CN109961476A - The localization method of the underground parking of view-based access control model - Google Patents
The localization method of the underground parking of view-based access control model Download PDFInfo
- Publication number
- CN109961476A CN109961476A CN201711418435.8A CN201711418435A CN109961476A CN 109961476 A CN109961476 A CN 109961476A CN 201711418435 A CN201711418435 A CN 201711418435A CN 109961476 A CN109961476 A CN 109961476A
- Authority
- CN
- China
- Prior art keywords
- point
- matrix
- access control
- control model
- based access
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
-
- 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
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
- G06T2207/30264—Parking
Abstract
The localization method of the underground parking of view-based access control model, comprising: S1 acquires parking lot environmental information；S2 carries out feature extraction and tracking to parking lot environmental information；S3 carries out characteristic matching with the information after tracking to extracting；S4 chooses key frame；S5 calculates the pose of vehicle in parking lot；S6 calculates the global pose of vehicle under world coordinate system.Underground parking information abundant can be obtained by camera, not by the interference of the factors such as external signal；In addition, monocular cam has, structure is simple, movement is flexible, is easy to demarcate, is at low cost, being easy many advantages, such as buying and installation.
Description
Technical field
The invention belongs to detection technique field of stopping, specifically a kind of positioning side of the underground parking of view-based access control model
Method.
Background technique
In recent years, with the continuous progress of science and technology and the improvement of people's living standards, more and more people select purchase private
Family's vehicle.With the increase of automobile, the requirement to parking lot is also higher and higher, even more than one layer of the parking lot especially having.
In order to avoid car owner gets lost during finding vehicle, need to know accurate location of the automobile in parking lot.At present
Localization method mainly have: 1) based on the location technology of GPS；2) location technology based on bluetooth 4.0；3) fixed based on wireless network
Position technology；4) based on the ranging localization technology of the sensors such as infrared, ultrasonic laser.
But the above location technology is applied under the environment of underground parking there are the limitation of its own,
1) based on the location technology of GPS；The pinpoint key of GPS system is that distance between satellite and receiver
Accurate calculating, according to fixed mode: distance=speed × time, the time determine after, speed press electromagnetic wave propagation speed
It is fixed.It is well known that the spread speed of electromagnetic wave in a vacuum is quickly, but atmosphere is not vacuum state, and signal will be ionized
Layer and tropospheric heavy interference.And GPS system can only carry out average computation to this, therefore exist certainly in certain specific regions
Error is embodied in the following aspects: the precision of GPS system is meter level, and error is larger, and precision is not high；Signal is weak.One
A little remote places or viaduct, high building nearby can all fail.3 are easy to be influenced by weather, at cloudy day, rainy day, search less than star.
Especially indoors and in the environment of underground parking, entirely without GPS signal, positioning failure.
2) based on the location technology of bluetooth.Bluetooth technology is positioned by measuring signal intensity.This is a kind of short distance
The Radio Transmission Technology of low-power consumption installs bluetooth local area network access point appropriate indoors, network configuration at based on multi-user
Basic network connection mode, and guaranteeing bluetooth local area network access point always is this main equipment, so that it may obtain user's
Location information.Bluetooth technology is mainly used in small range positioning.Bluetooth indoor positioning technologies biggest advantage be equipment volume it is small,
It is easily integrated into PDA, PC and mobile phone, therefore is easy to popularize.Theoretically, Bluetooth function shifting is integrated with for holding
The user of dynamic terminal device, as long as the Bluetooth function of equipment is opened, bluetooth indoor locating system can carry out position to it and sentence
It is disconnected.It is easy discovering device when making indoor short distance positioning using the technology and signal transmission is not influenced by sighting distance.Its deficiency exists
It is more expensive in the price of bluetooth devices and equipment, and for complicated space environment, the stability of Bluetooth system is slightly worse, by
Noise signal interference is big.Especially in 100*100 meters of planar range, the precision of bluetooth wireless location is at 5 meters or so, error
It is larger.
3) it is based on wireless network location technology；Positioning, monitor and chase after on a large scale for complexity may be implemented in Wireless LAN
Track task, and network node self poisoning is basis and the premise of most of applications.Current popular Wi-Fi positioning is nothing
A kind of location solution of the IEEE802.11 of line local area network series standard.But the system is passed using experience test and signal
The mode that model combines is broadcast, needs to carry out base station installation, and is needed using identical bottom wireless network structure, and price
Valuableness, there is presently no a large amount of universal.
4) it is based on infrared confirming orientation technology；The principle of infrared confirming orientation technology positioning is: infrared ray transmitting is modulated infrared
Ray is positioned by optical sensor reception.Although infrared ray has relatively high positioning accuracy, due to light
It cannot pass through barrier, so that infrared-ray is only capable of line-of-sight propagation.Straight line sighting distance and this shorter two big major defect of transmission range
Make the poor effect of its indoor positioning.It cannot be worked normally when mark is placed in pocket or has wall and other are blocked,
It needs that receiving antenna is installed in each space, cost is higher.Therefore, infrared ray is only suitable for short distance propagation, and is easy glimmering
Light interference in light lamp or room, there is limitation in accurate positioning.
5) ultrasonic wave location technology；Ultrasonic distance measurement mainly uses reflective telemetry, true by triangulation location scheduling algorithm
The position of earnest body, i.e. transmitting ultrasonic wave simultaneously receive the echo generated by measured object, are counted according to the time difference of echo and transmitted wave
Calculate testing distance.Some then uses unidirectional telemetry, and ultrasonic positioning system can be by several transponders and a main ranging
Device composition, main range finder are placed on testee, to the transponder transmitting that position is fixed under the action of microcomputer command signal
The radio signal of same frequency, transponder emit ultrasonic signal to main range finder simultaneously after receiving radio signal, obtain
The distance between main range finder and each transponder.There are 3 or 3 or more transponders not on the same line to make when simultaneously
When response, the position under the two-dimensional coordinate system where testee can be determined according to relevant calculation.Ultrasonic wave positioning is whole
Positioning accuracy is higher, and structure is simple, but ultrasonic wave is influenced very greatly by multipath effect and non-line-of-sight propagation, while needing a large amount of
Bottom hardware facility investment, cost are too high.
In order to avoid car owner gets lost during finding vehicle, in the underground parking environment of no GPS signal,
Need to know accurate location of the automobile in parking lot, therefore the location information for rapidly and accurately obtaining vehicle itself also becomes people
One of the problem of becoming more concerned with.
Summary of the invention
For disadvantages mentioned above of the existing technology and deficiency, the present invention provides a kind of underground parkings of view-based access control model
Localization method, underground parking information abundant can be obtained by camera, not by the interference of the factors such as external signal；Separately
Outside, monocular cam has many advantages, such as structure is simple, it is flexible to move, is easy to demarcate, is at low cost, being easy buying and install.
To achieve the above object, the present invention provides a kind of localization method of the underground parking of view-based access control model, comprising:
S1 acquires parking lot environmental information；
S2 carries out feature extraction and tracking to parking lot environmental information；
S3 carries out characteristic matching with the information after tracking to extracting；
S4 chooses key frame；
S5 calculates the pose of vehicle in parking lot；
S6 calculates the global pose of vehicle under world coordinate system.
Further, in step S1 method particularly includes: set up a monocular cam in front of the car, make camera optical axis
With vehicle body parallel.
Further, feature extraction is carried out to parking lot environmental information in step S2 method particularly includes:
S21: when obtaining image sequence, unified size change over is carried out, image sequence is sampled；
S22: using K grades of image pyramids, extracts FAST angle point,
S23: being divided into grid for every layer of pyramid, at least extracts L angle point in every lattice；
S24: if angle point number < L, improves threshold value, extraction is re-started；
S25: according to the FAST angle point extracted, direction and ORB Feature Descriptor are calculated using BRIEF algorithm.
Further, direction is calculated using BRIEF algorithm specifically:
Step 1. is round O using key point P as the center of circle, by radius of d；
Step 2. a certain mode in circle O chooses N number of point pair；
Step 3. defining operation T
Wherein, IAIndicate the gray value of A, IBIndicate the gray value of B；
Obtained result is combined by step 4. respectively to the point chosen to T operation is carried out.
Further, characteristic matching is carried out to the information after extracting and tracking in step S3 specifically:
S31. point in initializing randomly selects N to matching double points in given matching double points；
S32. fundamental matrix F is calculated by interior point；
S33. to matching double points remaining in matching double points, them are calculated at a distance from fundamental matrix, if result is small
Mr. Yu's threshold value then determines that it is asymmetric match point, rejects to it；
S34. previous step is repeated, until obtaining the final match point that arest neighbors match point is this feature.
Further, key frame is chosen in step S4 specifically: when interior points are greater than certain amount, determine that the frame is
Key frame.
As further, the pose of vehicle in parking lot is calculated in step S5 specifically:
All characteristic points are normalized first, and basis matrix and essential matrix are then solved according to key frame respectively；
Basis matrix are as follows:
X'Fx=0
Wherein,It is any pair of match point of two images；When giving enough match points, with the formula
To calculate unknown basis matrix F；
Essential matrix after normalization are as follows:
E=t × R=[t]x·R
Wherein E indicates that essential matrix, t indicate translation vector, [t]xIndicate the antisymmetric matrix of t, spin matrix R.
As further, the score of essential matrix is indicated with SH, SF indicates the score of basis matrix, sentences according to following
Cover half type determined, if:
Select that essential matrix acquires if ratio is greater than 0.45 as a result, if ratio is less than or equal to 0.45 selection basis
The result that Matrix Calculating obtains.
As further, the global pose of vehicle under S6 calculating world coordinate system specifically:
According to matched three-dimensional point to, basis matrix and essential matrix, the point in the corresponding three-dimensional world of characteristic point is solved,
The global pose of vehicle i.e. under world coordinate system；
The kinematic parameter of obtained each frame is accumulated, and obtains the global pose of vehicle movement under world coordinate system, i.e.,
The camera position of real time position and corner information in parking lot, note n moment is denoted as Ck, the camera position at k-1 moment is denoted as
Ck-1, wherein Ck=Ck-1Tk,k-1, motion profile of the vehicle in underground parking is rebuild at this time, and waits next frame image defeated
Enter, then starts the cycle over repetition step from step 1.
The present invention due to using the technology described above, can obtain following technical effect: obtain image sequence when
It waits, carries out unified size change over, image sequence is unified into size while carrying out gray proces, reduce calculation amount, improve speed
Degree；Using image pyramid, FAST angle point is extracted, wherein pyramid is divided into grid, single mapping distribution is ensure that, makes vehicle
Characteristic point is able to detect that having vibration during the motion, improves the validity of feature extraction.
Underground parking information abundant can be obtained by camera, not by the interference of the factors such as external signal；In addition,
Monocular cam has many advantages, such as structure is simple, it is flexible to move, is easy to demarcate, is at low cost, being easy buying and install.
Detailed description of the invention
The present invention shares 1 width of attached drawing:
Fig. 1 is the localization method flow chart of the underground parking of view-based access control model；
Specific embodiment
Below with reference to the embodiments and with reference to the accompanying drawing technical scheme of the present invention will be further explained in detail.
Present embodiments provide a kind of localization method of the underground parking of view-based access control model, comprising:
S1 acquires parking lot environmental information: setting up a monocular cam in the positive front end of vehicle, makes camera optical axis and vehicle
Body is parallel；Therefore vehicle, can be with the environmental information of very big viewing angles to parking lot in the traveling process of underground parking.
Meanwhile it is 30fps that sampling frame per second, which is arranged, remembers that in the image at the n-th moment be in, the image at n-1 moment is in-1；
S2 carries out feature extraction and tracking to parking lot environmental information, specifically:
S21: using no for normal video when no matter acquiring parking lot video, when obtaining image sequence,
Unified size change over is carried out, image sequence is sampled；It is unified for 640*480 size while carrying out gray proces, in this way,
It in the case where guaranteeing precision, can significantly reduce ORB extracted amount, improve system acquisition speed；
S22: using 8 grades of image pyramids, extracts FAST angle point, by many experiments discovery the case where scale is 1.25
Under, extraction effect is best, therefore the scale that uses of the application is 1.25；
S23: in order to ensure single mapping distribution, every layer of pyramid is divided into grid, obtains every layer of grid by many experiments
Optimal threshold is 50, at least extracts 5 angle points in every lattice；
S24: if angle point number < 5, improve threshold value, extraction is re-started；
S25: according to the FAST angle point extracted, direction and ORB Feature Descriptor are calculated using BRIEF algorithm；
S3 carries out characteristic matching with the information after tracking to extracting, specifically:
S31. point in initializing randomly selects 4 pairs of matching double points in given matching double points；
S32. fundamental matrix F is calculated by interior point；
S33. to matching double points remaining in matching double points, them are calculated at a distance from fundamental matrix, if result is small
Mr. Yu's threshold value sets the threshold value as 0.7 herein, then determines that it is asymmetric match point, reject to it；
S34. previous step is repeated, until obtaining the final match point that arest neighbors match point is this feature；
S4 chooses key frame；In order to reduce subsequent calculation amount, the calculation processing speed of positioning system is improved, needs to pick
Except invalid frame, subsequent image processing, the criterion of key frame only are carried out to representative key frame are as follows: when interior points are big
When certain amount, when such as > 70%, determine that the frame is key frame；
S5 calculates the pose of vehicle in parking lot:
All characteristic points are normalized first, and basis matrix and essential matrix are then solved according to key frame respectively；
Basis matrix are as follows:
X'Fx=0
Wherein,It is any pair of match point of two images；When giving enough match points, such as larger than etc.
In 7 pairs, unknown basis matrix F is calculated with the formula；
Optionally, x=(x, y, 1)TWith x'=(x', y', 1)T, then each group of match point provides the unknown element about F
One linear equation, coefficient can be indicated easily with the coordinate of known point x and x'.Specifically, correspond to a pair
Point x=(x, y, 1)TWith x'=(x', y', 1)TEquation be xx'f11+yx'f12+x'f13+xy'f21+yy'f22+y'f23+xf31+
yf32+f33=0 indicates to be made of the element of F with vector f, and by 9 n dimensional vector ns of row sequencing arrangement, then can use the formula
It indicates.
(xx', yx', x', xy', yy', y', x, y, 1) f=0
From the matched geometry of n group point, we can obtain following system of linear equations,
Thus system of linear equations is known, if the order of matrix A is 8, existence and unique solution, the basis matrix that can guarantee is kept
It is constant.
Essential matrix is the special shape of the fundamental matrix under normalized image coordinate, therefore, the essential square after normalization
Battle array are as follows:
E=t × R=[t]x·R
Wherein E indicates that essential matrix, t indicate translation vector, [t]xIndicate the antisymmetric matrix of t, spin matrix R.
Optionally, the solution of essential matrix can pass through following equation
It acquires.WhereinWithIt respectively indicates in adjacent two field pictures, the homogeneous seat of same group of characteristic point
Mark.
The score of essential matrix is indicated with SH, SF indicates the score of basis matrix, is determined according to following decision model,
If:
Select that essential matrix acquires if ratio is greater than 0.45 as a result, if ratio is less than or equal to 0.45 selection basis
The result that Matrix Calculating obtains.
Embodiment 2
As the supplement to embodiment 1, the above method further include: S6 calculates the global pose of vehicle under world coordinate system
Are as follows: according to matched three-dimensional point to, basis matrix and essential matrix, solve the point in the corresponding three-dimensional world of characteristic point, i.e. generation
The global pose of vehicle under boundary's coordinate system；
If U=W (U, V, 1)TIndicate that the point under homogeneous coordinate system, W are a scale factor.
If X=(x, y, z, 1)TThat indicate is the three-dimensional point in the corresponding world of U point, Pi TIt is transformation matrix [R1,2|t1,2]
The i-th row element.
Four systems of linear equations about point X are obtained, representation is AX=0, and wherein A is the matrix of 4*4, is then carried out
Singular value decomposition calculates four groups of solutions for acquiring X: (R1, t1), (R1, t2), (R2, t1), (R2, t2).
Tn and Rn constitutes Tn, Rn-1.
WhereinWhat is indicated is the transformation matrix at n moment and n-1 moment monocular cam position.
The kinematic parameter of obtained each frame is accumulated, the global pose of vehicle movement under world coordinate system is obtained,
I.e. in the real time position and corner information in parking lot, remember that the monocular cam position at n moment is Ck, the monocular camera shooting at k-1 moment
Head position is Ck-1, wherein Ck=Ck-1Tk,k-1, motion profile of the vehicle in underground parking is rebuild at this time, and waits next frame
Image input, then repetition step is started the cycle over from step S1.
Embodiment 3
As the supplement to embodiment 1 or 2, step S25: calculating direction using BRIEF algorithm, specifically:
Step 1. is round O using key point P as the center of circle, by radius of d；
Step 2. a certain mode in circle O chooses N number of point pair；Here for convenience of description, N=4, and the application is in parking lot
N takes 512 in the application of positioning；Assuming that 4 points currently chosen are respectively labeled as: P1(A,B)、P2(A,B)、P3(A,B)、P4(A,
B)；
Step 3. defining operation T
Wherein, IAIndicate the gray value of A, IBIndicate the gray value of B；
Obtained result is combined by step 4. respectively to the point chosen to T operation is carried out.
If: T (P1(A, B))=1, T (P2(A, B))=0, T (P3(A, B))=1, T (P4(A, B))=1 final retouch
State son are as follows: 1011
Optionally, such as the description of obtained characteristic point A, B is as follows,
A:1010, B:1011
Our given thresholds are 75%, and when the similarity of description of A and B is greater than 75%, we judge that A, B are identical
Characteristic point, i.e. this 2 successful match.
The application is method based on computer vision, and Lai Shixian is determined in underground parking without the environment of GPS signal
Position, can specifically be divided into following three parts: be when driving, to utilize the view in vehicle-mounted monocular camera acquisition parking lot first
Frequency information obtains videograph of the vehicle in the traveling process of parking lot；Acquire video information after carry out ORB feature extraction with
Matching；Then key frame is obtained, essential matrix and basis matrix are then tracked and solved to key frame, to be rotated
The solution of matrix and translation vector carries out the estimation of vehicle attitude finally according to the parameter of accumulation, obtains under world coordinate system
Vehicle overall situation pose.
When obtaining image sequence, unified size change over is carried out, image sequence is unified for 640*480 size simultaneously
Gray proces are carried out, calculation amount is reduced, improve speed；Using 8 grades of image pyramids, FAST angle point is extracted, wherein by pyramid
It is divided into grid, ensure that single mapping distribution, so that vehicle is able to detect that characteristic point having vibration during the motion, improve
The validity of feature extraction.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
Anyone skilled in the art within the technical scope of the present disclosure, according to the technique and scheme of the present invention and its
Inventive concept is subject to equivalent substitution or change, should be covered by the protection scope of the present invention.
Claims (9)
1. the localization method of the underground parking of view-based access control model characterized by comprising
S1 acquires parking lot environmental information；
S2 carries out feature extraction and tracking to parking lot environmental information；
S3 carries out characteristic matching with the information after tracking to extracting；
S4 chooses key frame；
S5 calculates the pose of vehicle in parking lot；
S6 calculates the global pose of vehicle under world coordinate system.
2. the localization method of the underground parking of view-based access control model according to claim 1, which is characterized in that specific in step S1
Method are as follows: set up a monocular cam in front of the car, make camera optical axis and vehicle body parallel.
3. the localization method of the underground parking of view-based access control model according to claim 1, which is characterized in that stopping in step S2
Parking lot environmental information carries out feature extraction method particularly includes:
S21: when obtaining image sequence, unified size change over is carried out, image sequence is sampled；
S22: using K grades of image pyramids, extracts FAST angle point；
S23: every layer of pyramid is divided into grid, every lattice at least extract L angle point；
S24: if angle point number < L, improves threshold value, extraction is re-started；
S25: according to the FAST angle point extracted, direction and ORB Feature Descriptor are calculated using BRIEF algorithm.
4. the localization method of the underground parking of view-based access control model according to claim 3, which is characterized in that calculated using BRIEF
Method calculates direction specifically:
Step 1. is round O using key point P as the center of circle, by radius of d；
Step 2. a certain mode in circle O chooses N number of point pair；
Step 3. defining operation T
Wherein, IAIndicate the gray value of A, IBIndicate the gray value of B；
Obtained result is combined by step 4. respectively to the point chosen to T operation is carried out.
5. the localization method of the underground parking of view-based access control model according to claim 4, which is characterized in that mentioning in step S3
It takes and carries out characteristic matching with the information after tracking specifically: obtain the final matching of this feature using arest neighbors matching point methods
Point.
6. the localization method of the underground parking of view-based access control model according to claim 1, which is characterized in that chosen in step S4
Key frame specifically: when interior points are greater than certain amount, determine that the frame is key frame.
7. the localization method of the underground parking of view-based access control model according to claim 1, which is characterized in that calculated in step S5
The pose of vehicle in parking lot specifically:
All characteristic points are normalized first, and basis matrix and essential matrix are then solved according to key frame respectively；
Basis matrix are as follows:
X'Fx=0
Wherein,It is any pair of match point of two images；When giving enough match points, counted with the formula
Unknown basis matrix F；
Essential matrix after normalization are as follows:
E=t × R=[t]x·R
Wherein E indicates that essential matrix, t indicate translation vector, [t]xIndicate the antisymmetric matrix of t, spin matrix R.
8. the localization method of the underground parking of view-based access control model according to claim 7, which is characterized in that indicate essence with SH
The score of matrix, the score of SF expression basis matrix, is determined according to following decision model, if:
Select that essential matrix acquires if ratio is greater than 0.45 as a result, if ratio is less than or equal to 0.45 selection basis matrix
The result acquired.
9. the localization method of the underground parking of view-based access control model according to claim 1, which is characterized in that S6 calculates the world and sits
Mark is the global pose of lower vehicle specifically: according to matched three-dimensional point to, basis matrix and essential matrix, solves characteristic point pair
Point in the three-dimensional world answered, accumulates the pose parameter of each frame, obtains the overall situation of vehicle movement under world coordinate system
Pose, i.e. real time position and corner information in parking lot.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711418435.8A CN109961476A (en) | 2017-12-25 | 2017-12-25 | The localization method of the underground parking of view-based access control model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711418435.8A CN109961476A (en) | 2017-12-25 | 2017-12-25 | The localization method of the underground parking of view-based access control model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109961476A true CN109961476A (en) | 2019-07-02 |
Family
ID=67020810
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711418435.8A Pending CN109961476A (en) | 2017-12-25 | 2017-12-25 | The localization method of the underground parking of view-based access control model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109961476A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111854678A (en) * | 2020-07-17 | 2020-10-30 | 浙江工业大学 | Pose measurement method based on semantic segmentation and Kalman filtering under monocular vision |
| CN112289058A (en) * | 2020-09-30 | 2021-01-29 | 上海汽车电器总厂有限公司 | Parking space management system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101581569A (en) * | 2009-06-17 | 2009-11-18 | 北京信息科技大学 | Calibrating method of structural parameters of binocular visual sensing system |
| CN102261916A (en) * | 2011-07-05 | 2011-11-30 | 大连理工大学 | Vision-based lunar rover positioning method in sandy environment |
| CN102435188A (en) * | 2011-09-15 | 2012-05-02 | 南京航空航天大学 | Monocular vision/inertia autonomous navigation method for indoor environment |
| CN102999759A (en) * | 2012-11-07 | 2013-03-27 | 东南大学 | Light stream based vehicle motion state estimating method |
-
2017
- 2017-12-25 CN CN201711418435.8A patent/CN109961476A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101581569A (en) * | 2009-06-17 | 2009-11-18 | 北京信息科技大学 | Calibrating method of structural parameters of binocular visual sensing system |
| CN102261916A (en) * | 2011-07-05 | 2011-11-30 | 大连理工大学 | Vision-based lunar rover positioning method in sandy environment |
| CN102435188A (en) * | 2011-09-15 | 2012-05-02 | 南京航空航天大学 | Monocular vision/inertia autonomous navigation method for indoor environment |
| CN102999759A (en) * | 2012-11-07 | 2013-03-27 | 东南大学 | Light stream based vehicle motion state estimating method |
Non-Patent Citations (2)
| Title |
|---|
| RAUL MUR-ARTAL: "ORB-SLAM: a versatile and accurate monocular SLAM system", 《IEEE TRANSACTIONS ON ROBOTICS》 * |
| 朱永丰: "基于ORB特征的单目视觉定位算法研究", 《计算机科学》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111854678A (en) * | 2020-07-17 | 2020-10-30 | 浙江工业大学 | Pose measurement method based on semantic segmentation and Kalman filtering under monocular vision |
| CN111854678B (en) * | 2020-07-17 | 2022-02-15 | 浙江工业大学 | Pose measurement method based on semantic segmentation and Kalman filtering under monocular vision |
| CN112289058A (en) * | 2020-09-30 | 2021-01-29 | 上海汽车电器总厂有限公司 | Parking space management system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106793086B (en) | Indoor positioning method | |
| CN105547305B (en) | A kind of pose calculation method based on wireless location and laser map match | |
| CN110118549B (en) | Multi-source information fusion positioning method and device | |
| CN107339990B (en) | Multi-mode fusion positioning system and method | |
| CN105588566B (en) | A kind of indoor locating system merged based on bluetooth with MEMS and method | |
| CN104732518B (en) | A kind of PTAM improved methods based on intelligent robot terrain surface specifications | |
| CN105371870B (en) | A kind of in-orbit accuracy measurement method of star sensor based on star chart data | |
| CN108632761B (en) | Indoor positioning method based on particle filter algorithm | |
| CN103363988B (en) | A kind of method utilizing intelligent mobile phone sensor to realize the navigation of earth magnetism indoor positioning | |
| CN103561462B (en) | Indoor positioning system and method totally based on smart mobile terminal platform | |
| CN103649680B (en) | Sensor location for 3D scanning | |
| CN103119611B (en) | The method and apparatus of the location based on image | |
| Xiao et al. | Comparison and analysis of indoor wireless positioning techniques | |
| CN106908821A (en) | A kind of indoor and outdoor seamless positioning changing method | |
| CN106291517A (en) | The indoor cloud robot angle localization method optimized with visual information based on position | |
| CN103017762B (en) | The extraterrestrial target fast Acquisition localization method of ground photo-electric telescope | |
| CN103994765B (en) | Positioning method of inertial sensor | |
| CN109540144A (en) | A kind of indoor orientation method and device | |
| CN105203023A (en) | One-stop calibration method for arrangement parameters of vehicle-mounted three-dimensional laser scanning system | |
| CN108413966A (en) | Localization method based on a variety of sensing ranging technology indoor locating systems | |
| Zhao et al. | An RSSI gradient-based AP localization algorithm | |
| CN107917880A (en) | A kind of height of cloud base inversion method based on ground cloud atlas | |
| Xiao et al. | Abnormal behavior detection scheme of UAV using recurrent neural networks | |
| CN106304330A (en) | A kind of radio frequency tomography localization method alleviating background electromagnetic wave action | |
| CN106370160A (en) | Robot indoor positioning system and 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: 20190702 |
|
| RJ01 | Rejection of invention patent application after publication |