CN201870808U - Guide device for blind people based on stereoscopic vision - Google Patents
Guide device for blind people based on stereoscopic vision Download PDFInfo
- Publication number
- CN201870808U CN201870808U CN2010206265022U CN201020626502U CN201870808U CN 201870808 U CN201870808 U CN 201870808U CN 2010206265022 U CN2010206265022 U CN 2010206265022U CN 201020626502 U CN201020626502 U CN 201020626502U CN 201870808 U CN201870808 U CN 201870808U
- Authority
- CN
- China
- Prior art keywords
- central processing
- processing module
- blind
- image
- guide device
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Traffic Control Systems (AREA)
Abstract
The utility model relates to a guide device for blind people based on stereoscopic vision, which comprises a binocular camera, a central processing module and a voice output module. The left ocular camera and the right ocular camera of the binocular camera are connected with the central processing module, the central processing module is provided with image processing software for processing of obstacles, traffic lights and road departure, and the output end of the central processing module is connected with the voice output module. A user can wear the device on the head to freely walk indoors and outdoors and safely cross intersections according to voice prompts sent by the voice output module after starting the device.
Description
Technical field
This utility model belongs to field of machine vision, particularly a kind of blind-guide device based on stereoscopic vision.
Background technology
Five or six thousand ten thousand blind persons are arranged in the world approximately, and China's blind person's quantity accounts for 1/4th of world blind person sum.The blind person is disadvantaged group, needs the more help of society, makes them improve the ability of living on one's own life, and has better quality of life.The assisting blind walking is a very important aspect.Research and develop the instrument of multiple assisting blind walking both at home and abroad, but still had many problems.For example: the equipment of assisting blind identification traffic light, the requirement that has is installed sound-producing device at each crossing, the requirement that has control traffic lights, this can increase cost, and can cause interference to other people.The equipment of most of assisting blind walking adopts ultrasound wave, infrared ray or radar to come the detecting obstacles thing, but these detection methods have directivity, can only survey the barrier in the certain angle scope, can only survey the barrier in the certain distance scope, and can't distinguish the size of barrier, can't discern traffic light.Can only there be significant limitation in the blind man navigation system based on computer vision that has along predefined route walking.Artificial retina implanted prosthetics cost height, technical sophistication is difficult for promoting.The guiding robot needs the complicated algorithm support, and cost is higher, and road surface, the leaping over obstacles of walking stair and inclination had certain difficulty.
Xie Yizheng has proposed employing TOSHIBA notebook computer, QuickCam
TMPro 4000CCD and earphone are realized blind person's aid system (seeing list of references: Xie Yizheng for details, " with the real blind person's accessory system that does of stereoscopic vision ", State Central Univ.'s information Graduate School of Engineering master thesis, Taiwan, in July, 2006).Xie Yizheng proposes to utilize " road surface region growing method " to cut out the zone, road surface of environment earlier, carry out " image quantification " then, image comparison is done according to quantized imagery zone piece in zone, non-road surface, utilize " method of hiving off " to remove image incidental information, set up out the foreign environment information, detect barrier, and utilize " comparison of three-dimensional zone " and " unbecoming hiving off " to obtain the distance of barrier.Blind person's aid system total algorithm complexity of Xie Yizheng, the CCD photographic head of employing can't adapt to the variation of indoor and outdoor available light power, can cause not fogging clearly in strong sunlight, backlight and when light is dark at dusk, can meet difficulty when outside work.The obstacle distance measuring method complexity that he adopts, the range finding accuracy rate is low.He only surveys experiment at the indoor static-obstacle quality testing of having done, and outdoor auxiliary walking, road profile is extracted, is departed from aspects such as road early warning, dynamic barrier detection and traffic light detection and study.
The utility model content
The technical problems to be solved in the utility model is to overcome above-mentioned the deficiencies in the prior art, a kind of blind-guide device based on stereoscopic vision is provided, and this device can adapt to the strong and weak variation of natural light, and by voice suggestion, assisting blind is walked without barrier in indoor and outdoor, and safety is passed through the intersection.
The technical solution of the utility model is as follows:
A kind of blind-guide device based on stereoscopic vision, comprise binocular camera, central processing module and voice output module, the left lens camera of described binocular camera links to each other with described central processing module with right lens camera, described central processing module has the barrier processing, traffic lights is handled and departed from road and is treated to main image processing software, and the outfan of this central processing module links to each other with described voice output module.
Described binocular camera is super wide dynamic for adopting, and high light suppresses, low-illuminance cameras.
Described binocular camera employing dynamic range is 1: 400 a low-illuminance cameras.
It is the super wide dynamic low-illuminance cameras of 0.05Lux that described binocular camera adopts minimal illumination.
Described central processing module adopts the embedded machine vision image processing platform.
Described voice output module is voice module or the earphone that described central processing module carries.
Utilize above-mentioned method of walking, comprise the following steps: based on the blind-guide device assisting blind of stereoscopic vision
1. the left lens camera of described binocular camera, right lens camera are installed in blind person's medicated cap the place ahead symmetrically, the center distance of overhead contour and left lens camera, right lens camera is fixed, this left side lens camera, right lens camera link to each other with described central processing module, the outfan of described central processing module links to each other with described voice output module, with the earphone plug of voice module in blind person's earhole;
2. start described blind-guide device, left lens camera, right lens camera are gathered traffic information synchronously in real time; The video image that central processing module absorbs left lens camera, right lens camera respectively with the speed of per second 2-8 frame becomes the image sequence of successive framing and keeps in;
3. described central processing module at first is converted to gray level image with the image sequence of the successive framing of left lens camera or right lens camera collection, adopt medium filtering to carry out the image pretreatment then, and every two field picture carried out Canny rim detection and Hough change detection straight line, extract road edge and also judge whether to be road edge:
When being road edge, carry out step simultaneously 4., 5., 6.;
When being not road edge, carry out step simultaneously 4., 5.;
4. barrier is handled: the consecutive image frame sequence that described central processing module is gathered monocular-camera, and the trapezoidal area-of-interest selecting on the image on the user direction of advance extracts barrier region to this zone according to gray value, extracts the barrier target; After detecting the barrier target, the same two field picture that described central processing module is gathered another monocular-camera of binocular camera, the distance of employing parallax method dyscalculia thing;
5. traffic light are handled: described central processing module is to the consecutive image frame sequence of left lens camera or the collection of right lens camera, carry out traffic light according to template gray scale coupling and handle, provide the voice suggestion of traffic light color and signal lights position then by voice module;
6. departing from road handles: described central processing module judges that according to the position and the inclination angle of road edge line in image that extracts user departs from the situation of road, provides voice suggestion.
Calculate described detected straight length, and judge: if this length greater than threshold value, and at the angle of the perpendicular bisector of left and right figure cathetus and image at 0 °～90 °, then this straight line is a road edge.
Trapezoidal area-of-interest during described barrier is handled is the practical situation setting according to user.
Template matching was by the testee prototype is provided during described traffic light were handled, and the similarity between all relevant positions of calculation template and each position of image is sought target object in the image of importing.
Technique effect of the present utility model:
Compare with above-mentioned prior art, this utility model can adapt to the strong and weak variation of natural light owing to adopted low-illuminance cameras; Particularly dynamic range 1: 400 and minimal illumination are the selecting for use of super wide dynamic low-illuminance cameras of 0.05Lux, make this device adapt to the strong sunlight or the working condition of faint light at dusk; In binocular stereo vision barrier range finding, adopt ask for the method that similarity mates after, method is simple for the walking of this device assisting blind, be easy to realization; This device also has the static state of detection, dynamic barrier, detects road profile, and the identification traffic light carry out functions such as indoor and outdoor navigation.
Description of drawings
Fig. 1 is the structured flowchart of this utility model based on the blind-guide device of stereoscopic vision
Fig. 2 is this utility model device main program flow chart
Fig. 3 is this utility model device road-edge detection program flow diagram
Fig. 4 is this utility model device detection of obstacles program flow diagram
Fig. 5 is that the coordinate and the ROI zone of image is provided with
The specific embodiment
Below in conjunction with accompanying drawing the blind-guide device based on stereoscopic vision is described in detail.
See also Fig. 1 earlier, Fig. 1 is the structured flowchart of this utility model based on the blind-guide device of stereoscopic vision, as seen from the figure, this utility model is based on the blind-guide device of stereoscopic vision, comprise binocular camera 1, central processing module 2 and voice output module 3, the left lens camera 11 of described binocular camera, right lens camera 12 is installed in blind person's medicated cap the place ahead symmetrically, this left side lens camera 11 links to each other with described central processing module 2 with right lens camera 12, described central processing module 2 has barrier to be handled, traffic lights is handled and is departed from road and is treated to main image processing software, and the outfan of this central processing module 2 links to each other with described voice output module 3.
This binocular camera 1 adopts outstanding scholar to pacify the super wide dynamic low-illuminance cameras of JSA-WZS31480CE, super-wide dynamic range 1: 400,520/570 line superelevation horizontal definition, has the BLC function, the numeral noise is eliminated (SDNR) function, image sensor: 1/3 inch, the total pixel of CCD: 795 (level) 596 (vertically) of X (pal mode)/811 (level) X 508 (vertically) (TSC-system formula), scanning system: 625 lines, 50/second (pal mode)/525 lines, 60/second (TSC-system formula), minimal illumination: 0.05Lux (F1.2,5600 ° of K), starlight pattern: 0.001Lux (F1.2).
This binocular camera 1 is transferred to central processing module 2 to the real-time road condition information that collects.Central processing module 2 is handled the video image of camera acquisition by image processing software, and result is exported by voice module 3.
This utility model comprises the following steps: based on the method for the blind-guide device assisting blind walking of stereoscopic vision
1. left lens camera 11, the right lens camera 12 with described binocular camera 1 is installed in blind person's medicated cap the place ahead symmetrically, the center distance of overhead contour and left lens camera 11, right lens camera 12 is fixed, this left side lens camera 11, right lens camera 12 link to each other with described central processing module 2, the outfan of described central processing module 2 links to each other with described voice output module 3, with the earphone plug of voice module in blind person's earhole;
2. start described blind-guide device, left lens camera 11, right lens camera 11 are gathered traffic information synchronously in real time; The video image that central processing module 2 absorbs left lens camera 11, right lens camera 12 respectively with the speed of per second 2-8 frame becomes the image sequence of successive framing and keeps in;
The image sequence of the successive framing that 3. described central processing module 2 is at first gathered left lens camera 11 or right lens camera 12 is converted to gray level image, adopt medium filtering to carry out the image pretreatment then, and every two field picture carried out Canny rim detection and Hough change detection straight line, extract road edge and also judge whether to be road edge:
When being road edge, carry out step simultaneously 4., 5., 6.;
When being not road edge, carry out step simultaneously 4., 5.;
4. barrier is handled: the consecutive image frame sequence that 2 pairs of monocular-cameras of described central processing module (left lens camera 11 or right lens camera 12) are gathered, at the trapezoidal area-of-interest of selecting on the image on the user direction of advance, barrier region is extracted according to gray value in this zone, extract the barrier target; After detecting the barrier target, the same two field picture that 2 pairs of binocular cameras of described central processing module, 1 another monocular-camera (right lens camera 12 or left lens camera 11) is gathered, the distance of employing parallax method dyscalculia thing;
5. traffic light are handled: the consecutive image frame sequence that 2 pairs of left lens cameras 11 of described central processing module or right lens camera 12 are gathered, carry out traffic light according to template gray scale coupling and handle, provide the voice suggestion of traffic light color and signal lights position then by voice module;
6. departing from road handles: described central processing module 2 judges that according to the position and the inclination angle of road edge line in image that extracts user departs from the situation of road, provides voice suggestion.
Elaborate below:
Fig. 2 is this utility model device main program flow chart, and after device started, binocular camera 1 was gathered video information in real time, is input to central processing module 2.Central processing module 2 becomes the consecutive image frame sequence with the speed of per second 2～8 frames with video.The image that binocular camera 1 is gathered carries out pretreatment, carries out road-edge detection then:
As shown in Figure 3, the flow process of road-edge detection is: the consecutive image frame sequence that binocular camera 1 is gathered carries out colored conversion to gray scale, adopts medium filtering to carry out pretreatment then, removes the noise in the image.Then image is become left and right sides two halves image, carry out Canny rim detection and Hough conversion straight-line detection respectively.Detected straight line is judged, if length (is set according to practical situation by the programming personnel greater than threshold value, be set to 200 herein), and the angle of straight line and image perpendicular bisector is in (0 °～90 °) scope, think that then this straight line is the road edge line, the inclination angle and the zero that provide the road edge line arrive collinear distance.According to the road edge line that left and right sides two halves image extracts, can determine the road area scope.
If detect the road edge line in the image, move 3 subprograms simultaneously:
(a) barrier is handled and is comprised detection of obstacles and obstacle distance measurement (seeing subprogram declaration for details)
(b) traffic light are handled (seeing subprogram declaration for details)
(c) depart from road and handle (seeing subprogram declaration for details).
If do not detect road edge line (for example walking on the square) in the image, then move 2 subprograms simultaneously:
(a) barrier is handled and is comprised detection of obstacles and obstacle distance measurement (seeing subprogram declaration for details)
(b) traffic light are handled (seeing subprogram declaration for details);
Each subprogram activation returns and reads the next frame image after finishing.
Subprogram declaration:
Subprogram (a): barrier is handled: as shown in Figure 2, at first carry out detection of obstacles:
The flow process of detection of obstacles is at first imported a two field picture of the consecutive image frame sequence of monocular-camera (for example left lens camera 11) shooting as shown in Figure 4.Area-of-interest (ROI zone) according to the practical situation setting, as shown in Figure 5, is set to the trapezoid area that the user direction of advance is made of ABCD herein by the software design personnel.The coordinate of A, B, C, D is respectively (180,200), (180,440), (480,40) and (480,600).The image file in ROI zone is stored in the central processing module 2 in advance by the programming personnel.Input picture and ROI zone are got common factor, extract the barrier target according to the difference of barrier and road surface gray value.Gray value extracted region is within the specific limits come out, and the gray value scope is set according to practical situation by the programming personnel, is set to 1-120 herein.Then corrosion treatment is done in the zone that extracts, again close regional connectivity.The connected region area is judged area is dropped less than the zone of threshold value (set according to practical situation by the programming personnel, be set to 300 herein).The area that extracts is transformed into the rectangular area greater than the connected region of threshold value.Calculate the gray value deviation of each rectangular area and image, deviation is a barrier greater than the regional determination of threshold value (set according to practical situation by the programming personnel, be set to 5 herein).
If detect barrier, to carry out obstacle distance and measure, the method that obstacle distance is measured is as follows:
After in the image that monocular-camera (supposing left lens camera 11 herein) is taken, detecting the barrier target, get the same two field picture that another video camera in the binocular camera (promptly being right lens camera 12) is taken herein, adopt parallax method to ask obstacle distance.Promptly in the same two field picture that right lens camera 12 is taken, on the position of (or being called row-coordinate) of identical abscissa with the barrier target, seek the optimal match point of barrier target.With the barrier target is the rectangular window at selected one 3 * 3 of center, is that similarity is measured with the gray scale correlation function in this two field picture that right lens camera 12 is taken, as follows formula:
Wherein (x y) is (x, y) similarity of Dui Ying two neighborhood windows to S.I (x+x ', the y+y ') gray value that to be the order picture locate at (x+x ', y+y ').
It is the average gray of target image neighborhood window; T (x ', y ') is the gray value that source images (x ', y ') is located,
It is the average gray of source images neighborhood window.When seeking the corresponding point of characteristic point in the left image in right image, right image is a target image, and left image is a source images.With the barrier target is that 3 * 3 windows at center are called the neighborhood window, is called left neighborhood window corresponding to the neighborhood window of left image, is called right neighborhood window corresponding to the neighborhood window of right image.The point of similarity maximum is the point of barrier target correspondence in right figure.Calculate then barrier target and its corresponding point about coordinate poor of two width of cloth images, and convert this parallax in the world coordinates distance.Formula is tried to achieve obstacle distance below utilizing:
Wherein: f is the focal length of two photographic head of binocular camera, and d is the distance of two photographic head of binocular camera, x
lAnd x
rBe respectively the barrier target about row coordinate in two width of cloth images.s
xBe the Pixel Dimensions convergent-divergent on the x direction, pixel unit is transformed into iu (as: rice).Then the disturbance in judgement object distance from whether less than threshold value (for example 5 meters).If less than threshold value, voice module provides the barrier prompting.
Subprogram (b): traffic light are handled.
The processing of traffic light adopts the method for template matching to realize.The template matching technology is by providing the testee prototype for device, the similarity between the relevant pose of all of calculation template and each position of image is sought target object in the image of input.The template matching of traffic light uses a template image to describe searched green light.The template image file of green light is stored in the central processing module in advance by the programming personnel.The real time imaging frame sequence and the template image of camera acquisition mate, and matching similarity judges that in threshold range the match is successful, and voice module provides the prompting of signal lamp color and green light centre coordinate position in image.Formula below the similarity of template matching adopts:
Wherein (r c) is similarity between present image and the template to s.The gray value of each point is that (u, v), the gray value when the template zone moves on to the image current location is that (r+u c+v), calculates a scalar value as similarity based on these gray values to f to t in the template.If template is identical with image, the similarity that obtains is 0.If image is different with template, similarity will be greater than 0.Threshold size by the software design personnel according to practical situation setting (be set to 20 herein, the match is successful less than 20 o'clock to show the similarity error).
Subprogram (c): depart from road and handle.
According to the result of road-edge detection, the inclination angle and the zero that obtain the road edge line arrive collinear distance.As shown in Figure 5, the inclination angle is the angle α of road edge line L and image perpendicular bisector, and zero (0,0) is D to the distance of L, and α and D are obtained by road-edge detection.The intersection point of road edge line L and image lower limb is P.E point and F point are respectively 1/4 and 3/4 places of image lower limb.For the image of 640 * 480 pixels, the E point coordinates is (480,160), and the F point coordinates is (480,480).The width R of image is 480 pixels.According to the triangle similarity relation, the x axial coordinate x that P is ordered
pCalculate by following formula:
When α greater than 60 °, or the P point judges route deviation taken place between E, F the time, provides corresponding voice suggestion.
This utility model device is through on probation and test shows, and the central processing module of this utility model device becomes the consecutive image frame sequence with the speed of per second 2,3,4,5,6,7,8 frames with video and all is suitable for.This utility model device can adapt to that natural light is strong and weak to be changed, by the voice suggestion of device, but assisting blind walk without barrier in indoor and outdoor, safety is passed through the intersection.
Claims (6)
1. blind-guide device based on stereoscopic vision, it is characterized in that this device comprises binocular camera, central processing module and voice output module, the left lens camera of described binocular camera links to each other with described central processing module with right lens camera, described central processing module has the barrier processing, traffic lights is handled and departed from road and is treated to main image processing software, and the outfan of this central processing module links to each other with described voice output module.
2. the blind-guide device based on stereoscopic vision according to claim 1 is characterized in that described binocular camera is super wide dynamic for adopting, and high light suppresses, low-illuminance cameras.
3. the blind-guide device based on stereoscopic vision according to claim 1 and 2 is characterized in that described binocular camera employing dynamic range is 1: 400 a super wide dynamic low-illuminance cameras.
4. according to each described blind-guide device of claim 1 to 3, it is characterized in that it is the super wide dynamic low-illuminance cameras of 0.05Lux that described binocular camera adopts minimal illumination based on stereoscopic vision.
5. the blind-guide device based on stereoscopic vision according to claim 1 is characterized in that described central processing module adopts the embedded machine vision image processing platform.
6. the blind-guide device based on stereoscopic vision according to claim 1 is characterized in that described voice output module is voice module or the earphone that described central processing module carries.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010206265022U CN201870808U (en) | 2010-11-24 | 2010-11-24 | Guide device for blind people based on stereoscopic vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010206265022U CN201870808U (en) | 2010-11-24 | 2010-11-24 | Guide device for blind people based on stereoscopic vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201870808U true CN201870808U (en) | 2011-06-22 |
Family
ID=44158745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010206265022U Expired - Fee Related CN201870808U (en) | 2010-11-24 | 2010-11-24 | Guide device for blind people based on stereoscopic vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201870808U (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104316919A (en) * | 2014-10-20 | 2015-01-28 | 贝壳网际（北京）安全技术有限公司 | System and method for warning road condition, mobile terminal and radar detection apparatus |
| CN105997447A (en) * | 2016-04-27 | 2016-10-12 | 绍兴文理学院 | Blind guiding robot with wheel leg structure and utilization method for same |
| CN106859929A (en) * | 2017-01-25 | 2017-06-20 | 上海集成电路研发中心有限公司 | A kind of Multifunctional blind person guiding instrument based on binocular vision |
| CN107065920A (en) * | 2016-12-26 | 2017-08-18 | 歌尔科技有限公司 | Avoidance obstacle method, device and unmanned plane |
-
2010
- 2010-11-24 CN CN2010206265022U patent/CN201870808U/en not_active Expired - Fee Related
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104316919A (en) * | 2014-10-20 | 2015-01-28 | 贝壳网际（北京）安全技术有限公司 | System and method for warning road condition, mobile terminal and radar detection apparatus |
| CN104316919B (en) * | 2014-10-20 | 2017-12-29 | 贝壳网际（北京）安全技术有限公司 | Road conditions system for prompting, method, mobile terminal and radar detection apparatus |
| CN105997447A (en) * | 2016-04-27 | 2016-10-12 | 绍兴文理学院 | Blind guiding robot with wheel leg structure and utilization method for same |
| CN107065920A (en) * | 2016-12-26 | 2017-08-18 | 歌尔科技有限公司 | Avoidance obstacle method, device and unmanned plane |
| CN106859929A (en) * | 2017-01-25 | 2017-06-20 | 上海集成电路研发中心有限公司 | A kind of Multifunctional blind person guiding instrument based on binocular vision |
| CN106859929B (en) * | 2017-01-25 | 2019-11-22 | 上海集成电路研发中心有限公司 | A kind of Multifunctional blind person guiding instrument based on binocular vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101999972B (en) | Stereoscopic vision based auxiliary walking device for blindmen and auxiliary method thereof | |
| CN102389361B (en) | Blindman outdoor support system based on computer vision | |
| CN106214437B (en) | A kind of intelligent blind auxiliary eyeglasses | |
| CN106859929B (en) | A kind of Multifunctional blind person guiding instrument based on binocular vision | |
| CN103868460B (en) | Binocular stereo vision method for automatic measurement based on parallax optimized algorithm | |
| CN106203390B (en) | A kind of intelligent blind auxiliary system | |
| CN103413313B (en) | The binocular vision navigation system of electrically-based robot and method | |
| CA2950791C (en) | Binocular visual navigation system and method based on power robot | |
| CN104036488B (en) | Binocular vision-based human body posture and action research method | |
| WO2020029444A1 (en) | Method and system for detecting attention of driver while driving | |
| CN105956586A (en) | Intelligent tracking system based on TOF 3D camera | |
| CN107397658B (en) | Multi-scale full-convolution network and visual blind guiding method and device | |
| CN103900494B (en) | For the homologous points fast matching method of binocular vision 3 D measurement | |
| CN104899869A (en) | Plane and barrier detection method based on RGB-D camera and attitude sensor | |
| CN106918299A (en) | A kind of line-structured light machine vision tire wear measuring method | |
| CN104751146B (en) | A kind of indoor human body detection method based on 3D point cloud image | |
| CN201870808U (en) | Guide device for blind people based on stereoscopic vision | |
| CN106778713A (en) | The iris identification device and method of a kind of dynamic human eye tracking | |
| CN104318237A (en) | Fatigue driving warning method based on face identification | |
| CN105014675B (en) | A kind of narrow space intelligent mobile robot vision navigation system and method | |
| CN105182320A (en) | Depth measurement-based vehicle distance detection method | |
| CN107225570A (en) | The barrier-avoiding method and device of intelligent robot | |
| KR102014228B1 (en) | Path detection system based on solar-blind ultraviolet optical signal | |
| JP2002024986A (en) | Pedestrian detector | |
| CN109164802A (en) | A kind of robot maze traveling method, device and robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110622 Termination date: 20111124 |