microphone signals using head position information from an optical head tracker. These signals are then actively controlled to give robust attenuation at the ear positions, even as the listener moves their head. Feedforward control provides selective attenuation of noise and broadband attenuation of around 20 dB is measured up to excitation frequencies of 1 kHz under favourable conditions, with head tracking achieved in a few seconds. The active control performance is thus comparable with that achieved with active headphones, but without the listener having anything attached to their head.

An elemental approach to modelling the mechanics of the cochlea.

The motion along the basilar membrane in the cochlea is due to the interaction between the micromechanical behaviour of the organ of Corti and the fluid movement in the scalae. By dividing the length of the cochlea into a finite number of elements and assuming a given radial distribution of the basilar membrane motion for each element, a set of equations can be separately derived for the micromechanics and for the fluid coupling. These equations can then be combined, using matrix methods, to give the fully coupled response. This elemental approach reduces to the classical transmission line model if the micromechanics are assumed to be locally-reacting and the fluid coupling is assumed to be entirely one-dimensional, but is also valid without these assumptions. The elemental model is most easily formulated in the frequency domain, assuming quasi-linear behaviour, but a time domain formulation, using state space method, can readily incorporate local nonlinearities in the micromechanics. Examples of programs are included for the elemental model of a human cochlea that can be readily modified for other species.

Fitting pole-zero micromechanical models to cochlear response measurements.

An efficient way of describing the linear micromechanical response of the cochlea is in terms of its poles and zeros. Pole-zero models with local scaling symmetry are derived for both one and two degree-of-freedom micromechanical systems. These elements are then used in a model of the coupled cochlea, which is optimised to minimise the mean square difference between its frequency response and that measured on the basilar membrane inside the mouse cochlea by Lee, Raphael, Xia, Kim, Grillet, Applegate, Ellerbee Bowden, and Oghalai [(2016) J. Neurosci. 36, 8160-8173] and Oghalai Lab [(2015). https://oghalailab.stanford.edu], at different excitation levels. A model with two degree-of-freedom micromechanics generally fits the measurements better than a model with single degree-of-freedom micromechanics, particularly at low excitations where the cochlea is active, except post-mortem conditions, when the cochlea is passive. The model with the best overall fit to the data is found to be one with two degree-of-freedom micromechanics and 3D fluid coupling. Although a unique lumped parameter network cannot be inferred from such a pole-zero description, these fitted results help indicate what properties such a network should have.

Combining the remote microphone technique with head-tracking for local active sound control.

This paper describes practical integration of the remote microphone technique with a head-tracking device in a local active noise control system. The formulation is first reviewed for the optimized observation filter and nearfield pressure estimation. The attenuation performance and stability of an adaptive active headrest system combined with the remote microphone technique are then studied. The accuracy of the nearfield estimation and the effect of the head-tracking on the control performance are investigated in real-time experiments. The regularization factor of the observation filter is selected as a trade-off between its accuracy and its robustness. The integrated active headrest system is used to estimate and attenuate disturbance signals at a listener's ears from a single tonal primary source, while a commercial head-tracking device detects and provides the real-time head position to the active headrest system whose responses are updated accordingly.

A linearly tapered box model of the cochlea.

A box shape with constant area is often used to represent the complex geometry in the cochlea, although variation of the fluid chambers areas is known to be more complicated. This variation is accounted for here by an "effective area," given by the harmonic mean of upper and lower chamber area from previous measurements. The square root of this effective area varies linearly along the cochleae in the investigated mammalian species. This suggests the use of a linearly tapered box model in which the fluid chamber width and height are equal, but decrease linearly along its length. The basilar membrane (BM) width is assumed to increase linearly along the model. An analytic form of the far-field fluid pressure difference due to BM motion is derived for this tapered model. The distributions of the passive BM response are calculated using both the tapered and uniform models and compared with human and mouse measurements. The discrepancy between the models is frequency-dependent and becomes small at low frequencies. The tapered model developed here shows a reasonable fit to experimental measurements, when the cochleae are cadaver or driven at high sound pressure level, and provides a convenient way to incorporate cochlear geometrical variations.

Modelling the effect of round window stiffness on residual hearing after cochlear implantation.

Preservation of residual hearing after cochlear implantation is now considered an important goal of surgery. However, studies indicate an average post-operative hearing loss of around 20 dB at low frequencies. One factor which may contribute to post-operative hearing loss, but which has received little attention in the literature to date, is the increased stiffness of the round window, due to the physical presence of the cochlear implant, and to its subsequent thickening or to bone growth around it. A finite element model was used to estimate that there is approximately a 100-fold increase in the round window stiffness due to a cochlear implant passing through it. A lumped element model was then developed to study the effects of this change in stiffness on the acoustic response of the cochlea. As the round window stiffness increases, the effects of the cochlear and vestibular aqueducts become more important. An increase of round window stiffness by a factor of 10 is predicted to have little effect on residual hearing, but increasing this stiffness by a factor of 100 reduces the acoustic sensitivity of the cochlea by about 20 dB, below 1 kHz, in reasonable agreement with the observed loss in residual hearing after implantation. It is also shown that the effect of this stiffening could be reduced by incorporating a small gas bubble within the cochlear implant.

Finite-element model of the active organ of Corti.

The cochlear amplifier that provides our hearing with its extraordinary sensitivity and selectivity is thought to be the result of an active biomechanical process within the sensory auditory organ, the organ of Corti. Although imaging techniques are developing rapidly, it is not currently possible, in a fully active cochlea, to obtain detailed measurements of the motion of individual elements within a cross section of the organ of Corti. This motion is predicted using a two-dimensional finite-element model. The various solid components are modelled using elastic elements, the outer hair cells (OHCs) as piezoelectric elements and the perilymph and endolymph as viscous and nearly incompressible fluid elements. The model is validated by comparison with existing measurements of the motions within the passive organ of Corti, calculated when it is driven either acoustically, by the fluid pressure or electrically, by excitation of the OHCs. The transverse basilar membrane (BM) motion and the shearing motion between the tectorial membrane and the reticular lamina are calculated for these two excitation modes. The fully active response of the BM to acoustic excitation is predicted using a linear superposition of the calculated responses and an assumed frequency response for the OHC feedback.

Efficient time-domain simulation of nonlinear, state-space, transmission-line models of the cochlea (L).

Nonlinear models of the cochlea are best implemented in the time domain, but their computational demands usually limit the duration of the simulations that can reasonably be performed. This letter presents a modified state space method and its application to an example nonlinear one-dimensional transmission-line cochlear model. The sparsity pattern of the individual matrices for this alternative formulation allows the use of significantly faster numerical algorithms. Combined with a more efficient implementation of the saturating nonlinearity, the computational speed of this modified state space method is more than 40 times faster than that of the original formulation.

Modeling local active sound control with remote sensors in spatially random pressure fields.

A general formulation is presented for the optimum controller in an active system for local sound control in a spatially random primary field. The sound field in a control region is selectively attenuated using secondary sources, driven by reference sensors, all of which are potentially remote from this control region. It is shown that the optimal controller is formed of the combination of a least-squares estimation of the primary source signals from the reference signals, and a least-squares controller driven by the primary source signals themselves. The optimum controller is also calculated using the remote microphone technique, in both the frequency and the time domains. The sound field under control is assumed to be stationary and generated by an array of primary sources, whose source strengths are specified using a spectral density matrix. This can easily be used to synthesize a diffuse primary field, if the primary sources are uncorrelated and far from the control region, but can also generate primary fields dominated by contributions from a particular direction, for example, which is shown to significantly affect the shape of the resulting zone of quiet.

Comparing methods of modeling near field fluid coupling in the cochlea.

As well as generating the far field pressure, which allows wave propagation in the cochlea, the vibration of an individual element of the basilar membrane (BM) will also generate a near field pressure, which increases its mass and gives rise to local longitudinal coupling. This paper compares the efficiency and accuracy of a number of different methods of calculating the near field pressure distribution, and explores the connections between them. In particular it is shown that a common approximation to the wavenumber description of the near field pressure is equivalent, in the spatial domain, to an exponential decay away from the point of excitation. Two important properties of the near field pressure are its maximum amplitude, which is finite if the vibrating element has a finite length, and the value of its spatial integral, which determines the added mass on the BM due to the fluid loading. These properties are calculated as a function of the BM width relative to the width of the fluid chamber. By parameterizing the near field pressure variation in this way, it can be readily incorporated into coupled models of the cochlea, without the considerable computational expense of calculating the full three dimensional pressure field.

Modified protein expression in the tectorial membrane of the cochlea reveals roles for the striated sheet matrix.

The tectorial membrane (TM) of the mammalian cochlea is a complex extracellular matrix which, in response to acoustic stimulation, displaces the hair bundles of outer hair cells (OHCs), thereby initiating sensory transduction and amplification. Here, using TM segments from the basal, high-frequency region of the cochleae of genetically modified mice (including models of human hereditary deafness) with missing or modified TM proteins, we demonstrate that frequency-dependent stiffening is associated with the striated sheet matrix (SSM). Frequency-dependent stiffening largely disappeared in all three TM mutations studied where the SSM was absent either entirely or at least from the stiffest part of the TM overlying the OHCs. In all three TM mutations, dissipation of energy is decreased at low (<8 kHz) and increased at high (>8 kHz) stimulus frequencies. The SSM is composed of polypeptides carrying fixed charges, and electrostatic interaction between them may account for frequency-dependent stiffness changes in the material properties of the TM. Through comparison with previous in vivo measurements, it is proposed that implementation of frequency-dependent stiffening of the TM in the OHC attachment region facilitates interaction among tones, backward transmission of energy, and amplification in the cochlea.

Modelling the effects of cochlear implant current focusing.

OBJECTIVES: A finite element model of the human coiled cochlea was used to model the voltage distribution due to stimulation by the individual electrodes of a cochlear implant. The scalar position of the electrode array was also varied in order to investigate its effect on the voltage distribution. Multi-electrode current focusing methods were then investigated, with the aim of increasing spatial selectivity. METHODS: Simultaneous current focusing is initially achieved, as in previous publications, by calculating the input currents to the 22 electrodes that best separates the voltages at these electrode positions. The benefits of this electrode focusing strategy do not, however, entirely carry over to the predicted voltage distributions at the position of the spiral ganglion cells, where excitation is believed to occur. A novel focusing strategy is then simulated, which compensates for the impedances between the currents at the electrode sites and the voltage distribution directly at the position of the spiral ganglion cells. RESULTS: The new strategy produces much better focusing at the sites of the spiral ganglion cells, as expected, but at the cost of increased current requirements. Regularization was introduced in order to reduce current requirements, which also reduced the sensitivity of the solution to uncertainties in the impedance matrix, so that improved focusing was achieved with similar current requirements to that of electrode focusing. DISCUSSION: Although such focusing strategies cannot be achieved in practice at the moment, since the responses from the electrodes to the neural sites cannot be determined with currently available recording methods, these results do support the feasibility of a more effective focusing strategy, which may provide improved spectral resolution leading to improved perception of sound.