Method and system for correcting image of spray gun in converting furnace

Abstract

The invention provides a correction method and system for a spray gun image in a converting furnace, and the method comprises the steps: constructing an initial neural network model, and obtaining a final neural network model after the training of a training data set; acquiring a real-time in-furnace image, and acquiring a real-time distortion parameter data set through the real-time in-furnace image and the final neural network model; acquiring a first pixel coordinate of an original pixel point in the real-time in-furnace image, and converting the first pixel coordinate into a first physical coordinate; performing position adjustment on the first physical coordinate based on the real-time distortion parameter data set to obtain a second physical coordinate, and converting the second physical coordinate into a second pixel coordinate; and obtaining an adjustment pixel value corresponding to the second pixel coordinate, and forming a corrected image through the second pixel coordinate and the adjustment pixel value. Through the mode, the judgment error caused by directly determining the suspension position of the spray gun through the real-time in-furnace image is avoided, an accurate basis is provided for identification of the suspension position of the spray gun, and the blowing efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image correction, and in particular to a correction method and system for a spray gun image in a blowing furnace. Background Art

[0002] Blowing is an important process in pyrometallurgy, in which high-pressure, high-speed industrial pure oxygen or air is sprayed onto the surface of the molten crude metal through a spray gun in the furnace, so that the impurities are oxidized into gases such as sulfur dioxide and carbon monoxide and escape, or become oxides and slag with the added flux, so as to obtain purer metal.

[0003] By controlling the distance between the spray gun and the melt surface, the air outlet of the spray gun is always suspended 200mm above the melt surface, which can effectively improve the blowing efficiency. In the actual production process, the spray gun will be worn out, causing the spray gun to be suspended too high above the melt surface. By shooting the image inside the furnace with a camera, the suspension position of the spray gun can be obtained in real time, and then the distance between the spray gun and the melt surface can be adjusted in real time to ensure the hammering efficiency.

[0004] However, during the shooting process, the image inside the furnace is distorted to a certain extent due to the design structure of the camera itself, which leads to errors in the judgment of the suspension position of the spray gun, affecting the efficiency of blowing. Summary of the invention

[0005] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a method and system for correcting the spray gun image in a blowing furnace, aiming to solve the technical problem in the prior art that when the suspension position of the spray gun is acquired by shooting the image in the furnace with a camera, the image in the furnace is distorted to a certain extent due to the structural problem of the camera itself, resulting in errors in the judgment of the suspension position of the spray gun, thereby affecting the blowing efficiency.

[0006] In order to achieve the above objectives, in a first aspect, an embodiment of the present application provides a method for correcting a spray gun image in a blowing furnace, comprising the following steps:

[0007] Constructing an initial neural network model and a training data set, wherein the training data set includes a simulated distortion parameter data set and a simulated distorted image, and training the initial neural network model through the training data set to obtain a final neural network model;

[0008] Acquire a real-time in-furnace image in the blowing furnace by means of a photographing device, and use the real-time in-furnace image as an input value of the final neural network model to acquire a real-time distortion parameter data set;

[0009] Acquire first pixel coordinates of all original pixel points in the real-time furnace image, and convert the first pixel coordinates into first physical coordinates based on the parameters of the shooting device;

[0010] Adjusting the position of the first physical coordinate based on the real-time distortion parameter data set to obtain a second physical coordinate, and converting the second physical coordinate into a second pixel coordinate, wherein the second pixel coordinate corresponds to the first pixel coordinate;

[0011] An adjusted pixel value corresponding to the second pixel coordinate is obtained, and a corrected image is formed by the second pixel coordinate and the adjusted pixel value.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: when considering the factors causing the distortion of the real-time in-furnace image, the final neural network model formed by training the initial neural network model can simultaneously consider all the distortion factors in the real-time in-furnace image with distortion, providing fast and accurate data support for subsequent distortion correction, thereby improving the efficiency of distortion correction; after obtaining the real-time distortion parameter data group, the real-time in-furnace image is subjected to distortion correction through the real-time distortion parameter data group, thereby avoiding the judgment error caused by directly determining the suspension position of the spray gun through the real-time in-furnace image, providing a more accurate basis for identifying the suspension position of the spray gun, and improving the blowing efficiency.

[0013] Furthermore, the steps of constructing the initial neural network model are specifically as follows:

[0014] The input layer, composite processing layer, fully connected layer and output layer are constructed in sequence to form an initial neural network model.

[0015] Furthermore, the training data set is constructed, and the training data set includes a simulated distortion parameter data set and a simulated distorted image. The step of training the initial neural network model through the training data set includes:

[0016] Acquire past furnace images, set a simulated distortion parameter data group, perform distortion processing on the past furnace images by using the simulated distortion parameter data group to acquire simulated distorted images, wherein the simulated distortion parameter data group and the simulated distorted images constitute a training data set;

[0017] The simulated distorted image is used as the input value of the initial neural network model, and the simulated distortion parameter data set is used as the output value of the initial neural network model to train the initial neural network model.

[0018] Furthermore, the formula for obtaining the first physical coordinate is:

[0019]

[0020] Among them, x w Represents the horizontal coordinate of the first physical coordinate, y wrepresents the ordinate of the first physical coordinate, i represents the abscissa of the first pixel coordinate, j represents the ordinate of the first pixel coordinate, cx represents the central abscissa of the real-time image inside the furnace, cy represents the central ordinate of the real-time image inside the furnace, fx represents the effective focal length of the shooting device in the abscissa direction, and fy represents the effective focal length of the shooting device in the ordinate direction.

[0021] Furthermore, the real-time distortion parameter data set includes a first radial distortion parameter, a second radial distortion parameter, a third radial distortion parameter, a first tangential distortion parameter and a second tangential distortion parameter.

[0022] Furthermore, the calculation formula of the second physical coordinate is:

[0023]

[0024] Among them, x w Represents the horizontal coordinate of the first physical coordinate, y w Indicates the ordinate of the first physical coordinate, x n Represents the horizontal coordinate of the second physical coordinate, y n represents the ordinate of the second physical coordinate, k1 represents the first radial distortion parameter, k2 represents the second radial distortion parameter, k3 represents the third radial distortion parameter, p1 represents the first tangential distortion parameter, p2 represents the second tangential distortion parameter, and r represents the distance between the original pixel point and the center point of the real-time furnace image.

[0025] Furthermore, the step of obtaining the adjusted pixel value corresponding to the second pixel coordinate includes:

[0026] Restoring the second pixel coordinates to the real-time furnace image to obtain a mapping pixel point, and determining whether the coordinate value of the mapping pixel point is an integer;

[0027] If the coordinate value of the mapped pixel point is an integer, the pixel value of the mapped pixel point is determined as the adjusted pixel value;

[0028] If the coordinate value of the mapped pixel point is a non-integer, the pixel points adjacent to the mapped pixel point are queried to obtain a number of adjacent pixel points, and the adjusted pixel value is obtained based on the pixel values ​​of the several adjacent pixel points.

[0029] In a second aspect, an embodiment of the present application provides a system for correcting a spray gun image in a blowing furnace, which is applied to the method for correcting a spray gun image in a blowing furnace as described in the first aspect above, and the system comprises:

[0030] A training module is used to construct an initial neural network model and a training data set, wherein the training data set includes a simulated distortion parameter data set and a simulated distorted image, and the initial neural network model is trained by the training data set to obtain a final neural network model;

[0031] An acquisition module, used for acquiring a real-time in-furnace image in the blowing furnace through a photographing device, and using the real-time in-furnace image as an input value of the final neural network model to acquire a real-time distortion parameter data set;

[0032] A first adjustment module, configured to obtain first pixel coordinates of all original pixel points in the real-time furnace image, and convert the first pixel coordinates into first physical coordinates based on the parameters of the shooting device;

[0033] a second adjustment module, configured to adjust the position of the first physical coordinate based on the real-time distortion parameter data group to obtain a second physical coordinate, and convert the second physical coordinate into a second pixel coordinate, wherein the second pixel coordinate corresponds to the first pixel coordinate;

[0034] The execution module is used to obtain the adjusted pixel value corresponding to the second pixel coordinate, and form a corrected image through the second pixel coordinate and the adjusted pixel value.

[0035] In a third aspect, an embodiment of the present application provides a computer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for correcting the spray gun image in the blowing furnace as described in the first aspect above is implemented.

[0036] In a fourth aspect, an embodiment of the present application provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for correcting the spray gun image in a blowing furnace as described in the first aspect above. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flow chart of the method for correcting the spray gun image in the blowing furnace in the first embodiment of the present invention;

[0038] Figure 2 It is a structural block diagram of a system for correcting a spray gun image in a blowing furnace according to a second embodiment of the present invention;

[0039] The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0040] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0041] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0043] See also Figure 1 The first embodiment of the present invention provides a method for correcting the image of a spray gun in a converting furnace, which is used to correct the image in the furnace obtained by a camera, wherein the camera is arranged on the side wall of the converting furnace, and an observation port corresponding to the position of the camera is provided on the converting furnace. In this embodiment, the camera includes a support platform connected to the side wall of the converting furnace in a vertically movable manner. Specifically, two guide rails extending vertically are provided on the side wall of the converting furnace, and the two guide rails are located on opposite sides of the observation port. The guide rails are slidably connected to a lifting block, and the top of the two lifting blocks is connected to the support platform. A sliding groove is provided on the support platform, and the sliding groove extends from one lifting block to the other lifting block. Preferably, the projection of the sliding groove on the side wall of the converting furnace is perpendicular to the guide rail, and a sliding block is slidably connected in the sliding groove. A camera is provided on the sliding block. Preferably, a spherical block is rotatably connected to the sliding block, and the end of the spherical block facing away from the sliding block is connected to the camera to achieve angle adjustment of the camera. It can be understood that by adopting the above structure, the position and angle of the camera can be adjusted, so as to obtain the image inside the furnace of the converting furnace more accurately. In some embodiments, a cover box is provided on the support platform, and the camera is located in the cover box. The cover box includes an openable door, and the door is located at one end of the cover box facing the side wall of the converting furnace. By providing the cover box, the camera can be protected when it is not in use.

[0044] The method for correcting the spray gun image in the blowing furnace comprises the following steps:

[0045] S10: constructing an initial neural network model and a training data set, wherein the training data set includes a simulated distortion parameter data set and a simulated distorted image, and training the initial neural network model through the training data set to obtain a final neural network model;

[0046] Specifically, the step S10 includes:

[0047] S110: constructing an input layer, a composite processing layer, a fully connected layer, and an output layer in sequence to form an initial neural network model;

[0048] The composite processing layer includes a plurality of connection layers. In this embodiment, the composite processing layer includes three connection layers, and the connection layers include convolution layers and pooling layers connected in sequence. The convolution layer is activated by a relu activation function.

[0049] S120: Acquire past furnace images, set a simulated distortion parameter data group, and perform distortion processing on the past furnace images using the simulated distortion parameter data group to obtain simulated distorted images, wherein the simulated distortion parameter data group and the simulated distorted images constitute a training data set.

[0050] Preferably, the previous in-furnace images are internal images of the blowing furnace obtained by the shooting device, and the simulated distortion parameter data group can be set accordingly based on experience, and the simulated distortion parameter data group includes a first simulated radial distortion parameter, a second simulated radial distortion parameter, a third simulated radial distortion parameter, a first simulated tangential distortion parameter and a second simulated tangential distortion parameter.

[0051] S130: using the simulated distorted image as an input value of the initial neural network model, and using the simulated distortion parameter data set as an output value of the initial neural network model, to train the initial neural network model;

[0052] The purpose of training the initial neural network model is to make the output of the initial neural network model closer to the value of the simulated distortion parameter data set. During training, a loss function is constructed based on the predicted value output by the initial neural network model and the simulated distortion parameter data set, and the weight parameter of the composite processing layer is adjusted by the loss function. Several simulated distortion images can be generated by several simulated distortion parameter data sets, and the output of the final neural network model can be made more accurate by more training data. It can be understood that the final neural network model is used to output a real-time distortion parameter data set based on the real-time furnace image.

[0053] S20: acquiring a real-time in-furnace image in the blowing furnace by means of a photographing device, and using the real-time in-furnace image as an input value of the final neural network model to obtain a real-time distortion parameter data set;

[0054] The real-time distortion parameter data set includes a first radial distortion parameter, a second radial distortion parameter, a third radial distortion parameter, a first tangential distortion parameter and a second tangential distortion parameter. It should be noted that the contents of the simulation distortion parameter data set and the real-time distortion parameter data set are consistent, and only the values ​​are different.

[0055] S30: Acquire first pixel coordinates of all original pixel points in the real-time furnace image, and convert the first pixel coordinates into first physical coordinates based on the parameters of the shooting device;

[0056] The formula for obtaining the first physical coordinate is:

[0057]

[0058] Among them, x w Represents the horizontal coordinate of the first physical coordinate, y w represents the ordinate of the first physical coordinate, i represents the abscissa of the first pixel coordinate, j represents the ordinate of the first pixel coordinate, cx represents the central abscissa of the real-time image inside the furnace, cy represents the central ordinate of the real-time image inside the furnace, fx represents the effective focal length of the shooting device in the abscissa direction, and fy represents the effective focal length of the shooting device in the ordinate direction.

[0059] S40: adjusting the position of the first physical coordinate based on the real-time distortion parameter data group to obtain a second physical coordinate, and converting the second physical coordinate into a second pixel coordinate, where the second pixel coordinate corresponds to the first pixel coordinate;

[0060] The calculation formula of the second physical coordinate is:

[0061]

[0062] Among them, x w Represents the horizontal coordinate of the first physical coordinate, y w Indicates the ordinate of the first physical coordinate, x n Represents the horizontal coordinate of the second physical coordinate, y n represents the ordinate of the second physical coordinate, k1 represents the first radial distortion parameter, k2 represents the second radial distortion parameter, k3 represents the third radial distortion parameter, p1 represents the first tangential distortion parameter, p2 represents the second tangential distortion parameter, and r represents the distance between the original pixel point and the center point of the real-time furnace image.

[0063] The principle of obtaining the second pixel coordinates through the second physical coordinates is the same as that of obtaining the first physical coordinates through the first pixel coordinates, except that the calculation method is opposite, which will not be described again here.

[0064] S50: Acquire an adjusted pixel value corresponding to the second pixel coordinate, and form a corrected image through the second pixel coordinate and the adjusted pixel value;

[0065] The real-time furnace image is a digital image, and its pixel value is defined only when the coordinates are integers. Therefore, step S50 includes:

[0066] S510: Restoring the second pixel coordinates to the real-time furnace image to obtain a mapping pixel point, and determining whether the coordinate value of the mapping pixel point is an integer;

[0067] S520: If the coordinate value of the mapped pixel point is an integer, determine the pixel value of the mapped pixel point as an adjusted pixel value;

[0068] S530: If the coordinate value of the mapped pixel point is a non-integer, query the pixel points adjacent to the mapped pixel point to obtain a number of adjacent pixel points, and obtain the adjusted pixel value based on the pixel values ​​of the several adjacent pixel points.

[0069] Specifically, the distance between the adjacent pixel point and the mapping pixel point is determined, and the pixel value of the adjacent pixel point that is closest to the mapping pixel point is used as the adjustment pixel value.

[0070] When considering the factors that cause the distortion of the real-time in-furnace image, the final neural network model formed by training the initial neural network model can simultaneously consider all distortion factors in the real-time in-furnace image with distortion, provide fast and accurate data support for subsequent distortion correction, and improve the efficiency of distortion correction; after obtaining the real-time distortion parameter data group, the real-time in-furnace image is subjected to distortion correction through the real-time distortion parameter data group, avoiding the judgment error caused by directly determining the suspension position of the spray gun through the real-time in-furnace image, providing a more accurate basis for identifying the suspension position of the spray gun, and improving the blowing efficiency.

[0071] See also Figure 2The second embodiment of the present invention provides a system for correcting the image of a spray gun in a blowing furnace. The system is applied to the method for correcting the image of a spray gun in a blowing furnace described in the above embodiment. The descriptions that have been made will not be repeated. As used below, the terms "module", "unit", "subunit" and the like can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0072] The system comprises:

[0073] A training module 10 is used to construct an initial neural network model and a training data set, wherein the training data set includes a simulated distortion parameter data set and a simulated distorted image, and the initial neural network model is trained by the training data set to obtain a final neural network model;

[0074] The driving module 10 comprises:

[0075] The first unit is used to sequentially construct an input layer, a composite processing layer, a fully connected layer, and an output layer to form an initial neural network model;

[0076] The second unit is used to obtain previous furnace images, set a simulated distortion parameter data group, and perform distortion processing on the previous furnace images by using the simulated distortion parameter data group to obtain simulated distorted images, wherein the simulated distortion parameter data group and the simulated distorted images constitute a training data set;

[0077] A third unit is used to use the simulated distorted image as an input value of the initial neural network model and the simulated distortion parameter data group as an output value of the initial neural network model to train the initial neural network model;

[0078] An acquisition module 20, for acquiring a real-time furnace image in the blowing furnace through a photographing device, and using the real-time furnace image as an input value of the final neural network model to acquire a real-time distortion parameter data set;

[0079] A first adjustment module 30, configured to obtain first pixel coordinates of all original pixel points in the real-time furnace image, and convert the first pixel coordinates into first physical coordinates based on the parameters of the shooting device;

[0080] A second adjustment module 40 is used to adjust the position of the first physical coordinate based on the real-time distortion parameter data group to obtain a second physical coordinate, and convert the second physical coordinate into a second pixel coordinate, where the second pixel coordinate corresponds to the first pixel coordinate;

[0081] An execution module 50 is used to obtain an adjusted pixel value corresponding to the second pixel coordinate, and form a corrected image through the second pixel coordinate and the adjusted pixel value;

[0082] The execution module 50 includes:

[0083] A fourth unit is used to restore the second pixel coordinates to the real-time furnace image to obtain a mapping pixel point, and determine whether the coordinate value of the mapping pixel point is an integer;

[0084] A fifth unit, configured to determine the pixel value of the mapping pixel point as the adjustment pixel value if the coordinate value of the mapping pixel point is an integer;

[0085] The sixth unit is used for querying the pixel points adjacent to the mapped pixel point to obtain a number of adjacent pixel points if the coordinate value of the mapped pixel point is a non-integer, and obtaining the adjusted pixel value based on the pixel values ​​of the several adjacent pixel points.

[0086] The present invention also provides a computer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for correcting the spray gun image in the blowing furnace as described in the above technical solution is implemented.

[0087] The present invention also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for correcting the spray gun image in the blowing furnace as described in the above technical solution is implemented.

[0088] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0089] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for correcting a spray gun image in a blowing furnace, characterized in that: The following steps are involved: Constructing an initial neural network model and a training data set, wherein the training data set includes a simulated distortion parameter data set and a simulated distorted image, and training the initial neural network model through the training data set to obtain a final neural network model; Acquire a real-time in-furnace image in the blowing furnace by means of a photographing device, and use the real-time in-furnace image as an input value of the final neural network model to acquire a real-time distortion parameter data set; Acquire first pixel coordinates of all original pixel points in the real-time furnace image, and convert the first pixel coordinates into first physical coordinates based on the parameters of the shooting device; Adjusting the position of the first physical coordinate based on the real-time distortion parameter data set to obtain a second physical coordinate, and converting the second physical coordinate into a second pixel coordinate, wherein the second pixel coordinate corresponds to the first pixel coordinate; An adjusted pixel value corresponding to the second pixel coordinate is obtained, and a corrected image is formed by the second pixel coordinate and the adjusted pixel value.

2. The method for correcting the spray gun image in a blowing furnace according to claim 1, characterized in that: The steps of constructing the initial neural network model are specifically as follows: The input layer, composite processing layer, fully connected layer and output layer are constructed in sequence to form an initial neural network model.

3. The method for correcting the spray gun image in a blowing furnace according to claim 1, characterized in that: The training data set is constructed, the training data set includes a simulated distortion parameter data set and a simulated distorted image, and the step of training the initial neural network model through the training data set includes: Acquire past furnace images, set a simulated distortion parameter data group, perform distortion processing on the past furnace images by using the simulated distortion parameter data group to acquire simulated distorted images, wherein the simulated distortion parameter data group and the simulated distorted images constitute a training data set; The simulated distorted image is used as the input value of the initial neural network model, and the simulated distortion parameter data set is used as the output value of the initial neural network model to train the initial neural network model.

4. The method for correcting the spray gun image in a blowing furnace according to claim 1, characterized in that: The formula for obtaining the first physical coordinate is: Among them, x w Represents the horizontal coordinate of the first physical coordinate, y w represents the ordinate of the first physical coordinate, i represents the abscissa of the first pixel coordinate, J represents the ordinate of the first pixel coordinate, c,x represents the abscissa of the center of the real-time furnace image, cy represents the ordinate of the center of the real-time furnace image, fx represents the effective focal length of the shooting device in the abscissa direction, and fy represents the effective focal length of the shooting device in the ordinate direction.

5. The method for correcting the spray gun image in a blowing furnace according to claim 1, characterized in that: The real-time distortion parameter data set includes a first radial distortion parameter, a second radial distortion parameter, a third radial distortion parameter, a first tangential distortion parameter, and a second tangential distortion parameter.

6. The method for correcting the spray gun image in a blowing furnace according to claim 5, characterized in that: The calculation formula of the second physical coordinate is: Among them, x w Represents the horizontal coordinate of the first physical coordinate, y w Indicates the ordinate of the first physical coordinate, x n Represents the horizontal coordinate of the second physical coordinate, y n represents the ordinate of the second physical coordinate, k1 represents the first radial distortion parameter, k2 represents the second radial distortion parameter, k3 represents the third radial distortion parameter, p1 represents the first tangential distortion parameter, p2 represents the second tangential distortion parameter, and r represents the distance between the original pixel point and the center point of the real-time furnace image.

7. The method for correcting the spray gun image in a blowing furnace according to claim 1, characterized in that: The step of obtaining the adjusted pixel value corresponding to the second pixel coordinate comprises: Restoring the second pixel coordinates to the real-time furnace image to obtain a mapping pixel point, and determining whether the coordinate value of the mapping pixel point is an integer; If the coordinate value of the mapped pixel point is an integer, the pixel value of the mapped pixel point is determined as the adjusted pixel value; If the coordinate value of the mapped pixel point is a non-integer, the pixel points adjacent to the mapped pixel point are queried to obtain a number of adjacent pixel points, and the adjusted pixel value is obtained based on the pixel values ​​of the several adjacent pixel points.

8. A system for correcting a spray gun image in a blowing furnace, applied to the method for correcting a spray gun image in a blowing furnace as claimed in any one of claims 1 to 7, characterized in that: The system comprises: A training module is used to construct an initial neural network model and a training data set, wherein the training data set includes a simulated distortion parameter data set and a simulated distorted image, and the initial neural network model is trained by the training data set to obtain a final neural network model; An acquisition module, used for acquiring a real-time in-furnace image in the blowing furnace through a photographing device, and using the real-time in-furnace image as an input value of the final neural network model to acquire a real-time distortion parameter data set; A first adjustment module, configured to obtain first pixel coordinates of all original pixel points in the real-time furnace image, and convert the first pixel coordinates into first physical coordinates based on the parameters of the shooting device; a second adjustment module, configured to adjust the position of the first physical coordinate based on the real-time distortion parameter data group to obtain a second physical coordinate, and convert the second physical coordinate into a second pixel coordinate, wherein the second pixel coordinate corresponds to the first pixel coordinate; The execution module is used to obtain the adjusted pixel value corresponding to the second pixel coordinate, and form a corrected image through the second pixel coordinate and the adjusted pixel value.

9. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for correcting a spray gun image in a blowing furnace according to any one of claims 1 to 7 is implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for correcting a lance image in a blowing furnace according to any one of claims 1 to 7 is implemented.

Images