CN106092086B - A kind of quick, high robust robot indoor orientation method based on panoramic vision - Google Patents
A kind of quick, high robust robot indoor orientation method based on panoramic vision Download PDFInfo
- Publication number
- CN106092086B CN106092086B CN201610407064.2A CN201610407064A CN106092086B CN 106092086 B CN106092086 B CN 106092086B CN 201610407064 A CN201610407064 A CN 201610407064A CN 106092086 B CN106092086 B CN 106092086B
- Authority
- CN
- China
- Prior art keywords
- road sign
- robot
- circle
- center
- road
- 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.)
- Active
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
-
- 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
- G01C21/20—Instruments for performing navigational calculations
- G01C21/206—Instruments for performing navigational calculations specially adapted for indoor navigation
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Image Analysis (AREA)
- Manipulator (AREA)
Abstract
The invention belongs to mobile robot visual field of locating technology, and in particular to a kind of quick, high robust robot indoor orientation method based on panoramic vision.The present invention includes:(1) road sign based on SURF characteristic points is designed;(2) quickly the road sign of design is accurately identified using improved SURF Feature Points Matchings algorithm in panoramic picture;(3) according to the recognition result for three road signs being arranged in environment calculate robot place two-dimensional space accurate location.The present invention realizes the indoor positioning function of mobile robot, suitable for the robot localization task under complex indoor environment, it can be widely applied in the navigation of home-services robot, industrial robot, the rapidity and robustness of algorithm can guarantee that robot has more accurate stationkeeping ability during the motion.
Description
Technical field
The invention belongs to mobile robot visual field of locating technology, and in particular to it is a kind of based on panoramic vision it is quick,
The robot indoor orientation method of high robust.
Background technology
The self-positioning problem of mobile robot is always the underlying issue and hot issue of robot research field.Machine regards
Feel the advantages such as contain much information by it, is cheap, is widely used in the research of robot localization.
In traditional vision positioning method, the localization method based on physical feature usually requires to extract a large amount of feature,
Algorithm operation quantity is larger, does not have real-time, and be vulnerable to light variation or environmental change influenced.Determined based on artificial landmark
The characteristics of position method then has higher flexibility, passes through manually-set road sign in contrast can be targetedly to road sign
It is identified, reduces the design difficulty of recognizer, improve efficiency of algorithm.
In the selection of sensor, omni-directional visual can obtain 360 degree of comprehensive visual field, with traditional visual imaging
Equipment, which is compared, can obtain more environmental informations, therefore obtain significant progress for the research of the technology in recent years.So
And panorama camera institute the image collected is the image for having distortion, therefore usually positioned using the colouring information of road sign,
This is just relatively high to the requirement of the light of environment, and requires the interference that same color is not present in environment, poor robustness, it is difficult to using
Under more complicated environment.
The method that the present invention designs solves the indoor positioning rapidity and robustness problem of mobile robot, establishes whole
A indoor locating system has important reference for robot indoor positioning, may be directly applied to home services machine
The fields such as device people, industrial robot.
Invention content
The object of the present invention is to provide quick, the high robusts based on panoramic vision under a kind of complex indoor environment
Robot indoor orientation method.
The object of the present invention is achieved like this:
(1) road sign based on SURF characteristic points is designed;
(1.1) utilize SURF feature point extractions algorithm to circle, triangle, quadrangle, pentagonal basis geometry into
Row characteristic point detects, and records the feature points of each shape;
(1.2) these shapes are positioned under different environment, it is imaged using panoramic vision, recorded every in image
The matching rate of the characteristic point and shape exemplary feature point of geometry;
(1.3) it is analyzed by the data to record, comprehensive design goes out readily identified road sign shape;
(2) it is accurate quickly to be carried out to the road sign of design using improved SURF Feature Points Matchings algorithm in panoramic picture
Identification;
(2.1) according to refraction-reflection panorama camera image-forming principle, it is circle that there are one with image center of fiqure in panoramic picture
The circle of the heart, the object point being located in same level with curved surface mirror foci will be imaged on this circle, using the circle characteristic by road
Mark is set on the wall higher than video camera, by pixel within the circle zero setting when characteristic point detects;
(2.2) a circular detection window is set, is determined by experiment the correct radial of window, further reduces feature
Point detection zone, simplifies region of search, shortens operation time;
(2.3) pass through the extraction to road sign outer profile, it is determined that the profile of road sign center circle, by the circle for calculating center circle
The heart realizes being accurately positioned for road sign;
(2.4) successfully after identification road sign using the road sign recognized as new template, real-time update template;
(3) according to the recognition result for three road signs being arranged in environment calculate robot place two-dimensional space standard
True position;
(3.1) three road signs for being higher than robot are placed in the environment, and three road signs are identified, determine it in panorama
Coordinate in image calculates the angle with panorama center of fiqure;
(3.2) two equation of a circles are determined according to two angle values and known road sign actual coordinate;
(3.3) two equation of a circle simultaneous are acquired into two intersection points, one of them is the coordinate of intermediate road sign, another is
The two-dimensional coordinate of robot.
The beneficial effects of the present invention are:
The present invention realizes the indoor positioning function of mobile robot, the robot localization being suitable under complex indoor environment
Task can be widely applied in the navigation of home-services robot, industrial robot, and the rapidity and robustness of algorithm can guarantee
Robot has more accurate stationkeeping ability during the motion.
Description of the drawings
Fig. 1 is the road sign designed in the present invention;
Fig. 2 is the improved landmark identification algorithm flow chart based on SURF characteristic matchings in the present invention.
Fig. 3 is the system structure diagram of the present invention;
Specific implementation mode
The present invention is described in detail below in conjunction with the accompanying drawings.
The present invention devises a kind of quick, high robust mobile robot indoor orientation method based on artificial landmark,
Including artificial landmark design, landmark identification algorithm and location algorithm based on panoramic vision.For color road sign by light intensity shadow
It rings big problem and devises the road sign for being conducive to Feature point recognition, it is proposed that the artificial landmark recognizer of feature based point, profit
Landmark identification is carried out with the shape information (characteristic point) rather than colouring information of road sign, overcome color recognizer is influenced by light intensity
The problem of;Caused by the factors such as big, pattern distortion, visual angle change and environment complexity accidentally for panoramic picture feature extraction operand
Identification problem is improved on the basis of traditional characteristic point matching algorithm, by reducing detection zone, setting detection window, reality
When more the methods of new template devise a kind of fast and accurately landmark identification algorithm;By the way that three road signs are arranged indoors, on road
The location algorithm of robot is realized on the basis of mark recognizer.The present invention realizes the indoor positioning work(of mobile robot
Can, the robot localization task being suitable under complex indoor environment can be widely applied to home-services robot, industrial robot
Navigation in, the rapidity and robustness of algorithm can guarantee that robot has more accurate stationkeeping ability during the motion.
Quick, high robust robot indoor orientation method based on panoramic vision, devises the shape of artificial landmark,
Road sign in panoramic picture is identified using improved SURF Feature Points Matchings algorithm, by the position for recognizing three road signs
Carry out absolute fix of the robot in two-dimensional space.Wherein:
(1) the road sign design based on SURF feature point detection algorithms, the road sign of design have more characteristic point, are easy to know
Not and it is accurately positioned;
(2) traditional SURF Feature Points Matching algorithms are improved, improve the standard of the rapidity and identification of algorithm operation
True property, quickly accurately identifies the road sign of design in panoramic picture;
(3) according to the recognition result for three road signs being arranged in environment calculate robot place two-dimensional space standard
True position.
In some embodiments, the design of artificial landmark is specially:
(1) basis geometries such as circle, triangle, quadrangle, pentagon are carried out using SURF feature point extractions algorithm
Characteristic point detects, and records the feature points of each shape;
(2) these shapes are positioned under different environment, it is imaged using panoramic vision, recorded every several in image
The matching rate of the characteristic point and shape exemplary feature point of what shape;
(3) by the data to record carry out it is analysis integrated design readily identified road sign shape, which has not
It is easy to the characteristic point misidentified and characteristic point is more is conducive to identify, the shape at the road sign center in addition designed is conducive to pair
It is accurately positioned.
In some embodiments, improved SURF Feature Points Matchings algorithm is specially:
(1) according to the image-forming principle of refraction-reflection panorama camera, it is circle that there are one with image center of fiqure in panoramic picture
The circle of the heart, the object point being located in same level with curved surface mirror foci will be imaged on this circle, using the circle characteristic by road
Mark is set on the wall higher than video camera, by pixel within the circle zero setting when characteristic point detects, is reduced region of search, is eliminated
It interferes part;
(2) it is provided with a circular detection window, the correct radial of window is determined through experimentation, further reduces
Characteristic point detection zone, simplifies region of search as far as possible, shortens operation time;
(3) pass through the extraction to road sign outer profile, it is determined that the profile of road sign center circle, by the center of circle for calculating center circle
Realize being accurately positioned for road sign.
(4) successfully after identification road sign using the road sign recognized as new template, real-time update template.
In some embodiments, localization method, which designs, is specially:
(1) three road signs for being higher than robot are placed in the environment, utilize three roads of landmark identification method pair of the present invention
Mark is identified, and determines its coordinate in panoramic picture, calculates the angle with panorama center of fiqure;
(2) two equation of a circles are determined according to two angle values and known road sign actual coordinate;
(3) two equation of a circle simultaneous are acquired into two intersection points, one of them is the coordinate of intermediate road sign, another is machine
The two-dimensional coordinate of device people.
(1) the matched road sign design of feature based point
The road sign for being provided with some basic configurations first is tested, including annulus, triangle, rectangle, pentagon, six sides
Shape and five-pointed star, it is to seek central point for convenience to be positioned to select axisymmetric shape, size be 290mm ×
400mm.These road signs are placed in three different indoor environments, panoramic vision acquisition image, detection characteristic point are utilized.It will
Road sign is placed in the position remote away from camera 3m, and the characteristic point detected and template are matched, according to correct of characteristic point
Go out with quantitative analysis:The feature points being correctly matched to are directly proportional to its angle point quantity.Therefore road sign as shown in Figure 1 is devised
Shape.Symmetrical shape ensures each visual angle resolution all having the same;There are the guarantee of more angle point is higher in road sign
Discrimination;The circle at the center target that is conducive to satisfy the need is accurately positioned.
(2) the improved landmark identification algorithm based on SURF Feature Points Matchings
In order to improve the speed of service and recognition success rate of algorithm, lane marker detection region is reduced using a series of means.It is dynamic
State updates road sign masterplate to eliminate the influence that pattern distortion is brought, and final design recognizer flow chart is referring to Fig. 2.
First, according to panorama camera image-forming principle, curved mirror ideally, there are one to scheme in panoramic picture
As the circle that center of fiqure is the center of circle, the object point being located in same level with curved surface mirror foci will be imaged on this circle.Indoor unit
The space of device people is plane, then the pixel on the circle is always the object point in same level, higher than this height
The corresponding pixel of object is located above the circle.In order to avoid blocking for ground object, the placing height of road sign is more than in robot
The mounting height of panorama camera.Therefore can using this characteristic using the pixel outside the circle as characteristic point detection zone,
Rest of pixels Ignore All both improves the speed of service in this way, and eliminates a part of noise spot, improves matching rate.
Then, one circular detection window is set using the road sign center recognized as the center of circle, by the region in the circle
As the region for detecting characteristic point next time, detection zone is further reduced.According to image-forming principle and mobile robot
Actual speed determines displacement distance of the robot under motion conditions in the road sign unit interval in panoramic picture, is set as P picture
Element, if the landmark identification algorithm cycle of operation is t, then circle detection windows radius is P*t, and it is 305 pixels to acquire P, takes t=
1s, the detection window through testing the radius do not influence the identification of road sign in robot motion, further improve recognition speed,
Eliminate part noise spot.
Next, existing under distortion and different visual angles for panoramic picture, distortion shape is different, and the present invention is by real time more
The method of new road sign template, using the road sign where the characteristic point being matched to every time as new mould after initializing road sign template
Plate, with the method for reducing detection zone is closed, in the case where execution cycle is short, robot motion's a small distance will update
Template is able to ensure the accuracy of template matches.
Finally, after determining the region of road sign, in order to obtain higher positioning accuracy, it is thus necessary to determine that road sign center institute
Exact position, i.e., the center of circle of circle in the road sign that designs herein is positioned.First to the region that detects from background
Separation, utilizes the marginal information of Canny operator extraction road signs.The pixel shared by profile due to road sign center circle is seldom and distorts
It is more serious, so being difficult with the means such as Hough loop truss detects center circle.The present invention is maximum by search area first
Profile finds the outer edge of road sign, finds out the center of fiqure of road sign entirety on this basis, and center circle is nearest with a distance from center of fiqure
One profile, therefore minimum one group of closed pixel sequence with a distance from center of fiqure is the center circle of road sign, finds out the profile
Center is both the position at road sign center.
(3) location algorithm
Referring to Fig. 3, the system is by three or more road signs, mobile robot and data processing system equipped with panoramic camera
Composition.The artificial landmark set using three realizes robot localization using the triangle polyester fibre algorithm based on panoramic vision.Three
Angle location algorithm is the artificial landmark that three or more are placed in experimental enviroment, using three road signs in panoramic picture relative to
The angular relationship of robot according to geometrical principle so that it is determined that robot two-dimensional space coordinate.Its algorithm principle is:If
The coordinate of known two road signs, P1 (x1, y1), P2 (x2, y2) and from robot to the angle a two road sign lines1Also may be used
With identification road sign success after directly acquire, then a circle can be determined according to these three amounts, according to same arc to angle of circumference
Equal principle, robot must be on this circles, and condition that this satisfactory foot is following:
1) P1 (x1, y1), P2 (x2, y2) meet the equation of circle;
2) angle of robot to two road signs is equal to a1。
Any two road sign and their angle in this way then corresponds to a round equation, when the road sign number in image is
N, and when N >=3, N (N-1)/2 equation of a circle can be obtained.Due to being multiple road signs in terms of same viewpoint, viewpoint one is scheduled on these
On round intersection point, the intersection point for solving these equation of a circles is the coordinate of the viewpoint i.e. the coordinate of camera optical center, i.e. machine
People position.
Claims (1)
1. a kind of quick, high robust robot indoor orientation method based on panoramic vision, which is characterized in that including as follows
Step:
(1) road sign based on SURF feature point extraction algorithms is designed;
(1.1) circle, triangle, quadrangle, pentagonal basis geometry are carried out using SURF feature point extractions algorithm special
Sign point detection, records the feature points of each shape;
(1.2) these shapes are positioned under different environment, it is imaged using panoramic vision, record every geometry in image
The matching rate of the characteristic point and shape exemplary feature point of shape;
(1.3) it is analyzed by the data to record, comprehensive design goes out readily identified road sign shape;
(2) quickly the road sign of design is accurately known using improved SURF Feature Points Matchings algorithm in panoramic picture
Not;
(2.1) according to refraction-reflection panorama camera image-forming principle, there are one using image center of fiqure as the center of circle in panoramic picture
Circle, the object point being located in same level with curved surface mirror foci will be imaged on this circle, be set road sign using the circle characteristic
It is placed higher than on the wall of video camera, by pixel within the circle zero setting when characteristic point detects;
(2.2) a circular detection window is set, the correct radial of window is determined by experiment, further reduces characteristic point inspection
Region is surveyed, region of search is simplified, shortens operation time;
(2.3) pass through the extraction to road sign outer profile, it is determined that the profile of road sign center circle, it is real by the center of circle for calculating center circle
Existing road sign is accurately positioned;
(2.4) successfully after identification road sign using the road sign recognized as new template, real-time update template;
(3) basis calculates robot in the accurate position of place two-dimensional space to the recognition result for three road signs being arranged in environment
It sets;
(3.1) three road signs for being higher than robot are placed in the environment, and three road signs are identified, determine it in panoramic picture
In coordinate, calculate the angle with panorama center of fiqure;
(3.2) two equation of a circles are determined according to two angle values and known road sign actual coordinate;
(3.3) two equation of a circle simultaneous are acquired into two intersection points, one of them is the coordinate of intermediate road sign, another is machine
The two-dimensional coordinate of people.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610407064.2A CN106092086B (en) | 2016-06-12 | 2016-06-12 | A kind of quick, high robust robot indoor orientation method based on panoramic vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610407064.2A CN106092086B (en) | 2016-06-12 | 2016-06-12 | A kind of quick, high robust robot indoor orientation method based on panoramic vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106092086A CN106092086A (en) | 2016-11-09 |
CN106092086B true CN106092086B (en) | 2018-08-31 |
Family
ID=57228617
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610407064.2A Active CN106092086B (en) | 2016-06-12 | 2016-06-12 | A kind of quick, high robust robot indoor orientation method based on panoramic vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106092086B (en) |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108122252A (en) * | 2016-11-26 | 2018-06-05 | 沈阳新松机器人自动化股份有限公司 | A kind of image processing method and relevant device based on panoramic vision robot localization |
CN106709942B (en) * | 2016-12-13 | 2020-05-19 | 广州智能装备研究院有限公司 | Panorama image mismatching elimination method based on characteristic azimuth angle |
CN106767833B (en) * | 2017-01-22 | 2019-11-19 | 电子科技大学 | A kind of robot localization method merging RGBD depth transducer and encoder |
CN106908040B (en) * | 2017-03-06 | 2019-06-14 | 哈尔滨工程大学 | A kind of binocular panorama visual robot autonomous localization method based on SURF algorithm |
CN107145906B (en) * | 2017-05-02 | 2020-06-16 | 哈尔滨工程大学 | Mobile robot indoor rapid homing method based on panoramic visual imaging system |
CN107478228A (en) * | 2017-07-13 | 2017-12-15 | 杭州品铂科技有限公司 | A kind of indoor orientation method |
CN108446725A (en) * | 2018-03-12 | 2018-08-24 | 杭州师范大学 | A kind of Image Feature Matching method of feature based triangle |
CN109099915B (en) * | 2018-06-27 | 2020-12-25 | 未来机器人(深圳)有限公司 | Mobile robot positioning method, mobile robot positioning device, computer equipment and storage medium |
CN109141432B (en) * | 2018-09-19 | 2020-04-10 | 西安科技大学 | Indoor positioning navigation method based on image space and panoramic assistance |
CN109974722B (en) * | 2019-04-12 | 2020-09-15 | 珠海市一微半导体有限公司 | Map updating control method and map updating control system of visual robot |
CN110332938B (en) * | 2019-06-20 | 2023-03-10 | 南京航空航天大学 | Indoor monocular self-positioning method based on mobile phone |
CN110928311B (en) * | 2019-12-16 | 2021-11-19 | 哈尔滨工业大学 | Indoor mobile robot navigation method based on linear features under panoramic camera |
CN111966041B (en) * | 2020-08-26 | 2021-10-08 | 珠海格力电器股份有限公司 | Robot control method and device |
CN113791377B (en) * | 2021-09-09 | 2024-04-12 | 中国科学院微小卫星创新研究院 | Positioning method based on angle measurement |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102135429A (en) * | 2010-12-29 | 2011-07-27 | 东南大学 | Robot indoor positioning and navigating method based on vision |
CN103901895A (en) * | 2014-04-18 | 2014-07-02 | 江苏久祥汽车电器集团有限公司 | Target positioning method based on unscented FastSLAM algorithm and matching optimization and robot |
CN104007760A (en) * | 2014-04-22 | 2014-08-27 | 济南大学 | Self-positioning method in visual navigation of autonomous robot |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102006062061B4 (en) * | 2006-12-29 | 2010-06-10 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Apparatus, method and computer program for determining a position based on a camera image from a camera |
TWI407280B (en) * | 2009-08-20 | 2013-09-01 | Nat Univ Tsing Hua | Automatic searching system and method |
-
2016
- 2016-06-12 CN CN201610407064.2A patent/CN106092086B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102135429A (en) * | 2010-12-29 | 2011-07-27 | 东南大学 | Robot indoor positioning and navigating method based on vision |
CN103901895A (en) * | 2014-04-18 | 2014-07-02 | 江苏久祥汽车电器集团有限公司 | Target positioning method based on unscented FastSLAM algorithm and matching optimization and robot |
CN104007760A (en) * | 2014-04-22 | 2014-08-27 | 济南大学 | Self-positioning method in visual navigation of autonomous robot |
Non-Patent Citations (2)
Title |
---|
单目视觉自然路标辅助的移动机器人定位方法;陈明芽等;《浙江大学学报(工学版)》;20140228;第48卷(第2期);全文 * |
基于EKF的全景视觉机器人SLAM算法;王开宇等;《计算机应用研究》;20131130;第30卷(第11期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN106092086A (en) | 2016-11-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106092086B (en) | A kind of quick, high robust robot indoor orientation method based on panoramic vision | |
CN105841687B (en) | indoor positioning method and system | |
US10060739B2 (en) | Method for determining a position and orientation offset of a geodetic surveying device and such a surveying device | |
Robertson et al. | An Image-Based System for Urban Navigation. | |
US6917702B2 (en) | Calibration of multiple cameras for a turntable-based 3D scanner | |
US7398928B2 (en) | Coded target and photogrammetry method using such targets | |
CN103256920B (en) | Determining tilt angle and tilt direction using image processing | |
CN108022264B (en) | Method and equipment for determining camera pose | |
CN106767810B (en) | Indoor positioning method and system based on WIFI and visual information of mobile terminal | |
Herráez et al. | 3D modeling by means of videogrammetry and laser scanners for reverse engineering | |
US7190826B2 (en) | Measuring the location of objects arranged on a surface, using multi-camera photogrammetry | |
CN106295512B (en) | Vision data base construction method and indoor orientation method in more correction lines room based on mark | |
KR100698534B1 (en) | Landmark for location recognition of mobile robot and location recognition device and method using same | |
CN105358937A (en) | Positioning method for a surveying instrument and said surveying instrument | |
CN106908040B (en) | A kind of binocular panorama visual robot autonomous localization method based on SURF algorithm | |
CN109471096B (en) | Multi-sensor target matching method and device and automobile | |
US20150269734A1 (en) | Apparatus and method for recognizing location of object | |
CN108007388A (en) | A kind of turntable angle high precision online measuring method based on machine vision | |
CN109919975B (en) | Wide-area monitoring moving target association method based on coordinate calibration | |
CN102773862A (en) | Quick and accurate locating system used for indoor mobile robot and working method thereof | |
CN109949365A (en) | Vehicle designated position parking method and system based on road surface characteristic point | |
CN108022265B (en) | Method, equipment and system for determining pose of infrared camera | |
CN102834691A (en) | Surveying method | |
Chu et al. | GPS refinement and camera orientation estimation from a single image and a 2D map | |
Liu et al. | Deep-learning and depth-map based approach for detection and 3-D localization of small traffic signs |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |