CN102543072A - System and method for real-time detection of fatigue - Google Patents
System and method for real-time detection of fatigue Download PDFInfo
- Publication number
- CN102543072A CN102543072A CN2010105753771A CN201010575377A CN102543072A CN 102543072 A CN102543072 A CN 102543072A CN 2010105753771 A CN2010105753771 A CN 2010105753771A CN 201010575377 A CN201010575377 A CN 201010575377A CN 102543072 A CN102543072 A CN 102543072A
- Authority
- CN
- China
- Prior art keywords
- voice
- fatigue
- real
- voice signal
- microphone
- 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.)
- Pending
Links
Images
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a system and a method for real-time detection of fatigue in the field of voice processing. The system for real-time detection of fatigue comprises a microphone used for collecting voice, wherein the microphone transmits the collected voice signal to a processor for detecting fatigue, the processor comprises a voice collecting module used for acquiring the voice collected by the microphone, the voice collecting module transmits the collected voice signal to a voice pre-processing module to carry out low-pass filtration, the pre-processed voice signal enters a characteristic extracting module to undergo MFCC (Mel Frequency Cepstrum Coefficient) parameter extraction, a reference template is extracted from the extracted parameter and then trained in a neural network, and the trained voice signal and a pre-processed voice sample are input into the neural network to be tested so as to obtain an experimental result. According to the system and the method for real-time detection of fatigue, the fatigue driving is warned in advance to reduce the accidents caused by the fatigue driving. The system and the method for real-time detection of fatigue, provided by the invention, are suitable for detecting fatigue under all kinds of driving environments.
Description
Technical field
The present invention relates to the computing machine embedded system and use and the speech processes field particularly a kind of fatigue detecting system and method.
Background technology
Fatigue is a kind of spontaneous phenomenon, is a kind of self-control and the defencive function of human body.There is data to show, in the traffic hazard that highway takes place, has over half owing to long-time fatigue driving or finding target dullness do not concentrate driver attention even reason such as doze causes.For reducing the accident of this respect, the fatigue strength test just has crucial meaning.The tired inducement that also often becomes brain and heart disease as detect the condition of oneself in real time through simple method, for prevent disease, reduces artificial accident and also has positive meaning.
The detection method of fatigue strength may be summarized to be objective and subjective two aspects.The domestic method of mainly taking subjective evaluation and test; Mainly wait the testee's that tests and assesses degree of fatigue according to self record sheet, sleep quality record sheet, personal behavior record sheet; Though subjective evaluation method uses simple; But be difficult to quantize tired grade and degree, the understanding because of each one has evident difference again, and its result often can not be satisfactory.External then mainly take the method for objective test and appraisal, retina detection based on the detection of behavioural characteristic, head position detection, direction of visual lines detection etc. are arranged and based on the EEG signal detection of the detection of physiological parameter, ECG signal detection, beat pulse detection, saliva detection, other physiological signal detection etc.Its weak point is: these methods be although can understand people's fatigue state to a certain extent, but also are not clear especially to everyone tired psychology, physiological attribute, and the Changing Pattern under the fatigue state is difficult to sum up and concludes; Present most of detection algorithm is because of the restriction of its testing conditions and the influence of complex environment, and detecting effect can not be entirely satisfactory; Cost performance is a problem that needs to be resolved hurrily, if cost too greatly then be difficult to widespread use.
Summary of the invention
To the objective of the invention is to the traffic hazard that causes because of fatigue driving in order solving, a kind of tired real-time detecting system and method to be provided, make it can in time detect driver's fatigue conditions, reduce accident rate.
For solving above technical matters; A kind of tired real-time detecting system provided by the present invention and method; Comprise that the microphone that is used to gather sound, said microphone give the transmission of sound signals that collects the processor that is used to carry out fatigue detecting; Said processor comprises and is used to obtain the sound collection module that microphone collects sound, and said sound collection module flows to the sound pre-processing module with the voice signal that collects and carries out LPF, and pretreated voice signal gets into characteristic extracting module and carries out the MFCC parameter extraction; From the parameter of extracting, extract reference template again; Put into neural network again and train, voice signal and pretreated speech samples input neural network after the training are tested, draw experimental result.
When the present invention worked, the collection of voice signal was accomplished through software, and the voice signal that collects carries out the sound pre-service again; Carry out LPF, filtering is higher than the signal content of 1/2 sampling rate, from pretreated signal, extracts reference template; Fatigue strength is the 1-5 level from low to high; Put into neural network then and train, test pretreated voice signal input neural network subsequently, the contrast reference template draws experimental result.The present invention is applicable to the fatigue detecting under the various driving environments.
As improvement of the present invention, said voice collecting process is accomplished through Cooledit software, and the voice of recording are preserved with the wave form.
As further improvement of the present invention, as experimental subjects, each digital speech is respectively recorded 40 in the morning 4:00,10:00 and afternoon 4:00, four periods of 10:00 respectively with vowel [a :], and totally 160 digital speechs are as the data source of experiment.
Description of drawings
Fig. 1 is a workflow diagram of the present invention.
Embodiment
As shown in Figure 1; A kind of tired real-time detecting system and method; Comprise that the microphone that is used to gather sound, said microphone give the transmission of sound signals that collects the processor that is used to carry out fatigue detecting; Said processor comprises and is used to obtain the sound collection module that microphone collects sound, and said sound collection module flows to the sound pre-processing module with the voice signal that collects and carries out LPF, and pretreated voice signal gets into characteristic extracting module and carries out the MFCC parameter extraction; From the parameter of extracting, extract reference template again; Put into neural network again and train, voice signal and pretreated speech samples input neural network after the training are tested, draw experimental result.
In the work; The collection of voice signal is accomplished through Cooledit software, and the voice signal that collects carries out the sound pre-service again, carries out LPF; Filtering is higher than the signal content of 1/2 sampling rate; As experimental subjects, each digital speech is respectively recorded 40 in the morning 4:00,10:00 and afternoon 4:00, four periods of 10:00 respectively with vowel [a :], and totally 160 digital speechs are as the data source of experiment; From pretreated signal, extract 10 reference templates; Fatigue strength is the 1-5 level from low to high; Put into neural network then and train, test pretreated 160 voice signal input neural networks subsequently, the contrast reference template draws experimental result.
Except that the foregoing description, the present invention can also have other embodiments.All employings are equal to the technical scheme of replacement or equivalent transformation formation, all drop in the protection domain of requirement of the present invention.
Claims (3)
1. tired real-time detecting system and method; It is characterized in that: comprise that the microphone that is used to gather sound, said microphone give the transmission of sound signals that collects the processor that is used to carry out fatigue detecting; Said processor comprises and is used to obtain the sound collection module that microphone collects sound; Said sound collection module flows to the sound pre-processing module with the voice signal that collects and carries out LPF; Pretreated voice signal gets into characteristic extracting module and carries out the MFCC parameter extraction, from the parameter of extracting, extracts reference template again, puts into neural network again and trains; Voice signal and pretreated speech samples input neural network after the training are tested, drawn experimental result.
2. a kind of tired real-time detecting system according to claim 1 and method is characterized in that, said voice collecting process is accomplished through Cooledit software, and the voice of recording are preserved with the wave form.
3. a kind of tired real-time detecting system according to claim 1 and 2 and method; It is characterized in that; With vowel [a :] as experimental subjects; Each digital speech is respectively recorded 40 in the morning 4:00,10:00 and afternoon 4:00, four periods of 10:00 respectively, and totally 160 digital speechs are as the data source of experiment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105753771A CN102543072A (en) | 2010-12-07 | 2010-12-07 | System and method for real-time detection of fatigue |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105753771A CN102543072A (en) | 2010-12-07 | 2010-12-07 | System and method for real-time detection of fatigue |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102543072A true CN102543072A (en) | 2012-07-04 |
Family
ID=46349812
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010105753771A Pending CN102543072A (en) | 2010-12-07 | 2010-12-07 | System and method for real-time detection of fatigue |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102543072A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106828098A (en) * | 2016-12-22 | 2017-06-13 | 吴中区穹窿山倪源交通器材经营部 | A kind of driver's nerves reaction monitoring system |
| CN110580908A (en) * | 2019-09-29 | 2019-12-17 | 出门问问信息科技有限公司 | command word detection method and device supporting different languages |
-
2010
- 2010-12-07 CN CN2010105753771A patent/CN102543072A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106828098A (en) * | 2016-12-22 | 2017-06-13 | 吴中区穹窿山倪源交通器材经营部 | A kind of driver's nerves reaction monitoring system |
| CN106828098B (en) * | 2016-12-22 | 2019-02-01 | 威马汽车科技集团有限公司 | A kind of driver's nerves reaction monitoring system |
| CN110580908A (en) * | 2019-09-29 | 2019-12-17 | 出门问问信息科技有限公司 | command word detection method and device supporting different languages |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021174618A1 (en) | Training method for electroencephalography mode classification model, classification method and system | |
| CN108765876A (en) | Driving fatigue depth analysis early warning system based on multimode signal and method | |
| CN105286890B (en) | Driver doze state monitoring method based on electroencephalogram signals | |
| CN105877766A (en) | Mental state detection system and method based on multiple physiological signal fusion | |
| CN110604565A (en) | Brain health screening method based on portable EEG equipment | |
| CN106691474A (en) | Brain electrical signal and physiological signal fused fatigue detection system | |
| CN102940490B (en) | Method for extracting motor imagery electroencephalogram signal feature based on non-linear dynamics | |
| CN105147248A (en) | Physiological information-based depressive disorder evaluation system and evaluation method thereof | |
| CN103405225B (en) | A kind of pain that obtains feels the method for evaluation metrics, device and equipment | |
| Cohen-McFarlane et al. | Comparison of silence removal methods for the identification of audio cough events | |
| Hu et al. | A real-time electroencephalogram (EEG) based individual identification interface for mobile security in ubiquitous environment | |
| Turnip et al. | Deception detection of EEG-P300 component classified by SVM method | |
| CN113208629A (en) | Alzheimer disease screening method and system based on EEG signal | |
| CN105700687A (en) | Single-trial electroencephalogram P300 component detection method based on folding HDCA algorithm | |
| CN107865638A (en) | Computer-readable recording medium, built-in earplug detection means | |
| CN103892797A (en) | Signal processing method and device for sleep structure analysis | |
| CN113208620A (en) | Sleep stage based Alzheimer disease screening method and system | |
| CN102543072A (en) | System and method for real-time detection of fatigue | |
| CN113080971A (en) | Method and system for judging fatigue state by detecting blink signals | |
| CN106618486B (en) | Sleep state identification method and system in intelligent sleep assistance | |
| CN103622707A (en) | Traffic safety detecting system and method | |
| Gao et al. | An ICA/HHT hybrid approach for automatic ocular artifact correction | |
| Rahman et al. | EMG signal segmentation to predict driver’s vigilance state | |
| CN106580320A (en) | EEG signal feature analysis method and system based on visual inducement | |
| Tian et al. | A review of fatigue driving detection based on electrooculogram |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120704 |