JP7743367B2 - Molten metal occupancy rate measuring device and method - Google Patents

Description

The present invention relates to a molten metal occupancy rate measuring device and a molten metal occupancy rate measuring method for determining the occupancy rate of molten metal in a vessel containing the molten metal.

Slag floats on the surface of molten metal, such as molten pig iron removed from a blast furnace into a ladle, or molten steel removed from a converter into a ladle. This slag is removed because it affects product quality and yield. To remove this slag, it is necessary to determine whether it is molten metal or slag. For example, an operator (worker) refers to a sample of slag and removes any that are determined to be slag. In this case, there is a risk that the determination will vary depending on the subjective judgment of each operator, and that the slag may not be removed appropriately. For this reason, an objective method of determining slag is desired, and one such technique is disclosed in Patent Document 1.

The slag quantification method disclosed in Patent Document 1 includes a brightness value calculation step of capturing an image of the surface of molten metal contained in a vessel using an imaging device and determining the brightness value in each pixel region of the captured image through image processing; a slag surface temperature calculation step of determining the relationship between brightness values and radiance in advance and determining the surface temperature of the slag in each pixel region using the brightness values determined in the brightness value calculation step based on this relationship and a temperature conversion equation; a slag thickness calculation step of estimating the slag thickness for each pixel region using the slag surface temperature for each pixel region determined in the slag surface temperature calculation step and a heat conduction equation; and a slag calculation step of determining the slag volume or total weight in the vessel based on the molten metal surface area per pixel region and the physical properties of the slag.

Japanese Patent Application Laid-Open No. 2020-112429

The slag quantification method disclosed in Patent Document 1 calculates the slag volume or total slag weight in a vessel by estimating the slag thickness for each pixel based on the relationship between the brightness value and radiance calculated in advance and the slag surface temperature calculated from a temperature conversion equation. However, there is room for improvement in terms of accuracy. In particular, when so-called thin slag occurs, the slag quantification method disclosed in Patent Document 1 has difficulty detecting the thin slag, which may result in a decrease in accuracy. Ordinary slag is composed of high-temperature molten oxides, such as silicon dioxide (SiO ₂ ), aluminum oxide (alumina, Al ₂ O ₃ ), and calcium oxide (CaO), as well as graphite precipitated from the molten iron. It exhibits a black color formed from a viscous liquid phase and powder. On the other hand, thin slag has a liquid phase with significantly less graphite than ordinary slag, and its color is lighter and appears similar to that of molten iron at the surface. For this reason, it is more difficult to distinguish thin slag from molten iron than normal slag, making it difficult to detect thin slag.

The present invention was made in consideration of the above-mentioned circumstances, and its purpose is to provide a molten metal occupancy rate measuring device and a molten metal occupancy rate measuring method that can more accurately determine the occupancy rate of an image area that contains molten metal, rather than slag.

After extensive investigation, the inventors have found that the above-mentioned object can be achieved by the present invention. Specifically, a molten metal occupancy rate measuring device according to one aspect of the present invention includes an imaging unit that generates a color image of a bird's-eye view of the interior of a vessel containing molten metal; a first region extraction unit that determines, based on a green image of the green component in the color image generated by the imaging unit, a slag image region determined to be an image region containing slag and a molten metal candidate image region determined to be a candidate image region containing the molten metal; a second region extraction unit that determines, based on a red image of the red component in the color image generated by the imaging unit, a slag molten metal image region determined to be an image region containing at least one of the slag and the molten metal from the slag image region and the molten metal candidate image region determined by the first region extraction unit; and an occupancy rate processing unit that calculates the occupancy rate of the image region containing the molten metal as a bare metal area rate based on the slag image region and the molten metal candidate image region determined by the first region extraction unit and the slag molten metal image region determined by the second region extraction unit. Preferably, in the above-described molten metal occupancy rate measuring device, the occupancy rate processing unit determines a residual image area by excluding the slag molten metal image area determined by the second area extraction unit from the slag image area and molten metal candidate image area determined by the first area extraction unit, determines the image area by excluding the residual image area from the molten metal candidate image area determined by the first area extraction unit as the molten metal image area, and determines the occupancy rate of the determined molten metal image area as the bare metal area ratio.

The inventors studied a color image of the interior of a vessel containing molten metal and discovered that more accurate determination of the molten metal area could be achieved by using a color image obtained by dividing the color image into its respective color components, rather than the color image itself. Based on this finding, the molten metal occupancy rate measuring device calculates the occupancy rate of the image area containing the molten metal as the bare metal area rate, thereby more accurately determining the occupancy rate of the image area containing molten metal, rather than slag. In other words, the molten metal occupancy rate measuring device can more accurately detect image areas containing slag (including normal slag and thin slag).

In another aspect, the above-mentioned molten metal occupancy rate measuring device further includes a calculation execution determination unit that determines whether or not to calculate the bare metal area rate based on the brightness value within a predetermined execution determination image area in the color image generated by the imaging unit.

The bare metal area ratio is determined primarily to properly remove slag, so it only needs to be determined during the slag removal process, and does not need to be determined when the slag removal process is not in progress. Furthermore, the slag removal process is often performed by tilting the container containing the molten metal. If the imaging unit is fixedly positioned relative to the container, tilting the container changes the position of the image area capturing the interior of the container in the color image generated by the imaging unit, thereby changing the brightness value in a predetermined image area (the execution determination image area). The molten metal occupancy rate measuring device further includes a calculation execution determination unit, which can determine whether the slag removal process is in progress, making it possible to determine the bare metal area ratio only during the slag removal process.

In another aspect, in the above-described molten metal occupancy rate measuring device, the first region extraction unit determines the slag image region and the molten metal candidate image region within a predetermined extracted image region in the color image, and the second region extraction unit determines the slag molten metal image region within the extracted image region in the color image. The device further includes a region adjustment unit that adjusts the extracted image regions used by the first and second region extraction units based on the color image generated by the imaging unit. Preferably, in the above-described molten metal occupancy rate measuring device, the extracted image region is represented by a mask image that masks an image, and the first region extraction unit extracts the extracted image region by masking the color image or a green image of the green component in the color image with the mask image, and determines the slag image region and the molten metal candidate image region within this extracted extracted image region. The second region extraction unit extracts the extracted image region by masking the color image or a red image of the red component in the color image with the mask image, and determines the slag molten metal image region within this extracted extracted image region. Preferably, in the above-described molten metal occupancy rate measuring device, the region adjustment unit selects one extracted image area from a plurality of different extracted image areas as the extracted image area to be used by each of the first and second region extraction units based on the color image generated by the imaging unit. Preferably, in the above-described molten metal occupancy rate measuring device, the imaging unit is positioned so that its optical axis is fixed, the extracted image area is an internal image area that captures the internal area of the container as viewed from above by the imaging unit and is represented by a mask image that masks the image, the plurality of extracted image areas are represented by a plurality of mask images corresponding to each of a plurality of tilt angles of the container, and the region adjustment unit selects, from a plurality of mask images corresponding to each of the plurality of extracted image areas, a mask image that best matches the internal image area capturing the internal area of the container in the color image generated by the imaging unit as the mask image for the extracted image area to be used by each of the first and second region extraction units.

The slag removal operation is not always performed with the container tilted at a fixed angle. The molten metal occupancy rate measurement device further includes an area adjustment unit, which makes it possible to adjust the extracted image area according to the actual angle at which the container is tilted, thereby more accurately determining the occupancy rate of the image area that contains molten metal, rather than slag.

A molten metal occupancy rate measurement method according to another aspect of the present invention includes an imaging step of generating a color image of a bird's-eye view of the interior of a vessel containing molten metal; a first region extraction step of determining, based on a green image of the green component in the color image generated in the imaging step, a slag image region determined to be an image region containing slag and a molten metal candidate image region determined to be a candidate image region containing the molten metal; a second region extraction step of determining, based on a red image of the red component in the color image generated in the imaging step, a slag molten metal image region determined to be an image region containing at least one of the slag and the molten metal from the slag image region and the molten metal candidate image region determined in the first region extraction step; and an occupancy rate processing step of determining, as a bare metal area rate, the occupancy rate of the image region containing the molten metal based on the slag image region and the molten metal candidate image region determined in the first region extraction step and the slag molten metal image region determined in the second region extraction step.

Based on the above-mentioned findings, this molten metal occupancy rate measurement method calculates the occupancy rate of the image area where the molten metal is reflected as the bare metal area rate, so it can more accurately determine the occupancy rate of the image area where molten metal, rather than slag, is reflected. In other words, the above-mentioned molten metal occupancy rate measurement method can more accurately detect the image area where slag is reflected.

A molten metal occupancy measurement device according to another aspect of the present invention comprises an imaging unit that generates a color image of the interior of a vessel containing molten metal, a first region extraction unit that determines a candidate molten metal image region that is determined to be a candidate image region that captures the molten metal based on a green image of the green component in the color image generated by the imaging unit, a second region extraction unit that determines a thin slag image region that is determined to be an image region that captures the thin slag from the candidate molten metal image region determined by the first region extraction unit based on a red image of the red component in the color image generated by the imaging unit, and a occupancy processing unit that determines the occupancy of the image region that captures the molten metal as a bare metal area ratio based on the candidate molten metal image region determined by the first region extraction unit and the thin slag image region determined by the second region extraction unit, wherein the thin slag is formed from the molten oxide without including the graphite powder, as opposed to ordinary slag that is formed from a molten oxide and graphite powder. Preferably, in the above-described molten metal occupancy rate measurement device, the occupancy rate processing unit determines the image area obtained by excluding the thin slag image area obtained by the second area extraction unit from the molten metal candidate image area obtained by the first area extraction unit as the molten metal image area in which the molten metal is captured, and determines the occupancy rate of the determined molten metal image area as the bare metal area ratio.

Based on the above findings, this type of molten metal occupancy measurement device calculates the occupancy of the image area where the molten metal is reflected as the bare metal area ratio, and can therefore more accurately determine the occupancy of the image area where molten metal, rather than slag, is reflected. In other words, the above-mentioned molten metal occupancy measurement device can more accurately detect image areas where slag is reflected. Furthermore, since the red image can distinguish between slag molten metal image areas and thin slag image areas, the second area extraction unit can be defined from the perspective of slag molten metal image areas, but can also be defined from the perspective of thin slag image areas.

In another aspect, the above-mentioned molten metal occupancy rate measuring device further includes a calculation execution determination unit that determines whether or not to calculate the bare metal area rate based on the brightness value within a predetermined execution determination image area in the color image generated by the imaging unit.

This type of molten metal area ratio measuring device further includes a calculation execution determination unit, which makes it possible to determine whether or not the slag removal operation is in progress, and to calculate the bare metal area ratio only during the slag removal operation.

In another aspect, in the above-described molten metal occupancy rate measuring device, the first region extraction unit determines the molten metal candidate image region within a predetermined extracted image region in the color image, and the second region extraction unit determines the thin slag image region within the extracted image region in the color image. The device further includes a region adjustment unit that adjusts the extracted image regions used by the first and second region extraction units based on the color image generated by the imaging unit. Preferably, in the above-described molten metal occupancy rate measuring device, the extracted image region is represented by a mask image that masks an image, and the first region extraction unit extracts the extracted image region by masking the color image or a green image of the green component in the color image with the mask image and determines the molten metal candidate image region within the extracted extracted image region. The second region extraction unit extracts the extracted image region by masking the color image or a red image of the red component in the color image with the mask image and determines the thin slag image region within the extracted extracted image region. Preferably, in the above-described molten metal occupancy rate measuring device, the region adjustment unit selects one extracted image area from a plurality of different extracted image areas as the extracted image area to be used by each of the first and second region extraction units based on the color image generated by the imaging unit. Preferably, in the above-described molten metal occupancy rate measuring device, the imaging unit is positioned so that its optical axis is fixed, the extracted image area is an internal image area that captures the internal area of the container as viewed from above by the imaging unit and is represented by a mask image that masks the image, the plurality of extracted image areas are represented by a plurality of mask images corresponding to each of a plurality of tilt angles of the container, and the region adjustment unit selects, from a plurality of mask images corresponding to each of the plurality of extracted image areas, a mask image that best matches the internal image area capturing the internal area of the container in the color image generated by the imaging unit as the mask image for the extracted image area to be used by each of the first and second region extraction units.

This type of molten metal occupancy rate measuring device further includes an area adjustment unit, which makes it possible to adjust the extracted image area according to the actual tilt angle of the container, thereby more accurately determining the occupancy rate of the image area that contains molten metal, rather than slag.

A molten metal occupancy measurement method according to another aspect of the present invention includes an imaging step of generating a color image of the interior of a vessel containing molten metal, a first region extraction step of determining a molten metal candidate image region determined to be a candidate image region that captures the molten metal based on a green image of the green component in the color image generated in the imaging step, a second region extraction step of determining a thin slag image region determined to be an image region that captures the thin slag from the molten metal candidate image region determined in the first region extraction step based on a red image of the red component in the color image generated by the imaging unit, and a region extraction processing step of determining the region occupancy of the image region that captures the molten metal as a bare metal area ratio based on the molten metal candidate image region determined in the first region extraction step and the thin slag image region determined in the second region extraction step. The thin slag is formed from a molten oxide that does not contain graphite powder, as opposed to ordinary slag formed from a molten oxide and graphite powder.

Based on the above-mentioned findings, this molten metal occupancy rate measurement method calculates the occupancy rate of the image area where the molten metal is reflected as the bare metal area rate, so it can more accurately determine the occupancy rate of the image area where molten metal, rather than slag, is reflected. In other words, the above-mentioned molten metal occupancy rate measurement method can more accurately detect the image area where slag is reflected.

The molten metal occupancy rate measuring device and molten metal occupancy rate measuring method of the present invention can more accurately determine the occupancy rate of an image area that contains molten metal, rather than slag.

1 is a block diagram showing the configuration of a molten metal occupancy rate measuring device according to an embodiment. 3 is a diagram illustrating the arrangement of an imaging unit with respect to a chicken pot, which is an example of a container, in the molten metal occupancy rate measuring device. FIG. 4 is a diagram for explaining a calculation execution determination unit of the molten metal occupancy rate measuring device. FIG. 3 is a diagram for explaining an area adjusting unit of the molten metal occupancy rate measuring device. FIG. 10A and 10B are diagrams for explaining a mask image used in the region adjustment unit; 3A and 3B are diagrams showing an example of a color image generated by an imaging unit of the molten metal occupancy rate measuring device and a color image generated from the color image. 10A and 10B are diagrams for explaining a first method for extracting a slag image area and a molten metal candidate image area in the molten metal occupancy rate measuring device. 10A and 10B are diagrams for explaining a second extraction method of a slag image area and a molten metal candidate image area in the molten metal occupancy rate measuring device. 4 is a diagram for explaining a method for extracting a slag molten metal image region in the molten metal occupancy rate measuring device. FIG. 4 is a flowchart showing the operation of the molten metal occupancy rate measuring device. FIG. 10 is a diagram showing the change over time in the bare bath area ratio as an example. As an example, this is an explanatory diagram showing the distribution of slag and molten iron with respect to pixel values for each color image at each temperature.

One or more embodiments of the present invention will be described below with reference to the drawings. However, the scope of the invention is not limited to the disclosed embodiments. Components with the same reference numerals in each drawing are identical components, and their description will be omitted where appropriate. In this specification, generic reference numerals are used without subscripts to refer to components, and subscripted reference numerals are used to refer to individual components.

In one embodiment, the molten metal occupancy rate measuring device includes an imaging unit that generates a color image of the interior of a vessel containing molten metal, a first region extraction unit that determines, based on a green image of the green component in the color image generated by the imaging unit, a slag image region that is determined to be an image region containing slag and a molten metal candidate image region that is determined to be a candidate image region containing the molten metal, a second region extraction unit that determines, based on a red image of the red component in the color image generated by the imaging unit, a slag molten metal image region that is determined to be an image region containing at least one of the slag and the molten metal from the slag image region and the molten metal candidate image region determined by the first region extraction unit, and a occupancy rate processing unit that determines, as a bare metal area rate, the occupancy rate of the image region containing the molten metal based on the slag image region and the molten metal candidate image region determined by the first region extraction unit and the slag molten metal image region determined by the second region extraction unit. Alternatively, the molten metal occupancy rate measuring device includes an imaging unit that generates a color image of the inside of a container containing molten metal, a first area extraction unit that determines a molten metal candidate image area that is determined to be a candidate image area that captures the molten metal based on a green image of the green component in the color image generated by the imaging unit, a second area extraction unit that determines a thin slag image area that is determined to be an image area that captures the thin slag among the molten metal candidate image areas determined by the first area extraction unit based on a red image of the red component in the color image generated by the imaging unit, and an occupancy rate processing unit that calculates the occupancy rate of the image area that captures the molten metal as a bare metal area rate based on the molten metal candidate image area determined by the first area extraction unit and the thin slag image area determined by the second area extraction unit, and the thin slag is formed from the molten oxide without containing the graphite powder, unlike ordinary slag formed from a molten oxide and graphite powder. Below, we will explain in more detail this molten metal occupancy rate measurement device and the molten metal occupancy rate measurement method implemented in it.

Figure 1 is a block diagram showing the configuration of a molten metal occupancy rate measuring device in an embodiment. Figure 2 is a diagram illustrating the arrangement of an imaging unit relative to a ladle, which is an example of a vessel, in the molten metal occupancy rate measuring device. Figure 2A shows a side view. Figure 2B is a diagram illustrating the height of the imaging unit relative to the ladle in a vertical plane, and Figure 2C is a diagram illustrating the orientation (angle) of the imaging unit relative to the ladle in a horizontal plane. Figure 3 is a diagram illustrating the calculation execution determination unit of the molten metal occupancy rate measuring device. Figure 3A shows an example of a color image generated by the imaging unit 1 when slag removal work is being performed (during slag removal work), and Figure 3B shows an example of a color image generated by the imaging unit 1 when slag removal work is not being performed (during non-slag removal work). Figure 4 is a diagram illustrating the area adjustment unit of the molten metal occupancy rate measuring device. FIG. 4A is a diagram illustrating the starting point SP of the comparison line CLt in the color image generated by the image capture unit 1, and FIG. 4B is a diagram illustrating the comparison line CLt in the color image generated by the image capture unit 1. FIG. 5 is a diagram illustrating a mask image used in the region adjustment unit. FIG. 5A shows, as an example, a mask image MPa when the inclination angle of the ladle IA is 32°, and FIG. 5F shows, as an example, a color image PCa generated by the image capture unit 1 when the inclination angle of the ladle IA is 32°. FIG. 5B shows, as an example, a mask image MPb when the inclination angle of the ladle IA is 35°, and FIG. 5G shows, as an example, a color image PCb generated by the image capture unit 1 when the inclination angle of the ladle IA is 35°. FIG. 5C shows, as an example, a mask image MPc when the inclination angle of the ladle IA is 38°, and FIG. 5H shows, as an example, a color image PCc generated by the image capture unit 1 when the inclination angle of the ladle IA is 38°. FIG. 5D shows an example of a mask image MPd when the inclination angle of the ladle IA is 41°, and FIG. 5I shows an example of a color image PCd generated by the image capture unit 1 when the inclination angle of the ladle IA is 41°. FIG. 5E shows an example of a mask image MPe when the inclination angle of the ladle IA is 44°, and FIG. 5J shows an example of a color image PCe generated by the image capture unit 1 when the inclination angle of the ladle IA is 44°. FIG. 6 shows an example of a color image generated by the image capture unit of the molten metal occupancy rate measurement device and a color image generated from the color image. FIG. 6A shows an example of the color image, FIG. 6B shows an example of a green image generated from the color image shown in FIG. 6A, and FIG. 6C shows an example of a red image generated from the color image shown in FIG. 6A. FIG. 7 is a diagram for explaining a first method for extracting a slag image region and a molten metal candidate image region in the molten metal occupancy rate measurement device. FIG. 7A is a diagram illustrating a method for distinguishing between slag image regions and molten metal candidate image regions using a first region determination threshold Thr1, and FIG. 7B is a diagram illustrating a first method for generating the first region determination threshold Thr1. FIG. 8 is a diagram illustrating a second method for extracting slag image regions and molten metal candidate image regions in the molten metal occupancy rate measurement device. FIG. 8A is a diagram illustrating a method for distinguishing between slag image regions and molten metal candidate image regions using a first region determination threshold Thr1, and FIG. 8B is a diagram illustrating a second method for generating the first region determination threshold Thr1. FIG. 9 is a diagram illustrating a method for extracting slag and molten metal image regions in the molten metal occupancy rate measurement device.

In this embodiment, the molten metal occupancy measurement device D includes, for example, an imaging unit 1, a control processing unit 2, an input unit 3, an output unit 4, an interface unit (IF unit) 5, and a memory unit 6, as shown in FIG. 1.

The imaging unit 1 is connected to the control processing unit 2 and, under the control of the control processing unit 2, generates a color image of the interior of a vessel containing molten metal. The molten metal may be any molten metal (including alloys) with floating slag on its surface, such as molten pig iron. The imaging unit 1 is, for example, a color digital camera equipped with an imaging optical system that forms an optical image of the target on a predetermined imaging plane, an area image sensor whose light-receiving surface is aligned with the imaging plane and converts the optical image of the target into an electrical signal, and an image processing unit that processes the output of the area image sensor to generate image data representing the image of the target. The vessel may be any vessel, but one example is a ladle. For example, as shown in Figure 2, the imaging unit 1 is positioned above the cylindrical, bottomed ladle IA so that the imaging unit can provide an overhead view of the interior of the ladle IA, with its optical axis aligned with the central axis of the ladle IA in its non-slag-removing position. Here, the imaging unit 1 is preferably positioned at a height relative to the dragger DR that can avoid the dust and smoke generated when the dragger DR removes sludge, as shown in Figure 2B, and is preferably positioned at an angle relative to the dragger DR that can avoid the dust and smoke, as shown in Figure 2C, for example.

The input unit 3 is connected to the control processing unit 2 and is a device that inputs various commands, such as a command to start monitoring, and various data required to operate the molten metal occupancy rate measuring device D, such as the temperature of the molten metal, into the molten metal occupancy rate measuring device D. This device may be, for example, a number of input switches assigned with specific functions, a keyboard, or a mouse. The output unit 4 is connected to the control processing unit 2 and is a device that outputs the commands, data, bare metal area ratio, etc. input from the input unit 3 in accordance with the control of the control processing unit 2. This device may be, for example, a display device such as a CRT display, LCD (liquid crystal display), or organic EL display, or a printing device such as a printer.

The input unit 3 and output unit 4 may be configured as a touch panel. In this touch panel configuration, the input unit 3 is a position input device, such as a resistive or capacitive type, that detects and inputs an operation position, and the output unit 4 is a display device. In this touch panel, a position input device is provided on the display surface of the display device, and one or more input content candidates that can be input to the display device are displayed. When a user touches the display position showing the input content they want to input, the position is detected by the position input device, and the display content displayed at the detected position is input to the molten metal occupancy measurement device D as the user's operation input content. Such a touch panel makes it easy for users to intuitively understand input operations, providing a molten metal occupancy measurement device D that is easy for users to use.

The IF unit 5 is connected to the control processing unit 2 and is a circuit that inputs and outputs data to and from external devices under the control of the control processing unit 2. Examples of such circuits include an RS-232C serial communication interface circuit, an interface circuit using the Bluetooth (registered trademark) standard, and an interface circuit using the USB standard. The IF unit 5 may also be a communications interface circuit that transmits and receives communications signals to and from external devices, such as a data communications card or a communications interface circuit conforming to the IEEE 802.11 standard.

The storage unit 6 is connected to the control processing unit 2 and is a circuit that stores various predetermined programs and various predetermined data under the control of the control processing unit 2. The various predetermined programs include, for example, a control processing program, which controls each of the units 1, 3-6 of the molten metal occupancy rate measuring device D according to the function of each unit; a calculation execution determination program that determines whether to obtain a bare metal area ratio based on the brightness value within a predetermined execution determination image area in the color image generated by the imaging unit 1; a first area extraction program that determines a slag image area determined to be an image area in which slag is captured and a molten metal candidate image area determined to be a candidate image area in which molten metal is captured based on the green image of the green component in the color image generated by the imaging unit 1; and a red component in the color image generated by the imaging unit 1. The programs include a second area extraction program that determines, based on an image, a slag molten metal image area that is determined to be an image area that captures at least one of the slag and the molten metal from among the slag image area and the molten metal candidate image area determined by the first area extraction program; an area adjustment program that adjusts the extracted image areas used in each of the first and second area extraction programs based on a color image generated by the imaging unit 1; and an occupancy processing program that determines the occupancy rate of the image area that captures the molten metal as a bare metal area rate based on the slag image area and the molten metal candidate image area determined by the first area extraction program and the slag molten metal image area determined by the second area extraction program. The first region extraction program may be a program for determining a molten metal candidate image region determined to be a candidate image region containing the molten metal based on a green image of a green component in the color image generated by the imaging unit 1. The second region extraction program may be a program for determining a thin slag image region determined to be a candidate image region containing the thin slag based on a red image of a red component in the color image generated by the imaging unit 1. The occupancy rate processing program may be a program for determining the occupancy rate of the image region containing the molten metal as a bare metal area ratio based on the molten metal candidate image region determined by the first region extraction program and the thin slag image region determined by the second region extraction program. The various predetermined data include, for example, data required for executing each of these programs, such as a mask image representing the extracted image region, various thresholds, the color image captured by the imaging unit 1, and a color image of a predetermined color component. Such storage unit 6 may include, for example, a non-volatile storage element such as ROM (Read Only Memory) or a rewritable non-volatile storage element such as EEPROM (Electrically Erasable Programmable Read Only Memory). Storage unit 6 also includes RAM (Random Access Memory), which serves as the working memory of control processing unit 2 and stores data generated during execution of the specified program. Storage unit 6 may also be configured with a hard disk drive with a relatively large storage capacity.

The control processing unit 2 is a circuit that controls each of the units 1, 3-6 of the molten metal occupancy rate measuring device D according to the function of each unit, and calculates the occupancy rate of the image area capturing the molten metal as the bare metal area rate. The control processing unit 2 is configured, for example, with a CPU (Central Processing Unit) and its peripheral circuits. By executing a control processing program, the control processing unit 2 functionally comprises a control unit 21, an arithmetic execution determination unit 22, an area adjustment unit 23, a first area extraction unit 24, a second area extraction unit 25, and an occupancy rate processing unit 26.

The control unit 21 controls each of the units 1, 3-6 of the molten metal occupancy rate measuring device D according to the function of each unit, and is responsible for the overall control of the molten metal occupancy rate measuring device D. When determining the occupancy rate of the image area capturing the molten metal (bare metal area rate), the control unit 21 causes the imaging unit 1 to generate a color image showing an overhead view of the inside of the vessel containing the molten metal, and acquires the color image from the imaging unit 1. The generation and acquisition of this color image is performed repeatedly, for example, at a predetermined sampling interval.

The calculation execution determination unit 22 determines whether or not to calculate the bare bath area ratio based on the brightness value within a specified execution determination image area in the color image generated by the imaging unit 1.

The molten metal occupancy rate measuring device D may continuously and repeatedly determine the bare metal area ratio at predetermined sampling intervals while the blast furnace is in operation. However, since the bare metal area ratio is determined primarily to properly remove slag, it only needs to be determined during the slag removal process, and does not need to be determined when the slag removal process is not in progress. Furthermore, the slag removal process is often performed by tilting the vessel containing the molten metal (in the example shown in Figure 2, ladle IA). If the imaging unit 1 is fixedly positioned relative to the vessel, tilting the vessel will change the position of the image area capturing the interior of the vessel in the color image generated by the imaging unit 1. Therefore, in the color image generated by the imaging unit 1, a predetermined image region whose brightness value changes between when the slag removal operation is being performed (during slag removal operation) and when the slag removal operation is not being performed (during non-slag removal operation) is set as the execution determination image region, so that it is possible to determine whether the slag removal operation is being performed or not based on the change in brightness value of the execution determination image region, and to determine whether the bare metal area ratio is to be calculated or not. For example, as shown in Figure 3, a predetermined region (region of interest) ROI is set in advance in the color image generated by the imaging unit 1 as an image region to be subjected to image processing, and as shown in Figure 3A, a boundary line BL is set at the boundary between the region where the interior of the ladle IA is reflected and the region where the interior of the ladle IA is not reflected during the slag removal operation, and the region including the region where the interior of the ladle IA is reflected from this boundary line BL is set as the execution determination image region RC. Alternatively, for example, since the slag removal operation is performed while the vessel is storing molten metal, a boundary line BL is set so as to mainly include the area in which the interior of the ladle IA is reflected during the slag removal operation, and an area including the area in which the interior of the ladle IA is reflected from this boundary line BL is set as the execution determination image area RC. When the execution determination image area RC is set in this manner, the brightness value within this execution determination image area RC becomes relatively high during the slag removal operation because the image shown in Figure 3A is generated, and becomes relatively low during the non-slag removal operation because the image shown in Figure 3B is generated. For this reason, a brightness value threshold (execution determination threshold) Thc for distinguishing whether the slag removal work is in progress or the non-slag removal work is in progress is appropriately set in advance, and the calculation execution determination unit 22 compares the brightness value of the execution determination image area in the color image generated by the imaging unit 1 with the execution determination threshold Thc, and if the brightness value of the execution determination image area is equal to or greater than the execution determination threshold Thc, determines that it is time to determine the bare water area ratio, and if the brightness value of the execution determination image area is not equal to or greater than the execution determination threshold Thc, determines that it is not time to determine the bare water area ratio (if the brightness value of the execution determination image area is less than the execution determination threshold Thc, the bare water area ratio is not to be determined). When it is determined that the bare molten metal area ratio needs to be calculated, the molten metal occupancy ratio measuring device D operates the imaging unit 1, control unit 21, calculation execution determination unit 22, region adjustment unit 23, first region extraction unit 24, second region extraction unit 25, and occupancy ratio processing unit 26, and causes the occupancy ratio processing unit 26 to calculate the bare molten metal area ratio through information processing described below. On the other hand, when it is determined that the bare molten metal area ratio needs not to be calculated, the molten metal occupancy ratio measuring device D operates the imaging unit 1, control unit 21, and calculation execution determination unit 22 to monitor whether or not the bare molten metal area ratio needs to be calculated, and stops the operation of at least the occupancy ratio processing unit 26, preventing the occupancy ratio processing unit 26 from calculating the bare molten metal area ratio.

The area adjustment unit 23 adjusts the extracted image areas used by the first and second area extraction units 24, 25 based on the color image generated by the imaging unit 1. More specifically, there are multiple extracted image areas that are different from each other, and the area adjustment unit 23 selects one extracted image area from the multiple extracted image areas as the extracted image area to be used by each of the first and second area extraction units 24, 25 based on the color image generated by the imaging unit 1.

More specifically, the extracted image area is an internal image area that captures the internal area of the vessel viewed from above by the imaging unit 1, and is represented by a mask image that masks the image. By masking the target image of the mask with the mask image, the internal image area is extracted from the target image, and the remaining area excluding the internal image area is changed to a predetermined single color to generate a mask-processed image. In the example shown in Figure 2, the imaging unit 1 is positioned so that its optical axis is fixed. Therefore, when ladle IA, an example of the vessel, is tilted, the position of the internal image area in the color image generated by the imaging unit 1 changes. Therefore, the multiple extracted image areas are represented by multiple mask images corresponding to multiple tilt angles of ladle IA. In the molten metal occupancy rate measuring device D, multiple mask images corresponding to multiple different tilt angles are prepared in advance, and these multiple mask images are stored in the memory unit 6. For example, mask images MPa to MPe with inclination angles of 32°, 35°, 38°, 41°, and 44° shown in Figures 5A to 5E are created in advance by referencing color images MPf to MPj with inclination angles of 32°, 35°, 38°, 41°, and 44° shown in Figures 5F to 5J, and stored in memory 6. From among the mask images corresponding to the plurality of extracted image regions, region adjustment unit 23 selects the mask image that best matches the internal image region capturing the internal region of the container in the color image generated by imaging unit 1 as the mask image of the extracted image region to be used in first and second region extraction units 24 and 25, respectively. More specifically, in selecting this mask image, the contour of the internal image region in the color image generated by imaging unit 1 is compared with the contour of the extracted image region of the mask image, and the mask image having the contour of the extracted image region that best matches the contour of the internal image region is selected. For example, as shown in FIG. 4 , in a planar view, an XY Cartesian coordinate system is established in which the lower left vertex of the color image CP generated by the imaging unit 1 is the coordinate origin, the horizontal direction is the X axis, and the vertical direction is the Y axis. The region adjustment unit 23 extracts, from the region of interest ROI in the color image generated by the imaging unit 1, a pixel with the smallest X coordinate value and a brightness value equal to or greater than a predetermined threshold, as a start point SP. In the example shown in FIG. 4A , point (x1, y1) is extracted as the start point SP. Next, the region adjustment unit 23 extracts a predetermined length of the contour of the internal image region in the color image generated by the imaging unit 1 from the extracted start point as a comparison line of the internal image region. The contour can be extracted, for example, by image processing using an edge filter that extracts edges. In the example shown in FIG. 4B , a portion of the contour of the internal image region is extracted as the comparison line CLt of the internal image region. The predetermined length is appropriately set in advance to a length that allows for analyzing the correlation between the contour of the internal image region and the contour of the extracted image region. The region adjustment unit 23 then calculates the correlation value between each comparison line in each mask image similarly extracted from each mask image stored in the memory unit 6 and the comparison line in the internal image region, selects the comparison line in the mask image that gives the highest correlation value, and selects the mask image containing this selected comparison line as the mask image of the extracted image region to be used by the first and second region extraction units 24, 25. In calculating the correlation value, for example, the distance to the comparison line in the mask image is calculated for each of multiple points on the comparison line in the internal image region, and the sum of these calculated distances is calculated as the correlation value; in this case, the smaller the correlation value (the sum), the higher the correlation. Note that the comparison lines in each mask image may be calculated in advance and stored in the memory unit 6 for use as described above.

The first region extraction unit 24 determines a slag image region determined to be an image region containing slag and a molten metal candidate image region determined to be a candidate image region containing molten metal, based on the green image of the green component in the color image generated by the imaging unit 1. The green image is an image in which the green pixel value (G value) of each pixel in the color image is used as the pixel value of each pixel. For example, the green image shown in Figure 6B is generated from the color image generated by the imaging unit 1 shown in Figure 6A. More specifically, the first region extraction unit 24 determines the slag image region and the molten metal candidate image region within a predetermined extracted image region in the color image generated by the imaging unit 1.

More specifically, the first region extraction unit 24 extracts the extracted image region by masking the color image generated by the imaging unit 1 or the green image of the green component in the color image with the mask image, and then determines the slag image region and the molten metal candidate image region within this extracted extracted image region. For example, the first region extraction unit 24 extracts the extracted image region by masking the color image generated by the imaging unit 1 with the mask image. In the masking process, the pixel values of each pixel are extracted directly from the color image corresponding to the extracted image region and set as the pixel values of each pixel in the extracted image region, while the pixel values of each pixel in the remaining region excluding the extracted image region are set to 0. Therefore, as a result of the masking process, an image is generated in which the extracted image region is set as the color image of the extracted image region and the remaining region is set as black. Next, the first region extraction unit 24 generates a green image of the green component in this generated image. This generates a masked green image in which the green pixel value (G value) of each pixel in the color image of the extracted image region is set as the pixel value of each pixel, while the remaining region remains black. The first region extraction unit 24 then determines the slag image region and the molten metal candidate image region within the extracted image region in the masked green image. Alternatively, for example, the first region extraction unit 24 extracts the extracted image region by masking a green image based on the color image generated by the imaging unit 1 with the mask image. In the masking process, the pixel values of each pixel in the green image corresponding to the extracted image region are extracted as they are and set as the pixel values of each pixel in the extracted image region, while the pixel values of each pixel in the remaining region excluding the extracted image region are set to 0. Therefore, as a result of the masking process, a masked green image similar to that described above is generated, in which the extracted image region is set as the green image of the extracted image region and the remaining region is set as black. The first region extraction unit 24 then determines the slag image region and molten metal candidate image region within the extracted image region in the green image after the generated mask processing. In other words, either the generation of the green image or the mask processing can be performed first.

According to the inventor's findings, the green image can be divided into slag image regions, which are determined to be image regions containing slag, and other candidate molten metal image regions, which are determined to be candidate image regions containing molten metal. Therefore, the slag image regions and the candidate molten metal image regions are extracted using a threshold (first region determination threshold) Thr1, as shown in Figures 7A and 8A, respectively. More specifically, the first region extraction unit 24 compares the pixel value (G value) of each pixel within the extracted image region in the green image after processing the generated mask with the first region determination threshold Thr1. If the pixel value (G value) of the pixel is greater than (or equal to or greater than) the first region determination threshold Thr1, the first region extraction unit 24 determines that the pixel is a pixel in the candidate molten metal image region. If the pixel value (G value) of the pixel is less than (or equal to) the first region determination threshold Thr1, the first region extraction unit 24 determines that the pixel is a pixel in the candidate molten metal image region. If thin slag or flames are captured in the color image, the image area containing the thin slag or flames is included in the molten metal candidate image area. In this embodiment, the thin slag is defined as slag formed from molten oxides without graphite powder, as opposed to ordinary slag formed from molten oxides and graphite powder. The first region determination threshold Thr1 is changed depending on the temperature of the molten metal contained in the container. For example, as shown in FIG. 7B , information (temperature threshold correspondence information) representing the correspondence between temperature and the first region determination threshold Thr1 is created in advance, and the created temperature threshold correspondence is stored in the memory unit 6. The temperature threshold correspondence may be expressed in the form of a function formula, as shown in FIG. 7B , or in the form of a table (not shown). The first region extraction unit 24 determines a first region determination threshold Thr1 corresponding to, for example, a temperature input by an operator (user) via the input unit 3 or a temperature measured by a temperature sensor (e.g., a thermocouple) that is further provided in the molten metal occupancy measurement device D for measuring the temperature of the molten metal contained in the container, from the temperature threshold correspondence information stored in the storage unit 6. The determined first region determination threshold Thr1 is used to determine the slag image region and the molten metal candidate image region. Alternatively, as shown in FIG. 8B , the first region extraction unit 24 creates a histogram representing the frequency of each pixel value (G value) (class) within the extracted image region in the green image after mask processing, determines the most frequent brightness peak (class), determines the brightness value (peak brightness value) Bp of the determined brightness peak, and determines a brightness value that is smaller than the determined peak brightness value Bp by a predetermined value as the first region determination threshold Thr1. For example, the first region extraction unit 24 multiplies the calculated brightness peak value Bp by a predetermined scaling factor α (0.5<α<1) and calculates the multiplication result as the first region determination threshold Thr1. Since the histogram profile changes depending on the temperature of the molten metal contained in the container, the first region determination threshold Thr1 generated by this second generation method will change depending on the temperature of the molten metal contained in the container.

The second region extraction unit 25, based on the red image of the red component in the color image generated by the imaging unit 1, determines a slag molten metal image region that is determined to be an image region that captures at least one of the slag and the molten metal from among the slag image region and molten metal candidate image region determined by the first region extraction unit 24. The red image is an image in which the red pixel value (R value) of each pixel in the color image is used as the pixel value of each pixel. For example, the red image shown in Figure 6C is generated from the color image generated by the imaging unit 1 shown in Figure 6A. More specifically, the second region extraction unit 25 determines the slag molten metal image region within a predetermined extracted image region in the color image generated by the imaging unit 1.

More specifically, the second region extraction unit 25 extracts the extracted image region by masking the color image generated by the imaging unit 1 or the red image of the red component in the color image with the mask image, and then determines the molten slag metal image region within this extracted extracted image region. For example, the second region extraction unit 25 extracts the extracted image region by masking the color image generated by the imaging unit 1 with the mask image, as described above. Next, the second region extraction unit 25 generates a red image after the mask processing by taking the red pixel value (R value) of each pixel in the color image of the extracted image region as the pixel value of each pixel and leaving the remaining region black. The second region extraction unit 25 then determines the molten slag metal image region within the extracted image region in the red image after the mask processing. Alternatively, for example, the second region extraction unit 25 extracts the extracted image region by masking the red image based on the color image generated by the imaging unit 1 with the mask image. In the masking process, the pixel values of each pixel are extracted directly from the red image corresponding to the extracted image area and set as the pixel values of the pixels in the extracted image area, while the pixel values of each pixel in the remaining area excluding the extracted image area are set to 0. Therefore, as a result of the masking process, a post-mask red image is generated in which the extracted image area is set as the red image of the extracted image area and the remaining area is set as black. The second area extraction unit 25 then determines the molten slag metal image area within the extracted image area in the generated post-mask red image. In other words, either the red image generation or the masking process can be performed first.

The red image can be divided into a slag/molten metal image area, which is determined to be an image area containing at least one of the slag and the molten metal, and a thin slag image area, which is determined to be an image area containing thin slag. Therefore, the slag/molten metal image area is extracted using a threshold (second region determination threshold) Thr2, as shown in Figure 9. More specifically, for each pixel within the extracted image area in the red image after processing the generated mask, the second region extraction unit 25 compares the pixel value (R value) of the pixel with the second region determination threshold Thr2. If the pixel value (R value) of the pixel is greater than (or equal to or greater than) the second region determination threshold Thr2, the pixel is determined to be a pixel in the thin slag image area, which is determined to be an image area containing thin slag. If the pixel value (R value) of the pixel is less than (or equal to) the second region determination threshold Thr2, the pixel is determined to be a pixel in the slag/molten metal image area. If a flame is captured in the color image, the image area containing the flame is included in the thin slag image area. This second area determination threshold Thr2 is set appropriately in advance, for example, from multiple samples.

The occupancy rate processing unit 26 calculates the occupancy rate of the image area containing molten metal as the bare metal area rate based on the slag image area and molten metal candidate image area calculated by the first area extraction unit 24 and the slag molten metal image area calculated by the second area extraction unit 25. More specifically, the occupancy rate processing unit 26 calculates the residual image area by excluding the slag molten metal image area calculated by the second area extraction unit 25 from the slag image area and molten metal candidate image area calculated by the first area extraction unit 24, calculates the image area by excluding the residual image area from the molten metal candidate image area calculated by the first area extraction unit 24 as the molten metal image area, and calculates the occupancy rate of the calculated molten metal image area as the bare metal area rate. If molten metal, slag, thin slag, and flames are reflected in the color image, the slag image area reflects the slag, the molten metal, thin slag, and flames are reflected in the molten metal candidate image area, and the slag and molten metal are reflected in the slag-molten metal image area. Therefore, since the thin slag and flames are reflected in the residual image area, the image area in which the molten metal is reflected (molten metal image area) can be determined by subtracting the residual image area from the molten metal candidate image area, and the bare molten metal area ratio, which is the occupancy rate of the image area in which the molten metal is reflected, can be determined ((bare molten metal area ratio) = (area of molten metal image area) / ((area of slag image area) + (area of molten metal candidate image area))). Since the area of an image area can be expressed, for example, by the number of pixels, the bare molten metal area ratio can be determined based on the number of pixels in each image area.

As can be seen from the above, the first area extraction unit 24 may determine a candidate molten metal image area that is determined to be a candidate image area that captures the molten metal based on a green image of the green component in the color image generated by the imaging unit 1, and the second area extraction unit 25 may determine a thin slag image area that is determined to be an image area that captures the thin slag from among the candidate molten metal image areas determined by the first area extraction unit 24 based on a red image of the red component in the color image generated by the imaging unit 1. In this case, the occupancy rate processing unit 26 may determine the occupancy rate of the image area that captures the molten metal as a bare metal area rate based on the candidate molten metal image area determined by the first area extraction unit 24 and the thin slag image area determined by the second area extraction unit 25. More specifically, the first region extraction unit 24 extracts the extracted image region by masking the color image generated by the imaging unit 1 or the green image of the green component in the color image with the mask image, and determines the molten metal candidate image region within this extracted extracted image region. The second region extraction unit 25 extracts the extracted image region by masking the color image generated by the imaging unit 1 or the red image of the red component in the color image with the mask image, and determines the thin slag image region within this extracted extracted image region. The occupancy rate processing unit 26 determines the image region obtained by excluding the thin slag image region obtained by the second region extraction unit 25 from the molten metal candidate image region obtained by the first region extraction unit 24 as the molten metal image region containing the molten metal, and determines the occupancy rate of the determined molten metal image region as the bare metal area ratio.

The control processing unit 2, input unit 3, output unit 4, IF unit 5, and memory unit 6 can be configured, for example, by a desktop or laptop computer. The computers that make up each of these units 2-6 can be located, for example, in an operation room in a plant such as a blast furnace or converter, and can be incorporated into a console (or can serve as the console), or can be separate from the console.

Next, the operation of this embodiment will be described. Figure 10 is a flowchart showing the operation of the molten metal occupancy rate measuring device. Figure 11 is a diagram showing, as an example, the change in bare molten metal area ratio over time. The horizontal axis of Figure 11 represents elapsed time, and the vertical axis represents bare molten metal area ratio [%].

When the power is turned on, the molten metal occupancy rate measuring device D configured as described above initializes the necessary components and begins operation. By executing the control processing program, the control processing unit 2 functionally configures a control unit 21, a calculation execution determination unit 22, a region adjustment unit 23, a first region extraction unit 24, a second region extraction unit 25, and a occupancy rate processing unit 26.

When the sampling timing arrives, as shown in Figure 10, the molten metal occupancy measurement device D first causes the control unit 21 of the control processing unit 2 to generate a color image in the imaging unit 1, and acquires the color image from the imaging unit (S1).

The molten metal occupancy rate measuring device D then uses the calculation execution determination unit 22 of the control processing unit 2 to determine whether or not to calculate the bare molten metal area ratio based on the brightness value within a predetermined execution determination image region in the color image generated by the imaging unit 1 in process S1 (S2). If the result of this determination is that the bare molten metal area ratio should be calculated (Yes), the molten metal occupancy rate measuring device D then executes process S3. On the other hand, if the result of the above determination is that the bare molten metal area ratio should not be calculated (No), the molten metal occupancy rate measuring device D then executes process S8.

In this process S3, the molten metal occupancy rate measuring device D uses the area adjustment unit 23 of the control processing unit 2 to adjust the extracted image areas used by the first and second area extraction units 24 and 25, respectively, based on the color image generated by the imaging unit 1 in process S1.

Next, the molten metal occupancy rate measuring device D uses the first area extraction unit 24 of the control processing unit 2 to determine the slag image area and molten metal candidate image area for the green image within the extracted image area adjusted in process S3, based on the color image generated by the imaging unit 1 in process S1 (S4).

Next, the molten metal occupancy rate measuring device D uses the second area extraction unit 25 of the control processing unit 2 to determine the slag molten metal image area for the red image within the extracted image area adjusted in process S3, based on the color image generated by the imaging unit 1 in process S1 (S5).

Next, the molten metal occupancy rate measuring device D, using the occupancy rate processing unit 26 of the control processing unit 2, calculates the occupancy rate of the image area containing the molten metal as the bare metal area rate based on the slag image area and molten metal candidate image area calculated in process S4 and the slag molten metal image area calculated in process S5 (S6).

In process S4, the first region extraction unit 24 determines the molten metal candidate image region, in process S5 the second region extraction unit 25 determines the thin slag image region, and in process S6 the occupancy rate processing unit 26 may determine the bare metal area ratio based on these molten metal candidate image region and thin slag image region.

Next, the molten metal occupancy rate measuring device D outputs the bare molten metal area ratio calculated in step S6 to the output unit 4 via the control unit 21 of the control processing unit 2 (S7). For example, an example display device of the output unit 4 displays the bare molten metal area ratio versus the elapsed time from the start, as shown in FIG. 11. If necessary, the bare molten metal area ratio calculated in step S6 may be output to an external device via the IF unit 5.

The molten metal occupancy rate measuring device D then determines, via the control processing unit 2, whether or not to terminate this process. If the result of this determination is that the process should be terminated (Yes), the molten metal occupancy rate measuring device D terminates this process. For example, if the input unit 3 has received a command instructing termination, the molten metal occupancy rate measuring device D determines that the process should be terminated. On the other hand, if the result of this determination is that the process should not be terminated (No), the molten metal occupancy rate measuring device D returns the process to process S1.

The operator (user) references the bare molten metal area ratio output to the output unit 4 and determines that the slag removal work is complete if this bare molten metal area ratio is below a threshold value corresponding to a preset slag amount. The determination of whether the slag removal work is complete may also be performed by the molten metal occupancy ratio measuring device D, in which case the molten metal occupancy ratio measuring device D also outputs the determination result of whether the slag removal work is complete to the output unit 4.

As described above, the molten metal occupancy rate measuring device D in the embodiment and the molten metal occupancy rate measuring method implemented therein calculate the occupancy rate of the image area where the molten metal is reflected as the bare metal area ratio based on the following findings, and therefore can more accurately calculate the occupancy rate of the image area where molten metal, rather than slag, is reflected. In other words, the molten metal occupancy rate measuring device can more accurately detect image areas where slag (including normal slag and thin slag) is reflected.

The inventors studied a color image of the interior of a vessel containing molten metal, and discovered that better accuracy could be achieved by determining the area of molten metal based on a color image obtained by dividing the color image into its respective color components, rather than on the color image itself.

Figure 12 is an explanatory diagram showing, as an example, the distribution of slag and molten iron relative to pixel values for each color image at each temperature. The top row of Figure 12 primarily shows the case where the molten iron temperature is 1262°C, the middle row of Figure 12 primarily shows the case where the molten iron temperature is 1301°C, and the bottom row of Figure 12 primarily shows the case where the molten iron temperature is 1350°C. In plan view, the first figures from the left in Figure 12 are color images overlooking the inside of the ladle (for convenience of illustration, Figure 12 shows grayscale monochrome images). The second figures from the left in Figure 12 are histograms showing the distribution of slag and molten iron relative to pixel values in the red image, with the horizontal axis representing the red pixel value (R value) (class) and the vertical axis representing frequency. The third figures from the left in Figure 12 are histograms showing the distribution of slag and molten iron relative to pixel values in the green image, with the horizontal axis representing the green pixel value (G value) (class) and the vertical axis representing frequency. The fourth figures from the left in Figure 12 are histograms showing the distribution of slag and molten iron relative to pixel values in the blue image, with the horizontal axis representing the blue pixel value (B value) (class) and the vertical axis representing frequency.

As can be seen from the third diagram from the left in Figure 12, at all temperatures, slag is distributed with a bias towards relatively small green pixel values (G values), while molten iron is distributed with a bias towards relatively large green pixel values (G values). Therefore, the image can be separated into image regions that primarily contain slag and other image regions that primarily contain molten metal using the first region determination threshold Thr1 described above.

On the other hand, as shown in the second figures from the left in Figure 12, the brightness of the thin slag saturates in the red component at all temperatures, resulting in a red pixel value of 255 at 256 gradations. Therefore, the red component can be used to effectively separate the slag molten metal image area and the thin slag image area. Therefore, compared to the slag quantification method disclosed in Patent Document 1, which does not separate the image into color components, the molten metal occupancy rate measuring device D and molten metal occupancy rate measuring method of this embodiment are more accurate.

The bare metal area ratio is determined primarily to properly remove slag, so it only needs to be determined during the slag removal process, and does not need to be determined when the slag removal process is not in progress. Furthermore, the slag removal process is often performed by tilting the container containing the molten metal. If the imaging unit is fixedly positioned relative to the container, tilting the container changes the position of the image area capturing the interior of the container in the color image generated by the imaging unit, thereby changing the brightness value in a predetermined image area (the execution determination image area). The molten metal occupancy ratio measuring device D and molten metal occupancy ratio measuring method perform a calculation execution determination, making it possible to determine whether the slag removal process is in progress and to determine the bare metal area ratio only during the slag removal process.

The slag removal operation is not always performed with the container tilted at a fixed angle. The molten metal occupancy rate measuring device D and molten metal occupancy rate measuring method perform area adjustment, making it possible to adjust the extracted image area according to the actual angle at which the container is tilted, thereby more accurately determining the occupancy rate of the image area that contains molten metal, rather than slag.

In order to express the present invention, the present invention has been appropriately and sufficiently described above through embodiments with reference to the drawings. However, it should be recognized that those skilled in the art could easily modify and/or improve the above-described embodiments. Therefore, unless modifications or improvements made by those skilled in the art deviate from the scope of the claims set forth in the claims, such modifications or improvements are deemed to be encompassed within the scope of the claims.

D Molten metal occupancy rate measuring device 1 Imaging unit 2 Control processing unit 6 Memory unit 21 Control unit 22 Calculation execution determination unit 23 Area adjustment unit 24 First area extraction unit 25 Second area extraction unit 26 Occupancy rate processing unit

Claims (8)

an imaging unit that generates a color image of the interior of a vessel containing molten metal;
a first area extraction unit that determines a slag image area that is determined to be an image area that includes slag and a molten metal candidate image area that is determined to be a candidate image area that includes the molten metal, based on a green image of a green component in the color image generated by the imaging unit;
A second area extraction unit extracts a slag molten metal image area that is determined to be an image area that includes at least one of the slag and the molten metal, from among the slag image area and the molten metal candidate image area obtained by the first area extraction unit, based on a red image of the red component in the color image generated by the imaging unit;
and an occupancy rate processing unit that calculates the occupancy rate of the image area in which the molten metal is captured as a bare metal area rate based on the slag image area and the molten metal candidate image area calculated by the first area extraction unit and the slag molten metal image area calculated by the second area extraction unit.
Molten metal occupancy rate measuring device. The image capturing unit further includes a calculation execution determination unit that determines whether or not to calculate the bare bath area ratio based on a luminance value within a predetermined execution determination image region in the color image generated by the image capturing unit.
The molten metal occupancy rate measuring device according to claim 1 . The first region extraction unit obtains the slag image region and the molten metal candidate image region within a predetermined extracted image region in the color image,
The second region extraction unit obtains the molten metal slag image region within the extracted image region in the color image,
further comprising a region adjusting unit that adjusts the extracted image regions used by the first and second region extracting units based on the color image generated by the imaging unit;
The molten metal occupancy rate measuring device according to claim 1 . an imaging step of generating a color image of an overhead view of the interior of a vessel containing molten metal;
a first area extraction step of determining a slag image area determined to be an image area containing slag and a molten metal candidate image area determined to be a candidate image area containing the molten metal, based on a green image of a green component in the color image generated in the imaging step;
A second area extraction step of determining a slag molten metal image area that is determined to be an image area that includes at least one of the slag and the molten metal from among the slag image area and the molten metal candidate image area determined in the first area extraction step based on a red image of the red component in the color image generated in the imaging step;
and an occupancy rate processing step of calculating an occupancy rate of the image area including the molten metal as a bare metal area rate based on the slag image area and the molten metal candidate image area calculated in the first area extraction step and the slag molten metal image area calculated in the second area extraction step.
Molten metal occupancy rate measurement method. an imaging unit that generates a color image of the interior of a vessel containing molten metal;
a first area extraction unit that determines a molten metal candidate image area that is determined to be a candidate image area that includes the molten metal based on a green image of a green component in the color image generated by the imaging unit;
A second area extraction unit that determines a thin slag image area that is determined to be an image area that includes thin slag, from the molten metal candidate image area determined by the first area extraction unit, based on a red image of the red component in the color image generated by the imaging unit;
an occupancy rate processing unit that calculates the occupancy rate of the image area in which the molten metal is captured as a bare metal area rate based on the molten metal candidate image area calculated by the first area extraction unit and the thin slag image area calculated by the second area extraction unit;
The thin slag is formed from a melt of the oxides without the graphite powder, as opposed to a conventional slag formed from a melt of the oxides and graphite powder.
Molten metal occupancy rate measuring device. The image capturing unit further includes a calculation execution determination unit that determines whether or not to calculate the bare bath area ratio based on a luminance value within a predetermined execution determination image region in the color image generated by the image capturing unit.
The molten metal occupancy rate measuring device according to claim 5. the first region extraction unit obtains the molten metal candidate image region within a predetermined extracted image region in the color image;
The second region extraction unit obtains the thin slag image region within the extracted image region in the color image,
further comprising a region adjusting unit that adjusts the extracted image regions used by the first and second region extracting units based on the color image generated by the imaging unit;
The molten metal occupancy rate measuring device according to claim 5. an imaging step of generating a color image of an overhead view of the interior of a vessel containing molten metal;
a first area extraction step of determining a molten metal candidate image area that is determined to be a candidate image area that includes the molten metal based on a green image of a green component in the color image generated in the imaging step;
A second area extraction process for determining a thin slag image area, which is determined to be an image area including thin slag, from the molten metal candidate image area determined in the first area extraction process based on a red image of the red component in the color image generated in the imaging process;
and an occupancy rate processing step of determining an occupancy rate of the image area in which the molten metal is projected as a bare molten metal area rate based on the molten metal candidate image area determined in the first area extraction step and the thin slag image area determined in the second area extraction step,
The thin slag is formed from a melt of the oxides without the graphite powder, as opposed to a conventional slag formed from a melt of the oxides and graphite powder.
Molten metal occupancy rate measurement method.