CN105956618A - Converter steelmaking blowing state recognition system and method based on image dynamic and static characteristics - Google Patents
Converter steelmaking blowing state recognition system and method based on image dynamic and static characteristics Download PDFInfo
- Publication number
- CN105956618A CN105956618A CN201610272037.9A CN201610272037A CN105956618A CN 105956618 A CN105956618 A CN 105956618A CN 201610272037 A CN201610272037 A CN 201610272037A CN 105956618 A CN105956618 A CN 105956618A
- Authority
- CN
- China
- Prior art keywords
- image
- flame
- pixel
- module
- information
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a converter steelmaking blowing state recognition system and a method based on image dynamic and static characteristics, and the system comprises an image acquisition module, a color image flame region segmentation module, a color flame image dynamic and static characteristics expression and description module, a blowing state classification module based on image characteristics and a host computer monitoring module. The method comprises the steps: an acquired original image frame is inputted through a video, pixels which are similar to flame are measured and reserved, and segmentation is performed; four kinds of image characteristics including brightness characteristics, chroma characteristics, texture characteristics and optical flow field dynamic texture characteristics are extracted separately, classification and recognition are performed on the inputted characteristics through a generalized regression neural network, so that blowing state judgment and an end point prediction function are reached. According to the invention, the system and the method can adapt a characteristic of short stable state moment belonged to flame, and has high recognition precision and low cost, the system and the method can be used for steel-making converters in different sizes, the production efficiency can be improved, and the raw material waste can be reduced.
Description
Technical field
The present invention relates to a kind of pneumatic steelmaking blowing state recognition system and method moving static nature based on image, belong to
Metallurgical automation technology field.
Background technology
Steel and iron industry is the important raw and processed materials industry supporting the national economic development, and China is that iron and steel maximum in the world produces
State.In the total output of steel, the average quantum of converter steel yield reaches 70%.Terminal point control is a crucial behaviour in converter later stage
Making, refer to control the phosphorus content of molten steel and temperature reaches the requirement of tapping, the most real-time bessemerizes endpoint always
One of difficult problem of steel and iron industry, accurate forecast terminal is to improving steel mill's production efficiency, reducing the energy and waste of raw materials, raising steel
Material measurer is significant.
Carry out to about 85% in master's stage of blowing of blowing, adjust two by the method measuring carbon content and temperature value and blow plan
Slightly, using sublance detection and micro-judgment is most commonly seen data detection method, and sublance carries out thermometric and takes in immersing molten bath
Carbon, or master worker's visually sampling of falling stove detection, according to the data regulation blowing oxygen quantity measured and the addition of converter raw material.
For the endpoint in later stage of blowing, the situation of change that some iron and steel enterprises use iraser to occur through furnace gas judges end
The photodetection method of point, and by measuring the eudiometry etc. of furnace gas chemical composition, the detection equipment used by these methods
Want for a long time high temperature, corrosion environment in work, the Demarcate Gas cycle is short, and sampling head is changed frequently, the operation and maintenance of equipment
Relatively costly, it is difficult to promote the use of in steelmaking converter industry.
Sublance control has higher accuracy of detection, but is generally used in the converter of more than 120t, it is difficult to meet China
Based on the present situation of medium and small steel mill, and can not realize measuring continuously.Intelligence terminal point determining method using molten steel detection data as
Mode input vector, with target molten steel composition, temperature and blowing oxygen quantity as output vector, sets up based on moulds such as ELM, CBR, GMDH
The prediction system of cooking end point of type, intelligence End-point Prediction method mostly utilizes the blowing data of actual acquisition in steelmaking process, from principle
On there is preferable real-time, but existing method all exists for obtaining accurate data and causing cost and increase or data acquisition
The problem of aspect such as not in time.
Along with developing rapidly and the raising of computer disposal performance of digital image processing techniques, manually to see that fire is as base
The image recognition of plinth judges to have obtained rapid development and application for converter terminal.The most existing based on gray level co-occurrence matrixes
Method, based on color averaging method etc..Existing method achieves certain effect in terms of describing flame static nature, but have ignored
The multidate information of flame, and the actual experience seeing fire also indicates that, and flame dynamic features can embody the play of blowing different phase
Earthquake intensity, can be as the reflection of oxidation of coal speed.Therefore, merge the static state of flame and behavioral characteristics and then realize blowing end point
Accurately identify to improve production efficiency and reduce waste of raw materials have important real value and meaning.
Summary of the invention
In order to overcome manually see fire judge terminal bring inaccurate, re-blow often, efficiency is low, and waste of raw materials etc. is asked
Topic, the present invention provides a kind of pneumatic steelmaking blowing state recognition system and method moving static nature based on image, and avoids
Use the equipment manufacturing cost such as flue gas analysis, photoanalysis high, and traditional flame image recognition does not accounts for dynamic characteristic and affects
The problems such as discrimination.
The present invention is realized by following technical proposal: a kind of pneumatic steelmaking blowing state knowledge moving static nature based on image
Other system, including image capture module, coloured image flame region segmentation module, colored flames image sound state character representation with
Describing module, blowing state classification module based on characteristics of image and ipc monitor module:
Image capture module is used for captured in real-time flame conditions, and the video signal of shooting is stored and transmitted to cromogram
As flame region splits module；
Coloured image flame region segmentation module is for being divided into single color image frames by the video signal of collection, right
Flame background part in coloured image is split with flame body part；
Colored flames image sound state character representation and describing module are for gathering the bright of segmentation gained flame body part
Degree information, chrominance information, texture information and optical flow field multidate information；
Blowing state classification module based on characteristics of image is for by colored flames image sound state character representation and description
The various information that module is gathered is set up as input and is identified model, and after arranging output state, gathers data sample set pair institute
Build identification model to be trained；And then the new input content gathered is judged output state, blowing is entered the output shape in latter stage
State sends to ipc monitor module；
Ipc monitor module is used for sending signal prompt.
Described image capture module also includes photographic head, memorizer and signal transmission device.
Another object of the present invention is to provide a kind of pneumatic steelmaking blowing state recognition moving static nature based on image
Method, through the following step:
(1) color video camera is installed so that it is keep consistent with the position manually seeing fire and angle, the position of fixing camera
Put, the parameter such as focal length；The purpose of fixing camera be to ensure that the coloured image of picked-up interframe have feature comparability and
The accuracy that behavioral characteristics is described；Image size is set as suitable with human eye perception, and the too small effect characteristics of image extracts
Effect, image is excessive affects algorithm process speed；The real time video signals that color video camera is gathered passes through data acquisition card
It is connected with datatron；
(2) video signal of collection is divided into single color image frames by datatron, to the flame background in coloured image
Split with flame body, will change to uniform L*a*b* space by RGB color space, after note red (255,0,0) conversion be
RLab, yellow (255,255,0) change after into YLab, white (255,255,255) change after into WLabIf P is in converted images
Individual pixel, D is the Euclidean distance between pixel, according to formula
Then segmentation strategy is:
In formula, DPIRefer to the color value of chromatism between one pixel of the image after color space conversion and reference point pixel；I
Referring to an image pixel after RGB color is transformed into Lab color, this pixel has three components, is L respectively, a, b；T is
Control the threshold value of similarity；
Again willThe part of ＜ T is as flame body, and other parts are split as flame background；
(3) it is used for gathering monochrome information, chrominance information, texture information, Yi Jiguang by step (2) gained flame body part
Flow field multidate information:
1. collection monochrome information B:
If (i, j) for the flame region pixel after segmentation, then brightness B=(∑ I (i, j))/(Count), wherein B for I
For brightness, Count is area pixel number, and i, j are the position of pixel, coordinate the most in the picture；
2. collection chrominance information S:
The color third moment taking each component in rgb space can effectively reflect the oxidation order of blowing molten bath in period element, single
Third moment under component i is
In formula, pi,jThe probability occurred for the pixel that gray scale in coloured image i-th Color Channel component is j, i is for representing
The i-th component of coloured image, j is the gray value representing image, and si is the color moment feature of image, and μ i is relevant parameter；
3. collection texture information ASM:
Grey scale difference statistical method is utilized to describe static Texture complication, if (x is y) a bit in image, with point (x+
Gray scale difference value between Δ x, y+ Δ y) is gΔ((x, y) (x+ Δ x, y+ Δ y), if the institute of grey scale difference is likely for-g for x, y)=g
Value is m level, make point (x, y) to flame image region in move, accumulative gΔ(x y) takes the number of times of different value, makes gΔ
(x, rectangular histogram y), from rectangular histogram, gΔ(x, y) probability of value is pΔ(i)；
Use the uniformity coefficient extracting angle direction second moment ASM reflection gradation of image distribution, if close pixel ash
Angle value differs greatly, then ASM value is the biggest, illustrates that texture is the most coarse；
Textural characteristics is relevant with the degree of oxidation of chemical element in molten bath, and the most violent grain roughness that burns is the highest；
4. collection optical flow field multidate information Ent:
In order to describe the flame stroboscopic feature of converting process Flame multidate information, particularly later stage, it is necessary first to set up
The descriptive model that flame dynamically changes, uses optical flow field to set up scitillation process, is small fortune according to the basic assumption of optical flow computation
Moving and brightness constancy, (point constant for pixel intensity is existed between consecutive frame by x, y, t)=I (x+dx, y+dy, t+dt) to obtain I
Limit region intraconnections, construct interframe light flow graph F；F have recorded the flame violent degree in consecutive variations process, and F is carried out spy
Levy expression and description can reflect the information such as stroboscopic；
Wherein, x, y refer to the coordinate at image midpoint, and t is to be the time shaft of consecutive image, this is because video is by continuously
Image construction on time shaft, dx represents the displacement of x-axis, and dy represents the displacement of y-axis, and dt represents the displacement of t axle, and I is video
Stream is the picture frame of t when the time；
To interframe light flow graph, using gray level co-occurrence matrixes to describe its feature, during it calculates, direction takes 0 °, 45 °, 90 °,
135 °, step-length step=1 is to adapt to the feature of the random microtexture of flame, if M is the gray level co-occurrence matrixes obtained, uses entropy
Describe its feature, calculateMultidate information for flame；
Wherein, the coordinate of pixel during i, j are image；
(4) use generalized regression nerve networks to set up and identify model, existing characteristic vector V=that step (3) gathers [B, S,
ASM, Ent] for inputting, with the state in which numeral that blows for output: " 1 " represents the initial stage, and " 2 " represent mid-term, and " 3 " represent end
Phase；The generalized regression nerve networks set up is four inputs and single output, and hidden layer neuron excitation function is Gaussian function, output
For linear mapping function；By monochrome information, chrominance information, texture information and the optical flow field multidate information of known flame body
Export with it that data are corresponding to be listed, set up set of data samples；It is trained, by the output data in training building identification model
Sort out by status number；
(5) real time video signals step (1) gathered is after step (2) and (3) process, and inputs as characteristic vector
In identification model after step (4) training, its output is the identification of blowing state, when the State-output numeral that blows is 3, and generation
Table blowing enters latter stage, sends signal prompt.
The point of described step (3) (x+ Δ x, y+ Δ y) be value pixel (x, y) point around, this point and value pixel (x,
Y) distance is Δ x and Δ y.
Beneficial effects of the present invention: the present invention is directed to existing based on flame image identification bessemerize state identification method
Deficiency, propose merge sound state characteristics of image state recognition scheme；It is particularly well-suited to below 120t converter, or because of limited
Use the test environment such as flue gas, sublance.The present invention is advantageous in that and is extracted flame dynamic features and combines static nature
The identification of row blowing state, it is possible to adapt to the of short duration stable state flash problem that has of flame, have higher accuracy of identification with low become
This advantage, can be used for the steelmaking converter of different scales, improves production efficiency and reduces waste of raw materials.The present invention solves existing
There is terminal point determining in convertor steelmaking process inaccurate, and use the methods such as flue gas analysis, photoanalysis, sublance analysis to have
The problem that operation and maintenance cost is high, solves single static graphical analysis in existing employing flame image recognition methods simultaneously,
Existing characteristics changes greatly, the problem that accuracy rate is the highest.
Accompanying drawing explanation
Fig. 1 is the structural representation that the present invention moves the pneumatic steelmaking blowing state recognition system of static nature based on image；
Fig. 2 is the schematic flow sheet that the present invention moves the pneumatic steelmaking blowing state identification method of static nature based on image.
Specific embodiments
Embodiments of the invention are described more fully below, and the example of described embodiment is shown in the drawings.Retouch with reference to accompanying drawing
The embodiment stated is exemplary, is only used for explaining the present invention, and is not construed as limiting the claims.
As it is shown in figure 1, move the pneumatic steelmaking blowing state recognition system of static nature based on image, including image acquisition mould
Block 1, coloured image flame region split module 2, colored flames image sound state character representation and describing module 3, based on image
The blowing state classification module 4 of feature and ipc monitor module 5:
Image capture module 1 is for captured in real-time flame conditions, and the video signal of shooting is stored and transmitted to colored
Image flame region segmentation module；Image capture module also includes photographic head, memorizer and signal transmission device；
Coloured image flame region segmentation module 2 is used for the video signal of collection is divided into single color image frames,
Flame background part in coloured image is split with flame body part；
Colored flames image sound state character representation and describing module 3 are for gathering the bright of segmentation gained flame body part
Degree information, chrominance information, texture information and optical flow field multidate information；
Blowing state classification module 4 based on characteristics of image is for by colored flames image sound state character representation and description
The various information that module is gathered is set up as input and is identified model, and after arranging output state, gathers data sample set pair institute
Build identification model to be trained；And then the new input content gathered is judged output state, blowing is entered the output shape in latter stage
State sends to ipc monitor module；
Ipc monitor module 5 is used for sending signal prompt.
Such as Fig. 2, move the pneumatic steelmaking blowing state identification method of static nature based on image, through the following step:
(1) color video camera is installed so that it is keep consistent with the position manually seeing fire and angle, the position of fixing camera
Put, the parameter such as focal length；The purpose of fixing camera be to ensure that the coloured image of picked-up interframe have feature comparability and
The accuracy that behavioral characteristics is described；Image size is set as suitable with human eye perception, and the too small effect characteristics of image extracts
Effect, image is excessive affects algorithm process speed；The real time video signals that color video camera is gathered passes through data acquisition card
It is connected with datatron；
(2) video signal of collection is divided into single color image frames by datatron, to the flame background in coloured image with
Flame body is split, and will change to uniform L*a*b* space by RGB color space, after note red (255,0,0) conversion is
RLab, yellow (255,255,0) change after into YLab, white (255,255,255) change after into WLabIf P is one in converted images
Pixel, D is the Euclidean distance between pixel, according to formulaThen
Segmentation strategy is:
In formula, DPIRefer to the color value of chromatism between one pixel of the image after color space conversion and reference point pixel；I
Referring to an image pixel after RGB color is transformed into Lab color, this pixel has three components, is L respectively, a, b；T is
Control the threshold value of similarity；
Again willThe part of ＜ T is as flame body, and other parts are split as flame background；
(3) it is used for gathering monochrome information, chrominance information, texture information, Yi Jiguang by step (2) gained flame body part
Flow field multidate information:
1. collection monochrome information B:
If (i, j) for the flame region pixel after segmentation, then brightness B=(∑ I (i, j))/(Count), wherein B for I
For brightness, Count is area pixel number, and i, j are the position of pixel, coordinate the most in the picture；
2. collection chrominance information S:
The color third moment taking each component in rgb space can effectively reflect the oxidation order of blowing molten bath in period element, single
Third moment under component i is
In formula, pi,jThe probability occurred for the pixel that gray scale in coloured image i-th Color Channel component is j, i is for representing
The i-th component of coloured image, j is the gray value representing image, and si is the color moment feature of image, and μ i is relevant parameter；
3. collection texture information ASM:
Grey scale difference statistical method is utilized to describe static Texture complication, if (x is y) a bit in image, with point (x+
Gray scale difference value between Δ x, y+ Δ y) is gΔ((x, y) (x+ Δ x, y+ Δ y), if the institute of grey scale difference is likely for-g for x, y)=g
Value is m level, make point (x, y) to flame image region in move, accumulative gΔ(x y) takes the number of times of different value, makes gΔ
(x, rectangular histogram y), from rectangular histogram, gΔ(x, y) probability of value is pΔ(i)；(x+ Δ x, y+ Δ y) is value picture at its midpoint
(x, y) point around, (x, distance y) is Δ x and Δ y to element for this point and value pixel；
Use the uniformity coefficient extracting angle direction second moment ASM reflection gradation of image distribution, if close pixel ash
Angle value differs greatly, then ASM value is the biggest, illustrates that texture is the most coarse；
Textural characteristics is relevant with the degree of oxidation of chemical element in molten bath, and the most violent grain roughness that burns is the highest；
4. collection optical flow field multidate information Ent:
In order to describe the flame stroboscopic feature of converting process Flame multidate information, particularly later stage, it is necessary first to set up
The descriptive model that flame dynamically changes, uses optical flow field to set up scitillation process, is small fortune according to the basic assumption of optical flow computation
Moving and brightness constancy, (point constant for pixel intensity is existed between consecutive frame by x, y, t)=I (x+dx, y+dy, t+dt) to obtain I
Limit region intraconnections, construct interframe light flow graph F；F have recorded the flame violent degree in consecutive variations process, and F is carried out spy
Levy expression and description can reflect the information such as stroboscopic；
Wherein, x, y refer to the coordinate at image midpoint, and t is to be the time shaft of consecutive image, this is because video is by continuously
Image construction on time shaft, dx represents the displacement of x-axis, and dy represents the displacement of y-axis, and dt represents the displacement of t axle, and I is video
Stream is the picture frame of t when the time；
To interframe light flow graph, using gray level co-occurrence matrixes to describe its feature, during it calculates, direction takes 0 °, 45 °, 90 °,
135 °, step-length step=1 is to adapt to the feature of the random microtexture of flame, if M is the gray level co-occurrence matrixes obtained, uses entropy
Describe its feature, calculateMultidate information for flame；
Wherein, the coordinate of pixel during i, j are image；
(4) use generalized regression nerve networks to set up and identify model, existing characteristic vector V=that step (3) gathers [B, S,
ASM, Ent] for inputting, with the state in which numeral that blows for output: " 1 " represents the initial stage, and " 2 " represent mid-term, and " 3 " represent end
Phase；The generalized regression nerve networks set up is four inputs and single output, and hidden layer neuron excitation function is Gaussian function, output
For linear mapping function；By monochrome information, chrominance information, texture information and the optical flow field multidate information of known flame body
Export with it that data are corresponding to be listed, set up the image video signal no less than 10 heats as set of data samples；And by therein
Every piece image carries out the mark of blowing state, is trained, by the output data in training by state building identification model
Numeral is sorted out；
(5) real time video signals step (1) gathered is after step (2) and (3) process, and inputs as characteristic vector
In identification model after step (4) training, its output is the identification of blowing state, when the State-output numeral that blows is 3, and generation
Table blowing enters latter stage, sends signal prompt.
Claims (4)
1. the pneumatic steelmaking blowing state recognition system moving static nature based on image, it is characterised in that include image acquisition
Module, coloured image flame region split module, colored flames image sound state character representation and describing module, based on image spy
The blowing state classification module levied and ipc monitor module:
Image capture module is used for captured in real-time flame conditions, and the video signal of shooting is stored and transmitted to coloured image fire
Flame region segmentation module；
Coloured image flame region segmentation module is for being divided into single color image frames by the video signal of collection, to colour
Flame background part in image is split with flame body part；
Colored flames image sound state character representation and describing module are for gathering the brightness letter of segmentation gained flame body part
Breath, chrominance information, texture information and optical flow field multidate information；
Blowing state classification module based on characteristics of image is for by colored flames image sound state character representation and describing module
The various information gathered is set up as input and is identified model, and after arranging output state, gathers the built knowledge of data sample set pair
Other model is trained；And then the new input content gathered is judged output state, the output state that blowing enters latter stage is sent out
Deliver to ipc monitor module；
Ipc monitor module is used for sending signal prompt.
2. according to the pneumatic steelmaking blowing state recognition system moving static nature based on image shown in claim 1, its feature
It is: described image capture module also includes photographic head, memorizer and signal transmission device.
3. the pneumatic steelmaking blowing state identification method moving static nature based on image, it is characterised in that through following step
Rapid:
(1) color video camera is installed so that it is keep consistent with the position manually seeing fire and angle, the position of fixing camera, Jiao
Away from；The real time video signals that color video camera is gathered is connected with datatron by data acquisition card；
(2) video signal of collection is divided into single color image frames by datatron, to the flame background in coloured image with
Flame body is split, and will change to uniform L*a*b* space by RGB color space, after note red (255,0,0) conversion
For RLab, yellow (255,255,0) change after into YLab, white (255,255,255) change after into WLabIf P is converted images
In a pixel, D is the Euclidean distance between pixel, according to formula
Then segmentation strategy is:
In formula, DPIRefer to the color value of chromatism between one pixel of the image after color space conversion and reference point pixel；I refers to
An image pixel after RGB color is transformed into Lab color, this pixel has three components, is L respectively, a, b；T is for controlling
The threshold value of similarity；
Again willPart as flame body, other parts are split as flame background；
(3) it is used for gathering monochrome information, chrominance information, texture information and optical flow field by step (2) gained flame body part
Multidate information:
1. collection monochrome information B:
If (i, j) for the flame region pixel after segmentation, then (∑ I (i, j))/(Count), wherein B is bright to brightness B=to I
Degree feature, Count is area pixel number, and i, j are the position of pixel, coordinate the most in the picture；
2. collection chrominance information S:
Third moment under simple component i is
In formula, pi,jThe probability occurred for the pixel that gray scale in coloured image i-th Color Channel component is j, i is for representing colour
The i-th component of image, j is the gray value representing image, and si is the color moment feature of image, and μ i is relevant parameter；
3. collection texture information ASM:
Grey scale difference statistical method is utilized to describe static Texture complication, if (x is y) a bit in image, with point (x+ Δ x, y
Gray scale difference value between+Δ y) is gΔ(x, y)=g (and x, y)-g (x+ Δ x, y+ Δ y), if institute's likely value of grey scale difference
For m level, make point (x, y) to flame image region in move, accumulative gΔ(x y) takes the number of times of different value, makes gΔ(x,y)
Rectangular histogram, from rectangular histogram, gΔ(x, y) probability of value is pΔ(i)；
Use the uniformity coefficient extracting angle direction second moment ASM reflection gradation of image distribution, if close grey scale pixel value
Differ greatly, then ASM value is the biggest, illustrates that texture is the most coarse；
Textural characteristics is relevant with the degree of oxidation of chemical element in molten bath, and the most violent grain roughness that burns is the highest；
4. collection optical flow field multidate information Ent:
Set up the descriptive model that flame dynamically changes, use optical flow field to set up scitillation process, according to the basic assumption of optical flow computation
For small movements and brightness constancy, (x, y, t)=I (x+dx, y+dy, t+dt), by pixel intensity perseverance between consecutive frame to obtain I
Fixed point is limiting region intraconnections, constructs interframe light flow graph F；F have recorded the flame violent degree in consecutive variations process, right
F carries out character representation and description can reflect the information such as stroboscopic；
Wherein, x, y refer to the coordinate at image midpoint, and t is to be the time shaft of consecutive image, this is because video is by continuous time
Image construction on axle, dx represents the displacement of x-axis, and dy represents the displacement of y-axis, and dt represents the displacement of t axle, and I is in video flowing
It is the picture frame of t when the time；
To interframe light flow graph, using gray level co-occurrence matrixes to describe its feature, during it calculates, direction takes 0 °, 45 °, 90 °,
135 °, step-length step=1 is to adapt to the feature of the random microtexture of flame, if M is the gray level co-occurrence matrixes obtained, uses entropy
Describe its feature, calculateMultidate information for flame；
Wherein, the coordinate of pixel during i, j are image；
(4) use generalized regression nerve networks to set up and identify model, existing characteristic vector V=that step (3) gathers [B, S, ASM,
Ent] for inputting, with the state in which numeral that blows for output: " 1 " represents the initial stage, and " 2 " represent mid-term, and " 3 " represent latter stage；Build
Vertical generalized regression nerve networks is four inputs and single output, and hidden layer neuron excitation function is Gaussian function, is output as line
Property mapping function；By monochrome information, chrominance information, texture information and the optical flow field multidate information of known flame body and its
Output data correspondence is listed, and sets up set of data samples；It is trained building identification model, the output data in training are pressed shape
State numeral is sorted out；
(5) real time video signals step (1) gathered is after step (2) and (3) process, as characteristic vector input step
(4) in the identification model after training, its output is the identification of blowing state, and when the State-output numeral that blows is 3, representative is blown
Refining enters latter stage, sends signal prompt.
4. according to the pneumatic steelmaking blowing state identification method moving static nature based on image shown in claim 3, its feature
Be: the point of described step (3) (x+ Δ x, y+ Δ y) be value pixel (x, y) point around, this point and value pixel (x, y)
Distance be Δ x and Δ y.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610272037.9A CN105956618B (en) | 2016-04-27 | 2016-04-27 | Converter steelmaking blowing state identification system and method based on image dynamic and static characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610272037.9A CN105956618B (en) | 2016-04-27 | 2016-04-27 | Converter steelmaking blowing state identification system and method based on image dynamic and static characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105956618A true CN105956618A (en) | 2016-09-21 |
| CN105956618B CN105956618B (en) | 2021-12-03 |
Family
ID=56916587
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610272037.9A Active CN105956618B (en) | 2016-04-27 | 2016-04-27 | Converter steelmaking blowing state identification system and method based on image dynamic and static characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105956618B (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109598200A (en) * | 2018-11-01 | 2019-04-09 | 云南昆钢电子信息科技有限公司 | A kind of digital image recognition system and method for hot-metal bottle tank number |
| CN110309973A (en) * | 2019-07-01 | 2019-10-08 | 中冶赛迪重庆信息技术有限公司 | A kind of converter splash prediction technique and system based on video intelligent algorithm |
| CN110378936A (en) * | 2019-07-30 | 2019-10-25 | 北京字节跳动网络技术有限公司 | Optical flow computation method, apparatus and electronic equipment |
| CN110796046A (en) * | 2019-10-17 | 2020-02-14 | 武汉科技大学 | Intelligent steel slag detection method and system based on convolutional neural network |
| CN111104856A (en) * | 2019-11-18 | 2020-05-05 | 中冶赛迪技术研究中心有限公司 | Converter smelting splash monitoring method, system, storage medium and equipment |
| CN111222360A (en) * | 2018-11-23 | 2020-06-02 | 隆基绿能科技股份有限公司 | Method and device for detecting melting state of silicon material and storage medium |
| CN111340116A (en) * | 2020-02-27 | 2020-06-26 | 中冶赛迪重庆信息技术有限公司 | Converter flame identification method and system, electronic equipment and medium |
| CN111476244A (en) * | 2020-05-29 | 2020-07-31 | 宝钢湛江钢铁有限公司 | System and method for intelligently identifying number of scrap steel trough for converter |
| CN111598905A (en) * | 2020-05-13 | 2020-08-28 | 云垦智能科技(上海)有限公司 | Method for identifying type of blast furnace flame by using image segmentation technology |
| CN112981135A (en) * | 2021-02-06 | 2021-06-18 | 楚雄滇中有色金属有限责任公司 | Method for judging end point of slagging period of converter copper smelting |
| CN113033705A (en) * | 2021-04-22 | 2021-06-25 | 江西理工大学 | Intelligent judgment and verification method for copper converter blowing slagging period end point based on pattern recognition |
| CN113033704A (en) * | 2021-04-22 | 2021-06-25 | 江西理工大学 | Intelligent judging method for copper converter converting copper making period end point based on pattern recognition |
| CN113158818A (en) * | 2021-03-29 | 2021-07-23 | 青岛海尔科技有限公司 | Method, device and equipment for identifying fake video |
| CN113592760A (en) * | 2020-04-30 | 2021-11-02 | 昆明理工大学 | Converter endpoint carbon content prediction method based on flame image texture features |
| CN115339879A (en) * | 2022-10-19 | 2022-11-15 | 昆明理工大学 | Intelligent conveying and tracking method and system for small long and square billets based on machine vision |
| CN115880490A (en) * | 2022-11-21 | 2023-03-31 | 广东石油化工学院 | Flame segmentation method based on isolated forest |
| CN116402813A (en) * | 2023-06-07 | 2023-07-07 | 江苏太湖锅炉股份有限公司 | Neural network-based copper converter converting copper-making period end point judging method |
| CN118671096A (en) * | 2024-08-15 | 2024-09-20 | 四川宏安兴盛电子科技有限公司 | PCB copper deposition electroplating quality real-time monitoring method and system based on video image processing |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002074370A (en) * | 2000-08-28 | 2002-03-15 | Ntt Data Corp | System and method for monitoring based on moving image and computer readable recording medium |
| US20030025794A1 (en) * | 2001-07-31 | 2003-02-06 | Hirofumi Fujii | Moving object detecting method, apparatus and computer program product |
| CN101393603A (en) * | 2008-10-09 | 2009-03-25 | 浙江大学 | Method for recognizing and detecting tunnel fire disaster flame |
| CN103116746A (en) * | 2013-03-08 | 2013-05-22 | 中国科学技术大学 | Video flame detecting method based on multi-feature fusion technology |
-
2016
- 2016-04-27 CN CN201610272037.9A patent/CN105956618B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002074370A (en) * | 2000-08-28 | 2002-03-15 | Ntt Data Corp | System and method for monitoring based on moving image and computer readable recording medium |
| US20030025794A1 (en) * | 2001-07-31 | 2003-02-06 | Hirofumi Fujii | Moving object detecting method, apparatus and computer program product |
| CN101393603A (en) * | 2008-10-09 | 2009-03-25 | 浙江大学 | Method for recognizing and detecting tunnel fire disaster flame |
| CN103116746A (en) * | 2013-03-08 | 2013-05-22 | 中国科学技术大学 | Video flame detecting method based on multi-feature fusion technology |
Non-Patent Citations (1)
| Title |
|---|
| 刘辉: "《转炉炼钢吹炼数据预测中火焰图像多特征提取方法研究》", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109598200B (en) * | 2018-11-01 | 2023-03-03 | 云南昆钢电子信息科技有限公司 | Intelligent image identification system and method for molten iron tank number |
| CN109598200A (en) * | 2018-11-01 | 2019-04-09 | 云南昆钢电子信息科技有限公司 | A kind of digital image recognition system and method for hot-metal bottle tank number |
| CN111222360B (en) * | 2018-11-23 | 2023-10-31 | 隆基绿能科技股份有限公司 | Method, equipment and storage medium for detecting molten state of silicon material |
| CN111222360A (en) * | 2018-11-23 | 2020-06-02 | 隆基绿能科技股份有限公司 | Method and device for detecting melting state of silicon material and storage medium |
| CN110309973A (en) * | 2019-07-01 | 2019-10-08 | 中冶赛迪重庆信息技术有限公司 | A kind of converter splash prediction technique and system based on video intelligent algorithm |
| CN110378936A (en) * | 2019-07-30 | 2019-10-25 | 北京字节跳动网络技术有限公司 | Optical flow computation method, apparatus and electronic equipment |
| CN110378936B (en) * | 2019-07-30 | 2021-11-05 | 北京字节跳动网络技术有限公司 | Optical flow calculation method and device and electronic equipment |
| CN110796046A (en) * | 2019-10-17 | 2020-02-14 | 武汉科技大学 | Intelligent steel slag detection method and system based on convolutional neural network |
| CN110796046B (en) * | 2019-10-17 | 2023-10-10 | 武汉科技大学 | Intelligent steel slag detection method and system based on convolutional neural network |
| CN111104856A (en) * | 2019-11-18 | 2020-05-05 | 中冶赛迪技术研究中心有限公司 | Converter smelting splash monitoring method, system, storage medium and equipment |
| CN111340116A (en) * | 2020-02-27 | 2020-06-26 | 中冶赛迪重庆信息技术有限公司 | Converter flame identification method and system, electronic equipment and medium |
| CN113592760B (en) * | 2020-04-30 | 2024-04-16 | 昆明理工大学 | Converter endpoint carbon content prediction method based on flame image texture features |
| CN113592760A (en) * | 2020-04-30 | 2021-11-02 | 昆明理工大学 | Converter endpoint carbon content prediction method based on flame image texture features |
| CN111598905A (en) * | 2020-05-13 | 2020-08-28 | 云垦智能科技(上海)有限公司 | Method for identifying type of blast furnace flame by using image segmentation technology |
| CN111476244A (en) * | 2020-05-29 | 2020-07-31 | 宝钢湛江钢铁有限公司 | System and method for intelligently identifying number of scrap steel trough for converter |
| CN112981135A (en) * | 2021-02-06 | 2021-06-18 | 楚雄滇中有色金属有限责任公司 | Method for judging end point of slagging period of converter copper smelting |
| CN112981135B (en) * | 2021-02-06 | 2022-09-27 | 楚雄滇中有色金属有限责任公司 | Method for judging end point of slagging period of converter copper smelting |
| CN113158818A (en) * | 2021-03-29 | 2021-07-23 | 青岛海尔科技有限公司 | Method, device and equipment for identifying fake video |
| CN113033704A (en) * | 2021-04-22 | 2021-06-25 | 江西理工大学 | Intelligent judging method for copper converter converting copper making period end point based on pattern recognition |
| CN113033704B (en) * | 2021-04-22 | 2023-11-07 | 江西理工大学 | Intelligent judging method and system for copper converter converting copper-making final point based on pattern recognition |
| CN113033705A (en) * | 2021-04-22 | 2021-06-25 | 江西理工大学 | Intelligent judgment and verification method for copper converter blowing slagging period end point based on pattern recognition |
| CN115339879A (en) * | 2022-10-19 | 2022-11-15 | 昆明理工大学 | Intelligent conveying and tracking method and system for small long and square billets based on machine vision |
| CN115880490A (en) * | 2022-11-21 | 2023-03-31 | 广东石油化工学院 | Flame segmentation method based on isolated forest |
| CN115880490B (en) * | 2022-11-21 | 2023-10-27 | 广东石油化工学院 | Flame segmentation method based on isolated forest |
| CN116402813A (en) * | 2023-06-07 | 2023-07-07 | 江苏太湖锅炉股份有限公司 | Neural network-based copper converter converting copper-making period end point judging method |
| CN116402813B (en) * | 2023-06-07 | 2023-08-04 | 江苏太湖锅炉股份有限公司 | Neural network-based copper converter converting copper-making period end point judging method |
| CN118671096A (en) * | 2024-08-15 | 2024-09-20 | 四川宏安兴盛电子科技有限公司 | PCB copper deposition electroplating quality real-time monitoring method and system based on video image processing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105956618B (en) | 2021-12-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105956618A (en) | Converter steelmaking blowing state recognition system and method based on image dynamic and static characteristics | |
| CN205845067U (en) | The pneumatic steelmaking blowing state recognition system of static nature is moved based on image | |
| CN105678332B (en) | Converter steelmaking end point judgment method and system based on flame image CNN recognition modeling | |
| CN106228150B (en) | Smog detection method based on video image | |
| CN101561262B (en) | Surface roughness on line measurement method under uncertain condition | |
| CN101819024B (en) | Machine vision-based two-dimensional displacement detection method | |
| CN106886216A (en) | Robot automatic tracking method and system based on RGBD Face datections | |
| CN109598200B (en) | Intelligent image identification system and method for molten iron tank number | |
| CN103571994B (en) | Infrared steel slag detection method of converter | |
| CN108346147B (en) | Technical method for quickly, automatically and accurately identifying coal rock micro-components | |
| US6571228B1 (en) | Hybrid neural networks for color identification | |
| CN106483143A (en) | A kind of solar energy photovoltaic panel dust stratification on-Line Monitor Device and its detection method | |
| CN111401246A (en) | Smoke concentration detection method, device, equipment and storage medium | |
| CN112862769A (en) | Blast furnace slag iron ratio online intelligent monitoring method and system | |
| CN112541427B (en) | Identification and material quantity evaluation method for high-quality heavy steel scrap | |
| CN104531936B (en) | Converter molten steel carbon content On-line Measuring Method based on Flame Image Characteristics | |
| Zhang et al. | Detection method for pulverized coal injection and particles in the tuyere raceway using image processing | |
| CN109815957A (en) | A kind of character recognition method based on color image under complex background | |
| CN110935646A (en) | Full-automatic crab grading system based on image recognition | |
| CN117783011A (en) | Intelligent quality control system for fruit juice production line | |
| CN116402813B (en) | Neural network-based copper converter converting copper-making period end point judging method | |
| CN102968644A (en) | Method for predicting smelting finishing point of argon-oxygen refined iron alloy | |
| CN104050678A (en) | Underwater monitoring color image quality measurement method | |
| CN215520976U (en) | Cable tunnel personnel are detained monitoring system | |
| CN105427335B (en) | A kind of detection of continuous band-shaped porous metal material plating leakage defect and the method positioned |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |