Communication location detection system and communication location detection method

Abstract

It accurately detects the location where communication is occurring. [Solution] The processing device 10 includes a proximity determination processing unit 11 that determines that there is proximity between people when one or more other people are within a predetermined range centered on the position of person H determined by the positioning device 5, and calculates the average position of the positions of multiple people within the predetermined range; a communication occurrence determination unit 12 that, when the proximity determination processing unit 11 determines that there is proximity between people, acquires the sound pressure level detected by the sound pressure level sensor 6 closest to the average position, and further determines that communication has occurred between people within the predetermined range if the value based on the sound pressure level is equal to or greater than a predetermined value; and an output unit 13 that, when the communication occurrence determination unit 12 determines that communication has occurred, outputs the predetermined range, the positions of multiple people, or the average position as the communication occurrence location.

Description

Technical Field

[0001] The present invention relates to a communication occurrence location detection system and a communication occurrence location detection method that can accurately detect a communication occurrence location.

Background Art

[0002] Conversations occurring within a building space are important as indicators for measuring the amount of people's communication and intellectual productivity. Detecting sound is an effective method as a means for determining whether there are people in a certain space or whether communication is actively taking place. If such a sound detection method is established, for example, it becomes possible to measure the communication activity level within an office or measure the utilization rate of the communication pace in public facilities and the like.

[0003] In Patent Document 1, in a conversation recognition recording system that recognizes and records whether a conversation has occurred in each of spaces divided into a plurality of sections, a voice determination unit that determines whether the sound collected in each section contains voice, a person determination unit that determines whether a plurality of people exist around each section, and in the voice determination unit, when it is determined that the sound collected in a certain section contains voice and in the person determination unit, when it is determined that a plurality of people exist around the section, a recording unit that recognizes and records that a conversation has occurred in the section is disclosed.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] By the way, Patent Document 1 has the problem that, because it uses a human presence sensor installed in a space to record conversations, the detection granularity of the conversation location (communication location) is coarse.

[0006] Furthermore, Patent Document 1 determines the presence or absence of conversation based on the position of the sound pressure sensor, which means it is dependent on the density of sound pressure sensors installed, and therefore has the problem of not being able to detect the actual location where conversation (communication) occurs in detail.

[0007] The present invention has been made in view of the above, and aims to provide a communication location detection system and a communication location detection method that can detect the location of communication with high accuracy. [Means for solving the problem]

[0008] To solve the above-mentioned problems and achieve the objective, the communication location detection system according to the present invention is a communication location detection system capable of detecting the location of communication in a target space where communication between people is possible, comprising: a positioning system for positioning a person in the target space; a sound pressure level sensor for detecting the sound pressure level in the target space; and a processing device for detecting the location of communication based on the position of a person positioned by the positioning system and the sound pressure level detected by the sound pressure level sensor, wherein the processing device comprises: a proximity determination processing unit that determines that there is proximity between people when one or more other people are within a predetermined range centered on the position of the person, and calculates the average position of the positions of the multiple people within the predetermined range; a communication occurrence determination unit that, when the proximity determination processing unit determines that there is proximity between people, acquires the sound pressure level detected by the sound pressure level sensor closest to the average position, and further determines that communication has occurred between people within the predetermined range if the value based on the sound pressure level is greater than or equal to a predetermined value; and an output unit that, when the communication occurrence determination unit determines that communication has occurred, outputs the predetermined range, the positions of the multiple people, or the average position as the communication occurrence location.

[0009] Furthermore, the communication location detection system according to the present invention is characterized in that, in the above invention, when the predetermined range in which multiple people are present partially overlaps and is continuous, the proximity determination processing unit groups the entire range of the partially overlapping and continuous predetermined range as one predetermined range and determines the average position of the multiple people within the grouped predetermined range.

[0010] Furthermore, the communication location detection system according to the present invention is characterized in that the output unit displays an image at a corresponding position on a display screen that shows the target space in a plan view, indicating the predetermined range that is the location of the communication, the positions of the multiple people, or the average position.

[0011] Furthermore, the communication location detection system according to the present invention is characterized in that the output unit outputs the communication location with person identification information attached.

[0012] Furthermore, the communication occurrence location detection system according to the present invention is characterized in that, in the above invention, the value based on the sound pressure level in the communication occurrence determination unit is the difference obtained by calculating the A-weighted equivalent sound pressure level in a set time interval band and the octave band equivalent sound pressure level with a center frequency of 500 Hz, and subtracting the A-weighted equivalent sound pressure level from the octave band equivalent sound pressure level with a center frequency of 500 Hz in the set time interval band.

[0013] Furthermore, the communication location detection system according to the present invention is characterized in that the processing device repeatedly performs the operation at predetermined time intervals and records the detection information for each predetermined time interval along with the time.

[0014] Furthermore, the present invention provides a method for detecting a location where communication occurs, comprising: a positioning system that positions a person within a target space where person-to-person communication is possible; a sound pressure level sensor that detects the sound pressure level within the target space; and a processing device that detects the location where communication occurs based on the position of the person positioned by the positioning system and the sound pressure level detected by the sound pressure level sensor. The method for detecting a location where communication occurs includes: a proximity determination processing step in which the processing device determines that there is proximity between people if one or more other people are within a predetermined range centered on the position of the person it has positioned, and determines the average position of the positions of the multiple people within the predetermined range; a communication occurrence determination step in which, if it is determined that there is proximity between people by the proximity determination processing step, it obtains the sound pressure level detected by the sound pressure level sensor closest to the average position, and if the value based on the sound pressure level is greater than or equal to a predetermined value, it determines that communication has occurred between people within the predetermined range; and an output step in which, if it is determined that communication has occurred by the communication occurrence determination step, it outputs the predetermined range, the positions of the multiple people, or the average position as the location where communication occurred.

[0015] Furthermore, the communication location detection method according to the present invention is characterized in that, in the proximity determination processing step, if the predetermined ranges in which multiple people exist partially overlap and are continuous, the entire range of the partially overlapping and continuous predetermined ranges is grouped as one predetermined range, and the average position of the multiple people within the grouped predetermined range is determined. [Effects of the Invention]

[0016] According to the present invention, the location where communication is occurring can be detected with high accuracy using a positioning system and a sound pressure level sensor. [Brief explanation of the drawing]

[0017] [Figure 1]FIG. 1 is a schematic diagram showing the general configuration of a communication occurrence location detection system according to the present embodiment. [Figure 2] FIG. 2 is a diagram showing an example of specifying a communication occurrence location by a processing device. [Figure 3] FIG. 3 is a diagram showing an example of output of a communication occurrence location. [Figure 4] FIG. 4 is a diagram showing an example of calculating an average position when a predetermined range where a plurality of people exist partially overlaps and is continuous. [Figure 5] FIG. 5 is a flowchart showing a communication occurrence location detection processing procedure according to the present embodiment. [Figure 6] FIG. 6 is a diagram showing frequency characteristics of sound pressure levels when conversation is being held and when no conversation is being held in a target space. [Figure 7] FIG. 7 is a diagram showing a frequency distribution of differences when conversation is being held and when no conversation is being held in a target space, respectively. [Figure 8] FIG. 8 is a diagram showing the relationship between a set value of a threshold and determination accuracy.

Embodiments for Carrying Out the Invention

[0018] Hereinafter, embodiments for carrying out the present invention will be described with reference to the accompanying drawings.

[0019] <Overall Configuration> FIG. 1 is a schematic diagram showing the general configuration of a communication occurrence location detection system 1 according to the present embodiment. As shown in FIG. 1, the communication occurrence location detection system 1 detects the location where communication occurs in a room 2 which is a target space. In the room 2, positioning tags 3 carried by a person H, a plurality of locators 4 provided on the ceiling, and a plurality of sound pressure level sensors 6 are arranged.

[0020] Locator 4 receives a signal from positioning tag 3 and transmits it to positioning device 5 outside room 2. Positioning device 5 then sends the position of positioning tag 3 as the position of person H to processing device 10. These positioning tag 3, locator 4, and positioning device 5 function as a positioning system. Positioning tag 3 is, for example, a BLE (registered trademark) tag and is carried by person H. Positioning tag 3 may also be an RFID tag, a Wi-Fi (registered trademark) tag, or even one using ultrasound or geomagnetism. Sound pressure level sensor 6 detects the A-weighted sound pressure level of room 2 and sends it to processing device 10.

[0021] The processing unit 10 includes a proximity detection processing unit 11, a communication occurrence detection unit 12, and an output unit 13. It detects the location of communication based on the position of person H determined by the positioning system and the sound pressure level detected by the sound pressure level sensor 6. In practice, the processing unit 10 stores programs corresponding to the proximity detection processing unit 11, the communication occurrence detection unit 12, and the output unit 13 in a storage device such as a non-volatile memory or a magnetic disk drive. These programs are then loaded into memory and executed by the CPU (Central Processing Unit), thereby executing the corresponding processes.

[0022] The proximity determination processing unit 11 determines that there is proximity between people if one or more other people are within a predetermined range centered on the position of person H, and calculates the average position of the multiple people H within the predetermined range.

[0023] When the proximity determination processing unit 11 determines that there is proximity of person H, the communication occurrence determination unit 12 acquires the sound pressure level detected by the sound pressure level sensor 6 closest to the average position, and further determines that communication has occurred between people H within a predetermined range if the value based on the sound pressure level is equal to or greater than a predetermined value.

[0024] If the communication occurrence determination unit 12 determines that communication has occurred, the output unit 13 outputs a predetermined range, the positions of multiple people H, or the average position as the communication occurrence location. For example, the output unit 13 displays an image on an external output device 20 at the corresponding position on a display screen that shows a plan view of the target space, room 2, indicating the predetermined range, the positions of multiple people, or the average position of the communication occurrence location.

[0025] <An example of identifying the location where communication occurred> Figure 2 shows an example of how the processing unit 10 identifies the location of a communication. The positioning system measures the position of person H in room 2 at set time intervals ΔT. At this time, the positioning system links the measured position of person H with the identification information (ID) in the positioning tag 3. In addition, the sound pressure level sensors 6 (6-1, 6-2) detect the sound pressure level at time intervals ΔT.

[0026] As shown in Figure 2, the proximity determination processing unit 11 selects one person H1 whose position has been measured and determines whether or not another person H exists within a predetermined range E centered on person H1's measured position L1. In Figure 2, another person H2 is located at measured position L2 within the predetermined range E. Therefore, the proximity determination processing unit 11 determines that since multiple people H1 and H2 exist within the predetermined range E, people H1 and H2 are in close proximity within the predetermined range E. The proximity determination processing unit 11 then calculates the average position Lc (average coordinate position) of person H1's measured position L1 and person H2's measured position L2. If there are three or more people H within the predetermined range E, the proximity determination processing unit 11 calculates the average position from the measured positions of each person H.

[0027] Subsequently, the communication occurrence determination unit 12 selects the sound pressure level sensor 6 closest to the average position Lc. In Figure 2, the distance d1 between the average position Lc and sound pressure level sensor 6-1 is shorter than the distance d2 between the average position Lc and sound pressure level sensor 6-2, so sound pressure level sensor 6-1 is selected. The communication occurrence determination unit 12 then acquires the sound pressure level from sound pressure level sensor 6-1, and if the value associated with this sound pressure level is greater than or equal to a predetermined value, it determines that communication has occurred between people H1 and H2 within the predetermined range E.

[0028] <Output Processing> Figure 3 shows an example of the output for communication occurrence locations. As shown in Figure 3, the output unit 13 plots the average location Lc, which is the communication occurrence location, as a mark M at the corresponding position on the display screen D, which is a plan view of room 2. The shape, color, and pattern of this mark M are arbitrary. Mark M may also be displayed at time intervals ΔT, or it may be displayed in the history. Furthermore, identification information of person H may be displayed near mark M, or the identification information of person H may be displayed when mark M is selected. In any case, the average location Lc and identification information are stored in a storage device (not shown) at time intervals ΔT.

[0029] Mark M may be plotted on the positioning locations of people within a predetermined range E, for example, positioning location L1 of person H1 and positioning location L2 of person H2 as shown in Figure 2. Furthermore, instead of Mark M, the predetermined range E where communication occurred may be displayed on the display screen D. When displaying the predetermined range E, it is preferable to display the range E in overlapping areas. By displaying the overlapping areas of the predetermined range E in varying shades, it is possible to recognize areas where communication occurs frequently.

[0030] Here, the average position Lc mentioned above was for one predetermined range E, but if the predetermined range E containing multiple people H overlaps and is continuous in part, the proximity determination processing unit 11 may group the entire overlapping and continuous predetermined range E as one predetermined range E and use the average position of the positioning positions of the multiple people H within the grouped predetermined range E as the average position.

[0031] Figure 4 shows an example of calculating the average position when a predetermined range E containing multiple people H partially overlaps and is continuous. As shown in Figure 4, first, it is determined whether other people H exist within a predetermined range E11 centered on the positioning position L11 of person H11. Within the predetermined range E11, other people H12, H13, and H14 exist at positioning positions L11, L12, and L13, respectively. Next, it is determined whether other people H exist within the predetermined ranges E12, E13, and E14 centered on positioning positions L11, L12, and L13. Then, another person H15 exists at positioning position L15 within the predetermined ranges E13 and E14. Furthermore, when it is determined whether other people H exist within the predetermined range E15 centered on positioning position L15, no other people H exist. As a result, it can be determined that people H11 to H15 are in close proximity at positioning positions L11 to L15, respectively. As a result, the average position Lc of positioning positions L11 to L15 can be obtained.

[0032] Furthermore, if the value based on the sound pressure level detected by the sound pressure level sensor 6 closest to this average position Lc is greater than or equal to a predetermined value, it is determined that communication has occurred between people within the predetermined range E11 to E15. In this case, the output may be the average position Lc of the positioning positions L11 to L15, the positioning positions of each person H11 to H15, or the predetermined range E10, which is the entire range of the predetermined range E11 to E15.

[0033] <Communication location detection process> Figure 5 is a flowchart showing the communication location detection processing procedure according to this embodiment. As shown in Figure 5, first, the position of all people in room 2, which is the target space, is measured within a time interval ΔT, and the position of each person is obtained (step S101), and the sound pressure level measured by the sound pressure level sensor 6 within the time interval ΔT is obtained (step S102).

[0034] Subsequently, the processing unit 10 selects one person whose position has been determined (step S103). Then, it determines whether or not other people exist within a predetermined range centered on the position of the selected person (step S104). If no other people exist within the predetermined range (step S104: No), the process proceeds to step S109. On the other hand, if other people exist within the predetermined range (step S104: Yes), the average position of each person's position within the predetermined range is calculated (step S105).

[0035] Furthermore, the system selects the sound pressure level sensor closest to the average position (step S106) and determines whether the value based on the sound pressure level of the selected sound pressure level sensor is greater than or equal to a predetermined value (step S107). If the value based on the sound pressure level is not greater than or equal to the predetermined value (step S107: No), the system proceeds to step S109. On the other hand, if the value based on the sound pressure level is greater than or equal to the predetermined value (step S107: Yes), the system determines that communication has occurred at the average position (step S108) and proceeds to step S109.

[0036] In step S109, it is determined whether or not there are any unprocessed individuals. If there are unprocessed individuals (step S109: Yes), the next unprocessed individual is selected (step S110), and the process proceeds to step S104 to continue the communication location detection process for the next individual. On the other hand, if there are no unprocessed individuals (step S109: No), the average location of communication occurrences is output to the output device 20 (step S111), and this process ends. The above process is repeated at time intervals ΔT.

[0037] <Specific examples of communication detection based on sound pressure level> The above method of determining whether communication has occurred is based on whether a value derived from the sound pressure level is above a predetermined value. Specifically, based on the sound collected by the sound pressure level sensor 6, the A-weighted equivalent sound pressure level (LAeq) in the set time interval ΔT band and the octave band equivalent sound pressure level (L500eq) at a center frequency of 500Hz are calculated. If the difference obtained by subtracting the A-weighted equivalent sound pressure level (LAeq) from the octave band equivalent sound pressure level (L500eq) at a center frequency of 500Hz is above a preset threshold (determined value), it is determined that communication has occurred. In this case, it is possible to detect the presence or absence of sound with high accuracy without using the sound pressure level value or fluctuation range.

[0038] Figure 6 shows the frequency characteristics of sound pressure levels in Room 2, the target space, under two conditions: when conversation (communication) was taking place and when no conversation was taking place. The frequency characteristics in Figure 6 were measured by continuously measuring the octave band equivalent sound pressure level for one minute. There were a total of 5622 measurement samples, of which 1406 were taken when conversation was taking place and 4216 were taken when no conversation was taking place. Figure 6 shows the mean value and mean value ± standard deviation of the measurement results for the target space under two conditions: when conversation was taking place and when no conversation was taking place. Note that the sound pressure level at each frequency is a weighted value using A-weighting.

[0039] Figure 6 shows that when conversation is taking place during the one minute in which the equivalent sound pressure level was measured, i.e., when speech is present, the 500Hz octave band sound pressure level exhibits a dominant frequency characteristic. On the other hand, when no conversation is taking place, i.e., when no speech is present, this characteristic is not observed. These measurement results indicate that speech has its main frequency components in the 500Hz octave band.

[0040] Conversely, if the 500Hz octave band sound pressure level is prominent in the frequency characteristics of the measured sound, it suggests that the measured sound likely contains speech.

[0041] Here, let's consider the relationship between A-weighted sound pressure level and 500Hz octave band sound pressure level. If we assume that the measured sound has frequency components only in the 500Hz octave band, then according to the definition of A-weighted sound pressure level, the difference Ldiff will be 3.2dB.

[0042] In reality, since sound has frequency components in bands other than the 500Hz octave band, the A-weighted sound pressure level and the difference Ldiff will never be exactly 3.2dB. However, the closer the difference Ldiff is to 3.2dB, the more dominant the 500Hz octave band sound pressure level of the measured sound is.

[0043] The difference Ldiff was calculated from the measurement results shown in Figure 6, and Figure 7 shows the frequency distribution of the difference Ldiff in the case where conversation was taking place and the case where no conversation was taking place in the target space.

[0044] Figure 7 shows that when conversation is taking place, the Ldiff difference is larger compared to when no conversation is taking place, meaning the Ldiff difference is close to 3.2 dB.

[0045] The accuracy of the determination changes depending on the threshold value LT set to determine whether speech is present in the sound measured from the difference Ldiff. Figure 8 shows the relationship between the threshold value LT and the determination accuracy.

[0046] In the measurement example shown here, when the threshold LT was set to -1dB, the accuracy of determining whether a conversation was taking place or not was approximately 90%.

[0047] Setting the threshold LT to a large value increases the probability of misidentifying a measured sound as not containing speech when it actually does. Conversely, setting the threshold LT to a small value increases the probability of misidentifying a measured sound as containing speech when it does not. The threshold LT needs to be set taking into account the influence of ambient noise and other factors besides speech.

[0048] In the measurement example shown here, the equivalent sound pressure level is measured over one minute, but the measurement time for the equivalent sound pressure level is not limited to one minute. The measurement time for the equivalent sound pressure level should be set according to the required frequency of sound detection, but generally, it is appropriate to set it to about 10 seconds to 10 minutes. Note that if the measurement time for the equivalent sound pressure level is set to a long duration, if sound is present only for a short period within the measurement time, it may be determined that no sound is present.

[0049] In this embodiment, since communication occurrence is determined by sound pressure level based on such technical grounds, complex data processing is unnecessary in principle.

[0050] It should be noted that the configurations illustrated in the above embodiments and modifications are functionally schematic and do not necessarily have to be physically represented as shown. In other words, the forms of distribution and integration of each device and component are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various usage situations. [Explanation of symbols]

[0051] 1. Communication Occurrence Location Detection System 2 rooms 3. Positioning Tag 4 Locators 5. Positioning device 6,6-1,6-2 Sound pressure level sensor 10 Processing Unit 11. Proximity detection processing unit 12 Communication occurrence determination unit 13 Output section 20 Output device d1,d2 distance E, E11~E15 predetermined range H,H1,H2,H11~H15 people L1, L2, L11~L15 Positioning location Lc average position Ldiff difference LT threshold M mark ΔT time interval

Claims

1. A communication location detection system that can detect the location where communication occurs in a target space where communication between people is possible, A positioning system for determining the location of a person within the aforementioned target space, A sound pressure level sensor for detecting the sound pressure level in the target space, A processing device for detecting the location where the communication occurred based on the location of the person determined by the positioning system and the sound pressure level detected by the sound pressure level sensor, Equipped with, The aforementioned processing apparatus is A proximity determination processing unit determines that there is proximity between people if one or more other people are within a predetermined range centered on the position of the person whose position has been determined, and calculates the average position of the multiple people within the predetermined range. When the proximity detection processing unit determines that there is a person in proximity, it acquires the sound pressure level detected by the sound pressure level sensor closest to the average position, and if the value based on the sound pressure level is greater than or equal to a predetermined value, it determines that communication has occurred between people within the predetermined range. When the communication occurrence determination unit determines that communication has occurred, the output unit outputs the predetermined range, the locations of the multiple people, or the average location as the communication occurrence location. A communication location detection system characterized by having the following features.

2. The proximity determination processing unit, when the predetermined ranges in which multiple people exist partially overlap and are continuous, groups the entire range of the partially overlapping and continuous predetermined ranges as one predetermined range, and determines the average position of the multiple people within the grouped predetermined range, characterized in that it determines the location of a communication occurrence according to claim 1.

3. The communication location detection system according to claim 1 or 2, characterized in that the output unit displays an image indicating the predetermined range where communication occurred, the locations of the multiple people, or the average location at a corresponding position on a display screen that views the target space in plan.

4. The communication location detection system according to claim 3, characterized in that the output unit outputs the location of the communication with the identification information of a person attached.

5. The communication occurrence location detection system according to claim 1, characterized in that the value based on the sound pressure level in the communication occurrence determination unit is the difference obtained by calculating the A-weighted equivalent sound pressure level in a set time interval band and the octave band equivalent sound pressure level with a center frequency of 500 Hz, and subtracting the A-weighted equivalent sound pressure level from the octave band equivalent sound pressure level with a center frequency of 500 Hz in a set time interval band.

6. The communication location detection system according to claim 1, characterized in that the processing device is performed repeatedly at predetermined unit time intervals and the detection information for each predetermined unit time interval is recorded along with the time.

7. A positioning system that determines the location of a person within a target space where person-to-person communication is possible, A sound pressure level sensor for detecting the sound pressure level in the target space, A processing device for detecting the location where the communication occurred based on the location of the person determined by the positioning system and the sound pressure level detected by the sound pressure level sensor, A method for detecting the location where the communication occurs, comprising: The aforementioned processing apparatus is A proximity determination process step that determines that there is proximity between people if one or more other people are within a predetermined range centered on the position of the person whose position has been determined, and calculates the average position of the positions of multiple people within the predetermined range, If the proximity detection process determines that there is a person in proximity, the communication occurrence determination step obtains the sound pressure level detected by the sound pressure level sensor closest to the average position, and if the value based on the sound pressure level is greater than or equal to a predetermined value, it determines that communication has occurred between people within the predetermined range. If the communication occurrence determination step determines that communication has occurred, the output step outputs the predetermined range, the locations of the multiple people, or the average location as the communication occurrence location. A method for detecting the location of a communication event, characterized by including the following:

8. The method for detecting a location where communication occurs, according to claim 7, is characterized in that, if the predetermined ranges in which multiple people exist partially overlap and are continuous, the proximity determination process step involves grouping the entire range of the partially overlapping and continuous predetermined ranges into a single predetermined range, and determining the average position of the multiple people within the grouped predetermined range.

Images