CN106767757A - Nearly eye display device gesture recognition method and system based on database - Google Patents
Nearly eye display device gesture recognition method and system based on database Download PDFInfo
- Publication number
- CN106767757A CN106767757A CN201611132662.XA CN201611132662A CN106767757A CN 106767757 A CN106767757 A CN 106767757A CN 201611132662 A CN201611132662 A CN 201611132662A CN 106767757 A CN106767757 A CN 106767757A
- Authority
- CN
- China
- Prior art keywords
- display device
- light source
- image
- eye display
- nearly eye
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Position Input By Displaying (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The present invention provides a kind of nearly eye display device gesture recognition system based on database, including nearly eye display device and observation device, and the observation device is arranged on the top in located space region, and the main detection of the observation device is directed downward.Compared with prior art, light-emitting device is arranged on the present invention top of nearly eye display device, reduces the probability that the test light that light-emitting device when using sends is blocked, and increased the degree of accuracy and the continuity of measurement.A kind of method by setting up database He call database, there is provided new attitude measurement means, is independent of gyroscope and is capable of achieving attitude measurement.
Description
Technical field
The present invention relates to space orientation field, more specifically to a kind of nearly eye display device appearance based on database
State recognition methods and system.
Background technology
Space orientation is typically positioned and calculated using the pattern of optics or ultrasonic wave, is derived by setting up model and treated
Survey the locus of object.General nearly eye display device gesture recognition method and system based on database using infrared point and
The mode that light sensation camera is received determines the locus of object, and measures the angle information of object according to gyroscope, red
In the front end of nearly eye display device, in positioning, light sensation camera catches the position of infrared point and then extrapolates user exterior point
Physical coordinates.This localization method is widely used in current nearly eye display device, and the localization method as main flow has
The features such as reaction is fast, seizure is simple.But meanwhile, this localization method has certain limitation, on the one hand in position fixing process,
IR is often blocked by the pedestrian of surrounding or article, causes of short duration to position；On the other hand, this space orientation side
Method depends on gyroscope unduly, and locating effect can be then had a strong impact on if there is gyroscope drift.
The content of the invention
In order to the nearly eye display device gesture recognition method and system that solve to be currently based on database are easily blocked and excessively
The defect of gyroscope is relied on, the present invention provides a kind of nearly eye based on database of the gyroscope that is difficult to be blocked, be independent of and shows
Device gesture recognition method and system.
The technical solution adopted for the present invention to solve the technical problems is：A kind of nearly eye display dress based on database is provided
Gesture recognition method is put, is comprised the following steps：
S1：Calculate user locus coordinate；
S2：According to user locus coordinate call to should space coordinates image data base；
S3：Observation device shoots the image of nearly eye display device, and image is processed；
S4：Described image after treatment is contrasted with the described image database for calling, the number the most close with it is found out
According to storehouse image, the corresponding angle coordinate of the database images is the angle coordinate of the nearly eye display device.
Preferably, positioning start before it is described observation device determine user multiple space coordinates pose presentation data simultaneously
Storage is in described image database.
Preferably, it is PnP algorithms to calculate user locus and sit calibration method.
Preferably, the observation device shoots the image of lateral sources and longitudinal light source, and image to shooting is carried out
Treatment, the image of the image, the image of rearmounted light source and the lateral sources of the frequency separation front located light source according to light, and judge
The length of the image of the front located light source, the image of the rearmounted light source and the lateral sources image.
A kind of nearly eye display device gesture recognition system based on database is provided, including nearly eye display device and observation are filled
Put, the observation device is arranged on the top in located space region, the main detection of the observation device is directed downward.
Preferably, the nearly eye display device includes lateral sources, longitudinal light source and front panel, the lateral sources and institute
State longitudinal light source to intersect, longitudinal light source includes front located light source and rearmounted light source, the front located light source is positioned close to
One end of the front panel, the rearmounted light source is identical with the frequency of the light that the lateral sources are launched, the front located light source with
The frequency of the light of the rearmounted light source transmitting is different.
Preferably, the nearly eye display device includes front panel and two cross one another headbands, and the headband is near institute
State and be provided with the direction of front panel mark light source.
Compared with prior art, light-emitting device is arranged on the present invention top of nearly eye display device, is reduced when using
The probability that the test light that light-emitting device sends is blocked, increased the degree of accuracy and the continuity of measurement.By setting up database
A kind of method with database is called, there is provided new attitude measurement means, is independent of gyroscope and is capable of achieving attitude measurement.It is logical
The characteristics of image crossed during identification lateral sources judge characteristics of image and matching database with the method for longitudinal light source length, indirect
Go out the attitude angle information of user's head, the feature for being different from other light sources glow frequency by front located light source may determine that horizontal stroke
To light source and the direction of longitudinal light source, when preventing test, direction is chaotic influences measurement accuracy.By the side for recognizing headband length
Method judges the characteristics of image in characteristics of image and matching database, and the attitude angle information of user's head is drawn indirectly, passes through
Mark light source is set, and direction is chaotic when preventing test influences measurement accuracy.Located space area is arranged on by by observation device
The top in domain, can effectively prevent observation device from cannot be observed because being blocked, the stability of the positioning of raising.
Brief description of the drawings
Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing：
Fig. 1 is prior art gesture recognition method and system structure diagram；
Fig. 2 is prior art gesture recognition method and system principle schematic diagram；
Fig. 3 is nearly eye display device gesture recognition method and system structure diagram of the present invention based on database；
Fig. 4 is nearly eye display device gesture recognition method and system principle schematic diagram of the present invention based on database；
Fig. 5 a are to observe the not inclined image schematic diagram of user's head that device is photographed；
Fig. 5 b are the image schematic diagrames that user's head that observation device is photographed leans forward；
Fig. 5 c are the image schematic diagrames for observing user's head layback that device is photographed；
Fig. 5 d are to observe the right-oblique image schematic diagram of user's head that device is photographed；
Fig. 5 e are to observe the left-oblique image schematic diagram of user's head that device is photographed；
Fig. 6 is that nearly eye display device gesture recognition method and system second embodiment principle of the present invention based on database show
It is intended to.
Specific embodiment
In order to the nearly eye display device gesture recognition method and system that solve to be currently based on database are easily blocked and excessively
The defect of gyroscope is relied on, the present invention provides a kind of nearly eye based on database of the gyroscope that is difficult to be blocked, be independent of and shows
Device gesture recognition method and system.
In order to be more clearly understood to technical characteristic of the invention, purpose and effect, now compare accompanying drawing and describe in detail
Specific embodiment of the invention.
Fig. 1-Fig. 2 shows that prior art is based on the nearly eye display device gesture recognition method and system of database, including
Nearly eye display device 10 and observation device 15, nearly eye display device 10 include setting the front panel of the nearly front end of eye display device 10
11, multiple light-emitting devices 13 are provided with front panel 11.In position fixing process, light-emitting device 13 is to emission infrared light.Observation
The infrared light of the observation transmitting of light-emitting device 13 of device 15, and the infrared light image for arriving according to the observation judges nearly eye display device 10
Horizontal coordinate, i.e. x-axis and y-axis coordinate as shown in the figure.In the existing nearly eye display device gesture recognition method based on database
And in system, observation device 15 is generally located on front or the front position on the upper side of nearly eye display device 10 to be easy to catch red
Outside line, when the pedestrian of surrounding walked from the front of nearly eye display device 10 it is out-of-date, nearly eye display device 10 and observation device 15 between
Light path be easily cut off, causing to observe device 15 cannot judge the position of nearly eye display device 10.
Fig. 3 shows the nearly eye display dress of nearly eye display device gesture recognition method and system of the present invention based on database
20 top view is put, nearly eye display device 20 includes lateral sources 21 and longitudinal light source 22, lateral sources 21 and longitudinal light source 22
Intersect, longitudinal light source 22 includes front located light source 221 and rearmounted light source 222, front located light source 221 is positioned close to front panel 23
One end.Lateral sources 21 are identical with the frequency of the light that rearmounted light source 222 is launched, and front located light source 221 is launched with rearmounted light source 222
Light frequency it is different.Lateral sources 21 and longitudinal light source 22 can be arranged on the headband 24 of nearly eye display device 20, direction
The top of nearly eye display device 20.
Fig. 4 shows nearly eye display device gesture recognition method and system principle schematic diagram of the present invention based on database.
The present invention includes nearly eye display device 20 and observation device 25, and observation device 25 is arranged on the top in located space region, including
At least one camera, its main detection is directed downward.After user wears nearly eye display device 20, observation device 25 can
To be clearly observable the top of nearly eye display device 20.Lateral sources 21 and the light of longitudinal transmitting of light source 22 can be observed dress
Put 25 to observe, image can be processed after the observation shooting image of device 25, the frequency separation front located light source 221 according to light
Image, the image of rearmounted light source 222 and lateral sources 21 image, and judge the length of each image, each image is not
Same length data combines and can form different patterns, the unique angle information of each pattern correspondence.
Fig. 5 a -5e shows the image of the observation shooting of device 25 and the corresponding relation of user's attitude.Wherein, Fig. 5 a are
The not inclined image schematic diagram of user's head that observation device is photographed；Fig. 5 b are to observe user's head that device is photographed
The image schematic diagram for leaning forward；Fig. 5 c are the image schematic diagrames for observing user's head layback that device is photographed；Fig. 5 d are observations
The right-oblique image schematic diagram of user's head that device is photographed；Fig. 5 e are to observe user's head "Left"-deviationist that device is photographed
Oblique image schematic diagram.When there is inclination all around in certain locus head in user, its gradient correspondence
Unique observation image.It should be pointed out that user is in the case of the identical attitude of different spatial, light source is in observation device 25
In formed image be different, it is therefore necessary to first with observation device 25 determine user multiple space coordinates attitude figure
Picture data are simultaneously stored in image data base, and when in use, the space coordinates correspondence according to user finds out this in database
The corresponding image data base of space coordinates, chooses the light source closest with the image observed in corresponding image data base
Image, the corresponding angle coordinate position of the light source image is the angle coordinate position of user.
After positioning starts, user wears nearly eye display device 20, is calculated according to algorithms most in use such as PnP algorithms first
User locus coordinate, according to user locus coordinate call to should space coordinates image data base, observation
Device 25 shoots the image of nearly eye display device 20, and image is processed.By itself and the figure for calling after the completion of image procossing
As database contrast, the database images the most close with it are found out, the corresponding angle coordinate of the database images is nearly eye
The angle coordinate of display device 20.
Fig. 6 shows that nearly eye display device gesture recognition method and system second embodiment of the present invention based on database are former
Reason schematic diagram.Second embodiment of the invention is essentially identical with first embodiment, and difference is that second embodiment is real with first
Apply example setting overhead different, in a second embodiment, nearly eye display device 30 includes two cross one another headbands 34, hands over
The headband 34 of fork can be observed device 25 and observe, mark light is provided with the direction of front panel 33 with 34 in cross head
Source 39, can be processed image after the observation shooting image of device 25, and the image according to mark light source 39 distinguishes cross head band
All directions, and judge the length of each headband image, the combination image of the length according to each image simultaneously utilizes above-mentioned number
The method judged according to storehouse can be determined that the attitude of user.
Compared with prior art, light-emitting device is arranged on the present invention top of nearly eye display device 20, is reduced and is used
When the probability that is blocked of the test light that sends of light-emitting device, increased the degree of accuracy and the continuity of measurement.By setting up data
Storehouse and the method for calling database, there is provided a kind of new attitude measurement means, are independent of gyroscope and are capable of achieving attitude measurement.
Characteristics of image in judging characteristics of image and matching database by the method for recognizing lateral sources 21 and longitudinal length of light source 22,
The attitude angle information of user's head is drawn indirectly, and the feature of other light sources glow frequency is different from by front located light source 221
The direction of lateral sources 21 and longitudinal light source 22 is may determine that, direction is chaotic when preventing test influences measurement accuracy.By knowing
The method of the other length of headband 34 judges the characteristics of image in characteristics of image and matching database, and the appearance of user's head is drawn indirectly
State angle information, by setting mark light source 39, when preventing test, direction is chaotic influences measurement accuracy.By the way that device will be observed
25 tops for being arranged on located space region, can effectively prevent observation device 25 from cannot be observed because being blocked, and raising is determined
The stability of position.
Embodiments of the invention are described above in conjunction with accompanying drawing, but the invention is not limited in above-mentioned specific
Implementation method, above-mentioned specific embodiment is only schematical, rather than restricted, one of ordinary skill in the art
Under enlightenment of the invention, in the case of present inventive concept and scope of the claimed protection is not departed from, can also make a lot
Form, these are belonged within protection of the invention.
Claims (7)
1. a kind of nearly eye display device gesture recognition method based on database, it is characterised in that：Comprise the following steps：
S1：Calculate user locus coordinate；
S2：According to user locus coordinate call to should space coordinates image data base；
S3：Observation device shoots the image of nearly eye display device, and image is processed；
S4：Described image after treatment is contrasted with the described image database for calling, the database the most close with it is found out
Image, the corresponding angle coordinate of the database images is the angle coordinate of the nearly eye display device.
2. the nearly eye display device gesture recognition method based on database according to claim 1, it is characterised in that：Fixed
Position starts the preceding observation device and determines the pose presentation data of user's multiple space coordinates and store in described image data
In storehouse.
3. the nearly eye display device gesture recognition method based on database according to claim 1, it is characterised in that：Calculate
It is PnP algorithms that calibration method is sat in user locus.
4. nearly eye display device gesture recognition method and system based on database according to claim 1, its feature exist
In：The observation device shoots the image of lateral sources and longitudinal light source, and image to shooting is processed, according to light
The image of the image of frequency separation front located light source, the image of rearmounted light source and the lateral sources, and judge the front located light source
Image, the image of the rearmounted light source and the lateral sources image length.
5. a kind of nearly eye display device gesture recognition system based on database, it is characterised in that：Including nearly eye display device and
Observation device, the observation device is arranged on the top in located space region, and the main detection of the observation device is directed downward.
6. the nearly eye display device gesture recognition system based on database according to claim 5, it is characterised in that：It is described
Nearly eye display device includes lateral sources, longitudinal light source and front panel, and the lateral sources and longitudinal light source intersect,
Longitudinal light source includes front located light source and rearmounted light source, and the front located light source is positioned close to one end of the front panel, institute
State that rearmounted light source is identical with the frequency of the light that the lateral sources are launched, the light that the front located light source is launched with the rearmounted light source
Frequency it is different.
7. the nearly eye display device gesture recognition system based on database according to claim 5, it is characterised in that：It is described
Nearly eye display device includes front panel and two cross one another headbands, and the headband sets up near the side of the front panel
There is mark light source.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611132662.XA CN106767757A (en) | 2016-12-09 | 2016-12-09 | Nearly eye display device gesture recognition method and system based on database |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611132662.XA CN106767757A (en) | 2016-12-09 | 2016-12-09 | Nearly eye display device gesture recognition method and system based on database |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106767757A true CN106767757A (en) | 2017-05-31 |
Family
ID=58879554
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611132662.XA Pending CN106767757A (en) | 2016-12-09 | 2016-12-09 | Nearly eye display device gesture recognition method and system based on database |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106767757A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6324453B1 (en) * | 1998-12-31 | 2001-11-27 | Automotive Technologies International, Inc. | Methods for determining the identification and position of and monitoring objects in a vehicle |
| CN101233534A (en) * | 2005-07-27 | 2008-07-30 | 村田机械株式会社 | Operation system and operation instruction method in operation system |
| CN201385620Y (en) * | 2009-03-02 | 2010-01-20 | 徐胜 | Cab visual disturbance eliminating device |
| US20100177929A1 (en) * | 2009-01-12 | 2010-07-15 | Kurtz Andrew F | Enhanced safety during laser projection |
| CN102156537A (en) * | 2010-02-11 | 2011-08-17 | 三星电子株式会社 | Equipment and method for detecting head posture |
| CN104392241A (en) * | 2014-11-05 | 2015-03-04 | 电子科技大学 | Mixed regression-based head pose estimation method |
| JP2015087921A (en) * | 2013-10-30 | 2015-05-07 | セイコーエプソン株式会社 | Head-mounted display device and method for controlling head-mounted display device |
| US20150138069A1 (en) * | 2012-05-17 | 2015-05-21 | The University Of North Carolina At Chapel Hill | Methods, systems, and computer readable media for unified scene acquisition and pose tracking in a wearable display |
| CN105069976A (en) * | 2015-07-28 | 2015-11-18 | 南京工程学院 | Integrated fatigue detection and driving record system and fatigue detection method |
| CN105872390A (en) * | 2016-06-15 | 2016-08-17 | 南京快脚兽软件科技有限公司 | Equipment for efficiently checking peripheral environment images |
| CN106020451A (en) * | 2016-05-10 | 2016-10-12 | 罗阳军 | Positioning and movement identification system |
-
2016
- 2016-12-09 CN CN201611132662.XA patent/CN106767757A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6324453B1 (en) * | 1998-12-31 | 2001-11-27 | Automotive Technologies International, Inc. | Methods for determining the identification and position of and monitoring objects in a vehicle |
| CN101233534A (en) * | 2005-07-27 | 2008-07-30 | 村田机械株式会社 | Operation system and operation instruction method in operation system |
| US20100177929A1 (en) * | 2009-01-12 | 2010-07-15 | Kurtz Andrew F | Enhanced safety during laser projection |
| CN201385620Y (en) * | 2009-03-02 | 2010-01-20 | 徐胜 | Cab visual disturbance eliminating device |
| CN102156537A (en) * | 2010-02-11 | 2011-08-17 | 三星电子株式会社 | Equipment and method for detecting head posture |
| US20150138069A1 (en) * | 2012-05-17 | 2015-05-21 | The University Of North Carolina At Chapel Hill | Methods, systems, and computer readable media for unified scene acquisition and pose tracking in a wearable display |
| JP2015087921A (en) * | 2013-10-30 | 2015-05-07 | セイコーエプソン株式会社 | Head-mounted display device and method for controlling head-mounted display device |
| CN104392241A (en) * | 2014-11-05 | 2015-03-04 | 电子科技大学 | Mixed regression-based head pose estimation method |
| CN105069976A (en) * | 2015-07-28 | 2015-11-18 | 南京工程学院 | Integrated fatigue detection and driving record system and fatigue detection method |
| CN106020451A (en) * | 2016-05-10 | 2016-10-12 | 罗阳军 | Positioning and movement identification system |
| CN105872390A (en) * | 2016-06-15 | 2016-08-17 | 南京快脚兽软件科技有限公司 | Equipment for efficiently checking peripheral environment images |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI662433B (en) | Living body detection method and device, and identity authentication method and device | |
| CN106933343B (en) | Apparatus and method for recognizing gestures in virtual reality headset | |
| US9921052B2 (en) | Information processing device and information processing method | |
| US11024251B2 (en) | Image processing apparatus and image processing method | |
| US20160104298A1 (en) | Apparatus and method for poomsae recognition and dan promotion test of taekwondo based on skeleton of human body using depth camera | |
| US20150294158A1 (en) | Method and System for Tracking Objects | |
| US9317739B2 (en) | Information processing device and information processing method | |
| US9573052B2 (en) | Game device, control method for a game device, and non-transitory information storage medium | |
| WO2016008265A1 (en) | Method and apparatus for locating position | |
| US20120219179A1 (en) | Computer-readable storage medium, image processing apparatus, image processing system, and image processing method | |
| EP3188075A1 (en) | Apparatus and method for recognizing hand gestures in a virtual reality headset | |
| JP2006185109A (en) | Image measurement device and image measurement method | |
| US20180185703A1 (en) | Motion-gesture detection equipment, treadmill equipment and motion-gesture detection method | |
| US20230091749A1 (en) | Maintaining localization and orientation of electronic headset after loss of slam tracking | |
| CN107168378A (en) | A kind of system and method for Automatic-searching object | |
| US20190073810A1 (en) | Flow line display system, flow line display method, and program recording medium | |
| US10606372B2 (en) | System and method for interaction with a computer implemented interactive application using a depth aware camera | |
| JP2019219904A (en) | Program, recognition apparatus, and recognition method | |
| CN106774992A (en) | The point recognition methods of virtual reality space location feature | |
| JP2017169167A (en) | Search support device, search support method and search support program | |
| KR101396488B1 (en) | Apparatus for signal input and method thereof | |
| CN106767757A (en) | Nearly eye display device gesture recognition method and system based on database | |
| US11944897B2 (en) | Device including plurality of markers | |
| US20160092743A1 (en) | Apparatus and method for measuring a gaze | |
| JP6463712B2 (en) | Target search device, target search method, and target search program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170531 |
|
| WD01 | Invention patent application deemed withdrawn after publication |