CN109938973A - A kind of visually impaired person's air navigation aid and system - Google Patents
A kind of visually impaired person's air navigation aid and system Download PDFInfo
- Publication number
- CN109938973A CN109938973A CN201910247664.0A CN201910247664A CN109938973A CN 109938973 A CN109938973 A CN 109938973A CN 201910247664 A CN201910247664 A CN 201910247664A CN 109938973 A CN109938973 A CN 109938973A
- Authority
- CN
- China
- Prior art keywords
- navigation
- information
- visually impaired
- model
- navigation model
- 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
Landscapes
- Navigation (AREA)
Abstract
The embodiment of the present invention provides a kind of visually impaired person's air navigation aid and system, and wherein method includes: the present image for obtaining vision sensing equipment acquisition；By the present image and current path information input to navigation model, the posture information of the navigation model output is obtained；The navigation model is obtained based on the training of sample image, sample path information, sample posture information and sample identification；Voice Navigation is carried out to visually impaired person based on the posture information.Method and system provided in an embodiment of the present invention, by by present image and current path information input into navigation model, posture information is obtained to navigate to visually impaired person, navigation is refine to the guidance to visually impaired person's movement posture, ensure that visually impaired person can receive detailed navigation information, it can walk along guidance path under foreign environment or complex environment and the avoiding obstacles that succeed, ensure that the walking safety of visually impaired person, expand the application range of visually impaired person's navigation.
Description
Technical field
The present embodiments relate to technical field of computer vision more particularly to a kind of visually impaired person's air navigation aid and it is
System.
Background technique
Dysopia drastically influences the life of visually impaired person, especially restricts the outdoor activities of visually impaired person.How to protect
The safe independent ambulation of visually impaired person is demonstrate,proved, is had received widespread attention.
With the development of airmanship, existing visually impaired person's air navigation aid can pass through GPS (Global
Positioning System, global positioning system) low cost, high accuracy three-dimensional position are provided, blind person place is obtained in real time
The latitude and longitude coordinates in geographical location, and view-based access control model sensing device obtains scene at the moment, carries out real time distance and orientation judgement, into
And navigation information is obtained, and navigation information is informed by visually impaired person by voice, realize the navigation for being directed to visually impaired person.
However, the navigation information that above-mentioned visually impaired person's air navigation aid is supplied to visually impaired person is extremely limited, can only support
Visually impaired person walks in the case where being familiar with environment.When under foreign environment or complex environment, visually impaired person is difficult to limited
Safe independent ambulation under navigation information.
Summary of the invention
The embodiment of the present invention provides a kind of visually impaired person's air navigation aid and system, to solve existing visually impaired person's navigation
The navigation information that method can provide is limited, the small problem of application range.
In a first aspect, the embodiment of the present invention provides a kind of visually impaired person's air navigation aid, comprising:
Obtain the present image of vision sensing equipment acquisition；
By the present image and current path information input to navigation model, the posture of the navigation model output is obtained
Information；The navigation model is obtained based on sample image, sample path information, sample posture information and sample identification training
's；
Voice Navigation is carried out to visually impaired person based on the posture information.
Second aspect, the embodiment of the present invention provide a kind of visually impaired person's navigation system, including portable intelligent terminal and vision
Sensing device；Wherein, the portable intelligent terminal is for executing the method as provided by first aspect.
The third aspect, the embodiment of the present invention provide a kind of visually impaired person's navigation device, comprising:
Image acquisition units, for obtaining the present image of vision sensing equipment acquisition；
Posture acquiring unit, for by the present image and current path information input to navigation model, described in acquisition
The posture information of navigation model output；The navigation model be based on sample image, sample path information, sample posture information and
Sample identification training obtains；
Navigation elements, for carrying out Voice Navigation to visually impaired person based on the posture information.
Fourth aspect, the embodiment of the present invention provide a kind of electronic equipment, including processor, communication interface, memory and total
Line, wherein processor, communication interface, memory complete mutual communication by bus, and processor can call in memory
Logical order, to execute as provided by first aspect the step of method.
5th aspect, the embodiment of the present invention provide a kind of non-transient computer readable storage medium, are stored thereon with calculating
Machine program is realized as provided by first aspect when the computer program is executed by processor the step of method.
A kind of visually impaired person's air navigation aid and system provided in an embodiment of the present invention, by by present image and current path
Information input obtains posture information and navigates to visually impaired person, navigation is refine to and is moved to visually impaired person into navigation model
The guidance to gesture, it is ensured that visually impaired person can receive detailed navigation information, even if under foreign environment or complex environment
Also it can walk along guidance path and the avoiding obstacles that succeed, ensure that the walking safety of visually impaired person, expand visually impaired person
The application range of navigation.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the flow diagram of visually impaired person's air navigation aid provided in an embodiment of the present invention；
Fig. 2 is the structural schematic diagram of visually impaired person's navigation system provided in an embodiment of the present invention；
Fig. 3 is the structural schematic diagram of visually impaired person's navigation device provided in an embodiment of the present invention；
Fig. 4 is the structural schematic diagram of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
Existing visually impaired person's air navigation aid be capable of providing it is extremely limited to the navigation information of visually impaired person, be only capable of support view
Barrier personnel walk in the case where being familiar with environment, and usable range is small.It is led in view of the above-mentioned problems, the embodiment of the present invention provides a kind of visually impaired person
Boat method.Fig. 1 is the flow diagram of visually impaired person's air navigation aid provided in an embodiment of the present invention, as shown in Figure 1, this method
Executing subject can be the portable intelligents terminal such as smart phone, Intelligent bracelet, can also be cloud server, the present invention to this not
Make specific limit.This method comprises:
Step 110, the present image of vision sensing equipment acquisition is obtained.
Herein, vision sensing equipment can be monocular vision sensing device, be also possible to be integrated with two or more views
Feel the binocular vision sensing device or multi-vision visual sensing device of sensing device.Vision sensing equipment specifically can be CMOS
(Complementary Metal Oxide Semiconductor, complementary metal oxide semiconductor) camera, can also be red
Outer camera or CMOS camera and infrared camera combine, and the present invention is not especially limit this.Visual sensing dress
The forward image for acquiring visually impaired person's walking path is set, to simulate the image that human eye is observed in the process of walking.
Present image is i.e. in being directed to visually impaired person's navigation procedure, vision sensing equipment people visually impaired of collected current time
The forward image of member's walking path.
Step 120, present image and current path information input are obtained into the posture of navigation model output to navigation model
Information；Navigation model is obtained based on the training of sample image, sample path information, sample posture information and sample identification.
Specifically, current path information is that current time visually impaired person needs the direction advanced, for example, straight trip, turn left or
It turns right, current path information, which can be, to be directly acquired by the existing map software for having navigation feature, such as Google
Figure, Baidu map etc., are also possible to the current location information based on visually impaired person and pre-set guidance path matches to obtain
, the present invention is not especially limit this.
Navigation model is used to judge that visually impaired person should execute at current time based on present image and current path information
Specific movement posture.After obtaining present image, by present image and current path information input to navigation model, it can obtain
The posture information exported to navigation model, posture information are used to indicate the movement posture that current time visually impaired person should execute,
Such as keep straight on, in another example turning left 45 °, in another example traversing two step to the right, the present invention is not especially limit this.
In addition, can also train in advance before executing step 120 and obtain navigation model, it specifically can be in the following way
Training obtains: firstly, collecting great amount of samples image, sample path information, sample posture information and sample identification.Wherein, sample
Image is the forward image for the walking path that personnel are collected in practical walking process by vision sensing equipment, herein
Personnel can be visually impaired person, be also possible to non-visually impaired person.Sample path information is vision sensing equipment capturing sample image
At the time of personnel need the direction advanced；Personnel are practical at the time of sample posture information is vision sensing equipment capturing sample image
The movement posture of execution, herein, sample posture information can have by personnel with accelerometer of carrying or gyroscope etc.
The equipment of standby human body attitude detection function collects；Sample identification be used to indicate vision sensing equipment capturing sample image when
The personnel at quarter execute the walking that sample posture information causes as a result, for example other than personnel go to guidance path or knocking obstacle
Object, then sample identification is negative, when personnel walking on guidance path and success avoiding obstacles, then sample identification is positive, then example
Deadline as personnel walk is shorter, then the absolute value of sample identification is higher, and the present invention is not especially limit this.
Sample path information, sample posture information and the sample at moment are corresponded to based on sample image and sample image immediately
Mark is trained initial model, to obtain navigation model.Wherein, initial model can be single neural network model,
It is also possible to the combination of multiple neural network models, the embodiment of the present invention does not make specific limit to the type of initial model and structure
It is fixed.
Step 130, Voice Navigation is carried out to visually impaired person based on posture information.
Specifically, after the posture information for obtaining navigation model output, voice is carried out to visually impaired person based on posture information and is led
Boat indicates the movement posture walking that visually impaired person indicate according to posture information by voice, so that it is guaranteed that visually impaired person can be
Walking and the avoiding obstacles that can succeed on guidance path.Herein, Voice Navigation is carried out to visually impaired person, people visually impaired can be passed through
Member wear earphone to visually impaired person propagate navigation voice, can also by the portable such as smart phone of visually impaired person,
Intelligent bracelet etc. has sound outer playing function smart machine and navigation voice is broadcast to visually impaired person.
Method provided in an embodiment of the present invention, by by present image and current path information input into navigation model,
It obtains posture information to navigate to visually impaired person, navigation is refine to the guidance to visually impaired person's movement posture, it is ensured that visually impaired
Personnel can receive detailed navigation information, can walk simultaneously along guidance path under foreign environment or complex environment
Success avoiding obstacles ensure that the walking safety of visually impaired person, expand the application range of visually impaired person's navigation.
Based on the above embodiment, navigation model specifically includes straight trip navigation model, bend navigation model.Accordingly, step
120 specifically include: step 121, being chosen from straight trip navigation model and bend navigation model when leading based on current path information
Model plane type；Step 122, present image is input in Present navigation model, obtains the posture information of Present navigation model output.
Specifically, straight trip navigation model is used to judge visually impaired person based on present image when current path information is straight trip
In the specific movement posture that current time should execute, bend navigation model is used to be based on when current path information is turning to work as
Preceding image judges the specific movement posture that visually impaired person should execute at current time.Present navigation model is based on current path
The model that information is chosen.
If current path information is straight trip, the navigation model that will keep straight on is as Present navigation model；If current path information
For turning, such as turn left, turn right etc., then using bend navigation model as Present navigation model.After obtaining Present navigation model,
Present image is input to Present navigation model, present image is analyzed by Present navigation model, and then exports posture letter
Breath.
In addition, training in advance straight trip navigation model and bend navigation mould can also be obtained respectively before executing step 121
Type.Can specifically train in the following way obtain straight trip navigation model: firstly, collect great amount of samples routing information for straight trip when pair
Sample image, sample posture information and the sample identification answered.Corresponding sample graph when being immediately straight trip based on sample path information
Picture, sample posture information and sample identification are trained initial model, to obtain straight trip navigation model.Similarly, specifically may be used
Training obtains bend navigation model in the following way: firstly, collecting corresponding sample when great amount of samples routing information is turning
Image, sample posture information and sample identification.Corresponding sample image, sample appearance when being immediately turning based on sample path information
State information and sample identification are trained initial model, to obtain bend navigation model.
Method provided in an embodiment of the present invention chooses corresponding Present navigation model based on current path information and obtains posture
Information is obtained by refining navigation model for the higher navigation model of single-pathway information navigation precision, to improve difference
Navigation accuracy under routing information.
Based on any of the above-described embodiment, bend navigation model includes that bend starts model and bend and terminates model.
Specifically, bend starts model for judging people visually impaired based on present image when current path information is turns
Whether member comes into bend, if starts to turn；Bend terminates model for when current path information is turning, based on working as
Preceding image judges whether visually impaired person has completed turning and entered forthright, if terminates turning.
Accordingly, step 121 specifically includes: if previous routing information is straight trip, current path information is straight trip, then will be straight
Row navigation model is as Present navigation model；If previous routing information is straight trip, current path information is turning, then opens bend
Beginning model is as Present navigation model；If previous routing information is turning, current path information is straight trip, then bend is terminated mould
Type is as Present navigation model.
Specifically, Present navigation model is chosen based on previous routing information and current path information.Herein, previous path letter
Routing information before breath i.e. current time.If previous routing information and current path information are straight trip, straight trip is navigated
Model is as Present navigation model；If previous routing information is straight trip, current path information is turning, then bend is started model
As Present navigation model, so that bend, which starts model, to judge whether visually impaired person comes into based on present image
Bend, if needs start to turn, and specific posture information are determined, to instruct the posture of visually impaired person to be adjusted to from straight-going state
Turn condition；If previous routing information is turning, current path information is to execute, then bend is terminated model as Present navigation
Model, so that bend terminates model can judge whether visually impaired person has completed turning and entered directly based on present image
Road, if can terminate to turn, determine specific posture information, to instruct the posture of visually impaired person to be adjusted to straight from turn condition
Row state.
Method provided in an embodiment of the present invention, starting model and bend by the way that bend navigation model is refined as bend terminates
Model, can the further work that is undertaken of refined model, improve the navigation accuracy of single model.
Based on any of the above-described embodiment, before step 120 further include: step 101, obtain working as orientation sensing device acquisition
Prelocalization information；Step 102, current path information is obtained based on current location information and guidance path.
Specifically, orientation sensing device can be GPS module, is also possible to WiFi module, bluetooth module etc. and is used for interior
The positioning device of positioning, the present invention is not especially limit this.Orientation sensing device is for acquiring working as visually impaired person
Prelocalization information, current location information is used to reflect the position at visually impaired person's current time herein.Obtaining current location information
Afterwards, current path information is obtained based on current location information and guidance path.Herein, guidance path, that is, preset needs pair
The path that visually impaired person navigates.There are many ways to obtaining current path information, for example, directly by current location information and
Guidance path is input in the existing map software for having navigation feature, obtains the current path information of map software output.
Based on any of the above-described embodiment, before step 102 further include: obtain navigation task voice；To navigation task voice
Speech recognition is carried out, destination is obtained；Guidance path is obtained based on destination.
Specifically, navigation task voice is the voice data for including destination, visually impaired before visually impaired person's trip
Personnel can set destination by way of voice.After obtaining navigation task voice, navigation task voice is carried out
Speech recognition obtains the corresponding text of navigation task voice, such as " I will go to convenience store ".Immediately from the corresponding text of navigation voice
Target location is extracted in word, i.e. visually impaired person it is expected the path termination navigated.After obtaining destination, based on current fixed
The path that destination is gone to from current location, i.e. guidance path are planned in position information and destination.
Method provided in an embodiment of the present invention obtains guidance path by carrying out speech recognition to navigation task voice, makes
Visually impaired person can be with self-service setting navigation path, and then independent startup navigates and arrives safely destination,
Existing navigation scheme visually impaired mostly uses CMOS vision sensing equipment to acquire image.But CMOS vision sensing equipment
It can only use, can not especially carry out the work at night under the environment of light conditions difference, therefore limit people visually impaired on daytime
The outside environment of member.In this regard, being based on any of the above-described embodiment, vision sensing equipment includes infrared camera in the embodiment of the present invention；
Accordingly, present image includes current thermal-induced imagery.
Specifically, infrared camera is the camera for recording the light issued by infrared light supply, and infrared camera is often applied
In night vision monitoring field.The forward image that visually impaired person's walking path is acquired by setting infrared camera is able to solve CMOS view
Feel that sensing device can not be the problem of working at night.Accordingly, image, that is, thermal-induced imagery of infrared camera acquisition, it is current infrared
Thermal image refers in visually impaired person's navigation procedure, the collected current time visually impaired person walking path of infrared camera
Forward image.
It should be noted that the vision sensing equipment in the embodiment of the present invention can not only include infrared camera.Further include
CMOS vision sensing equipment.Accordingly, present image may include current thermal-induced imagery, can also include current RGB image
Or the image of extended formatting.Herein, the thermal-induced imagery at current thermal-induced imagery, that is, current time.
Method provided in an embodiment of the present invention solves existing air navigation aid and is limited by CMOS vision sensing equipment, only
It can be the problem of working by day, visually impaired person's navigation expands the application of visually impaired person's navigation without being limited by time and light
Range.
Based on any of the above-described embodiment, sample posture information is acquired based on Inertial Measurement Unit.
Specifically, Inertial Measurement Unit (Inertial Measurement Unit, IMU) is measurement object three-axis attitude
The device of angle and acceleration.In general, an Inertial Measurement Unit contains three uniaxial accelerometers and three uniaxial
Gyroscope, accelerometer detection object founds the acceleration signal of three axis in carrier coordinate system unification and independence, and gyroscope detects carrier phase
For the angular velocity signal of navigational coordinate system, it is achieved in the measurement of the angular speed and acceleration of object in three dimensions, into
And calculate the posture of object.Inertial Measurement Unit is applied to the acquisition of sample posture information, i.e., in the practical walking of personnel
Pass through the movement posture of the self-contained Inertial Measurement Unit collector of personnel in the process.Based on Inertial Measurement Unit acquisition
Sample posture information, reliability is high, and the navigation model that thus training obtains can export more accurate posture information, into one
Step improves the accuracy and reliability of visually impaired person's navigation.
Based on any of the above-described embodiment, a kind of visually impaired person's air navigation aid includes the following steps:
Firstly, collecting great amount of samples image, sample path information, sample posture information and sample identification.Herein, sample graph
As including sample RGB image and sample thermal-induced imagery.
With namely based on sample path information, being the sample arm of straight trip based on sample path information and previous routing information
The corresponding sample image of diameter information, sample posture information and sample identification building straight trip training set are based on sample path information
Turning, previous routing information are the corresponding sample image of sample path information, sample posture information and the sample identification structure of straight trip
It builds turning and starts training set, be straight trip based on sample path information, previous routing information is that the sample path information of turning is corresponding
Sample image, sample posture information and sample identification building turning terminate training set.It is based respectively on straight trip training set, turning is opened
Beginning training set and turning terminate training set and are trained to initial model, obtain navigation model.Navigation model includes straight trip navigation
Model, turning, which start model and turning, terminates model.
Then, if receiving navigation task voice, speech recognition is carried out to navigation task voice, obtains destination
Point, and guidance path is obtained based on destination, start to navigate.
The present image of infrared camera acquisition is obtained in real time, and obtains the current location information of GPS module acquisition.Herein,
Present image includes current RGB image and current thermal-induced imagery.After obtaining current location information, by current location information and
Guidance path is input in Baidu map, obtains the current path information of Baidu map output.
Based on current path information and previous routing information, model and turning terminate since straight trip navigation model, turning
Present navigation model is chosen in model.Then, present image is input in Present navigation model, it is defeated obtains Present navigation model
Posture information out.
After obtaining posture information, the earphone worn by visually impaired person propagates navigation voice, instruction view to visually impaired person
Barrier personnel walk according to the movement posture that posture information indicates, so that it is guaranteed that visually impaired person can walk on guidance path and energy
Enough successful avoiding obstacles.
Method provided in an embodiment of the present invention, by by present image and current path information input into navigation model,
It obtains posture information to navigate to visually impaired person, navigation is refine to the guidance to visually impaired person's movement posture, it is ensured that visually impaired
Personnel can receive detailed navigation information, can walk simultaneously along guidance path under foreign environment or complex environment
Success avoiding obstacles ensure that the walking safety of visually impaired person, expand the application range of visually impaired person's navigation.
Based on any of the above-described embodiment, this method is tested.Experiment scene size is 200m*200m, including five
Material is the road of cement, and having a lot of social connections is 2 to 3 meters.In experimentation, progress sample collection work first is then based on sample graph
Picture, sample path information, sample posture information and sample identification are trained initial model and obtain navigation model after optimizing.
Instruction, which is obtained, after the image input navigation model based on camera acquisition immediately carries out walking and avoidance.According to test result, this hair
The method that bright embodiment provides, in the case where not occurring that illumination acute variation, environment scene are strange etc. not to acquire data,
The error at the midpoint of navigation accuracy distance objective route substantially may be used within 10cm when having pedestrian and other barriers occur
To accomplish effective avoidance.
Based on any of the above-described embodiment, this method is tested.Experiment scene size is 600m*600m, and experiment exists
A plurality of material is slabstone, ring park executes on one week road, and having a lot of social connections is 3 to 4 meters, whole process about 1.8km.It is first in experimentation
Sample collection work is first carried out, then based on sample image, sample path information, sample posture information and sample identification to initial
Model is trained and obtains navigation model after optimizing.It is instructed after the image input navigation model based on camera acquisition immediately
Carry out walking and avoidance.According to test result, is there is not illumination acute variation, ring in method provided in an embodiment of the present invention
Border scene it is strange it is equal do not acquire data in the case where, the error at the midpoint of navigation accuracy distance objective route 10cm with
It is interior, it can accomplish effective avoidance substantially when having pedestrian and other barriers occur.
Based on any of the above-described embodiment, this method is tested.Experiment scene size is 2500m*1800m, experiment
Road is integrated the red running special lane in ring garden, and having a lot of social connections is 2 to 3 meters.In experimentation, progress sample collection work first,
Then initial model is trained and is optimized based on sample image, sample path information, sample posture information and sample identification
After obtain navigation model.Instruction, which is obtained, after the image input navigation model based on camera acquisition immediately carries out walking and avoidance.Root
According to test result, method provided in an embodiment of the present invention, illumination acute variation, environment scene are strange etc. not to be acquired not occurring
In the case where data, the error at the midpoint of navigation accuracy distance objective route is having pedestrian and other obstacles within 5cm
Substantially it can accomplish effective avoidance when object occurs.
Existing visually impaired person's air navigation aid often relies on cloud server, needs vision sensing equipment and orientation sensing
Present image and current location information are transferred to cloud server to device by module by wireless communication respectively, and cloud server is again
Navigation information is returned after carrying out the operation such as image recognition, orientation detection, data transmission needs elapsed time, leads to real-time of navigating
Difference.And if cloud server breaks down, all visually impaired person's navigation service users will receive influence, reliability
It is low.In this regard, the embodiment of the invention provides a kind of visually impaired person's navigation system.
Based on any of the above-described embodiment, Fig. 2 is the structural representation of visually impaired person's navigation system provided in an embodiment of the present invention
Figure, as shown in Fig. 2, the system includes portable intelligent terminal 210 and vision sensing equipment 220；Wherein, portable intelligent terminal 210
The method provided for executing any of the above-described embodiment.
Specifically, portable intelligent terminal 210 be smart phone, Intelligent bracelet etc. can portable intelligent terminal set
It is standby.Portable intelligent terminal 210 can be carried out data transmission with vision sensing equipment 220 by wired or wireless mode, such as
Vision sensing equipment 220 is installed on the glasses of visually impaired person's wearing, while wireless communication module is also installed on glasses, vision
By the present image collected, module is sent on smart phone sensing device by wireless communication, and smart phone is receiving
After present image, by the navigation module stored in present image and current path information input to smart phone, and obtains and lead
The posture information of model plane block output, plays to view for the navigation voice generated based on posture information by sound outer playing function immediately
Barrier personnel, or the earphone worn by visually impaired person play navigation voice to visually impaired person.
System provided in an embodiment of the present invention executes visually impaired person's air navigation aid by portable intelligent terminal, effectively save
Data transmission period, improves the real-time and stability of navigation.Furthermore, it is only necessary to be equipped with vision sensing equipment, and portable
Navigation model is laid on intelligent terminal, visually impaired person's navigation can be realized, low in cost, structure is simple, easy to carry, reliability
Height facilitates the application of visually impaired person's navigation.
Based on any of the above-described embodiment, Fig. 3 is the structural representation of visually impaired person's navigation device provided in an embodiment of the present invention
Figure, as shown in figure 3, the device includes image acquisition units 310, posture acquiring unit 320 and navigation elements 330；
Wherein, image acquisition units 310 are used to obtain the present image of vision sensing equipment acquisition；
Posture acquiring unit 320 is used to the present image and current path information input obtaining institute to navigation model
State the posture information of navigation model output；The navigation model is based on sample image, sample path information, sample posture information
It is obtained with sample identification training；
Navigation elements 330 are used to carry out Voice Navigation to visually impaired person based on the posture information.
Device provided in an embodiment of the present invention, by by present image and current path information input into navigation model,
It obtains posture information to navigate to visually impaired person, navigation is refine to the guidance to visually impaired person's movement posture, it is ensured that visually impaired
Personnel can receive detailed navigation information, can walk simultaneously along guidance path under foreign environment or complex environment
Success avoiding obstacles ensure that the walking safety of visually impaired person, expand the application range of visually impaired person's navigation.
Based on any of the above-described embodiment, the navigation model specifically includes straight trip navigation model, bend navigation model；
Accordingly, posture acquiring unit 320 specifically includes model and chooses subelement and posture output subelement；
Wherein, model is chosen subelement and is used for based on the current path information from the straight trip navigation model and described curved
Present navigation model is chosen in road navigation model；
Posture output subelement obtains described current for the present image to be input in the Present navigation model
The posture information of navigation model output.
Based on any of the above-described embodiment, the bend navigation model includes that bend starts model and bend and terminates model；
Accordingly, model is chosen subelement and is specifically used for:
If previous routing information be straight trip, the current path information be straight trip, then using the straight trip navigation model as
The Present navigation model；
If the previous routing information is straight trip, the current path information is turning, then the bend is started model
As the Present navigation model；
If the previous routing information is turning, the current path information is straight trip, then the bend is terminated model
As the Present navigation model.
Based on any of the above-described embodiment, which further includes current path acquiring unit；Current path acquiring unit is specific
For:
Obtain the current location information of orientation sensing device acquisition；
Current path information is obtained based on the current location information and guidance path.
Based on any of the above-described embodiment, which further includes guidance path acquiring unit；Guidance path acquiring unit is specific
For:
Obtain navigation task voice；
Speech recognition is carried out to the navigation task voice, obtains destination；
Guidance path is obtained based on the destination.
Based on any of the above-described embodiment, the vision sensing equipment includes infrared camera；Accordingly, the present image packet
Include current RGB image and current thermal-induced imagery.
Based on any of the above-described embodiment, the sample posture information is acquired based on Inertial Measurement Unit.
Fig. 4 is the entity structure schematic diagram of electronic equipment provided in an embodiment of the present invention, as shown in figure 4, the electronic equipment
It may include: processor (processor) 401,402, memory communication interface (Communications Interface)
(memory) 403 and communication bus 404, wherein processor 401, communication interface 402, memory 403 pass through communication bus 404
Complete mutual communication.Processor 401 can call the meter that is stored on memory 403 and can run on processor 401
Calculation machine program, to execute visually impaired person's air navigation aid of the various embodiments described above offer, for example, obtain vision sensing equipment and adopt
The present image of collection；By the present image and current path information input to navigation model, the navigation model output is obtained
Posture information；The navigation model is based on sample image, sample path information, sample posture information and sample identification training
It obtains；Voice Navigation is carried out to visually impaired person based on the posture information.
In addition, the logical order in above-mentioned memory 403 can be realized by way of SFU software functional unit and conduct
Independent product when selling or using, can store in a computer readable storage medium.Based on this understanding, originally
The technical solution of the inventive embodiments substantially part of the part that contributes to existing technology or the technical solution in other words
It can be embodied in the form of software products, which is stored in a storage medium, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes the present invention respectively
The all or part of the steps of a embodiment the method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory
(ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk
Etc. the various media that can store program code.
The embodiment of the present invention also provides a kind of non-transient computer readable storage medium, is stored thereon with computer program,
The computer program is implemented to carry out visually impaired person's air navigation aid of the various embodiments described above offer when being executed by processor, such as wraps
It includes: obtaining the present image of vision sensing equipment acquisition；By the present image and current path information input to navigation model,
Obtain the posture information of the navigation model output；The navigation model is based on sample image, sample path information, sample appearance
What state information and sample identification training obtained；Voice Navigation is carried out to visually impaired person based on the posture information.
The apparatus embodiments described above are merely exemplary, wherein described, unit can as illustrated by the separation member
It is physically separated with being or may not be, component shown as a unit may or may not be physics list
Member, it can it is in one place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of the modules achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness
Labour in the case where, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features；
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (10)
1. a kind of visually impaired person's air navigation aid characterized by comprising
Obtain the present image of vision sensing equipment acquisition；
By the present image and current path information input to navigation model, the posture letter of the navigation model output is obtained
Breath；The navigation model is obtained based on the training of sample image, sample path information, sample posture information and sample identification；
Voice Navigation is carried out to visually impaired person based on the posture information.
2. the method according to claim 1, wherein the navigation model specifically includes straight trip navigation model and curved
Road navigation model；
Accordingly, it is described by the present image and current path information input into navigation model, obtain the navigation model
The posture information of output, specifically includes:
Based on the current path information, Present navigation mould is chosen from the straight trip navigation model and the bend navigation model
Type；
The present image is input to the Present navigation model, obtains the posture letter of the Present navigation model output
Breath.
3. according to the method described in claim 2, it is characterized in that, the bend navigation model includes that bend starts model and curved
Road terminates model；
Accordingly, described to be based on the current path information, it is selected from the straight trip navigation model and the bend navigation model
Present navigation model is taken, is specifically included:
If previous routing information is straight trip, the current path information is straight trip, then using the straight trip navigation model as described in
Present navigation model；
If the previous routing information be straight trip, the current path information be turning, then using the bend start model as
The Present navigation model；
If the previous routing information be turning, the current path information be straight trip, then using the bend terminate model as
The Present navigation model.
4. the method according to claim 1, wherein described by the present image and current path information input
Into navigation model, the posture information of the navigation model output is obtained, before further include:
Obtain the current location information of orientation sensing device acquisition；
Current path information is obtained based on the current location information and guidance path.
5. according to the method described in claim 4, it is characterized in that, described obtained based on the current location information and guidance path
Current path information is taken, before further include:
Obtain navigation task voice；
Speech recognition is carried out to the navigation task voice, obtains destination；
The guidance path is obtained based on the destination.
6. the method according to any one of claims 1 to 5, which is characterized in that the vision sensing equipment includes infrared
Camera；Accordingly, the present image includes current thermal-induced imagery.
7. the method according to any one of claims 1 to 5, which is characterized in that the sample posture information is based on used
Property measuring unit acquisition.
8. a kind of visually impaired person's navigation system, which is characterized in that including portable intelligent terminal and vision sensing equipment；Wherein, institute
Portable intelligent terminal is stated for executing the method as described in claim 1 to 7 is any.
9. a kind of electronic equipment, which is characterized in that including processor, communication interface, memory and bus, wherein processor leads to
Believe that interface, memory complete mutual communication by bus, processor can call the logical order in memory, to execute
Method as described in claim 1 to 7 is any.
10. a kind of non-transient computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer
The method as described in claim 1 to 7 is any is realized when program is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910247664.0A CN109938973A (en) | 2019-03-29 | 2019-03-29 | A kind of visually impaired person's air navigation aid and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910247664.0A CN109938973A (en) | 2019-03-29 | 2019-03-29 | A kind of visually impaired person's air navigation aid and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109938973A true CN109938973A (en) | 2019-06-28 |
Family
ID=67012873
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910247664.0A Pending CN109938973A (en) | 2019-03-29 | 2019-03-29 | A kind of visually impaired person's air navigation aid and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109938973A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111195191A (en) * | 2019-11-22 | 2020-05-26 | 韩凤明 | Intelligent system and method for guiding travel of blind people |
| CN111329735A (en) * | 2020-02-21 | 2020-06-26 | 北京理工大学 | Blind guiding method, device and system |
| CN116698045A (en) * | 2023-08-02 | 2023-09-05 | 国政通科技有限公司 | Walking auxiliary navigation method and system for vision disturbance people in nursing home |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9508269B2 (en) * | 2010-08-27 | 2016-11-29 | Echo-Sense Inc. | Remote guidance system |
| CN106375448A (en) * | 2016-09-05 | 2017-02-01 | 腾讯科技（深圳）有限公司 | Image processing method, device and system |
| CN107091646A (en) * | 2017-03-22 | 2017-08-25 | 北京小度信息科技有限公司 | Navigation control method, device and movable equipment |
| CN107328424A (en) * | 2017-07-12 | 2017-11-07 | 三星电子（中国）研发中心 | Air navigation aid and device |
| CN108245384A (en) * | 2017-12-12 | 2018-07-06 | 清华大学苏州汽车研究院（吴江） | Binocular vision apparatus for guiding blind based on enhancing study |
-
2019
- 2019-03-29 CN CN201910247664.0A patent/CN109938973A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9508269B2 (en) * | 2010-08-27 | 2016-11-29 | Echo-Sense Inc. | Remote guidance system |
| CN106375448A (en) * | 2016-09-05 | 2017-02-01 | 腾讯科技（深圳）有限公司 | Image processing method, device and system |
| CN107091646A (en) * | 2017-03-22 | 2017-08-25 | 北京小度信息科技有限公司 | Navigation control method, device and movable equipment |
| CN107328424A (en) * | 2017-07-12 | 2017-11-07 | 三星电子（中国）研发中心 | Air navigation aid and device |
| CN108245384A (en) * | 2017-12-12 | 2018-07-06 | 清华大学苏州汽车研究院（吴江） | Binocular vision apparatus for guiding blind based on enhancing study |
Non-Patent Citations (4)
| Title |
|---|
| 周庆瑞等: "移动机器人视觉导航控制研究", 《计算机工程与应用》 * |
| 张海波，原魁，周庆瑞: "基于路径识别的移动机器人视觉导航", 《中国图象图形学报》 * |
| 董丹煌，方志刚，徐洁: "视障人士户外行走辅助***研究", 《计算机工程与应用》 * |
| 赵瑜,种兰祥,张万绪: "基于计算机视觉的移动机器人导航", 《现代电子技》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111195191A (en) * | 2019-11-22 | 2020-05-26 | 韩凤明 | Intelligent system and method for guiding travel of blind people |
| CN111329735A (en) * | 2020-02-21 | 2020-06-26 | 北京理工大学 | Blind guiding method, device and system |
| CN111329735B (en) * | 2020-02-21 | 2021-09-28 | 北京理工大学 | Blind guiding method, device and system |
| CN116698045A (en) * | 2023-08-02 | 2023-09-05 | 国政通科技有限公司 | Walking auxiliary navigation method and system for vision disturbance people in nursing home |
| CN116698045B (en) * | 2023-08-02 | 2023-11-10 | 国政通科技有限公司 | Walking auxiliary navigation method and system for vision disturbance people in nursing home |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6897668B2 (en) | Information processing method and information processing equipment | |
| JP2021515286A (en) | Systems and methods for anonymizing navigation information | |
| CN106289290A (en) | A kind of path guiding system and method | |
| CN109429518A (en) | Automatic Pilot traffic forecast based on map image | |
| CN110378483A (en) | The system and method for training machine learning model being deployed on analog platform | |
| CN110347145A (en) | Perception for automatic driving vehicle assists | |
| CN108732589A (en) | The training data of Object identifying is used for using 3D LIDAR and positioning automatic collection | |
| CN109253735B (en) | Path planning method, device and storage medium | |
| CN107144285A (en) | Posture information determines method, device and movable equipment | |
| CN108303103A (en) | The determination method and apparatus in target track | |
| CN109219551A (en) | Condition of road surface head up display | |
| CN109937343A (en) | Appraisal framework for the prediction locus in automatic driving vehicle traffic forecast | |
| CN109938973A (en) | A kind of visually impaired person's air navigation aid and system | |
| CN110160542A (en) | The localization method and device of lane line, storage medium, electronic device | |
| CN108957505A (en) | A kind of localization method, positioning system and portable intelligent wearable device | |
| EP4050449A1 (en) | Method and device for robot positioning, smart robot, and storage medium | |
| CN106197429A (en) | A kind of Multi-information acquisition location equipment and system | |
| CN103968824A (en) | Method for discovering augmented reality target, and terminal | |
| CN104049742A (en) | Three dimensional fingertip tracking | |
| CN109491378A (en) | The route guiding system based on roadway segment of automatic driving vehicle | |
| CN110686694A (en) | Navigation method, navigation device, wearable electronic equipment and computer readable storage medium | |
| CN106774351A (en) | Have the positioning navigation device and robot of virtual rail function | |
| WO2020114214A1 (en) | Blind guiding method and apparatus, storage medium and electronic device | |
| CN111494175A (en) | Navigation method based on head-mounted equipment, head-mounted equipment and storage medium | |
| JP7040827B1 (en) | Search support system and rescue support 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190628 |