• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The present invention concerns a method for determining the direction of the head of a person, which method comprises the step of creating (S1) an image of said head comprising a desired view of the head, which method further comprises the steps of: creating (S2) a deep data image (D1, D2, D3) of said head; extracting (S3) deep data from defined points (DP) on the deep data image of the head; using said extracted deep data for determining (S4) said direction of the head by utilizing a classification algorithm. The present invention also concerns a system for determining the direction of the head of a person. The present invention also concerns a motor vehicle. The present invention also concerns a computer program and a computer program product.
    公开号:SE1250377A1
    申请号:SE1250377
    申请日:2012-04-16
    公开日:2013-10-17
    发明作者:Jonatan Olofsson；Joseph Ah-King
    申请人:Scania Cv Ab；
    IPC主号:
    专利说明:

    Other systems for monitoring the driver's inattention are known, where detection of inattention has been accomplished through the use of a plurality of cameras and IR technology. eye direction for assessment of A problem with the use of brightness for determining the direction of the head is that the precision in the calculation of the direction of the head is limited.
    OBJECT OF THE INVENTION An object of the present invention is to provide a method for determining the direction of the head of a person who has high precision and does not require specific information about head shape.
    An object of the present invention is to provide a system for determining the direction of the head of a person who has high precision and does not require specific information about head shape.
    SUMMARY OF THE INVENTION These and other objects, which will be apparent from the following description, are accomplished by a method, system, motor vehicle, computer program, and computer program product of the kind initially described and further having the features set forth in the characterizing part of the appended independent claims. 9, 17, 18 and 19. Preferred embodiments of the method and system are defined in appended dependent claims 2-8 and 10-16.
    According to the invention, the objects are achieved with a method for determining the direction of the head of a person comprising the step of creating an image of said head including a desired view of the head, comprising the steps of: creating an in-depth data image of said head; extracting depth data from certain points of the depth data image of the head; use said extracted depth data to assess the direction of said head by using a classification algorithm. This enables precise assessments of the direction of the head in a person, where the assessment can be achieved on the basis of general cases and consequently the direction of the head can be determined in heads with very varying head shape. This is made possible, among other things, by said depth data from said determined points containing more relevant information for assessing the direction of the head than, for example, a light image, whereby the precision and consequently the possibility of correctly determining the direction of the head in a person is significantly improved. any specific information about the shape of the face, placement of eyes, nose and / or mouth is required.
    According to one method, a classification algorithm during a learning phase is continuously improved with respect to said embodiment of said judgment by feedback performing evaluations for a number of different principal forms to adapt the judgment's judgment ability to any principal form. In this case, the determination of the direction of the head is improved by means of said depth data from determined points.
    According to one embodiment of the method, said classification algorithm is based on properties of artificial neural networks. By using artificial network effective improvement of classification algorithm during said learning phase. neurally enabling said According to one embodiment of the method, said classification of the direction of the head comprises substantially straight forward, facing right, facing left, facing upwards and facing downwards. Hereby there are clear definitions of head direction to facilitate the said determination of the direction of the head. According to one embodiment, the method further comprises the step of determining the position of said head in the created depth data image. facilitates the creation of depth data image of the head.
    Hereby According to an embodiment of the method, the step of determining the position of said head in the created depth data image includes the step of creating a two-dimensional head contour as a basis for thereby enabling efficient creation of depth data image for efficient extraction of depth data from certain points of the depth data image of said judgment.
    According to an embodiment of the method, said head belongs to a vehicle driver present in a vehicle driver's compartment. By thus determining the direction of the head of a vehicle driver, it is possible to effectively assess any inattention of the driver that may affect the driver's ability to drive the vehicle safely, whereby measures can be taken to draw the driver's attention to the prevailing risk. Inattention may, for example, be due to fatigue of the driver.
    According to one embodiment of the method, information concerning the direction of said head and changes thereof is used to assess the presence of inattention in said person. In this way, the information about inattention can be used to draw the person's attention, thus preventing mistakes being made by the person, preventing accidents and / or improving the person's performance.
    According to the invention, the objects are achieved with a system for determining the direction of the head of a person including means for creating an image of said head including a desired view of the head, comprising means for creating a depth data image of said head; means for extracting depth data from specific points of the depth data image of the head; means for using said extracted depth data for judging the direction of said head by utilizing a classification algorithm. This enables precise assessments of the direction of the head in a person, where the assessment can be achieved on the basis of general cases and consequently the direction of the head can be determined in heads with very varying head shape. This is made possible, among other things, by said depth data from said determined points containing more relevant information for assessing the direction of the head than, for example, a light image, whereby the precision and consequently the possibility of correctly determining the direction of the head in a person is significantly improved. any specific information about the shape of the face, placement of eyes, nose and / or mouth is required.
    According to an embodiment of the system, during a learning phase, said classification algorithm is arranged to continuously improve with respect to said assessment by performing feedback for a number of different main forms with feedback in order to adapt the judgment's judgment ability to any main form. In this case, the determination of the direction of the head is improved by means of said depth data from determined points.
    According to one embodiment of the system, said classification algorithm is based on properties of artificial neural networks. By using artificial classification algorithm during said learning phase. neural network enables effective improvement of said According to one embodiment of the system, said classification of the direction of the head comprises substantially straight forward, facing right, facing left, facing upwards and facing downwards. Thereby, there are clear definitions of head direction to facilitate said determination of head direction.
    According to one embodiment, the system further comprises means for determining the position of said head in the created depth data image. facilitates the creation of depth data image of the head.
    According to an embodiment of the system, said means for determining the position of said head in the created depth data image comprises means for creating a two-dimensional head contour as a basis for said assessment. This enables efficient creation of depth data image for efficient extraction of depth data from specific points of the depth data image of the head. According to an embodiment of the system, said head belongs to a vehicle driver present in a vehicle driver's compartment. By thus determining the direction of the head of a vehicle driver, it is possible to effectively assess any inattention of the driver that may affect the driver's ability to drive the vehicle safely, whereby measures can be taken to draw the driver's attention to the prevailing risk. Inattention may, for example, be due to fatigue of the driver.
    According to an embodiment of the system, information concerning the direction of said head and changes thereof is intended to be used to assess the presence of inattention in said person. In this way, the information about inattention can be used to draw the person's attention, thus preventing mistakes from being made by the person, preventing accidents and / or improving the person's performance.
    DESCRIPTION OF THE DRAWINGS The present invention will be better understood with reference to the following detailed description read in conjunction with the accompanying drawings, in which like reference numerals refer to like parts throughout the many views, and in which: Fig. 1 schematically illustrates a motor vehicle according to an embodiment of the present invention; invention; Fig. 2 schematically illustrates a system for determining the direction of the head of a person according to an embodiment of the present invention; Fig. 3 schematically illustrates a template of a generic head; Figs. 4a-c schematically illustrate depth data images of persons looking in different directions with specific points from which depth data is extracted; Fig. 5 schematically illustrates a block diagram of a method for determining the direction of the head of a person according to an embodiment of the present invention; and Fig. 6 schematically illustrates a computer according to an embodiment of the present invention.
    DESCRIPTION OF EMBODIMENT FOFWIER Here, the term "link" refers to a communication link which may be a physical line, such as an optoelectronic communication line, or a non-physical line, such as a wireless connection, for example a radio or microwave link.
    Fig. 1 schematically illustrates a motor vehicle 1 according to an embodiment of the present invention. The exemplary vehicle 1 consists of a heavy vehicle in the form of a truck with a cab 2. The vehicle can alternatively consist of any suitable vehicle such as a bus or a passenger car. The vehicle comprises a system I according to the present invention.
    Fig. 2 schematically illustrates a system I for determining the direction of the head of a person according to an embodiment of the present invention.
    System I comprises an electronic control unit 100 or computer for said determination of the direction of the head of a person.
    The system 1 further comprises means 110 for creating a depth data image.
    Said means 110 for creating a depth data image comprises a camera means 110 for depth data imaging. According to a variant, the camera means for depth data image consists of a 3D camera means, where according to a variant the camera means consists of Kinect® by the Microsoft Corportation.
    According to a variant, said camera means 110 is arranged in a driver's compartment such as a driver's cab. In this case, the camera means 110 is arranged in the driver's compartment to image an in-depth data image of the driver's head. In this case, the camera member 110 is directed towards the driver's face. According to a variant, the camera means 110 is arranged at a distance from and obliquely above the driver.
    The system further includes means 120 for utilizing directional head information. Said means 120 may include warning means such as alarm means, vibration means, flashing means / light means and / or activating means for activating braking means of the vehicle for the driver's attention. Said warning means may be dependent on other information and be activated when certain criteria are met, such as, when the system is arranged in a vehicle, an obstacle along the route of the vehicle or the like. Accordingly, according to this variant, information concerning the direction of the head and changes thereof is intended to be used to assess the presence of inattention in the person and by means of said means utilize the information to draw the person's attention. In this way, the information about inattention can be used to draw the person's attention so as to prevent mistakes being made in the person, prevent accidents, for example when driving a vehicle and / or improve the person's performance.
    The electronic control unit 100 is signal connected to said camera means 110 for depth data imaging via a link 10. The electronic control unit 100 is arranged via the link 10 to receive a signal from said camera means 1 representing depth data.
    The electronic control unit 100 comprises means 102 for pre-processing said depth data image. Said means 102 for pre-processing said depth data image includes any suitable image processing means. According to a variant, said means 102 for pretreating said depth data image includes low-pass filtering to reduce noise and to reduce details of the depth data image. Reduction of details of the depth data image is done when only the shape and direction of the head are of interest. According to a variant, said means 102 for pre-processing said depth data image comprises means for creating a contour image of the head including an edge detector. According to a variant, a so-called Canny edge detector which is a previously known algorithm, where an image is taken as described with a single value at each pixel, for example grayscale value or depth value, is used as input and performs a number of steps in accordance with the algorithm for creating the contour image. .
    The electronic control unit 100 comprises means 104 for determining the position of said head in the created depth data image. Said means 104 for determining the position of said head in the created depth data image includes means for creating a two-dimensional head contour as a basis for said assessment. Said means 104 for determining the position of said head in the created depth data image comprises according to a variant so-called chamfer matching which is a previously known algorithm, where the contour image is used to match a binary template to it. According to a variant, a distance map of the above-mentioned created contour image is created, said distance map of the contour image being used to find the position in which the distance of the binary template is minimized.
    According to a variant, a generic main template is created, where the fit of the main template is obtained according to a variant by studying how well it matches a selection of a number of different main images in a number of people where the people look in different directions at different main images. The performance of the generic main template is intended to be independent of the direction of the person's head.
    Fig. 3 shows an example of a generic main template.
    The electronic controller 100 includes means 106 for extracting depth data from specific points of the depth data image of the head. When the position of the head in the created depth image is known, depth data can be extracted from the depth data image. According to one embodiment, said determined points are constituted by a set of points distributed according to a predetermined pattern.
    According to an embodiment shown in Figs. 4a-c, said depth data points DP are divided into three rows where the underlying row is slightly offset to the right relative to the above rows. According to the embodiment shown in Figs. 4a-c, each row has five points and the total number of points is consequently fifteen. The number of points can be any suitable number and the distribution of the points can be any suitable distribution.
    The coordinates of these rows are calculated by the known positions of the determined main template.
    The electronic controller 100 includes means 108 for using said extracted depth data to assess the direction of said head by utilizing a classification algorithm. This takes advantage of the fact that said predetermined points of depth data contain information about the direction of the head. Said classification algorithm is used to find out how said depth data correlates with different directions of a person's head.
    Said classification algorithm is based on properties of means for identifying linear relationships between input data in the form of a person's head shape and output data in the form of different directions of the person's head, where said output data according to one embodiment is classified by: the person looking left, the person looks forward, and the person looks right.
    Said output can be classified by any suitable direction of the person's head, including the person looking left, the person looking forward, the person looking right, the person looking up, the person looking backwards, the person looking backwards. Furthermore, said output data can be classified by continuous distribution of the direction of the head, i.e. a continuous distribution of different angles of the head including left, right, right, back and angles in between, and / or up, straight and down and angles in between. Accordingly, according to one embodiment, said classification algorithm is configured to identify any distribution of the direction of the head, i.e. identification of the head facing straight forward and any angle of the head relative to the head facing straight forward including the head facing straight forward, the head facing: left, right, up, down, backward and angles therebetween.
    According to a preferred embodiment, said classification algorithm is based on alternative embodiments, said classification algorithm is based on properties of properties of artificial neural networks. According to the so-called support vector machine, AdaBoost or Bayesian network.
    Said classification algorithm is, during a learning test, arranged to continuously improve with respect to said assessment by performing with feedback assessments for a number of different main forms in order to adapt the algorithm's judgment to any main form.
    Said input data in the form of main shapes, where the main shapes in learning consist of main shapes which have substantially different shapes. Furthermore, according to a variant, an additional main form is used for validation.
    The artificial neural network is designed to generalize the relationship between input data in the form of head shapes and output data in the form of head direction. Two hidden layers, a first hidden layer and a second hidden layer, are a sufficient number of hidden layers, as any continuous or discrete function can be stored in two hidden layers, provided that they are large enough. The variables considered in the construction of the network are number of neurons in the first hidden layer, number of neurons in the second hidden layer, learning speed n and amount of movement o. Optimization of the electronic control unit 100 is consequently arranged to determine said direction based on said extracted depth data by utilization of a classification algorithm. The electronic control unit 100 is signal connected to said means 120 for utilizing information about the direction of the head via a link 20. The electronic control unit 100 is arranged via the link 20 to send a signal to said means 120 for utilizing information about the head. direction representing the main direction data from said determined direction of the head for utilizing information about the direction of the head, where activation takes place according to the above based on said information, i.e. based on the direction of the head, as well as any other information, to draw the person's attention to any inattention.
    Fig. 3 schematically illustrates an embodiment of a generic main template HT for optimizing the fit of different head shapes in persons whose head direction is to be determined. The main template HT according to this embodiment includes a number of distinct points P1, P2, an algorithm being configured to draw lines L between adjacent points P1, P2. Each point is described by an x-coordinate and a y-coordinate. According to this embodiment, the main template 42 includes distinct points. The main template HT can have any suitable number of distinct points.
    Fig. 4a-c schematically illustrates depth data images D1, D2, D3 of different people looking in different directions, ie. different direction of the head H1, H2, H3 in different persons with different head shape, with definite points DP from which depth data is extracted. The persons according to Figs. 4a-c are filtered depth data images D1, D2, D3 of drivers sitting in a driver's compartment of a vehicle. Here, the distribution of said points is DP from which depth data is extracted three rows of five points, where the rows are slightly offset from each other so that the underlying rows of points are offset to the right relative to the overlying row of points. As mentioned, this is just an example. The points can be any number and distributed in any way.
    Fig. 4a shows a driver looking to the left, Fig. 4b shows a driver looking straight ahead, and Fig. 4c shows a driver looking to the right. Fig. 5 schematically illustrates a block diagram of a method for determining the direction of the head of a person according to an embodiment of the present invention.
    According to one embodiment, the method for determining the direction of the head of a person comprises a first step S1. In this step, an image of said head is created including a desired view of the head.
    According to one embodiment, the method for determining the direction of the head of a person comprises a second step S2. In this step, a depth data image of said head is created.
    According to one embodiment, the method for determining the direction of the head of a person comprises a third step S3. In this step, depth data is extracted from specific points of the depth data image by the header.
    According to one embodiment, the method for determining the direction of the head of a person comprises a fourth step S4. In this step, said extracted depth data is used for judging the direction of said head by using a classification algorithm.
    Referring to Fig. 6, there is shown a diagram of an embodiment of a device 500. The controller 100 described with reference to Fig. 2 may in one embodiment include the device 500. The device 500 includes a non-volatile memory 520, a data processing unit 510, and a read / write memory 550.
    The non-volatile memory 520 has a first memory portion 530 in which a computer program, such as an operating system, is stored to control the operation of the device 500. Further, the device 500 includes a bus controller, a serial communication port, I / O means, a A / D converter, a time and date input and transfer unit, an event counter and an interrupt controller (not shown). The non-volatile memory 520 also has a second memory portion 540. A computer program P is provided which includes routines for determining the direction of the head of a person according to the innovative method.
    The program P includes routines for creating an image of said head including a desired view of the head. The program P includes routines for creating a depth data image of said head. The program P includes routines for extracting depth data from specific points of the depth data range of the header. The program P comprises routines for using said extracted depth data for assessing the direction of said head by using a classification algorithm. The program P can be stored in an executable manner or in a compressed manner in a memory 560 and / or in a read / write memory 550.
    When it is described that the data processing unit 510 performs a certain function, it is to be understood that the data processing unit 510 performs a certain part of the program which is stored in the memory 560, or a certain part of the program which is stored in the read / write memory 550.
    The data processing device 510 can communicate with a data port 599 via a data bus 515. The non-volatile memory 520 is intended for communication with the data processing unit 510 via a data bus 512. The separate memory 560 is intended to communicate with the data processing unit 510 via a data bus 511. the data processing unit 510 via a data bus 514. To the data port 599, e.g. the links connected to the control unit 100 are connected.
    The read / write memory 550 is arranged to communicate with When data is received on the data port 599, it is temporarily stored in the second memory part 540. When the received input data has been temporarily stored, the data processing unit 510 is arranged to perform code execution in a manner described above. The received signals on the data port 599 can be used by the device 500 to create an image of said head including a desired view of the head. The received signals on the data port 599 may be used by the device 500 to create a deep data image of said head; extract depth data from certain points of the depth data image of the head. The received signals on the data port 599 can be used by the device 500 to extract depth data from certain points of the depth data image of the header. The received signals on the data port 599 can be used by the device 500 to use said extracted depth data for judging the direction of said head by using a classification algorithm.
    Parts of the methods described herein may be performed by the device 500 by means of the data processing unit 510 running the program stored in the memory 560 or the read / write memory 550. When the device 500 runs the program, the methods described herein are executed.
    The above description of the preferred embodiments of the present invention has been provided for illustrative and descriptive purposes. It is not intended to be exhaustive or to limit the invention to the variations described. Obviously, many modifications and variations will occur to those skilled in the art. The embodiments have been selected and described to best explain the principles of the invention and its practical applications, thereby enabling one skilled in the art to make the first invention for different embodiments and with the various modifications suitable for the intended use.
    权利要求:
    Claims (19)
    [1]
    Method for determining the head including the step of creating (S1) an image of said head including a direction of a person desired view of the head, characterized by the steps of: creating (S2) a depth data image (D1, D2, D3) of said head; extracting (S3) depth data from specific points (DP) of the depth data image of the head; use said extracted depth data for judging (S4) the direction of said head by using a classification algorithm.
    [2]
    A method according to claim 1, wherein said classification algorithm during a learning phase is continuously improved with respect to said assessment by performing feedback for a number of different principal forms with feedback to adapt the judgment's judgment to any principal form.
    [3]
    A method according to claim 1 or 2, wherein said classification algorithm is based on properties of artificial neural networks.
    [4]
    A method according to any one of claims 1-3, wherein said classification of the direction of the head comprises substantially straight forward, facing right, facing left, facing upwards and / or facing downwards.
    [5]
    A method according to any one of claims 1-4, further comprising the step of determining the position of said head in the created depth data image (D1, D2, D3).
    [6]
    The method of claim 5, wherein the step of determining the position of said head in the created depth data image (D1, D2, D3) comprises the step of creating a two-dimensional head contour as a basis for said assessment.
    [7]
    A method according to any one of claims 1-6, wherein said head belongs to a vehicle driver present in a vehicle driver's compartment. 10 15 20 25 17
    [8]
    A method according to any one of claims 1-7, wherein information concerning the direction of said head and changes thereof is used to assess the presence of inattention in said person.
    [9]
    System including means (110) for creating an image of said head (I) for determining the direction of the head of a person including a desired view of the head, characterized by means (100, 110) for: creating a depth data image (D1 , D2, D3) of said head; means (100, 106) for extracting depth data from certain points (DP) of the depth data image (D1, D2, D3) of the head; means (100, 108) for using said extracted depth data to assess the direction of said head by utilizing a classification algorithm.
    [10]
    A system according to claim 9, wherein during a learning phase, said classification algorithm is arranged to continuously improve with respect to said assessment by performing feedback for a number of different main forms with feedback to adapt the judgment's judgment to any main form.
    [11]
    A system according to claim 9 or 10, wherein said classification algorithm is based on properties of artificial neural networks.
    [12]
    A system according to any one of claims 9-11, wherein said classification of the direction of the head includes substantially straight forward, turned to the right, turned to the left, turned upwards and / or turned downwards.
    [13]
    A system according to any one of claims 9-12, further comprising means (100, 104) for determining the position of said head in the created depth data image (D1, D2, D3).
    [14]
    The system of claim 13, wherein said means (100, 104) for determining the position of said head in the created depth data image (D1, D2, D3) includes means for creating a two-dimensional head contour as a basis for said judgment. 10 18
    [15]
    A system according to any one of claims 9-14, wherein said head belongs to a vehicle driver present in a vehicle driver's compartment.
    [16]
    A system according to any one of claims 9-15, wherein information concerning the direction of said head and changes thereof is intended to be used to assess the presence of inattention in said person.
    [17]
    Motor vehicle (1) including a system (I) according to any one of claims 9-16.
    [18]
    A computer program (P) for determining the direction of the head of a person, wherein said computer program (P) comprises program code which, when run by an electronic control unit (100) or another computer (500) connected to the electronic control unit (100), able the electronic control unit (100) to perform the steps according to claims 1-8.
    [19]
    A computer program product comprising a digital storage medium storing the computer program (P) according to claim 18.
    类似技术:
    公开号 | 公开日 | 专利标题
    CN107031508B|2019-08-06|Back-sight visual system and its application method for vehicle
    CN106485233A|2017-03-08|Drivable region detection method, device and electronic equipment
    US9767368B2|2017-09-19|Method and system for adaptive ray based scene analysis of semantic traffic spaces and vehicle equipped with such system
    JP5132750B2|2013-01-30|Behavior-based learning of visual characteristics from real-world traffic scenes for driver assistance systems
    CN102320301B|2015-12-09|For the method making the ride characteristic of vehicle adapt to chaufeur conversion
    US20170297488A1|2017-10-19|Surround view camera system for object detection and tracking
    US20160217335A1|2016-07-28|Stixel estimation and road scene segmentation using deep learning
    US10290080B2|2019-05-14|Method for displaying a vehicle environment of a vehicle
    US11216673B2|2022-01-04|Direct vehicle detection as 3D bounding boxes using neural network image processing
    CN106537180A|2017-03-22|Method for mitigating radar sensor limitations with video camera input for active braking for pedestrians
    CN105719311A|2016-06-29|Vehicle System For Detecting Object And Operation Method Thereof
    US20160307026A1|2016-10-20|Stereoscopic object detection device and stereoscopic object detection method
    US10552692B2|2020-02-04|Color learning
    SE1250377A1|2013-10-17|Procedure and system for determining the direction of the head of the person
    CN107161145A|2017-09-15|Method and control device for avoiding rear-end collision
    CN106570451A|2017-04-19|Self-recognition of autonomous vehicles in mirrored or reflective surfaces
    CN110770810A|2020-02-07|Ambient monitoring of autonomous vehicles
    US20210034925A1|2021-02-04|Obstacle recognition assistance device, obstacle recognition assistance method, and storage medium
    JP6918425B2|2021-08-11|Devices, methods, and programs for detecting boundaries contained in captured images
    CN109720352A|2019-05-07|Vehicle drive auxiliary control method and equipment
    CN111052735B|2022-01-04|Image processing apparatus, image processing method, and image display system
    JP2012181691A|2012-09-20|Method for recognizing patten on road surface, and information recording device for vehicle
    WO2020008787A1|2020-01-09|Marker recognition method for camera device, and marker recognition device
    JP7005762B2|2022-02-10|Sign recognition method of camera device and sign recognition device
    US11069049B2|2021-07-20|Division line detection device and division line detection method
    同族专利:
    公开号 | 公开日
    EP2654020A1|2013-10-23|
    EP2654020B1|2018-08-01|
    SE536976C2|2014-11-18|
    BR102013008598A2|2016-01-26|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US7620202B2|2003-06-12|2009-11-17|Honda Motor Co., Ltd.|Target orientation estimation using depth sensing|CN107944415A|2017-12-06|2018-04-20|董伟|A kind of human eye notice detection method based on deep learning algorithm|
    CN112758001A|2021-01-27|2021-05-07|四川长虹电器股份有限公司|TOF-based vehicle lamp follow-up control method|
    法律状态:
    优先权:
    申请号 | 申请日 | 专利标题
    SE1250377A|SE536976C2|2012-04-16|2012-04-16|Procedure and system for determining information on the head of a vehicle driver|SE1250377A| SE536976C2|2012-04-16|2012-04-16|Procedure and system for determining information on the head of a vehicle driver|
    EP13161540.3A| EP2654020B1|2012-04-16|2013-03-28|Method and system for determining the direction of the head of a person|
    BR102013008598A| BR102013008598A2|2012-04-16|2013-04-09|Method and system for determining the direction of a person's head|
    [返回顶部]