Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art obtain under not making creative work premise, broadly fall into the scope of protection of the invention.
The embodiment of the present invention provides the control method of a kind of starting up of terminal, as it is shown in figure 1, include:
S101, by each shooting moment in the N number of shooting moment in Preset Time, the image of what the first photographic head and second camera shot respectively simultaneously comprise gesture operation synthesizes N number of three-dimensional image;
S102, the human body contour outline extracted in the three-dimensional image corresponding with the first image, wherein, described first image is any one image of the first photographic head shooting described in N number of shooting moment;
S103, on the three-dimensional image corresponding with described first image, obtain the range information corresponding with at least one pixel in described human body contour outline;
Whether the difference between the range information that at least one pixel that in N number of three-dimensional image described in S104, comparison, each three-dimensional image is determined is corresponding is in preset threshold range;
S105, if so, then generate start-up command, and control terminal according to described start-up command and open.
nullThe embodiment of the present invention provides a kind of starting up of terminal control method and system,By at least one the image comprising human body that the first photographic head and second camera synchronization shoot respectively is synthesized three-dimensional image,And three-dimensional image corresponding to two dimensional image based on described first photographic head shooting obtains the range information that in human body contour outline, at least one pixel is corresponding,Whether the difference between the range information that relatively pixel of described N number of three-dimensional image is determined afterwards is in preset threshold range,Before can determine that whether user is in terminal by this result of the comparison,And if the difference of N number of range information is in certain threshold range,Then illustrate that position and relative and terminal the distance of user do not have saltus step,Such as can determine whether out whether user is look at TV and keeps a period of time,Relative improves the accuracy identifying that customer location generates start-up command,Namely when the difference of the plurality of range information meets preset threshold range,Can determine that user has the intention of viewing terminal such as TV,And further generate start-up command,Control terminal to open,Compared with prior art,Eliminate mode that whether infrared detection technology perception have users easily by surrounding environment influence,Accuracy of identification and the problem such as sensitivity is poor,The control method of this starting up of terminal and system are found range by dual camera three-dimensional information and in conjunction with human bioequivalence algorithm,Can determine that the start of user is intended to,Automatic control terminal is opened,Ensure that high real-time simultaneously,In high precision,The manipulation increasing substantially user is experienced.
A kind of processor that executive agent is terminal of the control method of the starting up of terminal of the embodiment of the present invention, this terminal can be TV, computer etc., this is not construed as limiting by the embodiment of the present invention, this first photographic head and second camera are for obtaining the image of human body, and this first photographic head and second camera can be the photographic head arranged in terminal.
In the embodiment of the present invention, whether this first photographic head and second camera sensing user be before terminal, first photographic head and second camera can periodically shoot several photos and carry out human bioequivalence, if before finding that user occurs in photographic head, obtain at least one the image comprising user's human body, user can be static, can also be mobile, additionally, it is manually entered user also by user and moves the start information controlling terminal, start user move the startup button of identification technology as user presses to arrange in terminal remote control, after getting the enabled instruction that described startup button triggers again, processor controls described first photographic head and second camera obtains at least one the image that user moves.
Wherein, Preset Time refers to be needed monitoring user and carries out the time that n times shooting is required, and Preset Time can also set in advance, can by as described in Preset Time be set to 2s-5s;The intervalometer that specifically can pass through to be arranged in described processor is to realize.Within the time period of 2s-5s, the image containing human body got is buffered in the memorizer of terminal by the sequencing obtained, when needs identify, obtained from memorizer by processor, owing to the first photographic head and second camera can shoot 10 ~ 60 picture frames in 1s, preferably, it is 25 ~ 30 picture frames, the human body shot due to the first photographic head and second camera is probably a dynamic process, therefore each two field picture frame is discrepant, therefore when selecting synthesis three-dimensional image, by choosing the two field picture that the first photographic head and second camera shoot at synchronization, difference between three-dimensional image and the actual user's gesture that so can avoid the formation of, improve identification accuracy.If user selects to stand still, then the first second camera can only shoot one in Preset Time or shoot multiple selections one input basis as follow-up identification process.
Wherein, optionally, shooting performance according to photographic head, M shooting moment is altogether comprised in Preset Time, each shooting moment the first photographic head and second camera have shot photo, the image the comprising human body synthesis M that the first photographic head and second camera described in M shooting moment shoot respectively simultaneously can be chosen and open three-dimensional image, it is also possible to the synthesis N choosing N number of shooting moment shooting opens three-dimensional image, wherein M >=N;
Image is a pictures of photographic head shooting, and picture frame is then a series of pictures being continuously shot in the set time, and picture frame sequence is made up of a series of images.
Certainly, when selecting synthesis three-dimensional image, each image can be selected in several images that the first photographic head is continuously shot and several the images that second camera is continuously shot all to synthesize three-dimensional image (wherein, the time of every image of second camera shooting is all corresponding with the photo shot at synchronization in the first photographic head).
Wherein, the mode of at least one the image synthesis three-dimensional image comprising human body for the first photographic head and second camera are shot respectively at synchronization, it is not belonging to the primary object of the present invention, there is multiple implementation in the prior art, this is not defined by the embodiment of the present invention, due to all identical with principle with the mode of every image synthesis three-dimensional image that second camera shoots in Preset Time for the first photographic head, the embodiment of the present invention only illustrates for the second image and the 3rd image, wherein, second image and the 3rd image respectively in Preset Time by the image that synchronization shoots respectively at least one of the first photographic head and the first photographic head, not there is any indicative implication.
Exemplary, as in figure 2 it is shown, step S101 can be accomplished by,
S1011, obtain each pixel of described second image;
Wherein, for obtaining the concrete mode of each pixel of the second image, the embodiment of the present invention does not repeat them here, it is possible to realized by prior art, for instance, particle filter.
After getting each pixel of the second image, coordinate system can be set with described second image and the 3rd image, then each pixel on the second image and the 3rd image all can represent by the form of coordinate, as shown in Figure 3 a with shown in Fig. 3 b, certainly can also there are other modes in order to pixel corresponding on uniquely tagged the second image and the 3rd image, the embodiment of the present invention does not repeat them here.
It should be noted that, when obtaining three-dimensional image, can also first extract the human body contour outline of described second image, after extracting human body human body contour outline, obtain each pixel in the human body contour outline of described second image, perform step S1012 based on each pixel in each described human body human body contour outline, so can improve accuracy of identification further, it is to avoid in three-dimensional image, introduce background or interference.
S1012, centered by each pixel of described second image, pixel sets up preset window;Wherein, described preset window comprises according to predeterminable range, M pixel centered by described central pixel point;
Fig. 3 a is that in the second image, centered by any one pixel, pixel sets up the schematic diagram of preset window, its preset window can be passed through centered by described central pixel point, extending the region that L long measure comprises described central pixel point surrounding (upper and lower, left, by) is each, namely described predeterminable range is that the 2L then each pixel of above-mentioned M is all pixels respectively extending in the region that L long measure comprises with described central pixel point surrounding;The specific size of described L is not defined by the embodiment of the present invention, it is possible to the precision reached according to actual needs is set.
S1013, obtain the pixel value of described preset window;
Owing to comprising M pixel in preset window, therefore the summation that pixel value is M pixel gray value of described preset window, the concrete mode embodiment of the present invention for calculating the gray value of each pixel does not repeat them here, such as, if described preset window be centered by any one pixel pixel to each pixel of from left to right, then comprising 5 pixels in this preset window, the pixel value of this preset window is the summation of 5 pixel gray values.
S1014, pixel value according to described preset window, extracting the region minimum with the value differences value of described preset window from described 3rd image is target area, as shown in Figure 3 b;
Owing to setting up preset window for the second each pixel of image kind, and the mode of the target area found from described 3rd image according to the pixel value of preset window is all identical with principle, therefore the embodiment of the present invention only illustrates for the first pixel, this first pixel is any one pixel in the second image, does not have indicative implication.
Exemplary, as shown in Figure 4, step S1014 can be accomplished by:
S10141, determine described first pixel coordinate in described second image, and set up the first preset window centered by described first pixel;As shown in Figure 3 a;
S10142, when keep described first pixel vertical coordinate constant, each candidate region is chosen from described 3rd image, the window size of described candidate region is identical with described first preset window size, and described candidate region is that centered by any one pixel, pixel is set up in described 3rd image, the vertical coordinate of each pixel in described candidate region is identical with the vertical coordinate of described first pixel;
Wherein, the window size of described candidate region or window distance refer to any one central pixel point in candidate region, according to predeterminable range 2L, centered by described central pixel point, extend the region that L long measure comprises described central pixel point surrounding (upper and lower, left, by) is each;
S10143, calculating the pixel value of each described candidate region, described pixel value refers to the gray value sum of all pixels in candidate region;
S10144, candidate region minimum for the difference value of the pixel value of described candidate region Yu the pixel value of described preset window is defined as target area.
Wherein, when getting the coordinate of the first pixel, described first pixel can be pointed to the direction of the second image from the 3rd image, when keeping vertical coordinate constant, first pixel is traveled through any one pixel in described 3rd image, and can to extract the region minimum with the value differences value of described preset window from the 3rd image by SAD (SumofAbsoluteDifference) or SSD (SumofSquaredDifference) algorithm matching mode be target area, d point as shown in Figure 3 c.
Certainly, in order to reduce amount of calculation, after the coordinate getting the first pixel, it is possible to identical with described first pixel vertical coordinate from described 3rd image, be more than or equal to the candidate region of abscissa is chosen target area.
Certainly, the embodiment of the present invention can also based on the 3rd image, choosing the region minimum with the value differences of the preset window of any one pixel structure in the 3rd image in the second image is target area, now, the direction of the 3rd image should be pointed to according to the second image, when keeping vertical coordinate constant, the preset window constituted by each pixel in the 3rd image travels through the candidate region of described second image, to obtain target area.
S1015, determine the central pixel point of each described target area;
S1016, each central pixel point of described second image is mated with the central pixel point of described target area, obtain the three-dimensional image corresponding with described second image.
Preferably, in order to improve accuracy of identification, need to extract the human body contour outline in described first image, on the basis of this human body contour outline, obtain the Pixel Information of each pixel, and from three-dimensional image, obtain corresponding pixel range information, owing to the human body of user should be at same plane, thus have close pixel range information, therefore before recognition, the pixel distance that human body in three-dimensional image is corresponding can be carried out averaging operation, so that the interference information such as the human body in human body contour outline and background are easily separated, thus the high-precision human body extracting user.
Further, the human body contour outline in the three-dimensional image that described extraction the first image is corresponding, including:
S1021, the horizontal histogram that the three-dimensional image corresponding with the first image is set up range information and longitudinal rectangular histogram;
S1022, carry out based on described horizontal histogram and described longitudinal rectangular histogram method of least square algorithm lines detection process;
S1023, horizontal histogram after processing through lines detection extract there is the horizontal straight line of identical vertical coordinate, and extract longitudinal straight line with identical abscissa in longitudinal rectangular histogram.
S1024, obtain the human body contour outline of three-dimensional image corresponding to described first image according to described horizontal straight line and described longitudinal straight line.
The mode extracted for human body contour outline has multiple, and the embodiment of the present invention does not repeat them here, exemplary, and the method can by adopting eight neighborhood search method to realize.
nullFor step S104,After the range information that each three-dimensional image is determined in getting N number of three-dimensional image at least one pixel is corresponding,Need the difference between the N number of range information of comparison whether in preset threshold range,Such as,In Preset Time 5 seconds,First photographic head and second camera have taken 10 two dimensional images respectively,Every pair of youngster's two dimensional image is synthetically derived three-dimensional image,10 range informations corresponding to human body contour outline can be finally got from three-dimensional image,If the difference that the numerical value of the distance of these 10 human body contour outlines is each other is in a default threshold range,Such as 10cm,Then show that user moves or static before terminal in a less range of activity,Namely further demonstrate that user has at least stopped Preset Time 5 seconds before terminal precedent such as television set,In this case,Show that user has the intention of viewing TV.Then continue executing with S105 step.
Further, also included before step S105 generates start-up command:
S1041: the image comprising human body of N number of shooting moment the first photographic head shooting is carried out eye recognition, obtains the eye profile of user;
Two dimensional image carries out eye recognition, existing technology has multiple implementation, for instance need the first image is carried out skin color segmentation;Described first image after carrying out skin color segmentation is carried out rim detection;Get the eye profile of user further.
S1042: identifying the eye profile variations information that in described N number of shooting moment, adjacent moment is corresponding frame by frame, and mate with human eye feature storehouse, described human eye feature storehouse prestores the human eye action of user;
S1043: choose the human eye action minimum with described adjacent human eye profile variations information gap from described human eye feature storehouse as target human eye action;
S1044: if described target human eye action meets default start-up command requirement, then generate start-up command.
Assume that user sleeps before television set, then the eye profile of user is always Guan Bi, now can determine whether that user is not intended to see TV, then need not generate starting up's instruction.If user normally opens eyes before television set, then in N number of shooting moment, at least photograph an eye profile when user opens one's eyes.
Specifically when identifying the eye motion of user, track algorithm can be passed through according to the human eye profile variations information between the multiple adjacent two dimensional image got, such as, user's human eye action that JPDA wave filter (JPDAF), multiple hypotheis tracking (MHT) algorithm, dynamic multidigit allocation algorithm etc. prestore with human eye feature storehouse is mated, the target human eye action corresponding to identify current user's eye profile, as opened eyes, or rapid eye movements etc., and perform operational order corresponding with described target human eye action.Generate corresponding start-up command.Such as, it is rapid eye movements that system identification goes out the human eye action of user, the target human eye action of its correspondence meets default start-up command and requires that (start-up command requires that corresponding human eye action can include rapid eye movements, all the time open eyes, or normal eye opening shows that user is look at TV, and have start wish), then after system identification, corresponding start-up command can be generated.
Those skilled in the art should know, similar to automatic turn-on function, corresponding to television auto power-off, still this recognition methods is gone for, if television set is in playing process, user's rapid eye movements detected by photographic head, it is also possible to show that user wishes closing television or makes television standby, then now can transmit a signal to the power management module of TV, perform shutdown command.
The embodiment of the present invention additionally provides the control system of a kind of starting up of terminal, as shown in Figure 5, the control method of each function starting up of terminal a kind of with the above embodiment of the present invention in the control system of this kind of starting up of terminal is corresponding, specifically being referred to the description of the above embodiment of the present invention, the embodiment of the present invention does not repeat them here.As shown in Figure 5, the control system of this kind of starting up of terminal, it is applied to terminal 60, including: it is set in parallel in the first photographic head 601 in terminal and second camera 602, operates in the image processing system 603 on described terminal handler, image identification system 604 and execution system 605;
Wherein, described first photographic head 601 and the second shooting 602 are on same level line;
Described first photographic head 601 and the second shooting 602, for shooting at least one the image comprising human body at Preset Time;
Described image processing system 603, for by each shooting moment in the N number of shooting moment in Preset Time, what the first photographic head and second camera shot respectively simultaneously comprises the image N number of three-dimensional image of synthesis of human body;
Described image identification system 604, for extracting the human body contour outline in the three-dimensional image that the first image is corresponding, wherein, described first image is any one image of the first photographic head shooting described in N number of shooting moment;
On the three-dimensional image corresponding with described first image, obtain the range information corresponding with at least one pixel in described human body contour outline;
Whether the difference between the range information that at least one pixel that relatively in described N number of three-dimensional image, each three-dimensional image is determined is corresponding is in preset threshold range;
Described execution system 605, for when judged result is for being, generating start-up command, and control terminal unlatching according to described start-up command.
nullThe embodiment of the present invention provides a kind of starting up of terminal to control system,By at least one the image comprising human body that the first photographic head and second camera synchronization shoot respectively is synthesized three-dimensional image,And three-dimensional image corresponding to two dimensional image based on described first photographic head shooting obtains the range information that in human body contour outline, at least one pixel is corresponding,Whether the difference between the range information that relatively pixel of described N number of three-dimensional image is determined afterwards is in preset threshold range,Before can determine that whether user is in terminal by this result of the comparison,And if the difference of N number of range information is in certain threshold range,Then illustrate that position and relative and terminal the distance of user do not have saltus step,Such as can determine whether out whether user is look at TV and keeps a period of time,Relative improves the accuracy identifying that customer location generates start-up command,Namely when the difference of the plurality of range information meets preset threshold range,Can determine that user has the intention of viewing terminal such as TV,And further generate start-up command,Control terminal to open,Compared with prior art,Eliminate mode that whether infrared detection technology perception have users easily by surrounding environment influence,Accuracy of identification and the problem such as sensitivity is poor,The control method of this starting up of terminal and system are found range by dual camera three-dimensional information and in conjunction with human bioequivalence algorithm,Can determine that the start of user is intended to,Automatic control terminal is opened,Ensure that high real-time simultaneously,In high precision,The manipulation increasing substantially user is experienced.
Optionally, as shown in Figure 6, described image processing system 603 includes:
First acquiring unit 6031, for obtaining each pixel of described second image;
Set up unit 6032, set up preset window for pixel centered by each pixel of described second image;Wherein, described preset window comprises according to predeterminable range, M pixel centered by described central pixel point;
Second acquisition unit 6033, for obtaining the pixel value of described preset window
Extraction unit 6034, for the pixel value according to described preset window, extracting the region minimum with the value differences value of described preset window from described 3rd image is target area;
Determine unit 6035, for determining the central pixel point of each described target area;
Generate unit 6036, for each central pixel point of described second image being mated with the central pixel point of described target area, obtain the three-dimensional image corresponding with described second image.
Optionally, described extraction unit 6034 includes:
Determine module, for determining described first pixel coordinate in described second image, and set up the first preset window centered by described first pixel;
Choose module, for when keeping described first pixel vertical coordinate constant, select from described 3rd image and the described first identical all candidate regions of preset window size, described candidate region is that centered by any one pixel, pixel is set up in described 3rd image, and the vertical coordinate of each pixel in described candidate region is identical with the vertical coordinate of described first pixel;
Computing module, for calculating the pixel value of each described candidate region, described pixel value refers to the gray value sum of all pixels in candidate region;
Determination module, for being defined as target area by candidate region minimum with the value differences value of described first preset window in the pixel value of described all candidate regions.
Optionally, described image identification system 604 includes contours extract unit and pixel extraction unit, described contours extract unit specifically for:
The three-dimensional image corresponding with the first image is set up the horizontal histogram of range information and longitudinal rectangular histogram;
The lines detection carrying out method of least square algorithm based on described horizontal histogram and described longitudinal rectangular histogram processes;
Horizontal histogram after processing through lines detection extracts the horizontal straight line with identical vertical coordinate, and extracts longitudinal straight line with identical abscissa in longitudinal rectangular histogram;
Obtain the human body contour outline of three-dimensional image corresponding to described first image with described longitudinal straight line according to described horizontal straight line.
Optionally, described image identification system 604 also includes recognition unit, and described recognition unit includes:
Human eye analysis module, for the image comprising human body of N number of shooting moment the first photographic head shooting is carried out eye recognition, obtains the eye profile of user;
Human eye matching module, for identifying the eye profile variations information that in described N number of shooting moment, adjacent moment is corresponding frame by frame, and mates with human eye feature storehouse, and described human eye feature storehouse prestores the human eye action of user;
Object selection module, chooses the human eye action minimum with described adjacent human eye profile variations information gap as target human eye action from described human eye feature storehouse;
Instruction control module, for when described target human eye action meets default start-up command requirement, controlling described execution system and generate start-up command.
In several embodiments provided herein, it should be understood that disclosed system, apparatus and method, it is possible to realize by another way.Such as, device embodiment described above is merely schematic, such as, the division of described unit, being only a kind of logic function to divide, actual can have other dividing mode when realizing, for instance multiple unit or assembly can in conjunction with or be desirably integrated into another system, or some features can ignore, or do not perform.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be through INDIRECT COUPLING or the communication connection of some interfaces, device or unit, it is possible to be electrical, machinery or other form.
The described unit illustrated as separating component can be or may not be physically separate, and the parts shown as unit can be or may not be physical location, namely may be located at a place, or can also be distributed on multiple NE.Some or all of unit therein can be selected according to the actual needs to realize the purpose of the present embodiment scheme.
It addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it is also possible to be that the independent physics of unit includes, it is also possible to two or more unit are integrated in a unit.Above-mentioned integrated unit both can adopt the form of hardware to realize, it would however also be possible to employ hardware adds the form of SFU software functional unit and realizes.
The above-mentioned integrated unit realized with the form of SFU software functional unit, it is possible to be stored in a computer read/write memory medium.Above-mentioned SFU software functional unit is stored in a storage medium, including some instructions with so that a computer equipment (can be personal computer, server, or the network equipment etc.) performs the part steps of method described in each embodiment of the present invention.And aforesaid storage medium includes: USB flash disk, portable hard drive, read only memory (Read-OnlyMemory, be called for short ROM), random access memory (RandomAccessMemory, be called for short RAM), the various media that can store program code such as magnetic disc or CD.
Last it is noted that above example is only in order to illustrate technical scheme, it is not intended to limit;Although the present invention being described in detail with reference to previous embodiment, it will be understood by those within the art that: the technical scheme described in foregoing embodiments still can be modified by it, or wherein portion of techniques feature is carried out equivalent replacement;And these amendments or replacement, do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.