CN107015193B - A kind of binocular CCD vision mine movable object localization method and system - Google Patents
A kind of binocular CCD vision mine movable object localization method and system Download PDFInfo
- Publication number
- CN107015193B CN107015193B CN201710252567.1A CN201710252567A CN107015193B CN 107015193 B CN107015193 B CN 107015193B CN 201710252567 A CN201710252567 A CN 201710252567A CN 107015193 B CN107015193 B CN 107015193B
- Authority
- CN
- China
- Prior art keywords
- target
- mine
- binocular ccd
- base station
- mobile target
- 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
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/0009—Transmission of position information to remote stations
-
- 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
- G01C21/005—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching
Abstract
The invention discloses a kind of binocular CCD vision mine movable object localization method and systems, the system for realizing this method includes uphole equipment and downhole hardware, uphole equipment includes base station controller, location-server, Ethernet switch and monitor terminal, and downhole hardware includes mine intrinsic safety type base station, wireless radio frequency identification mark.Implement the method and step of the system are as follows: (1) position of mobile target is estimated using RSSI algorithm;(2) matching characteristic is carried out to the wireless radio frequency identification mark of mobile target and extracts training;(3) matching of ORB algorithm characteristics is carried out to the image after sampling;(4) stereo calibration is carried out to mobile target left images using binocular CCD visual sensor;(5) the inside and outside ginseng matrix for obtaining binocular CCD visual sensor demarcates the world coordinates of mobile target;(6) position correction is carried out to the world coordinates information of mobile target, obtains the accurate location information that mine moves down moving-target.The present invention solves the problems, such as that mine NLOS environment moves down the accurate positioning of moving-target, improves the reliability and robustness of system.
Description
Technical field
The present invention relates to wireless communication techniques and computer vision technique, more particularly to a kind of binocular CCD vision mine
Moving objects location method and system.
Background technique
Coal is the important basic energy resource in China, and the energy resource structure based on coal will not change within the long duration, with
National economy development and coal demand it is increasing, adjoint Safety of Coal Mine Production accident is also being increasing.Cause
This, the requirement of mine safety is also higher and higher, this accuracy for Moving objects locations such as mine operation personnel, operating equipments
Higher requirements are also raised with reliability.
Currently, had it is some based on the location technologies such as RFID, WIFI, Zigbee and triangle center coordination, TOA,
The location algorithms such as AOA are applied in terms of personnel in the pit's positioning, but above-mentioned location technology and localization method, in underground moving mesh
There are problems in mark positioning accuracy: on the one hand since NLOS environmental electromagnetic wave non-direct-view in underground is propagated and multi-path jamming, WIFI,
The real-time of the location technologies such as Zigbee, accuracy susceptible, and due to mine laneway radio magnetic wave transmission loss is big,
Signal decaying is more serious, and using the location algorithms such as triangle center coordination, TOA, AOA, there are large errors, cannot achieve mine
The accurate positioning of the mobile target of well.On the other hand, RFID location technology can only carry out the disengaging identification of underground moving target, can not
The two-dimensional localization of mobile target is realized, especially, when multiple mobile mesh occur simultaneously in underground entrance and the same place of working face
When mark, RFID identification will cause missing inspection or identify inaccurate problem, it is also difficult to is accurately positioned.
CCD vision positioning technology is widely used in industrial non-cpntact measurement, has positioning accuracy height, anti-interference ability
By force, the features such as target image can be obtained at a distance, moreover, with CCD visual sensor and its measuring technique industrial application at
The continuous popularization of fruit is also more and more paid attention in the application of underground coal mine Moving objects location.
Summary of the invention
The technical problems to be solved by the present invention are: in view of the above problems, proposing a kind of binocular CCD vision mine
Moving objects location method and system realize that mine movable target is real-time, is accurately positioned.
The technical scheme is that proposing a kind of binocular CCD vision mine movable object localization method and system, realize
The system of this method includes uphole equipment and downhole hardware, and uphole equipment includes base station controller, Ethernet switch, positioning clothes
Business device and monitor terminal, downhole hardware include mine intrinsic safety type base station, wireless radio frequency identification mark.Uphole equipment passes through optical link
It is communicated with the downhole hardware.
Mine intrinsic safety type base station has the wireless interfaces such as LTE, WIFI, UWB, for wirelessly penetrating for underground moving target
Frequency identifies;
The mine intrinsic safety type embedded in base station binocular CCD visual sensor, the left and right for moving down moving-target for production wells are vertical
Body image information, and ORB characteristic matching and stereo calibration are carried out to it;
Mine intrinsic safety type base station passes through LTE wireless network and optical link for the position of binocular CCD vision sensor calibration
Confidence breath is sent to the location-server on well;
The wireless radio frequency identification mark is mounted in underground moving target, for moving as distinguishing mark
The RSSI radio frequency identification of target, feature extracting and matching;
The binocular CCD vision mine movable object localization method and system, implementation step include:
Step 1, mine intrinsic safety type base station are by receiving the radio frequency identification being mounted in mobile target
Label signal estimates the position of mobile target using RSSI algorithm.
Step 2, the image information that the decision threshold Thr for defining binocular CCD visual identity is binocular CCD visual sensor are adopted
Sample maximum distance.
Step 3 is less than or equal to decision threshold to mobile distance of the target away from the base station when mine intrinsic safety type base station monitors, i.e.,
dRssiWhen≤Thr, binocular CCD visual sensor carries out image information samples with apart from calculating to mobile target, otherwise, returns to step
Rapid 1 re-evaluates the position of mobile target.
Step 4, binocular CCD visual sensor carry out feature extraction training to wireless radio frequency identification mark using ORB algorithm,
And it is matched according to extraction feature and the collected mobile target identification markers feature of binocular CCD visual sensor.
Step 5, when mobile target signature successful match, binocular CCD visual sensor controls mobile target image
Stereo calibration obtains the inside and outside parameter matrix and parallax mean value of binocular CCD visual sensor, and uses binocular CCD visual sensing
Device fusion calculation goes out the world coordinates information of mobile target, and otherwise return step 1 re-evaluates the position of mobile target.
Step 6, according to mine intrinsic safety type base station present position, school is carried out to the world coordinates information of mobile target
Just, the final position coordinate information of mobile target is obtained.
Mine intrinsic safety type base station moves down the received signal strength of moving-target using RSSI algorithm monitoring well and to underground
Moving objects location, according to propagation loss theoretical modelBinocular CCD vision is calculated to pass
Sensor analyzes distance Thr to the image information samples of mobile target, wherein Pr(dRssi) it is signal transmission distance in subsurface environment
For dRssiWhen signal strength, Pr(d0) it is that signal transmission distance is d in ideal space state0When signal strength, k is signal
Strength retrogression's coefficient,I=1 ..., NLoss, to survey NLossSecondary underground signal strength retrogression coefficient.
The binocular CCD vision mine movable object localization method and system are realized and carry out ORB spy to underground moving target
The step of sign matching, further comprises:
Step 1 detects characteristic point using ORB algorithm, detects underground moving target feature point using FAST operator.
Step 2 adds directional information to the underground moving target feature point detected, constitutes oFAST.
Step 3 adds directional information to the underground moving Corner detected using gray scale centroid method.ORB takes ash
Degree centroid method gives the angle point addition directional information detected.Rosin defines neighborhood square: mpq=∑X, yxpyqI (x, y), mass center are as follows:Characteristic point and mass center angle are defined as the direction of FAST characteristic point: θ=atan2 (m01, m10)。
Step 4 carries out feature point description to the underground moving target detected using BRIEF description.
Step 5 adds rotational invariance to underground moving target feature point description detected, constitutes Steer
BRIEF not merely with simple, the quick advantage of calculating of BRIEF, and solves BRIEF itself and does not have rotational invariance
The characteristics of.S × S sized images are defined, algorithm extracts feature point description symbolWherein p
It (x) is smooth Image neighborhood P later at x=(u, v)TThe gray value at place.For (xi, yi) n position pair, Steer BRIEF
In (xi, yi) at, for the feature set of any n position pairUtilize spin matrix RθRotation matching point, obtains
With directive feature set Sθ=RθS。
Step 6, the sub- Hamming distance of description for calculating two matching double points carry out characteristic matching judgement from Ham.After rotation
Binary system criterion sub- g is describedn(p, θ) :=fn(p)|(xi, yi)∈Sθ, two feature point description need to be only calculated in matching
Hamming distance from Ham, carry out characteristic matching judgement.
Step 7 searches for the minimum picture of the n correlation set using greedy algorithm from all possible block of pixels centering
Mobile target signature is retrieved and matched to plain block pair, adjudicates characteristic matching result.
The binocular CCD vision mine movable object localization method and system carry out stereo calibration to underground moving target
The step of further comprise:
After binocular CCD visual sensor matches the success of underground moving target by ORB algorithm characteristics, regarded using binocular CCD
Feel that sensor positions collected mobile target, image pixel coordinates systemWherein, (XW, YW, ZW) it is world coordinates, u, v are image slices
Plain coordinate system coordinate, the x of image physical coordinates system, y-axis are parallel to pixel coordinate system u, v axis, dx,dyIt is horizontal and vertical respectively
The pixel spacing in direction, ZCFor the optical axis of binocular CCD visual sensor, f is focal length;
Step 1, binocular CCD visual sensor carry out left and right stereo calibration to collected mobile target image, obtain double
The inside and outside parameter matrix of mesh CCD visual sensor.Pass through camera coordinates system and world coordinate system calibration estimation binocular CCD vision
The outer ginseng matrix of sensorWherein R is 3 × 3 spin matrixs, 0TIt is 1 × 3 translation matrix, by the binocular
CCD visual sensor inner parameter and mobile target image or so stereo calibration solve the internal reference square of binocular CCD visual sensor
Battle arrayWherein fx,fyFor effective focal length.
Step 2, the parallax mean value for obtaining binocular CCD visual sensor, seek view according to the matching double points that Stereo matching obtains
Poor di, i=1,2 ..., NVer, and seek parallax diMean value
Step 3, the world coordinates that mobile target is sought by image pixel coordinates system equation, according to inside and outside parameter matrix and double
Mesh CCD visual sensor parallax dVer, solve coordinate (X in the world coordinate system of mobile targetW, YW, ZW), wherein ZWFor space
Mobile target is positioned to the vertical range of binocular CCD visual sensor, as depth information.
Step 4, the world coordinate system coordinate that mobile target is corrected by base station coordinates, built-in binocular CCD visual sensor
Mine intrinsic safety type base station in world coordinate system coordinate (0,0,0), position coordinate (X in the world coordinate system of mobile targetW,
YW, ZW), and mine laneway physical location locating for base station is combined to carry out coordinate position correction to the world coordinates information of mobile target,
The final position information of mine movable target is obtained, realizes the accurate positioning to mobile target.
The beneficial effects of the present invention are:
The invention is based on RSSI and ORB algorithm, passes through the RSSI radio frequency identification and binocular CCD to underground moving target
Vision calibration blends, and provides a kind of binocular CCD vision mine movable object localization method and system.The present invention solves existing
It with the presence of location technology missing inspection and identifies the problems such as inaccurate and positioning accuracy is lower, especially overcomes under the NLOS environment of underground
Target occlusion lead to not to improve mobile target identification and precise positioning problem the real-time of underground moving target positioning with
Accuracy and system robustness.System accuracy is high, strong antijamming capability, is suitable for various complex environments and limits space
Mobile object real-time tracking and accurate positioning.
Detailed description of the invention
Fig. 1 is the system composition schematic diagram of a kind of binocular CCD vision mine movable object localization method and system.
Fig. 2 is the positioning flow figure of a kind of binocular CCD vision mine movable object localization method and system.
Fig. 3 is a kind of binocular CCD vision mine movable object localization method and the ORB algorithm flow chart that system is taken.
Fig. 4 is a kind of binocular CCD vision mine movable object localization method and the binocular CCD location algorithm stream that system is taken
Cheng Tu.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawing to tool of the invention
Body embodiment is described in detail.
Fig. 1 is the system composition of binocular CCD mine movable object localization method and system.Shown in referring to Fig.1, realize above-mentioned
The system of mine localization method specifically includes that wireless radio frequency identification mark (101), built-in binocular CCD intrinsic safety type locating base station
(102), Ethernet switch (103), base station controller (104), location-server (105), monitor terminal (106).
Wherein, wireless radio frequency identification mark (101) transmitting radio frequency signal can be identified by intrinsic safety type locating base station, be used for
Coarse localization, the identification for carrying out characteristic matching to the label extract training, are used for binocular CCD visual sensor ORB algorithm characteristics
Matching.Binocular CCD intrinsic safety type locating base station (102), intrinsic safety type locating base station are based on propagation loss theoretical model, are calculated using RSSI
Method identifies the wireless radio frequency identification mark in mobile target and obtains the radio-frequency information of mine movable target, and it is fixed to be built in intrinsic safety type
Position base station binocular CCD visual sensor acquire in a triggered move target image and carry out characteristic matching and stereo calibration into
And realize the accurate positioning to mobile target.Ethernet switch (103) is realized for the aggregation node of positioning system network
The interconnection of positioning system and terrestrial communication networks.Base station controller (104) is used for underground wireless network management, radio resource pipe
The management of reason and mine intrinsic safety base station controls and monitors and the Handoff Control of underground moving target etc..Location-server
(105), for handling, storing mine movable target real-time position information, Location based service is provided for underground moving target, it is real
The now positioning to underground moving target and monitoring function, and service is consulted and transferred to the historical information for providing monitor terminal.Prison
Control terminal (106), for receiving image and real-time positioning information from location-server, by image and real-time positioning information
Carry out visualization display.
Fig. 2 is the positioning flow figure of binocular CCD mine movable object localization method and system.Referring to shown in Fig. 2, binocular
The positioning flow step and function of CCD mine movable object localization method are described as follows: (1) measurement of signal attenuation coefficient and calibration
(201), logarithm-normality propagation loss theoretical model is taken in the mine intrinsic safety type base station of in-built CCD visual sensor The measurement of signal attenuation coefficient and calibration are carried out, since the attenuation coefficient of mine signal is wanted
It is apparently higher than environment on well, carries out actual measurement N in mineLossIt is secondary to averageAs mine signal attenuation coefficient;
(2) training (202) is extracted in wireless radio frequency identification mark feature identification, for being mounted on the radio frequency label of mine movable target
Identification carries out the identification of matching characteristic and extracts to train, and provides matching characteristic for the ORB algorithmic match of target;(3) RSSI algorithm
Coarse localization obtains mobile target and identifies wirelessly penetrating in mobile target using the RSSI algorithm of base station away from base station distance (203)
Frequency identification tag takes logarithm-normality propagation loss theoretical model to estimate roughly the rough position of mine movable target;
(4) it adjudicates mobile target and whether is less than or equal to binocular CCD sampled distance Thr (203) away from base station distance, by adjudicating mobile target
Whether it is less than or equal to binocular CCD sampled distance Thr away from base station distance, triggers binocular CCD when the condition is satisfied to mobile target figure
As acquisition, the position for taking RSSI to re-evaluate mobile target is otherwise returned to;(5) triggering binocular CCD adopts mobile target image
Collect (205), when base station collects mobile target range base station distance dRssiIt can be into less than or equal to threshold value binocular CCD visual sensor
Row image information samples analyze distance Thr, can trigger the mine intrinsic safety type base station of built-in binocular CCD visual sensor, by double
Mesh CCD visual sensor carries out the image information samples of mobile target;(6) ORB algorithm to mobile target signature matching whether at
Function (206), when mine intrinsic safety type base station monitors are less than or equal to decision threshold, i.e. d to mobile distance of the target away from the base stationRssi≤
When Thr, binocular CCD visual sensor carries out image information samples with apart from calculating to mobile target, otherwise returns and takes RSSI
Re-evaluate the position of mobile target;(7) mobile coordinates of targets calibration (207), obtains according to image pixel coordinates system coordinate formula
Inside and outside parameter matrix, parallax mean value d required for taking coordinate to demarcateVer, demarcate the world coordinates of mobile target;(8) base station corrects
(208) are completed in positioning, mine intrinsic safety type base station coordinate system coordinate in world coordinate system of built-in binocular CCD visual sensor
(0,0,0) positions coordinate (X in the world coordinate system of mobile targetW, YW, ZW), and combine mine laneway actual bit locating for base station
It sets and coordinate position correction is carried out to the world coordinates information of mine movable target, obtain the final position letter of mine movable target
Breath realizes the accurate positioning to mobile target.
Fig. 3 is binocular CCD mine movable object localization method and the ORB algorithm flow chart that system is taken.As shown in figure 3,
The ORB algorithm flow step that binocular CCD mine movable object localization method is taken includes: that (1) inputs mobile target image information
(301), using the collected mobile target image of binocular CCD visual sensor as the matched input of ORB algorithm characteristics;(2)
Characteristic point detection (302) are carried out using oFAST operator, takes and oFAST operator, benefit is formed to FAST operator addition rotational invariance
Characteristic point detection is carried out with oFAST operator, and gray scale centroid method is used to increase matching local invariant for the characteristic point detected
Property;(3) characteristic point is described (303) using Steer BRIEF, Steer BRIEF algorithm is simple using the calculating of BRIEF
Single, quick advantage carries out feature description, and solves BRIEF itself not having to the characteristic point that oFAST algorithm detects
The shortcomings that rotational invariance;(4) construction feature collection and binary system criterion describe sub (304), define S × S sized images, the region
It is interior to choose any n position feature to (xi, yi) constitutive characteristic collectionAnd in S × S sized images, building is calculated
Method extracts feature point description symbolSteer BRIEF is in (xi, yi) at, for any n
The feature set of a position pairUtilize spin matrix RθRotation matching point is obtained with directive feature set Sθ=
RθS, postrotational binary system criterion describe sub- gn(p, θ) :=fn(p)|(xi, yi)∈Sθ;(5) threshold value of characteristic matching is set
Haming distance (305) repeatedly obtains the threshold value Haming of same characteristic features images match using the training of same characteristic features images match
Distance is simultaneously averaged, to obtain the matched threshold value Haming distance of best features, the setting matched threshold value Haming of this feature away from
From mean value as feature to the judgement standard whether matched;(6) greedy retrieval meets characteristic matching block of pixels (306), uses greediness
Algorithm detection is less than or equal to the block of pixels of setting Haming distance threshold, retrieves the result of characteristic matching;(7) feature is exported
With search result (307), according to greedy algorithm retrieve characteristic matching as a result, judgement characteristic matching result.
Fig. 4 is binocular CCD mine movable object localization method and the binocular CCD location algorithm process that system is taken.Such as Fig. 4
Shown, the binocular CCD location algorithm process that above-mentioned mine localization method is taken specifically includes that (1) binocular CCD adopts mobile target
Collect left images (401), acquires left images by moving down moving-target to mine using binocular CCD visual sensor.(2) binocular
CCD carries out stereo calibration (402) to mobile target left images, by the inner parameter of binocular CCD visual sensor and double
Mesh CCD visual sensor carries out left and right stereo calibration to collected mobile target image, obtain camera center away from, focal length f,
Parallax di, inside and outside parameter matrix, if world coordinate system coordinate (XW, YW, ZW), camera coordinates system coordinate
Wherein, (XC, YC, ZC) it is camera coordinates system coordinate, R is 3 × 3 spin matrixs, 0TIt is 1 × 3 translation matrix, binocular can be passed through
CCD visual sensor carries out left and right stereo calibration to collected mobile target image, i.e., to camera coordinates system and world coordinates
System's calibration estimation is outer to join matrixImage physical coordinates systemWherein, ZCFor
Optical axis, x, y are image physical coordinates system coordinates, and f is focal length, image pixel coordinates systemU, v are image pixel coordinates system coordinate, image physical coordinates
It is x, y-axis is parallel to pixel coordinate system u, v axis, dx,dyIt is the pixel spacing in horizontal and vertical direction respectively, is regarded using binocular CCD
Feel that sensor internal parameter and binocular CCD visual sensor carry out left and right stereo calibration to collected mobile target image and ask
Solve internal reference matrixWherein fx,fyFor effective focal length;(3) binocular CCD is obtained
Parallax (403), seek parallax d according to the matching double points that binocular CCD visual sensor Stereo matching obtainsi, i=1,2 ...,
NVer, and seek parallax diMean value(4) image pixel coordinates system equation seeks the world coordinates of mobile target
(404), the world coordinates (X of mobile target is acquired according to image pixel coordinates formulaW, YW, ZW), wherein ZWIt is exactly space orientation
Target is moved to the vertical range of binocular CCD visual sensor, also referred to as depth information;(5) it is corrected and is moved by base station coordinates
The world coordinate system coordinate (405) of target, the mine intrinsic safety type base station of built-in binocular CCD visual sensor is in world coordinate system
Coordinate (0,0,0) positions coordinate (X in the world coordinate system of mobile targetW, YW, ZW), and combine mine laneway locating for base station real
Border position carries out coordinate position correction to the world coordinates information of mobile target, obtains the final position letter of mine movable target
Breath realizes the accurate positioning to mobile target.
Obviously, those skilled in the art should be understood that localization method and system involved by the present invention and above-described embodiment
Each composition function, in addition to being applied to underground coal mine environment as the positioning of mine movable target, after suitably integrated or improvement
Suitable for the mobile monitoring of the non-coal mines such as nonmetallic and metal, tracking and positioning and downhole intelligent working face equipment
It is accurately positioned.It is mobile not limit non-coal mine in addition to underground coal mine Moving objects location, intelligent work face by the present invention in this way
The fields of communication technology such as monitoring and equipment accurate positioning.
The above content is the further descriptions for combining specific preferred embodiment mode to be the present invention, cannot recognize
Determine a specific embodiment of the invention and be only limitted to this, for those of ordinary skill in the art to which the present invention belongs, not
Under the premise of being detached from mentality of designing of the present invention, several simple replacements and change can be also carried out, all shall be regarded as belonging to the present invention
Protection scope involved in the claims submitted.
Claims (6)
1. a kind of binocular CCD vision mine movable object localization method, the system for realizing this method includes uphole equipment and underground
Device, which is characterized in that
The uphole equipment of the system includes base station controller, location-server, Ethernet switch and monitor terminal;
The downhole hardware of the system includes mine intrinsic safety type base station, wireless radio frequency identification mark;
The uphole equipment of the system is communicated by optical link with the downhole hardware;
The mine intrinsic safety type base station of the system has LTE, WIFI, UWB wireless interface, for wirelessly penetrating for underground moving target
Frequency identifies;
The mine intrinsic safety type embedded in base station binocular CCD visual sensor of the system, the left and right of moving-target is moved down for production wells
Stereo image information, and ORB characteristic matching and stereo calibration are carried out to it;
The mine intrinsic safety type base station of the system passes through LTE wireless network and optical link for binocular CCD vision sensor calibration
Location information is sent to the location-server on well;
The wireless radio frequency identification mark of the system is mounted in underground moving target, for moving as distinguishing mark
RSSI radio frequency identification, feature extraction and the training of moving-target;It is further characterized in that
The binocular CCD vision mine movable object localization method, step include:
(1) mine intrinsic safety type base station is by receiving the wireless radio frequency identification mark being mounted in mobile target letter
Number, the position of mobile target is estimated using RSSI algorithm;
(2) the decision threshold Thr for defining the binocular CCD visual identity is the image information samples of binocular CCD visual sensor
Maximum distance;
(3) when the mine intrinsic safety type base station monitors are less than or equal to decision threshold, i.e. d to mobile distance of the target away from the base stationRssi
When≤Thr, the binocular CCD visual sensor carries out image information samples with apart from calculating to mobile target, otherwise, returns to step
Suddenly (1) re-evaluates the position of mobile target;
(4) the binocular CCD visual sensor carries out feature extraction training to wireless radio frequency identification mark using ORB algorithm, and
It is matched according to extraction feature and the collected mobile target identification markers feature of binocular CCD visual sensor;
(5) when mobile target signature successful match, it is three-dimensional that the binocular CCD visual sensor carries out left and right to mobile target image
Calibration is obtained the inside and outside parameter matrix and parallax mean value of binocular CCD visual sensor, and is melted using binocular CCD visual sensor
Total world coordinates information for calculating mobile target, otherwise return step (1) re-evaluates the position of mobile target;
(6) according to mine intrinsic safety type base station present position, the world coordinates information of mobile target is corrected, obtains and moves
The final position coordinate information of moving-target.
2. binocular CCD vision mine movable object localization method according to claim 1, which is characterized in that
Mine intrinsic safety type base station moves down the received signal strength of moving-target and to underground moving using RSSI algorithm monitoring well
Target positioning, according to propagation loss theoretical modelCalculate binocular CCD visual sensor pair
The image information samples of mobile target analyze distance Thr, wherein Pr(dRssi) it is that signal transmission distance is d in subsurface environmentRssi
When signal strength, Pr(d0) it is that signal transmission distance is d in ideal space state0When signal strength, k declines for signal strength
Subtract coefficient,To survey NLossSecondary underground signal strength retrogression coefficient.
3. binocular CCD vision mine movable object localization method according to claim 1, which is characterized in that moved to underground
The step of moving-target progress ORB characteristic matching, further comprises:
(1) characteristic point is detected using ORB algorithm, underground moving target feature point is detected using FAST operator;
(2) directional information is added to the underground moving target feature point detected, constitutes oFAST;
(3) directional information is added to the underground moving Corner detected using gray scale centroid method;
(4) feature point description is carried out to the underground moving target detected using BRIEF description;
(5) to the underground moving target feature point description son addition rotational invariance detected, Steer BRIEF is constituted;
(6) the sub- Hamming distance of description of two matching double points is calculated from Ham, carries out characteristic matching judgement;
(7) the minimum block of pixels pair of the n correlation set using greedy algorithm from all possible block of pixels centering search,
Mobile target signature is retrieved and matched, characteristic matching result is adjudicated.
4. binocular CCD vision mine movable object localization method according to claim 1, which is characterized in that moved to underground
The step of moving-target progress stereo calibration, further comprises:
(1) the binocular CCD visual sensor carries out left and right stereo calibration to collected mobile target image, obtains binocular
The inside and outside parameter matrix of CCD visual sensor;
(2) the parallax mean value for obtaining the binocular CCD visual sensor, seeks parallax d according to the matching double points that Stereo matching obtainsi,
I=1,2 ..., NVer, and seek parallax diMean value
(3) image pixel coordinates system equation seeks the world coordinates of mobile target, is passed according to inside and outside parameter matrix and binocular CCD vision
Sensor parallax mean value dVer, solve coordinate (X in the world coordinate system of mobile targetW,YW,ZW), wherein ZWIt is mobile for space orientation
Target is to the vertical range of binocular CCD visual sensor, as depth information;
(4) the world coordinate system coordinate of mobile target, the mine of the built-in binocular CCD visual sensor are corrected by base station coordinates
Well intrinsic safety type base station coordinate (0,0,0) in world coordinate system positions coordinate (X in the world coordinate system of mobile targetW,YW,
ZW), and mine laneway physical location locating for base station is combined to carry out coordinate position correction to the world coordinates information of mobile target, it obtains
The final position information of mine movable target is taken, realizes the accurate positioning to mobile target.
5. binocular CCD vision mine movable object localization method according to claim 1, which is characterized in that
Base station controller, for underground wireless network management, the management of wireless resource management and mine intrinsic safety type base station, control and
The Handoff Control of monitoring and underground moving target;
Location-server, for handling, storing mine movable target real-time position information, it is mobile fixed to provide for underground moving target
Position service, realizes the positioning to underground moving target and monitoring function;
Ethernet switch realizes the positioning system and ground communication net for the aggregation node of the positioning system network
The interconnection of network;
Monitor terminal believes image and in real time positioning for receiving image and real-time positioning information from location-server
Breath carries out visualization display.
6. binocular CCD vision mine movable object localization method according to claim 1, which is characterized in that mine intrinsic safety
Type base station and wireless radio frequency identification mark are mining intrinsic safety type explosion-protection equipment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710252567.1A CN107015193B (en) | 2017-04-18 | 2017-04-18 | A kind of binocular CCD vision mine movable object localization method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710252567.1A CN107015193B (en) | 2017-04-18 | 2017-04-18 | A kind of binocular CCD vision mine movable object localization method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107015193A CN107015193A (en) | 2017-08-04 |
CN107015193B true CN107015193B (en) | 2019-10-11 |
Family
ID=59446916
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710252567.1A Active CN107015193B (en) | 2017-04-18 | 2017-04-18 | A kind of binocular CCD vision mine movable object localization method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107015193B (en) |
Families Citing this family (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107566981B (en) * | 2017-08-30 | 2020-02-04 | 山东大学 | Indoor high-precision positioning method, device and system based on optimal path |
CN108174440A (en) * | 2017-12-06 | 2018-06-15 | 山西宏安翔科技股份有限公司 | A kind of mine multifunction wireless communication base station |
CN107977022B (en) * | 2017-12-28 | 2020-11-13 | 国网福建省电力有限公司 | Automatic slot type facility inspection device and method based on angular point detection |
CN110008951B (en) * | 2019-03-14 | 2020-12-15 | 深兰科技(上海)有限公司 | Target detection method and device |
CN110058587B (en) * | 2019-03-18 | 2022-09-13 | 西安科技大学 | Coal mine fully-mechanized mining face inspection unmanned vehicle based on SLAM technology and autonomous inspection method |
CN110136172B (en) * | 2019-05-21 | 2023-04-07 | 中国矿业大学 | Detection method for wearing of underground protective equipment of miners |
CN110487262A (en) * | 2019-08-06 | 2019-11-22 | Oppo广东移动通信有限公司 | Indoor orientation method and system based on augmented reality equipment |
CN110428419A (en) * | 2019-09-02 | 2019-11-08 | 中国矿业大学(北京) | Mine positioning system based on mobile image identification |
CN110415302A (en) * | 2019-09-02 | 2019-11-05 | 中国矿业大学(北京) | Mine positioning system based on image recognition |
JP2022545969A (en) * | 2019-09-02 | 2022-11-01 | アイキュー、ワークス、リミテッド | System and method for event recognition |
CN110619664B (en) * | 2019-09-17 | 2023-06-27 | 武汉理工大学 | Laser pattern-assisted camera distance posture calculation method and server |
CN111542115B (en) * | 2020-05-18 | 2022-03-08 | 南京荣飞科技股份有限公司 | 125 k-based simulated reference tag positioning system and method thereof |
CN111852456B (en) * | 2020-07-29 | 2023-04-07 | 中国矿业大学 | Robust UWB (ultra wide band) underground anchor rod drilling positioning method based on factor graph |
CN112819770B (en) * | 2021-01-26 | 2022-11-22 | 中国人民解放军陆军军医大学第一附属医院 | Iodine contrast agent allergy monitoring method and system |
CN113175915B (en) * | 2021-04-16 | 2022-09-06 | 中国矿业大学 | Passive low-cost self-positioning device and positioning method for underground rescue robot |
CN113066134A (en) * | 2021-04-23 | 2021-07-02 | 深圳市商汤科技有限公司 | Calibration method and device of visual sensor, electronic equipment and storage medium |
CN113129378A (en) * | 2021-04-28 | 2021-07-16 | 北京市商汤科技开发有限公司 | Positioning method, positioning device, electronic equipment and storage medium |
CN113030860A (en) * | 2021-05-27 | 2021-06-25 | 东智安通(北京)科技有限公司 | Positioning method, device, equipment and medium based on RFID positioning label |
CN113950144B (en) * | 2021-10-01 | 2024-03-22 | 南宁市安普康商贸有限公司 | Monitoring method, system, device and computer program product |
CN114353782B (en) * | 2022-01-11 | 2023-06-20 | 华北理工大学 | Baseline-RFMDR-based underground positioning method and underground positioning device |
CN117528420A (en) * | 2023-11-14 | 2024-02-06 | 滕州郭庄矿业有限责任公司 | Mine personnel position detection system based on Internet of things |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8392065B2 (en) * | 2008-09-11 | 2013-03-05 | Deere & Company | Leader-follower semi-autonomous vehicle with operator on side |
CN101790179B (en) * | 2010-02-12 | 2012-07-11 | 中国矿业大学(北京) | Mine mobile communication system |
CN102848388A (en) * | 2012-04-05 | 2013-01-02 | 上海大学 | Service robot locating and grabbing method based on multiple sensors |
KR101343818B1 (en) * | 2012-10-15 | 2013-12-20 | 우리로광통신주식회사 | Monitoring camera system and method |
CN204168290U (en) * | 2014-04-09 | 2015-02-18 | 中国矿业大学(北京) | A kind of multi-mode mine mobile communication system |
CN105160321A (en) * | 2015-09-05 | 2015-12-16 | 深圳市飞思未来云媒体科技有限公司 | Vision-and-wireless-positioning-based mobile terminal identity verification method |
CN106408601B (en) * | 2016-09-26 | 2018-12-14 | 成都通甲优博科技有限责任公司 | A kind of binocular fusion localization method and device based on GPS |
-
2017
- 2017-04-18 CN CN201710252567.1A patent/CN107015193B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN107015193A (en) | 2017-08-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107015193B (en) | A kind of binocular CCD vision mine movable object localization method and system | |
CN107437044B (en) | Mine moving target tracking and positioning method | |
CN107328406B (en) | Method and system for positioning mine moving target based on multi-source sensor | |
CN107846258A (en) | A kind of unmanned plane system of defense | |
CN109581345A (en) | Object detecting and tracking method and system based on millimetre-wave radar | |
CN110514225A (en) | The calibrating external parameters and precise positioning method of Multi-sensor Fusion under a kind of mine | |
CN103995250B (en) | Radio-frequency (RF) tag trajectory track method | |
CN205232319U (en) | Camera and monitored control system | |
CN109583383A (en) | A kind of unmanned plane life detection method and system based on Multiple Source Sensor | |
CN109407697A (en) | A kind of unmanned plane pursuit movement goal systems and method based on binocular distance measurement | |
CN103796304B (en) | One kind is based on virtual training collection and markovian underground coal mine localization method | |
CN110346758A (en) | Ultra wide band positioning and alarming method for power applied to substation inspection personnel | |
CN104700408B (en) | A kind of indoor single goal localization method based on camera network | |
CN102680941A (en) | Method and system for ultra wide band positioning in coal mine | |
CN106646474A (en) | Unstructured road accidented barrier detection apparatus | |
CN112929826B (en) | Indoor hybrid positioning method based on UWB triangulation positioning and fingerprint information | |
CN202600134U (en) | Underground ultra wide band location system of coal mine | |
Peng et al. | Deep convolutional neural network for passive RFID tag localization via joint RSSI and PDOA fingerprint features | |
CN103116159A (en) | Multi-mode self-positioning networking radar life detection method and device | |
CN113705376B (en) | Personnel positioning method and system based on RFID and camera | |
CN112085003A (en) | Automatic identification method and device for abnormal behaviors in public places and camera equipment | |
Feng et al. | Visual map construction using RGB-D sensors for image-based localization in indoor environments | |
CN113534047A (en) | Equipment positioning method and device | |
Deng et al. | WiFi positioning based on user orientation estimation and smartphone carrying position recognition | |
CN114222240A (en) | Multi-source fusion positioning method based on particle filtering |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |