Summary of the invention
Based on the problems referred to above, it is an object of the invention to provide a kind of human dissection interactive system based on virtual reality,
Set up complete three-dimensional virtual environment by VR, make user and anthropometric dummy interact operation.This system includes:
Head-mounted display apparatus, is worn on user's head, makes user pass through the display in described head-mounted display apparatus
Viewing virtual environment and 3 D human body mathematical model;
Host computer system, for for described head-mounted display apparatus provide function ensure, described head-mounted display apparatus according to
Self is intelligent and automaticity utilizes host computer system;
Tracing system, for catching the motion of user, and returns to information described head-mounted display apparatus, thus creates
A kind of immersion VR experiences;
Controller, described controller is handheld device, for following the trail of action or the gesture of user;
Auxiliary visible system, including: visualization bulletin board and dynamic visual effect generator;
Described system also includes: by certain in host computer system real-time judge sensor location and virtual scene
Organization of human body position, if any contact or collision, then carries out different dynamics feedbacks according to the character of contact or colliding object.
Preferably, described tracing system includes built-in sensors, gyroscope and magnetometer, described sensor and HMD device
Separate, use Internal camera head or LED.
Preferably, described controller is also equipped with sense of touch trigger module, it is provided that tactile experience, makes user feel vibrations.
Preferably, described controller includes:
Left handle, is controlled described system by the control station of described left handle;
Right handles, is controlled described system by the functional keys of described right handles.
Preferably, described left handle includes:
VR receptor, gyroscope equipment and one piece of solid plate, solid plate and VR receptor form integral structure and hold in a left side
Hands, carries out level by gyroscope equipment and upset judges, breathes out the control in virtual scene when left handle is in level
Platform processed, closing control platform when left handle departs from level.
Further, the functional keys of described right handles includes: trigger, button, left direction button, right direction button.
Preferably, described trigger includes:
Trigger: trigger includes function activation button, can operate the control station of left handle,
Wherein, described operation includes, carries out capturing work, further to after arbitrary model contact of 6 groups of anthropometric dummies:
Press trigger to determine to be in and catch state, be in and catch the model under state can rotate with the rotation of handle;
Unclamping trigger and cancel crawl, model rests on changing coordinates point.
Preferably, described button has the model being contacted handle and carries out single restoring function.
Preferably, described LR-button has the model group being in cursor circle carries out left rotation and right rotation function, all in group
Model starts to rotate along scene Y coordinate.
Preferably, the characteristic of left and right handle includes: when moving coordinate in virtual scene, the attribute of detection collision in real time.
Preferably, described visualization bulletin board, when any model is in seized condition, is sat in the crawled model world
The target upper left corner occurs that transparent mode describes interface, shows the Chinese of current crawled model, English name, and other in interface
Model associated description.
Preferably, described dynamic visual effect generator is when all models stand on original coordinate points in scene, for dynamic
States model renders dynamic effect.
Preferably, the structure of described 3 D human body mathematical model comprises the following steps:
1) human medical image is obtained;
2) use Volume Rendering Techniques, set up human digit solution to model and cut open data base;
3) instructed by medical expert, according to individual variation and the deficiency of human body, adjust the deviation of human digit model, obtain
Obtain human digit model accurately;
4) by manual drawing step 3) the human digit model that obtains, set up organization of human body model accurately.
Preferably, step 2) set up mathematical model for build respectively according to men and women's size difference.
Preferably, the structure of above-mentioned human digit model uses time generation 3D digital modeling method.
Described system also includes: in order to reduce the spinning sensation of user in VR environment, use following arbitrary adjustment mode or its
Combination:
Static object of reference is added within sweep of the eye user;Suitably reduce the field range of user or avoid the frequent of the visual field
Conversion;Increase the refreshing frame per second of display picture;Prompting user is at the uniform velocity movable in VR environment.
According to above-mentioned each mode, the present invention can make user obtain more fine interactive experience accurately.In virtual scene
User is it will be seen that have and the 3D somatic data of normal person's peer body lattice, such as height, chest measurement, upper lower part of the body ratio etc.;With
Time directly can disassemble fractionation to organization of human body arbitrary in virtual scene by controller, and without again by extra
External device.Further, it is different from the single visual stereoscopic of existing virtual reality system or sense of touch three-dimensional, or haptic effect is simple
Simulation, the present invention is directed to different tissues and carries out different dynamics feedbacks, make user experience the sense of touch of different tissues, knot simultaneously
Structure character, form and suppleness etc..
Further, the model resource of the present invention breaks traditions three-dimensional reconstruction, is drawn by twice organization of human body,
Converting primitive medicine data, utilization time 3D digital modeling techniques from generation to generation and PBR workflow make and obtain accurate, standard
The anthropometric dummy structure changed.
Additionally, the present invention is from user's own experience, it is contemplated that the persistency that user uses, have employed effective adjustment
Measure, prevents and reduces the spinning sensation of user.
Detailed description of the invention
Hereinafter, with reference to the accompanying drawings to describe embodiments of the invention in detail.
Fig. 1 is present system structural representation.
Wherein, host computer system includes smart mobile phone, PC or video game machine.This medical anatomy interactive system tracing system can
The VR creating a kind of immersion experiences, such as, if user puts on HMD device and looks at the sky, HMD screen also can be to same side
To motion, in order to allow user see target.Tracing system can be integrated in HMD device, or the peripheral hardware as HMD device.For
Improve the accuracy of tracing system, sensor in tracing system can be separated with HMD device, and use Internal camera head or
LED.
Controller includes left and right handle, and wherein left handle includes: VR receptor and one piece of solid plate, the size of solid plate is
37cm*25cm.Solid plate requires light, and gravity is lighter.
Solid plate and left hand VR receptor form integral structure and hold in left hand, carry out water by the gyroscope equipment of handle
Gentle upset judges, breathes out the control station in virtual scene when handle is in level, and handle closes when departing from level
Close control station.In virtual scene, the control station of display is tabular operating board, and control station at least includes 6 model icons, point
Not: skeleton model, muscle model, Visceral models, Blood Model, neural model, lymph model, user can pass through right handles
The icon shown with control station interacts, and the icon such as shown control station selects, switches.Additionally, Visceral models
Translatable copying places single eyeball phantom and a complete set of brain model by a heart model, neural model.Meanwhile, by a left side
Handle has dynamic effect when activating control station and closing control platform.
The control station of left handle also includes that model button, model button have: activation, un-activation, three kinds of states of hovering,
While activator button, recall the model structure of correspondence, otherwise, activate model structure and then close.Have under each model button
One labeled button (totally 6), labeled button represents and chooses this model group.Labeled button is followed user and is clicked on, and judges
Now under which model group.By activate mark circle shown by labeled button can only by one single presented in.Simultaneously
Program is that each model group puts a box, utilizes right hand handle, utilizes what handle release went out to penetrate while pressing directionkeys
Line detects which group model chosen and rotates, and plays the function of mark circle.
Meanwhile, model button also includes 1 overall " Integral Restoration " button, can enter the model group being under mark circle
Row entirety reduction.Model button can also pass through ray detection model group, stirs other function key reset block mold groups.
Fig. 2 is the schematic diagram of right handles.
The function activation of right handles controls to include: trigger, button, left direction button, right direction button.
Wherein, trigger includes a function activation button, can operate the control station of left handle.Operation includes: to 6 groups
Carry out capturing work after arbitrary model contact of anthropometric dummy, when pressing trigger and determining to be in and catch state, be in and catch shape
Model under state can rotate with the rotation of handle, unclamps trigger and cancels crawl, and model rests on changing coordinates point.
Button, has the model being contacted handle and carries out single restoring function.
LR-button has the model group being in cursor circle carries out left rotation and right rotation function, and in group, all models are along scene Y
Coordinate starts to rotate.
Handle takes the form of true hands, or glove modalities.In grasping movement and click button actions, right handles
Form can in standby, grasping movement, refer to select in action and switch over.
The characteristic of left and right handle includes: when moving coordinate in virtual scene, the attribute of detection collision in real time.
Corresponding model information will be read when contacting or collide human tissue structure, and judge according to model information, and be given
FORCE FEEDBACK corresponding to corresponding model classification pulls the corresponding informance of power.FORCE FEEDBACK passes to hand receptor, and user is according to contact
Or the character of colliding object obtains different FORCE FEEDBACK information, the texture and the dynamics that reach real world stop sensation.By closing
The combination of reason, different force feedback technique, it is achieved feed back to the dynamics that user is different after human body different structure is analyzed
Conflict and experience.Certain human body locations of structures in sensor location and virtual scene is sentenced in real time by computer
Disconnected, if any collision, then carry out different dynamics feedbacks according to the character of colliding object.As collided obstruction dynamics and the fat of skeleton
Obstruction dynamics different, or the thickness sense of contact blood is the most different.
Auxiliary visible system, auxiliary visible system includes: visualization bulletin board and dynamic visual effect generator.
Visualization bulletin board, when any model is in seized condition, goes out in the upper left corner of crawled model world coordinates
Existing transparent mode describes interface, shows the Chinese of current crawled model, English name in interface, and other models are correlated with and are retouched
State.
Dynamic visual effect generator, when all models stand on original coordinate points in scene, renders dynamic for dynamic model
State effect.Human body has the organ being much kept in motion, the heart of state of such as beating, the lobe of the lung in breathing state, blood
Blood in Guan Liyou flowing, enterokinesia and gastric peristalsis etc..For increasing the mutual innervation in scene, when certain model departs from original seat
During punctuate, the model departing from coordinate points stops dynamic effect.
Fig. 3 is schematic diagram in 3D Digital human model, and 3D Digital human model is built by following steps:
1) obtain radiology department medical image, as cadaver sample resources bank and abundant human body CT MRI scan data etc.;
2) utilize 3D imaging system, use Volume Rendering Techniques based on light projecting algorithm, set up 3D Digital human model
Anatomical data storehouse, overlap three-dimensional digital model with complete and comprehensive, authoritative, fine anatomical data output men and women two;
3) instructed by medical expert, according to individual variation and the deficiency of human body, adjust the deviation of human digit model, obtain
Obtain human digit model accurately;
4) through the manual drawing step 3 of some fine arts personnel) model that obtains, set up out each a set of standard of men and women
Organization of human body.
Above-mentioned human digit model uses time generation 3D digital modeling techniques, and the establishment face number of average each model is 6000
(Mesh), on, texture mapping uses the full accuracy of 4096 (px) x 4096 (px) to make standard, to ensure each model
Fully reduction human body each system precision details, and arbitrarily can observe with 360 degree of angles.
By the 3D digital human body anatomical structure of dual drafting compared to the organization of human body only obtained by scanning, have more
Research and teaching.The threedimensional model directly reconstructing and being applied to system after scanning has individuation difference, and such human body is tied
Structure model can only represent body one by one, it is impossible to a capping.And the human digit model of Freehandhand-drawing is by somatic data storehouse, data
Obtain after many-sided resource consolidation such as scanning, three-dimensional reconstruction, expert's guidance, be that medical science most presses close to idealization and Medicine standard
Digital human body structural model.
It addition, in virtual scene, digital human body structural model can make user more finely interact body accurately
Test, user in virtual scene it will be seen that have and the 3D somatic data of normal person's peer body lattice, as height, chest measurement, on
Lower part of the body ratio etc..Directly can disassemble fractionation, no to organization of human body arbitrary in virtual scene by controller simultaneously
With again by extra external device.
Further, the model resource of the present invention breaks traditions three-dimensional reconstruction, by CG manufacturing technology, applies to
Medical education field.By primitive medicine data are converted, use time generation 3D digital modeling techniques and PBR workflow (base
Rendering in physical process) make accuracy, standardized anthropometric dummy structure.
Use primitive medicine data genaration primary mold, use time 3D digital modeling techniques drafting from generation to generation by modeling software
High-precision human model structure, organization of human body can be carried out high accuracy reduce and not by face numerical control system.Reference in conversion process
Many-sided specialized medical data, various dimensions, control model accuracy and standardization, set up out the organization of human body of perfect standard type.
Use PBR workflow, based on physical process render, simulate true illumination, make 4096 (px) x 4096
(px) high accuracy pinup picture.The organ structure drawn out is more nearly real human body, avoids conventional cadaver sample simultaneously and is difficult to
The drawback such as preserve, structure is unintelligible and live body difference is big.
The principle of secondary generation 3D model is to be mated by Gao Mo and low mould to produce multiple details pinup pictures, record in described pinup picture
Detailed information contained in high mould, and the normal pinup picture bakeed by high mould replies to the topic on low mould, to reduce the waste in face, simultaneously
Low mould page can show the visual effect of Gao Mo in engine in time, thus makes model performance more true, reaches movie-level essence
Degree, it is achieved the effect of high details, strong texture performance.
Above-mentioned model can be applicable to teaching shows and dissects training that the content that can require according to syllabus makes multiple
Sectional model, student can be by transparent or hiding operation, and observing and nursing internal anatomy and Morphological Features, to emphasis, difficulty
Point study position is simulated the operation of real anatomy, helps student to understand and grasps knowledge point, also improving teacher's simultaneously
Quality of instruction.
Fig. 4 and Fig. 5 is that the force feedback in present system describes schematic diagram.
As shown in Figure 4 and Figure 5, by the combination of rational, different force feedback techniques, it is achieved human body different structure is entered
Row feeds back to user different dynamics conflict after analyzing is experienced.Certain human body knot in sensor location and virtual scene
Structure position carries out real-time judge by computer, if any collision, then carries out different dynamics feedbacks according to the character of colliding object.
As different with the obstruction dynamics of fat in collided the obstruction dynamics of skeleton, or the thickness sense of contact blood is the most different.
Wherein, Fig. 4 shows the flow chart obtaining force feedback.
User touches anthropometric dummy by mobile both hands, when touching, activates and captures function, and right handles is in crawl shape
State, when user moves both hands again, just can show the movement locus capturing model further, obtain according to the action of user
The information of the corresponding model in human body or display capture the state of model.
Wherein, when user's both hands touch anthropometric dummy, carry out collision detection, and detection information is sent back to main frame system
System, to calculate feedback dynamics further, and returns to sense of touch trigger module by the dynamics of feedback, and sense of touch trigger module provides sense of touch
Experience, make user feel vibrations.
When user activates crawl function and is in seized condition, for dynamic organ, such as heart, blood vessel etc., pass through
Both hands coordinate and speed, calculate the force feedback information that human tissue structure is corresponding, and the force information of model returned to sense of touch
Trigger module, it is further provided tactile experience.
Fig. 5 further illustrates force feedback signal input/output relation in the system.
Virtual scene models show object receives the movement locus coordinate of hand exercise generation and carries out the dynamic posture of hand
Identify.Force feedback signal is by extracting counter stress feedback Resistance Pressure information and being sent to virtual scene mould after FORCE FEEDBACK signal processing
Type display object.The signal of head display simultaneously obtains the move of user by head display signal processing flow, also by it
It is sent to virtual scene models show object.Virtual scene models show object calculates the power to hand position further and returns
Feedforward information, and it is returned to sense of touch trigger module, it is further provided tactile experience.
In VR film, the image of acute variation is had easily to cause the spinning sensation of user in space.Meanwhile, these changes
Always attract the attention of user.With reference to Fig. 6, in the visual field of user, first area is that most suitable user observes, the secondth district
Territory belongs to remaining light region, the visual field, and user can be attracted because of the change of yellow area generation, thus changes Visual Observations Observations position.
3rd region is the insignificant part of vision, only when having picture generation acutely to reflect or the judgement of brain subconsciousness is dangerous information
People just can be made to have reflected.And now, the image of edge can be left in the basket, if reducing the visual field, Consumer's Experience is affected not
Greatly.Significantly picture change and subconscious danger signal source can be avoided in processing procedure so building at scenic picture,
Oneself is all the time in a safety and quiet environment to make brain think.Such environment avoids user significantly or quickly to move
Animation face, thus reduce user and produce the situation of dizzy effect.Therefore, can be by suitably reducing the user visual field (FOV), suitably subtracting
Lack the field range of user or avoid the frequent conversion etc. in the visual field, reducing the spinning sensation of user, strengthening Consumer's Experience.
It addition, VR equipment inter-frame delay does not catches up with the motion of people, when people has accomplished that corresponding motion picture does not also have
Refresh out, have small delay sense, i.e. sense organ and can produce spinning sensation with frame per second is asynchronous.If increasing equipment to picture
Refresh frame per second, as reached higher level (60Hz etc.), then user self moves at moving-head and virtual environment picture
Time, the delay of picture can be greatly reduced, to such an extent as to user is hardly it can be seen that effectively reduce spinning sensation.The present invention utilizes special
Processing mode reduce model face number while ensure that the precision of model, thus increase the refreshing frequency in virtual scene.
Can also user increase within sweep of the eye the health of oneself, a sterilizing room helmet or static object of reference frame (as
Head up displays) as object of reference, it is effectively reduced spinning sensation, feeling of immersion can also be increased in some cases.