CN109978953A - Method and system for target three-dimensional localization - Google Patents

Method and system for target three-dimensional localization Download PDF

Info

Publication number
CN109978953A
CN109978953A CN201910345409.XA CN201910345409A CN109978953A CN 109978953 A CN109978953 A CN 109978953A CN 201910345409 A CN201910345409 A CN 201910345409A CN 109978953 A CN109978953 A CN 109978953A
Authority
CN
China
Prior art keywords
target
axis coordinate
coordinate value
current location
value
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
Application number
CN201910345409.XA
Other languages
Chinese (zh)
Inventor
幸浩洋
罗迪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sichuan University
Original Assignee
Sichuan University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Sichuan University filed Critical Sichuan University
Publication of CN109978953A publication Critical patent/CN109978953A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]

Abstract

The invention discloses a kind of method and system for target three-dimensional localization, belong to a kind of positioning method, include the following steps: that step A, identification module identify that positioning target includes first object, the second target and third target with the presence or absence of positioning target in present image by intelligent algorithm；Step B, after identification module recognizes target, measurement current location is respectively the distance between with first object, the second target and third target；Step C, locating module calculates the D coordinates value of first object, the second target and third target according to the distance that identification module is surveyed, by being found in positioning target and determining known three target points, then after distinguishing the distance between measuring target point and current location, to calculate the D coordinates value of three target points, it can be used as the guidance benchmark of automatic Pilot, positioning can be completed without marking special marking in advance in target position or installing electronic tag, particularly suitable for the positioning of undercarriage tire.

Description

Method and system for target three-dimensional localization
Technical field
The present invention relates to a kind of localization methods, more specifically, the invention mainly relates to one kind to be used for target three-dimensional localization Method and system.
Background technique
With the development of artificial intelligence technology, automatic Pilot, automatic parking and automatic Pilot technology similar therewith by It gradually popularizes, to location technology, more stringent requirements are proposed for this, if to guide motor vehicle towards set target position in practice Advance, it is necessary first to the D coordinates value that target position is obtained by image scanning or other modes, for three-dimensional coordinate It is more accurate to measure, then guides the accuracy of automatic Pilot also higher.Currently, similar technique is mostly in advance on target position Special marking is marked, the D coordinates value of target position is obtained using these special markings of image recognition；Also have in advance in mesh Cursor position installs electronic tag, identifies that electronic tag achieves positioning by sensing device, but in some special environment Or when positioning in place, electronic tag can not be marked or installed in target position, and then can not be in distal end accurately measurement The D coordinates value of target position to limit automatic guidance and automatic Pilot technology in the application in these fields, thus has Necessity is further studied and is improved for this kind of three-dimensional localization techniques.
Summary of the invention
One of the aims of the present invention is to provide...in response to the above drawbacks a kind of method for target three-dimensional localization and it is System, can not be in the D coordinates value of the accurate measurement target position in distal end, limitation it is expected to solve similar technique in the prior art The technical problems such as application of the location technology in the particular surroundings or place.
In order to solve the above technical problems, the invention adopts the following technical scheme:
One aspect of the present invention provides a kind of method for target three-dimensional localization, and the method includes the following steps:
Step A, identification module identifies that described determines with the presence or absence of positioning target in present image by intelligent algorithm Position target includes first object, the second target and third target；
Step B, after identification module recognizes target, measurement current location respectively with first object, the second target and the The distance between three targets；
Step C, the distance that locating module is surveyed according to identification module, is calculate by the following formula out first object, the second target With the D coordinates value (x, y, z) of third target:
In above formula, X1For the X axis coordinate value of first object, y1For the Y axis coordinate value of first object, z1For first object Z axis coordinate value, X2For the X axis coordinate value of the second target, y2For the Y axis coordinate value of the second target, z2It is sat for the Z axis of the second target Scale value, X3For the X axis coordinate value of third target, y3For the Y axis coordinate value of third target, z3For the Z axis coordinate value of third target, X is the X axis coordinate value of current location, and y is the Y axis coordinate value of current location, and z is the Z axis coordinate value of current location, R1To work as Distance value, R of the front position to first object2Distance value, R for current location to the second target3For current location to third target Distance value.
Preferably, further technical solution is: the first object, the second target and third target are aircraft takeoffs and landings Three tires on frame.
Further technical solution is: identification module is distinguished by laser ranging or binocular camera in the step B Current location is measured respectively the distance between with first object, the second target and third target.
Further technical solution is: being identified in present image in the step A by Faster-R-CNN algorithm is It is not no while including first object, the second target and third target；The present image is acquired by video camera or camera.
Another aspect of the present invention provides a kind of for target three-dimensional localization, system, it is characterised in that the system Including identification module and locating module, the identification module accesses locating module, in which:
Identification module is used to identify by intelligent algorithm in present image with the presence or absence of positioning target, the positioning Target includes first object, the second target and third target；After identification module recognizes target, measurement current location respectively with The distance between first object, the second target and third target；
Locating module is used for the distance surveyed according to identification module, be calculate by the following formula out first object, the second target with The D coordinates value (x, y, z) of third target:
In above formula, X1For the X axis coordinate value of first object, y1For the Y axis coordinate value of first object, z1For first object Z axis coordinate value, X2For the X axis coordinate value of the second target, y2For the Y axis coordinate value of the second target, z2It is sat for the Z axis of the second target Scale value, X3For the X axis coordinate value of third target, y3For the Y axis coordinate value of third target, z3For the Z axis coordinate value of third target, X is the X axis coordinate value of current location, and y is the Y axis coordinate value of current location, and z is the Z axis coordinate value of current location, R1To work as Distance value, R of the front position to first object2Distance value, R for current location to the second target3For current location to third target Distance value.
Preferably, further technical solution is: the first object, the second target and third target are aircraft takeoffs and landings Three tires on frame.
Further technical solution is: the identification module is measured currently respectively by laser ranging or binocular camera Position is respectively the distance between with first object, the second target and third target.
Further technical solution is: the identification module is identified in present image by Faster-R-CNN algorithm is It is not no while including first object, the second target and third target；The identification module also accesses video camera or camera, for by The video camera or camera acquisition present image are simultaneously transmitted to the identification module.
Compared with prior art, beneficial effects of the present invention first is that: pass through positioning target in find and determine known to Three target points, then respectively after the distance between measuring target point and current location, to calculate three target point institutes D coordinates value in position can be used as the guidance benchmark of automatic Pilot with this, without marking special mark in advance in target position Positioning can be completed in note or installation electronic tag, is especially suitable for carrying out the positioning of undercarriage tire, while institute of the present invention A kind of method and system for target three-dimensional localization of offer are easy to operate, and field of employment and environment are not limited substantially System, is suitable for popularization and application.
Detailed description of the invention
Fig. 1 is the method flow diagram for illustrating one embodiment of the invention；
Fig. 2 is the schematic illustration for illustrating one embodiment of the invention；
Fig. 3 is the system structure schematic block diagram for illustrating one embodiment of the invention；
In figure, 1 it is first object, 2 is the second target, 3 is third target, 4 is current location.
Specific embodiment
The present invention is further elaborated with reference to the accompanying drawing.
Refering to what is shown in Fig. 1, one embodiment of the present of invention is a kind of method for target three-dimensional localization, the mesh of this method The D coordinates value by measuring and being calculated three target points in distal end, base as subsequent automatic guidance/automatic Pilot Standard, as follows on undercarriage being in three tires of product word arrangement (first tire 1 shown in figure, the second tire 2, Third tire 3), the implementation of this method is illustrated:
Step S1, identification module is identified in present image by intelligent algorithm with the presence or absence of positioning target, the positioning Target is three, i.e., is in three tires of product word arrangement on above-mentioned undercarriage, identification module judgement is current in this step Whether three tires are existed simultaneously in image；
Step S2, after identification module recognizes in present image while including three tires, measurement current location 4 is distinguished The distance between three tires, i.e. R1、R2、R3；It is specific as shown in Figure 2；
Step S3, the distance that locating module is surveyed according to identification module, since the mutual distance of three tires is also Known value, therefore be calculate by the following formula out three actual D coordinates values of tire, i.e., tri- axis of X, Y, Z coordinate value (x, y, Z):
In above formula, X1For the X axis coordinate value of first tire, y1For the Y axis coordinate value of first tire, z1For first tire Z axis coordinate value, X2For the X axis coordinate value of the second tire, y2For the Y axis coordinate value of the second tire, z2It is sat for the Z axis of the second tire Scale value, X3For the X axis coordinate value of third tire, y3For the Y axis coordinate value of third tire, z3For the Z axis coordinate value of third tire, X is the X axis coordinate value of current location, and y is the Y axis coordinate value of current location, and z is the Z axis coordinate value of current location, R1To work as Distance value, R of the front position to first tire2Distance value, R for current location to the second tire3For current location to third tire Distance value.
In the present embodiment, known three tires are found and determined by middle aboard, then measure tire respectively Behind the distance between current location, to calculate the D coordinates value of three tire positions, it can be used as automatically with this The practical institute of aircraft can be completed without marking special marking in advance in target position or installing electronic tag in the guidance benchmark of driving Positioning in position.
Simultaneously whether the intelligent algorithm in above-mentioned steps S1 identified in present image by Faster-R-CNN algorithm Include first tire 1, the second tire 2 and third tire 3；Conventional DV or photograph can be used in present image above-mentioned Acquired image, is transmitted in identification module by video camera or camera and carries out image recognition by machine acquisition.
Above-mentioned Faster-R-CNN is a kind of algorithm that similar neural network is identified for characteristics of image, and having can find The convolutional layer of characteristics of image generates the zonule comprising target using Region Proposal algorithm, chases after later using iteration Track determines the coordinate of target.In this way, the picture comprising undercarriage tire can be input to the nerve net of above-mentioned algorithm building Network, and then the image with tag along sort or the input of other data, there is corresponding label (output), by a large amount of training data, The design factor that can determine each neuron in hidden layer and output layer judges current figure using training data by neural network It whether include label (i.e. tire) as in, and then by whether there is undercarriage in above-mentioned identification module identification present image Tire carries out ranging convenient for next step.
In addition, above-mentioned steps S2 can be measured by laser ranging current location respectively with first tire 1, the second tire 2 with The distance between third tire 3, i.e. R1、R2、R3；Laser ranging generally emits laser by rangefinder, and laser is through object being measured Reflection after again received by rangefinder, rangefinder simultaneously the recording laser round-trip time, the one of the light velocity and the product of two-way time Half, it is exactly the distance between rangefinder and object being measured.In addition to this, binocular camera can be used in step S2 also to carry out Ranging uses twin-lens camera, the spacing (b) between known two camera lens, focal length (f), target P is in two cameras The position (xl, xr) of upper imaging, in the case of, longitudinal depth information z can be calculated through Depth z=f*b/ (xl-xr).Using Two ways above-mentioned can measure the distance between current location and first tire 1, the second tire 2 and third tire 3, practice In can be selected according to the actual situation.
Refering to what is shown in Fig. 3, another embodiment of the invention is a kind of for realizing above-mentioned target 3-D positioning method System, which includes identification module and locating module, and identification module need to access locating module, and identification module need to access photograph Camera or video camera, in which:
Identification module be used for by intelligent algorithm identification camera or the collected present image of video camera in whether In the presence of positioning target, and the positioning target includes first object 1, the second target 2 and third target 3；When identification module recognizes After target, measurement current location is transmitted to calmly respectively the distance between with first object 1, the second target 2 and third target 3 In the module of position, aforementioned measurement is generally realized that principle is as described above by laser range finder；
Locating module is used for the distance surveyed according to identification module, be calculate by the following formula out first object, the second target with The D coordinates value (x, y, z) of third target:
Likewise, in above formula, X1For the X axis coordinate value of first object, y1For the Y axis coordinate value of first object, z1It is The Z axis coordinate value of one target, X2For the X axis coordinate value of the second target, y2For the Y axis coordinate value of the second target, z2For the second mesh Target Z axis coordinate value, X3For the X axis coordinate value of third target, y3For the Y axis coordinate value of third target, z3For the Z of third target Axial coordinate value, X are the X axis coordinate value of current location, and y is the Y axis coordinate value of current location, and z is the Z axis coordinate of current location Value, R1Distance value, R for current location to first object2Distance value, R for current location to the second target3For current location To the distance value of third target.
Same as the previously described embodiments, first object 1, the second target 2 and third target 3 in the present embodiment are also above-mentioned Three tires on undercarriage；And identification module also identifies current image using above-mentioned Faster-R-CNN algorithm In whether simultaneously include first object, the second target and third target.
Scheme described in the above embodiments through the invention directly finds unique identification feature on guidance destination, By artificial intelligence identification, ranging, the D coordinates value that this identification feature is calculated, driven by the coordinate value as guidance Reference value, facilitate promoted automatic Pilot accuracy, and method of the invention be not necessarily to guidance destination on installation appoint What additional device, ensure that its popularity used.
Than that described above, it is also necessary to which explanation is " one embodiment " spoken of in the present specification, " another implementation Example ", " embodiment " etc. refer to that specific features, structure or the feature of embodiment description is combined to be included in the application summary Property description at least one embodiment in.It is not centainly to refer to the same reality that statement of the same race, which occur, in multiple places in the description Apply example.Furthermore, it is understood that is advocated is knot when describing a specific features, structure or feature in conjunction with any embodiment Other embodiments are closed to realize that this feature, structure or feature are also fallen within the scope of the present invention.
Although reference be made herein to invention has been described for multiple explanatory embodiments of the invention, however, it is to be understood that Those skilled in the art can be designed that a lot of other modification and implementations, these modifications and implementations will fall in this Shen It please be within disclosed scope and spirit.More specifically, disclose in the application, drawings and claims in the range of, can With the building block and/or a variety of variations and modifications of layout progress to theme combination layout.In addition to building block and/or layout Outside the variations and modifications of progress, to those skilled in the art, other purposes also be will be apparent.

Claims (8)

1. a kind of method for target three-dimensional localization, it is characterised in that the method includes the following steps:
Step A, identification module is identified in present image by intelligent algorithm with the presence or absence of positioning target, the positioning mesh Mark includes first object, the second target and third target；
Step B, after identification module recognizes target, measurement current location respectively with first object, the second target and third mesh The distance between mark；
Step C, the distance that locating module is surveyed according to identification module is calculate by the following formula out first object, the second target and the The D coordinates value (x, y, z) of three targets:
In above formula, X1For the X axis coordinate value of first object, y1For the Y axis coordinate value of first object, z1For the Z axis of first object Coordinate value, X2For the X axis coordinate value of the second target, y2For the Y axis coordinate value of the second target, z2For the Z axis coordinate of the second target Value, X3For the X axis coordinate value of third target, y3For the Y axis coordinate value of third target, z3For the Z axis coordinate value of third target, X For the X axis coordinate value of current location, y is the Y axis coordinate value of current location, and z is the Z axis coordinate value of current location, R1It is current Distance value, R of the position to first object2Distance value, R for current location to the second target3For current location to third target Distance value.
2. the method according to claim 1 for target three-dimensional localization, it is characterised in that: the first object, second Target and third target are three tires on undercarriage.
3. the method according to claim 1 or 2 for target three-dimensional localization, it is characterised in that: identified in the step B Module by laser ranging or binocular camera measure respectively current location respectively with first object, the second target and third target The distance between.
4. the method according to claim 3 for target three-dimensional localization, it is characterised in that: pass through in the step A Whether it includes first object, the second target and third target that Faster-R-CNN algorithm identifies in present image simultaneously；It is described to work as Preceding image is acquired by video camera or camera.
5. a kind of system for target three-dimensional localization, it is characterised in that the system includes identification module and locating module, The identification module accesses locating module, in which:
Identification module is used to identify by intelligent algorithm in present image with the presence or absence of positioning target, the positioning target Including first object, the second target and third target；After identification module recognizes target, measurement current location is respectively with first The distance between target, the second target and third target；
Locating module is used for the distance surveyed according to identification module, is calculate by the following formula out first object, the second target and third The D coordinates value (x, y, z) of target:
In above formula, X1For the X axis coordinate value of first object, y1For the Y axis coordinate value of first object, z1For the Z axis of first object Coordinate value, X2For the X axis coordinate value of the second target, y2For the Y axis coordinate value of the second target, z2For the Z axis coordinate of the second target Value, X3For the X axis coordinate value of third target, y3For the Y axis coordinate value of third target, z3For the Z axis coordinate value of third target, X For the X axis coordinate value of current location, y is the Y axis coordinate value of current location, and z is the Z axis coordinate value of current location, R1It is current Distance value, R of the position to first object2Distance value, R for current location to the second target3For current location to third target Distance value.
6. the system according to claim 5 for target three-dimensional localization, it is characterised in that: the first object, second Target and third target are three tires on undercarriage.
7. the system according to claim 5 or 6 for target three-dimensional localization, it is characterised in that: the identification module is logical It crosses laser ranging or binocular camera and measures current location respectively respectively between first object, the second target and third target Distance.
8. the method according to claim 7 for target three-dimensional localization, it is characterised in that: the identification module passes through Whether it includes first object, the second target and third target that Faster-R-CNN algorithm identifies in present image simultaneously；The knowledge Other module also accesses video camera or camera, for acquiring present image by the video camera or camera and being transmitted to the knowledge Other module.
CN201910345409.XA 2019-01-22 2019-04-26 Method and system for target three-dimensional localization Pending CN109978953A (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910057846 2019-01-22
CN2019100578461 2019-01-22

Publications (1)

Publication Number Publication Date
CN109978953A true CN109978953A (en) 2019-07-05

Family

ID=67086548

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910345409.XA Pending CN109978953A (en) 2019-01-22 2019-04-26 Method and system for target three-dimensional localization

Country Status (1)

Country Link
CN (1) CN109978953A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113052898A (en) * 2021-04-08 2021-06-29 四川大学华西医院 Point cloud and strong-reflection target real-time positioning method based on active binocular camera
CN114942026A (en) * 2022-06-01 2022-08-26 四川大学 Multimode three-dimensional image navigation system based on intelligent data

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1685199A (en) * 2002-11-13 2005-10-19 株式会社拓普康 Surveying instrument and electronic storage medium
CN105572682A (en) * 2015-12-17 2016-05-11 华中科技大学 Underground space distance measuring method and device
CN108986161A (en) * 2018-06-19 2018-12-11 亮风台（上海）信息科技有限公司 A kind of three dimensional space coordinate estimation method, device, terminal and storage medium

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1685199A (en) * 2002-11-13 2005-10-19 株式会社拓普康 Surveying instrument and electronic storage medium
CN105572682A (en) * 2015-12-17 2016-05-11 华中科技大学 Underground space distance measuring method and device
CN108986161A (en) * 2018-06-19 2018-12-11 亮风台（上海）信息科技有限公司 A kind of three dimensional space coordinate estimation method, device, terminal and storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
朱金成: "自主跟随机器人系统研究与实现", 《中国优秀硕士学位论文全文数据库信息科技辑》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113052898A (en) * 2021-04-08 2021-06-29 四川大学华西医院 Point cloud and strong-reflection target real-time positioning method based on active binocular camera
CN113052898B (en) * 2021-04-08 2022-07-12 四川大学华西医院 Point cloud and strong-reflection target real-time positioning method based on active binocular camera
CN114942026A (en) * 2022-06-01 2022-08-26 四川大学 Multimode three-dimensional image navigation system based on intelligent data

Similar Documents

Publication Publication Date Title
CN109931939B (en) Vehicle positioning method, device, equipment and computer readable storage medium
US9587948B2 (en) Method for determining the absolute position of a mobile unit, and mobile unit
EP3650814B1 (en) Vision augmented navigation
JP2022166143A (en) System and method for determining navigation parameter
CN110361027A (en) Robot path planning method based on single line laser radar Yu binocular camera data fusion
CN108406731A (en) A kind of positioning device, method and robot based on deep vision
US20190005667A1 (en) Ground Surface Estimation
CN110462343A (en) The automated graphics for vehicle based on map mark
WO2019173547A1 (en) Odometry system and method for tracking traffic lights
CN109949594A (en) Real-time traffic light recognition method
CN109154506A (en) For determining the method and system of the global position of the first terrestrial reference
WO2021006441A1 (en) Road sign information collection method using mobile mapping system
CN109541535A (en) A method of AGV indoor positioning and navigation based on UWB and vision SLAM
CN108594244B (en) Obstacle recognition transfer learning method based on stereoscopic vision and laser radar
CN113710988A (en) Method for detecting the functional capability of an environmental sensor, control unit and vehicle
CN111912416A (en) Method, device and equipment for positioning equipment
CN109444916A (en) The unmanned travelable area determining device of one kind and method
JP2018077162A (en) Vehicle position detection device, vehicle position detection method and computer program for vehicle position detection
WO2023072055A1 (en) Point cloud data processing method and system
CN109978953A (en) Method and system for target three-dimensional localization
JP2017181476A (en) Vehicle location detection device, vehicle location detection method and vehicle location detection-purpose computer program
CN111353453A (en) Obstacle detection method and apparatus for vehicle
CN110864692A (en) Pose determination method of heading machine
Laflamme et al. Driving datasets literature review
CN106225779A (en) Development machine alignment systems based on three laser labelling dot image and localization method

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

Application publication date: 20190705

RJ01 Rejection of invention patent application after publication