CN103760975A - Method for tracking and positioning faces and display system - Google Patents
Method for tracking and positioning faces and display system Download PDFInfo
- Publication number
- CN103760975A CN103760975A CN201410002101.2A CN201410002101A CN103760975A CN 103760975 A CN103760975 A CN 103760975A CN 201410002101 A CN201410002101 A CN 201410002101A CN 103760975 A CN103760975 A CN 103760975A
- Authority
- CN
- China
- Prior art keywords
- line segment
- actual
- point
- user
- display screen
- 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.)
- Granted
Links
Images
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method for tracking and positioning faces and a display system. The method includes S0, setting a first reference line segment, a second reference line segment, a third reference line segment and a fourth reference line segment which are at least four line segments formed by a plurality of reference feature points, and setting a comparison reference point; S1, forming a first actual line segment, a second actual line segment, a third actual line segment and a fourth actual line segment which correspond to the reference line segments, and setting an actual comparison point; S2, comparing a specific ratio of the first actual line segment to the second actual line segment to a specific ratio of the first reference line segment to the second reference line segment, and judging whether the specific ratio of the first actual line segment to the second actual line segment is identical to the specific ratio of the first reference line segment to the second reference line segment or not; S3, comparing a specific ratio of the third actual line segment to the fourth actual line segment to a specific ratio of the third reference line segment to the fourth reference line segment, and judging whether the specific ratio of the third actual line segment to the fourth actual line segment is identical to the specific ratio of the third reference line segment to the fourth reference line segment or not; S4, comparing the position of the actual comparison point to the position of the comparison reference point, and judging whether the position of the actual comparison point is identical to the position of the comparison reference point or not. The method and the display system have the advantage that the faces can be accurately tracked and positioned by the aid of the method and the display system.
Description
Technical field
The present invention relates to electronic device field, more particularly, relate to a kind of by adjusting method and the display system of screen position tracing and positioning people face.
Background technology
Electronic installation has the display screen of watching for user, and it is widely used in various terminating machines such as comprising notebook personal computer, desktop personal computer, all-in-one, only has when user is with respect to this electronic installation during in optimal viewing angle, and its viewing effect is just best.At present when the direction of electronic installation is constant user because certain need to change face when using this electronic installation, although also exist according to people's face change in location and adjust the method that the position of electronic installation display screen and the position of people's face adapt, but implementation procedure is all comparatively complicated, calculate comparatively loaded down with trivial details, be difficult to realize accurate location, affect user's impression, reduced user's the joyful sense of use.
Summary of the invention
The technical problem to be solved in the present invention is, for in prior art, exist according to people's face change in location, to adjust the method that electronic installation screen position and people's face position adapt comparatively complicated, be difficult to realize pinpoint defect, provide a kind of implementation procedure simple, and can be by adjusting the method for the accurate tracing and positioning people's face of screen position, described method comprises the steps:
S0: by camera collection user face image, extraction storage are distinguished corresponding a plurality of fixed reference feature points with the precalculated position in face image, according to predetermined rule by described a plurality of fixed reference feature points be connected between two form first with reference to line segment, second with reference to line segment, the 3rd with reference to line segment and the 4th with reference to line segment, record this first to fourth length with reference to line segment, and from described a plurality of fixed reference feature points, select a comparison reference point;
S1: again when the camera collection user face image, extract the fact characteristic point in the identical precalculated position in corresponding with described fixed reference feature point face image from the user's face image collecting; Described fact characteristic point is connected to form with described first to fourth between two with reference to the first actual line segment corresponding to line segment, the second actual line segment, the 3rd actual line segment and the 4th actual line segment, and in described fact characteristic point with described compare reference point corresponding the fact characteristic point in identical precalculated position in face image be set to actual specific to point;
S2: relatively whether the ratio of the first actual line segment and the second actual line segment is identical with reference to line segment and the second ratio with reference to line segment with first, if identical, carries out next sub-step; If not identical, adjust display screen and tilt forward and back angle;
S3: relatively whether the ratio of the 3rd actual line segment and the 4th actual line segment is identical with reference to line segment and the 4th ratio with reference to line segment with the 3rd, if identical, carries out next sub-step; If not identical, adjust display screen and horizontally rotate angle;
S4: relatively whether actual specific is identical with the position of comparing reference point to the position of point, if identical, adjusted, if not identical, adjusts display screen vertical height.
In the method for above-mentioned tracing and positioning people's face, described step S1 also comprises: if camera cannot collect all and described fixed reference feature, put identical fact characteristic point, provide the not prompting in setting range of user.
In the method for above-mentioned tracing and positioning people's face, between described step S1 and described step S2, also comprise:
S12: according to horizontal range and the vertical range that relatively actual specific goes out the position of point and the position calculation of comparing reference point, adjust successively display screen and horizontally rotate angle, vertical height and tilt forward and back angle, until meet: actual specific to the position of point with to compare reference point locations identical.
In the method for above-mentioned tracing and positioning people's face, described step S2 also comprises: if cannot when adjusting display screen and tilt forward and back angle and make the ratio of the first actual line segment and the second actual line segment identical with the second ratio with reference to line segment with reference to line segment with first, provide the not prompting in setting range of user;
In the method for above-mentioned tracing and positioning people's face, described step S3 also comprises: if cannot when adjusting display screen and horizontally rotate angle and make the ratio of the 3rd actual line segment and the 4th actual line segment identical with the 4th ratio with reference to line segment with reference to line segment with the 3rd, provide the not prompting in setting range of user;
In the method for above-mentioned tracing and positioning people's face, described step S4 also comprises: if cannot make to compare the position of reference point when identical to the position of putting with actual specific by adjusting display screen vertical height, provide the not prompting in setting range of user.
In the method for above-mentioned tracing and positioning people's face, a plurality of fixed reference feature points that obtain from user's face image of camera collection are end points X0 and the Y0 of user's eyeball position A0 and B0, nose position C0 and user's two corners of the mouths; Actual fixed reference feature point is end points X1 and the Y1 of user's eyeball position A1 and B1, nose position C1 and user's two corners of the mouths;
Wherein, described first is the line segment AB0 that A0 and B0 form with reference to line segment, and second is the line segment XY0 that X0 and Y0 form with reference to line segment, and the 3rd is the line segment AC0 that A0 and C0 form with reference to line segment, the 4th is the line segment BC0 that B0 and C0 form with reference to line segment, and comparison reference point is C0; The first actual line segment is the line segment AB1 that A1 and B1 form, and the second actual line segment is the line segment XY1 that X1 and Y1 form, and the 3rd actual line segment is the line segment AC1 that A1 and C1 form, and the 4th actual line segment is the line segment BC1 that B1 and C1 form, and actual specific is C1 to point.
In the method for above-mentioned tracing and positioning people's face, a plurality of fixed reference feature points that obtain from user's face image of camera collection are user's eyeball position A0 and B0, nose position C0, the end points X0 of user's two corners of the mouths and end points D0 and the E0 of Y0 and user's nose both sides; Fact characteristic point is user's eyeball position A1 and B1, nose position C1, the end points X1 of user's two corners of the mouths and end points D1 and the E1 of Y1 and user's nose both sides;
Wherein, described first is the line segment AB0 that A0 and B0 form with reference to line segment, and second is the line segment XY0 that X0 and Y0 form with reference to line segment, and the 3rd is the line segment AD0 that A0 and D0 form with reference to line segment, the 4th is the line segment BE0 that B0 and E0 form with reference to line segment, and comparison reference point is C0; Described the first actual line segment is the line segment AB1 that A1 and B1 form, and the second actual line segment is the line segment XY1 that X1 and Y1 form, and the 3rd actual line segment is the line segment AD1 that A1 and D1 form, and the 4th actual line segment is the line segment BE1 that B1 and E1 form, and actual specific is C1 to point.
A kind of display system is also provided, comprises display screen, camera, central processing unit and driver module; Described central processing unit comprises feature point extraction unit, the first comparing unit, the second comparing unit and the 3rd comparing unit;
Described camera is used for gathering user's face image, and is transferred to described central processing unit;
Described feature point extraction unit is used for receiving user's face image, extracts and puts identical fact characteristic point with described fixed reference feature; According to predetermined rule, described fact characteristic point is connected between two and is formed with described first to fourth with reference to the first actual line segment corresponding to line segment, the second actual line segment, the 3rd actual line segment and the 4th actual line segment, and in described fact characteristic point with described compare reference point corresponding the fact characteristic point in identical precalculated position in face image be set to actual specific to point; ;
Whether described the first comparing unit is identical with reference to line segment and the second ratio with reference to line segment with first for the ratio of comparison the first actual line segment and the second actual line segment, if not identical, the signal that output tilts forward and back angle for adjusting display screen is given described driver module;
Whether described the second comparing unit is identical with reference to line segment and the 4th ratio with reference to line segment with the 3rd for the ratio of comparison the 3rd actual line segment and the 4th actual line segment, if not identical, the signal that output horizontally rotates angle for adjusting display screen is given described driver module;
Described the 3rd comparing unit is for relatively whether actual specific is identical with the position of comparing reference point to the position of point, if not identical, the signal that output is used for adjusting display screen vertical height is to described driver module;
Described driver module is for rotating and/or lifting according to the signal driver display screen receiving.
Implement method and the display system of tracing and positioning people's face of the present invention, there is following beneficial effect: according in formed at least four the actual line segments of user's face image unique point of extracting, relatively whether the ratio of the first actual line segment and the second actual line segment is identical with reference to line segment and the second ratio with reference to line segment with first, relatively the ratio of the 3rd actual line segment and the 4th actual line segment with the 3rd with reference to line segment and whether the 4th ratio with reference to line segment identical and relatively actual specific to the position of point and compare the whether identical screen position of adjusting in position of reference point, reach the object of accurate tracing and positioning people's face, and implementation procedure is simple, can accurately locate people's face, to user, use and offer convenience.
Accompanying drawing explanation
Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing:
Fig. 1 is the process flow diagram of the method for a kind of tracing and positioning people's face of the present invention;
Fig. 2 is that the critical data of user's face image of providing of the present invention's the first preferred embodiment is put formed characteristic pattern;
Fig. 3 is the process flow diagram that the present invention's the first preferred embodiment provides;
Fig. 4 is that the critical data of user's face image of providing of the present invention's the second preferred embodiment is put formed characteristic pattern;
Fig. 5 is the process flow diagram that the present invention's the second preferred embodiment provides;
Fig. 6 is the structural representation of a kind of display system of the present invention.
Embodiment
For technical characterictic of the present invention, object and effect being had more clearly, understand, now contrast accompanying drawing and describe the specific embodiment of the present invention in detail.
Fig. 1 is the process flow diagram of the method for a kind of tracing and positioning people's face of the present invention, and method comprises the steps: mainly to comprise the steps:
S0: in advance by camera collection user face image, extraction storage are distinguished corresponding a plurality of fixed reference feature points and are stored at least four line segments that a plurality of fixed reference feature points are connected and formed between two according to pre-defined rule with the precalculated position in face image, be set to respectively first with reference to line segment, second with reference to line segment, the 3rd with reference to line segment and the 4th with reference to line segment, record first to fourth length with reference to line segment, and from these a plurality of fixed reference feature points, select a comparison reference point;
S1: again when the camera collection user face image, extract the fact characteristic point of the fact characteristic point in the identical precalculated position in corresponding with described fixed reference feature point face image from the user images collecting; According to the fixed reference feature point relating to reference to line segment, and the fact characteristic point corresponding with it connected, to form with described with reference to the first actual line segment corresponding to line segment, the second actual line segment, the 3rd actual line segment and the 4th actual line segment, and the unique point identical with comparing reference point in described fact characteristic point is set to actual contrast points; The fact characteristic point of extracting will identical with pre-stored fixed reference feature point, formed actual line segment with reference to the corresponding unique point of line segment, also want actual specific identical and that select to putting and compare an also same unique point that is decided to be user's face image of reference point.
S2: relatively whether the ratio of the first actual line segment and the second actual line segment is identical with reference to line segment and the second ratio with reference to line segment with first, if identical, carries out next sub-step; If not identical, adjust display screen and tilt forward and back angle and make it identical; If cannot be when adjusting display screen and tilt forward and back angle and make the ratio of the first actual line segment and the second actual line segment identical with the second ratio with reference to line segment with reference to line segment with first, illustrating that user comes back or bows has exceeded setting range, provides the not prompting in setting range of user.
S3: relatively whether the ratio of the 3rd actual line segment and the 4th actual line segment is identical with reference to line segment and the 4th ratio with reference to line segment with the 3rd, if identical, carries out next sub-step; If not identical, adjust display screen and horizontally rotate angle and make it identical; If cannot be when adjusting display screen and horizontally rotate angle and make the ratio of the 3rd actual line segment and the 4th actual line segment identical with the 4th ratio with reference to line segment with reference to line segment with the 3rd, user is described towards left side or right side and exceeded scope, provides the not prompting in setting range of user.
S4: relatively whether actual specific is identical with the position of comparing reference point to the position of point, if identical, adjusted, if not identical, adjust display screen vertical height and makes it identical.If cannot make to compare the position of reference point when identical to the position of point with actual specific by adjusting display screen vertical height, the too high or too low of user's seat is described and exceeded setting range, provide the not prompting in setting range of user.
In said process, the fixed reference feature point extracting must meet following condition according to the formed line segment of pre-defined rule: two line segments that tilt forward and back angle for adjusting display screen must be two horizontal line sections, the definition of horizontal line section is here: camera collection to user's facial image can be regarded as the rectangle identical with the screen size shape of display screen, form four end points, it is benchmark that the bottom two-end-point of screen of take forms line segment, it is that the line segment forming with the bottom two-end-point of screen is parallel that the fixed reference feature of the user face extracting is put formed line segment, the difference of Dan Yin different user face, can there is certain error, within the scope of certain error, all can think parallel, and when extract minutiae, can choose as far as possible as required the parallel as far as possible point of line segment forming with the bottom two-end-point of screen, for example, while extracting eyeball unique point, can select according to actual needs to depart from the allowed band of eyeball center point also can.Formed two line segments of unique point that horizontally rotate angle for adjusting display screen are to have two line segments that have certain angle, but the parallel line segment of line segment that all can not form for the bottom two-end-point with screen, such line segment can be defined as the line segment of two inclinations.
The line segment that the line segment of angle is two inclinations if tilt forward and back for adjusting display screen, the proportionate relationship of these two line segments is subject to tilting forward and back the impact that angle changes and is horizontally rotated again the impact that angle changes, if select two horizontal line sections, these two line segments are only tilted forward and back the impact of angle so, when this being tilted forward and back after angle adjusts, any two inclined line segment ratios that all the other form an angle all can be used as to adjust and horizontally rotate the foundation of angle, but can be preferably one group of vertical line segment.
Described step S1 also comprises: if the user's face image arriving according to camera collection, if cannot extract all and described fixed reference feature by existing face characteristic point extracting method, put identical fact characteristic point, the position that user is described has exceeded setting range, for example, user's side face, or it is far away to depart from display screen, can regard as in this case user and now not want to use the electronic installation that comprises this display screen as computer, provide the not prompting in setting range of user.
In addition between step S1 and step S2, also comprise:
S12: according to horizontal range and the vertical range that relatively actual specific goes out the position of point and the position calculation of comparing reference point, adjust successively display screen and horizontally rotate angle, vertical height and tilt forward and back angle, until meet: actual specific to the position of point with to compare reference point locations identical.
Apparently, camera, after receiving location requirement, gathers user's face image, and now user's face location is determined, the position and the actual line segment data that affect fact characteristic point are no longer user's face location, but the position of display screen.Each adjust display screen tilt forward and back angle horizontally rotate angle or vertical height after, camera all can Resurvey user face image, extract same fact characteristic point and form same actual line segment, therefore each fact characteristic point position or actual line segment data of extracting from face image after the position of display screen adjusted all can make corresponding changes until meet the demands.
Fig. 2 puts formed characteristic pattern for the critical data of user's face image that the present invention's the first preferred embodiment provides, first by user, select suitable position, by the camera collection user face image corresponding with precalculated position, a plurality of fixed reference feature points of the appointment of extracting by existing features of human face images extracting method are respectively user's eyeball position A0 and B0, end points X0 and the Y0 of nose position C0 and two corners of the mouths, according to these five fixed reference feature points, form four line segments, wherein, AB0 be fixed reference feature point A0 and B0 form first with reference to line segment, XY0 be fixed reference feature point X0 and Y0 form second with reference to line segment, AC0 be fixed reference feature point A0 and C0 form the 3rd with reference to line segment, BC0 be fixed reference feature point B0 and C0 form the 4th with reference to line segment, the comparison reference point of C0 for arranging.
Afterwards, at camera, receive after location requirement, gather user's face image, by existing features of human face images extracting method, extract and put identical fact characteristic point with fixed reference feature, this fact characteristic point is also user's eyeball position A1 and B1, end points X1 and the Y1 of nose position C1 and two corners of the mouths, put included user's facial feature points with the fixed reference feature of pre-stored and number is on all four, according to these five critical data points, form four line segments, wherein, AB1 is the first actual line segment that fact characteristic point A1 and B1 form, XY1 is the second actual line segment that fact characteristic point X1 and Y1 form, AC1 is the 3rd actual line segment that fact characteristic point A1 and C1 form, BC1 is the 4th actual line segment that fact characteristic point B1 and C1 form.The actual specific that C1 is setting is to point.
Wherein, for comparing the ratio of the first actual line segment and the second actual line segment, with first with reference to the line segment standard whether identical with the second ratio with reference to line segment, be: AB1/XY1=AB0/XY0, for comparing the ratio of the 3rd actual line segment and the 4th actual line segment, with the 3rd with reference to the line segment standard whether identical with the 4th ratio with reference to line segment, be: AD1/BE1=AD0/BE0, for comparing actual specific to the position standard whether identical with the position of comparing reference point of point is: C1 overlaps with C0.
As shown in Figure 3, be the process flow diagram that the present invention's the first preferred embodiment provides, the method flow diagram providing according to Fig. 1 can be summarized as following steps:
S0 ': by camera collection user face image, the fixed reference feature point that extracts and store is A0, B0, C0, X0, Y0 in advance; Four that form is AB0, XY0, AC0 and AD0 with reference to line segment, and comparison reference point is C0;
S1 ': camera is received after location requirement, again gathers user's face image, extraction fact characteristic point is A1, B1, C1, X1, Y1; Four actual line segments that form are AB1, XY1, AC1 and AD1, and actual specific is C1 to point;
S2 ': more whether meet: AB1/XY1=AB0/XY0; If meet, carry out next sub-step, if do not meet, adjust display screen and tilt forward and back angle, until meet this condition; If cannot when adjusting display screen and tilt forward and back angle and make it meet AB1/XY1=AB0/XY0, provide the not prompting in setting range of user;
S3 ': more whether meet: AC1/BC1=AC0/BC0; If meet, carry out next sub-step, if do not meet, adjust display screen and horizontally rotate angle until meet this condition; If cannot when adjusting display screen and horizontally rotate angle and make it meet AC1/BC1=AC0/BC0, provide the not prompting in setting range of user;
S4 ': more whether meet: C1 overlaps with C0, is that the position of C1 is identical with the position of C0; If meet, adjusted, without again display screen being adjusted, if do not meet, adjust display screen vertical height until meet this condition; If cannot when adjusting display screen vertical height and make it meet C1 and overlap with C0, provide the not prompting in setting range of user.
In above-mentioned steps, AB1/XY1=AB0/XY0, AC1/BC1=AC0/BC0 and C1 overlap with C0, and it is exactly equate and overlap within the specific limits that the equal and coincidence here can be defined as, and allows certain deviation, can set as required.Step S2 ' and step S3 ' are for more whether needing to adjust the angle of display screen, can change both orders, only need to meet the last display screen height of adjusting.
Between step S1 ' and step S2 ', also comprise step:
S12 ': at first according to actual specific to a C1 with compare the position of reference Point C 0, adjust successively display screen and horizontally rotate angle, vertical height and tilt forward and back angle, until meet: C1 overlaps with C0.This step is tentatively to adjust (also can be described as coarse adjustment), this step is the position of the pixel C1 that relatively photographs and C0 and the level, the vertical range x that calculate, y, the specific algorithm of this horizontal range, vertical range is: take and compare reference Point C 0 as initial point, the parallel straight line of line segment forming with the bottom two-end-point of screen is x axle, and perpendicular straight line is y axle.This x value is directly used for to adjust display screen and horizontally rotates angle, y value is used for adjusting display screen vertical height, if exceeded height control, adjusts display screen again and tilts forward and back angle.
The object of this step is that people's face is placed on to the relatively middle position of shooting, extract like this fact characteristic point and form AB1, more accurate during the data of the line segments such as AC1, because the edge of shooting easily produces distortion, but also can raising efficiency, the Yu Renlian position, position that can adjust quickly display screen adapts, to reach the object of tracing and positioning people face.
In above-mentioned steps S1 ', after receiving location requirement, if camera collection to user's face image in, cannot extract and allly put identical fact characteristic point with fixed reference feature, be equivalent to fact characteristic point not all in face image, illustrate that user's position has exceeded setting range, side face for example, or it is far away to depart from display screen, therefore can assert that in this case user does not intend to use this display screen, now provide the not prompting in setting range of user.
Therefore, the detailed process of implementing the present invention's the first preferred embodiment is as follows: by camera collection user face image, the user face fact characteristic point of extracting is A1, B1, C1, X1, Y1, first judges whether to extract all these unique points, if can not, be these fact characteristic point not all in face image, be that user's position has exceeded setting range, for example, user's side face, or it is far away to depart from display screen, this situation can assert that user does not intend to use this display screen, now provides the not prompting in setting range of user.
Then relatively whether AB1/XY1 equals AB0/XY0, equal can being defined as within the specific limits here just equates, this is for judging the angle of people's face and display screen vertical direction, by adjusting the angle that tilts forward and back of display screen, complete, for example, when user comes back, position near display screen can become large relatively, so AB1/XY1 is less than AB0/XY0, now need to adjust display screen and tilt forward and back angle, thereby change the data of line segment AB1 or XY1, so that the value of AB1/XY1 equals or levels off to AB0/XY0, if so cannot adjust to the position of requirement, illustrating that user comes back or bows has exceeded setting range, provide the not prompting in setting range of user, when user bows, AB1/XY1 is greater than AB0/XY0, adjust display screen and tilt forward and back angle, change the data of line segment AB1 or XY1, to complete the adjustment that tilts forward and back angle.
Equally, relatively whether AC1/BC1 equals AC0/BC0, equal can being defined as within the specific limits here just equates, this is for judging the angle of people's face and display screen horizontal direction, by adjusting display screen, horizontally rotates angle, can change the data of line segment AC1 or BC1, to complete the adjustment that horizontally rotates angle, if cannot adjust to the position of requirement, user is described towards left side or right side and exceeded setting range, provide the not prompting in setting range of user.Finally confirm whether C1 overlaps with C0, the coincidence here can be defined as within the specific limits and overlap, at this moment only need to adjust display screen vertical height, by adjusting the lifting/lowering of display screen, can change the position data of C1, it is overlapped with C0, if cannot adjust to the position of requirement, illustrate that user sits too high or too low and exceeded setting range, provides the not prompting in setting range of user.
Fig. 4 puts formed characteristic pattern for the critical data of user's face image that the present invention's the second preferred embodiment provides, first by user, select suitable position, by the camera collection user face image corresponding with precalculated position, a plurality of fixed reference feature points of the appointment of extracting by existing features of human face images extracting method are respectively the terminal A 0 ' at user's eyeball position A0 and B0 or two canthus and the terminal A 0 ' at B0 ' or two canthus ' and B0 ' ', nose position C0, end points X0 and the Y0 of the end points D0 of nose both sides and E0 and two corners of the mouths, as shown in the figure, according to these seven fixed reference feature points, form four line segments, wherein, take eyeball position A0 and B0 as example, AB0 be fixed reference feature point A0 and B0 form first with reference to line segment, XY0 be fixed reference feature point X0 and Y0 form second with reference to line segment, AD0 be fixed reference feature point A0 and D0 form the 3rd with reference to line segment, BE0 be fixed reference feature point B0 and E0 form the 4th with reference to line segment, the comparison reference point of C0 for arranging.
Afterwards, at camera, receive after location requirement, Real-time Collection user's face image, by existing features of human face images extracting method, extract and put identical fact characteristic point with fixed reference feature, this fact characteristic point is also the terminal A 1 ' at user's eyeball position A1 and the terminal A 1 ' at B1 or two canthus and B1 ' or two canthus ' and B1 ' ', nose position C1, end points X1 and the Y1 of the end points D1 of nose both sides and E1 and two corners of the mouths, put included user's facial feature points with the fixed reference feature of pre-stored and number is on all four, according to these seven critical data points, can form four line segments, wherein, take eyeball position A1 and B1 as example, AB1 is the first actual line segment that fact characteristic point A1 and B1 form, XY1 is the second actual line segment that fact characteristic point X1 and Y1 form, AD1 is the 3rd actual line segment that fact characteristic point A1 and D1 form, BE1 is the 4th actual line segment that fact characteristic point B1 and E1 form, the actual specific that C1 is setting is to point.
Wherein, for comparing the ratio of the first actual line segment and the second actual line segment, with first with reference to the line segment standard whether identical with the second ratio with reference to line segment, be: AB1/XY1=AB0/XY0, for comparing the ratio of the 3rd actual line segment and the 4th actual line segment, with the 3rd with reference to the line segment standard whether identical with the 4th ratio with reference to line segment, be: AD1/BE1=AD0/BE0, for comparing actual specific to the position standard whether identical with the position of comparing reference point of point is: C1 overlaps with C0.
As shown in Figure 5, be the process flow diagram of the second preferred embodiment provided by the invention, implementation step is the same with Fig. 3 principle, repeats no more herein.
The detailed process of now introducing enforcement the present invention the second preferred embodiment is as follows: by camera collection user face image, the user face fact characteristic point of extracting is A1, B1, C1, D1, E1, X1, Y1, first judge whether to extract all these unique points, if can not, these fact characteristic point are not all in face image, the position that is user has exceeded setting range, for example, user's side face, or it is far away to depart from display screen, this situation can assert that user does not intend to use this display screen, now provides the not prompting in setting range of user.
Then relatively whether AB1/XY1 equals AB0/XY0, equal can being defined as within the specific limits here just equates, this is for judging the angle of facial image and display screen vertical direction, by adjusting the angle that tilts forward and back of display screen, complete, for example, when user comes back, position near display screen can become large relatively, so AB1/XY1 is less than AB0/XY0, now need to adjust display screen and tilt forward and back angle, change the data of line segment AB1 or XY1, so that the value of AB1/XY1 equals or levels off to AB0/XY0, if so cannot adjust to the position of requirement, illustrating that user comes back or bows has exceeded setting range, provide the not prompting in setting range of user, when user bows, AB1/XY1 is greater than AB0/XY0, adjust screen position, change the data of line segment AB1 or XY1, to complete the adjustment that tilts forward and back angle.
Equally, relatively whether AD1/BE1 equals AD0/BE0, equal can being defined as within the specific limits here just equates, this is for judging the angle of people's face and display screen horizontal direction, by adjusting the angle that horizontally rotates of display screen, complete, by adjusting screen position, change the data of line segment AD1 or BE1, until equal or close to the ratio of AD0 and BE0, if cannot adjust to the position of requirement, user is described towards left side or right side and exceeded setting range, provides the not prompting in setting range of user.Finally confirm whether C1 overlaps with C0, the coincidence here can be defined as within the specific limits and overlap, and at this moment only needs to adjust display screen vertical height, by adjusting the vertical high data of display screen, can change the position data of C1, until overlap with C0.If cannot adjust to the position of requirement, illustrate that user sits too high or too low and exceeded setting range, provides the not prompting in setting range of user.
In the second preferred embodiment, the line segment AB0 that can also form by the fixed reference feature point that relatively whether the line segment AB1 of fact characteristic point formation and the ratio AB1/DE1 of DE1 equal to prestore and the ratio AB0/DE0 of DE0 are used as adjusting the standard that display screen tilts forward and back angle.Wherein, DE1 is the line segment that fact characteristic point D1 and E1 form, and DE0 is the line segment of D0 and E0 formation in fixed reference feature point.And the eyeball position A1 in fact characteristic point and B1 can be by terminal A 1 ' and B1 ' or the two canthus terminal A 1 ' at two canthus ' and B1 ' ' replace, all can realize the object of tracing and positioning people face.
In addition, also be not limited to above-mentioned several implementation, wherein, user's eyeball position A1 and B1, the end points X1 of nose position C1 and two corners of the mouths and Y1 capture than being easier to, easily explanation, hold intelligible user's facial feature points, but the unique point of people's face also comprise the tip of the brow and people medium, according to features of human face images grasping means, all can capture, only need to form at least two horizontal line sections, whether need as a comparison to adjust the foundation that display screen tilts forward and back angle, form two inclined line segments that form an angle, whether need as a comparison to adjust the foundation that display screen horizontally rotates angle, can be with image characteristic point as nose, whether places between the eyebrows etc. need to adjust the foundation of display screen vertical height as a comparison, but because nose relatively approaches people's face middle, adjustment process can be more simple and accurate.In a word, except above-mentioned these several examples of enumerating, also have more embodiment, do not repeat them here.
The method of implementing a kind of tracing and positioning people's face of the present invention, by coordinating certain hardware configuration, capable of regulating screen position and people's face position adapt, and realize tracing and positioning people face.As shown in Figure 6, this display system comprises display screen, also comprises camera 20, central processing unit 30 and driver module 40; Central processing unit 30 comprises feature point extraction unit 301, the first comparing unit 302, the second comparing unit 303 and the 3rd comparing unit 304.Wherein, central processing unit 30 is when user selects correct position, by camera collection user face image, extract in advance and store with the precalculated position in face image and distinguish corresponding a plurality of fixed reference feature points, according to predetermined rule, a plurality of reference point are connected to form to first to fourth between two with reference to line segment, and from these a plurality of fixed reference feature points, select a comparison reference point.
Camera 20 is installed on display screen, for gathering user's face image, and is transferred to central processing unit 30.
Feature point extraction unit 301, for receiving user's face image, extracts and puts identical fact characteristic point with fixed reference feature; According to the fixed reference feature point relating to reference to line segment, and the fact characteristic point corresponding with it connected, with form with reference to the first actual line segment corresponding to line segment, the second actual line segment, the 3rd actual line segment and the 4th actual line segment, and the unique point identical with comparing reference point in fact characteristic point is set to actual specific to point.
Whether the first comparing unit 302 is identical with reference to line segment and the second ratio with reference to line segment with first for the ratio of relatively the first actual line segment and the second actual line segment, if not identical, output is used for adjusting signal that display screen tilts forward and back angle to driver module 40.
Whether the second comparing unit 303 is identical with reference to line segment and the 4th ratio with reference to line segment with the 3rd for the ratio of relatively the 3rd actual line segment and the 4th actual line segment, if not identical, output is used for adjusting signal that display screen horizontally rotates angle to driver module 40.
The 3rd comparing unit 304 is for relatively whether actual specific is identical with the position of comparing reference point to the position of point, if not identical, the signal that output is used for adjusting display screen vertical height is to driver module 40.Its comparison and method of adjustment all realize according to above-mentioned tracing and positioning people's face method.
Driver module 40 is for rotating and/or lifting according to the signal driver display screen receiving.
Particularly, driver module 40 is by the signal that tilts forward and back angle for adjusting display screen receiving, horizontally rotate the signal of angle and export respectively the first to the 3rd motor to for adjusting the signal of display screen vertical height for adjusting display screen, and then adjust respectively display screen and tilt forward and back angle, horizontally rotate angle and vertical height, final and people's face position, the position of display screen is adapted, reach the object of tracing and positioning people face.
Method and the display system of a kind of tracing and positioning people's face of the present invention, be mainly used in the electronic installation of single camera, because of in existing electronic product, all-in-one for example, display, mostly the products such as mobile phone are to be furnished with a camera, therefore have very strong practicality, implementation method is simple, and can accurately locate people's face.
By reference to the accompanying drawings embodiments of the invention are described above; but the present invention is not limited to above-mentioned embodiment; above-mentioned embodiment is only schematic; rather than restrictive; those of ordinary skill in the art is under enlightenment of the present invention; not departing from the scope situation that aim of the present invention and claim protect, also can make a lot of forms, within these all belong to protection of the present invention.
Claims (9)
1. a method for tracing and positioning people face, is characterized in that, described method comprises the steps:
S0: by camera collection user face image, extraction storage are distinguished corresponding a plurality of fixed reference feature points with the precalculated position in face image, according to predetermined rule by described a plurality of fixed reference feature points be connected between two form first with reference to line segment, second with reference to line segment, the 3rd with reference to line segment and the 4th with reference to line segment, record this first to fourth length with reference to line segment, and from described a plurality of fixed reference feature points, select a comparison reference point;
S1: again when the camera collection user face image, extract the fact characteristic point in the identical precalculated position in corresponding with described fixed reference feature point face image from the user's face image collecting; Described fact characteristic point is connected to form with described first to fourth between two with reference to the first actual line segment corresponding to line segment, the second actual line segment, the 3rd actual line segment and the 4th actual line segment, and in described fact characteristic point with described compare reference point corresponding the fact characteristic point in identical precalculated position in face image be set to actual specific to point;
S2: relatively whether the ratio of the first actual line segment and the second actual line segment is identical with reference to line segment and the second ratio with reference to line segment with first, if identical, carries out next sub-step; If not identical, adjust display screen and tilt forward and back angle;
S3: relatively whether the ratio of the 3rd actual line segment and the 4th actual line segment is identical with reference to line segment and the 4th ratio with reference to line segment with the 3rd, if identical, carries out next sub-step; If not identical, adjust display screen and horizontally rotate angle;
S4: relatively whether actual specific is identical with the position of comparing reference point to the position of point, if identical, adjusted, if not identical, adjusts display screen vertical height.
2. according to the method for tracing and positioning people's face according to claim 1, it is characterized in that, described step S1 also comprises: if camera cannot collect all and described fixed reference feature, put identical fact characteristic point, provide the not prompting in setting range of user.
3. according to the method for tracing and positioning people's face according to claim 1, it is characterized in that, between described step S1 and described step S2, also comprise:
S12: according to horizontal range and the vertical range that relatively actual specific goes out the position of point and the position calculation of comparing reference point, adjust successively display screen and horizontally rotate angle, vertical height and tilt forward and back angle, until meet: actual specific to the position of point with to compare reference point locations identical.
4. the method for tracing and positioning people's face according to claim 1, it is characterized in that, described step S2 also comprises: if cannot when adjusting display screen and tilt forward and back angle and make the ratio of the first actual line segment and the second actual line segment identical with the second ratio with reference to line segment with reference to line segment with first, provide the not prompting in setting range of user.
5. the method for tracing and positioning people's face according to claim 1, it is characterized in that, described step S3 also comprises: if cannot when adjusting display screen and horizontally rotate angle and make the ratio of the 3rd actual line segment and the 4th actual line segment identical with the 4th ratio with reference to line segment with reference to line segment with the 3rd, provide the not prompting in setting range of user.
6. the method for tracing and positioning people's face according to claim 1, it is characterized in that, described step S4 also comprises: if cannot make to compare the position of reference point when identical to the position of putting with actual specific by adjusting display screen vertical height, provide the not prompting in setting range of user.
7. the method for tracing and positioning people's face according to claim 1, is characterized in that, a plurality of fixed reference feature points that extract from user's face image of camera collection are end points X0 and the Y0 of user's eyeball position A0 and B0, nose position C0 and user's two corners of the mouths; Actual fixed reference feature point is end points X1 and the Y1 of user's eyeball position A1 and B1, nose position C1 and user's two corners of the mouths;
Wherein, described first is the line segment AB0 that A0 and B0 form with reference to line segment, and second is the line segment XY0 that X0 and Y0 form with reference to line segment, and the 3rd is the line segment AC0 that A0 and C0 form with reference to line segment, the 4th is the line segment BC0 that B0 and C0 form with reference to line segment, and comparison reference point is C0; The first actual line segment is the line segment AB1 that A1 and B1 form, and the second actual line segment is the line segment XY1 that X1 and Y1 form, and the 3rd actual line segment is the line segment AC1 that A1 and C1 form, and the 4th actual line segment is the line segment BC1 that B1 and C1 form, and actual specific is C1 to point.
8. the method for tracing and positioning people's face according to claim 1, it is characterized in that, a plurality of fixed reference feature points that obtain from user's face image of camera collection are user's eyeball position A0 and B0, nose position C0, the end points X0 of user's two corners of the mouths and end points D0 and the E0 of Y0 and user's nose both sides; Fact characteristic point is user's eyeball position A1 and B1, nose position C1, the end points X1 of user's two corners of the mouths and end points D1 and the E1 of Y1 and user's nose both sides;
Wherein, described first is the line segment AB0 that A0 and B0 form with reference to line segment, and second is the line segment XY0 that X0 and Y0 form with reference to line segment, and the 3rd is the line segment AD0 that A0 and D0 form with reference to line segment, the 4th is the line segment BE0 that B0 and E0 form with reference to line segment, and comparison reference point is C0; Described the first actual line segment is the line segment AB1 that A1 and B1 form, and the second actual line segment is the line segment XY1 that X1 and Y1 form, and the 3rd actual line segment is the line segment AD1 that A1 and D1 form, and the 4th actual line segment is the line segment BE1 that B1 and E1 form, and actual specific is C1 to point.
9. a display system, is characterized in that, comprises display screen, camera, central processing unit and driver module; Described central processing unit comprises feature point extraction unit, the first comparing unit, the second comparing unit and the 3rd comparing unit;
Described camera is used for gathering user's face image, and is transferred to described central processing unit;
Described feature point extraction unit is used for receiving user's face image, extracts the fact characteristic point in the identical precalculated position in corresponding with fixed reference feature point face image; According to predetermined rule, described fact characteristic point is connected between two and is formed with first to fourth with reference to the first actual line segment corresponding to line segment, the second actual line segment, the 3rd actual line segment and the 4th actual line segment, and in described fact characteristic point with compare reference point corresponding the fact characteristic point in identical precalculated position in face image be set to actual specific to point;
Whether described the first comparing unit is identical with reference to line segment and the second ratio with reference to line segment with first for the ratio of comparison the first actual line segment and the second actual line segment, if not identical, the signal that output tilts forward and back angle for adjusting display screen is given described driver module;
Whether described the second comparing unit is identical with reference to line segment and the 4th ratio with reference to line segment with the 3rd for the ratio of comparison the 3rd actual line segment and the 4th actual line segment, if not identical, the signal that output horizontally rotates angle for adjusting display screen is given described driver module;
Described the 3rd comparing unit is for relatively whether actual specific is identical with the position of comparing reference point to the position of point, if not identical, the signal that output is used for adjusting display screen vertical height is to described driver module;
Described driver module is for rotating and/or lifting according to the signal driver display screen receiving.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410002101.2A CN103760975B (en) | 2014-01-02 | 2014-01-02 | A kind of method of tracing and positioning face and display system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410002101.2A CN103760975B (en) | 2014-01-02 | 2014-01-02 | A kind of method of tracing and positioning face and display system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103760975A true CN103760975A (en) | 2014-04-30 |
| CN103760975B CN103760975B (en) | 2017-01-04 |
Family
ID=50528227
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410002101.2A Active CN103760975B (en) | 2014-01-02 | 2014-01-02 | A kind of method of tracing and positioning face and display system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103760975B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105234940A (en) * | 2015-10-23 | 2016-01-13 | 上海思依暄机器人科技有限公司 | Robot and control method thereof |
| CN105242888A (en) * | 2014-07-10 | 2016-01-13 | 联想(北京)有限公司 | System control method and electronic device |
| CN105301789A (en) * | 2015-12-09 | 2016-02-03 | 深圳市中视典数字科技有限公司 | Stereoscopic display device following human eye positions |
| CN105701763A (en) * | 2015-12-30 | 2016-06-22 | 青岛海信移动通信技术股份有限公司 | Method and device for adjusting face image |
| CN106955122A (en) * | 2017-03-20 | 2017-07-18 | 苏州网颢信息科技有限公司 | Head 3 D detection method based on single camera module in radiotherapy |
| CN108134902A (en) * | 2017-12-25 | 2018-06-08 | 三星电子（中国）研发中心 | A kind of method and apparatus for the photo for shooting same background |
| CN109995994A (en) * | 2018-01-02 | 2019-07-09 | 上海银晨智能识别科技有限公司 | Camera system and its control device, method, the readable storage medium storing program for executing of automatic adjustment |
| CN110909662A (en) * | 2019-11-20 | 2020-03-24 | 湖南检信智能科技有限公司 | Face motion blur correction method |
| WO2020087751A1 (en) * | 2018-10-29 | 2020-05-07 | 歌尔股份有限公司 | Directional display method and apparatus for audio device and audio device |
| CN112869701A (en) * | 2021-01-11 | 2021-06-01 | 上海微创医疗机器人（集团）股份有限公司 | Sight line detection method, surgical robot system, control method, and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101458531A (en) * | 2007-12-12 | 2009-06-17 | 深圳富泰宏精密工业有限公司 | Display screen automatic adjustment system and method |
| CN101859176A (en) * | 2009-04-08 | 2010-10-13 | 群康科技(深圳)有限公司 | Screen control device and method thereof |
| US20100295827A1 (en) * | 2009-05-25 | 2010-11-25 | Lg Electronics Inc. | Display device and method of adjusting viewing angle thereof |
| CN102547209A (en) * | 2012-02-06 | 2012-07-04 | 华为技术有限公司 | Method and device for controlling videoconferencing equipment, and videoconferencing system |
| US8379059B2 (en) * | 2009-08-07 | 2013-02-19 | Fih (Hong Kong) Limited | Portable electronic device and method for adjusting display orientation of the portable electronic device |
| CN103092344A (en) * | 2013-01-11 | 2013-05-08 | 深圳市金立通信设备有限公司 | Method of controlling terminal screen to rotate and terminal |
-
2014
- 2014-01-02 CN CN201410002101.2A patent/CN103760975B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101458531A (en) * | 2007-12-12 | 2009-06-17 | 深圳富泰宏精密工业有限公司 | Display screen automatic adjustment system and method |
| CN101859176A (en) * | 2009-04-08 | 2010-10-13 | 群康科技(深圳)有限公司 | Screen control device and method thereof |
| US20100295827A1 (en) * | 2009-05-25 | 2010-11-25 | Lg Electronics Inc. | Display device and method of adjusting viewing angle thereof |
| US8379059B2 (en) * | 2009-08-07 | 2013-02-19 | Fih (Hong Kong) Limited | Portable electronic device and method for adjusting display orientation of the portable electronic device |
| CN102547209A (en) * | 2012-02-06 | 2012-07-04 | 华为技术有限公司 | Method and device for controlling videoconferencing equipment, and videoconferencing system |
| CN103092344A (en) * | 2013-01-11 | 2013-05-08 | 深圳市金立通信设备有限公司 | Method of controlling terminal screen to rotate and terminal |
Non-Patent Citations (1)
| Title |
|---|
| 吴巾一: "《人脸识别方法综述》", 《计算机应用研究》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105242888B (en) * | 2014-07-10 | 2018-10-12 | 联想(北京)有限公司 | A kind of system control method and electronic equipment |
| CN105242888A (en) * | 2014-07-10 | 2016-01-13 | 联想(北京)有限公司 | System control method and electronic device |
| CN105234940A (en) * | 2015-10-23 | 2016-01-13 | 上海思依暄机器人科技有限公司 | Robot and control method thereof |
| CN105301789A (en) * | 2015-12-09 | 2016-02-03 | 深圳市中视典数字科技有限公司 | Stereoscopic display device following human eye positions |
| CN105701763A (en) * | 2015-12-30 | 2016-06-22 | 青岛海信移动通信技术股份有限公司 | Method and device for adjusting face image |
| CN105701763B (en) * | 2015-12-30 | 2019-05-21 | 青岛海信移动通信技术股份有限公司 | A kind of method and device adjusting facial image |
| CN106955122A (en) * | 2017-03-20 | 2017-07-18 | 苏州网颢信息科技有限公司 | Head 3 D detection method based on single camera module in radiotherapy |
| CN108134902A (en) * | 2017-12-25 | 2018-06-08 | 三星电子（中国）研发中心 | A kind of method and apparatus for the photo for shooting same background |
| CN108134902B (en) * | 2017-12-25 | 2020-09-18 | 三星电子（中国）研发中心 | Method and device for shooting pictures with same background |
| CN109995994A (en) * | 2018-01-02 | 2019-07-09 | 上海银晨智能识别科技有限公司 | Camera system and its control device, method, the readable storage medium storing program for executing of automatic adjustment |
| WO2020087751A1 (en) * | 2018-10-29 | 2020-05-07 | 歌尔股份有限公司 | Directional display method and apparatus for audio device and audio device |
| US11551633B2 (en) | 2018-10-29 | 2023-01-10 | Goeriek Inc. | Directional display method and apparatus for audio device and audio device |
| CN110909662A (en) * | 2019-11-20 | 2020-03-24 | 湖南检信智能科技有限公司 | Face motion blur correction method |
| CN112869701A (en) * | 2021-01-11 | 2021-06-01 | 上海微创医疗机器人（集团）股份有限公司 | Sight line detection method, surgical robot system, control method, and storage medium |
| CN112869701B (en) * | 2021-01-11 | 2024-03-29 | 上海微创医疗机器人（集团）股份有限公司 | Sight line detection method, surgical robot system, control method, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103760975B (en) | 2017-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103760975A (en) | Method for tracking and positioning faces and display system | |
| US20110074822A1 (en) | Viewing Direction Determination Method, Viewing Direction Determination Apparatus, Image Processing Method, Image Processing Apparatus, Display Device and Electronic Device | |
| US20180225842A1 (en) | Method and apparatus for determining facial pose angle, and computer storage medium | |
| JP6211536B2 (en) | Apparatus and method for controlling rotation of display image | |
| US9195881B2 (en) | Image transformation apparatus and method | |
| CN103049084A (en) | Electronic device and method for adjusting display direction according to face direction | |
| CN105389491A (en) | Facial recognition authentication system including path parameters | |
| EP1501019A2 (en) | Apparatus and method for automatically compensating for an image gradient of a mobile communication terminal | |
| CN106295533A (en) | Optimization method, device and the camera terminal of a kind of image of autodyning | |
| CN103189899B (en) | Object display apparatus and object displaying method | |
| CN106250839A (en) | A kind of iris image perspective correction method, device and mobile terminal | |
| CN106600855A (en) | Payment device and method based on face recognition | |
| US20160062455A1 (en) | Electronic device and method for adjusting brightness of display screen | |
| WO2018058311A1 (en) | Control method, control device, and electronic device | |
| CN104216509B (en) | A kind of electronic equipment and information processing method | |
| CN108569214A (en) | The display system and its driving method of vehicle | |
| CN102043942A (en) | Visual direction judging method, image processing method, image processing device and display device | |
| US20190205635A1 (en) | Process for capturing content from a document | |
| CN109782913A (en) | A kind of method and device that control screen content is shown | |
| US20120293557A1 (en) | Object tracking apparatus, interactive image display system using object tracking apparatus, and methods thereof | |
| US8433444B2 (en) | Mirror and adjustment method therefor | |
| CN109752870A (en) | Electrophoresis electronic paper ghost detection system and detection method | |
| CN109840453B (en) | Face matching method and device | |
| CN104461002A (en) | Device and method for reducing vision fatigue of viewing mobile terminal equipment | |
| US20120026617A1 (en) | Mirror and adjustment method therefor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |