Apparatus and method for manufacturing steel sheet by using rolling mill

Abstract

A method for manufacturing a steel sheet by using a rolling mill, according to an embodiment, may comprise the steps of: obtaining an image from a camera that photographs a steel sheet coming out of an outlet of a continuous rolling mill; converting the obtained image into a top-view image; detecting a residual fragment from the top-view image and generating a masking image by masking a residual fragment area including the detected residual fragment; measuring the number and sizes of the residual fragments from the masking image; and determining processing of the residual fragments on the basis of the measured number and sizes of the residual fragments.

Description

Steel plate manufacturing apparatus and method using a rolling mill

[0001] The present invention relates to an apparatus and method for manufacturing steel plates using a rolling mill, and more specifically, to an apparatus and method for manufacturing steel plates using a rolling mill that detects and processes residual fragments generated during steel plate production in a rolling mill of a rolling process at the exit side of the rolling mill.

[0002] The finishing rolling mill for hot rolling is a continuous rolling mill consisting of 7 units, and it is an operation that passes steel plates through the corresponding rolling mills to produce the target thickness and width of the steel plate.

[0003] During these rolling operations, "plate breakage" accidents occur when steel plates tear due to a combination of factors, including variations in steel plate rigidity, temperature, and thickness, as well as operational inexperience. This generates steel plate fragments or chunks known as "scraps." If these scraps are small, the operator removes them; if they are large, they are transferred to processing facilities.

[0004] Currently, since operators visually check for the presence of such residue via cameras, verification may be delayed or missed depending on their condition, such as fatigue; this leads to equipment damage, product defects, and production stoppages.

[0005] One embodiment of the present invention aims to provide an apparatus and method for manufacturing steel sheets using a rolling mill that prevents a decline in operational productivity by automatically detecting residual material at the exit side of the rolling mill in a rolling process and providing corresponding processing guidance.

[0006] Among the embodiments, a method for manufacturing a steel plate using a rolling mill may include the steps of: acquiring an image from a camera that photographs a steel plate exiting from the exit side of a continuous rolling mill; converting the acquired image into a top-view image; detecting a residue in the top-view image and generating a masked image that masks a residue area containing the detected residue; measuring the number and size of the residue from the masked image; and determining the processing of the residue based on the measured number and size of the residue.

[0007] The method further includes the step of detecting the angle of the camera photographing the steel plate and determining whether to transform the acquired image based on the detected angle, and the step of detecting the angle of the camera photographing the steel plate and determining whether to transform the acquired image based on the detected angle may include the step of determining not to transform the acquired image only when the camera is photographing the steel plate from above and the angle of the camera is perpendicular to the cross-section of the steel plate on a plane.

[0008] The step of converting the above image into a top-view image may include the step of converting the above image into the top-view image using at least one different image processing technique including a perspective transformation.

[0009] The step of converting the above image into a top-view image may include setting four reference points in the above image, additionally specifying the location where the four reference points are converted into a top-view, and calculating the conversion coordinates between the reference points through IPM (Inverse Perspective Mapping).

[0010] The step of detecting debris in the planar image and generating a masked image that masks the debris area containing the detected debris may include the step of detecting the debris area using an artificial intelligence segmentation model, marking the debris area as 1 in the masked image, and marking the remaining area excluding the debris area as 0.

[0011] The step of measuring the number and size of the fragments from the masking image may include the step of grouping the pixels of the fragment area marked with 1 in the masking image to create a plurality of pixel groups, creating a bounding box for each of the plurality of pixel groups, and calculating the size and number of the bounding boxes.

[0012] The step of determining the processing of the fragments based on the number and size of the fragments measured above may include the step of determining the processing of the fragments based on the size of the bounding box, when the size of the bounding box is greater than or equal to a preset standard size.

[0013] The step of determining the processing of the residue based on the number and size of the measured residue may further include the step of processing with a coil catcher if the size of the bounding box is greater than or equal to a preset standard size, and withholding processing of the residue and delivering it to an operator if the size of the bounding box is smaller than the standard size.

[0014] Among the embodiments, the steel plate manufacturing device using a rolling mill is a steel plate manufacturing device using a rolling mill that executes program code loaded in one or more memory devices through one or more processors to automatically detect and process scraps using an image obtained by photographing a steel plate exiting the rolling mill, wherein the program code is executed to acquire an image from a camera that photographs a steel plate exiting the continuous rolling mill, converts the acquired image into a top-view image, detects scraps in the top-view image, generates a masked image that masks a scrap area containing the detected scraps, measures the number and size of the scraps from the masked image, and determines the processing of the scraps based on the measured number and size of the scraps.

[0015] The method further includes detecting the angle of the camera photographing the steel plate and determining whether to transform the acquired image based on the detected angle, and the method of detecting the angle of the camera photographing the steel plate and determining whether to transform the acquired image based on the detected angle may include determining not to transform the acquired image only when the camera is photographing the steel plate from above and the angle of the camera is perpendicular to the cross-section of the steel plate on a plane.

[0016] Converting the above image into a top-view image may include converting the image into the top-view image using at least one different image processing technique including a perspective transformation.

[0017] Converting the above image into a top-view image may include setting four reference points in the above image, additionally specifying the locations where the four reference points are converted into a top-view, and calculating the conversion coordinates between the reference points through IPM (Inverse Perspective Mapping).

[0018] Detecting debris in the planar image and generating a masked image that masks the debris area containing the detected debris may include using an artificial intelligence segmentation model to detect the debris area, marking the debris area as 1 in the masked image, and marking the remaining area excluding the debris area as 0.

[0019] Measuring the number and size of the fragments from the masked image may include grouping the pixels of the fragment area marked with 1 in the masked image to create a plurality of pixel groups, creating a bounding box for each of the plurality of pixel groups, and calculating the size and number of the bounding boxes.

[0020] Determining the processing of the fragments based on the number and size of the fragments measured above may include determining the processing of the fragments based on the size of the bounding box, when the size of the bounding box is greater than or equal to a preset standard size.

[0021] Determining the processing of the residue based on the number and size of the residue measured above may further include processing using a coil catcher if the size of the bounding box is greater than or equal to a preset standard size, and withholding processing of the residue and delivering it to an operator if the size of the bounding box is smaller than the standard size.

[0022] A steel plate manufacturing apparatus and method using a rolling mill according to one embodiment of the present invention can prevent operational accidents and reduce the time required to process residues compared to existing methods by automatically detecting residues at the exit side of a continuous rolling mill in a hot rolling process and providing guidance on how to process the residues.

[0023] FIGS. 1 and 2 schematically show a residue processing system of a hot rolling mill according to one embodiment of the present invention.

[0024] FIG. 3 is a block diagram of a steel plate manufacturing apparatus using a rolling mill according to one embodiment of the present invention.

[0025] FIG. 4 is a flowchart of a method for manufacturing a steel plate using a rolling mill according to one embodiment of the present invention.

[0026] FIG. 5 is a flowchart of a method for manufacturing a steel plate using a rolling mill according to one embodiment of the present invention.

[0027] FIG. 6 is a drawing for explaining a computing device according to an embodiment of the present invention.

[0028] A steel plate manufacturing apparatus and method using a rolling mill can automatically detect scraps and provide guidance on a method for processing said scraps by using an image obtained by photographing a steel plate exiting the rolling mill with a camera installed at the exit side of the rolling mill.

[0029] Embodiments of the present invention are described below with reference to the attached drawings so that those skilled in the art can easily implement them. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. Furthermore, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification are denoted by similar reference numerals.

[0030] Throughout the specification and claims, when a part is described as "comprising" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components. Terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but said components are not limited by said terms. Such terms are used solely for the purpose of distinguishing one component from another.

[0031] Terms such as "...part," "...unit," and "module" as used in the specification may refer to a unit capable of processing at least one function or operation described in this specification, and may be implemented as hardware or a circuit, software, or a combination of hardware or a circuit and software.

[0032] In addition, at least some components or functions of the steel plate manufacturing apparatus and method using a rolling mill according to the embodiments described below may be implemented as a program or software, and the program or software may be stored on a computer-readable medium.

[0033] The present invention proposes a method for detecting residual material generated during the passage of a sheet metal in a continuous rolling mill of a hot rolling process at the exit side of the rolling mill and determining a processing method.

[0034] In other words, the present invention relates to a method for automatically detecting scraps and guiding a method for processing said scraps by using an image obtained by photographing a steel plate exiting a rolling mill with a camera installed at the exit side of the rolling mill.

[0035] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0036] FIGS. 1 and 2 schematically show a system for processing residual pieces of a hot rolling mill during a process according to a method for manufacturing a steel plate using a rolling mill according to one embodiment of the present invention.

[0037] The present invention according to one embodiment includes an apparatus and method for manufacturing a steel plate using a rolling mill, and more specifically, may include an apparatus and method for processing residual fragments generated during the steel plate manufacturing process.

[0038] FIG. 1 is a block diagram of a waste processing system for a hot rolling mill. FIG. 2 is a drawing for explaining a waste processing system for a hot rolling mill according to one embodiment.

[0039] Referring to FIGS. 1 and 2, a residue processing system for a hot rolling mill may include a camera (10), a continuous rolling mill (20), and a steel plate manufacturing device (100, hereinafter referred to as a residue processing device) using the rolling mill.

[0040] The camera (10), the continuous rolling mill (20), and the scrap processing device (100) can be connected to each other via a wired or wireless network to communicate with one another.

[0041] The camera (10) can photograph the steel plate discharged from the exit side of the continuous rolling mill.

[0042] That is, the camera (10) can photograph the steel plate to obtain the first image containing the steel plate.

[0043] The position of the camera (10) can be variably determined so that the steel plate can be photographed from various angles.

[0044] For example, the camera (10) can take a frontal view of the cross-section of the steel plate and can also take a view at an oblique angle.

[0045] The camera (10) can transmit an image of the steel plate to the scrap processing device (100) via a network.

[0046] A continuous rolling mill (20) is a machine used in a rolling process that sequentially rolls metal into thin layers by placing it into multiple rolling rollers arranged vertically and moving simultaneously. The type of continuous rolling mill (20) is not particularly limited.

[0047] The continuous rolling mill (20) discharges steel plates to the exit side. The continuous rolling mill (20) is connected to the waste processing device (100) via a network and can transmit and receive various operation information, etc.

[0048] The debris processing device (100) can detect debris based on an artificial intelligence model in an image captured by a camera and determine the processing of the detected debris.

[0049] Scrap edge or tail scrap refers to unnecessary or defective ends generated during the steel plate production process, or excess parts that must be cut off during processing.

[0050] In FIG. 2, the camera (10) is installed at the outermost end of the continuous rolling mill (20) to photograph the steel plate coming out of the rolling mill.

[0051] The residue processing device (100) receives the captured image, determines whether residue is present in the image, and if residue is present, calculates the size and number of residues and transmits to the operator the residues to be processed by the operator (40) and the residues to be sent to the Coil Catcher (30) on the winding side.

[0052] FIG. 3 is a block diagram of a steel plate manufacturing apparatus using a rolling mill according to one embodiment of the present invention.

[0053] A residual processing device (100) according to one embodiment can execute program code or instructions loaded into one or more memory devices through one or more processors.

[0054] For example, the fragment processing device (100) may be implemented as a computing device (900) as described below in relation to FIG. 6. In this case, one or more processors may correspond to the processor (910) of the computing device (900), and one or more memory devices may correspond to the memory (930) of the computing device (900).

[0055] Program code or instructions can be executed by one or more processors to automatically detect and process scraps using images obtained by photographing steel plates exiting a rolling mill.

[0056] In this specification, the term "module" is used to logically distinguish these functions performed by program code or instructions.

[0057] Referring to FIG. 3, the debris processing device (100) may include an image conversion module (110), a masking module (120), a debris measurement module (130), and a debris processing determination module (140).

[0058] The image conversion module (110) can acquire an image from a camera that photographs a steel plate coming out of the continuous rolling mill.

[0059] The image conversion module (110) can convert the acquired image into a top-view image.

[0060] The image conversion module (110) can detect the angle of the camera capturing the steel plate. The image conversion module (110) can determine whether to convert the acquired image based on the detected angle.

[0061] The image conversion module (110) can decide not to convert the acquired image only when the camera is photographing the steel plate from above and the angle of the camera is perpendicular to the cross-section of the steel plate on a plane.

[0062] The image conversion module (110) can convert an image into a planar image using at least one different image processing technique including a perspective transformation.

[0063] The image conversion module (110) can set four reference points in the image, additionally specify the location where the four reference points are converted into a top-view, and calculate the conversion coordinates between the reference points through IPM (Inverse Perspective Mapping).

[0064] The image conversion module (110) can generate a planar image based on the conversion coordinates.

[0065] The masking module (120) can detect debris in a planar image. The masking module (120) can generate a masked image that masks the debris area containing the detected debris.

[0066] The masking module (120) can detect fragment areas using an artificial intelligence segmentation model.

[0067] The masking module (120) can mark the residual area in the masked image as 1 and the remaining area excluding the residual area as 0.

[0068] The debris measurement module (130) can measure the number and size of debris from the masking image.

[0069] The fragment measurement module (130) groups pixels of the fragment area marked with 1 in the masked image to create multiple pixel groups.

[0070] The fragment measurement module (130) can generate a bounding box for each of the multiple pixel groups.

[0071] The fragment measurement module (130) can calculate the size and number of bounding boxes.

[0072] The fragment measurement module (130) can detect the size of each fragment indicated by a plurality of bounding boxes.

[0073] The fragment measurement module (130) can count the number of fragments indicated by a plurality of bounding boxes.

[0074] That is, the fragment measurement module (130) can estimate the number and size of fragments based on the size and number of bounding boxes.

[0075] The debris processing decision module (140) can determine the processing of debris based on the number and size of the measured debris.

[0076] The residue processing decision module (140) determines the processing of residue when the size of the bounding box is greater than or equal to a preset standard size.

[0077] For example, the residual piece processing decision module (140) decides to process using a coil catcher when the size of the bounding box is larger than a preset reference size.

[0078] The residue processing decision module (140) may withhold processing of residue when the size of the bounding box is smaller than the reference size.

[0079] That is, the residue processing decision module (140) transfers the residue to a coil catcher for processing when the size of the bounding box is greater than or equal to a preset standard size, and transfers the residue to an operator for processing when the size of the bounding box is smaller than a preset standard size.

[0080] FIG. 4 is a flowchart of a method for manufacturing a steel plate using a rolling mill according to an embodiment of the present invention. The method for manufacturing a steel plate using a rolling mill according to FIG. 4 can be performed through the residual piece processing device (100) of FIG. 1.

[0081] In FIG. 4, the debris processing device (100) acquires an image captured by a camera (step S410).

[0082] The fragment processing device (100) converts the acquired image into a planar image using an artificial intelligence model (step S420).

[0083] The debris processing device (100) creates a masked image by masking the debris area containing debris in the planar image (step S430)

[0084] The debris processing device (100) detects debris from the masking image (step S440).

[0085] The debris processing device (100) can provide guidance on debris processing based on information about the detected debris (step S450).

[0086] FIG. 5 is a flowchart of a method for manufacturing a steel plate using a rolling mill according to an embodiment of the present invention. The method for manufacturing a steel plate using a rolling mill of FIG. 5 can be performed through the residual piece processing device (100) of FIG. 1.

[0087] In FIG. 5, the scrap processing device (100) can acquire an image from a camera that photographs a steel plate exiting from the continuous rolling mill (step S510).

[0088] The fragment processing device (100) can detect the angle of the camera that photographs the steel plate and determine whether to convert the acquired image based on the detected angle.

[0089] The fragment processing device (100) can decide not to convert the acquired image only when the camera is photographing the steel plate from above and the angle of the camera is perpendicular to the cross-section of the steel plate on a plane.

[0090] In other cases, the fragment processing device (100) can convert the acquired image into a top-view image (step S520).

[0091] The fragment processing device (100) can convert an image into a planar image using at least one image processing technique including perspective transformation.

[0092] In step S520, the fragment processing device (100) first switches the captured image to Top View to remove distortion of the image captured from a diagonal direction.

[0093] The debris processing device (100) sets four reference points in the captured image and additionally specifies the position when these four points are converted to Top View.

[0094] The fragment processing device (100) calculates a transformation matrix through the coordinate transformation between them using an Inverse Perspective Mapping function.

[0095] And, the fragment processing device (100) applies the transformation matrix to the entire image to perform a Top View transformation.

[0096] For example, the scrap processing device (100) can designate four vertices of the Roller Table through which the steel plate and scraps pass, automatically calculate the optimal Top View position so that the four points are positioned in the center of the Top View image, and then apply Inverse Perspective Mapping.

[0097] The debris processing device (100) can detect debris in a planar image and generate a masked image that masks the debris area containing the detected debris (step S530).

[0098] The debris processing device (100) can detect a debris region in a masked image using an artificial intelligence segmentation model, mark the debris region as 1, and mark the remaining area excluding the debris region as 0.

[0099] That is, in step S530, the debris processing device (100) can mask the area where debris exists in pixel units using a Segmentation AI model on the Top View image obtained in this way.

[0100] A segmentation AI model can be a classification or segmentation model trained on fragment samples. In other words, by learning from various fragments, a segmentation AI model can detect only fragments while ignoring the general steel plate and background.

[0101] Additionally, the residue processing device (100) can perform a filtering operation in which only values ​​in the corresponding detection area where the Red value is greater than the Green value and the Blue value are kept.

[0102] The debris processing device (100) can reduce false detections in the Segmentation AI model and increase the detection rate of debris through this.

[0103] The debris processing device (100) can measure the number and size of debris from the masking image (step S540).

[0104] The debris processing device (100) can group pixels of the debris area marked with 1 in the masked image to create multiple pixel groups and create a bounding box for each of the multiple pixel groups.

[0105] In step S540, the debris processing device (100) groups the pixels of the detection area to create a bounding box, and then calculates the number of debris and the area of ​​each.

[0106] When processing debris, since the shape, such as whether it is elongated on one side, is important along with the actual area of ​​the debris, the debris processing device (100) can calculate the pixel-based actual area and the bounding box-based area of ​​the debris together and transmit them to the driver.

[0107] The debris processing device (100) can determine a method of processing the debris based on the number and size of the measured debris (step S550).

[0108] The debris processing device (100) determines how to process the debris when the size of the bounding box is greater than or equal to a preset standard size.

[0109] The waste disposal device (100) can process the waste with a coil catcher if the size of the bounding box is larger than a preset standard size, and can withhold processing of the waste and deliver it to an operator if the size of the bounding box is smaller than the standard size.

[0110] In step S550, the debris processing device (100) provides debris processing guidance, and can guide the operator on a processing method according to a Rule based on the previously obtained information regarding the number of debris, actual area, and width. For example, the debris processing device (100) conveys the relevant information through a separate HMI screen.

[0111] FIG. 6 is a drawing for explaining a computing device according to an embodiment of the present invention.

[0112] Referring to FIG. 6, a steel plate manufacturing apparatus and method using a rolling mill according to embodiments can be implemented using a computing device (900).

[0113] The computing device (900) may include at least one of a processor (910), memory (930), user interface input device (940), user interface output device (950), and storage device (560) that communicate via a bus (920). The computing device (900) may also include a network interface (970) that is electrically connected to a network (90). The network interface (970) may transmit or receive signals to or from other entities via the network (90).

[0114] The processor (910) can be implemented in various types such as an MCU (Micro Controller Unit), AP (Application Processor), CPU (Central Processing Unit), GPU (Graphic Processing Unit), NPU (Neural Processing Unit), etc., and may be any semiconductor device that executes instructions stored in memory (930) or storage device (960). The processor (910) may be configured to implement the functions and methods described above in relation to FIGS. 1 to 5.

[0115] The memory (930) and storage device (960) may include various forms of volatile or non-volatile storage media. For example, the memory may include ROM (read-only memory) (931) and RAM (random access memory) (932). In this embodiment, the memory (930) may be located inside or outside the processor (910), and the memory (930) may be connected to the processor (910) through various known means.

[0116] In some embodiments, at least some components or functions of the steel plate manufacturing apparatus and method using a rolling mill according to the embodiments may be implemented as a program or software executed on a computing device (900), and the program or software may be stored on a computer-readable medium.

[0117] In some embodiments, at least some components or functions of the steel plate manufacturing apparatus and method using a rolling mill according to the embodiments may be implemented using hardware or circuits of a computing device (900), or may be implemented using separate hardware or circuits that can be electrically connected to the computing device (900).

[0118] Although embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements by those skilled in the art to which the present invention belongs, utilizing the basic concept of the present invention as defined in the following claims, also fall within the scope of the present invention.

[0119] The apparatus and method for manufacturing steel plates using a rolling mill have industrial applicability by automatically detecting residues at the exit of a continuous rolling mill in a hot rolling process and guiding the method of processing the residues, thereby preventing operational accidents through rapid residue processing by the operator and reducing the residue processing time compared to existing methods.

Claims

1. A step of acquiring an image from a camera that photographs a steel plate exiting from the exit side of a continuous rolling mill; A step of converting the acquired image into a top-view image; A step of detecting fragments in the above planar image and generating a masked image that masks the fragment area containing the detected fragments; A step of measuring the number and size of the fragments from the masking image; and A method for manufacturing a steel plate using a rolling mill, comprising the step of determining the treatment of the residue based on the number and size of the residues measured above.

2. In Paragraph 1, The method further includes the step of detecting the angle of the camera capturing the steel plate and determining whether to convert the acquired image based on the detected angle. The step of detecting the angle of the camera capturing the steel plate and determining whether to convert the acquired image based on the detected angle is A step comprising determining not to transform the acquired image only when the camera is photographing the steel plate from above and the angle of the camera is perpendicular to the cross-section of the steel plate on a plane, Method for manufacturing steel plates using a rolling mill.

3. In Paragraph 1, The step of converting the above image into a top-view image is, A method comprising the step of converting the image into the planar image using at least one different image processing technique including a perspective transformation. Method for manufacturing steel plates using a rolling mill.

4. In Paragraph 1, The step of converting the above image into a top-view image is, A method comprising the step of setting four reference points in the above image, additionally specifying the location where the four reference points are converted into a top-view, and calculating the converted coordinates between the reference points through IPM (Inverse Perspective Mapping). Method for manufacturing steel plates using a rolling mill.

5. In Paragraph 1, The step of detecting fragments in the above planar image and generating a masked image that masks the fragment area containing the detected fragments is as follows: A method comprising the step of detecting the fragment area using an artificial intelligence segmentation model, marking the fragment area as 1 in the masked image, and marking the remaining area excluding the fragment area as 0. Method for manufacturing steel plates using a rolling mill.

6. In Paragraph 5, The step of measuring the number and size of the fragments from the masked image is, A step of grouping pixels of the fragment area marked with 1 in the masked image to generate a plurality of pixel groups, and generating a bounding box for each of the plurality of pixel groups; and A method comprising the step of calculating the size and number of the bounding boxes. Method for manufacturing steel plates using a rolling mill.

7. In Paragraph 6, The step of determining the treatment of the residue based on the number and size of the residue measured above is: A step comprising determining the processing of the residue based on the size of the bounding box, if the size of the bounding box is greater than or equal to a preset standard size. Method for manufacturing steel plates using a rolling mill.

8. In Paragraph 7, The step of determining the treatment of the residue based on the number and size of the residue measured above is: If the size of the above bounding box is greater than or equal to a preset standard size, it is processed as a coil catcher, and If the size of the bounding box is smaller than the standard size, the method further includes the step of withholding processing of the residue and delivering it to the operator. Method for manufacturing steel plates using a rolling mill.

9. A steel plate manufacturing apparatus using a rolling mill that executes program code loaded into one or more memory devices through one or more processors to automatically detect and process scraps using an image obtained by photographing a steel plate exiting the rolling mill, The above program code is executed, An image is acquired from a camera that photographs a steel plate exiting the continuous rolling mill, and Convert the acquired image above into a top-view image, and Detecting fragments in the above planar image, and generating a masked image that masks the fragment area containing the detected fragments, The number and size of the fragments are measured from the masked image above, and A steel plate manufacturing apparatus using a rolling mill that determines the treatment of the residue based on the number and size of the residues measured above.

10. In Paragraph 9, It further includes detecting the angle of the camera capturing the steel plate and determining whether to convert the acquired image based on the detected angle, and Detecting the angle of the camera capturing the steel plate and determining whether to convert the acquired image based on the detected angle is, Determining not to transform the acquired image only when the camera is photographing the steel plate from above and the angle of the camera is perpendicular to the cross-section of the steel plate on a plane, Steel plate manufacturing device using a rolling mill.

11. In Paragraph 9, Converting the above image into a top-view image is, A method comprising converting the image into the planar image using at least one different image processing technique including a perspective transformation. Steel plate manufacturing device using a rolling mill.

12. In Paragraph 9, Converting the above image into a top-view image is, A method comprising setting four reference points in the above image, additionally specifying the location where the four reference points are converted to a top-view, and calculating the transformed coordinates between the reference points through IPM (Inverse Perspective Mapping). Steel plate manufacturing device using a rolling mill.

13. In Paragraph 9, Detecting fragments in the above planar image and generating a masked image that masks the fragment area containing the detected fragments is, The method includes detecting the fragment area using an artificial intelligence segmentation model, marking the fragment area as 1 in the masked image, and marking the remaining area excluding the fragment area as 0. Steel plate manufacturing device using a rolling mill.

14. In Paragraph 13, Measuring the number and size of the fragments from the masked image above is, In the above masked image, pixels of the fragment area marked with 1 are grouped to create a plurality of pixel groups, and a bounding box is created for each of the plurality of pixel groups, and including calculating the size and number of the bounding boxes above. Steel plate manufacturing device using a rolling mill.

15. In Paragraph 14, Determining the treatment of the residue based on the number and size of the residue measured above is, Based on the size of the bounding box, determining the processing of the residue when the size of the bounding box is greater than or equal to a preset standard size, Steel plate manufacturing device using a rolling mill.

16. In Paragraph 15, Determining the treatment of the residue based on the number and size of the residue measured above is, If the size of the above bounding box is greater than or equal to a preset standard size, it is processed using a coil catcher, and If the size of the bounding box is smaller than the standard size, the method further includes withholding processing of the residue and delivering it to the operator. Steel plate manufacturing device using a rolling mill.

Images