CN108764169A - A kind of driver's Emotion identification based on machine learning and display device and method - Google Patents
A kind of driver's Emotion identification based on machine learning and display device and method Download PDFInfo
- Publication number
- CN108764169A CN108764169A CN201810544319.9A CN201810544319A CN108764169A CN 108764169 A CN108764169 A CN 108764169A CN 201810544319 A CN201810544319 A CN 201810544319A CN 108764169 A CN108764169 A CN 108764169A
- Authority
- CN
- China
- Prior art keywords
- face
- model
- driver
- emotion identification
- image
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
- G06V20/597—Recognising the driver's state or behaviour, e.g. attention or drowsiness
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
Driver's Emotion identification that the present invention relates to a kind of based on machine learning and display device and method, described device includes face detection system, Emotion identification system and mood display system；The face detection system is connected to Emotion identification system, the face detection system is used to obtain the image information of driver's face, and processing is carried out to the image information of acquisition and obtains facial image model, and facial image model is sent to Emotion identification system by treated；The Emotion identification system is connected to mood display system, the Emotion identification system is used to extract the emotional characteristics point of facial image model, face Emotion identification in face information is obtained by Characteristic Analysis Model and mood disaggregated model as a result, and face Emotion identification result is sent to mood display system；The mood display system is used to, according to the different type of the face Emotion identification result received, show the emotional information of driver.It is also convenient for preferably controlling road conditions in driver, makes corresponding strategy.
Description
Technical field
The present invention relates to image real time transfer and machine learning fields, and in particular to a kind of driving based on machine learning
Member's Emotion identification and display device and method.
Background technology
With the arriving in big data epoch, the fields such as artificial intelligence, pattern-recognition, computer vision are also faced with huge choose
War, has thus more highlighted the importance of machine learning.Currently, using the face image data of magnanimity, in Face datection and table
Many model methods are had existed in the research of feelings identification, however are but more lacked for application of the Expression Recognition in traffic prompt
It is weary.In particular, the getting worse of the congestion phenomenon with urban transportation, and personal social pressures increase sharply, in drive the cross
Cheng Zhong, driver cannot recognize the emotional state of oneself, can not also obtain the emotional information of other drivers in time, so that handing over
Logical conflict frequently occurs.
Invention content
For this reason, it may be necessary to provide a kind of driver's Emotion identification based on machine learning and display device and method, solve existing
There is driver in driving procedure, can not recognize oneself emotional state in time, and is easy to cause what traffic conflict frequently occurred
Problem.
To achieve the above object, a kind of driver's Emotion identification based on machine learning is inventor provided to fill with display
It sets, including face detection system, Emotion identification system and mood display system；
The face detection system is connected to Emotion identification system, and the face detection system is for obtaining driver's face
Image information, and processing is carried out to the image information of acquisition and obtains facial image model, and will treated facial image mould
Type is sent to Emotion identification system；
The Emotion identification system is connected to mood display system, and the Emotion identification system is for extracting facial image mould
The emotional characteristics point of type, face Emotion identification in face information is obtained by Characteristic Analysis Model and mood disaggregated model as a result,
And face Emotion identification result is sent to mood display system；
The mood display system is used for the different type according to the face Emotion identification result received, shows driver
Emotional information.
It advanced optimizes, the face identification system includes image collection module, Face detection module and image preprocessing
Module；
Facial expression information of the described image acquisition module for obtaining driver in real time, and convert thereof into facial image
Information is sent to Face detection module；
The Face detection module is rejected and is made an uproar for detecting human face region to be treated in human face image information
Acoustic jamming, by treated, human face image information is sent to image pre-processing module；
Described image preprocessing module is used for by gathering method for normalizing and gray scale normalization method to treated people
Face image information is handled, and the facial image model of facial size and human face light information unification is obtained.
It advanced optimizes, the Emotion identification system includes face database, face emotional characteristics extraction module, people
Face emotional characteristics selecting module and face emotional characteristics sort module；
The face database is used to store the Characteristic Analysis Model and mood disaggregated model of human face image information；
The face emotional characteristics extraction module be used for by Gabor filtering methods to the facial image model that receives into
Row characteristic vector pickup obtains feature vector model, and the feature vector model of acquisition is sent to the selection of face emotional characteristics
Module；
The face emotional characteristics selecting module is used to carry out dimensionality reduction to the feature vector model extracted by PCA algorithms
Processing, deletes the redundancy data that obtain that treated, and will treated that data are sent to face emotional characteristics sort module；
The face classification modules are used for according to trained Characteristic Analysis Model and mood disaggregated model to receiving
To data classify, obtain current mood generic title, and item name is sent to mood display system.
It advanced optimizes, the Emotion identification system further includes image tranining database and image dynamic data base；
Described image tranining database is used to preserve for face emotional characteristics sort module training characteristics analysis model
Training image；
Described image dynamic data base is used to record the human face image information of the driver in certain time, and preserves the people
The classification results of face image information, and by the classification results of preservation in set time more new images tranining database, to face figure
As the Characteristic Analysis Model and mood disaggregated model of information optimize.
It advanced optimizes, the face Emotion identification knot that the mood display system is used to be sent according to Emotion identification system
Fruit shows mood classification and mood degree on output display screen.
Inventor additionally provides another technical solution：A kind of driver's Emotion identification based on machine learning and display side
Method includes the following steps：
The human face image information of driver is obtained, and processing is carried out to the human face image information of acquisition and obtains facial image mould
Type；
The emotional characteristics point for extracting facial image model obtains face letter by Feature Selection Model and mood disaggregated model
Face Emotion identification result in breath；
According to the different type of face Emotion identification result, the emotional information of driver is shown.
Advanced optimize, it is described " obtain driver human face image information, and to the human face image information of acquisition at
Reason obtains facial image model " it specifically includes：
The facial expression information of driver is obtained in real time, and is converted into human face image information；
Human face region to be treated in human face image information is detected, while cancelling noise interferes；
By set method for normalizing and gray scale normalization method, to treated, human face image information is handled, and is obtained
The facial image model of facial size and human face light information unification.
It advanced optimizes, it is described " the facial image model of extraction to be passed through into the feature extraction mould in face database
Type and mood disaggregated model obtain face Emotion identification result in face information " it specifically includes：
Characteristic vector pickup is carried out to the facial image model received by Gabor filtering methods, obtains feature vector
Model；
Dimension-reduction treatment is carried out to the feature vector model extracted by PCA algorithms, after deletion redundancy obtains processing
Data；
According in face database trained Characteristic Analysis Model and mood disaggregated model to receiving
Data are classified, and current mood generic title is obtained；
The face database is used to store the Feature Selection Model and tagsort model of human face image information.
It advanced optimizes, the Emotion identification system further includes image tranining database and image dynamic data base；
Described image tranining database is used to preserve the training image model for face emotional characteristics analysis module；
Described image dynamic data base is used to record the human face image information of the driver in certain time, and preserves the people
The classification results of face image information, and by the classification results of preservation in set time more new images tranining database, to face figure
As the Feature Selection Model and mood disaggregated model of information optimize.
It advanced optimizes, it is described " according to the different type of face Emotion identification result, showing the emotional information of driver "
Further include specifically：
According to face Emotion identification as a result, the mood classification and mood degree of display driver.
It is different from the prior art, above-mentioned technical proposal, when the emotional information for obtaining driver by the device, and to driving
Member mood shown, the driver of driving can be reminded, be also convenient on highway other drivers and traffic
Administrative staff preferably control road conditions, and make corresponding emergency policy.
Description of the drawings
Fig. 1 is a kind of structural representation of driver's Emotion identification and display based on machine learning described in specific implementation mode
Figure；
Fig. 2 is a kind of flow signal of driver's Emotion identification and display based on machine learning described in specific implementation mode
Figure.
Specific implementation mode
For the technology contents of technical solution, construction feature, the objects and the effects are described in detail, below in conjunction with specific reality
It applies example and attached drawing is coordinated to be explained in detail.
Referring to Fig. 1, driver's Emotion identification and display device based on machine learning described in the present embodiment, including face
Detecting system, Emotion identification system and mood display system；
The face detection system is connected to Emotion identification system, and the face detection system is used to obtain the people of driver
Face image information, and processing is carried out to the human face image information of acquisition and obtains facial image model, and will treated face figure
As model is sent to Emotion identification system；
The Emotion identification system is connected to mood display system, and the Emotion identification system is for extracting facial image mould
The emotional characteristics point of type, face Emotion identification in face information is obtained by Characteristic Analysis Model and mood disaggregated model as a result,
And face Emotion identification result is sent to mood display system；
The mood display system is used for the different type according to the face Emotion identification result received, shows driver
Emotional information.
Human face detection device, Emotion identification system and mood display system form organic whole, face by cascade system
Detecting system mainly obtains the human face image information for including driver by data collection station (such as vehicle-mounted front camera), and
Being obtained by image procossing has the facial image model for the face information that can be further effectively treated, by what is obtained after processing
Facial image model is sent to Emotion identification system；Emotion identification system is carried according to obtained facial image model by feature
Modulus type with the relevant effective emotional characteristics point extraction of mood, obtained by Characteristic Analysis Model and mood disaggregated model
The recognition result of face mood in facial image model, and after the recognition result of Emotion identification system acquisition face mood, it will
The recognition result is sent to mood display system, and mood display system is according to the different type of obtained recognition result, Ke Yitong
Cross the emotional information of vehicle-mounted display screen display driver.
Specifically, the face identification system includes image collection module, Face detection module and image pre-processing module；
Facial expression information of the described image acquisition module for obtaining driver in real time, and convert thereof into facial image
Information is sent to Face detection module；
The Face detection module is rejected and is made an uproar for detecting human face region to be treated in human face image information
Acoustic jamming, by treated, human face image information is sent to image pre-processing module；
Described image preprocessing module is used for by gathering method for normalizing and gray scale normalization method to treated people
Face image information is handled, and the facial image model of facial size and human face light information unification is obtained.
The Emotion identification system includes face database, face emotional characteristics extraction module, face emotional characteristics
Selecting module and face emotional characteristics sort module；
The face database is used to store the Characteristic Analysis Model and mood disaggregated model of human face image information；
The face emotional characteristics extraction module be used for by Gabor filtering methods to the facial image model that receives into
Row characteristic vector pickup obtains feature vector model, and the feature vector model of acquisition is sent to the selection of face emotional characteristics
Module；
The face emotional characteristics selecting module is used to carry out dimensionality reduction to the feature vector model extracted by PCA algorithms
Processing, deletes the redundancy data that obtain that treated, and will treated that data are sent to face emotional characteristics sort module；
Model and mood disaggregated model are to receiving about the face classification modules are used for according to trained feature point
To data classify, obtain current mood generic title, and item name is sent to mood display system.
When driver's Emotion identification and display device are in the state of operation, such as make the dress when driver opens the device
The state in operation is set, or when driver starts automobile, automatically turns on the state that the device makes the device be in operation；
Image collection module obtains the facial expression information of driver in real time, and converts thereof into human face image information, i.e., image obtains
Module obtains the facial image of current driver's in real time, and since the driver of a vehicle is relatively fixed, image collection module can be with
The image information of driver is accurately obtained, image collection module is that image obtains terminal, such as vehicle-mounted front camera；Wherein,
Image collection module can each 5 minutes image informations for obtaining a current driver；When what image collection module obtained works as
After the image information of preceding driver, it is sent to Face detection module.
After Face detection module receives the image information of current driver's, according to image information is obtained, face is detected
Human face region to be treated in image information determines human face region to be identified, rejects background information and complex environment is dry
It disturbs, and is sent to image pre-processing module and is further processed.
Image pre-processing module is further handled the human face image information after Face detection resume module, according to system
One mode carries out geometrical normalization to human face image information and gray scale normalization is handled, and obtains facial size and human face light letter
Unified facial image model is ceased, the facial image model of acquisition is sent to face emotional characteristics extraction module.
The facial image model received is carried out emotional characteristics point by face emotional characteristics extraction module using Gabor methods
Extraction obtains the feature vector of facial image model, and is sent to face emotional characteristics selecting module by feature vector is obtained.
Face emotional characteristics selecting module carries out dimension-reduction treatment to the feature vector received by PCA algorithms, deletes superfluous
Data that remaining information obtains that treated, will treated that data are sent into face emotional characteristics sort module.
When face emotional characteristics sort module is according to trained Characteristic Analysis Model pair in face database
Treated, and data are analyzed, and analysis result is carried out acquisition current driver's by trained mood disaggregated model
Emotion identification as a result, as described in current mood item name, be then sent to mood display system and carry out result to identification
It is shown.
For example, when driver is during driving, a running for encountering front is excessively slow, causes at the driver
In angry mood, and driver and the driver of other vehicles may also be unaware that the driver is in angry mood
In, driver may aggravate its angry degree, such as the whistle of vehicle below because of some behaviors of other automobiles, and work as
When driver is in the mood of indignation, it is dangerous to be easy to cause driving.And work as the emotional information that driver is obtained by the device,
And the mood of driver is shown, wherein driver can be shown to the mood situation of driver by vehicle-carrying display screen
It is checked, vehicle back can also be shown into the mood situation for being about to driver by the way that display screen is arranged behind vehicle
The drivers of other vehicles check, driving of the display screen by front vehicles can also be set in vehicle front by being arranged
Member knows the mood situation of the driver of the vehicle, or the display screen by the way that face forward and rear are arranged in the top of vehicle
The mood situation that display driver is carried out to the vehicle of front and back, can remind the driver of driving, be also convenient in public affairs
Other drivers and traffic administration personnel of road preferably control road conditions, and make corresponding emergency policy.
In the present embodiment, in order to improve the Emotion identification speed of the facial image to newly obtaining, the Emotion identification system
System further includes image tranining database and image dynamic data base；
Described image tranining database is used to preserve for face emotional characteristics sort module training characteristics analysis model
Training image；
Described image dynamic data base is used to record the human face image information of the driver in certain time, and preserves the people
The classification results of face image information, and so that image is instructed in set time more new images tranining database the classification results of preservation
Practice database to optimize the Characteristic Analysis Model and mood disaggregated model of human face image information.
The image information for obtaining driver and analysis are obtained its generic and are stored in image dynamic data base, image by device
Dynamic data base include each time the image information after Emotion identification and Emotion identification as a result, and all images have been sent to figure
As in tranining database, making it re-establish Characteristic Analysis Model and mood disaggregated model, and empty image dynamic data simultaneously
Library；And when image training module optimized to Characteristic Analysis Model and mood disaggregated model, train number for image
According to every piece image in library, feature point extraction is carried out using Gabor methods, to obtain different feature vectors；For each
The feature vector of width image carries out dimension-reduction treatment, to obtain new feature vector using PCA methods to it；It is instructed for image
Practice all images in database, according to the classification belonging to feature vector and the image, using machine learning to the power of feature
Value is iterated training, to obtain Characteristic Analysis Model and mood disaggregated model, realizes to Characteristic Analysis Model and mood point
The optimization of class model.
In the present embodiment, in order to help driver to recognize the mood of oneself, the mood display system is for basis
The face Emotion identification that Emotion identification system is sent on output display screen as a result, show mood classification and mood degree.It is logical
Mood classification and mood degree on output display screen by driver is crossed to show, can according to the mood classification of display and
Mood degree preferably helps driver to recognize the mood of oneself, can deal in time.
Referring to Fig. 2, in another embodiment, a kind of driver's Emotion identification based on machine learning and display side
Method includes the following steps：
Step S210：The human face image information of driver is obtained, and processing acquisition is carried out to the human face image information of acquisition
Facial image model；
Step S220：The emotional characteristics point for extracting facial image model, passes through Feature Selection Model and mood disaggregated model
Obtain face Emotion identification result in face information；
Step S230：According to the different type of face Emotion identification result, the emotional information of driver is shown.
The human face image information for including driver is obtained by data collection station (such as vehicle-mounted front camera), and is passed through
Image procossing obtains the facial image model with the face information that can be further effectively treated, the face that will be obtained after processing
Iconic model is sent to Emotion identification system；Emotion identification system passes through feature extraction mould according to obtained facial image model
Type carries out, with the relevant effective emotional characteristics point extraction of mood, face being obtained by Characteristic Analysis Model and mood disaggregated model
The recognition result of face mood in iconic model, and after the recognition result of Emotion identification system acquisition face mood, by the knowledge
Other result is sent to mood display system, and mood display system can pass through vehicle according to the different type of obtained recognition result
The emotional information of the display screen display driver of load.
Specifically, described " obtain the human face image information of driver, and carry out processing to the human face image information of acquisition and obtain
Obtain facial image model " it specifically includes：
The facial expression information of driver is obtained in real time, and is converted into human face image information；
Human face region to be treated in human face image information is detected, while cancelling noise interferes；
By set method for normalizing and gray scale normalization method, to treated, human face image information is handled, and is obtained
The facial image model of facial size and human face light information unification.
It is described " the facial image model of extraction to be classified by Feature Selection Model in face database and mood
Model obtains face Emotion identification result in face information " it specifically includes：
Characteristic vector pickup is carried out to the facial image model received by Gabor filtering methods, obtains feature vector
Model；
Dimension-reduction treatment is carried out to the feature vector model extracted by PCA algorithms, after deletion redundancy obtains processing
Data；
According in face database trained Characteristic Analysis Model and mood disaggregated model to receiving
Data are classified, and current mood generic title is obtained；
The face database is used to store the Feature Selection Model and tagsort model of human face image information.
When driver's Emotion identification is in the state of operation with display device, such as make the device when driver opens the device
State in operation, or when driver starts automobile, automatically turn on the state that the device makes the device be in operation；Figure
As acquisition module obtains the facial expression information of driver in real time, and human face image information is converted thereof into, i.e., image obtains mould
Block obtains the facial image of current driver's in real time, and since the driver of a vehicle is relatively fixed, image collection module can essence
The image information of driver is obtained accurately, and image collection module is that image obtains terminal, such as vehicle-mounted front camera；Wherein, scheme
As acquisition module can each 5 minutes image informations for obtaining a current driver；When image collection module obtain it is current
After the image information of driver, it is sent to Face detection module.
After Face detection module receives the image information of current driver's, according to image information is obtained, face is detected
Human face region to be treated in image information determines human face region to be identified, rejects background information and complex environment is dry
It disturbs, and is sent to image pre-processing module and is further processed.
Image pre-processing module is further handled the human face image information after Face detection resume module, according to system
One mode carries out geometrical normalization to human face image information and gray scale normalization is handled, and obtains facial size and human face light letter
Unified facial image model is ceased, the facial image model of acquisition is sent to face emotional characteristics extraction module.
The facial image model received is carried out emotional characteristics point by face emotional characteristics extraction module using Gabor methods
Extraction obtains the feature vector of facial image model, and is sent to face emotional characteristics selecting module by feature vector is obtained.
Face emotional characteristics selecting module carries out dimension-reduction treatment to the feature vector received by PCA algorithms, deletes superfluous
Data that remaining information obtains that treated, will treated that data are sent into face emotional characteristics sort module.
When face emotional characteristics sort module is according to trained Characteristic Analysis Model pair in face database
Treated, and data are analyzed, and analysis result is carried out acquisition current driver's by trained mood disaggregated model
Emotion identification as a result, as described in current mood item name, be then sent to mood display system and carry out result to identification
It is shown.
For example, when driver is during driving, a running for encountering front is excessively slow, causes at the driver
In angry mood, and driver and the driver of other vehicles may also be unaware that the driver is in angry mood
In, driver may aggravate its angry degree, such as the whistle of vehicle below because of some behaviors of other automobiles, and work as
When driver is in the mood of indignation, it is dangerous to be easy to cause driving.And work as the emotional information that driver is obtained by the device,
And the mood of driver is shown, wherein driver can be shown to the mood situation of driver by vehicle-carrying display screen
It is checked, vehicle back can also be shown into the mood situation for being about to driver by the way that display screen is arranged behind vehicle
The drivers of other vehicles check, driving of the display screen by front vehicles can also be set in vehicle front by being arranged
Member knows the mood situation of the driver of the vehicle, or the display screen by the way that face forward and rear are arranged in the top of vehicle
The mood situation that display driver is carried out to the vehicle of front and back, can remind the driver of driving, be also convenient in public affairs
Other drivers and traffic administration personnel of road preferably control road conditions, and make corresponding emergency policy.
In the present embodiment, in order to improve the Emotion identification speed of the facial image to newly obtaining, the Emotion identification system
System further includes image tranining database and image dynamic data base；
Described image tranining database is used to preserve the training image model for face emotional characteristics analysis module；
Described image dynamic data base is used to record the human face image information of the driver in certain time, and preserves the people
The classification results of face image information, and by the classification results of preservation in set time more new images tranining database, to face figure
As the Feature Selection Model and mood disaggregated model of information optimize.
Image dynamic data base include each time the image information after Emotion identification and Emotion identification as a result, and will be all
Image has been sent into image tranining database, so that it is re-established Characteristic Analysis Model and mood disaggregated model, and empty simultaneously
Image dynamic data base；And when image training module optimized to Characteristic Analysis Model and mood disaggregated model, needle
To per piece image, feature point extraction is carried out using Gabor methods in image tranining database, to obtain different features to
Amount；For the feature vector of every piece image, dimension-reduction treatment is carried out to it using PCA methods, to obtain new feature vector；
For all images in image tranining database, according to the classification belonging to feature vector and the image, using machine learning
Training is iterated to the weights of feature, to obtain Characteristic Analysis Model and mood disaggregated model, is realized to signature analysis mould
The optimization of type and mood disaggregated model.
In the present embodiment, described " according to face Emotion identification result in order to help driver to recognize the mood of oneself
Different type, show the emotional information of driver " further include specifically：According to face Emotion identification as a result, showing driver's
Mood classification and mood degree.It, can by showing the mood classification of driver and mood degree on output display screen
Preferably driver to be helped to recognize the mood of oneself according to the mood classification and mood degree of display, source can be made in time
Reason.
It should be noted that although the various embodiments described above have been described herein, it is not intended to limit
The scope of patent protection of the present invention.Therefore, based on the present invention innovative idea, to embodiment described herein carry out change and repair
Change, or using equivalent structure or equivalent flow shift made by description of the invention and accompanying drawing content, it directly or indirectly will be with
Upper technical solution is used in other related technical areas, is included within the scope of patent protection of the present invention.
Claims (10)
1. a kind of driver's Emotion identification and display device based on machine learning, which is characterized in that including face detection system,
Emotion identification system and mood display system；
The face detection system is connected to Emotion identification system, and the face detection system is used to obtain the figure of driver's face
As information, and processing is carried out to the image information of acquisition and obtains facial image model, and facial image model is sent out by treated
It send to mood identifying system；
The Emotion identification system is connected to mood display system, and the Emotion identification system is for extracting facial image model
Emotional characteristics point obtains face Emotion identification in face information as a result, and will by Characteristic Analysis Model and mood disaggregated model
Face Emotion identification result is sent to mood display system；
The mood display system is used to, according to the different type of the face Emotion identification result received, show the feelings of driver
Thread information.
2. driver's Emotion identification and display device based on machine learning according to claim 1, which is characterized in that described
Face identification system includes image collection module, Face detection module and image pre-processing module；
Facial expression information of the described image acquisition module for obtaining driver in real time, and convert thereof into human face image information
It is sent to Face detection module；
The Face detection module is for detecting human face region to be treated in human face image information, while cancelling noise is dry
It disturbs, by treated, human face image information is sent to image pre-processing module；
Described image preprocessing module is used for by gathering method for normalizing and gray scale normalization method to treated face figure
As information is handled, the facial image model of facial size and human face light information unification is obtained.
3. driver's Emotion identification and display device based on machine learning according to claim 1, which is characterized in that described
Emotion identification system includes face database, face emotional characteristics extraction module, face emotional characteristics selecting module and people
Face emotional characteristics sort module；
The face database is used to store the Characteristic Analysis Model and mood disaggregated model of human face image information；
The face emotional characteristics extraction module is used to carry out the facial image model received by Gabor filtering methods special
Sign vector extraction, obtains feature vector model, and the feature vector model of acquisition is sent to face emotional characteristics selecting module；
The face emotional characteristics selecting module is used to carry out at dimensionality reduction the feature vector model extracted by PCA algorithms
Reason, deletes the redundancy data that obtain that treated, and will treated that data are sent to face emotional characteristics sort module；
Model and mood disaggregated model are to receiving about the face classification modules are used for according to trained feature point
Data are classified, and obtain current mood generic title, and item name is sent to mood display system.
4. driver's Emotion identification and display device based on machine learning according to claim 3, which is characterized in that described
Emotion identification system further includes image tranining database and image dynamic data base；
Described image tranining database is used to preserve the training for face emotional characteristics sort module training characteristics analysis model
Image；
Described image dynamic data base is used to record the human face image information of the driver in certain time, and preserves the face figure
Facial image is believed in set time more new images tranining database as the classification results of information, and by the classification results of preservation
The Characteristic Analysis Model and mood disaggregated model of breath optimize.
5. driver's Emotion identification and display device based on machine learning according to claim 1, which is characterized in that described
The face Emotion identification that mood display system is used to be sent according to Emotion identification system on output display screen as a result, show mood
Classification and mood degree.
6. a kind of driver's Emotion identification and display methods based on machine learning, which is characterized in that include the following steps：
The human face image information of driver is obtained, and processing is carried out to the human face image information of acquisition and obtains facial image model；
The emotional characteristics point for extracting facial image model, is obtained by Feature Selection Model and mood disaggregated model in face information
Face Emotion identification result；
According to the different type of face Emotion identification result, the emotional information of driver is shown.
7. driver's Emotion identification and display methods based on machine learning according to claim 6, which is characterized in that described
" obtain the human face image information of driver, and carry out processing to the human face image information of acquisition and obtain facial image model " is specific
Including：
The facial expression information of driver is obtained in real time, and is converted into human face image information；
Human face region to be treated in human face image information is detected, while cancelling noise interferes；
By set method for normalizing and gray scale normalization method, to treated, human face image information is handled, and obtains face
The facial image model of size and human face light information unification.
8. driver's Emotion identification and display methods based on machine learning according to claim 6, which is characterized in that described
" the facial image model of extraction is obtained into face by Feature Selection Model in face database and mood disaggregated model
Face Emotion identification result in information " specifically includes：
Characteristic vector pickup is carried out to the facial image model received by Gabor filtering methods, obtains feature vector model；
Dimension-reduction treatment is carried out to the feature vector model that extracts by PCA algorithms, deletes the redundancy number that obtains that treated
According to；
According in face database trained Characteristic Analysis Model and mood disaggregated model to the data that receive
Classify, obtains current mood generic title；
The face database is used to store the Feature Selection Model and tagsort model of human face image information.
9. driver's Emotion identification and display methods based on machine learning according to claim 8, which is characterized in that described
Emotion identification system further includes image tranining database and image dynamic data base；
Described image tranining database is used to preserve the training image model for face emotional characteristics analysis module；
Described image dynamic data base is used to record the human face image information of the driver in certain time, and preserves the face figure
Facial image is believed in set time more new images tranining database as the classification results of information, and by the classification results of preservation
The Feature Selection Model and mood disaggregated model of breath optimize.
10. driver's Emotion identification and display methods based on machine learning according to claim 6, which is characterized in that institute
Stating " according to the different type of face Emotion identification result, showing the emotional information of driver " further includes specifically：
According to face Emotion identification as a result, the mood classification and mood degree of display driver.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810544319.9A CN108764169A (en) | 2018-05-31 | 2018-05-31 | A kind of driver's Emotion identification based on machine learning and display device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810544319.9A CN108764169A (en) | 2018-05-31 | 2018-05-31 | A kind of driver's Emotion identification based on machine learning and display device and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108764169A true CN108764169A (en) | 2018-11-06 |
Family
ID=64000819
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810544319.9A Pending CN108764169A (en) | 2018-05-31 | 2018-05-31 | A kind of driver's Emotion identification based on machine learning and display device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108764169A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109766771A (en) * | 2018-12-18 | 2019-05-17 | 深圳壹账通智能科技有限公司 | It can operation object control method, device, computer equipment and storage medium |
| CN109815817A (en) * | 2018-12-24 | 2019-05-28 | 北京新能源汽车股份有限公司 | Driver emotion recognition method and music pushing method |
| CN110197677A (en) * | 2019-05-16 | 2019-09-03 | 北京小米移动软件有限公司 | A kind of control method for playing back, device and playback equipment |
| CN110222623A (en) * | 2019-05-31 | 2019-09-10 | 深圳市恩钛控股有限公司 | Micro- expression analysis method and system |
| CN110944149A (en) * | 2019-11-12 | 2020-03-31 | 上海博泰悦臻电子设备制造有限公司 | Child care system and method for vehicle |
| CN111310730A (en) * | 2020-03-17 | 2020-06-19 | 扬州航盛科技有限公司 | Driving behavior early warning system based on facial expressions |
| CN111382608A (en) * | 2018-12-28 | 2020-07-07 | 广州盈可视电子科技有限公司 | Intelligent detection system with emotion recognition function |
| CN112257588A (en) * | 2020-10-22 | 2021-01-22 | 浙江合众新能源汽车有限公司 | Method for automatically identifying expression of driver |
| CN112927721A (en) * | 2019-12-06 | 2021-06-08 | 观致汽车有限公司 | Human-vehicle interaction method, system, vehicle and computer readable storage medium |
| CN113128534A (en) * | 2019-12-31 | 2021-07-16 | 北京中关村科金技术有限公司 | Method, device and storage medium for emotion recognition |
| CN113378733A (en) * | 2021-06-17 | 2021-09-10 | 杭州海亮优教教育科技有限公司 | System and device for constructing emotion diary and daily activity recognition |
| CN115359532A (en) * | 2022-08-23 | 2022-11-18 | 润芯微科技(江苏)有限公司 | Human face emotion capturing and outputting device based on 3D sensing |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103714660A (en) * | 2013-12-26 | 2014-04-09 | 苏州清研微视电子科技有限公司 | System for achieving fatigue driving judgment on basis of image processing and fusion between heart rate characteristic and expression characteristic |
| CN104537336A (en) * | 2014-12-17 | 2015-04-22 | 厦门立林科技有限公司 | Face identification method and system with self-learning function |
| CN104548309A (en) * | 2015-01-05 | 2015-04-29 | 浙江工业大学 | Device and method for adjusting driver emotional state through different affective characteristic music |
| CN105551499A (en) * | 2015-12-14 | 2016-05-04 | 渤海大学 | Emotion visualization method facing voice and facial expression signal |
| CN106529421A (en) * | 2016-10-21 | 2017-03-22 | 燕山大学 | Emotion and fatigue detecting auxiliary driving system based on hybrid brain computer interface technology |
| CN106627589A (en) * | 2016-12-27 | 2017-05-10 | 科世达（上海）管理有限公司 | Vehicle driving safety auxiliary method and system and vehicle |
| KR20170094836A (en) * | 2016-02-12 | 2017-08-22 | 한국전자통신연구원 | Apparatus and Method for recognizing a driver’s emotional state |
| CN107235045A (en) * | 2017-06-29 | 2017-10-10 | 吉林大学 | Consider physiology and the vehicle-mounted identification interactive system of driver road anger state of manipulation information |
| CN107766835A (en) * | 2017-11-06 | 2018-03-06 | 贵阳宏益房地产开发有限公司 | traffic safety detection method and device |
| CN107776579A (en) * | 2017-09-14 | 2018-03-09 | 中国第汽车股份有限公司 | A kind of direct feeling driver status alarm set |
-
2018
- 2018-05-31 CN CN201810544319.9A patent/CN108764169A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103714660A (en) * | 2013-12-26 | 2014-04-09 | 苏州清研微视电子科技有限公司 | System for achieving fatigue driving judgment on basis of image processing and fusion between heart rate characteristic and expression characteristic |
| CN104537336A (en) * | 2014-12-17 | 2015-04-22 | 厦门立林科技有限公司 | Face identification method and system with self-learning function |
| CN104548309A (en) * | 2015-01-05 | 2015-04-29 | 浙江工业大学 | Device and method for adjusting driver emotional state through different affective characteristic music |
| CN105551499A (en) * | 2015-12-14 | 2016-05-04 | 渤海大学 | Emotion visualization method facing voice and facial expression signal |
| KR20170094836A (en) * | 2016-02-12 | 2017-08-22 | 한국전자통신연구원 | Apparatus and Method for recognizing a driver’s emotional state |
| CN106529421A (en) * | 2016-10-21 | 2017-03-22 | 燕山大学 | Emotion and fatigue detecting auxiliary driving system based on hybrid brain computer interface technology |
| CN106627589A (en) * | 2016-12-27 | 2017-05-10 | 科世达（上海）管理有限公司 | Vehicle driving safety auxiliary method and system and vehicle |
| CN107235045A (en) * | 2017-06-29 | 2017-10-10 | 吉林大学 | Consider physiology and the vehicle-mounted identification interactive system of driver road anger state of manipulation information |
| CN107776579A (en) * | 2017-09-14 | 2018-03-09 | 中国第汽车股份有限公司 | A kind of direct feeling driver status alarm set |
| CN107766835A (en) * | 2017-11-06 | 2018-03-06 | 贵阳宏益房地产开发有限公司 | traffic safety detection method and device |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109766771A (en) * | 2018-12-18 | 2019-05-17 | 深圳壹账通智能科技有限公司 | It can operation object control method, device, computer equipment and storage medium |
| CN109815817A (en) * | 2018-12-24 | 2019-05-28 | 北京新能源汽车股份有限公司 | Driver emotion recognition method and music pushing method |
| CN111382608A (en) * | 2018-12-28 | 2020-07-07 | 广州盈可视电子科技有限公司 | Intelligent detection system with emotion recognition function |
| CN110197677A (en) * | 2019-05-16 | 2019-09-03 | 北京小米移动软件有限公司 | A kind of control method for playing back, device and playback equipment |
| CN110222623A (en) * | 2019-05-31 | 2019-09-10 | 深圳市恩钛控股有限公司 | Micro- expression analysis method and system |
| CN110944149A (en) * | 2019-11-12 | 2020-03-31 | 上海博泰悦臻电子设备制造有限公司 | Child care system and method for vehicle |
| CN112927721A (en) * | 2019-12-06 | 2021-06-08 | 观致汽车有限公司 | Human-vehicle interaction method, system, vehicle and computer readable storage medium |
| CN113128534A (en) * | 2019-12-31 | 2021-07-16 | 北京中关村科金技术有限公司 | Method, device and storage medium for emotion recognition |
| CN111310730A (en) * | 2020-03-17 | 2020-06-19 | 扬州航盛科技有限公司 | Driving behavior early warning system based on facial expressions |
| CN112257588A (en) * | 2020-10-22 | 2021-01-22 | 浙江合众新能源汽车有限公司 | Method for automatically identifying expression of driver |
| CN113378733A (en) * | 2021-06-17 | 2021-09-10 | 杭州海亮优教教育科技有限公司 | System and device for constructing emotion diary and daily activity recognition |
| CN115359532A (en) * | 2022-08-23 | 2022-11-18 | 润芯微科技(江苏)有限公司 | Human face emotion capturing and outputting device based on 3D sensing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108764169A (en) | A kind of driver's Emotion identification based on machine learning and display device and method | |
| Lu et al. | Driver action recognition using deformable and dilated faster R-CNN with optimized region proposals | |
| Moslemi et al. | Driver distraction recognition using 3d convolutional neural networks | |
| CN106611169B (en) | A kind of dangerous driving behavior real-time detection method based on deep learning | |
| Baheti et al. | Detection of distracted driver using convolutional neural network | |
| CN110059582B (en) | Driver behavior identification method based on multi-scale attention convolution neural network | |
| Sun et al. | Traffic sign detection and recognition based on convolutional neural network | |
| CN105956626A (en) | Deep learning based vehicle license plate position insensitive vehicle license plate recognition method | |
| CN110222596B (en) | Driver behavior analysis anti-cheating method based on vision | |
| CN106203330A (en) | A kind of vehicle classification method based on convolutional neural networks | |
| CN106469309A (en) | The method and apparatus of vehicle monitoring, processor, image capture device | |
| KR101845769B1 (en) | Car rear detection system using convolution neural network, and method thereof | |
| CN105740767A (en) | Driver road rage real-time identification and warning method based on facial features | |
| CN109635784A (en) | Traffic sign recognition method based on improved convolutional neural networks | |
| KR102105954B1 (en) | System and method for accident risk detection | |
| CN113920491A (en) | Fatigue detection system, method, medium and detection device based on facial skeleton model | |
| CN113642646A (en) | Image threat article classification and positioning method based on multiple attention and semantics | |
| Gan et al. | Bi-directional vectors from apex in cnn for micro-expression recognition | |
| CN104794432A (en) | Method and system for rapid cascade type car logo vision inspection and identification | |
| CN116129405A (en) | Method for identifying anger emotion of driver based on multi-mode hybrid fusion | |
| Chen | Traffic lights detection method based on the improved yolov5 network | |
| Al Nasim et al. | An automated approach for the recognition of bengali license plates | |
| CINAR et al. | Feature extraction and recognition on traffic sign images | |
| CN114120250B (en) | Video-based motor vehicle illegal manned detection method | |
| Padalia | Detection and Number Plate Recognition of Non-Helmeted Motorcyclists using YOLO |
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 |