NO20220168A1

NO20220168A1 - Method and structure for locating human face by in-vehicle camera

Info

Publication number: NO20220168A1
Application number: NO20220168A
Authority: NO
Inventors: Chenxuan Li; Qi Zhang; Kefeng Wang
Original assignee: Hozon New Energy Automobile Co Ltd
Priority date: 2019-08-13
Filing date: 2022-02-07
Publication date: 2022-02-07
Also published as: WO2021027701A1; CN110509924A

Description

METHOD AND STRUCTURE FOR POSITIONING HUMAN FACE

BY IN-VEHICLE CAMERA

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of in-vehicle cameras, and in particular, to a method and a structure for positioning a human face by an in-vehicle camera.

BACKGROUND

[0002] The development of image recognition technology has driven the application scenarios of cameras. At present, more and more application scenarios of cameras exist in automobiles, for example, for monitoring driver fatigue, recognizing driver expression, positioning driver eye, and the like.

[0003] When the camera used to photograph the driver's face is placed on the steering column of the steering wheel of the car, the coordinates of the driver's face detected by the camera are based on the camera coordinate system. In order to convert the coordinates of the driver's face to the vehicle coordinate system for other applications, it is also necessary to know the coordinates of the camera in the vehicle coordinate system, and then convert the camerabased coordinates of the driver's face to the vehicle coordinate system.

[0004] However, because the steering column of the steering wheel of the car generally has four mechanical adjustments, up, down, front and rear, the position of the in-vehicle camera on the steering column of the steering wheel is not fixed, so it is impossible to directly locate the coordinates of the camera in the vehicle coordinate system, and it is impossible to accurately locate the coordinates of the driver’s face in the vehicle coordinate system.

SUMMARY

[0005] The present disclosure is intended to solve the above-mentioned problems in the prior art, and provide a method and a structure for positioning a human face by an in-vehicle camera, such that the driver's face can be accurately and effectively positioned, thereby improving recognition accuracy of the state of the driver's face.

[0006] The object of the present disclosure can be realized through the following technical solutions:

[0007] A method for locating a face position by an in-vehicle camera, comprising the following steps:

[0008] method for positioning a human face by an in-vehicle camera, comprising steps of:

[0009] SI, determining a specific position of the in-vehicle camera by a fixed icon installed in a vehicle configured for being captured by the in-vehicle camera;

[0010] S2, determining a specific position of a driver's face by the specific position of the in-vehicle camera.

[0011] In the present disclosure, the specific position of the in-vehicle camera is firstly obtained accurately through the accurate position of the fixed icon fixed somewhere in the car and the identification of the relative position of the fixed icon by the in-vehicle camera; and then according to the accurate position of the in-vehicle camera, and the identification of the relative position of the driver's face by the in-vehicle camera, the specific position of the driver's face is finally obtained accurately.

[0012] In an embodiment, before step SI, the method further comprises: establishing an invehicle coordinate system, and obtaining coordinates of the fixed icon in the in-vehicle coordinate system.

[0013] In an embodiment, the step SI is specifically: obtaining the specific position of the fixed icon relative to the in-vehicle camera through an image recognition algorithm; and calculating coordinates of the in-vehicle camera in the in-vehicle coordinate system according to the coordinates of the fixed icon in the in-vehicle coordinate system.

[0014] In an embodiment, the step S2 is specifically: obtaining the specific position of the driver's face relative to the in-vehicle camera through the image recognition algorithm; and calculating coordinates of the driver's face in the in-vehicle coordinate system according to the coordinates of the in-vehicle camera in the in-vehicle coordinate system.

[0015] In an embodiment, said obtaining the specific position of the fixed icon relative to the in-vehicle camera through the image recognition algorithm is specifically: calculating distances of the fixed icon relative to the in-vehicle camera in X, Y and Z directions of the invehicle coordinate system, respectively.

[0016] In an embodiment, said obtaining the specific position of the driver's face relative to the in-vehicle camera through the image recognition algorithm is specifically: calculating distances of the driver's face relative to the in-vehicle camera in X, Y, Z directions of the invehicle coordinate system, respectively.

[0017] In an embodiment, the fixed icon is provided on a b-pillar on a side proximal to a driver's seat.

[0018] The present disclosure further provides a structure for positioning a human face by an in-vehicle camera, comprising an in-vehicle camera provided on a steering column of a steering wheel, a fixed icon provided in a capturing range of the in-vehicle camera, and a controller integrated in the in-vehicle camera or provided independently.

[0019] In an embodiment, the fixed icon is provided on a b-pillar on a side proximal to a driver's seat.

[0020] The present disclosure has the advantages that the method and the structure are simple and efective, and the accuracy of positioning the driver's face can be ensured by accurately positioning the in-vehicle camera, thereby improving the recognition accuracy of the in-vehicle camera for the driver's face state.

BRIEF DESCRIPTION OF DRAWINGS

[0021] FIG. 1 is the flow chart illustrating the method according to the present disclosure;

[0022] FIG. 2 is the coordinate diagram of the method according to the present disclosure;

[0023] FIG. 3 is a schematic diagram illustrating the structure according to the present disclosure.

[0024] In the figures: 1, in-vehicle camera; 2, driver's face; 3, fixed icon.

DESCRIPTION OF EMBODIMENTS

[0025] The present disclosure will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0026] A method for positioning a human face by an in-vehicle camera, comprising the following steps:

[0027] SI, determining a specific position of the in-vehicle camera by a fixed icon installed in a vehicle configured for being captured by the in-vehicle camera;

[0028] S2, determining a specific position of a driver's face by the specific position of the in-vehicle camera.

[0029] In the present method, the specific position of the in-vehicle camera is firstly obtained accurately through the accurate position of the fixed icon fixed somewhere in the car and the identification of the relative position of the fixed icon by the in-vehicle camera; and then according to the accurate position of the in-vehicle camera, and the identification of the relative position of the driver's face by the in-vehicle camera, the specific position of the driver's face is finally obtained accurately.

[0030] Specifically, before step 1, the method may further include: establishing an invehicle coordinate system, and obtaining coordinates of the fixed icon in the in-vehicle coordinate system. By establishing an in-vehicle coordinate system, the position coordinates of each component in the car are more intuitive and accurate, so as to facilitate the subsequent positioning of the in-vehicle camera and the driver's face.

[0031] The step SI may be specifically: obtaining the specific position of the fixed icon relative to the in-vehicle camera through an image recognition algorithm; and calculating coordinates of the in-vehicle camera in the in-vehicle coordinate system according to the coordinates of the fixed icon in the in-vehicle coordinate system. Similarly, the step S2 is specifically: obtaining the specific position of the driver's face relative to the in-vehicle camera through the image recognition algorithm; and calculating coordinates of the driver's face in the in-vehicle coordinate system according to the coordinates of the in-vehicle camera in the invehicle coordinate system. The above two steps respectively obtain the precise coordinates of the in-vehicle camera in the in-vehicle coordinate system, and the precise coordinates of the driver's face in the in-vehicle coordinate system.

[0032] Wherein, said obtaining the specific position of the fixed icon relative to the invehicle camera through the image recognition algorithm is specifically: calculating distances of the fixed icon relative to the in-vehicle camera in X, Y and Z directions of the in-vehicle coordinate system, respectively. Similarly, said obtaining the specific position of the driver's face relative to the in-vehicle camera through the image recognition algorithm is specifically: calculating distances of the driver's face relative to the in-vehicle camera in X, Y, Z directions of the in-vehicle coordinate system, respectively.

[0033] For example, as shown in FIG. 2, the larger coordinate system in the figure is the in-vehicle coordinate system, and the smaller coordinate system is the coordinate system with the in-vehicle camera as the origin (used to identify and obtain fixed icon and the distance between the driver's face from the in-vehicle camera). In the figure, 1 represents the in-vehicle camera, 2 represents the driver's face, and 3 represents the fixed icon. The calculation of the coordinates of the driver's face in the X direction in the in-vehicle coordinate system is taken as an example.

[0034] We want to obtain the position of the driver's face in the in-vehicle coordinate system, namely dx, and dx=dc+dz; where, dc is the position of the in-vehicle camera in the in-vehicle coordinate system, and dz is the position of the driver's face relative to the in-vehicle camera (which can be obtained through the camera's recognition and image recognition algorithms). dc=dxl-dzl; where dxl is the position of the fixed icon in the in-vehicle coordinate system, and dzl is the position of the fixed icon relative to the in-vehicle camera. To sum up, dx=dxldzl+dz. Similarly, we can calculate the Y and Z coordinate values of the driver's face in the invehicle coordinate system.

[0035] In addition, the present disclosure also provides a structure for positioning a human face by an in-vehicle camera, comprising an in-vehicle camera 1 provided on a steering column of a steering wheel, a fixed icon 2 provided in the capturing/illumination range of the in-vehicle camera 1, and a controller integrated in the in-vehicle camera or provided independently. The controller is mainly used to process the image information obtained by the in-vehicle camera, and obtain the position information of the driver's face and the state information of the driver's face. Specifically, the fixed icon is provided on the b-pillar on the side proximal to the driver's seat. The fixed icon is provided on the b-pillar, such that it is not easy to be blocked by the driver, and is convenient for the illumination and positioning of the camera in the car.

[0036] The above is only a preferred specific embodiment of the present disclosure, which is an implementation based on the overall concept of the present disclosure, and the protection scope of the present disclosure is not limited to this. Any changes or substitutions that can be easily conceived by any person skilled in the art within the technical scope disclosed by the present disclosure shall be included within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the appended claims.

Claims

a.s c maemeos:orosonn aumanace ann-vece camera,x coemcrosnnns saees on: vece, coenermurennor a seenccoson o en-vece camera a a

vece, cameeerram.nn a sec cacuoreson oe an-vreverc'se caacmeerae seccoson o en-

w. e meoorosonn aumanace an:n-vece camera accorno cam, sseeremn, aneo orea snen co,oren maeeso ouerexre cocmonrsnese esna-vescne co anornn-vaee scese cmoo.rnae

were.nee m seeoosr soescocnan: a ouamnannacee see acnnc-veoscoen ca omerea acxceorcnon roe caavmeo, en-vece camerarou anmae recon

oecne ceamne-vraencee cono-rvenac on aor acmcor ann caocuea n ee s csooerm.nae ssem cnoor coonraensa oes o xeen-cv e

woeree.ne

ee m seeooroesconn aumanace ann-vece camera accorno c er or rea nam avee, o s vecen

is-v

cavme'sce

raanc ceam

enesr said

n-evenro-vuca: oann e seccoson o erver'sace cee ccooee

r cmonoa

oar

an R

saec

seo

e sn

ms.eomn a accoornm aone ca cocourannae cso o en-

were.n sae m oeaonnoroes soencnc aousmoann oacee axnen-vece camera accorno cam, rou emae reconon aorms secca: cacocnu raeanvesoen-vece camera reaveoen-vece cameran, an recons o en-veacnece cso oorneaex sesecmon, respectively.

6. The method for positioning a human face by an in-vehicle camera according to claim 4, wherein said obtaining the specific position of the driver's face relative to the in-vehicle camera through the image recognition algorithm is specifically: calculating distances of the driver's face relative to the in-vehicle camera in X, Y, Z directions of the in-vehicle coordinate system, respectively.

7. The method for positioning a human face by an in-vehicle camera according to claim 1, wherein the fixed icon is provided on a b-pillar on a side proximal to a driver's seat.

8. A structure for positioning a human face by an in-vehicle camera, comprising an in-vehicle camera provided on a steering column of a steering wheel, a fixed icon provided in a capturing range of the in-vehicle camera, and a controller integrated in the in-vehicle camera or provided independently.

9. The structure for positioning a human face by an in-vehicle camera according to claim 8, wherein the fixed icon is provided on a b-pillar on a side proximal to a driver's seat.