CN104050454A - Movement gesture track obtaining method and system - Google Patents
Movement gesture track obtaining method and system Download PDFInfo
- Publication number
- CN104050454A CN104050454A CN201410290358.2A CN201410290358A CN104050454A CN 104050454 A CN104050454 A CN 104050454A CN 201410290358 A CN201410290358 A CN 201410290358A CN 104050454 A CN104050454 A CN 104050454A
- Authority
- CN
- China
- Prior art keywords
- colour
- image
- skin mask
- frame video
- video image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The invention is suitable for the field of man-machine interaction, and provides a movement gesture track obtaining method and system. The method comprises the steps of obtaining a current frame video image; analyzing the current frame video image and obtaining a skin color mask image of the current frame video image; comparing the skin color mask image of the current frame video image with a skin color mask image of a preset reference frame video image to obtain a skin color mask difference image; obtaining a first hand area according to the skin color mask difference image; diagnosing the first hand area and obtaining a movement gesture track from the final position of the hand. According to the movement gesture track obtaining method and system, the skin color mask difference image of the current frame video image and the reference frame video image is obtained, and a seed area for calculating the hand area is obtained from the skin color mask difference image, so that the seed area grows to be the whole hand area, calculation efficiency is high, and obtaining stability of the movement gesture is improved.
Description
Technical field
The invention belongs to field of human-computer interaction, relate in particular to a kind of motion gesture track acquisition methods and system.
Background technology
In recent years, dynamic gesture recognition technology has had application more and more widely in field of human-computer interaction.For example, on Intelligent television terminal, user can change the mouse cursor position of intelligent television by hand exercise, carry out game control or web page browsing, and the simulating keyboard behavior that can use gesture carrys out switching TV channel and regulates volume etc.In these application, an involved core is motion hand tracking problem, that is, how to stablize and follow the tracks of glibly operator's hand and obtain the movement locus of standard.The existing hand tracking based on computer vision, ubiquity following problem: 1, counting yield is low, tracking based on motion detection is easily subject to the factors such as background perturbation and illumination fluctuation and disturbs, and tracking based on characteristic matching, generally need to extract complex characteristic, also need to carry out many wheel couplings, algorithm complex is high, and huge operand makes it be difficult to operation glibly on embedded platform (as: intelligent television, smart mobile phone etc.); 2, poor stability, due to view transformation and the motion deformation of moving and causing, and the impact of neighbourhood noise (comprising the associated movement of other target travels, body local motion, elbow etc.), the mistakes such as motion gesture is easily lost in following the tracks of, drift.
Summary of the invention
The object of the embodiment of the present invention is to provide a kind of acquisition methods and system of man-machine interaction order, low to solve the counting yield of prior art, the problem of poor stability.
The embodiment of the present invention is achieved in that a kind of motion gesture track acquisition methods, said method comprising the steps of:
Obtain current frame video image;
Described current frame video image is analyzed, obtained the colour of skin mask image of described current frame video image;
The colour of skin mask image of the colour of skin mask image of described current frame video image and default reference frame video image is compared, obtain colour of skin mask difference image;
According to described colour of skin mask difference image, obtain the first hand region;
Described the first hand region is diagnosed, and is obtained motion gesture track from the result of diagnosis, comprising:
If it is normal that diagnostic result shows first area, using the peak of described the first hand region as representative point component movement gesture track; Or,
If diagnostic result shows abnormal, take the appearance features of described reference frame video image as according to obtaining the second hand region of described current frame video image, and using the peak of described the second hand region as representative point component movement gesture track.
Another object of the embodiment of the present invention is to provide a kind of motion gesture track to obtain system, and described system comprises:
Current frame video image acquiring unit, for obtaining current frame video image;
Colour of skin mask image acquiring unit, analyzes for the current frame video image that described current frame video image acquiring unit is obtained, and obtains the colour of skin mask image of described current frame video image;
Colour of skin mask difference image acquiring unit, for described colour of skin mask image being obtained to the colour of skin mask image of the current frame video image that subelement obtains and the colour of skin mask image of default reference frame video image is compared, obtains colour of skin mask difference image;
The first hand region acquiring unit, obtains the first hand region for the colour of skin mask difference image obtaining according to described colour of skin mask difference image acquiring unit;
Motion gesture track obtains single cloud, diagnoses, and obtain motion gesture track from final hand position for the first hand region that described the first hand region acquiring unit is obtained, and comprising:
If diagnostic result is normal, using the peak of described the first hand region as representative point component movement gesture track; Or,
If diagnostic result is abnormal, take the appearance features of described reference frame video image as according to obtaining the second hand region of described current frame video image, and using the peak of described the second hand region as representative point component movement gesture track.
The embodiment of the present invention, obtain the colour of skin mask difference image of current frame video image and reference frame video image, in this colour of skin mask difference image, obtain the seed region that calculates hand region, pass through region-growing method, make seed region be grown to whole hand region, counting yield is quick, and after obtaining hand region, by error diagnosis scheme, hand region is diagnosed, and use the appearance features of video image to revise there is abnormal hand region, improved the stability that motion gesture is obtained.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the process flow diagram of the motion gesture track acquisition methods that provides of the embodiment of the present invention;
Fig. 2 is the schematic diagram of the colour of skin mask difference image that provides of the embodiment of the present invention;
Fig. 3 is the schematic diagram with seed region formation hand region that the embodiment of the present invention provides;
Fig. 4 is the schematic diagram of the hand region obtained after diagnosis and error correction that the embodiment of the present invention provides;
Fig. 5 be the embodiment of the present invention provide using the peak of hand region as the schematic diagram of representative point component movement gesture track;
Fig. 6 is the structural drawing that motion hand track that the embodiment of the present invention provides obtains system.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
For technical solutions according to the invention are described, below by specific embodiment, describe.
Embodiment mono-
The process flow diagram that is illustrated in figure 1 the motion gesture track acquisition methods that the embodiment of the present invention provides, said method comprising the steps of:
In step S101, obtain current frame video image.
In embodiments of the present invention, carry out the video image that first terminal that motion gesture obtains obtains present frame, terminal can be obtained video image by the first-class video acquisition instrument of making a video recording.
In step S102, described current frame video image is analyzed, obtain the colour of skin mask image of described current frame video image.
In embodiments of the present invention, carry out terminal that motion gesture obtains after having obtained video image, this video image is calculated, obtain the colour of skin mask image of this video image.Described described current frame video image is analyzed, is obtained the step of the colour of skin mask image of described current frame video image, comprising:
1, obtain the pixel value of each point of described current frame video image.
In embodiments of the present invention, video image is by forming, and each point has its pixel value, carries out the pixel value that first terminal that motion gesture obtains obtains these points.Obtain the method for pixel value and use prior art, at this, do not repeat.
2, the pixel value of described each point and default complexion model are compared, obtain the colour of skin mask image of described current frame video image.
In embodiments of the present invention, in comparison process, the point being consistent with complexion model is designated as 1, does not want that with complexion model the point meeting is designated as 0, can conveniently carry out like this terminal that motion gesture obtains and generate fast colour of skin mask image.
As an optional embodiment of the present invention, described, the pixel value of described each point and default complexion model are compared, before obtaining the step of colour of skin mask image of described current frame video image, described method is further comprising the steps of:
Default complexion model.
In embodiments of the present invention, default complexion model can be the complexion model of training under line, also can make the complexion model of online design.If use online complexion model, need to there is an online complexion model of preparatory stage initialization; After model initialization completes, the video image that every width is newly obtained is used complexion model to detect, and obtains colour of skin mask image.
In step S103, the colour of skin mask image of the colour of skin mask image of described current frame video image and default reference frame video image is compared, obtain colour of skin mask difference image.
In embodiments of the present invention, carry out terminal that motion gesture obtains after having obtained the colour of skin mask image of current frame video image, the colour of skin mask image of this colour of skin mask image and default reference frame video image is compared (at certain point, if present frame mask image is labeled as 1 this point, and reference frame this point be labeled as 0, colour of skin mask difference image is labeled as 1 at this point; In all the other situations, colour of skin mask difference image is all 0 at the mark of this point), can obtain colour of skin mask difference image.Be illustrated in figure 2 the schematic diagram of colour of skin mask difference image.
It is to be noted: the selection of reference frame will be determined according to the computing velocity of equipment, if therefore terminal processes speed is fast, the present frame video image far away of should selecting to be separated by, if terminal processes speed is slow, the nearer video image of present frame of should selecting to be separated by, " distance " here determines by the frame number of being separated by between present frame and reference frame.
In step S104, according to described colour of skin mask difference image, obtain the first hand region.
In embodiments of the present invention, carry out the terminal that motion gesture obtains and can obtain hand region according to colour of skin mask difference image.The described step of obtaining the first hand region according to described colour of skin mask difference image comprises:
1, described colour of skin mask difference image is done to morphology opening operation, remove the noise prospect in described colour of skin mask difference image;
2, the described colour of skin mask difference image of removing after noise is carried out to UNICOM's regional analysis, by region parameter in described colour of skin mask difference image and the highest region of default moving region constraint condition matching degree, be set to hand exercise foreground area, described region parameter comprises: the length breadth ratio in described region area and described region;
In embodiments of the present invention, moving region constraint condition, relevant with factors such as " reference frame be separated by frame number ", " operator-camera distance range ", " user's hand exercise speed spans ", these factors can be called independent parameter.In order to determine the relevant value in the constraint condition of moving region, the variation relation between these values and independent parameter under line first, this relation can obtain by experiment statistics under line; In actual applications, according to concrete independent parameter value used on line, determine the relevant value in restriction relation.
3, take described hand exercise foreground area is seed region, uses region-growing method, obtains the first hand region that comprises described hand exercise foreground area in the colour of skin mask image of described current frame video image.
It is to be noted, why on the basis in sport foreground region, pass through region-growing method, sport foreground region is further expanded, be because: be limited by gesture motion speed, for doing two frame video image hands of colour of skin mask difference, may there is overlapping part, therefore corresponding colour of skin mask difference image possibly cannot comprise whole hand foreground area, but is a part for whole hand region by the known difference gained of the definition hand prospect of mask difference image.
Be illustrated in figure 3 the schematic diagram that forms hand region with seed region.
In step S105, described the first hand region is diagnosed, and obtained motion gesture track from final hand position, comprising: if diagnostic result for normal, is usingd the peak of described the first hand region as representative point component movement gesture track; Or,
If diagnostic result is abnormal, take the appearance features of described reference frame video image as according to obtaining the second hand region of described current frame video image, and using the peak of described the second hand region as representative point component movement gesture track.
In embodiments of the present invention, when the remarkable face of hand warp in motion process or class colour of skin object, hand prospect in colour of skin mask image can overlap or connect together with people's face/class colour of skin object prospect, and this there will be following situation: on colour of skin mask difference image, can not find the region that meets moving region constraint condition; Or the hand region obtaining by region-growing method comprises people's face prospect/class colour of skin prospect.Therefore after obtaining the first hand region, need to diagnose this first hand region, comprise:
If diagnostic result is normal, using the peak of described the first hand region as representative point component movement gesture track; Or,
If diagnostic result is abnormal, take the appearance features of described reference frame video image as according to obtaining the second hand region of described current frame video image, and using the peak of described the second hand region as representative point component movement gesture track.
After being as shown in Figure 4 after diagnosis and error correction, obtain the schematic diagram of hand region.
In Fig. 4, the 1st row represents the tracking results of t frame, and this frame, because do not note abnormalities in diagnostic procedure, has therefore only been done the tracking based on motion+Skin Color Information; The tracking results of motion+colour of skin gained is used in the 2nd row representative at t+1 frame, the first hand region still notes abnormalities after diagnosis, therefore need to further by apparent information, follow the tracks of; What the 3rd row was shown is in error correction procedure, use respectively the result of original graph and filtration figure gained, front two width figure have shown the result of following the tracks of based on original color image, red block mark the position of t+1 frame, can find out with actual target locations and have significant difference, rear two width show is to take the result that color filter figure follows the tracks of as source data, can find out that, in t+1 frame, the result of tracking is improved;
Be illustrated in figure 5 and using the peak of hand region as the schematic diagram of representative point component movement gesture track.
In Fig. 5, as a comparison, listed file names with and take the movement locus schematic diagram that hand central point forms as representative point; Two of the left side one row is to take the track of hand central point as representative point composition, two of the right one row is to take the track of hand peak as representative point composition, if can find out, take central point as representative point, when hand enters visual field at first, the situation that can exist tracing point to press together, the track total length of gained is also smaller, and evidence is not as the way abundance with end points forming curves.
The embodiment of the present invention, obtain the colour of skin mask difference image of current frame video image and reference frame video image, in this colour of skin mask difference image, obtain the seed region that calculates hand region, pass through region-growing method, make seed region be grown to whole hand region, counting yield is quick, and after obtaining hand region, by with relatively hand region being diagnosed of historical data, and use the appearance features of video image to revise hand region in the situation that appearance is abnormal, improved the stability that motion gesture is obtained.
As an optional embodiment of the present invention, in the described colour of skin mask image by the colour of skin mask image of described current frame video image and default reference frame video image, compare, before obtaining the step of colour of skin mask difference image, described method is further comprising the steps of:
Default reference frame video image.
Embodiment bis-
The motion hand track that being illustrated in figure 6 the embodiment of the present invention provides obtains the structural drawing of system, for convenience of explanation, the part relevant to the embodiment of the present invention is only shown, and comprising:
Current frame video image acquiring unit 601, for obtaining current frame video image.
In embodiments of the present invention, when first video image acquiring unit 601 obtains the video image of present frame, terminal can be obtained video image by the first-class video acquisition instrument of making a video recording.
Colour of skin mask image acquiring unit 602, analyzes for the current frame video image that described current frame video image acquiring unit 601 is obtained, and obtains the colour of skin mask image of described current frame video image.
In embodiments of the present invention, colour of skin mask image acquiring unit 602, after having obtained video image, calculates this video image, obtains the colour of skin mask image of this video image.Described colour of skin mask image acquiring unit 602, comprising:
Pixel value obtains subelement 6021, for obtaining the pixel value of each point of described current frame video image.
In embodiments of the present invention, video image is by forming, and each point has its pixel value, carries out the pixel value that first terminal that motion gesture obtains obtains these points.Obtain the method for pixel value and use prior art, at this, do not repeat.
Colour of skin mask image is obtained subelement 6022, for described pixel value being obtained to the pixel value of each point that subelement 6021 obtains and default complexion model, compares, and obtains the colour of skin mask image of described current frame video image.
In embodiments of the present invention, in comparison process, the point being consistent with complexion model is designated as 1, does not want that with complexion model the point meeting is designated as 0, can conveniently carry out like this terminal that motion gesture obtains and generate fast colour of skin mask image.
As an optional embodiment of the present invention, in described colour of skin mask image, obtain before subelement 6022 obtains colour of skin mask image, described colour of skin mask image acquiring unit 602 also comprises:
Complexion model is preset subelement 6023, for default complexion model.
In embodiments of the present invention, default complexion model can be the complexion model of training under line, also can make the complexion model of online design.If use online complexion model, need to there is an online complexion model of preparatory stage initialization; After model initialization completes, the video image that every width is newly obtained is used complexion model to detect, and obtains colour of skin mask image.
Colour of skin mask difference image acquiring unit 603, for described colour of skin mask image being obtained to the colour of skin mask image of the current frame video image that subelement 602 obtains and the colour of skin mask image of default reference frame video image is compared, obtain colour of skin mask difference image.
In embodiments of the present invention, carry out terminal that motion gesture obtains after having obtained the colour of skin mask image of current frame video image, the colour of skin mask image of this colour of skin mask image and default reference frame video image is compared (at certain point, if present frame mask image is labeled as 1 this point, and reference frame this point be labeled as 0, colour of skin mask difference image is labeled as 1 at this point; In all the other situations, colour of skin mask difference image is all 0 at the mark of this point), can obtain colour of skin mask difference image.Be illustrated in figure 2 the schematic diagram of colour of skin mask difference image.
It is to be noted: the selection of reference frame will be determined according to the computing velocity of equipment.If therefore terminal processes speed is fast, the present frame video image far away of should selecting to be separated by, if terminal processes speed is slow, the nearer video image of present frame of should selecting to be separated by.Here " distance " determines by the frame number of being separated by between present frame and reference frame.
The first hand region acquiring unit 604, obtains the first hand region for the colour of skin mask difference image obtaining according to described colour of skin mask difference image acquiring unit 603.
In embodiments of the present invention, carry out the terminal that motion gesture obtains and can obtain hand region according to colour of skin mask difference image.Described the first hand region acquiring unit 604, comprising:
Noise prospect is removed subelement 6041, for described colour of skin mask difference image is done to morphology opening operation, removes the noise prospect in described colour of skin mask difference image;
Hand exercise foreground area arranges subelement 6042, for the colour of skin mask difference image of described noise prospect being removed after subelement 6041 removal noises, carry out UNICOM's regional analysis, by region parameter in described colour of skin mask difference image and the highest region of default moving region constraint condition matching degree, be set to hand exercise foreground area, described region parameter comprises: the length breadth ratio in described region area and described region.
In embodiments of the present invention, moving region constraint condition, relevant with factors such as " reference frame be separated by frame number ", " operator-camera distance range ", " user's hand exercise speed spans ", these factors can be called independent parameter.In order to determine the relevant value in the constraint condition of moving region, the variation relation between these values and independent parameter under line first, this relation can obtain by experiment statistics under line; In actual applications, according to concrete independent parameter value used on line, determine the relevant value in restriction relation.
The first hand region is obtained subelement 6043, for take described hand exercise foreground area, the hand exercise foreground area that subelement 6042 arranges is set is seed region, use region-growing method, in the colour of skin mask image of described current frame video image, obtain the first hand region that comprises described hand exercise foreground area.
It is to be noted, why on the basis in sport foreground region, pass through region-growing method, sport foreground region is further expanded, be because:: be limited by gesture motion speed, for doing two frame video image hands of colour of skin mask difference, may there is overlapping part, therefore corresponding colour of skin mask difference image possibly cannot comprise whole hand foreground area, but is a part for whole hand region by the known difference gained of the definition hand prospect of mask difference image.
Motion gesture track obtains single cloud 605, diagnoses, and obtain motion gesture track from final hand position for the first hand region that described the first hand region acquiring unit 604 is obtained, and comprising:
If diagnostic result is normal, using the peak of described the first hand region as representative point component movement gesture track; Or,
If diagnostic result is abnormal, take the appearance features of described reference frame video image as according to obtaining the second hand region of described current frame video image, and using the peak of described the second hand region as representative point component movement gesture track.
In embodiments of the present invention, when the remarkable face of hand warp in motion process or class colour of skin object, hand prospect in colour of skin mask image can overlap or connect together with people's face/class colour of skin object prospect, and this there will be following situation: on colour of skin mask difference image, can not find the region that meets moving region constraint condition; Or the hand region obtaining by region-growing method comprises people's face prospect/class colour of skin prospect.Therefore after obtaining the first hand region, need to diagnose or further error correction this first hand region, comprise:
If diagnostic result is normal, using the peak of described the first hand region as representative point component movement gesture track; Or,
If diagnostic result is abnormal, take the appearance features of described reference frame video image as according to obtaining the second hand region of described current frame video image, and using the peak of described the second hand region as representative point component movement gesture track.
Be illustrated in figure 4 the schematic diagram of the hand region of obtaining after diagnosis and error correction.
In Fig. 4, the 1st row represents the tracking results of t frame, and this frame, because do not note abnormalities in diagnostic procedure, has therefore only been done the tracking based on motion+Skin Color Information; The tracking results of motion+colour of skin gained is used in the 2nd row representative at t+1 frame, the first hand region still notes abnormalities after diagnosis, therefore need to further by apparent information, follow the tracks of; What the 3rd row was shown is in error correction procedure, use respectively the result of original graph and filtration figure gained, front two width figure have shown the result of following the tracks of based on original color image, red block mark the position of t+1 frame, can find out with actual target locations and have significant difference, rear two width show is to take the result that color filter figure follows the tracks of as source data, can find out that, in t+1 frame, the result of tracking is improved.
Be illustrated in figure 5 and using the peak of hand region as the schematic diagram of representative point component movement gesture track.
As a comparison, in Fig. 5, listed file names with and take the movement locus schematic diagram that hand central point forms as representative point; Two of the left side one row is to take the track of hand central point as representative point composition, two of the right one row is to take the track of hand peak as representative point composition, if can find out, take central point as representative point, when hand enters visual field at first, the situation that can exist tracing point to press together, the track total length of gained is also smaller, and evidence is not as the way abundance with end points forming curves.
The embodiment of the present invention, obtain the colour of skin mask difference image of current frame video image and reference frame video image, in this colour of skin mask difference image, obtain the seed region that calculates hand region, pass through region-growing method, make seed region be grown to whole hand region, counting yield is quick, and after obtaining hand region, by with relatively hand region being diagnosed of historical data, and use the appearance features of video image to revise hand region in the situation that appearance is abnormal, improved the stability that motion gesture is obtained.
As an optional embodiment of the present invention, before described colour of skin mask difference image acquiring unit 603 obtains colour of skin mask difference image, described system also comprises:
Reference frame video image is preset unit 606, for default reference frame video image.
Those skilled in the art can be well understood to, for convenience and simplicity of description, only the division with above-mentioned each functional unit, module is illustrated, in practical application, can above-mentioned functions be distributed and by different functional units, module, completed as required, the inner structure that is described system is divided into different functional units or module, to complete all or part of function described above.Each functional unit in embodiment or module can be integrated in a processing unit, also can be that the independent physics of unit exists, also can be integrated in a unit two or more unit, above-mentioned integrated unit or module both can adopt the form of hardware to realize, and also can adopt the form of SFU software functional unit to realize.In addition, the concrete title of each functional unit, module also, just for the ease of mutual differentiation, is not limited to the application's protection domain.The specific works process of unit, module in said system, can, with reference to the corresponding process in preceding method embodiment, not repeat them here.
Those of ordinary skills it is also understood that, the all or part of step realizing in above-described embodiment method is to come the hardware that instruction is relevant to complete by program, described program can be in being stored in a computer read/write memory medium, described storage medium, comprises ROM/RAM, disk, CD etc.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any modifications of doing within the spirit and principles in the present invention, be equal to and replace and improvement etc., within all should being included in protection scope of the present invention.
Claims (10)
1. a motion gesture track acquisition methods, is characterized in that, described method comprises:
Obtain current frame video image;
Described current frame video image is analyzed, obtained the colour of skin mask image of described current frame video image;
The colour of skin mask image of the colour of skin mask image of described current frame video image and default reference frame video image is compared, obtain colour of skin mask difference image;
According to described colour of skin mask difference image, obtain the first hand region;
Described the first hand region is diagnosed, and is obtained motion gesture track from final hand position, comprising:
If diagnostic result is normal, using the peak of described the first hand region as representative point component movement gesture track; Or,
If diagnostic result is abnormal, take the appearance features of described reference frame video image as according to obtaining the second hand region of described current frame video image, and using the peak of described the second hand region as representative point component movement gesture track.
2. the method for claim 1, it is characterized in that, in the described colour of skin mask image by the colour of skin mask image of described current frame video image and default reference frame video image, compare, before obtaining the step of colour of skin mask difference image, described method is further comprising the steps of:
Default reference frame video image.
3. the method as described in claim 1～2 any one, is characterized in that, described described current frame video image is analyzed, and obtains the step of the colour of skin mask image of described current frame video image, comprising:
Obtain the pixel value of each point of described current frame video image;
The pixel value of described each point and default complexion model are compared, obtain the colour of skin mask image of described current frame video image.
4. method as claimed in claim 3, is characterized in that, described, the pixel value of described each point and default complexion model is compared, and before obtaining the step of colour of skin mask image of described current frame video image, described method is further comprising the steps of:
Default complexion model.
5. the method as described in claim 1～2 any one, is characterized in that, the described step of obtaining the first hand region according to described colour of skin mask difference image comprises:
Described colour of skin mask difference image is done to morphology opening operation, remove the noise prospect in described colour of skin mask difference image;
The described colour of skin mask difference image of removing after noise is carried out to UNICOM's regional analysis, by region parameter in described colour of skin mask difference image and the highest region of default moving region constraint condition matching degree, be set to hand exercise foreground area, described region parameter comprises: the length breadth ratio in described region area and described region;
The described hand exercise foreground area of take is seed region, uses region-growing method, obtains the first hand region that comprises described hand exercise foreground area in the colour of skin mask image of described current frame video image.
6. motion hand track obtains a system, it is characterized in that, described system comprises:
Current frame video image acquiring unit, for obtaining current frame video image;
Colour of skin mask image acquiring unit, analyzes for the current frame video image that described current frame video image acquiring unit is obtained, and obtains the colour of skin mask image of described current frame video image;
Colour of skin mask difference image acquiring unit, for described colour of skin mask image being obtained to the colour of skin mask image of the current frame video image that subelement obtains and the colour of skin mask image of default reference frame video image is compared, obtains colour of skin mask difference image;
The first hand region acquiring unit, obtains the first hand region for the colour of skin mask difference image obtaining according to described colour of skin mask difference image acquiring unit;
Motion gesture track obtains single cloud, diagnoses, and obtain motion gesture track from final hand position for the first hand region that described the first hand region acquiring unit is obtained, and comprising:
If diagnostic result is normal, using the peak of described the first hand region as representative point component movement gesture track; Or,
If diagnostic result is abnormal, take the appearance features of described reference frame video image as according to obtaining the second hand region of described current frame video image, and using the peak of described the second hand region as representative point component movement gesture track.
7. system as claimed in claim 6, rises and is characterised in that, before described colour of skin mask difference image acquiring unit obtains colour of skin mask difference image, described system also comprises:
Reference frame video image is preset unit, for default reference frame video image.
8. the system as described in claim 6～7 any one, is characterized in that, described colour of skin mask image acquiring unit, comprising:
Pixel value obtains subelement, for obtaining the pixel value of each point of described current frame video image;
Colour of skin mask image is obtained subelement, for described pixel value being obtained to the pixel value of each point that subelement obtains and default complexion model, compares, and obtains the colour of skin mask image of described current frame video image.
9. system as claimed in claim 8, is characterized in that, in described colour of skin mask image, obtains before subelement obtains colour of skin mask image, and described colour of skin mask image acquiring unit also comprises:
Complexion model is preset subelement, for default complexion model.
10. the system as described in claim 6～7 any one, is characterized in that, described the first hand region acquiring unit, comprising:
Noise prospect is removed subelement, for described colour of skin mask difference image is done to morphology opening operation, removes the noise prospect in described colour of skin mask difference image;
Hand exercise foreground area arranges subelement, for the colour of skin mask difference image of described noise prospect being removed after subelement removal noise, carry out UNICOM's regional analysis, by region parameter in described colour of skin mask difference image and the highest region of default moving region constraint condition matching degree, be set to hand exercise foreground area, described region parameter comprises: the length breadth ratio in described region area and described region;
The first hand region is obtained subelement, for take the hand exercise foreground area that described hand exercise foreground area arranges subelement setting, it is seed region, use region-growing method, in the colour of skin mask image of described current frame video image, obtain the first hand region that comprises described hand exercise foreground area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410290358.2A CN104050454B (en) | 2014-06-24 | 2014-06-24 | A kind of motion gesture track acquisition methods and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410290358.2A CN104050454B (en) | 2014-06-24 | 2014-06-24 | A kind of motion gesture track acquisition methods and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104050454A true CN104050454A (en) | 2014-09-17 |
| CN104050454B CN104050454B (en) | 2017-12-19 |
Family
ID=51503267
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410290358.2A Active CN104050454B (en) | 2014-06-24 | 2014-06-24 | A kind of motion gesture track acquisition methods and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104050454B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105825170A (en) * | 2016-03-10 | 2016-08-03 | 浙江生辉照明有限公司 | Detection method and apparatus for repeated gesture |
| CN106054650A (en) * | 2016-07-18 | 2016-10-26 | 汕头大学 | Novel intelligent household system and multi-gesture control method thereof |
| CN106127707A (en) * | 2016-06-23 | 2016-11-16 | 南京大学金陵学院 | A kind of processing method of supersonic blood image |
| CN108073851A (en) * | 2016-11-08 | 2018-05-25 | 株式会社理光 | A kind of method, apparatus and electronic equipment for capturing gesture identification |
| CN110956219A (en) * | 2019-12-09 | 2020-04-03 | 北京迈格威科技有限公司 | Video data processing method and device and electronic system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593022A (en) * | 2009-06-30 | 2009-12-02 | 华南理工大学 | A kind of quick human-computer interaction of following the tracks of based on finger tip |
| CN102324016A (en) * | 2011-05-27 | 2012-01-18 | 郝红卫 | Statistical method for high-density crowd flow |
| CN103376890A (en) * | 2012-04-16 | 2013-10-30 | 富士通株式会社 | Gesture remote control system based on vision |
| CN103455794A (en) * | 2013-08-23 | 2013-12-18 | 济南大学 | Dynamic gesture recognition method based on frame fusion technology |
| CN103514435A (en) * | 2012-06-29 | 2014-01-15 | 富士通株式会社 | Hand detection method and device |
| CN103679130A (en) * | 2012-09-24 | 2014-03-26 | 富士通株式会社 | Hand tracking method, hand tracking apparatus and gesture recognition system |
-
2014
- 2014-06-24 CN CN201410290358.2A patent/CN104050454B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593022A (en) * | 2009-06-30 | 2009-12-02 | 华南理工大学 | A kind of quick human-computer interaction of following the tracks of based on finger tip |
| CN102324016A (en) * | 2011-05-27 | 2012-01-18 | 郝红卫 | Statistical method for high-density crowd flow |
| CN103376890A (en) * | 2012-04-16 | 2013-10-30 | 富士通株式会社 | Gesture remote control system based on vision |
| CN103514435A (en) * | 2012-06-29 | 2014-01-15 | 富士通株式会社 | Hand detection method and device |
| CN103679130A (en) * | 2012-09-24 | 2014-03-26 | 富士通株式会社 | Hand tracking method, hand tracking apparatus and gesture recognition system |
| CN103455794A (en) * | 2013-08-23 | 2013-12-18 | 济南大学 | Dynamic gesture recognition method based on frame fusion technology |
Non-Patent Citations (1)
| Title |
|---|
| 杨青: "手势识别技术的研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105825170A (en) * | 2016-03-10 | 2016-08-03 | 浙江生辉照明有限公司 | Detection method and apparatus for repeated gesture |
| CN105825170B (en) * | 2016-03-10 | 2019-07-02 | 浙江生辉照明有限公司 | Toward the detection method and device of renaturation gesture |
| CN106127707A (en) * | 2016-06-23 | 2016-11-16 | 南京大学金陵学院 | A kind of processing method of supersonic blood image |
| CN106127707B (en) * | 2016-06-23 | 2019-04-16 | 南京大学金陵学院 | A kind of processing method of supersonic blood image |
| CN106054650A (en) * | 2016-07-18 | 2016-10-26 | 汕头大学 | Novel intelligent household system and multi-gesture control method thereof |
| CN108073851A (en) * | 2016-11-08 | 2018-05-25 | 株式会社理光 | A kind of method, apparatus and electronic equipment for capturing gesture identification |
| CN108073851B (en) * | 2016-11-08 | 2021-12-28 | 株式会社理光 | Grabbing gesture recognition method and device and electronic equipment |
| CN110956219A (en) * | 2019-12-09 | 2020-04-03 | 北京迈格威科技有限公司 | Video data processing method and device and electronic system |
| CN110956219B (en) * | 2019-12-09 | 2023-11-14 | 爱芯元智半导体(宁波)有限公司 | Video data processing method, device and electronic system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104050454B (en) | 2017-12-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10275653B2 (en) | Primary video objects with neighborhood reversibility | |
| EP2344980B1 (en) | Device, method and computer program for detecting a gesture in an image, and said device, method and computer program for controlling a device | |
| CN104050454A (en) | Movement gesture track obtaining method and system | |
| CN110362210B (en) | Human-computer interaction method and device integrating eye movement tracking and gesture recognition in virtual assembly | |
| CN105139004A (en) | Face expression identification method based on video sequences | |
| CN102981742A (en) | Gesture interaction system based on computer visions | |
| CN104123007A (en) | Multidimensional weighted 3D recognition method for dynamic gestures | |
| CN102609093A (en) | Method and device for controlling video playing by using gestures | |
| CN109145766A (en) | Model training method, device, recognition methods, electronic equipment and storage medium | |
| CN102316352B (en) | Stereo video depth image manufacturing method based on area communication image and apparatus thereof | |
| CN101807198A (en) | Video abstraction generating method based on sketch | |
| CN104007819A (en) | Gesture recognition method and device and Leap Motion system | |
| CN104732203A (en) | Emotion recognizing and tracking method based on video information | |
| CN103164865A (en) | Method and device of beautifying handwriting input | |
| CN106446862A (en) | Face detection method and system | |
| CN111652017B (en) | Dynamic gesture recognition method and system | |
| CN110119726A (en) | A kind of vehicle brand multi-angle recognition methods based on YOLOv3 model | |
| CN103246922B (en) | A kind of video abstraction generating method | |
| CN103413323B (en) | Based on the object tracking methods of component-level apparent model | |
| CN106204658A (en) | Moving image tracking and device | |
| CN110717154A (en) | Method and device for processing characteristics of motion trail and computer storage medium | |
| CN110738163A (en) | mine personnel illegal action recognition system | |
| CN103034851B (en) | The hand tracking means based on complexion model of self study and method | |
| CN109948450A (en) | A kind of user behavior detection method, device and storage medium based on image | |
| CN107748639A (en) | A kind of edit methods of curve, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |