A kind of positioning system and method based on more somatosensory devices
Technical field
The present invention relates to object location techniques field more particularly to a kind of positioning systems and side based on more somatosensory devices
Method.
Background technique
In the prior art, all targeted purposes of positioning system: as GPS positioning system has more long-term use to go through
History, but the technology has more apparent deviation and unstable, the in addition higher operating costs of GPS positioning system;It is such as infrared fixed
Position system is mainly used for pinpointing the touch-control identification in region, and the use of the system has significant limitation, such as to realize that region is big
Then cost is larger for the identification of area, is unfavorable for commercial operations.
Therefore, the existing technology needs to be improved and developed.
Summary of the invention
In view of above-mentioned deficiencies of the prior art, the purpose of the present invention is to provide a kind of positioning based on more somatosensory devices
System and method, it is intended to solve the problems, such as that positioning system stability is poor in the prior art.
Technical scheme is as follows:
A kind of positioning system based on more somatosensory devices, wherein include:
Several somatosensory devices are sent to control host for obtaining the skeleton data of identified object, and by the skeleton data;
Several control hosts, for the skeleton data received to be converted to location information, and by the location information
It is uploaded to server;
Server adjusts the position letter of identified object for the location information received to be converted in same world coordinate system
Breath, positions identified object;
Each somatosensory device control host corresponding with one is connected, and control host is connected with server.
The positioning system based on more somatosensory devices, wherein the somatosensory device is Kinect.
The positioning system based on more somatosensory devices, wherein the quantity of the somatosensory device is 3-6 platform.
A kind of localization method based on more somatosensory devices, wherein comprising steps of
The skeleton data of identified object is obtained by somatosensory device, and the skeleton data is sent to control host;
The skeleton data that the control host will receive is converted to location information, and the location information is uploaded to clothes
Business device;
The location information received is converted in same world coordinate system by the server, adjusts the position letter of identified object
Breath, positions identified object.
The localization method based on more somatosensory devices, wherein the location information includes displacement data and/or moves
Make data.
The localization method based on more somatosensory devices, wherein the step is obtained identified by somatosensory device
The skeleton data of object, and the skeleton data is sent to before control host further include:
Somatosensory device is in real time scanned the region in identification range, when there is object by the region then to the object into
Row track identification.
The localization method based on more somatosensory devices, wherein control host described in the step will receive
The skeleton data is converted to location information, and the location information is uploaded to server, specifically includes:
Judge whether the center of gravity of identified object shifts according to the skeleton data, when the center of gravity of identified object shifts
When, then the skeleton data is converted into displacement data;
Judge whether the limbs of identified object shift according to the skeleton data, when the limbs of identified object shift
When, then the skeleton data is converted into action data;
The displacement data, action data are uploaded to server.
The localization method based on more somatosensory devices, wherein the position that server described in the step will receive
Confidence breath is converted in same world coordinate system, is adjusted the location information of identified object, is positioned to identified object, specific to wrap
It includes:
The location information received is screened, skeleton data institute acquired in the somatosensory device nearest from identified object is chosen
The location information of conversion;
It after the location information is transformed into same world coordinate system, is transferred in corresponding scene, by adjusting identified object
Location information the identified object in scene is positioned.
The localization method based on more somatosensory devices, wherein the position that server described in the step will receive
Confidence breath is converted in same world coordinate system, is adjusted the location information of identified object, is positioned to identified object, wherein institute
The location information for adjusting identified object is stated, before positioning to identified object further include: to received location information into
Capable smooth, noise reduction process.
The utility model has the advantages that the positioning system and method provided by the present invention based on more somatosensory devices, using more stage body senses
Equipment simultaneously identifies the identified object in identification region, by the data conversion recognized at corresponding displacement data with
And action data, it is aggregated into after the displacement data and action data are uploaded onto the server in same world coordinate system, shape
At the location paths information of identified object, to realize the positioning to identified object.The positioning system has position stability height,
Positioning flexibly, is easy to commercialized feature.
Detailed description of the invention
Fig. 1 is that the present invention is based on the structural block diagrams of more somatosensory device positioning system preferred embodiments.
Fig. 2 is that the present invention is based on the flow charts of more somatosensory device localization method preferred embodiments.
Fig. 3 is that the present invention is based on the flow charts of step S20 in more somatosensory device localization method preferred embodiments.
Fig. 4 is that the present invention is based on the flow charts of step S30 in more somatosensory device localization method preferred embodiments.
Specific embodiment
The present invention provides positioning system and method based on more somatosensory devices, to make the purpose of the present invention, technical solution
And effect is clearer, clear, the present invention is described in more detail below.It should be appreciated that specific implementation described herein
Example is only used to explain the present invention, is not intended to limit the present invention.
Referring to Fig. 1, Fig. 1 is that the present invention is based on the structural block diagrams of more somatosensory device positioning system preferred embodiments, such as
Shown in Fig. 1, the positioning system comprising:
Several somatosensory devices 100 are sent to control master for obtaining the skeleton data of identified object, and by the skeleton data
Machine 200;Several control hosts 200, for the skeleton data received to be converted to location information, and by the position
Information is uploaded to server 300;Server 300 is adjusted for the location information received to be converted in same world coordinate system
The location information of whole identified object, positions identified object;Each somatosensory device 100 control master corresponding with one
Machine 200 is connected, and control host 200 is connected with server 300.
Positioning system in the prior art, has that there are stability is poor, applies in general item in short distance identification object
Upper commercial operations problem at high cost.
To solve the above-mentioned problems, the present invention provides a kind of positioning systems based on more somatosensory devices, since this is
System uses more somatosensory devices, is superimposed its identification region, extends, so as to effectively identify biggish area, meanwhile, lead to
The framework of the mature technology and server of crossing somatosensory device can be effectively stable the identified object of identification location information, improve positioning
The stability of system.Due to using more somatosensory devices, additionally it is possible to flexibly increase and decrease need to position according to the actual needs
Region.
Specifically, the control host corresponding with one of every somatosensory device in positioning system is connected, by controlling
Host processed controls the work of somatosensory device, and then every control host is connected with a unique server.
As a specific embodiment of the invention, the somatosensory device is Kinect, it is a kind of 3D body-sensing video camera,
It has imported the functions such as instant dynamic capture, image identification, microphone input, speech recognition, community interactive simultaneously.Pass through
Kinect included instant dynamic capture function, can identify the identified object in its identification region, obtain identified
The skeleton data of object.It controls host caller and skeleton data is converted into location information.Somatosensory device Kinect is set as 3-6
Platform (such as 5), i.e. control host corresponding with 5 is connected 5 Kinect respectively, 5 control hosts and unique server
Connection, the location information that server converts out control host, is aggregated into same world coordinate system, realizes nearby region
The position of interior identified object positions.
For above system, the present invention also provides a kind of localization methods based on more somatosensory devices, referring to Fig. 2, method
Comprising steps of
S10, the skeleton data that identified object is obtained by somatosensory device, and the skeleton data is sent to control host;
The skeleton data that S20, the control host will receive is converted to location information, and the location information is uploaded
To server;
The location information received is converted in same world coordinate system by S30, the server, adjusts the position of identified object
Information positions identified object.
Specifically, the quilt entered in its identification range is obtained by every somatosensory device Kinect set in system
It identifies that acquired skeleton data is transferred in corresponding control host by the skeleton data of object, every Kinect, controls host
Caller judges that the location information of identified object, control host will using the distance between skeleton point relationship and angled relationships
The location information identified is uploaded to coupled server, is converted, is aggregated into location information by server
In same world coordinate system locating for identified object, the location information of identified object is adjusted, identified object is positioned.
Further, in the present embodiment, the location information includes displacement data and/or action data.The somatosensory device
Kinect is in real time scanned the region in identification range, is then tracked to the object when there is object by the region
Identification, until identified object leaves identification region.
Referring to Fig. 3, the step S20 includes:
S201, judge whether the center of gravity of identified object shifts according to the skeleton data, when the center of gravity of identified object occurs
When offset, then the skeleton data is converted into displacement data;
S202, judge whether the limbs of identified object shift according to the skeleton data, when the limbs of identified object occur
When offset, then the skeleton data is converted into action data;
S203, the displacement data, action data are uploaded to server.
Before and after displacement data by obtaining identified object may determine that whether identified object is made in identification region
The movement of left and right, and the action data by obtaining the identified object may determine that identified object is not do corresponding move
Make.The position that can more accurately perceive identified object is obtained by displacement data to identified object and action data
It sets, it is more convenient that identified object is positioned.
Referring to Fig. 4, the step S30 includes:
S301, the location information received is screened, chooses bone acquired in the somatosensory device nearest from identified object
The location information that data are converted;
S302, it after the location information is transformed into same world coordinate system, is transferred in corresponding scene, by adjusting being known
The location information of other object positions the identified object in scene.
Wherein, when Kinect identifies identified object, since multiple devices work at the same time, every Kinect is adopted
The data collected have differences, and need to screen it.It can be by detecting between identified object and Kinect device
Distance, the data that Kinect for selecting the identified object of distance close is identified are as trust data.
Below with the citing of VR suspension bridge game item, to the positioning system provided by the present invention based on more somatosensory devices
And method is further explained explanation.
For Kinect, identification effective coverage is certain.There are two types of its region descriptions: Physical
Limits (physical limit) and Sweet spot (zone of comfort), the actual ability that the former represents inductor can be seen with him
Real estate, the latter represents the best interaction area of human body.Its recognition mode has including two kinds: i.e. half body mode and whole body mould
Formula uses whole body mode in this game item.For its default mode (whole body mode): physical limit range: 0.8 to 4
Rice;Comfort zone range: 1.2 to 3.5 meters.In addition, the visual angle Kinect size: level angle: 57.5 degree;Vertical angle: 43.5 degree.
For VR suspension bridge game item, since suspension bridge is 4 meters, high 2 meters or so long, so we can use whole body mode, with comfort zone
Range is standard, its identification region can be made to be superimposed along suspension bridge direction word order using 3 Kinect.
Then, when player walks on suspension bridge, Kinect can collect the skeleton data of player, and will be adopted
The player's skeleton data collected is transferred to control host, for cognizance code, is mainly closed using the distance between skeleton point
It is to judge which kind of behavior player is with angled relationships human skeleton data.It is exactly up and down generally in the game item
Variation, if center of gravity moves to left, and distance is more than given threshold value (such as 0.1m), is considered as target body and moves to left behavior, equally
, if center of gravity moves to right, and distance is more than given threshold value (such as 0.1m), is considered as target body and moves to right behavior.If weight
The heart does not move, that does not just move to left and move to right certainly behavior.Equally, when player makes squatting motion, whether detection leg
There is bending to a certain extent, judges the angle of 3 buttocks, knee, ankle lines whether less than 160 degree, if it is less, saying
Bright behavior of squatting down.
Because Kinect can get the skeleton data of player, the i.e. same movement 3 of same position respectively a certain position 3
It is had differences between data acquired in platform Kinect, selects bone acquired in nearest that Kinect with a distance from player at this time
Bone data are trust data, and the displacement data and action data which is converted are uploaded to as master data
Server, server are scaled in same world coordinate system after carrying out smooth, noise reduction process to received data, form quilt
It identifies the location paths information of object and is transferred to inside suspension bridge game, so that the player role inside suspension bridge game maintained like
Movement.In addition, always having Kinect all the time recognizes player, is not in Dead Core Problems, player is allowed to feel oneself
It is smoothly to move, the final positioning realized to player.
In conclusion the positioning system and method provided by the invention based on more somatosensory devices, is set by more stage body senses
It is standby that identified object in identification region is identified, identify the skeleton data of identified object, and the skeleton data is turned
It changes location information into, then the location information is uploaded to server, server carries out smooth, noise reduction to received data
It is scaled in same world coordinate system, forms the location paths information of identified object and is transferred to inside game after processing, so that
The movement that role corresponding with identified object inside game maintains like.Due to the position letter detected by multiple somatosensory devices
Breath is filtered and parses, and determines the higher positional value of confidence, carries out reduction data precision simultaneously for the position data
Etc. a series of processing, to effectively improve the stability of positioning system.
It should be understood that the application of the present invention is not limited to the above for those of ordinary skills can
With improvement or transformation based on the above description, all these modifications and variations all should belong to the guarantor of appended claims of the present invention
Protect range.