Anomaly detection systems, nurse call systems, intercom systems, and intercom systems for apartment buildings.

Abstract

This invention provides an anomaly detection system that reduces computational processing and enables rapid anomaly detection when detecting anomalies based on microphone sound input from a microphone. [Solution] The anomaly detection system 1 of this disclosure includes a sound image generation unit 15 that generates a first sound image G1 based on the waveform of microphone sound input from a microphone 5, a storage unit 242 that stores a second sound image G2 generated based on the waveform of a reference sound, a sound image comparison unit 16 that compares the first sound image G1 and the second sound image G2, and an anomaly detection unit 17 that detects a predetermined anomaly based on the comparison result by the sound image comparison unit 16.

Description

Technical Field

[0001] The present disclosure relates to an anomaly detection system, a nurse call system including the anomaly detection system, an intercom system, and an intercom system for collective housing. Background Art

[0002] Conventionally, techniques for detecting anomalies based on microphone sound input from a microphone are known. For example, Patent Document 1 proposes a technique for a notification system that inputs sounds around a house through a microphone and determines whether an event to be notified to a resident has occurred based on feature values extracted from the microphone sound. Prior Art Documents Patent Documents

[0003] Patent Document 1 Japanese Unexamined Patent Application Publication No. 2009-260849 Summary of the Invention Problem to be Solved by the Invention

[0004] However, when feature values are used to discriminate sounds, complex calculation processing such as waveform analysis is required, which results in long calculation time and may make it impossible to detect anomalies quickly.

[0005] Therefore, an object of the present disclosure is to provide an anomaly detection system that can reduce arithmetic processing and quickly detect anomalies when detecting anomalies based on microphone sound. Means for Solving the Problem

[0006] To solve the above problems, the anomaly detection system of this disclosure is characterized by comprising: sound image generation means for generating a first sound image based on the waveform of microphone sound input from a microphone; storage means for storing a second sound image generated based on the waveform of a reference sound; sound image comparison means for comparing the first sound image and the second sound image; and anomaly detection means for detecting a predetermined anomaly based on the comparison result by the sound image comparison means. [Effects of the Invention]

[0007] According to the anomaly detection system of this disclosure, anomalies are detected by comparing a first sound image and a second sound image, which represent the waveform of the sound. This eliminates the need for complex calculations such as waveform analysis, reduces computational processing, and enables rapid anomaly detection. [Brief explanation of the drawing]

[0008] [Figure 1] This is a schematic diagram of an anomaly detection system showing one embodiment of the present disclosure. [Figure 2] This is a system diagram of a nurse call system. [Figure 3] This is a block diagram of the control unit. [Figure 4] (a) A schematic diagram showing the first and second sound images, and (b) A schematic diagram showing modified versions of the first and second sound images. [Figure 5] This flowchart shows the operation of the anomaly detection system. [Figure 6] (a) System diagram of an intercom system for individual houses, (b) System diagram of an intercom system for apartment buildings. [Figure 7] This is a block diagram of the in-room master unit. [Modes for carrying out the invention]

[0009] Hereinafter, embodiments of this disclosure implemented as an anomaly detection system, a nurse call system, an intercom system, or an intercom system for apartment buildings will be described with reference to the drawings. In each embodiment, identical or similar components are denoted by the same reference numerals and their descriptions are omitted. [Examples]

[0010] As shown in Figure 1, the anomaly detection system 1 of Embodiment 1 is installed in the nurse call system 2. The anomaly detection system 1 generates a first sound image G1 based on the waveform of the sound input from the microphone 5 of the nurse call sub-unit 21, compares the first sound image G1 with a second sound image G2 generated based on the waveform of a reference sound, and detects a predetermined anomaly.

[0011] Here, the "reference sound" in Example 1 refers to the sound input from microphone 5 under normal conditions. The reference sound may be collected several times to generate multiple second sound images G2, and an average second sound image G2 may be generated using these multiple second sound images G2. "Normal" refers to, for example, a state in which the patient is lying down in bed and at rest. "Abnormal" includes, for example, falling out of bed, tripping, getting up, sitting up, getting out of bed, returning from getting out of bed, etc.

[0012] As shown in Figure 2, the nurse call system 2 consists of a management PC (nurse call server 25) that manages multiple wards, an antenna 27 that provides wireless LAN to multiple wards, and nurse call units U installed in each ward. Each nurse call unit U includes multiple nurse call slave units 21, a nurse call master unit 22, a control unit 24 that controls the entire nurse call system, a portable information terminal 26a such as a PHS carried by medical personnel, corridor lights 23 connected to the multiple nurse call slave units 21, a portable information terminal 26b such as a smartphone that can connect to the wireless LAN, and a HUB 28 that connects the nurse call units U to the nurse call line L1.

[0013] As shown in Figure 3, the control unit 24 is equipped with an abnormality detection system 1. The control unit 24 includes a call control unit 244 that controls calls from the nurse call sub-unit 21 and communication between the nurse call sub-unit 21 and the nurse call master unit 22, a CPU (Central Processing Unit) 241, a storage unit 242, an interface 243 that receives microphone sound input from the microphone 5 of the nurse call sub-unit 21, and a storage unit 242 that stores the second sound image G2.

[0014] The anomaly detection system 1 includes a sound image generation unit 15 that generates a first sound image G1 based on the waveform of the microphone sound input from the microphone 5, a storage unit 242 that stores a second sound image G2 generated based on the waveform of a reference sound, a sound image comparison unit 16 that compares the first sound image G1 and the second sound image G2, and an anomaly detection unit 17 that detects a predetermined anomaly based on the comparison result by the sound image comparison unit 16.

[0015] The sound-image generation unit 15 records the microphone sound input from the microphone 5 into the storage unit 242. The sound-image generation unit 15 starts recording when the microphone sound exceeds a predetermined threshold, and then stops recording when the microphone sound remains below the predetermined threshold for a predetermined period of time. Specifically, for example, recording starts when the microphone sound exceeds 60 decibels, and then stops recording when the microphone sound remains below 30 decibels for 3 seconds. After recording is complete, the sound-image generation unit 15 generates a first sound image G1 based on the waveform of the microphone sound. At this time, the sound-image generation unit 15 can represent the amplitude and period of the waveform by the brightness, hue, or shape of the color, or by using all of these elements.

[0016] The sound-image comparison unit 16 compares the first sound image G1 and the second sound image G2 and calculates the degree of difference. The degree of difference can be expressed as a percentage (%) or the like. The degree of difference can be calculated comprehensively using parameters such as waveform amplitude and period.

[0017] As shown in Fig. 4(a), the first sound image G1 is generated based on the sound waveform of the microphone sound. The second sound image G2 is generated based on the sound waveform of the reference sound. The recording time of the microphone sound and the reference sound may be different. Fig. 4(b) shows first and second sound images G1', G2' which are modified examples of the first and second sound images G1, G2. The first sound image G1' is generated based on the contour of the waveform of the microphone sound, and the second sound image G2' is generated based on the contour of the waveform of the reference sound.

[0018] The abnormality detection unit 17 detects an abnormality when the degree of difference exceeds a predetermined threshold. When the abnormality detection means detects an abnormality, the controller 24 notifies the nurse call master unit 22 of the abnormality, and the nurse call master unit 22 outputs the abnormality by sound, light, an image or the like. Note that the controller 24 may also notify the portable information terminals 26a and 26b of the abnormality.

[0019] The storage unit 242 is configured by a ROM (Read Only Memory) and a RAM (Random Access Memory). The ROM is a non-volatile memory or the like, and stores various programs, as well as files, parameters and the like used in the programs. The RAM temporarily stores data used when executing various programs. The abnormality detection system 1 is a program that is stored in the storage unit 242 and executed by the CPU 241 to function as the abnormality detection system 1.

[0020] Next, the operation of the abnormality detection system 1 having the above configuration will be described with reference to Fig. 5. First, the sound image generation unit 15 records the microphone sound input from the microphone 5 of the nurse call slave unit 21 (S1). Next, the sound image generation unit 15 generates the first sound image G1 based on the waveform of the recorded microphone sound (S2). The sound image comparison unit 16 reads out the second sound image G2 from the storage unit 242, and compares the first sound image G1 with the second sound image G2 (S3). The abnormality detection unit 17 detects an abnormality when the degree of difference exceeds the predetermined threshold (S5: Yes). On the other hand, when the degree of difference is smaller than the predetermined threshold, no abnormality is detected (S5: No).

[0021] According to the anomaly detection system 1 and the nurse call system 2 equipped with the anomaly detection system 1, since the system compares a first sound image G1 representing the waveform of the microphone sound with a second sound image representing the waveform of a reference sound, complex calculations such as sound wave analysis are unnecessary, and anomalies can be detected quickly. [Examples]

[0022] The anomaly detection system 1 of Example 2 is installed in an intercom system 3a for individual houses and an intercom system 3b for apartment buildings.

[0023] The intercom system 3a for individual houses shown in Figure 6(a) comprises an outdoor unit 34, a residential unit 35 equipped with an anomaly detection system 1, and an electric lock 31 that is locked and unlocked by the residential unit 35. The outdoor unit 34 includes a microphone 5. The outdoor unit 34 and the residential unit 35 are connected via an intercom line L2.

[0024] The apartment building intercom system 3b shown in Figure 6(b) comprises a main entrance unit 32, multiple entrance units 34 and multiple room master units 35 installed in each apartment, an electric lock 31 that is locked and unlocked by the room master unit 35, and a controller 33 that controls the entire main intercom system. The main entrance unit 32 includes a microphone 5. The main entrance unit 32 and the room master units 35 are connected via the controller 33 and the intercom line L2.

[0025] As shown in Figure 7, the room master unit 35 is equipped with an abnormality detection system 1. The room master unit 35 includes a communication unit 354 for communicating with the main entrance unit 32 or the entrance sub-unit 34, a CPU 351, an interface 353 for inputting microphone sound from the microphone 5, and a storage unit 352 for storing the second sound image G2. The room master unit 35 unlocks the electric lock 31 when the abnormality detection unit 17 does not detect an abnormality. If an abnormality is detected, the entrance sub-unit 34 or the main entrance unit 32 may output a message such as "Authentication failed."

[0026] Here, in Example 2, "reference tone" refers to "keywords pronounced in the resident's voice," etc. The "keyword" can be selected as, for example, "I'm home." The reference tone may be collected several times to generate multiple second-sound images G2, and an average second-sound image G2 may be generated using these multiple second-sound images G2. In this case, "abnormality" includes, for example, "keywords pronounced in the voice of someone other than the resident," "words other than keywords pronounced in the resident's voice," etc.

[0027] The memory unit 351 consists of ROM and RAM. ROM is a non-volatile memory that stores various programs, files used by those programs, and parameters. RAM temporarily stores data used when various programs are executed. The anomaly detection system 1 is a program stored in the memory unit 352 and executed by the CPU 351 to function as the anomaly detection system 1.

[0028] With the above configurations for individual dwellings (3a) and apartment buildings (3b), the system compares a first sound image (G1) representing the waveform of the microphone sound with a second sound image representing the waveform of a reference sound. This eliminates the need for complex calculations such as sound wave analysis, enabling rapid anomaly detection.

[0029] This disclosure is not limited to the embodiments described above, and it is possible to implement the invention by arbitrarily changing the configuration of each part without departing from the spirit of the invention. [Explanation of Symbols]

[0030] 1. Anomaly detection system 2. Nurse call system 3. Intercom systems (a: individual houses, apartment buildings) 5 microphones 15. Sound and Image Generation Unit 16. Sound and Image Comparison Section 17 Anomaly detection unit 21 Nurse call handset 22 Nurse call master unit 23 Corridor light 24 Control Unit 241 CPU 242 Storage section 243 Interface 244 Call Control Unit 25 Nurse call server 26. Mobile Information Terminals 27 Antennas 28 HUB 31 Electric lock 32. Entrance panel for all buildings 33. Control Unit 34 Door station 35. Room master unit 351 CPU 352 Storage section 353 Interface 354 Telephone section 355 Operation section 356 Monitors G1 First Sound Image G2 2nd sound image U Nurse Call Unit L1 Nurse call line L2 Intercom Line

Claims

1. An anomaly detection system comprising: sound image generation means for generating a first sound image based on the waveform of microphone sound input from a microphone; storage means for storing a second sound image generated based on the waveform of a reference sound; sound image comparison means for comparing the first sound image and the second sound image; and anomaly detection means for detecting a predetermined anomaly based on the comparison result by the sound image comparison means.

2. A nurse call system comprising the abnormality detection system described in claim 1, It comprises a nurse call sub-unit, a nurse call master unit, a control unit that controls the entire nurse call system, and a portable information terminal carried by medical personnel. The nurse call sub-unit includes the microphone, The controller includes the storage means, the sound-image generation means, the sound-image comparison means, and the anomaly detection means. The control unit is a nurse call system that notifies the nurse call master unit and / or the portable information terminal of an abnormality when the abnormality detection means detects an abnormality.

3. An intercom system comprising the abnormality detection system described in claim 1, It comprises an outdoor unit, a master unit in the living room, and an electric lock that is unlocked by the master unit in the living room, The aforementioned door station includes the microphone, The aforementioned room master unit includes the storage means, the sound-image generation means, the sound-image comparison means, and the anomaly detection means, The aforementioned room master unit is an intercom system that unlocks the electric lock when the abnormality detection means does not detect an abnormality.

4. An intercom system for an apartment building equipped with the abnormality detection system described in claim 1, The system comprises a main entrance unit, an entrance sub-unit, a room master unit, an electric lock unlocked by the room master unit, and a control unit that controls the entire main intercom system. The aforementioned entrance unit includes the microphone, The aforementioned room master unit includes the storage means, the sound-image generation means, the sound-image comparison means, and the anomaly detection means, The aforementioned room master unit is an intercom system that unlocks the electric lock when the abnormality detection means does not detect an abnormality.

Images