CN103462618A - Automobile driver fatigue detecting method based on steering wheel angle features - Google Patents
Automobile driver fatigue detecting method based on steering wheel angle features Download PDFInfo
- Publication number
- CN103462618A CN103462618A CN2013103961025A CN201310396102A CN103462618A CN 103462618 A CN103462618 A CN 103462618A CN 2013103961025 A CN2013103961025 A CN 2013103961025A CN 201310396102 A CN201310396102 A CN 201310396102A CN 103462618 A CN103462618 A CN 103462618A
- Authority
- CN
- China
- Prior art keywords
- steering wheel
- fatigue
- wheel angle
- data
- driving
- 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
Abstract
The invention discloses an automobile driver fatigue detecting method based on steering wheel angle features. The method is characterized by including: reading steering wheel angle data during traveling of an automobile through a steering wheel steering parameter reading instrument; performing feature vector extraction and normalization processing on the read data; building a fatigue driving model which is based on a support vector machine and optimized by a genetic algorithm and an Adaboost algorithm; developing fatigue driving software; inputting the steering wheel angle data into the fatigue driving software to detect fatigue state of a driver in real time. By the method, direct or indirect physical and psychological influence of a detecting system on the driver is eliminated, reliability and accuracy of fatigue driving state detection are increased, and false alarm rate is lowered.
Description
Technical field
The present invention relates to car steering wheel corner technical field of data processing, particularly relate to a kind of fatigue of automobile driver detection method based on the steering wheel angle characteristic.
Background technology
Driver tired driving is one of major reason that causes road traffic accident to occur always.Statistics shows, the road traffic accident caused because of fatigue driving accounts for 20% of sum, accounts for more than 40% of traffic accident.Therefore, whether detect in real time the driver in fatigue driving state, to reducing the generation of road traffic accident, improving the traffic safety level has very important meaning.
At present, generally the objective detection method of fatigue driving is summarized as to three classes: the driving fatigue detection method based on physiological signal, the driving fatigue detection method based on the physiological reaction feature and the driving fatigue detection method based on vehicle parameter.Driving fatigue detection method based on physiological signal mainly contains following two aspects: the 1) detection method based on EEG signals; 2) detection method based on electrocardiosignal.Driver Fatigue Detection based on the physiological reaction feature mainly contains the following aspects: the 1) detection method based on the monitoring head position; 2) detection method based on the monitoring eye state; 3) detection method based on the monitoring pupil diameter; 4) detection method based on monitoring mouth state.Driving fatigue detection method based on vehicle parameter mainly contains following two aspects: the 1) detection method based on the monitoring steering-wheel movement; 2) detection method based on the displacement of monitor vehicle side.
Research shows: in driving procedure, when driving fatigue occurs in the driver, reaction meeting dull, the action meeting of operation steering wheel is slack-off, the process that generally there will be long-time inoperation steering wheel and significantly revise suddenly, thereby the steering wheel operation changes and driving fatigue exists certain contacting, can detect well driver's driving fatigue state by the motion of monitoring steering wheel.
Although the driving fatigue detection method based on physiological signal can detect driving fatigue state, but this method need to be attached to electrode or other devices driver's specific position, can affect to driver's health and psychology, thereby bring more interference to driver's operating and controlling vehicle.Although the driving fatigue detection method based on the physiological reaction feature can realize real-time, non-contacting detection, the industrial equipments such as thermal camera commonly used, head position sensor can bring certain psychological burden to the driver.These bad mental emotion or burden all can be brought certain influence to driver's normal driving.In addition, during each driver fatigue, the physiological reaction feature difference is larger, and during actual the detection, rate of false alarm is relatively high.Yet, driving fatigue detection method based on vehicle parameter can not have any impact to driver's health and psychology, and the variation of vehicle parameter is substantially similar during each driver fatigue, particularly the variation of steering wheel angle parameter is more similar, and while therefore the method being detected for reality, rate of false alarm is relatively low.
In sum, the method that existing disclosed driver fatigue detects, all exist the driving direct or indirect to the driver to disturb, and rate of false alarm is relatively high while using in real time.Therefore, need at present a kind of to the driver without any disturbing and the lower driving fatigue detection method of rate of false alarm.
Summary of the invention
The object of the present invention is to provide a kind of fatigue of automobile driver detection method based on the steering wheel angle characteristic, utilize car steering wheel corner data processing technique to carry out the driving fatigue state judgement to the driver, and the prompting that gives the alarm in time when fatigue state occurs.Described method can realize effectively to the driver fatigue detecting without any interference, can greatly reduce the road traffic accident caused because of fatigue driving.
For achieving the above object, the present invention adopts following technical scheme:
The present invention proposes a kind of fatigue of automobile driver detection method based on the steering wheel angle characteristic, the method comprises:
Step 1: reading of steering wheel angle data, read instrument by the steering wheel turn around parameters and read the steering wheel angle data in the vehicle operating process.
Step 2: the processing of steering wheel angle data, the steering wheel angle data that obtain are carried out to characteristic vector extraction and normalized.
Step 3: set up the driving fatigue detection model based on support vector machine after genetic algorithm and Adaboost algorithm optimization.
Step 4: driving fatigue detects exploitation and the application of software, utilizes MATLAB GUI to develop driving fatigue and detects software.
Step 5: real-time detection and the prompting of driving fatigue state.Detect software by the steering wheel angle data being imported into to driving fatigue, detect in real time driver's fatigue state, and the prompting that in good time gives the alarm.
Wherein, the processing of the data of steering wheel angle described in step 2 comprises data is carried out to characteristic vector extraction and normalized.Utilize the autoregression model Auto-regression model (AR model) set up to carry out feature analysis, extraction to the data of obtaining in step 1, to find the inherent law between steering wheel angle characteristic and driving fatigue; And using the AR model parameter as the characteristic vector of steering wheel angle, using this as the pattern vector based on steering wheel angle Characteristics Detection driving fatigue.Then, the characteristic vector data that adopts the minimax method to obtain extraction carries out [1,1] interval normalized, to eliminate the difference of the order of magnitude between each dimension data.
In step 2, the AR model of steering wheel angle under normal condition is:
Wherein the residual error variance is
,
In step 2, the AR model of steering wheel angle under accurate fatigue state is:
Wherein the residual error variance is
,
In step 2, the AR model of steering wheel angle under fatigue state is:
Wherein the residual error variance is
,
.
Pattern vector based on steering wheel angle Characteristics Detection driving fatigue in step 2 specifically is expressed as:
Wherein:
be
the characteristic vector of individual signal;
be
the state of kind characteristic of correspondence is to flow control
individual parameter.
The statement formula that in step 2, the minimax method is carried out [1,1] interval normalized to characteristic vector data is:
Wherein:
for data sequence;
for the data sequence after normalization;
for the minimum number in data sequence;
for the maximum number in data sequence.
Wherein, set up the driving fatigue detection model based on support vector machine after genetic algorithm and Adaboost algorithm optimization described in step 3, its step is as follows:
The first step: training being chosen and processing with the test sample book data
This step is mainly obtained steering wheel angle equal samples data in the vehicle operating process by the steering wheel angle tester be arranged on laboratory vehicle, the experiment vehicle is automotive safety key lab of Jiangsu University experiment car, and what the steering wheel angle tester adopted is the ZX-2 type turn around parameters tester of Zibo AudioCodes electronics corporation.Travel is mainly incity, Zhenjiang main trunk road, and for fear of evening peak morning, running time is chosen as 9 o'clock to the 12 o'clock morning and 13 o'clock to 16 o'clock afternoon (midfeather 1 hour).Driver's degree of fatigue is divided into to three kinds of states, is respectively abnormal driving state, accurate fatigue driving state and fatigue driving state.This step, in conjunction with general type and the algorithm of support vector machine of detection model, is numbered or quantizes above-mentioned three kinds of states, and abnormal driving state is defined as to 1, and accurate fatigue state is defined as 2, and fatigue state is defined as 3.
By preceding step 2 data processing methods, the steering wheel angle data under above-mentioned three kinds of states are respectively got to 60 groups and carry out feature extraction and processing, the status number of the characteristic vector after processing and correspondence is as the sample of driving fatigue detection model, amount to 180 groups, wherein choose at random 120 groups of training samples, 60 groups of test sample books.
By step 2 data processing method, the steering wheel angle data under above-mentioned three kinds of states are carried out obtaining after characteristic vector extraction and processing, obtain the steering wheel angle time domain waveform figure under tri-kinds of states of Fig. 7, Fig. 8 and Fig. 9.
Second step: the selection of support vector machine kernel function
Selection of kernel function gaussian radial basis function in this step, concrete functional form is:
Wherein:
for sample data,
for parameter
The 3rd step: utilize training sample to be trained support vector machine, the driving fatigue detection model after being trained (SVM model)
This step is selected support vector machine Support Vector Machine(SVM) set up concrete driving fatigue detection model (SVM model).The LIBSVM workbox based on the MATLAB language version that the support vector machine workbox adopted is Taiwan professor Lin Zhiren exploitation, mainly apply the multi-category support vector machines one to one in this workbox.Best punishment parameter in support vector machine
with best kernel functional parameter
by cross validation method K-fold Cross Validation(K-CV) choose.Utilize the sample data in step 2 to be trained support vector machine, thus the driving fatigue detection model after being trained (SVM model).Set up the overall flow figure of driving fatigue detection model as shown in Figure 2.
With the driving fatigue detection model (SVM model) after training, test sample book is detected, testing result is in Table 1.
The testing result of table 1 SVM model to driving fatigue under each state
| ? | Abnormal driving state | Accurate fatigue is driven state | Fatigue driving state | Driving fatigue state |
| Number of |
20 | 20 | 20 | 60 |
| Detect number | 17 | 19 | 16 | 52 |
| Do not detect |
3 | 1 | 4 | 8 |
| Recall rate | 85.00% | 95.00% | 80.00% | 86.67% |
As shown in Table 1, detection model is 85.00% to recall rate under abnormal driving state, and aiming at the fatigue driving state recall rate is 95.00%, to the recall rate of fatigue driving state, is 80.00%, and to driving fatigue state, total recall rate is 86.67%.This shows, the driving fatigue detection model that this step is set up, can reflected well steering wheel angle characteristic and driving fatigue between relation, also can detect preferably driver's driving fatigue state.
The 4th step: the optimization (GA-SVM model) based on genetic algorithm to SVM driving fatigue detection model
This step is mainly the Algorithm(GA with genetic algorithm Genetic) the Support Vector Machines Optimized parameter, set up the driving fatigue detection model (GA-SVM model) based on GA-SVM.The algorithm flow of genetic algorithm optimization support vector machine parameter as shown in Figure 3.
The design of genetic algorithm in this step, the GAs Toolbox that adopts Sheffield,England university to develop, wherein:
1. initial population is set and fitness calculating.In this step, the coded system of genetic algorithm adopts binary coding, initializes population and adopts in workbox
function, the form of specifically calling of this function is as follows:
, wherein:
for initializing rear population;
quantity for population at individual;
for individual length; Fitness value in genetic algorithm adopts the classification accuracy of grader under the CV meaning.
2. select the operator design.The selection operator design of genetic algorithm adopts in workbox
function completes, this function to call form as follows:
, wherein:
for the population after selection operation;
be a character string, comprise a rudimentary choice function name, as
or
;
for comprising population
the fitness value of middle individuality;
for generation gap, shown the degree that the previous generation population is replicated.
3. crossover operator design.Intersection in genetic algorithm is calculated design and is adopted in workbox
function, this function to call form as follows:
, wherein:
for the population after interlace operation;
be a character string, comprise a rudimentary intersection function name, as
or
;
for crossover probability.
4. mutation operator design.Mutation operator design in genetic algorithm adopts in workbox
function, this function to call form as follows:
, wherein:
for the population after mutation operation;
for preoperative population;
for the variation probability.
5. operational factor determines.Determine the final value of genetic algorithm relevant parameter in this step, be respectively Population Size
be 20, chromosome length
be 20, stop algebraically
be 100, generation gap
be 0.9, crossover probability
be 0.7, the variation probability
be 0.035.
Realize support vector machine punishment parameter by genetic algorithm
and kernel functional parameter
optimization, finally set up the driving fatigue detection model (GA-SVM model) based on GA-SVM.Driving fatigue detection model based on GA-SVM (GA-SVM model) is detected test sample book, and testing result is in Table 2.
The testing result of table 2 GA-SVM model to driving fatigue under each state
| ? | Abnormal driving state | Accurate fatigue is driven state | Fatigue driving state | Driving fatigue state |
| Number of |
20 | 20 | 20 | 60 |
| Detect number | 17 | 20 | 17 | 54 |
| Do not detect |
3 | 0 | 3 | 6 |
| Recall rate | 85.00% | 100.00% | 85.00% | 90.00% |
As shown in Table 2, the GA-SVM model is 85.00% to recall rate under abnormal driving state, and aiming at the fatigue driving state recall rate is 100.00%, to the recall rate of fatigue driving state, is 85.00%, and to driving fatigue state, total recall rate is 90.00%.By the data in table 2 and table 1 Data Comparison, can find that the GA-SVM model has improved 3.33% to the recall rate of driving fatigue state, this shows that the GA-SVM model after genetic algorithm optimization more can detect driver's driving fatigue state exactly.
The 5th step: the further optimization based on the Adaboost algorithm to GA-SVM driving fatigue detection model
This step is mainly using the GA-SVM model as Weak Classifier, adopt the strong classifier of Adaboost algorithm design based on a plurality of Weak Classifiers, to reach the GA-SVM model of further optimization, thereby obtain final driving fatigue detection model (GA-SVM-Adaboost model).
SVM-Adaboost strong classifier basic thought, be using GA-SVM as Weak Classifier, the output of repetition training GA-SVM test sample book, and the thought by the Adaboost algorithm obtains the strong classifier that a plurality of GA-SVM Weak Classifiers form.The algorithm flow of SVM-Adaboost strong classifier as shown in Figure 4.Driving fatigue detection model based on GA-SVM-Adaboost (GA-SVM-Adaboost model) is detected test sample book, and testing result is in Table 3.
The testing result of table 3 GA-SVM-Adaboost model to driving fatigue under each state
| ? | Abnormal driving state | Accurate fatigue is driven state | Fatigue driving state | Driving fatigue state |
| Number of |
20 | 20 | 20 | 60 |
| Detect number | 18 | 20 | 17 | 55 |
| Do not detect |
2 | 0 | 3 | 5 |
| Recall rate | 90.00% | 100.00% | 85.00% | 91.67% |
As shown in Table 3, the GA-SVM-Adaboost model is 90.00% to recall rate under abnormal driving state, and aiming at the fatigue driving state recall rate is 100.00%, to the recall rate of fatigue driving state, is 85.00%, and to driving fatigue state, total recall rate is 91.67%.By the data in table 3 and table 2 Data Comparison, can find that the GA-SVM-Adaboost model has improved 1.67% to the recall rate of driving fatigue state, this shows that the precision of model has obtained further improving, and through the strengthened model of Adaboost, more can further detect exactly driver's driving fatigue state.
Wherein, driving fatigue described in step 4 detects exploitation and the application of software, mainly for the driving fatigue detection method based on the steering wheel angle characteristic is applied to the practice among, according to the data processing method in step 2 and the GA-SVM-Adaboost model in step 3, and utilize MATLAB GUI to develop driving fatigue and detect software.The interface of this software as shown in Figure 5, specifically comprises the functional keys module, and operational factor arranges module, real-time information prompting module, the content such as display module, image display module as a result.
The application process that this driving fatigue detects software is as follows: at first by step 1, obtain steering wheel angle related data (in the domestic vehicle, can read the corner related data by the steering wheel sensor is installed), then the data of obtaining are passed in " being written into data " functional module of this detection software in real time, then move the fatigue state that measuring ability can detect the driver in real time.This driving fatigue detects the use of software on vehicle, can be installed in automobile navigation DVD by prior embedding, can conveniently use with cost-saving like this.
Wherein, real-time detection and the prompting of driving fatigue state described in step 5, be mainly that the steering wheel angle related data will obtained in step 1 is passed in the driving fatigue detection software in step 4 in real time, can detect in real time driver's fatigue state.When detecting the driver for fatigue state, will give the alarm in real time and remind the driver to note.
Beneficial effect of the present invention is:
1, built the method frame that the fatigue of automobile driver based on steering wheel angle characteristic mode detects, for the steering wheel angle specificity analysis of multitude of different ways is laid a good foundation;
2, the accuracy rate detected in order to improve model, used genetic algorithm and Adaboost algorithm to be optimized detection model;
3, stop the direct or indirect impact of detection system on driver's health and psychology, improved reliability and accuracy that driving fatigue state detects, reduced rate of false alarm.
The accompanying drawing explanation
Fig. 1 is driver fatigue testing process schematic diagram in the present invention.
Fig. 2 is the overall flow figure that sets up the driving fatigue detection model.
The algorithm flow chart that Fig. 3 is genetic algorithm optimization support vector machine parameter.
The algorithm flow chart that Fig. 4 is the SVM-Adaboost strong classifier.
Fig. 5 is the surface chart that driving fatigue detects software.
Fig. 6 is driving fatigue testing result figure in embodiment.
Fig. 7 is the steering wheel angle time domain waveform figure under normal condition.
Fig. 8 steering wheel angle time domain waveform figure under fatigue state that is as the criterion.
Fig. 9 is the steering wheel angle time domain waveform figure under fatigue state.
The specific embodiment
Fig. 1 is driver fatigue testing process schematic diagram in the present invention, and as shown in Figure 1, Driver Fatigue Detection comprises:
Step 1: reading of steering wheel angle data, read instrument by the steering wheel turn around parameters and read the steering wheel angle data.
Step 2: the processing of steering wheel angle data, the steering wheel angle data that obtain are carried out to characteristic vector extraction and normalized.
Step 3: set up the driving fatigue detection model based on support vector machine after genetic algorithm and Adaboost algorithm optimization.
Step 4: driving fatigue detects exploitation and the application of software.Utilize MATLAB GUI to develop driving fatigue and detect software.
Step 5: real-time detection and the prompting of driving fatigue state.Detect software by the steering wheel angle data being imported into to driving fatigue, can detect in real time driver's fatigue state.
According to the flow process in Fig. 1, the driving fatigue that the steering wheel angle related data obtained in step 1 is passed in step 4 in real time detects in software, can detect in real time driver's fatigue state.When detecting the driver for fatigue state, will give the alarm in real time and remind the driver to note.In step 2 in flow chart of data processing and method, step 3 Establishment and optimization of detection model elaborated in front, the present embodiment is no longer set forth.The driving fatigue that the present embodiment mainly imports the data in step 1 in step 4 detects in software, feasibility, practicality and the accuracy of checking the method.
In the present embodiment, the data of step 1 read with the ZX-2 type turn around parameters tester of car by the experiment of automotive safety key lab of Jiangsu University.This has read 60 groups of data altogether, and the present embodiment only has been listed as 6 groups of representative data, as shown in table 4.
Table 4 part steering wheel angle data
The driving fatigue that 60 groups of steering wheel angle data obtaining in step 1 are imported in step 4 detects in software, show that testing result is is 93.00% to recall rate under abnormal driving state, aiming at the fatigue driving state recall rate is 100.00%, recall rate to fatigue driving state is 95.00%, and to driving fatigue state, total recall rate is that 96.00%(testing result figure is shown in Fig. 6).This testing result tables of data Benq is feasible and practicality in the fatigue of automobile driver detection method of steering wheel angle characteristic, and to the driver, without any interference, and the accuracy rate detected is higher, and rate of false alarm is relatively low.
The above; be only preferred embodiment of the present invention, be not intended to limit protection scope of the present invention, be to be understood that; the present invention is not limited to implementation as described herein, and the purpose that these implementations are described is to help those of skill in the art to put into practice the present invention.Any those of skill in the art are easy to be further improved without departing from the spirit and scope of the present invention and perfect, therefore the present invention only is subject to the restriction of content and the scope of the claims in the present invention, and its intention contains all alternative and equivalents that are included in the spirit and scope of the invention limited by claims.
Claims (6)
1. the fatigue of automobile driver detection method based on the steering wheel angle characteristic is characterized in that comprising the following steps:
Step 1: reading of steering wheel angle data, read instrument by the steering wheel turn around parameters and read steering wheel angle data in the vehicle operating process;
Step 2: the processing of steering wheel angle data, the steering wheel angle data that obtain are carried out to characteristic vector extraction and normalized;
Step 3: set up the driving fatigue detection model based on support vector machine after genetic algorithm and Adaboost algorithm optimization;
Step 4: driving fatigue detects exploitation and the application of software;
Step 5: real-time detection and the prompting of driving fatigue state, detect software by the steering wheel angle data being imported into to driving fatigue, can detect in real time driver's fatigue state, and the prompting that in good time gives the alarm.
2. a kind of fatigue of automobile driver detection method based on the steering wheel angle characteristic according to claim 1 is characterized in that: in described step 2, the processing of steering wheel angle data comprises data is carried out to characteristic vector extraction and normalized; The autoregression model Auto-regression model that the data utilization of obtaining in step 1 is set up carries out feature analysis and extraction to it, to find the inherent law between steering wheel angle characteristic and driving fatigue; Using Auto-regression model model parameter as the characteristic vector of steering wheel angle, using this as the pattern vector based on steering wheel angle Characteristics Detection driving fatigue; Then adopt the characteristic vector data that the minimax method obtains extraction to carry out [1,1] interval normalized, to eliminate the difference of the order of magnitude between each dimension data.
3. a kind of fatigue of automobile driver detection method based on the steering wheel angle characteristic according to claim 1 is characterized in that: the process of setting up the driving fatigue detection model based on support vector machine after genetic algorithm and Adaboost algorithm optimization in described step 3 is:
The first step: training being chosen and processing with the test sample book data;
Second step: the selection of support vector machine kernel function;
The 3rd step: utilize training sample to be trained support vector machine, the driving fatigue detection model after being trained, i.e. SVM model;
The 4th step: the optimization based on genetic algorithm to SVM driving fatigue detection model, i.e. GA-SVM model;
The 5th step: the further optimization based on the Adaboost algorithm to GA-SVM driving fatigue detection model.
4. a kind of fatigue of automobile driver detection method based on the steering wheel angle characteristic according to claim 1, it is characterized in that: in described step 4, the exploitation of driving fatigue detection software is according to the data processing method in step 2 and the GA-SVM-Adaboost model in step 3 with application, and the driving fatigue that utilizes MATLAB GUI to develop detects software.
5. a kind of fatigue of automobile driver detection method based on the steering wheel angle characteristic according to claim 4, it is characterized in that: the interface that described driving fatigue detects software specifically comprises the functional keys module, and operational factor arranges module, real-time information prompting module, display module and image display module as a result.
6. a kind of fatigue of automobile driver detection method based on the steering wheel angle characteristic according to claim 1, it is characterized in that: in described step 5, the real-time detection of driving fatigue state is that the steering wheel angle related data will obtained in step 1 is passed in the driving fatigue detection software in step 4 in real time with reminding, detect in real time driver's fatigue state, when detecting the driver for fatigue state, the real-time information prompting module gives the alarm in real time and reminds the driver to note.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013103961025A CN103462618A (en) | 2013-09-04 | 2013-09-04 | Automobile driver fatigue detecting method based on steering wheel angle features |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013103961025A CN103462618A (en) | 2013-09-04 | 2013-09-04 | Automobile driver fatigue detecting method based on steering wheel angle features |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103462618A true CN103462618A (en) | 2013-12-25 |
Family
ID=49787835
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013103961025A Pending CN103462618A (en) | 2013-09-04 | 2013-09-04 | Automobile driver fatigue detecting method based on steering wheel angle features |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103462618A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103824421A (en) * | 2014-03-26 | 2014-05-28 | 上海长安汽车工程技术有限公司 | System and method for detecting fatigue driving and alarming to ensure active safety |
| CN106448059A (en) * | 2016-06-06 | 2017-02-22 | 清华大学 | Wrist strap instrument based driver fatigue detection method |
| CN106446812A (en) * | 2016-09-13 | 2017-02-22 | 西安科技大学 | Driving state recognition method based on approximate entropy template matching |
| CN106650636A (en) * | 2016-11-30 | 2017-05-10 | 同济大学 | Brain machine interface-based device and method for monitoring driving vigilance in real time |
| CN107198526A (en) * | 2016-06-20 | 2017-09-26 | 山东理工大学 | A kind of method and system for aiding in driving |
| CN108596409A (en) * | 2018-07-16 | 2018-09-28 | 江苏智通交通科技有限公司 | The method for promoting traffic hazard personnel's accident risk prediction precision |
| CN109243006A (en) * | 2018-08-24 | 2019-01-18 | 深圳市国脉畅行科技股份有限公司 | Abnormal driving Activity recognition method, apparatus, computer equipment and storage medium |
| CN112308136A (en) * | 2020-10-29 | 2021-02-02 | 江苏大学 | SVM-Adaboost-based driving distraction detection method |
| CN114343638A (en) * | 2022-01-05 | 2022-04-15 | 河北体育学院 | Fatigue degree evaluation method and system based on multi-modal physiological parameter signals |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005143896A (en) * | 2003-11-17 | 2005-06-09 | Nissan Motor Co Ltd | Device for determining psychological state of driver |
| CN1873722A (en) * | 2006-04-07 | 2006-12-06 | 中山大学 | Safety caution system for driving automobile |
| US20070084661A1 (en) * | 2002-09-04 | 2007-04-19 | Volkswagen Aktiengesellschaft | Method and device for recognizing the level of awareness of a vehicle driver |
| JP2008079939A (en) * | 2006-09-28 | 2008-04-10 | Sanyo Electric Co Ltd | Physiological data extraction device, switch device and physiological state warning device |
| JP2010115227A (en) * | 2008-11-11 | 2010-05-27 | Sensor:Kk | Fatigue degree measuring and alarming apparatus for automobile |
| JP5018575B2 (en) * | 2008-03-11 | 2012-09-05 | トヨタ自動車株式会社 | Vehicle control device |
| CN202448833U (en) * | 2011-12-20 | 2012-09-26 | 长安大学 | Fatigue driving early warning device |
-
2013
- 2013-09-04 CN CN2013103961025A patent/CN103462618A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070084661A1 (en) * | 2002-09-04 | 2007-04-19 | Volkswagen Aktiengesellschaft | Method and device for recognizing the level of awareness of a vehicle driver |
| JP2005143896A (en) * | 2003-11-17 | 2005-06-09 | Nissan Motor Co Ltd | Device for determining psychological state of driver |
| CN1873722A (en) * | 2006-04-07 | 2006-12-06 | 中山大学 | Safety caution system for driving automobile |
| JP2008079939A (en) * | 2006-09-28 | 2008-04-10 | Sanyo Electric Co Ltd | Physiological data extraction device, switch device and physiological state warning device |
| JP5018575B2 (en) * | 2008-03-11 | 2012-09-05 | トヨタ自動車株式会社 | Vehicle control device |
| JP2010115227A (en) * | 2008-11-11 | 2010-05-27 | Sensor:Kk | Fatigue degree measuring and alarming apparatus for automobile |
| CN202448833U (en) * | 2011-12-20 | 2012-09-26 | 长安大学 | Fatigue driving early warning device |
Non-Patent Citations (5)
| Title |
|---|
| YANCHAO DONG 等: "Driver Inattention Monitoring System for Intelligent Vehicles: A Review", 《IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS》, vol. 12, no. 2, 30 June 2011 (2011-06-30), pages 596 - 615, XP 011325846, DOI: doi:10.1109/TITS.2010.2092770 * |
| 张利: "基于转向盘转角特性的驾驶疲劳检测方法研究", 《江苏大学硕士学位论文》, 3 September 2012 (2012-09-03) * |
| 谷也: "基于方向盘转角信号的驾驶员疲劳监测装置研制", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, no. 1, 31 December 2011 (2011-12-31) * |
| 魏武 等: "基于遗传算法的改进AdaBoost算法在汽车识别中的应用", 《公路交通科技》, vol. 27, no. 2, 28 February 2010 (2010-02-28) * |
| 鲁松 等: "驾驶员疲劳状态检测仿真研究", 《计算机仿真》, vol. 29, no. 11, 30 November 2012 (2012-11-30), pages 378 - 381 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103824421A (en) * | 2014-03-26 | 2014-05-28 | 上海长安汽车工程技术有限公司 | System and method for detecting fatigue driving and alarming to ensure active safety |
| CN106448059A (en) * | 2016-06-06 | 2017-02-22 | 清华大学 | Wrist strap instrument based driver fatigue detection method |
| CN107198526A (en) * | 2016-06-20 | 2017-09-26 | 山东理工大学 | A kind of method and system for aiding in driving |
| CN106446812A (en) * | 2016-09-13 | 2017-02-22 | 西安科技大学 | Driving state recognition method based on approximate entropy template matching |
| CN106446812B (en) * | 2016-09-13 | 2017-09-26 | 西安科技大学 | Driving condition discrimination method based on approximate entropy template matches |
| CN106650636A (en) * | 2016-11-30 | 2017-05-10 | 同济大学 | Brain machine interface-based device and method for monitoring driving vigilance in real time |
| CN108596409A (en) * | 2018-07-16 | 2018-09-28 | 江苏智通交通科技有限公司 | The method for promoting traffic hazard personnel's accident risk prediction precision |
| CN108596409B (en) * | 2018-07-16 | 2021-07-20 | 江苏智通交通科技有限公司 | Method for improving accident risk prediction precision of traffic hazard personnel |
| CN109243006A (en) * | 2018-08-24 | 2019-01-18 | 深圳市国脉畅行科技股份有限公司 | Abnormal driving Activity recognition method, apparatus, computer equipment and storage medium |
| CN112308136A (en) * | 2020-10-29 | 2021-02-02 | 江苏大学 | SVM-Adaboost-based driving distraction detection method |
| CN114343638A (en) * | 2022-01-05 | 2022-04-15 | 河北体育学院 | Fatigue degree evaluation method and system based on multi-modal physiological parameter signals |
| CN114343638B (en) * | 2022-01-05 | 2023-08-22 | 河北体育学院 | Fatigue degree assessment method and system based on multi-mode physiological parameter signals |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103462618A (en) | Automobile driver fatigue detecting method based on steering wheel angle features | |
| Jin et al. | Driver cognitive distraction detection using driving performance measures | |
| Hu et al. | Abnormal driving detection based on normalized driving behavior | |
| US20200039525A1 (en) | Method for detecting safety of driving behavior, apparatus, device and storage medium | |
| Liang et al. | Real-time detection of driver cognitive distraction using support vector machines | |
| Doshi et al. | On-road prediction of driver's intent with multimodal sensory cues | |
| CN110765807B (en) | Driving behavior analysis and processing method, device, equipment and storage medium | |
| CN110575163B (en) | Method and device for detecting driver distraction | |
| CN108259494A (en) | A kind of network attack detecting method and device | |
| Liao et al. | Detection of driver cognitive distraction: An SVM based real-time algorithm and its comparison study in typical driving scenarios | |
| CN103996287A (en) | Vehicle forced lane changing decision-making method based on decision-making tree model | |
| Koh et al. | Smartphone-based modeling and detection of aggressiveness reactions in senior drivers | |
| Lv et al. | Qualitative action recognition by wireless radio signals in human–machine systems | |
| Shirazi et al. | Detection of intoxicated drivers using online system identification of steering behavior | |
| CN108248610A (en) | A kind of monitoring of adaptive driving behavior of diverting one's attention and tampering devic | |
| CN105539026A (en) | Tire pressure detection system and method | |
| Ouyang et al. | An ensemble learning-based vehicle steering detector using smartphones | |
| CN108769104A (en) | A kind of road condition analyzing method for early warning based on onboard diagnostic system data | |
| Yin et al. | Driver danger-level monitoring system using multi-sourced big driving data | |
| CN106408032A (en) | Fatigue driving detection method based on corner of steering wheel | |
| Kanaan et al. | Using naturalistic vehicle-based data to predict distraction and environmental demand | |
| Ouyang et al. | Multiwave: A novel vehicle steering pattern detection method based on smartphones | |
| Wang et al. | Smartphone sensors-based abnormal driving behaviors detection: Serial-feature network | |
| KR20150031051A (en) | Apparatus for judging driver inattention and method thereof | |
| CN111717210B (en) | Detection method for separation of driver from steering wheel in relative static state of hands |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20131225 |