Dynamic hand language recognition device based on KinectTechnical field
The present invention relates to the application processing of Kinect, and in particular, to the dynamic hand language recognition device based on Kinect.
Background technique
With the fast development of computer technology in recent years, human-computer interaction becomes the important of computer technology research fieldComponent part.This technology mainly includes the identification of movement to the mankind, language, the identification of expression and sign language.Wherein sign language is knownThe a part not understood as Body Languages, there is an extremely important effect, on the one hand it be between deaf-mute and deaf-mute andMain exchange way between normal person, the important channel of another aspect human-computer interaction.
Traditional sign Language Recognition Method mainly has the methods of template matching, neural network, HMM, but these methods all existCertain deficiency needs to obtain the observation number in sampling feature vectors such as in training HMM model, and for differentIts depth image of sign language changes greatly, it is not easy to determine observation number.And it is frequent for the method for traditional Sign Language RecognitionIt is single use, this makes the precision of Sign Language Recognition and user experience not high.
Have at present by using data glove and obtain the feature and deliberate action of human hands, although by this methodAvailable higher recognition accuracy, but data glove has using complicated and expensive deficiency for nowPoint.
Summary of the invention
The object of the present invention is to provide a kind of the dynamic hand language recognition device based on Kinect, the dynamic based on KinectHand language recognition device overcomes that resolution in the prior art is low, problem complicated for operation, realizes resolution height, easy to operateEffect.
To achieve the goals above, the present invention provides a kind of dynamic hand language recognition device based on Kinect, this is based onThe dynamic hand language recognition device of Kinect includes: voice broadcast equipment, driving mechanism, robot body, Kinect device and controlCenter processed；Wherein, the voice broadcast equipment, driving mechanism and Kinect device are electrically connected to the control centre, institute's predicateSound casting equipment, driving mechanism, Kinect device and control centre are set on the robot body；The driving mechanism bandMove the robot body movement；The Kinect device carries out the identification of human body sign language, and is held according to the sign language of the human bodyRow voice broadcast.
Preferably, the robot body includes: top half and lower half portion；Wherein, the top half is set up inThe top of the lower half portion, and the voice broadcast equipment is set to the top of the top half, the Kinect deviceIt is set in the top half, and be collected towards multiple directions, the driving mechanism is set to the lower half portionBottom.
Preferably, the driving mechanism includes: four-wheel differentia driving machine, and the four-wheel differentia driving machine is set to the machineThe lower section of device human body to drive the movement of the robot body, and is electrically connected to the control centre, described in receivingThe control signal of control centre.
Preferably, the quantity of the Kinect device is multiple, and the dynamic sign language of the Kinect device acquires directionTowards multiple directions.
Preferably, the accommodating chamber matched with the control centre, and the receiving are provided on the robot bodyChamber being capable of backstop rainwater.
Preferably, the upper surface of the bottom of the robot body is additionally provided with battery rack, to hold battery, instituteIt states battery and is electrically connected to the voice broadcast equipment, Kinect device, driving mechanism and the control centre, to provide workVoltage.
Preferably, it is somebody's turn to do the dynamic hand language recognition device based on Kinect further include: radio transmitting device, the wireless transmissionDevice is connected to the control centre, and the human body sign language identified is wirelessly transmitted to outside.
Preferably, the Kinect device obtains human skeleton data and relevant depth information obtains hand position, benefitHand is detected and is tracked with Kinect, then extracts hand-characteristic, and pre-processed, finally extract gesture feature toParameter is measured, and carries out the dynamic Sign Language Recognition based on HMM and SVM.
According to the above technical scheme, the more traditional recognition methods of dynamic hand language recognition device of the invention has equipment costLower, easy to operate, accuracy of identification is higher, the preferable feature of user experience.More traditional dynamic sign Language Recognition Method, this hairThe bright method using HMM model and SVM models coupling, improves the accuracy of identification；Equipment used in the present invention isKinect has the characteristics that cost is relatively low and easy to operate compared to other relevant devices.
Other features and advantages of the present invention will the following detailed description will be given in the detailed implementation section.
Detailed description of the invention
The drawings are intended to provide a further understanding of the invention, and constitutes part of specification, with following toolBody embodiment is used to explain the present invention together, but is not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is the structural schematic diagram for illustrating a kind of indoor acoustic location device based on Kinect of the invention.
Description of symbols
1 robot body, 2 driving mechanism
11 top half, 12 lower half portion
3 voice broadcast equipment
Specific embodiment
Below in conjunction with attached drawing, detailed description of the preferred embodiments.It should be understood that this place is retouchedThe specific embodiment stated is merely to illustrate and explain the present invention, and is not intended to restrict the invention.
In the present invention, in the absence of explanation to the contrary, the noun of locality be included in the term " up and down " etc. is onlyRepresent orientation of the term under normal service condition, or be those skilled in the art understand that be commonly called as, and be not construed as to thisThe limitation of term.
The present invention provides a kind of dynamic hand language recognition device based on Kinect, which is characterized in that should be based on Kinect'sDynamic hand language recognition device includes: voice broadcast equipment, driving mechanism, robot body, Kinect device and control centre；ItsIn, the voice broadcast equipment, driving mechanism and Kinect device are electrically connected to the control centre, and the voice broadcast is setStandby, driving mechanism, Kinect device and control centre are set on the robot body；The driving mechanism drives the machineDevice human body movement；The Kinect device carries out the identification of human body sign language, and executes voice according to the sign language of the human body and broadcastReport.
According to the above technical scheme, the more traditional recognition methods of dynamic hand language recognition device of the invention has equipment costLower, easy to operate, accuracy of identification is higher, the preferable feature of user experience.More traditional dynamic sign Language Recognition Method, this hairThe bright method using HMM model and SVM models coupling, improves the accuracy of identification；Equipment used in the present invention isKinect has the characteristics that cost is relatively low and easy to operate compared to other relevant devices.
In a kind of specific embodiment of the invention, in order to realize upper and lower differentiation, facilitate the fixation of Kinect device,The robot body may include: top half and lower half portion；Wherein, the top half is set up in the lower half portionTop, and the voice broadcast equipment is set to the top of the top half, and the Kinect device is set on describedIt in half part, and is collected towards multiple directions, the driving mechanism is set to the bottom of the lower half portion.
In a kind of specific embodiment of the invention, the driving mechanism may include: four-wheel differentia driving machine, describedFour-wheel differentia driving machine is set to the lower section of the robot body, to drive the movement of the robot body, and is electrically connectedIt is connected to the control centre, to receive the control signal of the control centre.By above-mentioned mode, may be implemented to turn to fortuneIt is dynamic, driving movement may be implemented.
In a kind of specific embodiment of the invention, in order to realize the video data acquiring of multi-angle, the KinectThe quantity of device is multiple, and the dynamic sign language of the Kinect device acquires direction towards multiple directions.
In a kind of specific embodiment of the invention, in order to accommodate control centre, it is prevented to be soaked with rain, the machineThe accommodating chamber matched with the control centre is provided on device human body, and the accommodating chamber being capable of backstop rainwater.
In a kind of specific embodiment of the invention, in order to facilitate the placement of battery, the bottom of the robot bodyThe upper surface in portion is additionally provided with battery rack, to hold battery, the battery be electrically connected to the generation source device,Kinect device, driving mechanism and the control centre, to provide operating voltage.
In a kind of specific embodiment of the invention, being somebody's turn to do the dynamic hand language recognition device based on Kinect can also be wrappedInclude: radio transmitting device, the radio transmitting device are connected to the control centre, and the human body sign language identified is wirelessly passedTransport to outside.
By above-mentioned mode, the wireless transmission of data may be implemented, data can be allowed to transfer out.
In a kind of specific embodiment of the invention, the Kinect device obtains human skeleton data and correlation is deepDegree information obtains hand position, and hand is detected and tracked using Kinect, then extracts hand-characteristic, and located in advanceReason finally extracts gesture feature vector parameter, and carries out the dynamic Sign Language Recognition based on HMM and SVM.
The present invention is used for human-computer interaction.For exchanging between deaf-mute and between deaf-mute and normal person.And it can useIn deaf-mute or normal person's training sign language.
It is described the prefered embodiments of the present invention in detail above in conjunction with attached drawing, still, the present invention is not limited to above-mentioned realitiesThe detail in mode is applied, within the scope of the technical concept of the present invention, a variety of letters can be carried out to technical solution of the present inventionMonotropic type, these simple variants all belong to the scope of protection of the present invention.
It is further to note that specific technical features described in the above specific embodiments, in not lanceIn the case where shield, can be combined in any appropriate way, in order to avoid unnecessary repetition, the present invention to it is various canNo further explanation will be given for the combination of energy.
In addition, various embodiments of the present invention can be combined randomly, as long as it is without prejudice to originallyThe thought of invention, it should also be regarded as the disclosure of the present invention.