Environmental and polymeric acoustic dosimetry

Abstract

The invention relates to environmental and polymeric acoustic dosimetry. The invention provides a digital signal processing method for acoustic dosimetry of ambient sound. A strength of a microphone signal generated by a watch worn by a user is determined and converted into a sound sample having units for sound noise exposure. It is repeated multiple times to produce a time sequence of sound samples that is written to a secure database to which access is authorized by the user. Other aspects are also described and claimed to be protected.

Description

[0001] This application is a divisional application of the divisional application with application number 202111660986.1, application date June 1, 2020, and invention name “Environmental and Aggregate Acoustic Dosimetry”. The divisional application with application number 202111660986.1 is a divisional application of the invention patent application with application number 202010482766.3, application date June 1, 2020, and invention name “Environmental and Aggregate Acoustic Dosimetry”. Technical Field

[0002] One aspect of the disclosure herein relates to digital signal processing techniques for monitoring acoustic energy in an ambient listening environment of a user. Other aspects are also described, including a convergent acoustic dosimetry process. Background Art

[0003] Consumer electronic headphones have become increasingly popular with users because they reproduce media such as music, podcasts, and movie soundtracks with high fidelity while not disturbing others nearby. While the listening experience using headphone devices is enjoyable and the maximum sound output of headphone devices is limited according to hearing health safety standards, there is still a need to monitor the sound output of headphone devices over relatively long periods of time (such as days and weeks) as part of personal hearing health monitoring aimed at avoiding long-term exposure to loud sounds. BRIEF DESCRIPTION OF THE DRAWINGS

[0004] Various aspects of the present disclosure are described by way of example and are not limited to the illustrations of the various drawings, in which similar reference numerals indicate similar elements. It should be noted that reference to "one" or "an" aspect in the present disclosure is not necessarily the same aspect, and it means at least one. In addition, for the sake of brevity and to reduce the total number of drawings, a given drawing may be used to illustrate features of more than one aspect of the present disclosure, and for a given aspect, not all elements in the drawing may be required.

[0005] Figure 1 An exemplary audio system having a headphone arrangement and an audio source device is shown.

[0006] Figure 2 is a flow chart of a feed-forward type acoustic dosimetry process for headphone device monitoring.

[0007] Figure 3 An exemplary volume curve determined for an unknown headphone device is shown, relating volume step size to maximum SPL output.

[0008] Figure 4 is a flow chart of the ambient acoustic dosimetry process.

[0009] Figure 5A polymerized dosimetry process receiving multiple inputs from different devices is shown.

[0010] Figure 6 is a schematic diagram of the acoustic dosimetry process using a wind detector as contextual input.

[0011] Figure 7 is a schematic diagram of the acoustic dosimetry process using a water event detector as contextual input. DETAILED DESCRIPTION

[0012] This nonprovisional patent application claims the benefit of the earlier filing date of provisional application 62 / 855,956, filed on June 1, 2019.

[0013] Various aspects of the present disclosure will now be explained with reference to the accompanying drawings. Whenever the shapes, relative positions and other aspects of the components described are not clearly defined, the scope of the present invention is not limited to the components shown, which are only used for illustrative purposes. In addition, although many details are set forth, it should be understood that some aspects of the present disclosure can be implemented without these details. In other cases, well-known circuits, structures and technologies are not shown in detail to avoid obscuring the understanding of the description.

[0014] Figure 1An example of a tactile audio device 1 as part of an audio system is shown, in which an audio signal processing method for acoustic dosimetry of sounds produced by the tactile audio device 1 can be implemented. The tactile audio device 1 shown is an in-ear earbud (an in-ear headphone, which can be a sealed type with a flexible tip, or it can be a non-sealed type, also known as a loose fit). The tactile audio device 1 can be one of the two headphones (left and right) that constitute an earphone device. The method described below can be implemented for one or both of the headphones that constitute the earphone device. Alternative alternatives to the in-ear earbud (not shown) include supra-aural headphones and over-ear headphones. The tactile audio device 1 is shown being used by its user (also referred to as a listener or wearer). The tactile audio device 1 has a tactile sound transducer or speaker 2 (arranged and constructed to reproduce sound directly into the ear of its user), and optionally has an external microphone 3 (arranged and constructed to directly receive ambient sound) and an internal microphone 4 (arranged and constructed to directly receive sound reproduced by the speaker 2). These microphones may be integrated into their respective audio device housings. A microphone is not necessary for some of the methods described below (e.g., in the case of estimating sound exposure due to headphone playback). An internal microphone 4 may be used to accurately measure the in-ear sound pressure level SPL. In addition, there may be more than one external microphone 3, such as a microphone array integrated into a single housing of the in-ear audio device 1 that produces multiple channels (multiple sequences of digital audio frames); any reference here to evaluating the strength of "one" microphone signal is understood to be a more general reference to evaluating the strength of one or more pickup channels, for example, a single output of a single microphone, a multi-channel output of a microphone array, or a single output of a pickup beamforming process that receives a multi-channel output from a microphone array.

[0015] The methods described below are applicable to processing digital output audio signals that are essentially ready to drive the speaker 2. The output audio is the result of an audio rendering process that may begin with obtaining an input audio signal from a streaming music or podcast or movie service via a wired or wireless communication interface, such as over the Internet. The methods described below, as well as the audio rendering process and even the communication interface that delivers the input audio, are performed by various electronic hardware components, all of which may be integrated into the housing of the tactile-ear device 1. Thus, in the case where the housing of the tactile-ear device 1 has sufficient space and power (e.g., from a rechargeable battery), all of the electronic devices that obtain, process, and generate the transducer signals that drive the speaker 2 may be placed in the same housing. The electronic devices may include an audio amplifier that drives the speaker 2 with the output audio signal, an optional microphone sensing circuit or amplifier that receives optional microphone signals and converts them into the desired format for digital signal processing, and one or more digital processors (referred to herein as "processors") and memory (e.g., solid-state electronic data storage circuits), wherein the memory stores instructions for configuring the processor (e.g., instructions to be executed by the processor) to perform the digital signal processing tasks discussed in more detail below.

[0016] It is noted that some or substantially all of the electronics referred to below as implementing the acoustic dosimetry technique may also be located in another device separate from the tactile ear device 1. For example, where the tactile ear device 1 is a left earphone or a right earphone, the earphone may be connected to the tactile ear device 1 via a wired connection (e.g., a computer bus connector that also delivers power, in which case a power source may not be required in the earphone housing) or via a wireless connection (e.g., a Bluetooth link). Figure 1In the example of FIG. 1 , an audio source device 5 is shown as a smart phone. In both cases, the output audio (which contains user content, such as a movie soundtrack, music, or the voice of a far-end user during a phone call) has been rendered by one or more processors (“processors”) in the audio source device 5, and most, if not all, of the processing described below may be performed by (or at) the processor in the audio source device 5. For example, in the audio source device 5, the instructions that configure or program the processor may be part of the operating system software, or they may be part of application software that runs on top of the operating system program. Thus, one or more operations in the acoustic dosimetry process may be performed by or at the processor of a smart phone, smart watch, tablet computer, or other audio source device that may communicate with the headphone device. For example, the headphone device may be a peripheral device of the audio source device, or the headphone device and the audio device may be companion devices, for example, the headphones may be paired with or configured for use with the audio source device, and they may be considered personal devices for a particular user in the sense that both are personally configured for a particular user. The audio source device provides the user content output audio to the headphone device via a wired audio link or via a wireless audio link for playback via the speakers of the headphone device. Alternatively, one or more operations of the acoustic dosimetry process may be performed by a processor in one or both earphone housings of the earphone device or at other locations in the earphone device (provided the earphone device has sufficient computing resources to implement this).

[0017] During headphone playback of user content, the acoustic dosimetry process measures or estimates the in-ear SPL, for example, at or near the tympanic membrane reference point. In one aspect, the in-ear SPL is measured as follows. The in-ear SPL from the internal microphone 4 (see ) that picks up sound in the ear canal can be measured using, for example, laboratory calibration results. Figure 1 ) to an equivalent SPL, the laboratory calibration result includes correction factors to be applied to the microphone signal, such as equalization. This conversion of the microphone signal to the in-ear SPL can be performed during headphone playback (sound conversion is performed by the headphone device worn by the user). Alternatively, if the internal microphone 4 is not available, then Figure 2 The processing flow in can be used to estimate the in-ear SPL in a "feed-forward" manner.

[0018] refer to Figure 2, the process here begins with laboratory measurements using several different ear simulators and human models wearing the manufactured earphone devices to generate sound output sensitivity data for different types of ears and ear canals expected in the user group. In one aspect, a set of nominal output sensitivity data is selected for the expected user group. This factory calibration procedure can be performed for each manufactured earphone device, for example, during production testing, or it can be performed once for a group of similar earphone devices. Statistical techniques can be used to find the best fit for a group of similar earphone devices, and then the best fit is used for a large number of similar earphone devices. In any case, the calibration procedure assigns sound output sensitivity to each earphone device. The sound output sensitivity gives a mathematical relationship between the intensity of the audio signal driving the speaker 2 and the intensity of the resulting sound output of the speaker 2 worn by the expected user. It can vary according to the auditory frequency of interest, or it can be provided as a statistical measure over the entire audible frequency range, for example as a relationship between the average input and the average output. It can be stored as a data structure in a microelectronic memory within the housing of each earphone device. In one aspect, the tactile audio device 1 may provide its stored assigned output sensitivity data to the audio source device, which may include headphone acoustic output sensitivity and volume curve parameters, such as described below.

[0019] The in-ear audio device 1 may then be paired with a separate audio source device 5 (e.g., as a wireless headset device paired with a smartphone, smartwatch, or tablet via a wireless link, such as a Bluetooth link, or as a wired headset device connected to the audio source device via a wired link, such as a serial peripheral bus, such as a Universal Serial Bus, USB connection). A processor in the headset device will then receive the digital output audio via the paired connection with the audio source device, and will then drive the audio amplifier and speakers of the headset device to reproduce the output audio (also referred to as a playback signal, which contains user content, such as media such as music, podcasts, video games, or movie soundtracks) as sound.

[0020] The output sensitivity data may be transmitted from a memory in the tactile ear device 1 to a memory in the audio source device 5, for example via a Bluetooth link, where the output sensitivity data is used by a processor executing an acoustic dosimetry algorithm or program for monitoring by the earphone device. The process is as follows to estimate the in-ear sound pressure level SPL. See Figure 2, a processor in the audio source device 5 calculates a measure of the intensity of the output audio being played back (box 8). The measure can be, for example, a root mean square RMS value. It should be noted that the output audio is the result of an audio rendering process that performs a conventional audio signal processing operation chain on the input playback signal (including media such as music or movie soundtracks). These may include dynamic range adjustment, equalization, and gain adjustment for volume steps. The processor then converts the RMS value of such output audio to an in-ear SPL by applying the received acoustic output sensitivity data (for the currently used headphone device) to the RMS value (multiplying it by the acoustic output sensitivity data). For example, a dB full-scale RMS value is converted to an in-ear SPL dB value. More generally, this is referred to herein as a conversion based on output sensitivity to in-ear SPL (output sensitivity conversion 10).

[0021] Next, in box 11, the measure or estimate of the in-ear SPL is converted into units suitable for evaluating sound noise exposure (e.g., units specified by a hearing health safety standard, which may be a standard or generally defined indicator of permissible sound noise exposure for hearing health). For example, the in-ear SPL may be multiplied by a transfer function (which has been determined in a laboratory environment) that converts the in-ear SPL into an equivalent free field or diffuse field measurement of the sound that would be picked up by a virtual reference microphone located at a certain distance from the user, as defined by the hearing health safety standard. The transfer function may be between the eardrum reference point DRP and the reference microphone. The result of the conversion is referred to herein as a calculated sound sample, for example in units of SPL dBA (A-weighted decibels).

[0022] The sound samples may be repeatedly calculated over time, for example every second or other suitable interval during playback, and then accumulated into a batch or series of sound samples. The batch of sound samples is then written to a database 12 of health data stored in the memory of the audio source device 5. The database 12 may be secure in the sense that individual users or owners of the health data are required to authorize access to it. In addition to the sound samples, metadata, such as the model of the headphone device that reproduces the output audio as sound and the application from which the output audio is generated, may also be written as part of the batch. It should be noted here that the sound samples written to the database cannot be used to recover identification information about the media being played back, such as the title of a musical work or movie.

[0023] The stored or monitored sound samples may then be presented by an application or application 14 (also executed by a processor in the audio source device 5) for visualization on a graphical user interface of the audio source device. Figure 2, a health application as an example of application 14 may be authorized to access a locally stored health database 12 to retrieve sound samples, and calculate various statistical measures of the collected sound samples, such as Leq dBA (average) within a specific time interval. The health application can then "display" to the user their sound exposure due to the playback of the in-ear audio device 1. The health application can also visualize to the user which parts of the sound samples are generated by which applications (such as music applications, video game applications, and movie players), and which models of in-ear audio devices produce which sound samples. It is expected that users may use several different models of headphone devices to listen at different volume steps or with different media, such as in-ear wired earbuds, in-ear wireless earbuds, and on-ear headphones. This available information can be monitored and reported to the user by the health application. Other ways of reporting useful information about such collected sound samples to the user (acoustic dosimetry) are also possible.

[0024] In the event that the factory calibrated output sensitivity information is not available, the processor in the audio source device 5 may still detect that an unknown in-ear audio device is being used to convert the output audio into sound (e.g., detecting that a computer bus connector has been plugged into the audio source device, or that a Bluetooth link has been established with a headphone unit). The processor uses this information to select a volume curve. Figure 3 As shown, a volume curve can be selected that best fits the characteristics of a number of different types of commercially available headphone devices (and their published sound output characteristics). These headphone devices may have been tested to measure the relationship between their sound output (e.g., in-ear SPL) and volume step size, and at the same time a nominal audio signal processing chain (e.g., specific to a particular application that produces the current output audio in the audio source device) is applied. The current volume step size (at which the current output audio is produced) is then applied to the selected volume curve to obtain an estimate of the maximum SPL assigned to the connected unknown headphone device at that volume step size. An output sensitivity relationship between RMS and in-ear SPL derived for the unknown headphone device is then determined based on the maximum SPL given by the selected volume curve. The derived output sensitivity relationship is then inserted into the output sensitivity relationship described above and in Figure 2 In the processing flow shown in , a sound sample is calculated in SPL dBA units (or other units suitable for assessing sound exposure, such as noise exposure calculations based on standard or commonly defined hearing health sound standards).

[0025] It should be noted that the in-ear SPL measurements or estimates obtained as above should be adjusted in the presence of acoustic energy generated by the loudspeaker 2 from outside the playback signal. For example, the headphone device may have an acoustic noise cancellation ANC capability, an ambient sound transparency capability (wherein ambient sound is picked up during playback and actively reproduced to provide the wearer with a sense of awareness of its environment), or an ambient sound enhancement capability (wherein ambient sound is picked up and amplified when reproduced in order to compensate for hearing loss or meet the hearing preferences of the wearer). In such cases, the headphone device may electronically identify such capabilities during initial pairing with the audio source device, in response to which the acoustic dosimetry process may be configured to adjust its in-ear SPL measurements or estimates accordingly to account for any such added acoustic energy. Alternatively, since such additional acoustic energy is picked up to some extent by the internal microphone 4, it will be inherently taken into account in the in-ear SPL measurement made using the signal from the internal microphone 4.

[0026] Now go to Figure 4 , which shows an example of how to perform acoustic dosimetry on the sounds in the user's surroundings (also called ambient sound or noise exposure), such as sounds from speakers, vehicle sounds, and sounds from any other source in the surrounding environment. Figure 1 The external microphone, which was shown in the example of the present invention as being embedded on the outer surface of the earphone housing and can therefore perform direct sound pickup in the surrounding environment, is now embedded or integrated in the housing of a smart watch 16, which is worn on the user's wrist, as shown. The smart watch 16 can be paired for wireless digital communication with an audio source device 5 (e.g., a user's smartphone or tablet). A calculation is now performed on the microphone signal generated by the audio device worn by the user, which is similar to the calculation described above in Figure 2 The calculations described in are performed on the output audio to determine its strength (e.g., RMS). For example, Figure 4 This signal generated by the external microphone 3 is shown. It is noted here that there may be more than one external microphone 3, such as a microphone array integrated into a single housing of the audio source device 5 or smartwatch 16, producing multiple channels as a sequence of multiple digital audio frames; any reference here to evaluating the strength of "one" microphone signal is understood to be a more general reference to evaluating the strength of one or more pickup channels, for example, a single output of a single microphone, a multi-channel output of a microphone array, or a single output of a pickup beamforming process that receives a multi-channel output from a microphone array.

[0027] The desired result of ambient acoustic dosimetry is to calculate a sound sample of the ambient environment by converting the strength of the microphone signal into appropriate units, such as defined by the Health Hearing Safety Standard, for example, in units of SPL dBA. In one aspect, these calculations are performed by a low-power or auxiliary processor integrated in the housing of the smart watch 16, which helps to reduce power consumption (an important goal for smaller, lower energy density batteries that power smart watches). An additional benefit of having the sound sample calculations performed by the low-power or auxiliary processor (which is also used to detect voice trigger phrases in the signal from the external microphone 3) is that the external microphone signal is only buffered for a short time interval (e.g., about one second) and therefore does not capture any private conversations that may be present in the ambient environment. The low-power processor or auxiliary processor is a processor that continuously processes the external microphone signal to detect a voice trigger, while the high-power processor or main processor in the smart watch is in a low-power state (e.g., a sleep state). Once a voice trigger is detected, the high-power processor or main processor in the smart watch transitions to a high-power state (e.g., a wake-up state) in which it can perform the more power-intensive tasks of the virtual assistant program.

[0028] Figure 4 The process flow for ambient / environmental acoustic dosimetry is compared to Figure 2 There are some differences in the headphone device acoustic dosimetry or headphone audio exposure presented in . These differences include replacing the user content playback signal with the external microphone signal, replacing the output sensitivity conversion 10 with the input sensitivity conversion 18 (e.g., a conversion that relates the dB full scale of the signal produced by the external microphone 3 to the SPL incident on the external microphone 3), and modifying block 11 to use the external microphone 3 instead of the reference microphone of the DRP and hearing health safety standard (which is in Figure 4 The free field or diffuse field transfer function between φ(x) and φ(y) is shown as a dashed line in FIG.

[0029] Figure 4 The aspects shown in also include storing the health database 12 to which the sound samples are written in batches as described above in the memory of the smart watch 16, rather than in the memory of the audio source device 5. In addition, a visualization application 17 is also executed by the high-power processor or main processor in the smart watch 16, which generates a visual effect that will be displayed to the user on the screen of the smart watch, showing the level of ambient sound exposure and its corresponding duration. The visualization application 17 can also generate a visual or tactile notification 19 in the smart watch 16 when it determines that the collected sound samples exceed a loudness threshold (the sound is too loud for too long) within a time interval threshold, which thresholds may be defined by hearing health safety standards. The notification 19 notifies the user that their surrounding environment may be too loud for hearing health, and is also referred to herein as a "loud environment" notification.

[0030] Another aspect of the present disclosure is an "aggregate" acoustic dosimetry process that can be performed by a programmed digital processor, which estimates the overall sound or hearing exposure ("overall sound exposure") for a particular user. The calculation of the overall sound exposure is a holistic view in the sense that the processor collects sound measurement inputs from microphones in multiple audio devices, each of which may be, for example, within a radius of two meters from the user. Each of the audio devices has an independent housing (containing its corresponding one or more microphones) and is a different type of device, such as a desktop computer, a laptop computer, a tablet computer, a smart phone, a smart watch, and a headphone device. The audio device may be associated with a particular user in the sense of how smart phones, smart watches, and headphone devices are typically described as being "owned" by a particular user (e.g., configured to be personal to the user). One or more operations of the aggregate acoustic dosimetry process may be performed by a processor of a smart phone, smart watch, tablet computer, or other audio source device that can communicate with the headphone device. For example, the earphone device may be a peripheral device of the audio source device, or the earphone device and the audio device may be companion devices, for example, the earphone may be paired with the audio source device or configured for use with the audio source device, and they may be considered personal devices of a particular user in the sense that both are personally configured for a particular user. The audio source device provides user content output audio to the earphone device via a wired audio link or via a wireless audio link for playback through the speaker of the earphone device. Alternatively, one or more of these operations may be performed by a processor in one or both earphone housings of the earphone device or in other locations in the earphone device (provided that the earphone device has sufficient computing resources to implement and is able to communicate with the user's other devices to obtain the inputs required for the aggregation dose determination process). The aggregation process may interpret the sound measurement values ​​using other inputs (referred to herein as contextual inputs), which are sources of information that reference or clarify the situation in which the user finds himself (during a given sound measurement). The overall sound exposure is then stored as a time series of in-ear SPL values ​​(e.g., in a secure database that a particular user is authorized to access). Therefore, the overall sound exposure indicates the actual sound level to which the user's eardrum is exposed over a relatively long period of time (such as hours, days, or weeks).

[0031] like Figure 5As shown, the sound measurement inputs to the aggregate dosimetry process may include two or more of the following occurring at different times when the user is exposed to various sound levels: estimates or measurements of headphone playback sound; sound measurements of ambient noise made by an external microphone 3 integrated into the user's smartwatch 16; sound measurements made by an internal microphone 4 integrated into an in-ear audio device 1 (such as headphones), which is worn by the user and may represent i) headphone playback sound, or ii) ambient acoustic noise that has leaked into the user's ear through the headphones, or both; sound measurements of ambient noise made by an external microphone 3 integrated into the in-ear audio device 1; ambient noise measurements made by an external microphone 3 integrated into a smart phone or another audio source device 5 belonging to the user, such as during a phone call; and sound measurements of ambient noise made by an external microphone 3 of an audio device such as a desktop computer 19 or laptop computer or tablet computer. The processor may then convert these different sound measurement inputs into a common format, such as an in-ear SPL or hearing health safety standard (see, e.g., above). Figure 2 and Figure 4 ). The converted or common format SPL values ​​are stored as a time series of sound samples, also referred to herein as an aggregated dosimetry time series reflecting the actual in-ear SPL over time.

[0032] As part of the aggregate dosimetry process, in some cases, the processor may execute an algorithm to determine which of two or more simultaneously arriving inputs (e.g., time-stamped within the same time interval, or received simultaneously) will better or more accurately reflect the actual in-ear SPL. In one case, the input with the highest confidence level is selected, and other simultaneous inputs are ignored. In other cases, one of the sound measurement inputs is weighted more heavily than the other sound measurement inputs before being combined into a single input value (e.g., estimated in-ear SPL). For example, consider the case where a headphone playback estimate is received simultaneously with an ambient noise measurement, the headphone playback estimate being based on the current user-settable volume at which certain user content, such as a movie soundtrack or music, is played back through the headphones; in this case, the user is simultaneously exposed to various types of sounds because they are both listening to music playback on their headphones, but are also in a loud ambient environment that may still be heard despite passive attenuation by the worn headphone device.

[0033] The processor stores a time series of such estimated in-ear SPL values, which reflects the user's hearing exposure changing over time (e.g., over a period of hours, days, or weeks). The processor may analyze the time series to determine when the in-ear SPL exceeds a given level threshold within a given time interval, in response to which the processor signals that a "loud sound" notification 19 is immediately generated and presented to the user. In general, in response to the estimated in-ear SPL exceeding the level threshold during a short-term time range or interval (e.g., seconds) and during a long-term time range (e.g., an hour, hours, a day, days, a week), a loud sound notification 14 may be generated in one or more devices of the user. The loud sound notification 14 may, for example, include a tactile alarm on the user's smart watch 16 or on his or her smart phone or other audio source device 5. In addition to the tactile alarm, the loud sound notification 14 may also include a simultaneous voice alarm (e.g., "Consider reducing the volume by double-clicking") or an audible alarm (e.g., a warning sound) that can be played back through the user's headphones, smart phone, tablet, or laptop speakers. The loudness notification 14 may also include a concurrent visual alert, such as a pop-up menu on a display screen of any one or more of the user's devices.

[0034] A time series of sound samples (e.g., estimated in-ear SPL values) generated by an aggregate dosimetry process may be tagged (by a processor) with metadata specifying whether a particular time interval of the sound samples was primarily due to headphone playback or primarily due to ambient noise. The time interval and its metadata may be interpreted by the dosimetry process to determine that during the particular time interval, the user's ambient or environmental noise exceeded a given level threshold and a given duration threshold. Alternatively, the sequence and metadata may be interpreted by a processor to determine that during the particular time interval, the headphone playback exceeded a given level threshold and a given duration threshold. In response, in each case, a loud sound notification may be signaled.

[0035] The time series of sound samples may cover a period of hours, days, or weeks. Location information (as defined by, for example, a global positioning system, GPS, coordinates, or other means for defining a geographic location or location of a user) may be associated with or tagged with the time series. When tagged with location information, the time series of sound samples may be used by a processor to generate a map of locations that it has determined to be loud (e.g., where it has been determined that the ambient noise level and its duration exceeds hearing health standards).

[0036] In another aspect of the aggregate dosimetry process, the processor determines or records information about why or how the loud sound measurement (the in-ear SPL exceeded the level threshold and the duration threshold) occurred. Such contextual information may be received as one or more contextual inputs to the process (see Figure 5 ). For example, now refer to Figure 6 , one of the contextual inputs may be the output of a wind detector (a wind detection algorithm executed by a digital processor). In box 36, the wind detector processes each digital audio frame (e.g., 10 to 50 milliseconds long) of a sequence of received digital audio frames of a pickup channel (e.g., the output of a single microphone, or the output of a pickup beamforming process that receives multi-channel outputs from a microphone array) to generate an indication, such as a confidence level, a binary yes / no, or an estimated wind speed, as to whether a given frequency component in a particular audio frame is due to wind (rather than some ambient sound source). This contextual input from the wind detector may then be used by the aggregate dosimetry process in box 39 to decide whether to record the sound measurements received simultaneously in box 37, or to discard or alternatively reduce their weight as being due to wind (rather than due to loud sounds generated by sound sources in the surrounding environment).

[0037] To further ensure the accuracy of the calculated overall sound exposure, the aggregate acoustic dosimetry process may determine whether the user has active hearing protection and passive hearing protection in place during the sound measurement, and apply attenuation or reduction to the sound level measurements based on this. The active hearing protection may be acoustic noise canceling ANC, where an acoustic transducer (e.g., earpiece speaker, speaker in a headphone device) is driven to produce anti-noise that cancels ambient noise that would otherwise be heard by the user.

[0038] Passive hearing protection may include earplugs 20 (see Figure 5 ), and ear occlusion due to an in-ear audio device 1 such as headphones or due to a smartphone being held against the user's ear, and the result of attenuating ambient noise due to a partial or complete acoustic seal against the ear. In some cases, the processor may have predetermined stored knowledge of the amount to subtract from the sound measurements, the predetermined stored knowledge being based on knowledge of the particular type of passive hearing protection used in a given instance (e.g., sealing earplugs, loose fitting earplugs, and headphone arrangements, or for in-ear use with smartphones, all of which may be provided by the same entity that also provides the aggregate dosimetry process.) In other cases, the user may wear third-party hearing protection whose attenuating effect is unknown to the user and unknown to the provider of the acoustic dosimetry process.

[0039] Part of the aggregate dosimetry process can be a process for tracking usage of passive hearing protection (by a particular user) over time and recording such usage over time, wherein the usage of passive hearing protection is tracked such as whether the user is using the passive hearing protection at a concert, and the type of hearing protection (such as a pair of headphones or a pair of specialized earplugs), and what level of attenuation the hearing protection is designed to provide. This produces a time series of attenuation values ​​that can be applied to reduce the synchronized time series of ambient sound exposure measurements and headphone sound exposure measurements to produce a more accurate "net" sound exposure sample. In one case, the process can prompt the user to manually select from a list of the most popular hearing protection, including third-party hearing protection that is different from those hearing protection they are currently wearing from the provider of the aggregate dosimetry process.

[0040] Another difficulty in determining the correct amount of passive attenuation is the uncertainty of how the hearing protection physically fits the user's ear. For example, the fit of headphones or specialized earplugs to the user's ear strongly affects the amount of passive attenuation. When determining such attenuation values, care should be taken to avoid underestimating the sound dose to which the user is exposed.

[0041] Acoustic dosimetry based on processing touch images to discard apparent touch caused by water

[0042] Figure 5 Another input to the aggregate dosimetry process in can be the output of a water event detector (a water detection algorithm executed by a digital processor). The water event detector outputs an indication of the presence of water on a particular audio device of the user. The audio device can be, for example, a user's smart watch worn by the user while the user is in the shower or outside on a rainy day, or a user's smartphone or other portable consumer electronic device that may encounter water in the surrounding environment. This is used to provide context to the dosimetry process for when ambient sound level measurements made by the audio device are not accepted as valid samples. This is because the external acoustic port in the device is likely to be hit by water or continuously covered with water, so a microphone "listening" through that port is likely to generate an erroneous sound level measurement at that point in time.

[0043] Reference now Figure 7, which is a flow chart of a method for acoustic dosimetry in which a water event detector provides contextual input. The method may be performed by a digital processor and touch sensor subsystem within an audio device such as a smartwatch or smart phone. A digital touch image is acquired from a touch-sensitive surface of the touch sensor subsystem (e.g., a touch screen in the device) and digitally processed to discard apparent touches caused by water (box 35). An ambient sound level measurement is made using a microphone in the device (box 37). When mixed with other sound measurement inputs, a determination is made as to whether the ambient sound level measurement should be recorded as a valid sound exposure sample for acoustic dosimetry, or whether its weight should be reduced (box 38). This is done based on the results of the processed touch image, such as determining how much (if any) of the touch image is caused by water. In other words, the touch sensing surface in the device is being used to detect water events, which becomes a context for when measurements of ambient sound levels are not recorded (when using an "open to the environment" microphone within the audio device).

[0044] In one aspect, processing the touch image includes distinguishing between i) floating (or electrically ungrounded) water on the touch-sensitive surface and ii) a finger touch (which is electrically grounded). In another aspect, processing the touch image includes determining the presence of water on the touch-sensitive surface and classifying the determined water on the touch-sensitive surface as static (e.g., one or more droplets) or dynamic (e.g., the water is flowing, as when a user wears the device in the shower). In this aspect, if the water is classified as dynamic, the dosimetry process does not record the ambient sound level measurement, and if the water is classified as static, the ambient sound level measurement is recorded. In other words, if the water is classified as static, the ambient sound level measurement is recorded, but if the water is classified as dynamic, it is not recorded.

[0045] In yet another aspect, processing the touch includes determining an amount of water on the touch-sensitive surface by, for example, counting a number of pixels in the touch image by a processor. If the determined amount of water is greater than a threshold, then an ambient sound level measurement is not recorded, and if the determined amount of water is less than the threshold, then an ambient sound level measurement is recorded. If the recorded ambient sound level measurement remains above the threshold for at least a threshold amount of time, then the dosimetry process can send a loud ambient notification to its user.

[0046] The touch-sensitive surface and microphone may be integrated into the housing of the portable consumer electronic device. In the case where the water event detection process is being performed in a smartwatch (where the touch-sensitive surface and possibly the microphone are integrated into the smartwatch or a portion of the smartwatch), determining whether an ambient sound level measurement is recorded is further based on the processor having detected that the smartwatch is on the user's wrist (e.g., by executing an on-wrist detection algorithm).

[0047] A method for acoustic dosimetry may be performed as follows, wherein a water event is detected by a processor and then used as contextual input for deciding whether to record a concurrent loud event. A time sequence of touch images is acquired from a touch-sensitive surface and processed so as to detect floating water on the touch-sensitive surface as a time sequence of water events. A microphone is used to perform ambient sound level measurements as a time sequence and processed so as to detect loud sounds as a time sequence of loud events. In time intervals where the time sequence of water events is unrelated to the time sequence of loud events, loud events are recorded for acoustic dosimetry. However, in time intervals where water events are associated with loud events and the water events indicate a floating amount of water exceeding a threshold or the presence of dynamic water (water is flowing through the touch-sensitive surface), loud events are not recorded (or ignored for the purpose of acoustic dosimetry). For example, for a time interval where a water event is associated with a concurrent loud event, the processor forgoes presenting a user notification indicating exposure to a loud sound even if the loud event exceeds a level threshold for at least a threshold amount of time. Such water events may indicate a floating amount of water exceeding a threshold, or they may indicate the presence of dynamic water on the touch-sensitive surface.

[0048] The acoustic dosimetry process may continuously monitor the time series of recorded loud events, and wherein in response to the recorded loud events exceeding a level threshold for at least a threshold amount of time, the processor will signal a user's device (e.g., a user's smartwatch or a user's smart phone) to present a user notification indicating exposure to the loud sound. As described above, the user notification may include displaying text or graphics by the touch sensing surface.

[0049] See again Figure 5 , the aggregate acoustic dosimetry process is fed with various inputs, some of which are direct sound level measurements, while others provide the process with context about why or how the very loud sound event occurred. One such context input described above is the output of a wind detector. Another context input, also discussed above and which may be interpreted as a decision not to record an ambient sound level measurement or to reduce its weight when mixed with other sound level estimates, is the output of a water event detector. Other context inputs may also be present, such as the following: determining that the smartwatch is not on the user's wrist; determining that the smartwatch environment is windy; and determining the user context based on accelerometer / gyroscope / magnetic sensor (compass) signals in a smartwatch or smart phone, e.g., the user is walking, running, biking, riding, or driving a car. The context input may be interpreted by a processor to, for example, determine that a microphone used to make sound measurements is experiencing high airflow that affects the accuracy of the sound measurements. In addition to context inputs derived from sensors, context inputs may also be provided manually, such as a user's response to a voice prompt or visual prompt asking if the user is in water.

[0050] An article of manufacture, such as a microelectronic integrated circuit device, for headphone assembly playback acoustic dosimetry may include a memory having stored therein instructions that configure a processor of an audio source device to perform any one or more of the operations described above.

[0051] A digital signal processing method for performing acoustic dosimetry of headphone device playback at a processor of an audio source device, the method comprising: a) determining an in-ear sound pressure level (SPL) resulting from the conversion of output audio into sound by a headphone device worn by a user, wherein the headphone device is a peripheral device of the audio source device; b) converting the in-ear SPL into a sound sample having units for acoustic noise exposure; repeating a) and b) a plurality of times to generate a time series of sound samples; and writing the time series of sound samples to a secure database to which access is authorized by the user. Determining the in-ear SPL comprises measuring the in-ear SPL from an audio signal generated by an internal microphone of the headphone device. Determining the in-ear SPL comprises: receiving previously determined acoustic output sensitivity data stored in the headphone device from the headphone device; determining the intensity of the output audio; and multiplying the acoustic output sensitivity data by the intensity of the output audio. The method also comprises writing metadata of the time series of sound samples to a database, wherein the metadata identifies a model of the headphone device and an application from which the output audio is generated. The method also includes executing an application by a processor in the audio source device, the application accessing the time series of sound samples in the database, and calculating a statistical measure of the time series of sound samples, and visualizing the statistical measure through a graphical user interface of the audio source device. The method, wherein determining the in-ear SPL includes detecting that an earphone device is connected to the audio source device, but output sensitivity data of the connected earphone device is not available, and in response, selecting a volume curve; applying the current volume step to the volume curve to obtain a maximum in-ear SPL; and deriving an output sensitivity relationship based on the maximum in-ear SPL given by the volume curve. The method, wherein determining the in-ear SPL takes into account acoustic noise cancellation, ambient sound transparency, or ambient sound enhancement of the earphone device.

[0052] An audio source device having an earphone device playback acoustic dosimetry capability, the audio source device comprising: a processor; and a memory having instructions stored therein, the instructions configuring the processor to: a) determine an in-ear sound pressure level (SPL) resulting from the conversion of output audio into sound by an earphone device worn by a user, wherein the earphone device is a peripheral device of the audio source device, b) convert the in-ear SPL into a sound sample having a unit for acoustic noise exposure, repeat a) and b) a plurality of times to generate a time sequence of sound samples, and write the time sequence of sound samples to a secure database authorized by the user to access the same. The audio source device, wherein the processor determines the in-ear SPL by measuring the in-ear SPL from an audio signal generated by an internal microphone of the earphone device. The audio source device, wherein the processor determines the in-ear SPL by: receiving previously determined acoustic output sensitivity data stored in the earphone device from the earphone device; determining the intensity of the output audio; and multiplying the acoustic output sensitivity data by the intensity of the output audio. The audio source device, wherein the memory has additional instructions that configure the processor to write metadata of the time series of sound samples to a database, wherein the metadata identifies the model of the headphone device and the application from which the output audio is generated. The audio source device, wherein the memory has additional instructions that configure the processor to execute an application that accesses the time series of sound samples in the database and calculates statistical measures of the time series of sound samples, and visualizes the statistical measures through a graphical user interface of the audio source device. The audio source device, wherein the processor determines the in-ear SPL by the following operations: detecting that the headphone device is connected to the audio source device, but the output sensitivity data of the connected headphone device is not available, and in response, selecting a volume curve; applying the current volume step to the volume curve to obtain the maximum in-ear SPL; and deriving an output sensitivity relationship based on the maximum in-ear SPL given by the volume curve. The audio source device, wherein the processor takes into account the acoustic noise cancellation, ambient sound transparency or ambient sound enhancement of the headphone device when determining the in-ear SPL.

[0053] A smartwatch, comprising: a smartwatch housing having a touch-sensitive surface integrated therein, a microphone, a processor, and a memory having instructions stored therein, the instructions configuring the processor to perform a digital signal processing method for ambient sound acoustic dosimetry, wherein the processor acquires a touch image from the touch-sensitive surface, processes the touch image to discard apparent touches caused by water, performs ambient sound level measurements using the microphone concurrently with acquiring the touch image; and determines whether ambient sound level measurements are recorded as valid sound exposure samples for acoustic dosimetry based on the processed touch image. The smartwatch, wherein the processor processes the touch image by distinguishing between i) floating water on the touch-sensitive surface and ii) a finger touch. The smartwatch, wherein the processor processes the touch image by: determining the presence of water on the touch-sensitive surface; and classifying the determined water on the touch-sensitive surface as static or dynamic. In one aspect, if the water is classified as dynamic, the ambient sound level measurements are not recorded, and if the water is classified as static, the ambient sound level measurements are recorded. In another aspect, if the water is classified as static, the ambient sound level measurement is recorded, and if the water is classified as dynamic, it is not recorded. The smartwatch, wherein the processor processes the touch image by determining the amount of water on the touch-sensitive surface. The smartwatch, wherein the processor determines the amount of water by counting the number of pixels in the touch image. If the determined amount of water is greater than a threshold, the ambient sound level measurement is not recorded. The smartwatch, wherein the memory has additional instructions that configure the processor to record the ambient sound level measurement if the determined amount of water is less than a threshold.

[0054] As described above, one aspect of the present technology is the collection and use of data that can be obtained from specific and legitimate sources to monitor sound exposure when using headphones. The present disclosure contemplates that, in some instances, the collected data may include personal information data that uniquely identifies or can be used to identify a specific person. Such personal information data may include demographic data, location-based data, online identifiers, phone numbers, email addresses, home addresses, data or records related to the user's health or hearing loss level, date of birth, or any other personal information.

[0055] The present disclosure recognizes that the use of such personal information data in the present technology can be used to benefit users. For example, personal information data can be used to monitor long-term exposure to sound according to their preferences to promote hearing health.

[0056] The present disclosure envisions that entities responsible for collecting, analyzing, disclosing, transmitting, storing or otherwise using such personal information data will comply with established privacy policies and / or privacy practices. Specifically, such entities will be expected to implement and consistently apply privacy practices that are generally recognized as meeting or exceeding the requirements of the industry or government that maintain user privacy. Such information about the use of personal data should be highlighted and easily accessible to users, and should be updated as the collection and / or use of data changes. The user's personal information should be collected only for legal use. In addition, such collection / sharing should only occur after receiving user consent or other legal basis specified in applicable law. In addition, such entities should consider taking any necessary steps to defend and safeguard access to such personal information data and ensure that others who have access to personal information data comply with their privacy policies and processes. In addition, such entities may subject themselves to third-party assessments to demonstrate that they comply with widely accepted privacy policies and practices. In addition, policies and practices should be adjusted for specific types of personal information data collected and / or accessed, and applied to applicable laws and standards, including considerations that are exclusive to jurisdictions that can be used to impose higher standards. For example, in the United States, collection or access to certain health data may be governed by federal and / or state laws, such as the Health Insurance Portability and Accountability Act (HIPAA); while health data in other countries may be subject to other regulations and policies and should be handled accordingly.

[0057] Regardless of the foregoing, the present disclosure also contemplates implementations in which users selectively block use or access to personal information data. That is, the present disclosure contemplates providing hardware elements and / or software elements to prevent or block access to such personal information data. For example, the present technology can be configured to allow a user to choose to "opt in" or "opt out" of sound exposure processing.

[0058] In addition, it is an object of the present disclosure that personal information data should be managed and processed to minimize the risk of unintentional or unauthorized access or use. Risks can be minimized by limiting data collection and deleting data once it is no longer needed. In addition, and when applicable, including in certain health-related applications, data de-identification can be used to protect the privacy of users. De-identification can be facilitated by removing identifiers, controlling the amount or specificity of stored data (e.g., collecting location data at a city level rather than at an address level), controlling how data is stored (e.g., aggregating data across users), and / or other methods such as differential privacy, where appropriate.

[0059] Therefore, while the present disclosure broadly covers the use of personal information data to implement one or more of the various disclosed embodiments, the present disclosure also contemplates that various embodiments may be implemented without access to such personal information data. That is, various embodiments of the present technology will not fail to function properly due to the lack of all or part of such personal information data.

[0060] To assist the Patent Office and any reader of any patent issuing in this application in interpreting the appended claims, applicants wish to note that they do not intend for any of the appended claims or claim elements to invoke 35 U.S.C. 112(f) unless the words “means for” or “step for” are expressly used in a particular claim.

[0061] Although certain aspects have been described and shown in the drawings, it should be understood that these aspects are merely illustrative of the invention and not limiting, and the invention is not limited to the specific constructions and arrangements shown and described, as various other modifications may occur to those skilled in the art. Therefore, the description is to be regarded as illustrative and not limiting.

Claims

1. A digital signal processing method for acoustic dosimetry, the method comprising: receiving ambient sound measurements made using a microphone integrated into the user's audio device; receiving a first context input, the first context input being an output of a wind detector; receiving a second context input, the second context input being an output of a water event detector; determining whether to discard or reduce the weight of the ambient sound measurement based on the output of the wind detector and based on the output of the water event detector; Convert the ambient sound measurement value into an in-ear sound pressure level (SPL), and append the in-ear SPL to the stored time series; as well as The stored time series is analyzed to determine whether overall sound exposure for the user exceeds a threshold level within a threshold time interval.

2. The method of claim 1, wherein the user's audio device is a headphone set.

3. The method of claim 2, wherein the ambient sound measurement, the wind detector, the conversion to in-ear SPL and the analysis are performed by a processor in the headphone device. The method of claim 1 , wherein the user's audio device is a smart watch.

5. The method of claim 4, wherein the ambient sound measurement, the wind detector, the conversion to in-ear SPL, and the analysis are performed by a processor in a smartwatch.

6. The method of claim 4, wherein the ambient sound measurement, the wind detector, the conversion to in-ear SPL, and the analysis are performed by a low power or secondary processor in the smart watch when a high power or main processor in the smart watch is in a low power state.

7. The method of claim 1, wherein the stored time series is stored in a secure data storage device, access to which is authorized by the user.

8. The method of claim 7, wherein analyzing the stored time series comprises: An application is executed by the processor, the application accesses the stored time series, calculates statistical measures of the stored time series, and visualizes the statistical measures through a graphical user interface.

9. The method according to claim 1, further comprising: During an ambient sound measurement, it is determined whether the user has active hearing protection or passive hearing protection, or both, in place against their ears, and the ambient sound measurement is attenuated based on this determination.

10. The method of claim 1, wherein the wind detector processes a microphone signal generated by a microphone integrated in the audio device of the user, the audio device being a smart watch, and the water event detector processes a touch image acquired from a touch-sensitive surface in the smart watch.

11. An audio device having acoustic dosimetry capability, the audio device comprising: microphone; processor; as well as A memory having stored therein instructions that configure the processor to: receiving ambient sound measurements made using the microphone, receiving a first context input, the first context input being an output of a wind detector, receiving a second context input, the second context input being an output of a water event detector, determining whether to discard or reduce the weight of the ambient sound measurement based on the output of the wind detector and based on the output of the water event detector, converting the ambient sound measurements into in-ear sound pressure levels (SPLs) and appending the in-ear SPLs to the stored time series, and The stored time series is analyzed to determine whether overall sound exposure to the user exceeds a threshold level within a threshold time interval.

12. The audio device of claim 11, further comprising an earphone device housing, the microphone, the processor and the memory being integrated into the earphone device housing.

13. The audio device of claim 11, further comprising a smart watch housing, the microphone, the processor, and the memory being integrated into the smart watch housing.

14. An audio device according to claim 13, wherein the processor is a low power or auxiliary processor, and the ambient sound measurement, the wind detector, the conversion to in-ear SPL, and the analysis are performed by the low power or auxiliary processor when the high power or main processor in the smart watch is in a low power state.

15. The audio device of claim 11, wherein the stored time series is stored in a secure data storage device, access to which is authorized by the user.

16. The audio device of claim 15, wherein analyzing the stored time series comprises the processor: An application is executed that accesses the stored time series and calculates statistical measures of the stored time series and visualizes the statistical measures through a graphical user interface.

17. The audio device of claim 11, wherein the memory stores further instructions that configure the processor to: During an ambient sound measurement, it is determined whether the user has active hearing protection or passive hearing protection, or both, in place against their ears, and the ambient sound measurement is attenuated based on this determination.

18. The audio device of claim 11, further comprising: A smart watch housing, wherein the microphone, the processor and the memory are integrated into the smart phone housing together with a touch-sensitive surface, and wherein the wind detector processes a microphone signal generated by the microphone and the water event detector processes a touch image acquired from the touch-sensitive surface.

19. A method for acoustic dosimetry, comprising: Acquiring a touch image from the touch-sensitive surface and processing the touch image to discard apparent touches caused by water; Simultaneously with acquiring the touch image, using a microphone to measure the ambient sound level; as well as A determination is made based on processing the touch image whether an ambient sound level measurement is recorded as a valid sound exposure sample for acoustic dosimetry.

20. The method of claim 19, wherein processing the touch image comprises distinguishing between i) floating water on the touch-sensitive surface and ii) a finger touch.

21. The method of claim 19, wherein processing the touch image comprises: determining the presence of water on the touch-sensitive surface; as well as The determined water on the touch-sensitive surface is categorized as static or dynamic.

22. The method according to claim 21, wherein: The ambient sound level measurements are not recorded if the water is classified as dynamic and are recorded if the water is classified as static.

23. The method according to claim 21, further comprising: If the water is classified as static, then the ambient sound level measurement is recorded, whereas if the water is classified as dynamic, then the ambient sound level measurement is not recorded.

24. The method of claim 19, wherein processing the touch image comprises determining an amount of water on the touch-sensitive surface.

25. The method of claim 24, wherein determining the amount of water comprises counting the number of pixels in the touch image.

26. The method according to claim 24, wherein: The ambient sound level measurement is not recorded if the determined amount of water is greater than a threshold value.

27. The method of claim 24, further comprising: If the determined amount of water is less than the threshold, the ambient sound level measurement is recorded.

28. The method of claim 19, further comprising: If the recorded ambient sound level measurement remains above the threshold for at least a threshold amount of time, a loud ambient notification is sent to the user.

29. The method of claim 19, wherein the touch-sensitive surface is part of a smart watch, and wherein determining whether the ambient sound level measurement is recorded is further based on detecting whether the smart watch is on a user's wrist.

30. A smart watch, comprising: A smart watch housing having the following components integrated therein: Touch-sensitive surface, microphone, processor, and A memory having stored therein instructions configuring the processor to perform a digital signal processing method for ambient sound acoustic dosimetry, wherein the processor: acquiring a touch image from the touch-sensitive surface and processing the touch image to discard apparent touches caused by water; Simultaneously with acquiring the touch image, using the microphone to measure the ambient sound level; as well as A determination is made based on processing the touch image whether an ambient sound level measurement is recorded as a valid sound exposure sample for acoustic dosimetry.

31. The smart watch of claim 30, wherein the processor processes the touch image by distinguishing between i) floating water on the touch-sensitive surface and ii) a finger touch.

32. The smart watch according to claim 30, wherein the processor processes the touch image by: determining the presence of water on the touch-sensitive surface; and The determined water on the touch-sensitive surface is categorized as static or dynamic.

33. The smart watch according to claim 32, wherein: The ambient sound level measurements are not recorded if the water is classified as dynamic and are recorded if the water is classified as static.

34. The smart watch according to claim 32, wherein: The memory has further instructions that configure the processor to record the ambient sound level measurement if the water is classified as static and not record the ambient sound level measurement if the water is classified as dynamic.

35. The smart watch according to claim 30, wherein: The processor processes the touch image by determining an amount of water on the touch-sensitive surface.

36. The smart watch according to claim 35, wherein: The processor determines the amount of water by counting the number of pixels in the touch image.

37. The smart watch according to claim 35, wherein: The ambient sound level measurement is not recorded if the determined amount of water is greater than a threshold value.

38. The smart watch according to claim 35, wherein: The memory has further instructions that configure the processor to record the ambient sound level measurement if the determined amount of water is less than a threshold value.

Images