CN102620048A - Image monitoring method for monitoring safety of valves in industrial boiler field - Google Patents
Image monitoring method for monitoring safety of valves in industrial boiler field Download PDFInfo
- Publication number
- CN102620048A CN102620048A CN2012101094098A CN201210109409A CN102620048A CN 102620048 A CN102620048 A CN 102620048A CN 2012101094098 A CN2012101094098 A CN 2012101094098A CN 201210109409 A CN201210109409 A CN 201210109409A CN 102620048 A CN102620048 A CN 102620048A
- Authority
- CN
- China
- Prior art keywords
- white
- black
- image
- valve
- picture
- 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
- 238000000034 method Methods 0.000 title claims abstract description 55
- 238000012544 monitoring process Methods 0.000 title claims abstract description 42
- 238000012545 processing Methods 0.000 claims abstract description 38
- 238000006243 chemical reaction Methods 0.000 claims description 34
- 230000009545 invasion Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 230000007423 decrease Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 description 8
- 241001465754 Metazoa Species 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 241000669298 Pseudaulacaspis pentagona Species 0.000 description 2
- 241000316887 Saissetia oleae Species 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 239000004744 fabric Substances 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000004922 lacquer Substances 0.000 description 1
- 239000003595 mist Substances 0.000 description 1
- 238000010422 painting Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Abstract
The invention provides an image monitoring method for monitoring safety of valves in an industrial boiler field, which comprises the following steps of S100, installing an embedded type industrial camera and/or an industrial video camera and performing pre-arranging processing to valve regions in the field; S200, adopting the industrial camera and/or the industrial video camera to continuously acquiring original images of the valve regions in real time; and S300, analyzing, processing and identifying acquired original images in a processing period according to preset threshold values and monitoring the safety of the valve regions. By means of the method, the safety of the valve regions of an industrial boiler can be monitored automatically, and persons can be liberated from a monitoring system.
Description
Technical field
The present invention relates to a kind of security monitoring field of industrial boiler valve site, particularly relate to a kind of image monitoring method of industrial boiler valve site safety.
Background technique
Valve is the on-the-spot important component part of industrial boiler always; To valve operation whether reasonable, whether have the inoperative personnel to get into valve region, whether have the behavior of operated valve etc., be directly connected to the on-the-spot safety of whole industrial boiler and the quality of product.
Simultaneously; Industrial boiler valve site zone also comprises various connected elements; In the production run process along with corrosion on Equipment, aging be easy to generate the space, leak, the existence of leakage is related to the on-the-spot safety of whole industrial boiler and the normal operation of equipment equally.Therefore, the action need of valve is strictly controlled, and does one's utmost to avoid the arbitrarily generation of operation.
For the illegal or unreasonable operation of the personnel of stopping to valve, the normal operation of support equipment, the on-the-spot valve region of industrial boiler need be installed the operational circumstances that real-time monitoring system is monitored valve.
The on-the-spot main monitor mode that adopts of boiler is divided into two kinds at present: simulation monitoring and digital supervision.
Under the simulation monitoring situation, to display device, the staff monitors through observation display the monitory point image through cable transmission.
Under the digital supervision situation, image data passes on the display device of high-resolution through after the Computer Processing, and the staff can observe guarded region more clearly.
Yet no matter be simulation monitoring or digital supervision, these conventional monitor modes all are image to be transmitted with simple handle, and do not analyze further.Therefore need the people before display device, to carry out continual observation, but owing to unsurmountable reasons such as personnel's fatigue can't realize not intermittently observation in 24 hours, this brings leak for security monitoring of valve.
Summary of the invention
The object of the present invention is to provide a kind of image monitoring method of industrial boiler valve site safety, it can monitor industrial boiler valve region safety automatically, and personnel are freed from supervisory system.
For realizing the image monitoring method of a kind of industrial boiler valve site safety that the object of the invention provides, comprise the steps:
Step S100 installs built-in industrial camera and/or industrial camera in the valve region that industrial boiler is on-the-spot, and the valve site zone is provided with processing in advance;
Step S200 adopts industrial camera and/or industrial camera to gather the original image of valve region in real time, continuously;
Step S300 according to preset threshold value, in the processing cycle, carries out analysing and processing identification to the original image that collects, and valve region safety is monitored.
More excellent ground, said step S200 also comprises the steps:
Step S200 ' shows the original image remote transmission to real-time monitoring client.
More excellent ground, said step S300 comprises the steps:
Steps A, is carried out optimal threshold successively to original image and is cut apart method of iteration binaryzation, thinning processing, Hough conversion and angle calculation in the cycle in collecting treatment, finally obtains the residing angle of valve, thereby realizes the monitoring to valve region.
More excellent ground, said steps A comprises the steps:
Steps A 1 is carried out optimal threshold to original image and is cut apart the method for iteration binary conversion treatment, and image is divided into two parts of black and white;
Promptly pass through analysis to the original image overall intensity,, in original image, confirm a preferred threshold point; Be divided into two parts of black and white to image according to optimum threshold point; Wherein the gray value of stain is 0, and the gray value of white point is 255, is convenient to follow-up identification and handles;
Steps A 2 is carried out thinning processing to the image after the binaryzation;
If p0 is an image slices vegetarian refreshments of treating thinning processing, with the number of non-0 pixel in the 8-neighborhood of n (p0) remarked pixel point p0, s (p0) expression is in the sequence of order with p1, p2, p3, p4, p5, p6, p7, p8, p1, and pixel value changes to 255 number of times from 0;
To the stain of each 8-centre of neighbourhood in the raw image data array,, just this point is changed to white point if said stain satisfies following four conditions:
1)?2≤n(p0)≤6;
2)?s(p0)=1;
3) p1^p3^p7=0 or a s (p1)!=1;
4) p1^p3^p5=0 or a s (p3)!=1.
Wherein, mark ^ representes " logical AND " computing.
Steps A 3, utilization Hough conversion with refinement after remaining all stains from the right angled coordinates spatial alternation to polar coordinate space, obtain the angle angle of drawn white straight line and X axle forward on the valve through the maximum of searching accumulator.
More excellent ground, said steps A 3 comprises the steps:
Steps A 31 with corresponding pixel coordinate of each stain of image space after the thinning processing, is equivalent to a point in right angled coordinates space, transforms to polar coordinate space and becomes a curve.With polar coordinate space (ρ θ) is divided into many junior units, and each junior unit is established an accumulator;
Steps A 32, image space decline every bit (x, y) corresponding polar coordinate space curve ρ=x*cos θ+y*sin θ point-blank; Make θ equal 0 respectively, △ θ, 2 △ θ;, obtain corresponding ρ value, and calculate the number of times that drops on each junior unit respectively.
Steps A 33; After all black pixel point transformation calculations finish, the maximum value in the statistics accumulator element, and obtain corresponding θ value; This is the angle angle of the normal and the X axle forward of drawn white straight line on the valve, thereby can obtain the angle angle of white line and X axle.
More excellent ground among the said step S100, is provided with processing to the valve site zone in advance, comprises the steps:
Gather no mobiles through the picture of valve region as reference base picture, and cut apart the method for iteration binary conversion treatment as optimal threshold, obtain black and white picture, calculate the black/white pixel ratio P1 of benchmark black and white picture then;
Among the said step S300, the original image that collects is carried out analysing and processing identification, valve region safety is monitored, comprise the steps:
Step B1, the picture of gathering valve region with in real time, continuously is picture as a comparison, and cuts apart the method for iteration binary conversion treatment as optimal threshold, obtains black and white picture, calculates the black/white pixel ratio P2 of contrast black and white picture then; Step B2 calculates the black/white pixel ratio P2 of contrast black and white picture and the difference P2-P1 of the black/white pixel ratio P1 of benchmark black and white picture, and invades threshold value Trq with mobiles and compare;
If step B3 is P2-P1>Trq, then think the mobiles invasion take place; Otherwise, the mobiles invasion does not take place.
More excellent ground among the said step S100, is provided with processing to the valve site zone in advance, comprises the steps:
Gather not have block valve region picture as reference base picture, and cut apart the method for iteration binary conversion treatment as optimal threshold, obtain black and white picture, calculate the white/black ratio X1 of benchmark black and white picture then;
Among the said step S300, the original image that collects is carried out analysing and processing identification, valve region safety is monitored, comprise the steps:
Step C1, the picture of gathering valve region with in real time, continuously is picture as a comparison, and cuts apart the method for iteration binary conversion treatment as optimal threshold, obtains black and white picture, calculates the white/black ratio X2 of contrast black and white picture then;
Step C2 calculates the difference X2-X1 of white/black pixel ratio X1 of white/black pixel ratio X2 and the benchmark black and white picture of contrast black and white picture, and invades threshold value Txl with personnel and compare;
If step C3 is X2-X1>Txl, then think the generation valve leak; Otherwise, valve leak does not take place.
Beneficial effect of the present invention: the present invention is the image monitoring method of industrial boiler valve site safety; The valve region on-the-spot with industrial boiler is concrete monitoring target; Through the image information that collects is carried out image analysis, basis for estimation is provided for preventing the valve misoperation; And detect the personnel invasion of valve region, for the valve operation scene provides safety guarantee; Can discern the leakage situation of valve region, for apparatus maintenance with prevent gaining time of serious accident.In addition, can also detect the work angle of valve automatically, for the related personnel provides reference to the subsequent operation of valve.The final purpose that realizes the 24 hours uninterrupted valve of monitoring automatically region securities.
Description of drawings
Fig. 1 is the image monitoring method flow chart of industrial boiler valve site safety;
Fig. 2 is the monitoring example schematic that embodiment one detects the valve angle;
Fig. 3 is a black/white scale drawing behind the valve region image binaryzation;
Fig. 4 is a black/white scale drawing behind the valve region image binaryzation.
Embodiment
In order to make the object of the invention, technological scheme and advantage clearer,, the realization of the image monitoring method of industrial boiler valve site safety of the present invention is further elaborated below in conjunction with accompanying drawing.Should be appreciated that specific embodiment described herein only in order to explanation the present invention, and be not used in qualification the present invention.
The image monitoring method of a kind of industrial boiler valve site safety of the embodiment of the invention, as shown in Figure 1, comprise the steps:
Step S100 installs built-in industrial camera and/or industrial camera in the valve region that industrial boiler is on-the-spot, and the valve site zone is provided with processing in advance.
The on-the-spot valve of industrial boiler generally is painted eye-catching redness, in order to detect the angle of valve, in the embodiment of the invention; But as a kind of mode of execution; Saidly be provided with in advance that to handle be lacquer painting straight line on valve,, generally straight line painted white for the ease of identification.
But as a kind of mode of execution, in the embodiment of the invention, industrial camera is a PA300EM type industrial camera, and industrial camera can adopt M5018-MP type industry pick-up lens;
But as a kind of mode of execution, the M5018-MP camera is installed on the industrial camera PA300EM.
Preferably, but as a kind of mode of execution, said valve is joystick type valve or handwheel type valve.
Step S200 adopts industrial camera and/or industrial camera to gather the original image of valve region in real time, continuously.
Preferably, but as a kind of mode of execution, industrial camera apart from 5 meters on valve over, the size of gathering image is set to the 400*400 pixel, the valve region size that image is contained is just in time suitable.
Preferably, but as a kind of mode of execution, said step S200 also comprises the steps:
Step S200 ' shows the original image remote transmission to real-time monitoring client.
Step S300 according to preset threshold value, in the processing cycle, carries out analysing and processing identification to the original image that collects, and valve region safety is monitored.
Preferably, but as a kind of mode of execution, after collecting the valve region original image, process binary conversion treatment, thinning processing, processes such as contrast calculating, Hough conversion are analyzed identification to image.Pass through the image monitoring method of the industrial boiler valve site safety of three embodiment's further explain embodiment of the invention below.
Embodiment one:
But as a kind of mode of execution, the image monitoring method of the industrial boiler valve site safety of the embodiment of the invention one comprises the steps:
Step S110, industrial camera and/or industrial camera are gathered the original image of valve region in real time, continuously;
Step S120, carries out optimal threshold successively to original image and cuts apart method of iteration binaryzation, thinning processing, Hough conversion and angle calculation in the cycle in collecting treatment, finally obtains the work angle of valve, thereby realizes the monitoring to valve region.
Because the residing angle of valve has shown the aperture state of valve, can confirm the rationality of valve operation thus, so that the harm of in time finding and preventing misoperation to bring.
Preferably, but as a kind of mode of execution, said step S120 comprises the steps:
Step S211 carries out optimal threshold to original image and cuts apart the method for iteration binary conversion treatment, and image is divided into two parts of black and white;
Promptly pass through analysis to the original image overall intensity,, in original image, confirm a preferred threshold point; Be divided into two parts of black and white to image according to optimum threshold point; Wherein the gray value of stain is 0, and the gray value of white point is 255, is convenient to follow-up identification and handles;
Step S212 carries out thinning processing to the image after the binaryzation;
If p0 is an image slices vegetarian refreshments of treating thinning processing, with the number of non-0 pixel in the 8-neighborhood of n (p0) remarked pixel point p0, s (p0) expression is in the sequence of order with p1, p2, p3, p4, p5, p6, p7, p8, p1, and pixel value changes to 255 number of times from 0.
To the stain of each 8-centre of neighbourhood in the raw image data array,, just this point is changed to white point if said stain satisfies following four conditions:
1)?2≤n(p0)≤6;
2)?s(p0)=1;
3) p1^p3^p7=0 or a s (p1)!=1;
4) p1^p3^p5=0 or a s (p3)!=1.
Wherein, mark ^ representes " logical AND " computing.
Step S213, utilization Hough conversion with refinement after remaining all stains from the right angled coordinates spatial alternation to polar coordinate space, obtain the angle angle of drawn white straight line and X axle forward on the valve through the maximum of searching accumulator.
Hough transform-based present principles shows, a some curve of corresponding polar coordinate space in right angled coordinates space, the N bar curve of concurrent in N the corresponding polar coordinate space of point on the straight line of right angled coordinates space.
In the embodiment of the invention, but as a kind of mode of execution, said step S213 comprises the steps:
Step S2131, corresponding pixel coordinate of each stain of image space after the thinning processing is equivalent to the point in right angled coordinates space, transforms to polar coordinate space and becomes a curve.With polar coordinate space (ρ θ) is divided into many junior units, and each junior unit is established an accumulator.
Step S2132, image space decline every bit (x, y) corresponding polar coordinate space curve ρ=x*cos θ+y*sin θ point-blank; Make θ equal 0 respectively, △ θ, 2 △ θ;, obtain corresponding ρ value, and calculate the number of times that drops on each junior unit respectively.
Step S2133; After all black pixel points (only needing to calculate stain) transformation calculations finishes; Maximum value in the statistics accumulator element; And obtain corresponding θ value, this is the angle angle of the normal and the X axle forward of drawn white straight line on the valve, thereby just can learn the angle angle of white line and X axle.
Illustrate the process of measuring the valve angle below.
Step 1.1: adjustment camera direction, the image of the 400*400 that makes it to gather size is a picture centre with the valve center, shown in Fig. 2 (A);
Step 1.2: image is carried out optimal threshold cut apart the iteration binary conversion treatment, obtain image, shown in Fig. 2 (B);
Step 1.3: image is carried out thinning processing, obtain image, shown in Fig. 2 (C);
Step 1.4: for the convenience calculated with quick; Need to extract necessary angle identifying information; Promptly be that the certain length of side of center intercepting (decide by choosing according to the length of institute's pencilling of this length of side with the picture centre; Here elect 56 pixels as) square area as the object images of hough conversion, obtain image, shown in Fig. 2 (D);
Step 1.5: be equally divided into 720 parts at the θ coordinate direction with 360 ℃, 0.5 ℃ of each part correspondence; Each stain to obtaining on the image 2 (4) is done the hough conversion, and statistics polar coordinate space accumulator maximum value obtains its corresponding θ coordinate, and recognition result is 63.0 ℃, thereby the angle of white line and X axle forward is 90-63.0=27.0 ℃.
Test shows that the present invention measures the valve angle precision can reach 0.5 ℃.
Embodiment two:
But as a kind of mode of execution, the image monitoring method of the industrial boiler valve site safety of the embodiment of the invention two comprises the steps:
Step S120; The picture of gathering no mobiles (mobiles that comprises people, animal or other absence of vital signs) process valve region is as reference base picture; And cut apart the method for iteration binary conversion treatment as optimal threshold; Obtain black and white picture, calculate the black/white pixel ratio P1 of benchmark black and white picture then;
Step S121, the picture of gathering valve region with in real time, continuously is picture as a comparison, and cuts apart the method for iteration binary conversion treatment as optimal threshold, obtains black and white picture, calculates the black/white pixel ratio P2 of contrast black and white picture then; Step S122 calculates the black/white pixel ratio P2 of contrast black and white picture and the difference P2-P1 of the black/white pixel ratio P1 of benchmark black and white picture, and invades threshold value Trq with mobiles and compare.If P2-P1>Trq, then think the mobiles invasion take place.
Illustrate mobiles invasion recognition process below.
Step 2.1: confirm to differentiate the threshold value Trq whether the generation personnel invade.
Specific practice is: every separated 30s gathers the valve region image of width of cloth 400*400 size and cuts apart the method for iteration binary conversion treatment, the black/white pixel ratio P of image after the computing as optimal threshold.
24 hours collecting treatment of whole day are got off, and obtain 2880 data, and drawing with matlab, the result is as shown in Figure 3 in the back.As can beappreciated from fig. 3, the P value of gathering image during through valve region when no mobiles (mobiles that comprises people, animal or other absence of vital signs) does not fluctuate or fluctuates very little, in case personnel's process is arranged, the P value has significant increase.
Whether therefore can differentiate according to the variation of P value has personnel to invade.According to result shown in Figure 3, the threshold value Trq that differentiation personnel invasion is set is 0.3.
Step 2.2: the picture of gathering no mobiles (mobiles that comprises people, animal or other absence of vital signs) process valve region is as benchmark image; Size is 400*400; Image is carried out optimal threshold cut apart the method for iteration binary conversion treatment; Obtain binary image, calculate the black/white pixel ratio P1 of image;
Step 2.3: gather the valve region image of 400*400 size in real time and cut apart the method for iteration binary conversion treatment, the black/white pixel ratio P2 of image after the computing as optimal threshold;
Step 2.4: calculate P2-P1; If the result is greater than threshold value Trq; Wait for 2 seconds; Again gather the valve region image and make binary conversion treatment, the value of the black/white pixel ratio P2 of image and P2-P1 after the computing (purpose of waiting for 2 seconds is to get rid of the interference that has the people through this situation of valve region identification to be caused);
Step 2.5: compare the value of P2-P1 and the size of threshold value Trq once more.If P2-P1>Trq, show personnel's invasion has taken place; If ≤Trq shows the mobiles invasion does not take place P2-P1.
Embodiment three:
Step S130, gather not have block valve region picture as reference base picture, and cut apart the method for iteration binary conversion treatment as optimal threshold, obtain black and white picture, calculate the white/black ratio X1 of benchmark black and white picture then;
Step S131, the picture of gathering valve region with in real time, continuously is picture as a comparison, and cuts apart the method for iteration binary conversion treatment as optimal threshold, obtains black and white picture, calculates the white/black ratio X2 of contrast black and white picture then;
Step S132 calculates the difference X2-X1 of white/black pixel ratio X1 of white/black pixel ratio X2 and the benchmark black and white picture of contrast black and white picture, and invades threshold value Txl with personnel and compare;
If step S133 is X2-X1>Txl, then think the generation valve leak; Otherwise, valve leak does not take place.
Illustrate the diagnostic imaging process that valve region is leaked below:
Step 3.1: confirm whether differentiation the threshold value Txl of valve leak takes place.
Specific practice is: every separated 30s gathers the valve region image of width of cloth 400*400 size and cuts apart the method for iteration binary conversion treatment, the white/black pixel ratio X of image after the computing as optimal threshold.24 hours collecting treatment of whole day are got off, and obtain 2880 data, and drawing with matlab, the result is as shown in Figure 4 in the back.As can beappreciated from fig. 4, do not have the X value of gathering image when blocking when valve region and do not fluctuate or fluctuate very little, in case valve leak (simulating with the white mist of humidifier manufacturing under the situation of laboratory) takes place, the X value has significant increase.
Therefore can judge whether to take place valve leak according to the variation of X value.According to Fig. 3 result, it is 10 that the threshold value Txl that differentiates valve leak is set.
Step 3.2: gather not have block valve region picture as benchmark image, size is 400*400, image is carried out optimal threshold cut apart the method for iteration binary conversion treatment and obtain binary image, calculates the white/black pixel ratio X1 of image;
Step 3.3: gather the valve region image of 400*400 size in real time and cut apart the method for iteration binary conversion treatment, the white/black pixel ratio X2 of image after the computing as optimal threshold;
Step 3.4: calculate X2-X1,, gather the valve region image again and make binary conversion treatment, the value of the white/black pixel ratio X2 of image and X2-X1 after the computing (waiting for 2 seconds here also is in order to get rid of interference) if the result greater than threshold value Txl, waits for 2 seconds;
Step 3.5: compare the value of X2-X1 and the size of threshold value Txl once more.
Step 3.6: if X2-X1>Txl, show the mobiles invasion has taken place; If ≤Txl shows the mobiles invasion does not take place X2-X1.
In the embodiment of the invention, at first control camera and/or industrial camera and gather the valve region image, then image is carried out analysing and processing, image after the processing and signal can send through network interface, give and carry out control processing.
The image monitoring method of the industrial boiler valve site safety of the embodiment of the invention, the valve region on-the-spot with industrial boiler is concrete monitoring target, through the image information that collects is carried out image analysis, to preventing the valve misoperation basis for estimation is provided; And detect the personnel invasion of valve region, for the valve operation scene provides safety guarantee; Can discern the leakage situation of valve region, for apparatus maintenance with prevent gaining time of serious accident.Realize the purpose of the 24 hours uninterrupted valve of monitoring automatically region securities.
Should be noted that at last that obviously those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these revise and modification belongs within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and modification.
Claims (7)
1. the image monitoring method of an industrial boiler valve site safety is characterized in that, comprises the steps:
Step S100 installs built-in industrial camera and/or industrial camera in the valve region that industrial boiler is on-the-spot, and the valve site zone is provided with processing in advance;
Step S200 adopts industrial camera and/or industrial camera to gather the original image of valve region in real time, continuously;
Step S300 according to preset threshold value, in the processing cycle, carries out analysing and processing identification to the original image that collects, and valve region safety is monitored.
2. image monitoring method according to claim 1 is characterized in that said step S200 also comprises the steps:
Step S200 ' shows the original image remote transmission to real-time monitoring client.
3. image monitoring method according to claim 1 and 2 is characterized in that said step S300 comprises the steps:
Steps A, is carried out optimal threshold successively to original image and is cut apart method of iteration binaryzation, thinning processing, Hough conversion and angle calculation in the cycle in collecting treatment, finally obtains the residing angle of valve, thereby realizes the monitoring to valve region.
4. image monitoring method according to claim 3 is characterized in that said steps A comprises the steps:
Steps A 1 is carried out optimal threshold to original image and is cut apart the method for iteration binary conversion treatment, and image is divided into two parts of black and white;
Promptly through analysis to the original image overall intensity; In original image, confirm a preferred threshold point, be divided into two parts of black and white to image according to optimum threshold point, wherein the gray value of stain is 0; The gray value of white point is 255, is convenient to follow-up identification and handles;
Steps A 2 is carried out thinning processing to the image after the binaryzation;
If p0 is an image slices vegetarian refreshments of treating thinning processing, with the number of non-0 pixel in the 8-neighborhood of n (p0) remarked pixel point p0, s (p0) expression is in the sequence of order with p1, p2, p3, p4, p5, p6, p7, p8, p1, and pixel value changes to 255 number of times from 0;
To the stain of each 8-centre of neighbourhood in the raw image data array,, just this point is changed to white point if said stain satisfies following four conditions:
1)?2≤n(p0)≤6;
2)?s(p0)=1;
3) p1^p3^p7=0 or a s (p1)!=1;
4) p1^p3^p5=0 or a s (p3)!=1;
Wherein, mark ^ representes " logical AND " computing;
Steps A 3, utilization Hough conversion with refinement after remaining all stains from the right angled coordinates spatial alternation to polar coordinate space, obtain the angle angle of drawn white straight line and X axle forward on the valve through the maximum of searching accumulator.
5. image monitoring method according to claim 4 is characterized in that, said steps A 3 comprises the steps:
Steps A 31 with corresponding pixel coordinate of each stain of image space after the thinning processing, is equivalent to a point in right angled coordinates space, transforms to polar coordinate space and becomes a curve;
With polar coordinate space (ρ θ) is divided into many junior units, and each junior unit is established an accumulator;
Steps A 32, image space decline every bit (x, y) corresponding polar coordinate space curve ρ=x*cos θ+y*sin θ point-blank; Make θ equal 0 respectively, △ θ, 2 △ θ;, obtain corresponding ρ value, and calculate the number of times that drops on each junior unit respectively;
Steps A 33; After all black pixel point transformation calculations finish, the maximum value in the statistics accumulator element, and obtain corresponding θ value; This is the angle angle of the normal and the X axle forward of drawn white straight line on the valve, thereby just can learn the angle angle of white line and X axle.
6. image monitoring method according to claim 1 and 2 is characterized in that:
Among the said step S100, the valve site zone is provided with processing in advance, comprises the steps:
Gather no mobiles through the picture of valve region as reference base picture, and cut apart the method for iteration binary conversion treatment as optimal threshold, obtain black and white picture, calculate the black/white pixel ratio P1 of benchmark black and white picture then;
Among the said step S300, the original image that collects is carried out analysing and processing identification, valve region safety is monitored, comprise the steps:
Step B1, the picture of gathering valve region with in real time, continuously is picture as a comparison, and cuts apart the method for iteration binary conversion treatment as optimal threshold, obtains black and white picture, calculates the black/white pixel ratio P2 of contrast black and white picture then; Step B2 calculates the black/white pixel ratio P2 of contrast black and white picture and the difference P2-P1 of the black/white pixel ratio P1 of benchmark black and white picture, and invades threshold value Trq with mobiles and compare;
If step B3 is P2-P1>Trq, then think the mobiles invasion take place; Otherwise, the mobiles invasion does not take place.
7. image monitoring method according to claim 1 and 2 is characterized in that:
Among the said step S100, the valve site zone is provided with processing in advance, comprises the steps:
Gather not have block valve region picture as reference base picture, and cut apart the method for iteration binary conversion treatment as optimal threshold, obtain black and white picture, calculate the white/black ratio X1 of benchmark black and white picture then;
Among the said step S300, the original image that collects is carried out analysing and processing identification, valve region safety is monitored, comprise the steps:
Step C1, the picture of gathering valve region with in real time, continuously is picture as a comparison, and cuts apart the method for iteration binary conversion treatment as optimal threshold, obtains black and white picture, calculates the white/black ratio X2 of contrast black and white picture then;
Step C2 calculates the difference X2-X1 of white/black pixel ratio X1 of white/black pixel ratio X2 and the benchmark black and white picture of contrast black and white picture, and invades threshold value Txl with personnel and compare;
If step C3 is X2-X1>Txl, then think the generation valve leak; Otherwise, valve leak does not take place.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201210109409 CN102620048B (en) | 2012-04-13 | 2012-04-13 | Image monitoring method for monitoring safety of valves in industrial boiler field |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201210109409 CN102620048B (en) | 2012-04-13 | 2012-04-13 | Image monitoring method for monitoring safety of valves in industrial boiler field |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102620048A true CN102620048A (en) | 2012-08-01 |
| CN102620048B CN102620048B (en) | 2013-10-30 |
Family
ID=46560164
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201210109409 Expired - Fee Related CN102620048B (en) | 2012-04-13 | 2012-04-13 | Image monitoring method for monitoring safety of valves in industrial boiler field |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102620048B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102797727A (en) * | 2012-08-17 | 2012-11-28 | 国电联合动力技术有限公司 | Method and device for detecting oil leakage of hydraulic system of wind turbine based on CCD (Charge Coupled Device) |
| CN104251687A (en) * | 2014-10-11 | 2014-12-31 | 盐城工学院 | Part surface evenness detection method based on mirror image processing |
| CN106814661A (en) * | 2015-12-01 | 2017-06-09 | 佳霖科技股份有限公司 | Machine table monitoring interlocking system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4949288A (en) * | 1988-09-02 | 1990-08-14 | Movats Incorporated | Testing system for safety relief valves |
| CN101281079A (en) * | 2008-05-12 | 2008-10-08 | 北京理工大学 | Method and system for detecting and positioning leakage based on infrared imaging technique |
| CN101409004A (en) * | 2008-11-24 | 2009-04-15 | 浙江大学 | Safety defense monitoring method based on Symbian intelligent mobile phone platform |
| CN101840219A (en) * | 2009-09-30 | 2010-09-22 | 浙江大学 | Wetland park oriented security precaution integrated system and method |
-
2012
- 2012-04-13 CN CN 201210109409 patent/CN102620048B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4949288A (en) * | 1988-09-02 | 1990-08-14 | Movats Incorporated | Testing system for safety relief valves |
| CN101281079A (en) * | 2008-05-12 | 2008-10-08 | 北京理工大学 | Method and system for detecting and positioning leakage based on infrared imaging technique |
| CN101409004A (en) * | 2008-11-24 | 2009-04-15 | 浙江大学 | Safety defense monitoring method based on Symbian intelligent mobile phone platform |
| CN101840219A (en) * | 2009-09-30 | 2010-09-22 | 浙江大学 | Wetland park oriented security precaution integrated system and method |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102797727A (en) * | 2012-08-17 | 2012-11-28 | 国电联合动力技术有限公司 | Method and device for detecting oil leakage of hydraulic system of wind turbine based on CCD (Charge Coupled Device) |
| CN102797727B (en) * | 2012-08-17 | 2015-11-11 | 国电联合动力技术有限公司 | A kind of Wind turbines oil leakage of hydraulic system detecting method based on CCD and device |
| CN104251687A (en) * | 2014-10-11 | 2014-12-31 | 盐城工学院 | Part surface evenness detection method based on mirror image processing |
| CN104251687B (en) * | 2014-10-11 | 2017-05-03 | 盐城工学院 | Part surface evenness detection method based on mirror image processing |
| CN106814661A (en) * | 2015-12-01 | 2017-06-09 | 佳霖科技股份有限公司 | Machine table monitoring interlocking system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102620048B (en) | 2013-10-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101957325B (en) | Substation equipment appearance abnormality recognition method based on substation inspection robot | |
| US11906445B2 (en) | Automated defect detection for wire rope using image processing techniques | |
| CN105491349A (en) | Belt group abnormal state automatic touring detection system and detection method thereof | |
| CN104535356A (en) | Method and system for monitoring rope arrangement faults of drum steel wire rope on basis of machine vision | |
| CN104992270A (en) | Power transmission and transformation equipment state maintenance aid decision making system and method | |
| CN110378897A (en) | A kind of pantograph running state real-time monitoring method and device based on video | |
| CN111289529B (en) | Conveying belt tearing detection system and detection method based on AI intelligent analysis | |
| CN103996204A (en) | Analyzing system, analyzing program and analyzing method | |
| CN102620048B (en) | Image monitoring method for monitoring safety of valves in industrial boiler field | |
| CN111126196A (en) | Equipment oil leakage detection method | |
| WO2021020866A9 (en) | Image analysis system and method for remote monitoring | |
| CN115171361B (en) | Dangerous behavior intelligent detection and early warning method based on computer vision | |
| CN116152202A (en) | Equipment appearance detection system based on image recognition technology and infrared thermal imaging technology | |
| CN116366002A (en) | Intelligent operation and maintenance system and method for photovoltaic power station | |
| KR20190081691A (en) | System and Method for Abnormality Monitoring by Using Machine Learning Algorithm | |
| CN111812114A (en) | On-line monitoring method and monitoring system for oil smoke pipeline | |
| CN116840240A (en) | Visual detection system of power supply distributor | |
| CN113592916B (en) | Sintering machine trolley axle fault detection method and system | |
| CN109271828A (en) | The method and system of construction segregator barriers condition intelligent detection based on deep learning | |
| KR102265461B1 (en) | Abnormal Data Detection System in Manufacturing Process | |
| CN104794702B (en) | A kind of based on the image blurring detection method of electric power video under the special scenes of image complexity analytical approach | |
| CN116449761B (en) | Intelligent pump station intelligent control system and method | |
| CN111222428B (en) | Detection method for device fastening operation based on monitoring video | |
| KR102519988B1 (en) | Monitoring system for accident worker using camera | |
| CN216253056U (en) | Hydraulic station monitoring system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20131030 |