CN109951687A - A kind of underground coal mine environment realizing of Robot Vision system - Google Patents
A kind of underground coal mine environment realizing of Robot Vision system Download PDFInfo
- Publication number
- CN109951687A CN109951687A CN201910218626.2A CN201910218626A CN109951687A CN 109951687 A CN109951687 A CN 109951687A CN 201910218626 A CN201910218626 A CN 201910218626A CN 109951687 A CN109951687 A CN 109951687A
- Authority
- CN
- China
- Prior art keywords
- image
- coal mine
- underground coal
- image acquisition
- mist
- 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.)
- Granted
Links
Landscapes
- Studio Devices (AREA)
Abstract
The invention discloses a kind of underground coal mine environment realizing of Robot Vision systems, comprising: several first Image Acquisition terminals, several second Image Acquisition terminals, data line, power supply line, display unit;Several first Image Acquisition terminals, which are uniformly distributed, to be fixed on along coal mine well port to coal mine well bottom zone, and the first Image Acquisition terminal is used for collection of coal mine underground predeterminable area ambient image data;First Image Acquisition terminal and the second Image Acquisition terminal pass through power supply line and connect with ground power supply;Image of the second Image Acquisition terminal for the default corresponding movable object in collection of coal mine underground, and mobile tracking shooting is carried out to movable object;First Image Acquisition terminal and the second Image Acquisition terminal are shown the data of acquisition by the display unit that data line is transmitted to ground;The technical effect of track up can be carried out while being capable of providing coal mine underground monitoring video image to the mobile target of underground coal mine.
Description
Technical field
The present invention relates to mine supervision fields, and in particular, to a kind of underground coal mine environment realizing of Robot Vision system.
Background technique
Coal mine is the mankind in the region of the mining area exploitation coal resources rich in coal, is generally divided into underground mine and outdoor coal
Mine.Although coal mining sets up many safety codes, coal mining accident but still occurs, and coal mining accident can generate great
Casualties and economic loss.
Therefore, colliery scene needs real-time perfoming monitoring to ensure safety of coal mines, the especially underground coal mine of environment complexity, coal
Mine down space is narrow, insufficient light, and underground coal mine environment realizing of Robot Vision has certain difficulty, in the prior art mainly
Underground coal mine environment machine vision is realized using underground coal mine video monitoring system, but underground coal mine video monitoring system is only
The image being capable of providing in certain area can not carry out track up, the mobile target of underground to the mobile target of underground
Whether be normally on well monitoring personnel more to vision realize content.
Summary of the invention
The present invention provides a kind of underground coal mine environment realizing of Robot Vision systems, solve existing underground coal mine video
Monitoring system is merely able to provide the image in certain area, and the technology of track up can not be carried out to the mobile target of underground
Problem realizes system design rationally, can be mobile to underground coal mine while being capable of providing coal mine underground monitoring video image
The technical effect of target progress track up.
For achieving the above object, described this application provides a kind of underground coal mine environment realizing of Robot Vision system
System includes:
Several first Image Acquisition terminals, several second Image Acquisition terminals, data line, power supply line, display unit;
Several first Image Acquisition terminals, which are uniformly distributed, to be fixed on along coal mine well port to coal mine well bottom zone, and the first image is adopted
Collect terminal and is used for collection of coal mine underground predeterminable area ambient image data;First Image Acquisition terminal and the second Image Acquisition terminal
It is connect by power supply line with ground power supply;Second Image Acquisition terminal is for the default corresponding movable object in collection of coal mine underground
Image, and to movable object carry out mobile tracking shooting;First Image Acquisition terminal and the second Image Acquisition terminal will
The data of acquisition are shown by the display unit that data line is transmitted to ground.
Wherein, to be that several first Image Acquisition terminals are uniformly distributed be fixed on along coal mine well port to coal the principle of the present invention
Mine well bottom zone, can all areas to underground coal mine carry out Image Acquisition;Second Image Acquisition terminal is for acquiring
The image of the default corresponding movable object of underground coal mine, and mobile tracking shooting is carried out to movable object;Coal can be provided
Under mine track up can be carried out to the mobile target of underground coal mine while monitor video image.
Further, several second Image Acquisition terminals are evenly distributed on along coal mine well port to coal mine well bottom zone, each
Second Image Acquisition terminal passes through removable installation structure and is mounted on the coal mine borehole wall, and the second Image Acquisition terminal can be can
It is moved on mobile mounting structure, each second Image Acquisition terminal is equipped with corresponding pre-set image pickup area, default when having
When mobile target enters some second Image Acquisition terminal corresponding pre-set image pickup area, then the second Image Acquisition terminal
Image Acquisition is carried out to the movement target, after mobile target leaves image acquisition region, then the second Image Acquisition terminal stops
To the Image Acquisition of the movement target, when having multiple mobile targets while to enter the same second Image Acquisition terminal corresponding pre-
If when image acquisition region, which carries out Image Acquisition to multiple mobile targets simultaneously.
Further, the removable installation structure specifically includes:
Flexible guide rail, motor-driven carrier, controller, detection module;Second Image Acquisition terminal is fixed on motor-driven carrier
On vehicle body, motor-driven carrier is mounted on flexible guide rail and is slidably connected with flexible guide rail, and flexible guide rail is fixed on the coal mine borehole wall, inspection
It surveys module to be fixed on the vehicle body of motor-driven carrier, detection module is connect with controller;Default mobile target is equipped with matched signal
Sending module sends out the matched signal received when detection module, which detects, has matched signal in pre-set image pickup area
Controller is given, controller matches matched signal, and controller controls the second Image Acquisition to default shifting after successful match
Moving-target carries out Image Acquisition, and controls motor-driven carrier and carry out tracking movement to default mobile target;When default mobile target is moved
When dynamic pre-set image pickup area out, motor-driven carrier stops tracking, and resets to initial position.Preset when mobile target enters
After in image acquisition region, detection module can receive matched signal, then controller control motor-driven carrier to mobile target into
Line trace, tracking technique herein is target following technology in the prior art, and the application is without repeating.
Further, it is equipped at image defogging in the first image acquisition terminal and the second Image Acquisition terminal
Module is managed, described image defogging processing module specifically includes:
Acquisition unit, for obtaining underground coal mine ambient image;
Judging unit, for judging in underground coal mine ambient image with the presence or absence of mist;
Analytical unit, if in underground coal mine ambient image there are mist, to the mist of image in underground coal mine environment into
Row analysis, judges that the type of mist in underground coal mine ambient image, the type of mist include: the first monitored object underground coal mine environment
In mist, the mist on equipment camera lens is acquired in second of acquisition unit;
Defogging unit, if the type of mist is the first, is based on for carrying out defogging processing to underground coal mine ambient image
Defogging algorithm carries out defogging processing to underground coal mine ambient image;If the type of mist is second, to acquiring in acquisition unit
Equipment camera lens carries out wiping processing automatically;If the first has with second, two kinds of processing are carried out simultaneously.
Wherein, since underground coal mine environment is poor, moisture is heavier, is easy to produce fog, the present invention is first to mine supervision
Image carries out judging whether there is mist, then analyzes the mist in underground coal mine ambient image, judges underground coal mine environment
The type of mist in image, it is different from a kind of traditional only type of mist, fully consider that underground coal mine ambient image generates mist
Many reasons, then to underground coal mine ambient image carry out defogging processing, if the type of mist be the first, based on defogging calculate
Method carries out defogging processing to underground coal mine ambient image;If the type of mist is second, to acquisition equipment mirror in acquisition unit
Head carries out wiping processing automatically;If the first has with second, two kinds of processing are carried out simultaneously, so that final underground coal mine
Environment defogging treatment effect is preferable.
Further, the judging unit specifically includes:
Module is established, for, there are the underground coal mine ambient image of mist progress deep learning, establishing coal mine based on history
Lower ambient image fog judgment models;
Judgment module, for judging whether deposit in mine supervision image based on underground coal mine ambient image fog judgment models
In mist;
The analytical unit specifically includes:
First analysis module, for based in monitored object underground coal mine environment there are the underground coal mine ambient image of mist into
Row deep learning establishes the first underground coal mine ambient image fog type judgment models;
Second analysis module, for based on there are the underground coal mine ambient images of mist to carry out depth on acquisition equipment camera lens
It practises, establishes the second underground coal mine ambient image fog type judgment models;
Third analysis module, for being based on the first underground coal mine ambient image fog type judgment models and the second coal mine
Lower ambient image fog type judgment models, judge the type of the mist in underground coal mine ambient image.
Existing deep learning algorithm is used in the application, by the way that there are the progress of the underground coal mine ambient image of mist to history
Study, establishes underground coal mine ambient image fog judgment models, is judged by obtaining the feature of the image with mist accordingly
Image whether there is mist, and which is to be achieved by the prior art, and the application no longer repeats.
It is further, described that defogging processing is carried out to underground coal mine ambient image based on defogging algorithm, specifically:
Mist influence caused by image, Misty Image degradation model expression formula are described using Misty Image degradation model
Are as follows:
I (x)=J (x) e-rd (x)+A (1-e-rd (x)) (1)
Wherein, x is the space coordinate of image pixel, and r indicates that atmospheric scattering coefficient, d represent scenery depth, and A is that the whole world is big
Gas light ingredient;
Based on above-mentioned model, t (x) is used) to indicate transmissivity, it establishes and describes the mist graph model that mist image is formed:
I (x)=J (x) t (x)+A (1-t (x)); (2)
X is the space coordinate of image pixel, and I (x) is the image to defogging, and J (x) is fog free images, and A is global atmosphere light
Ingredient, t are refractive index, that is, atmospheric transmittance;
Estimate transmissivity t (x):
Obtain global atmosphere light A, then according to formula (2) it follows that
T (x)=A-I (x) A-J (x) (3)
The range of t (x) is [0,1], and the range of I (x) is [0,255], and the range of J (x) is [0,255];A and I (x) is
Know, the range of t (x) determined according to the range of J (x):
0≤J(x)≤255,0≤I(x)≤A,0≤J(x)≤A,0≤t(x)≤1 (4)
T (x) >=A-I (x) A-0=A-I (x) A=1-I (x) A (5)
Convolution (4) and formula (5) can obtain:
1-I(x)A≤t(x)≤1 (6)
Therefore just slightly estimate the calculation formula of transmissivity:
T (x)=1-I (x) A (7)
In order to guarantee the naturality of picture, increase a parameter w to adjust transmissivity:
T (x)=1-wI (x) A (8)
Iteration thought estimates transmissivity:
Formula (2) is converted to obtain:
J (x)=1t (x) I (x)-At (x)+A (9)
Image after estimating global atmosphere light A and transmissivity t (x), after defogging is calculated using formula (9);
Image uses auto contrast's enhancing or brightness enhancing or gamma correction image processing method pair after defogging
Image is handled.After defogging, obtained fog free images brightness can be partially dark, and auto contrast's enhancing, brightness can be used
Enhancing, the image processing methods such as gamma correction are handled, to obtain the fog free images of better effect.
Further, the acquisition unit is specially video monitoring system, using cleaning structure to acquiring in acquisition unit
Equipment camera lens carries out wiping processing automatically, and the cleaning structure includes:
Fixed plate, controller, mechanical arm, mounting disc, motor, cleaning axle, disk;
Fixed plate is fixed on video monitoring head shell upper surface, and controller is fixed in fixed plate, mechanical arm
One end is fixedly connected with fixed plate, and the mechanical arm other end can prolong to the camera lens of video monitoring camera under the control of the controller
It stretches, the mechanical arm other end is fixedly connected with the mounting disc back side, and motor is fixed on the positive Middle face of mounting disc, cleaning axle one end and motor
Rotation axis connection, the cleaning axle other end connect with the disk back side, and disk front is equipped with spongy layer;Motor is used in controller
It is rotated under control, disc-rotated, disk front spongy layer contacts rotation with the camera lens of video monitoring camera to view
The fog on the camera lens surface of frequency monitoring camera is purged.
Wherein, mechanical arm and motor are controlled by controller, mechanical arm drives disk to image close to video monitoring
The camera lens of head, motor is disc-rotated, is cleaned and is absorbed water using sponge, removes to the camera lens of video monitoring camera
Mist, controller control mechanical arm withdraws cleaning structure after the completion.
Further, it is preset with control program in the controller, for controlling displacement and the displacement of mechanical arm, so that
In cleaning, disk front spongy layer is contacted with the camera lens of video monitoring camera;Cleaning structure is withdrawn after the completion of cleaning, so that
Cleaning structure does not appear in the monitoring range of visibility of video monitoring camera;It can by controller command displacement and displacement
Accurately cleaned.
Pressure sensor is equipped in spongy layer, when pressure sensor detects that disk front spongy layer and video monitoring image
When the camera lens contact of head, the pressure that will test passes to controller, and controller controls motor and rotates predetermined amount, completes rotation
Cleaning structure is withdrawn using mechanical arm afterwards.
Further, humidity sensor, humidity in the spongy layer that humidity sensor will test are equipped in the spongy layer
Information passes to controller, and when humidity is greater than threshold value in the spongy layer detected, controller controls mechanical arm and disk is driven to move
It is dynamic, so that disk is mobile to video monitoring head shell, spongy layer is squeezed using video monitoring head shell, works as spongy layer
Interior humidity stops squeezing when being lower than threshold value, and is withdrawn cleaning structure using controller control mechanical arm.When moisture in spongy layer
When more, it is poor to will lead to clean effect, and therefore, when spongy layer excess moisture, controller controls mechanical arm and drives disk
It is mobile, so that disk is mobile to video monitoring head shell, spongy layer is squeezed using video monitoring head shell, it will be extra
Moisture squeeze out and then cleaned, ensure clean effect.
Further, the shell at the video monitoring cam lens is equipped with several infrared alignment transmitting modules, sea
The disk at continuous layer edge is equipped with several infrared alignment receiving modules, disk frontage dimension and video monitoring cam lens front
Size matching, infrared alignment transmitting module and infrared alignment receiving module correspond, infrared alignment transmitting module and infrared right
Quasi- receiving module is connect with controller, and when controller receives cleaning instruction, controller opens infrared alignment transmitting simultaneously
Module and infrared alignment receiving module, infrared alignment transmitting module is for emitting alignment infrared information, infrared alignment receiving module
It is directed at infrared information for receiving, when all infrared alignment receiving module receives what corresponding infrared alignment transmitting module was sent
When being directed at infrared information, then controller control motor is cleaned, without cleaning if alignment is unsuccessful.It solves existing
The incomplete technical problem of image defogging treatment effect that video monitor of mine system obtains, realizes for the fog in image
The different situations of generation are respectively calculated machine image defogging, and carry out wiping defogging processing to monitoring camera camera lens, protect
Hinder the technical effect of monitoring image defogging treatment effect.
One or more technical solution provided by the present application, has at least the following technical effects or advantages:
It solves existing underground coal mine video monitoring system to be merely able to provide the image in certain area, it can not be right
The mobile target of underground carries out the technical issues of track up, realizes system design rationally, is capable of providing coal mine underground monitoring
The technical effect of track up can be carried out while video image to the mobile target of underground coal mine.
Detailed description of the invention
Attached drawing described herein is used to provide to further understand the embodiment of the present invention, constitutes one of the application
Point, do not constitute the restriction to the embodiment of the present invention;
Fig. 1 is the composition schematic diagram of underground coal mine environment realizing of Robot Vision system in the application;
Fig. 2 is the structural schematic diagram of removable installation structure in the application;
Fig. 3 is the composition schematic diagram of image defogging processing module in the application;
Fig. 4 is the structural schematic diagram of cleaning structure in the application.
Specific embodiment
In the following description, numerous specific details are set forth in order to facilitate a full understanding of the present invention, still, the present invention may be used also
Implemented with being different from the other modes being described herein in range using other, therefore, protection scope of the present invention is not by under
The limitation of specific embodiment disclosed in face.
Referring to FIG. 1, this application provides a kind of underground coal mine environment realizing of Robot Vision system, the system comprises:
Several first Image Acquisition terminal a, several second Image Acquisition terminal b, data line c, power supply line, display are single
First d;
Several first Image Acquisition terminals, which are uniformly distributed, to be fixed on along coal mine well port to coal mine well bottom zone, and the first image is adopted
Collect terminal and is used for collection of coal mine underground predeterminable area ambient image data;First Image Acquisition terminal and the second Image Acquisition terminal
It is connect by power supply line with ground power supply;Second Image Acquisition terminal is for the default corresponding movable object in collection of coal mine underground
Image, and to movable object carry out mobile tracking shooting;First Image Acquisition terminal and the second Image Acquisition terminal will
The data of acquisition are shown by the display unit that data line is transmitted to ground.
Wherein, to be that several first Image Acquisition terminals are uniformly distributed be fixed on along coal mine well port to coal the principle of the present invention
Mine well bottom zone, can all areas to underground coal mine carry out Image Acquisition;Second Image Acquisition terminal is for acquiring
The image of the default corresponding movable object of underground coal mine, and mobile tracking shooting is carried out to movable object;Coal can be provided
Under mine track up can be carried out to the mobile target of underground coal mine while monitor video image.
Wherein, in the embodiment of the present application, several second Image Acquisition terminals are evenly distributed on along coal mine well port to coal mine
Well bottom zone, each second Image Acquisition terminal pass through removable installation structure and are mounted on the coal mine borehole wall, and the second image is adopted
Collection terminal can move in removable installation structure, and each second Image Acquisition terminal is equipped with corresponding pre-set image acquisition
Region, when there is default mobile target to enter the corresponding pre-set image pickup area of some second Image Acquisition terminal, then this
Two Image Acquisition terminals carry out Image Acquisition to the movement target, after mobile target leaves image acquisition region, then the second figure
As acquisition terminal stops to the Image Acquisition of the movement target, when there is multiple mobile targets, same second image of entrance is adopted simultaneously
When collecting the corresponding pre-set image pickup area of terminal, which carries out image simultaneously to multiple mobile targets and adopts
Collection.
Wherein, in the embodiment of the present application, referring to FIG. 2, the removable installation structure specifically includes:
Flexible guide rail 9, motor-driven carrier 10, controller 11, detection module 12;Second Image Acquisition terminal 13 is fixed on
On motor-driven carrier vehicle body, motor-driven carrier is mounted on flexible guide rail and is slidably connected with flexible guide rail, and flexible guide rail is fixed on coal mine
On the borehole wall, detection module is fixed on the vehicle body of motor-driven carrier, and detection module is connect with controller;It is set in default mobile target 15
There is matched signal sending module 14, when detection module, which detects, has matched signal in pre-set image pickup area, will receive
Matched signal be sent to controller, controller matches matched signal, after successful match controller control the second image
Acquisition carries out Image Acquisition to default mobile target, and controls motor-driven carrier and carry out tracking movement to default mobile target;When pre-
If mobile target moves out pre-set image pickup area, motor-driven carrier stops tracking, and resets to initial position.Work as movement
After target enters in pre-set image pickup area, detection module can receive matched signal, and then controller controls motor-driven carrier
Mobile target is tracked, tracking technique herein is target following technology in the prior art, and the application is without repeating.
Wherein, in the embodiment of the present application, in the first image acquisition terminal and the second Image Acquisition terminal
Equipped with image defogging processing module, referring to FIG. 3, described image defogging processing module specifically includes:
Acquisition unit, for obtaining underground coal mine ambient image;
Judging unit, for judging in underground coal mine ambient image with the presence or absence of mist;
Analytical unit, if in underground coal mine ambient image there are mist, to the mist of image in underground coal mine environment into
Row analysis, judges that the type of mist in underground coal mine ambient image, the type of mist include: the first monitored object underground coal mine environment
In mist, the mist on equipment camera lens is acquired in second of acquisition unit;
Defogging unit, if the type of mist is the first, is based on for carrying out defogging processing to underground coal mine ambient image
Defogging algorithm carries out defogging processing to underground coal mine ambient image;If the type of mist is second, to acquiring in acquisition unit
Equipment camera lens carries out wiping processing automatically;If the first has with second, two kinds of processing are carried out simultaneously.
Wherein, since underground coal mine environment is poor, moisture is heavier, is easy to produce fog, the present invention is first to mine supervision
Image carries out judging whether there is mist, then analyzes the mist in underground coal mine ambient image, judges underground coal mine environment
The type of mist in image, it is different from a kind of traditional only type of mist, fully consider that underground coal mine ambient image generates mist
Many reasons, then to underground coal mine ambient image carry out defogging processing, if the type of mist be the first, based on defogging calculate
Method carries out defogging processing to underground coal mine ambient image;If the type of mist is second, to acquisition equipment mirror in acquisition unit
Head carries out wiping processing automatically;If the first has with second, two kinds of processing are carried out simultaneously, so that final underground coal mine
Environment defogging treatment effect is preferable.
Wherein, in the embodiment of the present application, the judging unit specifically includes:
Module is established, for, there are the underground coal mine ambient image of mist progress deep learning, establishing coal mine based on history
Lower ambient image fog judgment models;
Judgment module, for judging whether deposit in mine supervision image based on underground coal mine ambient image fog judgment models
In mist;
The analytical unit specifically includes:
First analysis module, for based in monitored object underground coal mine environment there are the underground coal mine ambient image of mist into
Row deep learning establishes the first underground coal mine ambient image fog type judgment models;
Second analysis module, for based on there are the underground coal mine ambient images of mist to carry out depth on acquisition equipment camera lens
It practises, establishes the second underground coal mine ambient image fog type judgment models;
Third analysis module, for being based on the first underground coal mine ambient image fog type judgment models and the second coal mine
Lower ambient image fog type judgment models, judge the type of the mist in underground coal mine ambient image.
Existing deep learning algorithm is used in the application, by the way that there are the progress of the underground coal mine ambient image of mist to history
Study, establishes underground coal mine ambient image fog judgment models, is judged by obtaining the feature of the image with mist accordingly
Image whether there is mist, and which is to be achieved by the prior art, and the application no longer repeats.
It is further, described that defogging processing is carried out to underground coal mine ambient image based on defogging algorithm, specifically:
Mist influence caused by image, Misty Image degradation model expression formula are described using Misty Image degradation model
Are as follows:
I (x)=J (x) e-rd (x)+A (1-e-rd (x)) (1)
Wherein, x is the space coordinate of image pixel, and r indicates that atmospheric scattering coefficient, d represent scenery depth, and A is that the whole world is big
Gas light ingredient;
Based on above-mentioned model, t (x) is used) to indicate transmissivity, it establishes and describes the mist graph model that mist image is formed:
I (x)=J (x) t (x)+A (1-t (x)); (2)
X is the space coordinate of image pixel, and I (x) is the image to defogging, and J (x) is fog free images, and A is global atmosphere light
Ingredient, t are refractive index, that is, atmospheric transmittance;
Estimate transmissivity t (x):
Obtain global atmosphere light A, then according to formula (2) it follows that
T (x)=A-I (x) A-J (x) (3)
The range of t (x) is [0,1], and the range of I (x) is [0,255], and the range of J (x) is [0,255];A and I (x) is
Know, the range of t (x) determined according to the range of J (x):
0≤J(x)≤255,0≤I(x)≤A,0≤J(x)≤A,0≤t(x)≤1 (4)
T (x) >=A-I (x) A-0=A-I (x) A=1-I (x) A (5)
Convolution (4) and formula (5) can obtain:
1-I(x)A≤t(x)≤1 (6)
Therefore just slightly estimate the calculation formula of transmissivity:
T (x)=1-I (x) A (7)
In order to guarantee the naturality of picture, increase a parameter w to adjust transmissivity:
T (x)=1-wI (x) A (8)
Iteration thought estimates transmissivity:
Formula (2) is converted to obtain:
J (x)=1t (x) I (x)-At (x)+A (9)
Image after estimating global atmosphere light A and transmissivity t (x), after defogging is calculated using formula (9);
Image uses auto contrast's enhancing or brightness enhancing or gamma correction image processing method pair after defogging
Image is handled.After defogging, obtained fog free images brightness can be partially dark, and auto contrast's enhancing, brightness can be used
Enhancing, the image processing methods such as gamma correction are handled, to obtain the fog free images of better effect.
Wherein, in the embodiment of the present application, the acquisition unit is specially video monitoring system, using cleaning structure to adopting
Acquisition equipment camera lens carries out wiping processing automatically in collection unit, referring to FIG. 4, the cleaning structure includes:
Fixed plate 2, controller 3, mechanical arm 4, mounting disc 5, motor 6, cleaning axle 7, disk 8;
Fixed plate is fixed on video monitoring head shell upper surface, and controller is fixed in fixed plate, mechanical arm
One end is fixedly connected with fixed plate, and the mechanical arm other end can prolong to the camera lens of video monitoring camera under the control of the controller
It stretches, the mechanical arm other end is fixedly connected with the mounting disc back side, and motor is fixed on the positive Middle face of mounting disc, cleaning axle one end and motor
Rotation axis connection, the cleaning axle other end connect with the disk back side, and disk front is equipped with spongy layer;Motor is used in controller
It is rotated under control, disc-rotated, disk front spongy layer contacts rotation with the camera lens of video monitoring camera to view
The fog on the camera lens surface of frequency monitoring camera is purged.
Wherein, mechanical arm and motor are controlled by controller, mechanical arm drives disk to image close to video monitoring
The camera lens of head, motor is disc-rotated, is cleaned and is absorbed water using sponge, removes to the camera lens of video monitoring camera
Mist, controller control mechanical arm withdraws cleaning structure after the completion.
Wherein, in the embodiment of the present application, it is preset with control program in the controller, for controlling the displacement of mechanical arm
And displacement, so that disk front spongy layer is contacted with the camera lens of video monitoring camera in cleaning;It is withdrawn after the completion of cleaning
Cleaning structure, so that cleaning structure does not appear in the monitoring range of visibility of video monitoring camera;Pass through controller control bit
Shifting and displacement can be cleaned accurately.
Pressure sensor is equipped in spongy layer, when pressure sensor detects that disk front spongy layer and video monitoring image
When the camera lens contact of head, the pressure that will test passes to controller, and controller controls motor and rotates predetermined amount, completes rotation
Cleaning structure is withdrawn using mechanical arm afterwards.
Wherein, in the embodiment of the present application, humidity sensor is equipped in the spongy layer, what humidity sensor will test
Humidity information passes to controller in spongy layer, and when humidity is greater than threshold value in the spongy layer detected, controller control is mechanical
It is mobile that armband moves disk, so that disk is mobile to video monitoring head shell, squeezes sea using video monitoring head shell
Continuous layer stops squeezing when humidity is lower than threshold value in spongy layer, and is withdrawn cleaning structure using controller control mechanical arm.When
When moisture is more in spongy layer, it is poor to will lead to clean effect, and therefore, when spongy layer excess moisture, controller controls machine
It is mobile that tool armband moves disk, so that disk is mobile to video monitoring head shell, is squeezed using video monitoring head shell
Extra moisture is squeezed out and then is cleaned, ensures clean effect by spongy layer.
Wherein, in the embodiment of the present application, the shell at the video monitoring cam lens is equipped with several infrared right
The disk of quasi- transmitting module, spongy layer edge is equipped with several infrared alignment receiving modules, disk frontage dimension and video monitoring
The matching of cam lens frontage dimension, infrared alignment transmitting module and infrared alignment receiving module correspond, infrared alignment hair
It penetrates module and infrared alignment receiving module is connect with controller, when controller receives cleaning instruction, controller is opened simultaneously
Infrared alignment transmitting module and infrared alignment receiving module are opened, infrared alignment transmitting module is red for emitting alignment infrared information
Outer alignment receiving module is for receiving alignment infrared information, when all infrared alignment receiving module receives corresponding infrared alignment
When the alignment infrared information that transmitting module is sent, then controller control motor is cleaned, without clear if alignment is unsuccessful
It is clean.It solves the incomplete technical problem of image defogging treatment effect that existing video monitor of mine system obtains, realizes
For the different situations that the fog in image generates, be respectively calculated machine image defogging, and to monitoring camera camera lens into
Row wiping defogging processing, ensures the technical effect of monitoring image defogging treatment effect.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (10)
1. a kind of underground coal mine environment realizing of Robot Vision system, which is characterized in that the system comprises:
Several first Image Acquisition terminals, several second Image Acquisition terminals, data line, power supply line, display unit;
Several first Image Acquisition terminals, which are uniformly distributed, to be fixed on along coal mine well port to coal mine well bottom zone, and the first Image Acquisition is whole
End is used for collection of coal mine underground predeterminable area ambient image data;First Image Acquisition terminal and the second Image Acquisition terminal are logical
Power supply line is crossed to connect with ground power supply;Figure of the second Image Acquisition terminal for the default corresponding movable object in collection of coal mine underground
Picture, and mobile tracking shooting is carried out to movable object;First Image Acquisition terminal and the second Image Acquisition terminal will acquisitions
Data shown by the display unit that data line is transmitted to ground.
2. underground coal mine environment realizing of Robot Vision system according to claim 1, which is characterized in that several second images
Acquisition terminal is evenly distributed on along coal mine well port to coal mine well bottom zone, and each second Image Acquisition terminal passes through removable peace
Assembling structure is mounted on the coal mine borehole wall, and the second Image Acquisition terminal can move in removable installation structure, each second figure
It is whole when there is default mobile target to enter some second Image Acquisition as acquisition terminal is equipped with corresponding pre-set image pickup area
When holding corresponding pre-set image pickup area, then the second Image Acquisition terminal carries out Image Acquisition to the movement target, works as shifting
After moving-target leaves image acquisition region, then the second Image Acquisition terminal stops the Image Acquisition to the movement target, more when having
When a mobile target while pre-set image pickup area corresponding into the same second Image Acquisition terminal, which is adopted
Collection terminal carries out Image Acquisition to multiple mobile targets simultaneously.
3. underground coal mine environment realizing of Robot Vision system according to claim 2, which is characterized in that the removable peace
Assembling structure specifically includes:
Flexible guide rail, motor-driven carrier, controller, detection module;Second Image Acquisition terminal is fixed on motor-driven carrier vehicle body
On, motor-driven carrier is mounted on flexible guide rail and is slidably connected with flexible guide rail, and flexible guide rail is fixed on the coal mine borehole wall, detects mould
Block is fixed on the vehicle body of motor-driven carrier, and detection module is connect with controller;Default mobile target is sent equipped with matched signal
The matched signal received is sent to by module when detection module, which detects, has matched signal in pre-set image pickup area
Controller, controller match matched signal, and controller controls the second Image Acquisition to default mobile mesh after successful match
Mark carries out Image Acquisition, and controls motor-driven carrier and carry out tracking movement to default mobile target;When default mobile target moves out
When pre-set image pickup area, motor-driven carrier stops tracking, and resets to initial position.
4. underground coal mine environment realizing of Robot Vision system according to claim 1, which is characterized in that the first image
Image defogging processing module, described image defogging processing module tool are equipped in acquisition terminal and the second Image Acquisition terminal
Body includes:
Acquisition unit, for obtaining underground coal mine ambient image;
Judging unit, for judging in underground coal mine ambient image with the presence or absence of mist;
Analytical unit, if dividing for there are mists in underground coal mine ambient image the mist of image in underground coal mine environment
Analysis, judges that the type of mist in underground coal mine ambient image, the type of mist include: in the first monitored object underground coal mine environment
Mist acquires the mist on equipment camera lens in second of acquisition unit;
Defogging unit, if the type of mist is the first, is based on defogging for carrying out defogging processing to underground coal mine ambient image
Algorithm carries out defogging processing to underground coal mine ambient image;If the type of mist is second, to acquiring equipment in acquisition unit
Camera lens carries out wiping processing automatically;If the first has with second, two kinds of processing are carried out simultaneously.
5. underground coal mine environment realizing of Robot Vision system according to claim 4, which is characterized in that the judging unit
It specifically includes:
Module is established, for, there are the underground coal mine ambient image of mist progress deep learning, establishing underground coal mine ring based on history
Border image fog judgment models;
Judgment module, for judging to whether there is in mine supervision image based on underground coal mine ambient image fog judgment models
Mist;
The analytical unit specifically includes:
First analysis module, for based on there are the underground coal mine ambient images of mist to carry out deeply in monitored object underground coal mine environment
Degree study, establishes the first underground coal mine ambient image fog type judgment models;
Second analysis module, for based on acquisition equipment camera lens on there are the underground coal mine ambient image of mist carry out deep learning,
Establish the second underground coal mine ambient image fog type judgment models;
Third analysis module, for being based on the first underground coal mine ambient image fog type judgment models and the second underground coal mine ring
Border image fog type judgment models, judge the type of the mist in underground coal mine ambient image.
6. underground coal mine environment realizing of Robot Vision system according to claim 4, which is characterized in that described to be based on defogging
Algorithm carries out defogging processing to underground coal mine ambient image, specifically:
Mist influence caused by image, Misty Image degradation model expression formula are described using Misty Image degradation model are as follows:
I (x)=J (x) e-rd (x)+A (1-e-rd (x)) (1)
Wherein, x is the space coordinate of image pixel, and r indicates that atmospheric scattering coefficient, d represent scenery depth, and A is global atmosphere light
Ingredient;
Based on above-mentioned model, t (x) is used) to indicate transmissivity, it establishes and describes the mist graph model that mist image is formed:
I (x)=J (x) t (x)+A (1-t (x)); (2)
X is the space coordinate of image pixel, and I (x) is the image to defogging, and J (x) is fog free images, A be global atmosphere light at
Point, t is refractive index, that is, atmospheric transmittance;
Estimate transmissivity t (x):
Obtain global atmosphere light A, then according to formula (2) it follows that
T (x)=A-I (x) A-J (x) (3)
The range of t (x) is [0,1], and the range of I (x) is [0,255], and the range of J (x) is [0,255];A and I (x) is known
, the range of t (x) is determined according to the range of J (x):
0≤J(x)≤255,0≤I(x)≤A,0≤J(x)≤A,0≤t(x)≤1 (4)
T (x) >=A-I (x) A-0=A-I (x) A=1-I (x) A (5)
Convolution (4) and formula (5) can obtain:
1-I(x)A≤t(x)≤1 (6)
Therefore just slightly estimate the calculation formula of transmissivity:
T (x)=1-I (x) A (7)
In order to guarantee the naturality of picture, increase a parameter w to adjust transmissivity:
T (x)=1-wI (x) A (8)
Iteration thought estimates transmissivity:
Formula (2) is converted to obtain:
J (x)=1t (x) I (x)-At (x)+A (9)
Image after estimating global atmosphere light A and transmissivity t (x), after defogging is calculated using formula (9);
Image is after defogging, using auto contrast's enhancing or brightness enhancing or gamma correction image processing method to image
It is handled.
7. underground coal mine environment realizing of Robot Vision system according to claim 4, which is characterized in that the acquisition unit
Specially video monitoring system carries out wiping processing to acquisition equipment camera lens in acquisition unit using cleaning structure automatically, described
Cleaning structure includes:
Fixed plate, controller, mechanical arm, mounting disc, motor, cleaning axle, disk;
Fixed plate is fixed on video monitoring head shell upper surface, and controller is fixed in fixed plate, mechanical arm one end
It being fixedly connected with fixed plate, the mechanical arm other end can extend to the camera lens of video monitoring camera under the control of the controller,
The mechanical arm other end is fixedly connected with the mounting disc back side, and motor is fixed on the positive Middle face of mounting disc, cleaning axle one end and motor
Axis connection is rotated, the cleaning axle other end is connect with the disk back side, and disk front is equipped with spongy layer;Motor is used for the control in controller
It is rotated under system, disc-rotated, disk front spongy layer contacts rotation with the camera lens of video monitoring camera to video
The fog on the camera lens surface of monitoring camera is purged.
8. underground coal mine environment realizing of Robot Vision system according to claim 7, which is characterized in that in the controller
It is preset with control program, for controlling displacement and the displacement of mechanical arm, so that disk front spongy layer and video in cleaning
The camera lens of monitoring camera contacts;Cleaning structure is withdrawn after the completion of cleaning, so that cleaning structure does not appear in video monitoring camera shooting
In the monitoring range of visibility of head;Pressure sensor is equipped in spongy layer, when pressure sensor detect disk front spongy layer with
When the camera lens contact of video monitoring camera, the pressure that will test passes to controller, and it is default that controller controls motor rotation
Number is enclosed, utilizes mechanical arm withdrawal cleaning structure after completing rotation.
9. underground coal mine environment realizing of Robot Vision system according to claim 8, which is characterized in that in the spongy layer
Equipped with humidity sensor, humidity information passes to controller in the spongy layer that humidity sensor will test, when the sea detected
When humidity is greater than threshold value in continuous layer, controller controls mechanical arm and drives disk mobile, so that disk is to outside video monitoring camera
Shell is mobile, squeezes spongy layer using video monitoring head shell, stops squeezing when humidity is lower than threshold value in spongy layer, and benefit
Cleaning structure is withdrawn with controller control mechanical arm.
10. underground coal mine environment realizing of Robot Vision system according to claim 7, which is characterized in that the video prison
It controls the shell at cam lens and is equipped with several infrared alignment transmitting modules, the disk at spongy layer edge is equipped with several infrared
Be directed at receiving module, disk frontage dimension matches with video monitoring cam lens frontage dimension, infrared alignment transmitting module and
Infrared alignment receiving module corresponds, and infrared alignment transmitting module and infrared alignment receiving module are connect with controller, when
When controller receives cleaning instruction, controller opens infrared alignment transmitting module and infrared alignment receiving module simultaneously, infrared
It is directed at transmitting module and works as whole for receiving alignment infrared information for emitting alignment infrared information, infrared alignment receiving module
When infrared alignment receiving module receives the alignment infrared information that corresponding infrared alignment transmitting module is sent, then controller controls
Motor is cleaned, without cleaning if alignment is unsuccessful.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910218626.2A CN109951687B (en) | 2019-03-21 | 2019-03-21 | Machine vision implementation system for underground coal mine environment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910218626.2A CN109951687B (en) | 2019-03-21 | 2019-03-21 | Machine vision implementation system for underground coal mine environment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109951687A true CN109951687A (en) | 2019-06-28 |
CN109951687B CN109951687B (en) | 2021-04-23 |
Family
ID=67011181
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910218626.2A Expired - Fee Related CN109951687B (en) | 2019-03-21 | 2019-03-21 | Machine vision implementation system for underground coal mine environment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109951687B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110430402A (en) * | 2019-08-20 | 2019-11-08 | 福州市协成智慧科技有限公司 | A kind of Internet of Things vision optimization monitoring device |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060064286A1 (en) * | 2002-07-22 | 2006-03-23 | California Institute Of Technology | Multi-agent autonomous system |
CN201581930U (en) * | 2009-12-30 | 2010-09-15 | 扬州三鑫矿山成套设备有限公司 | Mining circulating robot |
CN204442554U (en) * | 2015-03-09 | 2015-07-01 | 石家庄学院 | A kind of real-time tracking system based on monopod video camera |
CN106851210A (en) * | 2017-03-01 | 2017-06-13 | 中国矿业大学(北京) | Development machine abnormal work and calamity forecast system based on image |
CN206668325U (en) * | 2017-03-21 | 2017-11-24 | 中国矿业大学(北京) | A kind of mine movable object real-time tracking and Precise Position System |
CN107958233A (en) * | 2017-12-26 | 2018-04-24 | 重庆轩腾科技有限公司 | Intelligent face recognition system |
CN108124089A (en) * | 2017-12-26 | 2018-06-05 | 重庆轩腾科技有限公司 | Intelligent monitor system based on monitoring camera-shooting |
-
2019
- 2019-03-21 CN CN201910218626.2A patent/CN109951687B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060064286A1 (en) * | 2002-07-22 | 2006-03-23 | California Institute Of Technology | Multi-agent autonomous system |
CN201581930U (en) * | 2009-12-30 | 2010-09-15 | 扬州三鑫矿山成套设备有限公司 | Mining circulating robot |
CN204442554U (en) * | 2015-03-09 | 2015-07-01 | 石家庄学院 | A kind of real-time tracking system based on monopod video camera |
CN106851210A (en) * | 2017-03-01 | 2017-06-13 | 中国矿业大学(北京) | Development machine abnormal work and calamity forecast system based on image |
CN206668325U (en) * | 2017-03-21 | 2017-11-24 | 中国矿业大学(北京) | A kind of mine movable object real-time tracking and Precise Position System |
CN107958233A (en) * | 2017-12-26 | 2018-04-24 | 重庆轩腾科技有限公司 | Intelligent face recognition system |
CN108124089A (en) * | 2017-12-26 | 2018-06-05 | 重庆轩腾科技有限公司 | Intelligent monitor system based on monitoring camera-shooting |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110430402A (en) * | 2019-08-20 | 2019-11-08 | 福州市协成智慧科技有限公司 | A kind of Internet of Things vision optimization monitoring device |
Also Published As
Publication number | Publication date |
---|---|
CN109951687B (en) | 2021-04-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10970859B2 (en) | Monitoring method and device for mobile target, monitoring system and mobile robot | |
US11324095B2 (en) | Automatic stage lighting tracking system and a control method therefor | |
CN109878552B (en) | Foreign matter monitoring devices between track traffic platform door and train based on machine vision | |
CN103885449B (en) | Intelligent vision based on multi-sensor cooperation process follows the tracks of the control method of wheeled robot | |
CN102006461B (en) | Joint tracking detection system for cameras | |
CN109905678A (en) | A kind of mine supervision image defogging processing system | |
CN103020983A (en) | Human-computer interaction device and method used for target tracking | |
CN107315410B (en) | Automatic obstacle removing method for robot | |
CN109934108B (en) | Multi-target and multi-type vehicle detection and distance measurement system and implementation method | |
CN104374376A (en) | Vehicle-mounted three-dimensional measurement system device and application thereof | |
CN103419944A (en) | Air bridge and automatic abutting method therefor | |
CN104952104A (en) | Three-dimensional human body gesture estimating method and device thereof | |
CN105373130A (en) | Special device accident on-site information detection system based on stereo modeling | |
CN112633131B (en) | Underground automatic tracking method based on deep learning video identification | |
CN112116633A (en) | Mine drilling counting method | |
CN105676884A (en) | Infrared thermal imaging searching/ tracking/ aiming device and method | |
CN116832380B (en) | Fire extinguishing method of fire extinguishing system based on multi-fire-source point evaluation decision | |
CN112270807A (en) | Old man early warning system that tumbles | |
CN109951687A (en) | A kind of underground coal mine environment realizing of Robot Vision system | |
CN112964180A (en) | Urban rail vehicle collector shoe detection system | |
CN106941580B (en) | The method and system that teacher student automatically tracks is realized based on single detective camera lens | |
CN105518490A (en) | Object detection method and device, remote control moving equipment, and aircraft | |
CN106815556A (en) | A kind of plane crowd hazards data collecting system of many data types | |
CN113194249A (en) | Moving object real-time tracking system and method based on camera | |
CN107538485A (en) | A kind of robot guidance method and system |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210423 |