CN115309109A - Steel plate rotation control method and device, storage medium and electronic equipment - Google Patents
Steel plate rotation control method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN115309109A CN115309109A CN202110500032.8A CN202110500032A CN115309109A CN 115309109 A CN115309109 A CN 115309109A CN 202110500032 A CN202110500032 A CN 202110500032A CN 115309109 A CN115309109 A CN 115309109A
- Authority
- CN
- China
- Prior art keywords
- steel plate
- image
- controlled
- control
- instruction
- 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
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
- G05B19/414—Structure of the control system, e.g. common controller or multiprocessor systems, interface to servo, programmable interface controller
- G05B19/4147—Structure of the control system, e.g. common controller or multiprocessor systems, interface to servo, programmable interface controller characterised by using a programmable interface controller [PIC]
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B21—MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
- B21B—ROLLING OF METAL
- B21B37/00—Control devices or methods specially adapted for metal-rolling mills or the work produced thereby
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B21—MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
- B21B—ROLLING OF METAL
- B21B39/00—Arrangements for moving, supporting, or positioning work, or controlling its movement, combined with or arranged in, or specially adapted for use in connection with, metal-rolling mills
- B21B39/20—Revolving, turning-over, or like manipulation of work, e.g. revolving in trio stands
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/34—Director, elements to supervisory
- G05B2219/34013—Servocontroller
Abstract
The embodiment of the application provides a steel plate rotation control method, a steel plate rotation control device, a storage medium and electronic equipment, wherein the method comprises the following steps: acquiring a steel plate image of a controlled steel plate positioned on the roller way; processing the steel plate image to obtain various target parameters; constructing an input vector according to the various target parameters, and inputting the input vector into a neural network model to obtain an output vector, wherein the output vector comprises at least one control parameter; and generating a corresponding control instruction according to the at least one control parameter, and controlling the rotation process of the controlled steel plate on the roller way according to the control instruction. According to the technical scheme, the traditional image processing method and the deep learning method are fully combined, the neural network model only needs to be trained fully, the model is good, the accuracy can be guaranteed, and various target parameters can comprehensively reflect the current condition of the controlled steel plate, so that the method can accurately control the rotating process of the controlled steel plate.
Description
Technical Field
The application relates to the technical field of intelligent industrial manufacturing, in particular to a steel plate rotation control method and device, a storage medium and electronic equipment.
Background
In the steel rolling process, a steel rotating operation of a steel plate is involved, that is, the steel plate on a roller table is rotated to adjust the angle of the steel plate. In the prior art, the control process of driving the steel to rotate only judges whether the rotation process of the steel plate is finished according to whether the angle of the steel plate is in place, and the judgment mode is too single and is not accurate enough.
Disclosure of Invention
The embodiment of the application provides a steel plate rotation control method and device, a storage medium and electronic equipment, and aims to solve the problem that the rotation of a steel plate cannot be accurately controlled in the prior art.
In a first aspect, an embodiment of the present application provides a steel plate rotation control method, including: acquiring a steel plate image of a controlled steel plate positioned on the roller way; processing the steel plate image to obtain various target parameters; constructing an input vector according to the various target parameters, and inputting the input vector into a neural network model to obtain an output vector, wherein the output vector comprises at least one control parameter; and generating a corresponding control instruction according to the at least one control parameter, and controlling the rotation process of the controlled steel plate on the roller way according to the control instruction.
In a second aspect, an embodiment of the present application provides a steel plate rotation control apparatus, including: the image acquisition module is used for acquiring a steel plate image of the controlled steel plate on the roller way; the image processing module is used for processing the steel plate image to obtain various target parameters; the simulation output module is used for constructing an input vector according to the various target parameters and inputting the input vector into a neural network model to obtain an output vector, wherein the output vector comprises at least one control parameter; and the instruction generating module is used for generating a corresponding control instruction according to the at least one control parameter and controlling the rotation process of the controlled steel plate on the roller way according to the control instruction.
In a third aspect, an embodiment of the present application provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method according to the first aspect is performed.
In a fourth aspect, an embodiment of the present application provides an electronic device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating over the bus when the electronic device is operating, the machine-readable instructions when executed by the processor performing the method of the first aspect.
By adopting the technical scheme provided by the application, when the steel plate is rotated, a plurality of target parameters are identified by utilizing the steel plate image collected on site, the plurality of target parameters are influence variables in the rotation process of the controlled steel plate, an output vector is obtained by utilizing an input vector constructed by the plurality of target parameters and combining a deep learning method, and the output vector comprises control parameters for controlling the controlled steel plate in the next step. According to the technical scheme, the traditional image processing method and the deep learning method are fully combined, the neural network model only needs to be trained fully, the model is good, the accuracy can be guaranteed, and various target parameters can comprehensively reflect the current condition of the controlled steel plate, so that the method can accurately control the rotation process of the controlled steel plate.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 shows a schematic diagram of a steel turning system provided by an embodiment of the present application;
FIG. 2 is a flowchart illustrating a steel plate rotation control method according to an embodiment of the present disclosure;
FIG. 3 is a flow chart illustrating the step 120 of the method for controlling the rotation of the steel plate;
FIG. 4 is a flowchart illustrating a preprocessing of a steel plate image in a steel plate rotation control method;
FIG. 5 is a detailed flowchart of a steel plate rotation control method provided by an embodiment of the present application in a specific embodiment;
FIG. 6 is a schematic diagram illustrating a steel plate rotation control apparatus provided in an embodiment of the present application;
fig. 7 shows a schematic diagram of an electronic device provided in an embodiment of the present application.
Detailed Description
The following detailed description of exemplary embodiments of the present application, taken in conjunction with the accompanying drawings, makes it apparent that the described embodiments are only some embodiments of the application, and are not exhaustive of all embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
To facilitate understanding of the technical solution, the present application first introduces the steel turning system in the embodiment.
As shown in fig. 1, the steel turning system in the present embodiment includes: cameras, rotation control systems, and PLC (Programmable Logic Controller) systems. When steel is rotated, the controlled steel plate is positioned on the roller way, a plurality of cameras are arranged above the roller way, and each camera is used for acquiring the steel plate image of the controlled steel plate in a corresponding area on the roller way in real time. The camera uploads the collected steel plate image to the rotation control system, the rotation control system identifies parameters such as the position and the angle of the controlled steel plate according to the steel plate image uploaded by the camera, and generates a corresponding control instruction through the trained neural network model by utilizing the identified parameters, and sends the control instruction to the PLC control system. And the PLC control system executes corresponding operation according to the control instruction to complete real-time and accurate control of the controlled steel plate.
Optionally, the PLC control system further feeds back the current rotation speed Sr of the motor, the specification U of the controlled steel plate, and the rotation time t of the controlled steel plate to the rotation control system. The rotation control system also generates a corresponding control instruction through a neural network model by utilizing the current rotating speed Sr of the motor, the specification U of the controlled steel plate and the rotating time t of the controlled steel plate which are fed back by the PLC control system.
Optionally, the steel turning system further includes an electronic fence system, and the electronic fence system detects whether the controlled steel plate is located in the set dangerous area, generates the state parameter a, and uploads the state parameter a to the rotation control system. The rotation control system also generates a control instruction through a neural network model by using the state parameter A uploaded by the electronic fence system.
Fig. 2 shows a flowchart of a steel plate rotation control method provided in the present embodiment, which describes the steps performed by the rotation control system. Referring to fig. 2, the method includes:
and 110, acquiring a steel plate image of the controlled steel plate on the roller way.
And step 120, processing the steel plate image to obtain various target parameters.
The camera is arranged above the controlled steel plate, when the steel is rotated, the camera collects the steel plate image of the controlled steel plate on the roller way in real time and transmits the steel plate image to the rotation control system, and the rotation control system processes the steel plate image so as to identify various required target parameters.
Wherein the output vector comprises at least one control parameter. The neural network model is pre-trained to automatically output a corresponding output vector based on parameters in the input vector.
And 140, generating a corresponding control instruction according to the control parameter in the output vector, and controlling the rotation process of the controlled steel plate on the roller way according to the control instruction.
In implementation, a plurality of target parameters are identified by using the steel plate images collected on site, the target parameters are influence variables in the rotation process of the controlled steel plate, an input vector constructed by the target parameters is used, and a deep learning method is combined to obtain an output vector, wherein the output vector comprises control parameters for controlling the controlled steel plate in the next step.
Deep learning: the neural network model is trained by learning the mapping relation between various target parameters and control parameters, so that the neural network model can automatically output corresponding results in practical application. The embodiment fully combines the traditional image processing method and the deep learning method, the neural network model only needs to be trained fully, the model is good, the accuracy can be ensured, and various target parameters can comprehensively reflect the current condition of the controlled steel plate, so that the method can accurately control the rotation process of the controlled steel plate.
Referring to fig. 3, in step 120, the plurality of target parameters at least include corner coordinates and an angle of the controlled steel plate on the roller table. The process of step 120 when embodied includes:
step 121: and acquiring a steel plate area in the steel plate image.
Optionally, the semantic segmentation network is used for obtaining the steel plate area in the steel plate image, and the semantic segmentation network is trained in advance, so that the steel plate area can be accurately identified.
And step 122, performing corner detection on the steel plate area by using a corner detection algorithm to obtain coordinates of each corner in the steel plate area.
Optionally, the process of performing corner detection on the steel plate area is as follows:
(1) and performing a gray scale integration operation on any one point (x, y) of the image f (x, y) (i.e., the image of the steel plate region acquired in step 121) and the original image through a window w (x, y) centered on the point:
E(u,v)＝∑w(x,y)[f(x+u,y+v)-f(x,y)] 2
when the window moving distance is very small, the model can be simplified and equivalent to an ellipse model:
(2) in the above formula, a and b change with the movement of the window w (x, y), corresponding to the major and minor axes λ 1 and λ 2 of the elliptical model, λ 1 > λ 2 or λ 2 > λ 1 when (x, y) is in the boundary region, λ 1 and λ 2 are both small when (x, y) is not in contact with the boundary or the corner point, and λ 1 and λ 2 are both large when (x, y) is in the corner point position, so that the corner point position in the steel plate region can be determined.
After detecting the angular points, the coordinates (p) of each angular point in the steel plate area can be obtained xi ,p yi ) And i denotes the ith corner point. For a quadrangular steel plate, detected corner points are four vertexes on the image.
Of course, other corner detection algorithms may be used to identify the corner coordinates.
And step 123, calculating the angle theta of the controlled steel plate on the roller way according to the coordinates of each corner point.
Optionally, the calculation process of the angle θ of the controlled steel plate on the roller way is as follows:
(1) and calculating the attitude angle of the controlled steel plate in the steel plate image according to the coordinates of each corner point
Specifically, the angular points are connected pairwise according to coordinates of the angular points, and two longest edges are removed to obtain a target quadrangle; determining one long edge of the target quadrangle, and recording the end point value of the long edge as (x) 1 ,y 1 )，(x 2 ,y 2 ) (ii) a Calculating the included angle of the long edge relative to the horizontal line of the steel plate image to obtain an attitude angle
The calculation formula is as follows:
(2) attitude angle to be obtained
Steel plate current converted into world coordinate systemThe angle alpha.
(3) And obtaining the angle theta of the controlled steel plate on the roller way according to the current angle alpha of the steel plate and the angle of the perpendicular bisector of the roller way under the world coordinate system.
Specifically, the difference between the current angle α of the steel plate and the angle β of the roll table perpendicular bisector in the world coordinate system, which is obtained in advance, is calculated, and the difference is the angle θ of the controlled steel plate on the roll table.
The coordinates (p) of each corner point are obtained through the above-mentioned steps 121-123 xi ,p yi ) And angle θ, in step 130, according to (p) xi ,p yi ) And the angle theta constitutes the input vector.
Optionally, in step 120, the multiple objective parameters may further include: center point coordinates (c) of steel plate area x ,c y ) The control method comprises one or more of a motion trend omega of an angular point, a parameter H for representing whether the angular point is located in a first dangerous area, a parameter A for representing whether a controlled steel plate is located in a second dangerous area, the current rotating speed Sr of a motor, the specification U of the controlled steel plate and the rotating time t of the controlled steel plate. The specification U of the controlled steel plate can comprise the length L, the width W and the mass m of the controlled steel plate.
Thus, step 120 may also include:
(1) Acquiring the center point coordinates (c) of the steel plate area x ，c y ). When the angular point detection algorithm is used for detecting the angular points of the steel plate area, the coordinates (p) of each angular point in the steel plate area can be obtained simultaneously xi ,p yi ) And center point coordinates (c) x ,c y )。
(2) And calculating the motion trend omega of the corner by using a corner tracking algorithm.
(3) Judging whether any corner point is positioned in the first dangerous area according to the coordinates of each corner point, and obtaining a parameter H according to a judgment result; and if any corner point is located in the first dangerous area, setting the value of the parameter H to be 1, otherwise, setting the value of the parameter H to be 0. The first dangerous area is an area facing the weld joint, if the angular point is facing the weld joint, the controlled steel plate may damage equipment, and therefore the parameter H is added to the input vector as one of the influence variables in the rotation process of the controlled steel plate.
(4) And acquiring a parameter A, wherein the parameter A is used for representing whether the controlled steel plate is positioned in the second dangerous area or not, and the parameter A can be uploaded by an electronic fence system. If the controlled steel plate is located in the second dangerous area, the value of the parameter A is 1, otherwise, the value is 0.
(5) Acquiring the current rotating speed Sr of the motor, the specification U of the controlled steel plate and the rotating time t of the controlled steel plate; sr, U and t are obtained by the feedback of a PLC control system.
In step 130, according to (p) xi ,p yi )、(c x ，c y ) Omega, theta, H, A, sr, U and t form an input vector of [ (p) xi ,p yi )，(c x ,c y )，ω，θ，H，A，Sr，U，t]。
Optionally, the neural network model is trained on input vectors [ (p) xi ,p yi )，(c x ,c y )，ω，θ，H，A，Sr，U，t]Automatically outputting corresponding output vector [ S, D, M ]]And S is a rotating speed control parameter used for controlling the rotating speed of the roller motor, D is a stop parameter used for controlling whether to stop the rotation of the controlled steel plate, and M is a control mode parameter used for switching the control mode. The output vector may include at least one control parameter, which may be one or more of the parameters S, D, M, and possibly other parameters.
The rotation speed control parameters S comprise Sa and Sb, the Sa is used for controlling the rotation speed of the motor of the A-type conical roller, the Sb is used for controlling the rotation speed of the motor of the B-type conical roller, and the two roller motors are respectively used for providing different rotation torques for the rollers on the roller way.
Optionally, the output vector includes a rotation speed control parameter S, the rotation control system generates a first instruction according to the rotation speed control parameter S, and issues the first instruction to the PLC control system, and the PLC control system drives the roller motor to rotate according to the corresponding rotation speed according to the first instruction.
Optionally, the output vector includes a stop parameter D, the rotation control system generates a second instruction according to a parameter value of the stop parameter D, and issues the second instruction to the PLC control system, and the PLC control system stops the rotation process of the controlled steel plate according to the second instruction and drives the clamp to clamp the controlled steel plate.
For example, the rotation control system generates the second instruction when detecting that the value of the parameter D in the output vector S, D, M is 1.
Optionally, the output vector includes a control mode parameter M, the rotation control system generates a third instruction according to a parameter value of the control mode parameter M, and issues the third instruction to the PLC control system, and the PLC control system switches the control mode of the controlled steel plate from the automatic control mode to the manual control mode according to the third instruction.
For example, when t >30s in the input vector, the value of the parameter M in the output vector output by the neural network model is 0, and the rotation control system generates a third instruction when detecting that the value of the parameter M in the output vector is 0, so that the controlled steel plate can be automatically stopped when the rotation timeout occurs.
After the steel plate rotates, the PLC control system transmits stop information to the rotation control system and the electronic fence system to wait for the next steel plate to enter.
Further, in order to solve the problem that the edge of the steel plate in the image is blurred due to high brightness of a part of the hot steel plate, the obtained steel plate image is preprocessed by using a highlight inhibition algorithm, a target image is obtained after preprocessing, and then the steel plate area is obtained according to the target image.
Fig. 4 shows a step of preprocessing an image of a steel plate, as shown in fig. 4, including:
In this case, the input steel sheet image f (x, y) is separated into an illumination component and a reflection component.
(1) First, a low-pass filter capable of maintaining boundary contour information is used to estimate the illumination component of the steel plate image
In the above expression, a is the maximum weight, S v (x, y) and S h (x, y) are boundary detectors in the horizontal and vertical directions, respectively, and are expressed as follows, where δ is a very small positive number, and h is a constant:
(2) estimating the reflected component
Comprises the following steps:
(3) when estimated
While standing
ε is a number greater than 0.
Wherein, the enhancement process of the irradiation component is as follows:
(1) and the distribution range of the irradiation component value is [ i ] max ，i min ]Normalizing the illumination component:
(2) enhanced with a modified gamma function:
i(x,y)＝g(z(x,y))＝(1-a(z(x,y)))·z(x,y) γ +a(z(x,y))·z(x,y)
wherein a (z (x, y)) is a weight function, a (z (x, y)) = z (x, y) 2 The parameter gamma is equal to 0.2,0.3]。
(3) Linear broadening of the linear contrast of the illumination component, the calculation formula is as follows:
wherein i * (x, y) is the enhanced illumination component, d high And d low The upper and lower saturation points after the enhancement processing are obtained according to the histogram statistics.
The enhancement process of the reflection component is as follows:
(1) using a modified sigmoid function:
where k ∈ [0.35,0.87], function c (z) is the weighting function:
wherein a is equal to 0.35,0.62, b is equal to 1.05,1.2, d is equal to 0.7, 1.5.
(2) When the reflection component is enhanced, the improved sigmoid function is adopted to carry out curve adjustment on the reflection component to obtain the enhanced reflection component r * (x, y), namely:
r * (x,y)＝e r(x,y)
Illumination component and inverseFusing the components to obtain a pre-enhanced image
Namely:
in step 240, each pixel point on the pre-enhanced image is traversed to search for highlight areas on the image.
In step 250, the gray scale value of the highlight region is suppressed to obtain the target image.
The gray-scale value of the highlight region to be suppressed in the image is [ C ] L ,C H ],C H Highest gray value of highlight pixel, C L ＝0.6C H The inhibition process of the high light region is as follows:
(1) traversing the pre-enhanced image
Finding a pixel grayscale value satisfying [ C L ,C H ]Searching a sub-highlight pixel point q meeting the condition in a dm multiplied by dm range taking p as a center; the secondary highlight pixel point q meets the condition: the gray value of the sub-highlight pixels must be smaller than the gray value of the highlight pixels:
the gray value of the sub-highlight pixel should be closest to the highlight pixel gray value.
(2) Gray scale value of sub-highlight pixel q
Replacing gray values of highlight pixels, i.e.
Then let C L ＝C L +1, go through the highlight pixel suppression process again until C L ＝C H I.e. the iterative suppression of the highlight pixels is completed.
(3) After the highlight inhibition, the histogram of the image has relatively concentrated and continuous one-section gray level non-pixel distribution, and the gray levels of the image pixels are redistributed by utilizing the empty gray levels to complete the histogram broadening processing. Namely, the suppression of the highlight area of the image is completed, and the target image is obtained.
The steel plate rotation control method in the present application will be described in detail below with specific examples. Referring to fig. 5, the embodiment specifically includes:
and step 310, acquiring a steel plate image acquired by a camera.
And 320, preprocessing the obtained steel plate image by using a highlight inhibition algorithm to obtain a target image.
And step 330, inputting the target image into a semantic segmentation network U-net to acquire a steel plate area on the image.
In general, the controlled steel plate is a quadrilateral steel plate, and in step 340, the coordinates (p) of the corner points are obtained together xi ,p y1 )、(p x2 ,p y2 )、(p x3 ,p y3 )、(p x4 ,p y4 ) And center point coordinates (c) x ,c y )。
And 350, calculating the motion trend omega of the corner by using a corner tracking algorithm.
And 360, calculating the angle theta of the controlled steel plate on the roller way according to the coordinates of each corner point.
And 380, acquiring a parameter A fed back by the electronic fence system, and acquiring the current rotating speeds Sra and Srb of the motor, the length L, the width W and the mass m of the controlled steel plate and the rotating time t of the controlled steel plate which are fed back by the PLC control system.
Wherein Sra is the current rotating speed of the A-type tapered roller motor, and Srb is the current rotating speed of the B-type tapered roller motor.
And step 410, generating a corresponding control instruction according to the output vector, and issuing the control instruction to the PLC control system to complete the control of the controlled steel plate.
To sum up, the technical scheme of the embodiment of the application has the following technical effects:
(1) And preprocessing the obtained steel plate image by using a highlight inhibition algorithm, obtaining a target image after preprocessing, and then sending the preprocessed target image into a semantic segmentation network, thereby solving the problem that the edge of the steel plate in the image is blurred due to higher brightness of part of hot steel plates.
(2) The rotation control system is fully combined with the traditional image processing method and the deep learning method, and the neural network model only needs to be trained fully and well, so that the accuracy can be ensured, and therefore, the method can accurately control the rotation process of the controlled steel plate.
(3) The rotation control system fully pays attention to the influence variable in the steel plate rotation process, in the prior art, the control process of driving the steel plate to rotate only judges whether the rotation process of the steel plate is finished according to whether the angle of the steel plate is in place, neglects other parameters needing attention in the steel plate rotation process, such as the length, the width and the quality of the steel plate, which are closely related to the rotation torque and the rotation static friction, when the rotation torque can not overcome the static friction, the steel plate can not be rotated, and the steel plate rotation is a continuous motion process, parameters such as the angular point motion trend, the current rotation speed of a motor and the like also need to pay attention constantly, so that the neural network model can adjust the rotation speed of the motor accordingly.
According to various objective parameters, the output vector [ Sa, sb, D, M ] is obtained through the neural network model, the steel plate rotating process can be adjusted in time according to real-time conditions (including steel plate positions, steel plate angles, motor rotating speeds, steel plate specifications and the like), effective rotation of steel plates with different specifications and sizes is achieved, meanwhile, the steel plates are controlled to stop in time through the stop parameter D according to dangerous conditions such as angular points and the position of the steel plates in dangerous areas, and equipment damage risks are reduced.
(4) In the embodiment, multiple target parameters are easy to obtain, complex calculation is not needed, and the current condition of the controlled steel plate can be accurately and comprehensively reflected in real time.
Based on the same inventive concept, an embodiment of the present application provides a steel plate rotation control apparatus, referring to fig. 6, the apparatus includes: an image acquisition module 510, an image processing module 520, a mimic output module 530, and an instruction generation module 540. The image acquisition module 510 is configured to acquire a steel plate image of a controlled steel plate on a roller way; the image processing module 520 is configured to process the steel plate image to obtain a plurality of target parameters; the simulation output module 530 is configured to construct an input vector according to the plurality of objective parameters, and input the input vector into a neural network model to obtain an output vector, wherein the output vector includes at least one control parameter; the instruction generating module 540 is configured to generate a corresponding control instruction according to the at least one control parameter, and control a rotation process of the controlled steel plate on the roller way according to the control instruction.
Optionally, the image processing module 520 is configured to obtain a steel plate area in the steel plate image; performing corner detection on the steel plate area by using a corner detection algorithm to obtain coordinates (p) of each corner in the steel plate area xi ,p yi ) I represents the ith corner point; calculating the angle theta of the controlled steel plate on a roller way according to the coordinates of each angular point; the mimic output module 530 is used to generate a function according to (p) xi ,p yi ) And θ constructs an input vector.
Optionally, the image processing module 520 is further configured to calculate an attitude angle of the controlled steel plate in the steel plate image according to the coordinates of each corner point
The attitude angle
Converting the current angle alpha of the steel plate into a world coordinate system; and obtaining the angle theta of the controlled steel plate on the roller way according to the current angle alpha of the steel plate and the angle of the perpendicular bisector of the roller way under the world coordinate system.
Optionally, the image processing module 520 is further configured to connect two corner points according to coordinates of each corner point, and remove two longest edges to obtain a target quadrangle; determining one long edge of the target quadrangle; calculating the included angle of the long edge relative to the horizontal line of the steel plate image to obtain the attitude angle
Optionally, the simulation output module 530 is used for acquiring coordinates (c) of a center point of the steel plate area x ,c y ) (ii) a Calculating the motion trend omega of the angular point by using an angular point tracking algorithm; judging whether any corner point is positioned in the first dangerous area according to the coordinates of each corner point, and obtaining a parameter H according to a judgment result; acquiring a parameter A, wherein the parameter A is used for representing whether a controlled steel plate is located in a second dangerous area; acquiring the current rotating speed Sr of the motor, the specification U of a controlled steel plate and the rotating time t of the controlled steel plate; according to (p) xi ,p yi )、(c x ,c y ) ω, θ, H, A, sr, U, and t construct an input vector.
Optionally, the image processing module 520 is configured to separate the steel plate image into an illumination component and a reflection component; enhancing the illumination component and the reflection component, respectively; fusing the enhanced illumination component and the enhanced reflection component to obtain a pre-enhanced image; traversing each pixel point on the pre-enhanced image to search a highlight area on the image; suppressing the gray value of the highlight area to obtain a target image; and performing semantic segmentation on the target image by using a semantic segmentation network to obtain a steel plate area in the target image.
Optionally, the at least one control parameter includes a rotation speed control parameter, and the instruction generating module 540 is configured to generate a first instruction according to the rotation speed control parameter, and issue the first instruction to a PLC control system, so that the PLC control system drives the roller motor to rotate according to a corresponding rotation speed according to the first instruction.
Optionally, the at least one control parameter includes a stop parameter, and the instruction generating module 540 is configured to generate a second instruction according to the stop parameter, and issue the second instruction to the PLC control system, so that the PLC control system stops a rotation process of the controlled steel plate according to the second instruction and drives the clamp to clamp the controlled steel plate.
Optionally, the at least one control parameter includes a control mode parameter, and the instruction generating module 540 is configured to generate a third instruction according to the control mode parameter, and issue the third instruction to the PLC control system, so that the PLC control system switches the control mode of the controlled steel plate from an automatic control mode to a manual control mode according to the third instruction.
It is understood that the steel plate rotation control device in the present embodiment, the implementation principle and the technical effects thereof have been described in the foregoing method embodiments, and for the sake of brief description, the corresponding description in the steel plate rotation control method may be referred to for what is not mentioned in the steel plate rotation control device.
Fig. 7 illustrates one possible structure of an electronic device 600 provided in an embodiment of the present application. Referring to fig. 7, the electronic device 600 includes: a processor 610, a memory 620, and a communication interface 630, which are interconnected and in communication with each other via a communication bus 640 and/or other form of connection mechanism (not shown).
The Memory 620 includes one or more (Only one shown in the figure), which may be, but not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an electrically Erasable Programmable Read-Only Memory (EEPROM), and the like. The processor 610, and possibly other components, may access, read, and/or write data to the memory 620.
The processor 610 includes one or more (only one shown) which may be an integrated circuit chip having signal processing capabilities. The Processor 610 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Micro Control Unit (MCU), a Network Processor (NP), or other conventional processors; the Application-Specific Processor may also be a special-purpose Processor, including a Graphics Processing Unit (GPU), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, and discrete hardware components. Also, when there are multiple processors 610, some of them may be general-purpose processors and others may be special-purpose processors.
The communication interface 630 includes one or more (only one is shown in the figure) devices, and can be used for directly or indirectly communicating with other devices so as to perform data interaction, such as receiving steel plate images uploaded by the camera through the communication interface 630, issuing control instructions to the PLC control system, and receiving parameters uploaded by the electronic fence system. Communication interface 630 may include an interface for wired and/or wireless communication.
One or more computer program instructions may be stored in the memory 620 and read and executed by the processor 610 to implement the steel plate rotation control method provided by the embodiments of the present application and other desired functions.
It will be appreciated that the configuration shown in fig. 7 is merely illustrative and that electronic device 600 may include more or fewer components than shown in fig. 7 or have a different configuration than shown in fig. 7. The components shown in fig. 7 may be implemented in hardware, software, or a combination thereof. The electronic device 600 may be a PC, a notebook computer, a tablet computer, a server, an embedded device, or the like, and the electronic device 600 is not limited to a single device, and may also be a combination of multiple devices or a cluster formed by a large number of devices. The rotation control system in the present embodiment may be disposed on the electronic apparatus 600.
Embodiments of the present application further provide a computer-readable storage medium, including but not limited to a disk memory, a CD-ROM, an optical storage, etc., on which computer program instructions are stored, and when the computer program instructions are read and executed by a processor of a computer, the computer program instructions execute the steel plate rotation control method provided by the embodiments of the present application. For example, the computer-readable storage medium may be embodied as memory 620 in electronic device 600 in FIG. 7.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the unit is only a logical division, and other divisions may be realized in practice. In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (12)
1. A steel plate rotation control method, characterized by comprising:
acquiring a steel plate image of a controlled steel plate positioned on the roller way;
processing the steel plate image to obtain various target parameters;
constructing an input vector according to the various target parameters, and inputting the input vector into a neural network model to obtain an output vector, wherein the output vector comprises at least one control parameter;
and generating a corresponding control instruction according to the at least one control parameter, and controlling the rotation process of the controlled steel plate on the roller way according to the control instruction.
2. The method of claim 1, wherein processing the steel sheet image to obtain a plurality of objective parameters, and constructing an input vector according to the plurality of objective parameters comprises:
acquiring a steel plate area in a steel plate image;
performing corner detection on the steel plate area by using a corner detection algorithm to obtain coordinates (p) of each corner in the steel plate area xi ，p yi ) I represents the ith corner point;
calculating the angle theta of the controlled steel plate on the roller way according to the coordinates of the angular points;
according to (p) xi ,p yi ) And θ constructs an input vector.
3. The method according to claim 2, wherein the calculating an angle θ of the controlled steel plate on a roller bed according to the coordinates of each corner point comprises:
calculating the attitude angle of the controlled steel plate in the steel plate image according to the coordinates of each corner point
Will the attitude angle
Converting the current angle alpha of the steel plate into a world coordinate system;
and obtaining the angle theta of the controlled steel plate on the roller way according to the current angle alpha of the steel plate and the angle of the perpendicular bisector of the roller way under the world coordinate system.
4. The method of claim 3, wherein said determining is based on saidCalculating the attitude angle of the controlled steel plate in the steel plate image by the coordinates of each angular point
The method comprises the following steps:
connecting the angular points pairwise according to the coordinates of the angular points, and removing two longest edges to obtain a target quadrangle;
determining one long side of the target quadrangle;
calculating the included angle of the long edge relative to the horizontal line of the steel plate image to obtain the attitude angle
5. The method according to claim 2, wherein the method is according to (p) xi ,p yi ) And θ constructing an input vector comprising:
obtaining coordinates of a center point of the steel plate region (c) x ,c y )；
Calculating the motion trend omega of the angular point by using an angular point tracking algorithm;
judging whether any corner point is positioned in the first dangerous area according to the coordinates of each corner point, and obtaining a parameter H according to a judgment result;
acquiring a parameter A, wherein the parameter A is used for representing whether a controlled steel plate is positioned in a second dangerous area;
acquiring the current rotating speed Sr of the motor, the specification U of a controlled steel plate and the rotating time t of the controlled steel plate;
according to (p) xi ,p yi )、(c x ,c y ) ω, θ, H, A, sr, U, and t construct an input vector.
6. The method of claim 2, wherein acquiring a steel plate area in the steel plate image comprises:
separating the steel sheet image into an illumination component and a reflection component;
enhancing the illumination component and the reflection component, respectively;
fusing the enhanced illumination component and the enhanced reflection component to obtain a pre-enhanced image;
traversing each pixel point on the pre-enhanced image to search a highlight area on the image;
suppressing the gray value of the highlight area to obtain a target image;
and performing semantic segmentation on the target image by using a semantic segmentation network to obtain a steel plate area in the target image.
7. The method according to any one of claims 1 to 6, wherein the at least one control parameter comprises a rotation speed control parameter, and the generating of the corresponding control command according to the at least one control parameter and the controlling of the rotation process of the controlled steel plate on the roller table according to the control command comprise:
and generating a first instruction according to the rotating speed control parameter, and issuing the first instruction to a PLC control system so that the PLC control system drives a roller motor to rotate according to the corresponding rotating speed according to the first instruction.
8. The method according to any one of claims 1 to 6, wherein the at least one control parameter comprises a stop parameter, and the generating of the corresponding control command according to the at least one control parameter and the controlling of the rotation process of the controlled steel plate on the roller table according to the control command comprises:
and generating a second instruction according to the stop parameter, and issuing the second instruction to a PLC control system so that the PLC control system stops the rotation process of the controlled steel plate according to the second instruction and drives a clamp to clamp the controlled steel plate.
9. The method according to any one of claims 1 to 6, wherein the at least one control parameter comprises a control mode parameter, and the generating of the corresponding control command according to the at least one control parameter and the controlling of the rotation process of the controlled steel plate on the roller table according to the control command comprises:
and generating a third instruction according to the control mode parameters, and issuing the third instruction to a PLC control system so that the PLC control system switches the control mode of the controlled steel plate from an automatic control mode to a manual control mode according to the third instruction.
10. A steel plate rotation control device characterized by comprising:
the image acquisition module is used for acquiring a steel plate image of the controlled steel plate on the roller way;
the image processing module is used for processing the steel plate image to obtain various target parameters;
the simulation output module is used for constructing an input vector according to the various target parameters and inputting the input vector into a neural network model to obtain an output vector, wherein the output vector comprises at least one control parameter;
and the instruction generating module is used for generating a corresponding control instruction according to the at least one control parameter and controlling the rotation process of the controlled steel plate on the roller way according to the control instruction.
11. A storage medium, characterized in that the storage medium has stored thereon a computer program which, when being executed by a processor, performs the method according to any one of claims 1-9.
12. An electronic device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating over the bus when the electronic device is operating, the machine-readable instructions, when executed by the processor, performing the method of any of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110500032.8A CN115309109A (en) | 2021-05-08 | 2021-05-08 | Steel plate rotation control method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110500032.8A CN115309109A (en) | 2021-05-08 | 2021-05-08 | Steel plate rotation control method and device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115309109A true CN115309109A (en) | 2022-11-08 |
Family
ID=83853495
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110500032.8A Pending CN115309109A (en) | 2021-05-08 | 2021-05-08 | Steel plate rotation control method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115309109A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117539209A (en) * | 2024-01-09 | 2024-02-09 | 东北大学 | Steel conversion control method, device, computer equipment and computer readable storage medium |
-
2021
- 2021-05-08 CN CN202110500032.8A patent/CN115309109A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117539209A (en) * | 2024-01-09 | 2024-02-09 | 东北大学 | Steel conversion control method, device, computer equipment and computer readable storage medium |
| CN117539209B (en) * | 2024-01-09 | 2024-03-15 | 东北大学 | Steel conversion control method, device, computer equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10872433B2 (en) | Vehicle environment modeling with a camera | |
| CN109345593B (en) | Camera posture detection method and device | |
| CN107179768B (en) | Obstacle identification method and device | |
| JP6946831B2 (en) | Information processing device and estimation method for estimating the line-of-sight direction of a person, and learning device and learning method | |
| CN110853100B (en) | Structured scene vision SLAM method based on improved point-line characteristics | |
| JP5070213B2 (en) | Vision processing device for network-based intelligent service robot, processing method thereof, and system using the same | |
| JP4931218B2 (en) | Imaging apparatus, object detection method, and attitude parameter calculation method | |
| US8811744B2 (en) | Method for determining frontal face pose | |
| WO2018177159A1 (en) | Method and system for determining position of moving object | |
| CN109872366B (en) | Method and device for detecting three-dimensional position of object | |
| US11049270B2 (en) | Method and apparatus for calculating depth map based on reliability | |
| JP2018522348A (en) | Method and system for estimating the three-dimensional posture of a sensor | |
| CN110493488B (en) | Video image stabilization method, video image stabilization device and computer readable storage medium | |
| EP2757527A1 (en) | System and method for distorted camera image correction | |
| WO2015072217A1 (en) | Moving body position estimation device and moving body position estimation method | |
| WO2018102990A1 (en) | System and method for rectifying a wide-angle image | |
| CN111756990B (en) | Image sensor control method, device and system | |
| CN116193108B (en) | Online self-calibration method, device, equipment and medium for camera | |
| JP2003271933A (en) | Face detector, face detecting method, and robot device | |
| CN115309109A (en) | Steel plate rotation control method and device, storage medium and electronic equipment | |
| CN110910349B (en) | Wind turbine state acquisition method based on aerial photography vision | |
| CN114170826B (en) | Automatic driving control method and device, electronic device and storage medium | |
| WO2022152050A1 (en) | Object detection method and apparatus, computer device and storage medium | |
| KR102541959B1 (en) | Elevator control system and method for controlling elevator which robot and human board | |
| CN107967691B (en) | Visual mileage calculation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |