Infection prevention warning device and method

Abstract

To provide an infection prevention warning device and a method thereof that promote the avoidance of crowded places, close-contact settings, and closed spaces, and the wearing of masks, which are advocated as virus infection prevention measures.SOLUTION: An infection prevention warning device, which prevents virus infection of a user in a crowded place, includes: a microphone 1 for collecting environmental sounds of a densely populated place and a user's conversation; voice processing means 4 for extracting a user's conversation signal from an output signal of the microphone 1; means for visualizing an output level of the voice processing means 4 as a risk level and displaying a warning; and means for determining an incentive for the user according to the risk level.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an infection prevention warning device and method that promotes avoiding crowded, close-contact, and enclosed spaces and wearing masks, which are proposed as virus infection prevention measures. [Background technology]

[0002] As infectious diseases spread, thorough preventative measures such as wearing masks, regular ventilation, and observing social distancing are considered important. In particular, national and local government guidelines call for avoiding the so-called "three Cs": (1) enclosed spaces with poor ventilation, (2) crowded places where many people gather, and (3) close-contact settings where people talk or speak in close proximity. Being in one of these places and encountering an infected person transmitting a virus or other illness increases the risk of infection, and various methods and devices have been proposed to measure the degree of enclosed, crowded, and close contact, and to implement ventilation and density avoidance measures.

[0003] Methods of measuring human density include infrared measurement, which uses infrared light to measure heat generated by the human body, and carbon dioxide measurement, which measures the amount of carbon dioxide (CO2) produced by people speaking (see, for example, Patent Document 1). Patent Document 2 also discloses a system and method for measuring human density using a video camera.

[0004] While it is important to measure crowd density and implement infection prevention measures, these measurement results do not directly translate into infection control measures, and more effective countermeasures are needed. Typical places with a high risk of infection include clubs, bars, restaurants, live music venues, karaoke, and event venues. It has been pointed out that removing masks and talking or speaking while eating or drinking is a factor that increases the risk of infection. Furthermore, COVID-19, also known as the novel coronavirus disease (COVID-19), is believed to spread through droplets and contact. In order to avoid droplet and contact transmission, which are factors that contribute to the spread of infection, not only the level of crowding but also the risk of infection, it is important to wear masks in crowded places and to limit conversation and speech. Therefore, means and methods for achieving this are needed. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2012-073989 [Patent Document 2] Japanese Patent Application Laid-Open No. 2014-115019 Summary of the Invention [Problem to be solved by the invention]

[0006] The present invention has been made in consideration of the above-mentioned circumstances, and aims to provide an infectious disease prevention warning device and method as described below. (1) A device and method for visualizing, displaying, and issuing warnings about the level of infection risk in crowded places based on conversation and vocalization, the duration of mask wearing, and the degree of crowding, with the aim of preventing droplet and contact infection of infectious diseases. (2) A warning device and method for linking the above-mentioned infection risk level warning to voluntary infection prevention and encouraging infection prevention. [Means for solving the problem]

[0007] The warning device according to the present invention is a warning device for preventing virus infection of users in crowded places, and includes: a microphone for collecting environmental sounds in the crowded place and conversations of the users; and audio processing means for extracting conversation signals of the users from output signals of the microphone. A mask wearing time measurement means for calculating the time the user wears a mask using a video camera installed in the crowded place, and a person density measurement means for detecting the person density using the video camera or a carbon dioxide sensor installed in the crowded place; the audio processing means, the mask wearing time measuring means and the crowd density measuring means The output level is visualized as a risk level, To the user A means of displaying a warning 、 and means for determining an incentive for the user in accordance with the risk level.

[0012] Furthermore, the warning method according to the present invention is a warning method for preventing virus infection of users in crowded places, and is characterized by comprising the steps of: extracting the user's conversation signal from a microphone; measuring the time the user wears a mask using a video camera; measuring the human density using a video camera or a carbon dioxide sensor; visualizing the output level of the conversation signal, the time the mask is worn, and / or the human density as a risk level and displaying a warning; and determining an incentive for the user according to the risk level.

[0013] Furthermore, the warning method according to the present invention can be configured to include an evaluation step in which the risk level is used as an evaluation criterion for the event and the restaurant. [Effects of the Invention]

[0014] The warning device and method of the present invention aim to prevent droplet infection at restaurants, events, etc. by measuring and visualizing speech levels, mask wearing time, and crowd density detection, thereby raising awareness of prevention among users. Furthermore, by configuring the device to provide incentives to users based on the results of these measurements, users can be encouraged to voluntarily suppress speech and wear masks, rather than being forced to do so, leading to infection prevention measures. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a block diagram illustrating the entire droplet infection warning device according to the present invention. [Figure 2] FIG. 1 is a schematic block diagram illustrating speech detection of the present invention. [Figure 3] FIG. 2 is an explanatory diagram of speech detection and equivalent speech level of the present invention. [Figure 4] FIG. 2 is a diagram showing an example of a display means of the present invention. [Figure 5] 1 is a flowchart illustrating an infection prevention method according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0016] The warning device of this embodiment will be described in detail below with reference to the drawings. Note that all drawings in the following examples are drawn as outlines or schematic diagrams for the purpose of explaining the present invention, and the actual dimensions and shapes are not particularly limited. Furthermore, the block circuit configuration, component configuration, materials, shapes, relative arrangements, and uses of the components are intended to be used as an infectious disease prevention warning device in restaurants, event venues, schools, assembly halls, etc., but unless otherwise specified, the technical scope of the invention is not intended to be limited to these.

[0017] Many infectious diseases are transmitted through droplets or contact caused by an infected person's speech, coughing, or sneezing, and it is known that the risk increases when people gather in crowded, enclosed rooms or venues, such as restaurants or event venues, and are in close contact with one another. When eating and drinking or at exciting events, it is difficult to avoid removing masks to talk, shouting, or cheering, despite warnings and appeals from restaurant or event organizers. This invention provides a warning device and method that uses various sensor means to detect and visualize mask wearing, talking, cheering, and other behaviors in close, enclosed, or crowded spaces, providing feedback to users and linking them to various incentives, thereby contributing to voluntary droplet infection prevention.

[0018] Figure 1 is a block diagram of the entire infection warning device according to the present invention. This example warning device is equipped with a microphone (mic) 1 that detects conversation levels, a video camera (thermo camera) 2 that detects body temperature, crowding, and mask wearing, and a carbon dioxide sensor 3 that detects crowding by carbon dioxide as input interfaces.

[0019] Conversations and vocalizations by patrons at restaurants and events are of particular importance in terms of virus infection, and microphone 1 acquires the conversation levels of the patrons. This microphone 1 is placed at a predetermined distance from the patrons, and is installed in a location that is roughly equidistant from the patrons, such as on the ceiling or in the center of a table within a patron group or dining room. The audio collected by microphone 1 is extracted by audio processing means 4, specifically focusing on the conversations and vocalizations of the patrons, and averaged as an equivalent audio level, and the conversation sound level data is output to dense level indexing means 5.

[0020] An example of a block diagram embodying the voice processing means 4 is shown in FIG. 2. In FIG. 2, input collected by microphone 1 is amplified by amplifier means 21 and supplied to voice filter means 22. Voice filter means 22 is composed of a bandpass filter that extracts a band of approximately 100 to 300 Hz, which corresponds to the male voice band of 100 to 200 Hz and the female voice band of 160 to 300 Hz in normal conversation sounds, as shown in FIG. 3. The output of voice filter means 22 is provided to noise removal means 23.

[0021] As shown in Figure 3, the collected voice-band audio signal contains ambient noise and impulsive noise such as sudden sounds and plosives. Therefore, these noises must be removed to detect the conversation level. The noise removal means 23 removes the impulsive noise using a high-cut filter. Furthermore, the ambient noise collected by the microphone 1 is appropriately amplified by the external noise processing means 24 and supplied to the noise removal means 23 in an inverted phase. The supplied external noise is combined with the input audio signal by the noise removal means 23 for noise cancellation. Restaurants and event venues, in particular, often contain a large amount of steady noise at a volume level equivalent to background music (BGM). In such cases, it is effective to perform BGM noise cancellation by combining the BGM sound source in an inverted phase with the input audio signal. The output signal from which the noise has been removed by the noise removal means 23 is supplied to the averaging means 26 as an equivalent audio level.

[0022] The equivalent sound level averaging means 26 averages the input conversation signal to calculate the equivalent sound level. Equivalent sound level is defined as the sound level of a continuous steady sound that gives a mean square sound pressure equal to the mean square sound pressure of the fluctuating noise within the measurement time, but there is no problem as long as the energy of conversational sound over a certain period of time is averaged and the equivalent sound level can be determined. The equivalent sound level shown in Figure 3 is an explanatory diagram of the average value of conversational sound extracted during a specific measurement time period (for example, the time period from entering the store to leaving the store).

[0023] The signal in which the conversational energy has been averaged by the equivalent sound level conversion means 26 has its level determined by the threshold determination means 27. Here, because noise values ​​measured by a general noise meter are targeted at outdoor environmental sounds that fluctuate irregularly and significantly, it is preferable to cancel out the surrounding environmental sounds and set a unique threshold specifically for the conversational sound level while limiting the sound to the conversational sound band. For example, a conversational sound level of 30 dB or less can be set as "recommended" (Level A), 30 to 50 dB as "caution" (Level B), and 50 dB or more as "dangerous" (Level C). These level settings are indexed to indicate how much the sound exceeds a reference value set for the measurement location, and are output as conversational sound level data.

[0024] The video camera (thermal camera) 2 in Figure 1 is composed of a visible light (color) video camera equipped with an infrared band body temperature detection function. If a user's body temperature detected by the user's body temperature detection means 6 is, for example, 37°C or higher, the user is excluded as having a fever and is barred or restricted from using or entering the restaurant or venue. The video camera 2 also uses a density detection means 8 to detect the density and closeness of users in a user group or dining room. Many methods for detecting the density and closeness using the density detection means 8 have already been disclosed, as in the prior art documents mentioned above, and any of these may be used. The detected human density and closeness are determined by a human density measurement means 9 according to a set threshold and provided to the density level indexing means 5 as human density distribution data.

[0025] Furthermore, video camera 2 first detects whether users are wearing masks and issues warnings or restricts entry to users who are not wearing masks. At the same time, video camera 2 is used to detect the time spent wearing a mask (or not wearing a mask). Wearing a mask is an important preventative measure against droplet infection, and wearing a mask is recommended whenever possible except when eating or drinking. Since wearing a mask at all times is particularly desirable at event venues where eating and drinking are not involved, detecting users without a mask can encourage them to wear one. Images of users captured by the color camera unit of video camera 2 are input to mask detection means 10. For mask wearers, mask wearing time measurement means 11 measures the time spent wearing a mask or not wearing a mask. This detection of mask wearing or not and measurement of the time spent wearing a mask (or not wearing a mask) can be performed using known image (face) recognition technology. The measured mask wearing time data is sent to density level indexing means 5.

[0026] Furthermore, the warning device of the present invention can be additionally equipped with a carbon dioxide density measuring means 12 that detects the degree of crowding using a carbon dioxide sensor 3. This carbon dioxide sensor 3 is typically an optical sensor using NDIR (Non-Dispersive Infra-Red), but any type, such as a semiconductor sensor, can be used as long as it can measure CO2 levels of approximately 350 to 3000 ppm. The carbon dioxide concentration is based on an outdoor air concentration of 415 to 450 ppm. A level of 800 ppm or less is considered sufficient ventilation, and government guidelines recommend ventilation to ensure levels do not exceed 1000 ppm. It is known that levels of 1500 ppm or more can cause drowsiness and fatigue, and levels of 2000 ppm or more can cause symptoms such as headaches and difficulty breathing. With reference to these, the carbon dioxide measurement means 12 provides the density level indexing means 5 with numerically quantified carbon dioxide density data that is classified into levels, such as "recommended" for 1000 ppm or less, "normal" for above 1500 ppm, "caution" for above 2000 ppm, and "danger" for above 2000 ppm.

[0027] The level indexing means 5 determines the infection risk level using the provided (1) conversation sound level data, (2) person density distribution data, (3) mask wearing time data, and (4) carbon dioxide density data. Each piece of data provided to the density level indexing means 5 is weighted and calculated to be indexed, indicating "safe," "caution," "danger," or a three- or five-level numerical classification, relative to a reference level. This weighting is set independently depending on the type, scale, and nature of the restaurant or event. From the perspective of preventing droplet infection, it is important to suppress conversation and vocalization and to comply with mask wearing requirements, so conversation sound level and mask wearing time are typically weighted highly. Additionally, at event venues and other locations, it is expected that person density distribution data and carbon dioxide density data will also be weighted highly.

[0028] The density level indexing means 5 weights data such as conversation sound level, human density distribution, mask wearing time, and carbon dioxide density, and the indexed signal is provided to the dense data processing means 13. The dense data processing means 13 processes each piece of data and the weighted overall data using a display method or calculation method according to the purpose of use. When reflecting this in the user's usage fee or incentive, the conversation level within the usage time, mask wearing time, etc. are displayed in stages, and an incentive index signal for the user corresponding to this is output to the output processing means 14.

[0029] The output processing means 14 performs arithmetic processing according to the intended use of the dense data. When the dense data is to be reflected in the user's fee structure or incentives, the system is configured to output discount rates and incentive details to a fee calculation device according to the indexed dense data, or to display the discount amount at the time of payment. At the same time, various dense data are indexed and supplied to the display means 16 as data signals such as density level and conversation level, and are displayed as graphs or digits. The people density distribution data captured by the video camera 2 is accompanied by video, and is supplied to the mapping visualization means 16 via the density level indexing means 5 and visualized. This visualization can display the camera image as is, or map the degree of people density using fonts, colors, etc., and provide the result to the display means 17.

[0030] Various display methods are possible depending on the purpose of use. Figures 4(A) and (B) show examples of display means 16 for restaurants and events. Figure 4(A) shows an example of a user-side monitor 40, primarily used to visualize and inform users of infection risk. Various acquired data related to infection risk are weighted and displayed on the user monitor 40 as a risk level 41. This display can be easily recognized by users without a monitor display. It can also be displayed as a level meter display showing the constantly changing status with an LED (Light Emitting Diode) lamp, or simply as a three-color LED lamp displaying green (safe), yellow (caution), and red (danger), or as a three- or five-level numerical display. In addition to the risk level for droplet infection, the carbon dioxide (CO2) detection level can be displayed in ppm using a CO2 display means 42, based on various acquired data, or by changing the color of the LED depending on the degree of risk. Similarly, the conversation sound level can be displayed in dB by the display means 43, or the degree of risk can be displayed by changing the color of an LED to call attention to the risk.

[0031] Furthermore, in the present invention, this risk data is used to display discount rates for usage fees and incentive displays 44 to raise users' awareness of infection prevention measures. By visualizing the risk level in this way and displaying a system that benefits users, users can be encouraged to voluntarily wear masks as much as possible and reduce the amount of conversation they engage in without being warned or reprimanded by the administrator. Furthermore, if the detected time period during which a mask is not worn exceeds a predetermined time, a message urging users to wear a mask may be displayed in the warning message field 45.

[0032] In addition to the above display, the administrator's monitor 50 can also display the population density distribution within the venue on a density distribution monitor 51 using people density distribution data acquired from the video camera 2 as a plot or a color-coded density distribution map. The camera can also be configured to display the number of hours since each user entered the venue for each individual user. This people density distribution display can be used by administrators to control people flow and avoid crowding by referring to the entry or visitor count display 53. This display also includes a ventilation warning signal 52. When the amount of carbon dioxide acquired from the CO2 sensor exceeds a predetermined value (e.g., 1000 ppm), the ventilation warning signal 52 issues a visual or audible warning, prompting the administrator to ventilate the venue.

[0033] As mentioned above, avoiding crowded, close contact, and closed spaces is important for preventing infection. This article describes a method for visualizing the risk level of infection for patrons of restaurants and event venues, encouraging them to voluntarily wear masks and refrain from conversation. Figure 5 is a flowchart of the infection risk prevention method according to the present invention. Means for detecting crowded, close contact, and closed spaces include acquiring conversation volume data using a microphone, detecting body temperature using a thermal camera, acquiring people density distribution data and mask wearing time data, and detecting carbon dioxide density using a carbon dioxide sensor. To do this, first, an input detection means is selected depending on which detection or data is required (Step 1; hereinafter referred to as S-1, and the same applies to the following steps).

[0034] When selecting a thermal camera, the thermal camera detects whether or not a mask is being worn (S-2). If a mask is not being worn (No), the user is warned to wear a mask or is excluded from use. If a mask is being worn (Yes), the temperature is then detected (S-3). If the temperature is above a predetermined temperature (e.g., 37°C) (No), use is restricted or the user is excluded from use. If the temperature is below the predetermined temperature (Yes), the user is permitted to enter the store (or enter). When users enter the store (or enter), the number of people entering is counted and the human density distribution is measured based on how many people are in which locations, and the process moves to the next step.

[0035] In microphone selection, only conversational voices are extracted from the collected ambient sound using a bandpass filter that passes frequencies between approximately 100 and 300 Hz (S-5). Next, ambient noise, including background music noise, and impulsive noise are removed from the extracted conversational voice band using a noise canceller (S-6). The noise-removed conversational voice signal is extracted as an equivalent voice level using an averaging circuit or similar (S-7). The voice signal averaged to the equivalent voice level is ranked using a set threshold (S-8). The data ranked using the threshold is then moved on to the next step.

[0036] When a carbon dioxide sensor is selected as the input method, the CO2 data acquired by the carbon dioxide sensor is ranked using threshold judgment and the process moves to the next step (S-9). Unlike the human density distribution obtained using a thermal camera, this CO2 data measures the overall density of a venue or store and is used as a ventilation warning to avoid enclosed spaces.

[0037] The conversation sound level data, human density distribution data, mask wearing time data, and human density data based on carbon dioxide levels acquired through each of the above input methods are weighted according to the purpose of use, venue, and store size, and the degree of crowding is indexed against a reference value (S-10). Each indexed data item is then combined and calculated to determine the crowding risk (S-11). Data determined to be a crowding risk is displayed and warned to users and administrators according to the level using color (green, yellow, red), audio (warning, etc.), numbers (ppm, dB, or 3-level or 5-level display), level meter, etc. (S-12). The crowd density distribution, including color and crowded areas, is also displayed on a monitor (S-12).

[0038] The risk level is displayed on a display monitor, and incentives and discount rates for users are reflected according to the risk level, such as the user's conversation level, the time spent wearing a mask, and the degree of crowding (S-13). In this way, users can recognize the risk of infection as a droplet infection risk level displayed in numbers, colors, and data, and by linking this to some kind of incentive or discount on the usage fee, they can voluntarily avoid crowding, comply with mask wearing, and limit their speaking without being forced to do so.

[0039] In the examples, the explanation was given assuming a droplet infection prevention warning device and infection prevention method for restaurants and event venues, but the present invention can also be applied to various viral infections in crowded places and facilities within the scope of the present invention. [Industrial Applicability]

[0040] The prevention and warning device and infection prevention method of the present invention make it possible to weight, visualize, display, and warn about various data necessary for preventing droplet infection and other diseases, and also to reflect this in users' usage fees and incentives to warn about and prevent infection.This will be effective when introduced in restaurants, event venues, etc., and will therefore have broad industrial applicability. [Explanation of symbols]

[0041] 1 microphone 2 Video camera (thermal camera) 3 Carbon dioxide (CO2) sensor 4. Audio processing means 5. Density level indexing method 6. Body temperature detection means 8. Means for detecting human density distribution 9. Means of measuring human density 10 Mask detection means 11. Method for measuring mask wearing time 12 Carbon dioxide density measuring means 13 Dense data processing means 14 Output Processing Means 15. Incentives and other measures 16 Mapping visualization methods 17 Display means 22 Speech filtering means 23 Noise removal methods 26 Averaging means (equivalent sound level averaging means) 27 Threshold determination means

Claims

1. A warning device for preventing virus infection among users in crowded places, a microphone for collecting environmental sounds in the crowded place and conversations between the users; a voice processing means for extracting a speech signal of the user from the output signal of the microphone; A mask wearing time measurement means for calculating the time the user wears a mask using a video camera installed in the crowded place; a crowd density measuring means for detecting crowd density using the video camera or a carbon dioxide sensor installed in the crowded place; a means for visualizing the output levels of the voice processing means, the mask wearing time measuring means, and the crowd density measuring means as a risk level and displaying a warning to the user; and means for determining an incentive for the user in accordance with the risk level.

2. A warning method for preventing virus infection of users in crowded places, comprising: extracting a speech signal from the user through a microphone; measuring the time the user wears the mask using a video camera; measuring the density of people using a video camera or a carbon dioxide sensor; a step of visualizing an output level according to any one or a combination of the conversation signal, the mask wearing time, and the human density as a risk level and displaying a warning; and determining an incentive for the user according to the risk level.

3. The warning method according to claim 2, further comprising an evaluation step of using the risk level as an evaluation criterion for events and restaurants.

Images