CN113247730A - Elevator passenger screaming detection method and system based on multi-dimensional features - Google Patents
Elevator passenger screaming detection method and system based on multi-dimensional features Download PDFInfo
- Publication number
- CN113247730A CN113247730A CN202110645178.1A CN202110645178A CN113247730A CN 113247730 A CN113247730 A CN 113247730A CN 202110645178 A CN202110645178 A CN 202110645178A CN 113247730 A CN113247730 A CN 113247730A
- Authority
- CN
- China
- Prior art keywords
- data
- elevator
- audio
- extracting
- imf
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0006—Monitoring devices or performance analysers
- B66B5/0018—Devices monitoring the operating condition of the elevator system
- B66B5/0031—Devices monitoring the operating condition of the elevator system for safety reasons
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0006—Monitoring devices or performance analysers
- B66B5/0012—Devices monitoring the users of the elevator system
Abstract
The invention relates to an elevator passenger screech detection method and system based on multi-dimensional characteristics, wherein the method comprises the following steps: a. acquiring acceleration data and audio data of elevator operation, and preprocessing the acquired data; b. analyzing the acceleration data to judge whether the elevator runs abnormally, and if so, extracting the audio features of the audio data; c. and analyzing the audio features to judge whether the phenomenon of screaming of passengers occurs in the elevator, and if so, sending an alarm signal. The invention can report the elevator safety accident in time by a method for detecting the passenger screaming, thereby ensuring the safety of the passenger.
Description
Technical Field
The invention relates to an elevator passenger screaming detection method and system based on multi-dimensional characteristics.
Background
In modern life, the elevator brings a high-efficiency convenient living environment for people, so that the elevator is more and more widely used. However, once the elevator breaks down in the operation process, if the elevator cannot be rescued in time, the life safety of passengers can be seriously threatened. Some fault judgment methods exist in the prior art, which can detect the noise condition, but have certain limitations. For example, patent CN105679313A discloses a solution that a gaussian mixture model matches voiceprint data to achieve noise detection. It can be seen that the scheme of the patent utilizes a relatively traditional algorithm, and only uses the voiceprint data, and the detection effect is not comprehensive.
Disclosure of Invention
The invention aims to provide an elevator passenger screaming detection method and system based on multi-dimensional characteristics.
In order to achieve the aim, the invention provides an elevator passenger screaming detection method and system based on multi-dimensional characteristics, wherein the method comprises the following steps:
a. acquiring acceleration data and audio data of elevator operation, and preprocessing the acquired data;
b. analyzing the acceleration data to judge whether the elevator runs abnormally, and if so, extracting the audio features of the audio data;
c. and analyzing the audio features to judge whether the phenomenon of screaming of passengers occurs in the elevator, and if so, sending an alarm signal.
According to one aspect of the invention, in the step (a), the preprocessing includes cleaning the data to remove the dirty data without audio frequency and with too large and/or too small acceleration values.
According to one aspect of the invention, in the step (a), the audio data is a single-channel audio signal extracted from elevator video monitoring data, the audio sampling frequency is 22.05kHz, and the duration is between 1 and 10 s.
According to an aspect of the invention, the preprocessing further includes performing residual calculation on the acceleration data to obtain a residual sequence x ═ x1，x2，x3，...，xn]。
According to one aspect of the invention, in the step (b), when the maximum and minimum difference min _ max of the residual error sequence is greater than 70, it is determined that the elevator operation is abnormal.
According to an aspect of the invention, in the step (b), the extracting of the audio features comprises CEEMD feature extraction and feature acquisition;
extracting time domain features for each layer of IMF of the audio data using CEEMD while performing the CEEMD feature extraction;
extracting IMF layer energy E reflecting signal time domain energy changeIMF：
Extracting an IMF layer energy ratio sigma reflecting energy distribution of different frequency scales:
wherein E isxIs the short-time energy of the frame signal, y (t) is the corresponding IMF time domain signal;
and when the characteristic acquisition is carried out, extracting Fbank and LPCC from each extracted IMF layer to obtain the frequency domain characteristic of each IMF layer.
According to an aspect of the invention, the extracting of the audio features further comprises extracting Fbank, LPCC, EIMFAnd the sum sigma is spliced to obtain the final input characteristic.
According to an aspect of the present invention, in the step (c), identifying whether a scream occurs in the elevator at the current time period by using a scream detection model, the scream detection model being a deep neural network model;
the deep neural network model comprises Conv1: 3256 Relu, Conv2: 3256 Relu, MaxPool1:2, Conv3: 3128 Relu, Conv4: 3128 Relu, MaxPool2:2, Dense1:512Relu and Dense2:2 Softmax;
wherein Conv is a convolutional neural network, Dense is a fully-connected neural network, the convolutional kernel size of the convolutional network is 3 x 3, 256 and 128 are filter numbers, Relu is an activation function, MaxPool2:2,2 is maximum pooling with 2 steps and 512 and 2 are hidden layer neuron numbers.
Elevator passenger screech detection system based on multidimensional characteristic includes:
the data acquisition module is used for acquiring elevator operation data in real time and preprocessing the data;
the audio characteristic acquisition module is used for extracting audio characteristics in the audio data;
the abnormal operation identification module is used for analyzing the acceleration data and judging whether the elevator operates abnormally or not;
and the scream detection module is used for detecting whether the scream phenomenon of passengers occurs in the elevator and selecting whether to give an alarm according to the detection result.
According to one aspect of the invention, the data acquisition module includes a monitoring camera and an acceleration sensor.
According to the scheme of the invention, video monitoring data of the elevator and acceleration data of the elevator running are collected. And analyzing the acceleration data in real time, calculating residual errors to find the elevator time interval of abnormal operation, intercepting the audio data of the elevator in the current time interval from the video monitoring data, analyzing, identifying whether passengers exist and further confirming whether screaming occurs.
Drawings
Fig. 1 schematically shows a flow chart of an elevator passenger screech detection method based on multi-dimensional features according to an embodiment of the present invention;
fig. 2 schematically shows an abnormal acceleration data (a), an abnormal acceleration residual (b), a normal acceleration data (c), and a normal acceleration residual (d) in an elevator passenger screech detection method based on multi-dimensional features according to an embodiment of the present invention;
fig. 3 schematically shows an audio feature extraction flow chart in a multi-dimensional feature-based elevator passenger screaming detection method according to an embodiment of the invention;
fig. 4 schematically shows a scream detection model structure diagram in an elevator passenger scream detection method based on multi-dimensional features according to an embodiment of the present invention.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
In describing embodiments of the present invention, the terms "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like are used in an orientation or positional relationship that is based on the orientation or positional relationship shown in the associated drawings, which is for convenience and simplicity of description only, and does not indicate or imply that the referenced device or element must have a particular orientation, be constructed and operated in a particular orientation, and thus, the above-described terms should not be construed as limiting the present invention.
The present invention is described in detail below with reference to the drawings and the specific embodiments, which are not repeated herein, but the embodiments of the present invention are not limited to the following embodiments.
Referring to fig. 1, the elevator passenger screech detection method based on multi-dimensional features of the invention first collects acceleration data and audio data of elevator operation and preprocesses the collected data. Analyzing the acceleration data in real time to judge whether the elevator runs abnormally, and if so, extracting the audio features of the audio data. And analyzing the audio features to judge whether the phenomenon of screaming of passengers occurs in the elevator, and if so, sending an alarm signal. Therefore, the elevator safety monitoring system monitors abnormal conditions in the elevator in real time based on the elevator running acceleration data and the audio data in the elevator running, and immediately sends out an alarm if the screaming condition of passengers occurs, and feeds back the alarm to related units, so that elevator safety accidents are reported in time, rescue is conveniently implemented in time, and the safety of the passengers is guaranteed.
In the present invention, the pre-processing is essentially a cleaning of the data to ensure that the collected data is available for subsequent testing. Specifically, for the acceleration data, if too large and/or too small acceleration values are collected, the data may be erroneous (i.e., dirty data), and therefore, the data should be deleted. In addition, the audio data of the invention is a single-channel audio signal extracted from the elevator video monitoring data, the audio sampling frequency is 22.05kHz, and the duration (or length) is different from 1 to 10 s. Therefore, if the collected video monitoring data does not contain audio, the data is also judged to be dirty data and deleted.
Referring to fig. 2, for the acceleration data, the preprocessing further includes performing residual calculation (or residual processing) to obtain a residual sequence x ═ x1，x2，x3，...，xn]Wherein x isnFor the nth value in the acceleration residual sequence of length n, the aim is to remove the influence of the acceleration and deceleration phase. After residual decomposition, the influence of the acceleration and deceleration stages is eliminated, and the residual of the real abnormal region is still large, as shown in fig. 2b and 2d, so that the acceleration with large residual can be defined as the abnormal acceleration, thereby distinguishing the abnormal acceleration from the normal acceleration as shown in fig. 2a and 2 c. Specifically, when the maximum and minimum value difference min _ max of the residual sequence x is greater than 70, the abnormal acceleration can be judged, and the abnormal operation of the elevator in the current time period can be further judged.
Referring to fig. 3, the extraction of audio features of the present invention includes CEEMD (complementary set empirical mode decomposition) feature extraction and feature acquisition. CEEMD is an adaptive spatio-temporal analysis method suitable for processing non-stationary non-linear sequences, similar to fourier transforms and wavelet decompositions, and is particularly suitable for analyzing natural signals, which are typically non-linear and non-stationary, such as audio signals. In CEEMD feature extraction, the CEEMD is used to extract time-domain features for each layer of IMF (implicit modal components) of the audio data (i.e., the active audio frame signal). Specifically, IMF layer energy E reflecting signal time domain energy change is extractedIMF：
And extracting an IMF layer energy ratio sigma reflecting energy distribution of different frequency scales:
wherein E isxFor the short-time energy of the frame signal, y (t) is the corresponding IMF time-domain signal, N represents the length of the time-domain signal, tth time unit.
When the characteristics are collected, the frequency domain characteristics of each IMF layer are obtained by extracting Fbank and LPCC from each extracted IMF layer (seven layers in total). Fbank is frequency domain data obtained by processing audio in a mode similar to human ears, and the performance of voice recognition can be improved. The LPCC is a frequency domain feature in abnormal sound recognition, and has the advantages of high reliability and strong robustness. Finally, the extracted Fbank, LPCC and E are processedIMFAnd the sum sigma is spliced to obtain the final input characteristic.
Referring to fig. 4, the present invention identifies whether a scream occurs in an elevator at a current time period using a scream detection model, wherein the scream detection model is a deep neural network model. The structure of the deep neural network model is shown in fig. 4, wherein Conv is a convolutional neural network, density is a fully-connected neural network, Conv1: 3256 Relu is a first layer convolutional network, the size of a convolution kernel is 3 × 3, 256 is the number of filters, Relu is an activation function, MaxPool: 2,2 is the maximum value pooling with the size of 2 and the step length of 2, Dense:512Relu is a fully-connected neural network with the hidden layer neuron number of 512, and the rest of the layer structures are analogized. In this embodiment, the deep neural network model includes Conv1: 3256 Relu, Conv2: 3256 Relu, MaxPool1:2, Conv3: 3128 Relu, Conv4: 3128 Relu, MaxPool2:2, Dense1:512Relu, and Dense2:2 Softmax.
The invention relates to an elevator passenger screaming detection system based on multi-dimensional characteristics, which comprises: the data acquisition module is used for acquiring elevator operation data in real time and preprocessing the data; the audio characteristic acquisition module is used for extracting audio characteristics in the audio data; the abnormal operation identification module is used for analyzing the acceleration data and judging whether the elevator operates abnormally or not; and the scream detection module is used for detecting whether the scream phenomenon of passengers occurs in the elevator and selecting whether to give an alarm according to a judgment result. Wherein, data acquisition module includes surveillance camera head and acceleration sensor, so can use the camera to carry out the record to the condition when the elevator moves, uses acceleration sensor to carry out the record to acceleration data simultaneously.
In conclusion, the elevator monitoring system based on the mass elevator collected data comprises elevator running acceleration data and audio data in elevator running, monitors abnormal conditions in the elevator in real time, analyzes various elevator audio data when abnormal conditions occur, and immediately gives an alarm if a passenger scream occurs. Therefore, the method has a wide application range, and the deep neural network model has a better detection effect compared with the traditional algorithm, and can accurately monitor screaming sounds (including sounds generated by terrorism, distress, crying and the like threatened by personal safety) sent by passengers in the elevator in real time.
The above description is only one embodiment of the present invention, and is not intended to limit the present invention, and it is apparent to those skilled in the art that various modifications and variations can be made in the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A multi-dimensional feature-based elevator passenger screaming detection method comprises the following steps:
a. acquiring acceleration data and audio data of elevator operation, and preprocessing the acquired data;
b. analyzing the acceleration data to judge whether the elevator runs abnormally, and if so, extracting the audio features of the audio data;
c. and analyzing the audio features to judge whether the phenomenon of screaming of passengers occurs in the elevator, and if so, sending an alarm signal.
2. The method of claim 1, wherein in step (a), the pre-processing comprises washing the data to remove dirty data with no audio and with too large and/or too small acceleration values.
3. The method of claim 1, wherein in step (a), the audio data is a single channel audio signal extracted from elevator video surveillance data, the audio sampling frequency is 22.05kHz, and the duration is between 1-10 s.
4. The method of claim 2, wherein the preprocessing further comprises performing a residual calculation on the acceleration data to obtain a residual sequence x ═ x [ ]1，x2，x3，...xn]。
5. The method according to claim 4, characterized in that in step (b), when the maximum and minimum difference min _ max of the residual error sequence is greater than 70, it is determined that the elevator operation is abnormal.
6. The method of claim 1, wherein in step (b), the extracting of the audio features comprises CEEMD feature extraction and feature acquisition;
extracting time domain features for each layer of IMF of the audio data using CEEMD while performing the CEEMD feature extraction;
extracting IMF layer energy E reflecting signal time domain energy changeIMF：
Extracting an IMF layer energy ratio sigma reflecting energy distribution of different frequency scales:
wherein E isxIs the short-time energy of the frame signal, y (t) is the corresponding IMF time domain signal;
and when the characteristic acquisition is carried out, extracting Fbank and LPCC from each extracted IMF layer to obtain the frequency domain characteristic of each IMF layer.
7. The method of claim 6, wherein the extracting of the audio features further comprises extracting Fbank, LPCC, EIMFAnd the sum sigma is spliced to obtain the final input characteristic.
8. The method of claim 1, wherein in the step (c), whether a screech occurs in the elevator at the current time period is identified using a screech detection model, the screech detection model being a deep neural network model;
the deep neural network model comprises Conv1: 3256 Relu, Conv2: 3256 Relu, MaxPool1:2, Conv3: 3128 Relu, Conv4: 3128 Relu, MaxPool2:2, Dense1:512Relu and Dense2:2 Softmax;
wherein Conv is a convolutional neural network, Dense is a fully-connected neural network, the convolutional kernel size of the convolutional network is 3 x 3, 256 and 128 are filter numbers, Relu is an activation function, MaxPool2:2,2 is maximum pooling with 2 steps and 512 and 2 are hidden layer neuron numbers.
9. A system for implementing the multi-dimensional feature-based elevator passenger screech detection method of any one of claims 1-8, comprising:
the data acquisition module is used for acquiring elevator operation data in real time and preprocessing the data;
the audio characteristic acquisition module is used for extracting audio characteristics in the audio data;
the abnormal operation identification module is used for analyzing the acceleration data and judging whether the elevator operates abnormally or not;
and the scream detection module is used for detecting whether the scream phenomenon of passengers occurs in the elevator and selecting whether to give an alarm according to the detection result.
10. The system of claim 9, wherein the data acquisition module comprises a monitoring camera and an acceleration sensor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110645178.1A CN113247730B (en) | 2021-06-10 | 2021-06-10 | Elevator passenger screaming detection method and system based on multi-dimensional features |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110645178.1A CN113247730B (en) | 2021-06-10 | 2021-06-10 | Elevator passenger screaming detection method and system based on multi-dimensional features |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113247730A true CN113247730A (en) | 2021-08-13 |
| CN113247730B CN113247730B (en) | 2022-11-08 |
Family
ID=77187267
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110645178.1A Active CN113247730B (en) | 2021-06-10 | 2021-06-10 | Elevator passenger screaming detection method and system based on multi-dimensional features |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113247730B (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004018174A (en) * | 2002-06-17 | 2004-01-22 | Hitachi Ltd | Moving body device for diagnosis |
| CN102897626A (en) * | 2012-11-02 | 2013-01-30 | 智眼天下通（北京）科技发展有限公司 | Elevator monitoring and alarming device |
| JP2013086889A (en) * | 2011-10-14 | 2013-05-13 | Hitachi Building Systems Co Ltd | Elevator control device |
| CN104655425A (en) * | 2015-03-06 | 2015-05-27 | 重庆大学 | Bearing fault classification diagnosis method based on sparse representation and LDM (large margin distribution machine) |
| US20160213286A1 (en) * | 2015-01-26 | 2016-07-28 | National Chiao Tung University | Electronic device, monitoring and feedback system on thoracoabdominal motion and method thereof |
| CN107522053A (en) * | 2017-07-11 | 2017-12-29 | 浙江新再灵科技股份有限公司 | A kind of elevator platform detecting system and method based on audio analysis |
| CN108382940A (en) * | 2018-03-16 | 2018-08-10 | 深圳市敢为特种设备物联网技术有限公司 | Lift running safety monitoring method, equipment and readable storage medium storing program for executing |
| CN108831439A (en) * | 2018-06-27 | 2018-11-16 | 广州视源电子科技股份有限公司 | Audio recognition method, device, equipment and system |
| CN109632311A (en) * | 2019-01-21 | 2019-04-16 | 北京化工大学 | A kind of adaptive acoustical signal Method for Bearing Fault Diagnosis |
| CN110033787A (en) * | 2018-01-12 | 2019-07-19 | 英特尔公司 | Trigger the audio event of video analysis |
| CN110189769A (en) * | 2019-05-23 | 2019-08-30 | 复钧智能科技(苏州)有限公司 | Abnormal sound detection method based on multiple convolutional neural networks models couplings |
| US20200196875A1 (en) * | 2017-05-22 | 2020-06-25 | Adaptive, Intelligent And Dynamic Brain Corporation (Aidbrain) | Method, module and system for analysis of physiological signals |
| CN111742365A (en) * | 2018-02-28 | 2020-10-02 | 罗伯特·博世有限公司 | System and method for audio event detection in a monitoring system |
| CN112320520A (en) * | 2020-11-09 | 2021-02-05 | 浙江新再灵科技股份有限公司 | Elevator abnormal vibration detection method based on residual error analysis |
| CN112693985A (en) * | 2020-12-10 | 2021-04-23 | 太原理工大学 | Non-invasive elevator state monitoring method fusing sensor data |
-
2021
- 2021-06-10 CN CN202110645178.1A patent/CN113247730B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004018174A (en) * | 2002-06-17 | 2004-01-22 | Hitachi Ltd | Moving body device for diagnosis |
| JP2013086889A (en) * | 2011-10-14 | 2013-05-13 | Hitachi Building Systems Co Ltd | Elevator control device |
| CN102897626A (en) * | 2012-11-02 | 2013-01-30 | 智眼天下通（北京）科技发展有限公司 | Elevator monitoring and alarming device |
| US20160213286A1 (en) * | 2015-01-26 | 2016-07-28 | National Chiao Tung University | Electronic device, monitoring and feedback system on thoracoabdominal motion and method thereof |
| CN104655425A (en) * | 2015-03-06 | 2015-05-27 | 重庆大学 | Bearing fault classification diagnosis method based on sparse representation and LDM (large margin distribution machine) |
| US20200196875A1 (en) * | 2017-05-22 | 2020-06-25 | Adaptive, Intelligent And Dynamic Brain Corporation (Aidbrain) | Method, module and system for analysis of physiological signals |
| CN107522053A (en) * | 2017-07-11 | 2017-12-29 | 浙江新再灵科技股份有限公司 | A kind of elevator platform detecting system and method based on audio analysis |
| CN110033787A (en) * | 2018-01-12 | 2019-07-19 | 英特尔公司 | Trigger the audio event of video analysis |
| CN111742365A (en) * | 2018-02-28 | 2020-10-02 | 罗伯特·博世有限公司 | System and method for audio event detection in a monitoring system |
| CN108382940A (en) * | 2018-03-16 | 2018-08-10 | 深圳市敢为特种设备物联网技术有限公司 | Lift running safety monitoring method, equipment and readable storage medium storing program for executing |
| CN108831439A (en) * | 2018-06-27 | 2018-11-16 | 广州视源电子科技股份有限公司 | Audio recognition method, device, equipment and system |
| CN109632311A (en) * | 2019-01-21 | 2019-04-16 | 北京化工大学 | A kind of adaptive acoustical signal Method for Bearing Fault Diagnosis |
| CN110189769A (en) * | 2019-05-23 | 2019-08-30 | 复钧智能科技(苏州)有限公司 | Abnormal sound detection method based on multiple convolutional neural networks models couplings |
| CN112320520A (en) * | 2020-11-09 | 2021-02-05 | 浙江新再灵科技股份有限公司 | Elevator abnormal vibration detection method based on residual error analysis |
| CN112693985A (en) * | 2020-12-10 | 2021-04-23 | 太原理工大学 | Non-invasive elevator state monitoring method fusing sensor data |
Non-Patent Citations (3)
| Title |
|---|
| 吴真玮等: "基于CEEMD和自适应中值滤波的大地电磁数据去噪研究", 《世界地质》 * |
| 杨元威等: "基于声音信号的高压断路器机械故障诊断方法", 《中国电机工程学报》 * |
| 梁仲明等: "基于CEEMD-参数优化Morlet小波变换的柴油机振动信号冲击特征提取方法", 《船舶工程》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113247730B (en) | 2022-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107527617A (en) | Monitoring method, apparatus and system based on voice recognition | |
| CN108648748B (en) | Acoustic event detection method under hospital noise environment | |
| CN109616140B (en) | Abnormal sound analysis system | |
| CN110706700B (en) | In-vehicle disturbance prevention alarm method and device, server and storage medium | |
| WO2018036281A1 (en) | Smart home monitoring method and system based on vibration detection | |
| CN108877146A (en) | It is a kind of that safety automatic-alarming devices and methods therefor is driven based on multiplying for intelligent sound identification | |
| CN111063162A (en) | Silent alarm method and device, computer equipment and storage medium | |
| CN112149638B (en) | Personnel identity recognition system construction and use method based on multi-modal biological characteristics | |
| KR20180122171A (en) | Sound event detection method using deep neural network and device using the method | |
| CN111436944B (en) | Falling detection method based on intelligent mobile terminal | |
| CN111354366A (en) | Abnormal sound detection method and abnormal sound detection device | |
| US6876964B1 (en) | Apparatus for detecting fatigue and doze by voice, and recording medium | |
| CN112078590B (en) | Driving behavior monitoring method and system | |
| CN112478975A (en) | Elevator door fault detection method based on audio features | |
| CN114436087B (en) | Deep learning-based elevator passenger door-pulling detection method and system | |
| Pan et al. | Cognitive acoustic analytics service for Internet of Things | |
| CN113345399A (en) | Method for monitoring sound of machine equipment in strong noise environment | |
| CN113247730B (en) | Elevator passenger screaming detection method and system based on multi-dimensional features | |
| CN112850408B (en) | Elevator emergency stop trapped person fault detection method based on multi-model fusion | |
| CN110472563B (en) | Vertical ladder vibration over-large fault diagnosis method based on wavelet packet decomposition and neural network | |
| Pramanick et al. | Deep learning based urban sound classification and ambulance siren detector using spectrogram | |
| CN112957044A (en) | Driver emotion recognition system based on double-layer neural network model | |
| CN112960506B (en) | Elevator warning sound detection system based on audio frequency characteristics | |
| Ghiurcau et al. | A modified TESPAR algorithm for wildlife sound classification | |
| CN113599052A (en) | Snore monitoring method and system based on deep learning algorithm and corresponding electric bed control method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |