CN101718527A - Method for calculating length of dry sand of tailings reservoir based on image recognition - Google Patents
Method for calculating length of dry sand of tailings reservoir based on image recognition Download PDFInfo
- Publication number
- CN101718527A CN101718527A CN200910248499A CN200910248499A CN101718527A CN 101718527 A CN101718527 A CN 101718527A CN 200910248499 A CN200910248499 A CN 200910248499A CN 200910248499 A CN200910248499 A CN 200910248499A CN 101718527 A CN101718527 A CN 101718527A
- Authority
- CN
- China
- Prior art keywords
- beach
- dam
- image
- dry sand
- dried
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The invention discloses a method for calculating the length of the dry sand of a tailings reservoir based on image recognition, which comprises the following steps: binarizing a dry sand image of the tailings reservoir through image recognition technology so that the dry sand image is more simple, a data size is reduced and the outlines of a dam body of a tailings dam and the dry sand are shown; processing and analyzing the binarized dry sand image and utilizing pixel detection technology to calculate and analyze the height of the dry sand to further analyze accurate required data results; and finally utilizing the recognized height of the dry sand and a slope to calculate the length of the dry sand of the tailings reservoir. In the method, the image recognition technology is utilized to accurately divide the dry sand image of the tailings reservoir; data obtained by analysis and calculation is accurate; the accuracy of the calculated length of the dry sand is relatively higher and real-time property is relatively stronger; and the requirements of the online and real-time monitoring of the length of the dry sand of the tailings reservoir are met.
Description
Technical field
The present invention relates to a kind of image recognition technology, relate to a kind of metal or non-metal mine dried beach length calculation measuring method specifically the mine tailing storehouse.
Background technology
Length of dry sand of tailings reservoir is to weigh the important indicator that can the mine tailing storehouse pass the flood period safely flood season, also is to estimate whether one of the key factor of safety of mine tailing storehouse.Seek a kind of suitable image processing techniques the real dam crest height in mine tailing storehouse is accurately handled, analyzes, drawn to the dry sand of tailings reservoir image, and then calculate the top priority that the dried beach of tailing dam length has just become exploitation mine tailing storehouse robotization safety monitoring system.
Measure dried beach length at present and mainly rely on manual type.When flood season, mine tailing Kuku district water level raises, and does beach length and shortens, and at this moment does beach length and whether reaches national tailing dam design specifications standard.Just need the mine tailing workman to go field survey to do beach length, not only error is big, precision is low for the dried beach length of hand dipping like this, and main is that the data that draw are not data real-time, that continuous coverage draws.Cause the mining area leader to be difficult in time grasp the every safety technique index in mine tailing storehouse, these factors all greatly influence the safety management in mine tailing storehouse.So seek a kind of method to the dried beach length in mine tailing storehouse carry out in real time, continuous detection is very to be necessary.Take place significant to the accident that reduces the mine tailing storehouse.
Summary of the invention
Do the defective that beach length exists at existing hand dipping, the invention provides that a kind of adaptability is good, calculated amount is few, recognition result is accurate, accuracy of identification is high.The dried beach length calculation method that the result is accurate, real-time is higher.
Solving the problems of the technologies described above the concrete technical measures of being taked is: a kind of method based on the image recognition calculating length of dry sand of tailings reservoir is characterized in that:
(1) observe to do the beach image: do the beach image and be by camera the resulting image of the camera angle of dam body unanimity, should reach more than 704 * 576 to the pixel of doing the beach image.
(2) colour difference on clear and definite tailing dam accumulation of dam and dried beach is drawn some spots X0 on the tailing dam accumulation of dam ... Xn utilizes C# program coordinate figure, the color value information of the drawn point of record automatically, and through X0 ... Xn strokes and dots Zn bar straight line;
(3) determine threshold value, picture carried out binary conversion treatment:
Utilize the alternative manner selected threshold, the average gray T0 to Tn that initial threshold is chosen for image i.e. drawn X0 on dam body ... the color value of Xn, utilize T0 to Tn that the pixel branch on every straight line is made two parts then, calculate two parts average gray separately, part less than T0 is TA, part greater than T0 is TB, utilize the C# programming to realize the automatic detection computations of color value of each coordinate points, T1 is replaced T0 as new global threshold, repeat above process, so iteration detects to the image bottom until automatic;
(4) do the beach height with the image calculation after the binaryzation:
(4.1) mark out the fill dam dam crest point coordinate X0 of each bar binaryzation line segment ... Xn, Y0 ... Yn;
(4.2) realize detecting automatically the accumulation of dam of each bar binaryzation line segment and the coordinate X ' 0 of dried beach separation with the C# programming ... X ' n, Y ' 0 ... Y ' n;
(4.3) mark out each bar from the fill dam dam crest between the binaryzation line segment of doing the beach separation apart from K0 ... Kn;
(4.4) utilize pixel detection technology to calculate the function corresponding relation of line segment between pixel and the extremely dried beach of the fill dam dam crest separation: x=H/k
X is the distance of a pixel representative, and H is the dried beach height of demarcating behind the field survey, and k is that the fill dam dam crest is to the sum of all pixels of doing the beach separation;
(4.5) calculate dried beach height number: (Y0-Y ' 0) * x=H;
(5) calculate dried beach length
The dried beach gradient J that utilizes the dried beach height H obtained and field survey is by formula: L=tanJ * HL is dried beach length.
Beneficial effect of the present invention: video done the image processing techniques of beach picture binaryzation and based on the image binaryzation partitioning algorithm of two dimensional gray threshold value, the dried beach height of doing in the image of beach can be separated from background effectively, be had good binaryzation result.Detecting error proves less than 2cm through field survey.Satisfy national standard to doing the detection error of beach height.This method has that adaptability is good, calculated amount is few, recognition result is accurate, the accuracy of identification advantages of higher.The dried beach length that the dried beach altitude gauge that goes out by image recognition is calculated, result not only accurately but also convenient, real-time is higher.
Description of drawings
Fig. 1 is the dry sand of tailings reservoir picture that does not carry out binaryzation;
Fig. 2 is the dry sand of tailings reservoir picture after the binaryzation.
Embodiment
Accompanying drawings the inventive method.
A kind of method based on the image recognition calculating length of dry sand of tailings reservoir, its concrete recognition methods is as follows:
At first dried beach picture such as the Fig. 1 to the mine tailing storehouse carries out binary conversion treatment, and binarization processing of images is exactly that gray values of pixel points on the image is set to 0 or 255, just entire image is presented tangible black and white effect.Image threshold divides that to cut be a kind of the most frequently used, also is the simplest image partition method simultaneously, and it is specially adapted to the image that target and background occupies the different grey-scale scope.It is amount of compressed data greatly not only, and has simplified analysis and treatment step greatly, therefore under many circumstances, is to carry out graphical analysis, feature extraction and pattern-recognition necessary image preprocessing process before.The purpose of image thresholdization is to carry out a division to collection of pixels according to gray level, and each subclass that obtains forms one and the corresponding zone of real scenery, and each intra-zone has consistent attribute, and the adjacent area layout has this consistent attribute.Such division can realize by choose one or more threshold values from gray level.Thresholding method is a kind of image Segmentation Technology based on the zone, and its ultimate principle is: by setting different characteristic threshold value, the image slices vegetarian refreshments is divided into some classes.Determine threshold value T, passing threshold T is two parts, i.e. background area and target area with image segmentation.
The beach image of dried beach height do to(for) tailing dam adopts following method to discern, and analyzes:
(1) observe to do the beach image: as shown in Figure 1, do the beach image and be by camera, should reach more than 704 * 576 to the pixel of doing the beach image to the resulting image of the camera angle of dam body unanimity.
(2) gray scale difference on clear and definite accumulation of dam and dried beach, colour difference, on the accumulation of dam of picture, draw some spots X0 ... Xn, utilize the automatically information such as coordinate figure, color value of the drawn point of record of C# program, and through this X0 ... Xn strokes and dots Zn bar straight line such as Fig. 2.
(3) determine threshold value, picture carried out binary conversion treatment:
Utilize the alternative manner selected threshold, initial threshold is chosen for average gray T0 to Tn (the i.e. drawn X0 on dam body of image ... the color value of Xn), utilize T0 to Tn that the pixel branch on every straight line is made two parts then, calculate two parts average gray separately, part less than T0 is TA, part greater than T0 is TB, utilize the C# programming to realize the automatic detection computations of color value of each coordinate points, T1 is replaced T0 as new global threshold, repeat above process, so iteration detects to doing image bottom, beach.Through test relatively, the method can obtain satisfactory result quickly, and it has kept the details of former figure preferably.
(4) utilize the image calculation after the binaryzation to do the beach height.
(4.1) as shown in Figure 2, mark out the fill dam dam crest point coordinate (X0 of each bar binaryzation line segment ... Xn, Y0 ... Yn) for example: the dam crest coordinate points is respectively: (14,170), (96,194), (180,223), (192,229), (239,245), (273,257);
(4.2) utilize C# programming realization to detect coordinate (X ' 0 of the accumulation of dam and the dried beach separation of each bar binaryzation line segment automatically ... X ' n, Y ' 0 ... Y ' n) for example doing beach and dam body separation coordinate is respectively: (14,265), (96,290), (180,317), (192,322), (239,336), (273,348);
(4.3) mark out each bar from the fill dam dam crest between the binaryzation line segment of doing the beach separation apart from K0 ... Kn for example marks dam crest and dried beach separation distance: K0=265-170=95, K1=290-194=96, K2=317-223=94, K3=322-229=93, K4=336-245=91, K5=348-257=91;
(4.4) utilize pixel detection technology to calculate the function corresponding relation of line segment between pixel and the extremely dried beach of the fill dam dam crest separation: x=H/k
X is the distance of a pixel representative, and H is the dried beach height of demarcating behind the field survey, and k is that the fill dam dam crest is to the sum of all pixels of doing the beach separation; For example the high H of the current heap of field survey is 240, then
x0＝H/k0＝240/95＝2.526；
x1＝H/k1＝240/96＝2.5；
x2＝H/k2＝240/94＝2.553；
x3＝H/k3＝240/93＝2.58；
x4＝H/k4＝240/91＝2.637；
(4.5) calculate dried beach height number: (Y0-Y ' 0) * x=H, utilize this formula can calculate the dry sand of tailings reservoir altitude information, will calculate in Y0 and Y ' the 0 substitution formula respectively, process is as follows:
H0＝(265-170)×2.526＝240；
H1＝(290-194)×2.5＝240；
H2＝(317-223)×2.553＝240；
H3＝(322-229)×2.58＝240；
H4＝(336-245)×2.637＝240；
H5＝(348-257)×2.637＝240。
If camera photographs new picture next time, calculate in the automatic substitution formula of the coordinate that identifies in the program, can conveniently calculate the beach heights of roofs.
(5) calculate dried beach length
With the resultant safe superelevation h of difference that does the measured reservoir level value of beach height H and ultrasonic liquid level gauge, the dried beach gradient J by field survey utilizes formula L=h/tanJ, and L is dried beach length.For example be through field survey dry sand of tailings reservoir slope: 0.004, doing the beach height H is 240 meters, safe superelevation h is 2 meters.Calculate in the substitution formula and do beach length L=2/0.004=500 rice.
Claims (1)
1. method based on the image recognition calculating length of dry sand of tailings reservoir is characterized in that:
(1) observe to do the beach image: do the beach image and be by camera the resulting image of the camera angle of dam body unanimity, should reach more than 704 * 576 to the pixel of doing the beach image;
(2) colour difference on clear and definite tailing dam accumulation of dam and dried beach is drawn some spots X0 on the tailing dam accumulation of dam ... Xn utilizes C# program coordinate figure, the color value information of the drawn point of record automatically, and through X0 ... Xn strokes and dots Zn bar straight line;
(3) determine threshold value, picture carried out binary conversion treatment:
Utilize the alternative manner selected threshold, the average gray T0 to Tn that initial threshold is chosen for image i.e. drawn X0 on dam body ... the color value of Xn, utilize T0 to Tn that the pixel branch on every straight line is made two parts then, calculate two parts average gray separately, part less than T0 is TA, part greater than T0 is TB, utilize the C# programming to realize the automatic detection computations of color value of each coordinate points, T1 is replaced T0 as new global threshold, repeat above process, so iteration detects to the image bottom until automatic;
(4) do the beach height with the image calculation after the binaryzation:
(4.1) mark out the fill dam dam crest point coordinate X0 of each bar binaryzation line segment ... Xn, Y0 ... Yn;
(4.2) realize detecting automatically the accumulation of dam of each bar binaryzation line segment and the coordinate X ' 0 of dried beach separation with the C# programming ... X ' n, Y ' 0 ... Y ' n;
(4.3) mark out each bar from the fill dam dam crest between the binaryzation line segment of doing the beach separation apart from K0 ... Kn;
(4.4) utilize pixel detection technology to calculate the function corresponding relation of line segment between pixel and the extremely dried beach of the fill dam dam crest separation: x=H/k
X is the distance of a pixel representative, and H is the dried beach height of demarcating behind the field survey, and k is that the fill dam dam crest is to the sum of all pixels of doing the beach separation;
(4.5) calculate dried beach height number: (Y0-Y ' 0) * x=H;
(5) calculate dried beach length
The dried beach gradient J that utilizes the dried beach height H obtained and field survey is by formula: L=tanJ * H
L is for doing beach length.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910248499A CN101718527A (en) | 2009-12-17 | 2009-12-17 | Method for calculating length of dry sand of tailings reservoir based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910248499A CN101718527A (en) | 2009-12-17 | 2009-12-17 | Method for calculating length of dry sand of tailings reservoir based on image recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101718527A true CN101718527A (en) | 2010-06-02 |
Family
ID=42433134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910248499A Pending CN101718527A (en) | 2009-12-17 | 2009-12-17 | Method for calculating length of dry sand of tailings reservoir based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101718527A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102589497A (en) * | 2012-02-29 | 2012-07-18 | 山东黄金矿业（玲珑）有限公司 | Method for measuring and calculating length of gold tailing pond dry beach in real time |
| CN102981517A (en) * | 2012-11-30 | 2013-03-20 | 中国有色金属长沙勘察设计研究院有限公司 | Dry beach safety detecting system |
| CN103196420A (en) * | 2013-03-01 | 2013-07-10 | 北京矿冶研究总院 | Method and system for measuring length of dry beach of tailing pond |
| CN103366543A (en) * | 2013-07-17 | 2013-10-23 | 深圳市粮食集团有限公司 | Loading control method and system based on detection of grain accumulation height |
| CN105319050A (en) * | 2015-09-10 | 2016-02-10 | 水利部交通运输部国家能源局南京水利科学研究院 | Test measuring system of riverbank lateral erosion collapse rate and measuring method thereof |
| CN110132200A (en) * | 2019-05-07 | 2019-08-16 | 四川安信科创科技有限公司 | Dry sand of tailings reservoir dynamic monitoring method and system based on Beidou and video identification |
-
2009
- 2009-12-17 CN CN200910248499A patent/CN101718527A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102589497A (en) * | 2012-02-29 | 2012-07-18 | 山东黄金矿业（玲珑）有限公司 | Method for measuring and calculating length of gold tailing pond dry beach in real time |
| CN102981517A (en) * | 2012-11-30 | 2013-03-20 | 中国有色金属长沙勘察设计研究院有限公司 | Dry beach safety detecting system |
| CN102981517B (en) * | 2012-11-30 | 2015-11-25 | 中国有色金属长沙勘察设计研究院有限公司 | Dry beach safety detecting system |
| CN103196420A (en) * | 2013-03-01 | 2013-07-10 | 北京矿冶研究总院 | Method and system for measuring length of dry beach of tailing pond |
| CN103196420B (en) * | 2013-03-01 | 2015-03-25 | 北京矿冶研究总院 | Method and system for measuring length of dry beach of tailing pond |
| CN103366543A (en) * | 2013-07-17 | 2013-10-23 | 深圳市粮食集团有限公司 | Loading control method and system based on detection of grain accumulation height |
| CN105319050A (en) * | 2015-09-10 | 2016-02-10 | 水利部交通运输部国家能源局南京水利科学研究院 | Test measuring system of riverbank lateral erosion collapse rate and measuring method thereof |
| CN105319050B (en) * | 2015-09-10 | 2017-12-15 | 水利部交通运输部国家能源局南京水利科学研究院 | The testing ＆ measuring system and its method for measurement of riverbank lateral erosion avalanche speed |
| CN110132200A (en) * | 2019-05-07 | 2019-08-16 | 四川安信科创科技有限公司 | Dry sand of tailings reservoir dynamic monitoring method and system based on Beidou and video identification |
| CN110132200B (en) * | 2019-05-07 | 2021-07-20 | 四川安信科创科技有限公司 | Tailing pond dry beach dynamic monitoring method and system based on Beidou and video identification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109764930B (en) | Water gauge water line visual detection method suitable for complex illumination conditions | |
| CN103499585B (en) | Based on noncontinuity lithium battery film defect inspection method and the device thereof of machine vision | |
| CN101718527A (en) | Method for calculating length of dry sand of tailings reservoir based on image recognition | |
| CN108921165B (en) | Water level identification method based on water gauge image | |
| CN107833203A (en) | A kind of horizontal plane identification and water level method for real-time measurement based on image procossing | |
| Hu et al. | A robust method for semi-automatic extraction of road centerlines using a piecewise parabolic model and least square template matching | |
| CN104183142A (en) | Traffic flow statistics method based on image visual processing technology | |
| CN107798293A (en) | A kind of crack on road detection means | |
| CN109376740A (en) | A kind of water gauge reading detection method based on video | |
| CN108470173B (en) | Ore particle segmentation method | |
| CN107727658A (en) | Pantograph crackle on-line measuring device and method based on image procossing | |
| CN104537651A (en) | Proportion detecting algorithm and system for cracks in road surface image | |
| CN111754460A (en) | Method, system and storage medium for automatically detecting gap of point switch | |
| Kumar et al. | An automated road roughness detection from mobile laser scanning data | |
| CN114067095A (en) | Water level identification method based on water gauge character detection and identification | |
| CN113392846A (en) | Water gauge water level monitoring method and system based on deep learning | |
| CN106599894A (en) | Method for identifying pole number of overhead line system based on image identification | |
| CN103852041A (en) | Method for online monitoring of dry beach length of tailing pond | |
| CN115690012A (en) | Detection method for wrong connection line of electric energy meter | |
| Zheng et al. | Rail detection based on LSD and the least square curve fitting | |
| CN108596968B (en) | Sleeper counting method based on track 3D depth image | |
| CN202562477U (en) | Device for detecting dangerous container stockpiling manner at container terminal storage yard | |
| CN107238837A (en) | A kind of drauht detection method | |
| CN105426845A (en) | Device and method for automatically identifying kinds of materials in material field | |
| CN113569943B (en) | Slag piece large-block early warning method, system and device based on deep neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20100602 |