CN104100256B - Method for measuring coal mine underground drilling depth based on image processing technology - Google Patents

Method for measuring coal mine underground drilling depth based on image processing technology Download PDF

Info

Publication number
CN104100256B
CN104100256B CN201310132223.9A CN201310132223A CN104100256B CN 104100256 B CN104100256 B CN 104100256B CN 201310132223 A CN201310132223 A CN 201310132223A CN 104100256 B CN104100256 B CN 104100256B
Authority
CN
China
Prior art keywords
frame
drilling
target
template
tracking
Prior art date
Application number
CN201310132223.9A
Other languages
Chinese (zh)
Other versions
CN104100256A (en
Inventor
董立红
李占利
张杰慧
Original Assignee
西安科技大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 西安科技大学 filed Critical 西安科技大学
Priority to CN201310132223.9A priority Critical patent/CN104100256B/en
Publication of CN104100256A publication Critical patent/CN104100256A/en
Application granted granted Critical
Publication of CN104100256B publication Critical patent/CN104100256B/en

Links

Abstract

The invention discloses an image processing video tracking technology. The technology specifically comprises tracking the movement of drills in a punching video, measuring and analyzing tracking paths to automatically calculate the number of drilling rods, and calculating the depth of drilling holes according to the number of the drilling rods. The measuring error caused by human factors can be avoided, the data can be processed in real time or after the event; the measuring security is excellent, and major hidden danger can be solved. A method for measuring the coal mine underground drilling depth based on an image processing technology is provided, and the technology for automatically tracking and identifying objects in the video with low-quality in the underground complex environment is provided; according to the method, the problems that the underground movement objects are complex, and underground workers block the drilling hole video can be solved; the complex light environment generated by underground mine lamps and head lamps of the workers can be processed; the tracking result automatic quantitative analysis and counting judging result precision is high, and the real-time performance of a system is high.

Description

Coal mine down-hole drilling depth measurement method based on image processing techniquess

Technical field

The present invention relates to image processing techniquess, and in particular to the coal mine down-hole drilling depth survey based on image processing techniquess Method.

Background technology

Gas is always the major hidden danger of Safety of Coal Mine Production.Maximum for gaseous mine, during coal production Potential safety hazard is gas accident.China's nineteen ninety to 2000, once dead 3 people above major gas accident proportion was year by year Rise, up to 45.61%, more than 40% is always held at later within 1996.Therefore, gas accident is China's coal-mine safety The high principal contradiction of accident, effective control gas accident is the key for solving China's coal-mine safety problem.Due to gas Accident it is very harmful, eliminating gas accident hidden danger needs to spend more time and expense, to high gas and outburst mine, machinery Change extractive equipment to be difficult to play effectiveness, coal roadway tunneling speed was generally all difficult to more than the 100m/ months, and stope yield is generally difficult With more than 1,000,000 t/a.Therefore, the threat of Gas Disaster accident also strongly limit coal production scale, production efficiency and economy The raising of benefit.The effective control of Gas Disaster is a critical problem for ensureing China coal industry sustainable development.

Therefore comprehensive control is carried out to coal mine gas particularly important, application at present is gas compared with wide, the preferable technology of effect Extraction technology.Gas pumping is the Main Means that colliery prevents Gas Outburst.At present for the measurement of drilling depth is also rested on The method of measurement while drilling, subsequent supervision cannot be carried out.And for the measurement after drilling still adopts manual method, there is safety hidden Suffer from.

The content of the invention

In order to solve the deficiencies in the prior art, the present invention adopts image processing techniquess, and drilling rod counting algorithm is applied automatically, surveys Amount squeezes into the quantity of drilling rod during drilling, and according to drilling rod quantity calculating drilling depth, it is to avoid anthropic factor is caused Measuring error, can in real time or afterwards processing data, and the safety for measuring is good, solves great potential safety hazard.

The technical scheme is that:Based on the coal mine down-hole drilling depth measurement method of image processing techniquess, including with Lower step:

Step one, calculates drill bit and pierces the frame number FramIn=of process and drill and take time automatically according to the frame per second number of video (s) * video frame rates (frame/s), and exit process frame number FrameOut=move back brill the time required to take time (s) * video frame rates (frame/s);

Step 2, selects the drill bit rectangle frame drilled as To Template, records position and the size of target rectangle frame And the target image of drill bit;

Step 3, the feature histogram weighted using kernel function according to average drifting target tracking algorism is describing target mould Plate, similarity measurement is carried out in every frame to To Template model and candidate target model, and along core histogram similarity Gradient direction iterative search target location, realizes target following;The target following is exactly the drill bit target position according to previous frame Put y0, the position y for making that distance is minimum or likeness coefficient is maximum is found in the current frame1. under conditions of core window width determines, Obtain target masterplate p and candidate target masterplate q and histogram model after, the matching distance between model is defined asWherein ρ is Bhattacharyya coefficients, by constantly along the gradient side of similarity measurement To new position is moved to until restraining, current new position is obtained;

Step 4, in drilling process the motion of drill bit for constantly jittery advance with retreat, so to tracking result away from Mark leave the right or normal track first using median smoothing process, the offset distance at certain moment t is d ' (t)=med { d (t-k, t+k) } after smoothing, its Middle k is filtering dimensional parameters, and by finding the notable jump of the To Template offset distance after smoothing single drilling rod is judged The division of motion, whole curve movement is rendered as Z-shaped repeatedly, and the criterion of significantly jump is defined as judging point In front and back three mean deviations distance is more than the 2/3 of whole movement locus (i.e. run of steel) in certain limit, i.e. meetThe time difference that vertical jump in succession twice need to be met simultaneously meets the time requirement of setting, Meet Ti+1-Ti> a* (FrameIn+FrameOut), a are then a drilling rod quantity between 0 to 1;

Step 5, drilling depth can be convenient for measuring according to the drilling well drilling bar quantity of analysis and learn:Depth=runs of steel * Drilling well drilling bar quantity.

Further improvement of the present invention includes:

The step 2 also include select drilling rod direction as direction of primary motion, to avoid complicated subsurface environment in do more physical exercises The interference of the generations such as target such as labour movement and circumstance of occlusion, record the parameter of direction line segment.

If the position of target is displaced to from the fortune selected in step 2 in the present frame that mean shift algorithm is traced into Dynamic direction certain distance, then it is assumed that present frame target is lost, the frame result is invalid, with the predictive value of persistent movement as target position Put.

The step 3 also include judging by gray feature, gradient direction feature and Corner Feature To Template and Similarity between candidate family.

The step 3 also includes defining target area, 3 centered on the geometric center of target area times target area face Long-pending areas are background area, calculate discrimination D to judge the differentiation degree of target and background, calculating principle be defined asWherein Ha and Hb represent respectively the feature histogram of target area and background area

The present invention adopts image processing video tracking technique, and to punching, the drill bit movement in video is tracked, measurement point Analysis pursuit path calculates the number of drilling rod automatically, and according to drilling rod quantity calculating drilling depth, it is to avoid anthropic factor is made Into measuring error, can in real time or afterwards processing data, and the safety for measuring is good, solves great potential safety hazard. Based on the coal mine down-hole drilling depth measurement method of image processing techniquess, to the target of low quality video in the complex environment of down-hole from Motion tracking technology of identification, the method can not only solve down-hole moving target complexity, the screening that underground labour produces to hole-drilling video Gear problem, and the complex illumination environment produced in down-hole mine lamp and workman's forehead lamp can also be tackled, while tracking result is certainly Dynamic quantitative analyses and counting judged result precision are very high, and the real-time of system is high.

Specific embodiment

The present invention is elaborated below.

1. video pre-filtering

The process of 1.1 videos

Choose and open with a series of pending drilling well video, read video inner parameter.

The automatic acquisition of 1.2 parameters

Drill bit is calculated according to the frame per second number of video automatically pierce the frame number FramIn=of process and drill (s) * videos that take time Frame per second (frame/s), and exit process frame number FrameOut=move back brill the time required to take time (s) * video frame rates (frame/s)

2. target following

The selection of 2.1 tracking targets and the direction of motion

Artificial mouse selects the drill bit rectangle frame drilled as To Template, and selects drilling rod direction as main motion side To, to avoid complicated subsurface environment in the multiple mobile object such as interference of the generation such as labour movement and circumstance of occlusion.Record target The parameter of rectangle frame and direction line segment.

The 2.2 average drifting target tracking algorisms based on multiple features

The feature histogram that average drifting target tracking algorism is weighted using kernel function describing target, to mesh in every frame Mark template model and candidate target model carry out similarity measurement, and along the gradient direction iterative search of core histogram similarity Target location, realizes target following.

A) selection of multiple features

Downhole video is the serious low resolution gray-scale maps of noise, is difficult to distinguish tracked brill using single gray feature Head and background, the ability that different feature descriptions distinguishes target and background distribution is different.Therefore, with reference to gray feature, gradient side The similarity between To Template and candidate family is judged to feature and Corner Feature.

Target area is defined, the areas of 3 centered on the center of target area times target area area are as background area Domain, calculates discrimination D to judge the differentiation degree of target and background, and the calculating principle of D is defined asWherein Ha With the feature histogram that Hb represents respectively target area and background area.

B) tracking of present frame target location

Target following is exactly the drill bit target location y according to previous frame0, find makes distance minimum or phase in the current frame Like the position y that property coefficient is maximum1. under conditions of core window width determines, obtain target masterplate p's and candidate target masterplate q and straight After square graph model, the matching distance between model is defined asWherein ρ is Bhattacharyya systems Number.New position is moved to by the gradient direction constantly along similarity measurement until convergence, obtains current new position.

C) correction of tracking position of object and model modification

If the position of the present frame that mean shift algorithm is traced into is displaced to from the direction of motion one selected in 2.1 steps Set a distance, then it is assumed that present frame target is lost, the frame result is invalid, and the predictive value with persistent movement is as target location.

3. automatic drilling rod method of counting

The movement locus of 3.1 smooth target followings

The motion of drill bit in drilling process is constantly jittery advance and retrogressing, so to tracking result apart from track elder generation Using median smoothing process, the offset distance at certain moment t is d ' (t)=med { d (t-k, t+k) } after smoothing, and wherein k is filtering Dimensional parameters.

3.2 drilling rods move back the judgement in bar cycle

Boring procedure for show as into-move back-...-cycle movement entered-move back, then show as on movement locus figure partially Move apart from periodically jump is presented, judge that single drilling rod is transported by finding the notable jump of the offset distance after smoothing Dynamic division, whole curve movement is rendered as Z-shaped repeatedly, and the criterion of significantly jump is defined as judging before point Afterwards three mean deviations distance is more than the 2/3 of whole movement locus (i.e. run of steel) in certain scope, i.e. meetNeed to meet while twice the time difference of vertical jump in succession meets the time of setting simultaneously Require, that is, meet Ti+1-Ti> a* (FrameIn+FrameOut), a is between 0 to 1.

4. drilling depth measurement and correction

Drilling depth can be convenient for measuring according to the drilling well drilling bar quantity of analysis and learn:Depth=run of steel * drilling well drilling bars Quantity.Due to down-hole complex illumination and movement environment, or the irregular operation of workman etc. may cause tracking to be lost, or bore The movement locus of head provide track following figure not fully to drill and moving back drill-through journey in system, can brightly see Go out tracking loss to drill the cycle with irregular, management personnel can verify simultaneously partial video section according to track following figure Correction result of calculation.

The ultimate principle and principal character and advantages of the present invention of the present invention has been shown and described above.The technology of the industry Personnel it should be appreciated that the present invention is not restricted to the described embodiments, the simply explanation described in above-described embodiment and description this The principle of invention, without departing from the spirit and scope of the present invention, the present invention also has various changes and modifications, these changes Change and improve clan to enter in scope of the claimed invention.The claimed scope of the invention by appending claims and its Equivalent thereof.

Claims (4)

1. the coal mine down-hole drilling depth measurement method of image processing techniquess is based on, it is characterised in that comprised the following steps:
Step one, calculates drill bit and pierces drill (s) * that take time of frame number FramIn=of process and regard automatically according to the frame per second number of video Frequency frame per second (frame/s), and exit process frame number FrameOut=move back brill the time required to take time (s) * video frame rates (frame/ s);
Step 2, selects the drill bit rectangle frame drilled as To Template P, record To Template P position and size and The target image of drill bit;
Step 3, the feature histogram weighted using kernel function according to average drifting target tracking algorism describing To Template P, Similarity measurement is carried out to To Template P and candidate target template q in every frame, and along the gradient side of core histogram similarity To iterative search target location, target following is realized;The target following is exactly the drill bit target location according to previous frame y0, the position y for making that distance is minimum or likeness coefficient is maximum is found in the current frame1, in the condition that core window width determines Under, after obtaining the histogram model of To Template p and candidate target template q, the matching distance between model is defined asWherein ρ is Bhattacharyya coefficients, by constantly along the gradient side of similarity measurement To new position is moved to until restraining, current new position is obtained;
Step 4, the motion of drill bit in drilling process is constantly jittery advance and retrogressing, so to tracking result apart from rail First using median smoothing process, the offset distance at certain moment t is d ' (t)=med { d (t-k, t+k) } to mark after smoothing, and wherein k is Filtering dimensional parameters, by finding the notable jump of the tracking result offset distance after smoothing drawing for single drilling rod motion is judged Point, while the time difference that need to meet vertical jump in succession twice meets the time requirement of setting, that is, meet Ti+1-Ti> a* (FrameIn+ FrameOut), a is a drilling rod quantity between 0 to 1;
Step 5, drilling depth can be convenient for measuring according to the drilling well drilling bar quantity of analysis and learn:Depth=run of steel * drilling wells Drilling rod quantity.
2. the coal mine down-hole drilling depth measurement method based on image processing techniquess according to claim 1, its feature exists Also include selecting drilling rod direction as direction of primary motion in, the step 2, to avoid complicated subsurface environment in multiple mobile object The interference of generation and circumstance of occlusion, record the parameter of direction line segment.
3. the coal mine down-hole drilling depth measurement method based on image processing techniquess according to claim 2, its feature exists In if the position of present frame that average drifting target tracking algorism is traced into is displaced to from the motion side selected in step 2 To certain distance, then it is assumed that present frame target is lost, the frame result is invalid, and the predictive value with persistent movement is as target location.
4. the coal mine down-hole drilling depth measurement method based on image processing techniquess according to claim 1, its feature exists In the step 3 also includes judging To Template p and candidate by gray feature, gradient direction feature and Corner Feature Similarity between To Template q.
CN201310132223.9A 2013-04-15 2013-04-15 Method for measuring coal mine underground drilling depth based on image processing technology CN104100256B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310132223.9A CN104100256B (en) 2013-04-15 2013-04-15 Method for measuring coal mine underground drilling depth based on image processing technology

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310132223.9A CN104100256B (en) 2013-04-15 2013-04-15 Method for measuring coal mine underground drilling depth based on image processing technology

Publications (2)

Publication Number Publication Date
CN104100256A CN104100256A (en) 2014-10-15
CN104100256B true CN104100256B (en) 2017-04-12

Family

ID=51668760

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310132223.9A CN104100256B (en) 2013-04-15 2013-04-15 Method for measuring coal mine underground drilling depth based on image processing technology

Country Status (1)

Country Link
CN (1) CN104100256B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105649607A (en) * 2015-12-31 2016-06-08 郑州光力科技股份有限公司 Mining while-drilling type borehole track detection method and system and drill rod
CN107833240B (en) * 2017-11-09 2020-04-17 华南农业大学 Target motion trajectory extraction and analysis method guided by multiple tracking clues
CN110259438B (en) * 2019-06-21 2020-12-04 精英数智科技股份有限公司 Intelligent monitoring method and device for coal mine water exploration and drainage and terminal equipment

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1818337A (en) * 2006-03-27 2006-08-16 天地科技股份有限公司 Testing method and apparatus for geological mechanics parameter under coal mine
CN2904512Y (en) * 2006-03-27 2007-05-23 天地科技股份有限公司 Test hole electronic sighting device for mine
CN101575983A (en) * 2009-02-27 2009-11-11 河南省煤层气开发利用有限公司 Directional fracturing permeability improvement outburst elimination method in coal mine and device thereof.
CN102174886A (en) * 2011-02-16 2011-09-07 中国地质大学(武汉) LWD (Logging While Drilling) real-time detection device and method of horizontal directional drilling depth of coal bed gas
CN102808610A (en) * 2012-08-06 2012-12-05 淮南同正科技有限公司 Real-time detection device and real-time detection method for coal mine downhole drilling depth
CN102944906A (en) * 2012-11-06 2013-02-27 山东科技大学 Precise search observation method for form and evolutionary process of crack of coal-mine roof
CN102968801A (en) * 2012-09-12 2013-03-13 浙江师范大学 Moving target tracking method based on photoelectric mixing combination transformation correlation
CN202866790U (en) * 2012-08-06 2013-04-10 淮南同正科技有限公司 Coal mine downhole drilling depth real-time detection device

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1818337A (en) * 2006-03-27 2006-08-16 天地科技股份有限公司 Testing method and apparatus for geological mechanics parameter under coal mine
CN2904512Y (en) * 2006-03-27 2007-05-23 天地科技股份有限公司 Test hole electronic sighting device for mine
CN101575983A (en) * 2009-02-27 2009-11-11 河南省煤层气开发利用有限公司 Directional fracturing permeability improvement outburst elimination method in coal mine and device thereof.
CN102174886A (en) * 2011-02-16 2011-09-07 中国地质大学(武汉) LWD (Logging While Drilling) real-time detection device and method of horizontal directional drilling depth of coal bed gas
CN102808610A (en) * 2012-08-06 2012-12-05 淮南同正科技有限公司 Real-time detection device and real-time detection method for coal mine downhole drilling depth
CN202866790U (en) * 2012-08-06 2013-04-10 淮南同正科技有限公司 Coal mine downhole drilling depth real-time detection device
CN102968801A (en) * 2012-09-12 2013-03-13 浙江师范大学 Moving target tracking method based on photoelectric mixing combination transformation correlation
CN102944906A (en) * 2012-11-06 2013-02-27 山东科技大学 Precise search observation method for form and evolutionary process of crack of coal-mine roof

Also Published As

Publication number Publication date
CN104100256A (en) 2014-10-15

Similar Documents

Publication Publication Date Title
CN105899749B (en) The stratum and interpretation of structure of deviated wellbore and horizontal hole
CN104612675B (en) A kind of carbonate formation Lithology while drilling method for quickly identifying
AU2010345083B2 (en) Method for using dynamic target region for well path/drill center optimization
Jafari et al. Spatial prediction of soil great groups by boosted regression trees using a limited point dataset in an arid region, southeastern Iran
CA2725923C (en) Heterogeneous earth models for a reservoir field
CN104730596B (en) A kind of discrete fractures modeling method based on multiple dimensioned factor constraints
US10318662B2 (en) Systems and methods for horizontal well correlation and geosteering
Coli et al. Evaluation of rock-mass permeability tensor and prediction of tunnel inflows by means of geostructural surveys and finite element seepage analysis
DE112013007332T5 (en) Probabilistic methodology for real-time drilling
CN102705010B (en) Working surface outburst risk evaluation index system analysis method
US20130341093A1 (en) Drilling risk avoidance
US9103936B2 (en) Method and apparatus for determining and displaying geosteering information
CN103529475A (en) Method for identifying and interpreting carbonate rock ancient karst reservoir layer three-dimensional structure
MX2015004000A (en) Determining a confidence value for a fracture plane.
CA2888037C (en) System, method and computer program product for a rug plot for geosteering applications
AU2009311619A1 (en) System and method for planning a drilling operation
CN103774989B (en) Treat that drilling well Trajectory Design horizontal well geosteering analyzes method with boring dynamic modeling
RU2009147272A (en) System and method for performing a drilling operation at a field
CN102918423A (en) Method for seismic hydrocarbon system analysis
CN105785471B (en) A kind of shock hazard evaluation method of the pre- working seam of mine
CN106869790B (en) A kind of quick fine geology guidance method of shale gas horizontal well
CN104847264B (en) A kind of method and apparatus realizing geosteering
CN103837908A (en) Rapid prospecting positioning method applicable to hidden sandstone-type uranium mine
CN103046868A (en) Integrated horizontal well geosteering method
CN106285476B (en) A kind of horizontal drilling Real-time Seismic geological syntheses guidance method

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20170412

Termination date: 20190415

CF01 Termination of patent right due to non-payment of annual fee