Abstract
Method, imaging system (5), data processing device (60) and system (10) for determination of a 3D information (90), especially of a point cloud (80) or of a 3D surface reconstruction (81) or of a 3D object (82), of an inner part (55) of a metallurgical vessel (50) or of a modification, the method comprising the steps of providing (100) a metallurgical vessel (50); capturing (110) a first optical image (21) of at least one first inner part (51) of the metallurgical vessel (50), from a first imaging device position (22) outside of the metallurgical vessel (50), with a first optical axis (23), by a first imaging device (20); capturing (120) a second optical image (31) of at least one second inner part (52) of the metallurgical vessel (50), from a second imaging device position (32) outside of the metallurgical vessel (50), with a second optical axis (33), by a second imaging device (30); calculating (130) a 3D information (90), such as a point cloud (80) or a 3D surface reconstruction (81) or a 3D object (82), of at least one inner part (55) of the metallurgical vessel (50) from at least the first optical image (21) and the second optical image (31), whereas the first optical image (21) is captured from a first fixed imaging device position (22) with a first fixed optical axis (23) and whereas the second optical image (31) is captured from a second fixed imaging device position (32) with a second fixed optical axis (33).
Claims
1. Method for determination of 3D information (90), of an inner part (55) of a metallurgical vessel (50), the method comprising: providing (100) the metallurgical vessel (50); capturing (110) a first optical image (21) of at least one first inner part (51) of the metallurgical vessel (50), from a first imaging device position (22) outside of the metallurgical vessel (50), by a first imaging device (20), where the first imaging device has a first optical axis (23); capturing (120) a second optical image (31) of at least one second inner part (52) of the metallurgical vessel (50), from a second imaging device position (32) outside of the metallurgical vessel (50), by a second imaging device (30), where the second imaging device has a second optical axis (33); calculating (130) the 3D information (90) of the inner part (55) of the metallurgical vessel (50) from at least the first optical image (21) and the second optical image (31).
2. Method for determination of a modification of an inner part (55) of a metallurgical vessel (50), the method comprising: providing (200) a metallurgical vessel (50); capturing (210) a first optical image (21) of at least one first inner part (51) of the metallurgical vessel (50), from a first imaging device position (22) outside of the metallurgical vessel (50) by a first imaging device (20), where the first imaging device (20) has a first optical axis (23); capturing (220) a second optical image (31) of at least one second inner part (52) of the metallurgical vessel (50), from a second imaging device position (32) outside of the metallurgical vessel (50) by a second optical imaging device (30), where the second imaging device has a second optical axis (33); calculating (230) 3D information (90) n of the inner part (55) of the metallurgical vessel (50) from at least the first optical image (21) and the second optical image (31); and determining (240) a modification of the inner part (55) of the metallurgical vessel (50) based on a comparison of the calculated 3D information (90) with a previously stored 3D information (90) of the metallurgical vessel (50).
3. Method according to claim 1, whereas the at least one first inner part (51) of the metallurgical vessel (50) in the first optical image (21) and the at least one second inner part (52) of the metallurgical vessel (50) in the second optical image (52) overlap, wherein a region of overlap relative to the-total image content of any one of the captured optical images is at least 50%.
4. Method according to claim 1, whereas capturing the first (21) and the second optical image (31) is done within 1000 milliseconds, by the imaging devices, which are synchronized.
5. Method according to claim 1, whereas neither of the first imaging device (20) nor the second imaging device (30) is mounted on a moveable manipulator or on a moveable arm of a robot.
6. Method according to claim 1, whereas the first imaging device (20) is calibrated and the second optical imaging device (30) is calibrated.
7. Method according to claim 1, whereas a distance from the metallurgical vessel (50) to the first imaging device (20) and the second imaging device (30) is in a range of 3 m to 30 m.
8. Method according to claim 1, whereas a ratio between a distance of the first imaging device position (22) and the second imaging device position (32) to a distance from the metallurgical vessel (50) to the first imaging device (20) or the second imaging device (30) is in a range of 0.03 to 0.7.
9. Imaging system (5) for determination of 3D information (90) of an inner part (55) of a metallurgical vessel (50), the imaging system comprising: a first imaging device (20) that captures a first optical image (21) of at least one first inner part (51) of the metallurgical vessel (50), from a first imaging device position (22) outside of the metallurgical vessel (50), where the first imaging device (20) has a first optical axis (23); a second imaging device (30) that captures a second optical image (31) of at least one second inner part (52) of the metallurgical vessel (50), from a second imaging device position (32) outside of the metallurgical vessel (50), where the second imaging device (30) has a second optical axis (33); a_data exchange device (61) connected to the first imaging device (20) and the second imaging device (30), the data exchange device (61) being programmed to perform acts comprising: receiving the first optical image (21) from the first imaging device (20); receiving the second optical image (31) from the second imaging device (30); sending the first optical image (21) to a data processing device (60); sending the second optical image (31),to the data processing device (60); characterized in that the first imaging device (20) is mounted at the first imaging device position (22); and the second imaging device (30) is mounted at the second imaging device position (32).
10. Imaging system (5) according to claim 9, whereas the first imaging device (20) is not mounted on a moveable manipulator or on a moveable arm of a robot; and whereas the second imaging device (30) is not mounted on a moveable manipulator or on a moveable arm of a robot.
11. Imaging system (5) according to claim 9 , whereas the first imaging device (20) is a calibrated first imaging device, and whereas the second imaging device (30) is a calibrated second optical imaging device (30).
12. Data processing device (60) for determination of 3D information (90) of an inner part (55) of a metallurgical vessel (50), the data processing device (60) programmed to perform acts comprising: Receiving receiving a first optical image (21) of at least one first inner part (51) of the metallurgical vessel (50), from an imaging system (5); Receiving receiving a second optical image (31) of at least one second inner part (52) of the metallurgical vessel (50) from the imaging system (5); calculating (130) the 3D information (90) of the inner part (55) of the metallurgical vessel (50) from at least the first optical image (21) and the second optical image (31).
13. Data processing device (60) for determination of a modification of an inner part (55) of a metallurgical vessel (50), the data processing device programmed to perform acts comprising: receiving a first optical image (21) of at least one first inner part (51) of the metallurgical vessel (50) from an imaging system (5); receiving a second optical image (31) of at least one second inner part (52) of the metallurgical vessel (50) from the imaging system (5); calculating (230) 3D information (90), of the inner part (55) of the metallurgical vessel (50) from at least the first optical image (21) and the second optical image (31); determining (240) a modification of the inner part (55) of the metallurgical vessel (50) based on a comparison of the calculated 3D information (90) with a previously stored 3D information (90) of the metallurgical vessel (50).
14. System (10) for determination of 3D information (90) of an inner part (55) of a metallurgical vessel (50), the system comprising: an imaging system (5) connected to a data processing device (60); whereas the data exchange device (61) of the imaging system (5) is programmed to perform acts comprising: sending a first optical image (21) of at least one inner part (51) of the metallurgical vessel (50) to the data processing device (60); a second optical image (31) of at least one inner part (51) of the metallurgical vessel (50) to the data processing device (60); Receiving receiving the 3D information (90); whereas the data processing device (60) is programmed to perform acts comprising: receiving the first optical image (21) from the data exchange device (61) of the imaging system (5); receiving the second optical image (31) from the data exchange device (61) of the imaging system (5); calculating (130) the 3D information of the inner part (55) of the metallurgical vessel (50) from at least the first optical image (21) and the second optical image (31); sending (140) the calculated 3D information (90) to the data exchange device (61) of the imaging system (5).
15. (canceled)
16. Data processing device (60) of claim 13, wherein a first imaging device (20) captures the first optical image (21) and a second imaging device (30) captures the second optical image (31).
17. Data processing device (60) of claim 16, wherein the first imaging device (20) is at a first fixed position outside of the metallurgical vessel (50) and the second imaging device (30) is at a second fixed position outside of the metallurgical vessel (50).
18. Data processing device of claim 13, the acts further comprising: sending (250) an output based on the determined modification of the at least one inner part (55) of the metallurgical vessel (50) to a data exchange device (61) of an imaging system (5).
19. Data processing device of claim 12, the acts further comprising: sending (140) the 3D information (90) of the inner part (55) of the metallurgical vessel (50) to an imaging system (5).
20. Method of claim 1, wherein the first imaging device position (22) is a first fixed position and the second imaging device position (32) is a second fixed position.
21. Data processing device of claim 12, wherein a first imaging device (20) captures the first optical image (21) and a second imaging device (30) captures the second optical image (31), and further wherein the first imaging device (20) is at a first fixed position outside of the metallurgical vessel (50) and the second imaging device (30) is at a second fixed position outside of the metallurgical vessel (50).
Description
[0103] Exemplary embodiments of the invention are explained in more detail by means of illustrations:
[0104] FIG. 1 shows an exemplary flowchart of the first embodiment.
[0105] FIG. 2 shows an exemplary flowchart of the second embodiment.
[0106] FIG. 3a shows a sketch of a point cloud.
[0107] FIG. 3b shows a sketch of a 3D surface reconstruction.
[0108] FIGS. 4a and 4b are reproductions of a calculated 3D point cloud.
[0109] FIG. 5 is a schematic setup of a system according to the sixth and seventh embodiment.
[0110] FIG. 6 is an alternative schematic setup of a system according to the sixth and seventh embodiment.
[0111] FIG. 1 shows a first embodiment of the invention of a method for determination of a 3D information (90), especially a point cloud (80), of an inner part (55) of a metallurgical vessel (50), in this example, the metallurgical vessel (50) is an unused ladle (50), that is a ladle (50) with a new refractory lining (57) at the inner part (on the inside) of the ladle (50). The method comprises the steps of providing (100) the ladle (50), capturing (110) a first optical image (21) of at least one first inner part (51) of the ladle (50), from a first fixed imaging device position (22) outside the ladle (50), with a first fixed optical axis (23), by a first imaging device (20), which is a digital full HD camera with 1,555 ,200 pixels; capturing (120) a second optical image (31) of at least one second inner part (52) of the ladle (50), from a second fixed imaging device position (32) outside the ladle (50), with a second fixed optical axis (33), by a second imaging device (30) which is a digital full HD camera with 1,555 ,200 pixels; capturing a third optical image (41) of at least one third inner part (53) of the ladle (50), from a third fixed imaging device position (42) outside the ladle (50), with a third fixed optical axis (43), by a third imaging device (40) which is a digital full HD camera with 1,555 ,200 pixels, whereas in this example the overlap between the first optical image (21) and the second optical image (31) is 70% and the overlap between the second optical image (31) and the third optical image (41) is 70% the overlap between the third optical image (41) and the first optical image (21) is 70%, then calculating (130) a 3D point cloud (80) containing 1,500 ,000 points P.sub.i (t0)= P.sub.i(x.sub.i, y.sub.i, z.sub.i) at a timestep t0, with i=1..1,500,000, of at least one inner part (55) of the ladle (50) from at least the first optical image (21), the second optical image (31) and the third optical image (41); further storing (140) in the memory of a computer and on a remote (network) drive in the form of a data structure (91), e.g. in the form of a computer-readable file a 3D information (90) of the ladle (50), the 3D information (90) comprising the 3D point cloud (80) of the at least one inner part (55) o the ladle (50), that is the points P.sub.i (t0) and optionally a colour information (e.g. C.sub.i) for each point P.sub.i, with i=1..1.000.000.
[0112] FIG. 2 shows a second embodiment of the invention of a method for determination of a modification of an inner part (55) of a metallurgical vessel (50), whereas in this example, the metallurgical vessel (50) is a used ladle (50), that is a ladle (50) with a worn refractory lining (57) at the inner part (on the inside) of the ladle (50). The method comprises the steps of providing (100) the ladle (50), capturing (110) a first optical image (21) of at least one first inner part (51) of the ladle (50), from a first fixed imaging device position (22) outside the ladle (50), with a first fixed optical axis (23), by a first imaging device (20), which is a digital full HD camera with 1,555 ,200 pixels; capturing (120) a second optical image (31) of at least one second inner part (52) of the metallurgical vessel (50), from a second fixed imaging device position (32) outside the ladle (50), with a second fixed optical axis (33), by a second imaging device (30) which is a digital full HD camera with 1,555 ,200 pixels; whereas in this example the overlap between the first optical image (21) and the second optical image (31) is 70%, then calculating (130) a 3D surface reconstruction (81) containing a mesh connecting 1,000 ,000 points P.sub.i (t1)= P.sub.i(x.sub.i, y.sub.i, z.sub.i) at a timestep t1, with i=1..1.000.000, of at least one inner part (55) of the metallurgical vessel (50) from at least the first optical image (21) and the second optical image (31). Further the method comprises determining (240) a modification, in this example the modification is the amount of wear of the lining since the previous furnace campaign of the ladle (50) of at least one inner part (55) of the ladle (50). This determination (240) is based on the comparison of the calculated 3D surface reconstruction (81) (timestep t=t1) with a previously stored 3D surface reconstruction (81) (timestep t=t0) of the ladle (50), whereas the previously stored 3D surface reconstruction (81) is loaded from a remote (network) drive in the form of a data structure (91), such as a computer-readable file containing the 3D surface reconstruction (81) from the same ladle (50) before the previous furnace campaign. Further the method comprises sending (250) an output based on the determined modification of a ladle (50). Here this output comprises a computer readable table with the absolute values (in mm) of the wear of the refractory lining (57) at several positions inside the ladle (50) (such as in the slag zone, or in the bottom of the ladle).
[0113] In an alternative embodiment, the same steps are followed as described above in relation to FIG. 2, except that the method step of determining (240) a modification is altered in this example such that the modification is the amount of wear of the lining since the first use of the ladle (50) and the resulting residual wall thickness of the refractory lining (57) of at least one inner part (55) of the ladle (50). This determination (240) is based on the comparison of the calculated 3D surface reconstruction (81) with the previously stored 3D surface reconstruction (81), which is loaded from a remote (network) drive in the form of a data structure (91), such a computer-readable file containing the 3D surface reconstruction (81) from the same but unused ladle (50), that is the same ladle (50) with a new refractory lining (57) and the initial wall thicknesses. In this alternative embodiment the method comprises sending (250) an output based on the determined modification of the ladle (50) to an imaging system (5). The imaging system (5) is programmed to receiving a 3D information (90) from a data processing device (60). The imagining system (5) can present the received 3D information on a data output device (70), e.g. a monitor, or further process the received data. Here this output comprises a computer readable table with the absolute values (in mm) of the residual wall thickness of the refractory lining (57) at several positions inside the ladle (50) (such as in the slag zone, or in the bottom of the ladle). Alternatively, the determination of a modification can be such, that a certain modification is given as a signal (65), e.g. a warning signal (65), when the residual wall thickness of the refractory lining (57) lies outside a certain threshold.
[0114] FIG. 3a shows a schematic point cloud comprising 16 points P.sub.i (t)= P.sub.i(x.sub.i, y.sub.i, z.sub.i), i=1..16, at a certain timestep t, in a representation in a coordinate system with a x, y and z axis. Each of the points P.sub.i are represented with three variables x.sub.i, y.sub.i, z.sub.i which denote the coordinates of the respective point. In that sense the Point P.sub.1 is represented by the coordinates x.sub.1, y.sub.1, z.sub.1. The points P.sub.i can be shown graphically as a 2D representation of a 3D body, e.g. on a computer screen. The 3D body can be viewed from different directions, by applying rotational transformations on the points P.sub.i. The points P.sub.i can be stored in the form of a data structure (91), such as a computer-readable file in a computer readable format, such as e.g. in a list, where each line contains the x, y and z coordinates of a respective point.
[0115] FIG. 3b shows a 3D surface reconstruction (81). The 3D surface reconstruction (81) comprises a mesh / a grid connecting the 3D point cloud (80), thus the point cloud (80) with points P.sub.i (t)= P.sub.i(x.sub.i, y.sub.i, z.sub.i) span out the reconstructed surface (81).
[0116] FIG. 4a and FIG. 4b shows two different sections of the same calculated 3D information (80) of an inner part (55) of a ladle (50) (here the 3D point cloud (80) of the refractory lining (57) is shown) obtained by the present invention. The 3D point cloud (80) overall comprises 10 million points P.sub.i (t)= P.sub.i(x.sub.i, y.sub.i, z.sub.i), i=1..10.000.000. FIG. 4a and FIG. 4b show a 2D representation of this 3D point cloud (80). By applying several matrix operations onto this point cloud (80), that is on each point P.sub.i, is it possible to tilt the inner ladle (50) 3D point cloud (80) or to zoom into the ladle (50) and to inspect in detail the surface of the refractory lining (57). Here, FIG. 4a shows an overview on the whole 3D point cloud (80), while FIG. 4b shows a section (rotated and zoomed in) of the same 3D point cloud (80), revealing details of the refractory lining (57). As shown above in connection with the second embodiment, it is also possible to determine a modification of a ladle (50) based on the calculated 3D point cloud (80), such as e.g. the determination of wear during a certain furnace campaign. The 3D point cloud (80) shown in FIGS. 4a and 4b can e.g. be stored on a remote (network) drive in the form of a data structure (91), such as a computer-readable file. The computer readable file comprises the 3D information (90), in this example the 3D information (90) comprises the 3D point cloud (80) in the form of a list, where each line contains the x, y and z coordinates of a respective point P.sub.i, and further an intensity information for each point P.sub.i.
[0117] FIG. 5 shows a schematic setup of a system (10) according to the invention. A system (10) for determination of a 3D information (90) of an inner part (55) of a metallurgical vessel (50) is shown, the system comprises an imaging system (5) connected to a data processing device (60). The imaging system (5) comprises a first imaging device (20) for capturing a first optical image (21) of at least one first inner part (51) of the metallurgical vessel (50) from a first fixed imaging device position (22), with a first fixed optical axis (23), which is a digital 4K camera with 8,294 ,400 pixels, and a second imaging device (30) for capturing a second optical image (31) of at least one second inner part (52) of the metallurgical vessel (50) from a second fixed imaging device position (32), with a second fixed optical axis (33), which is a digital 4K camera with 8,294 ,400 pixels and a third imaging device (40) for capturing a third optical image (41) of at least one third inner part (53) of the metallurgical vessel (50) from a third fixed imaging device position (42), with a third fixed optical axis (43), which is a digital 4K camera with 8,294 ,400 pixels. The imaging system (5) further includes a data exchange device (61) adapted to receiving the optical images (21, 31, 41) from the imaging device (20, 30, 40) and sending the optical images (21, 31, 41) to the data processing device (60). The first imaging device (20), the second imaging device (30) and the third imaging device (40) are mounted at a fixed position on a frame (15) connected to the floor, such that the imaging devices (20, 30, 40) are targeted towards the ladle (50) resting at the ladle repair place. The first fixed optical axis (23), the second fixed optical axis (33) and the third fixed optical axis (43) are intersecting at the ladle repair place and the respective angle (α) between the directions defined by these axes is approximately 7°. The distance between each of the first fixed imaging device position (22), the second fixed imaging device position (32) and the third fixed imaging device position (42) are 2 m, the three positions being placed in the corners of an equilateral triangle of 2 m side length. A ladle (50) was used as the metallurgical vessel (50), the ladle (50) is placed at the ladle repair place, the ladle (50) is in the hot state, with a temperature of approximately 800° C. of the surface of the refractory lining. The distance of the first imaging device (20), the second imaging device (30) and the third imaging device (40) from the ladle (50) (that is to the nearest part of the ladle (50), which is the ring of the ladle (50)) is for all imaging devices (20, 30, 40) 10 m. This yields a ratio between the distance of the first imaging device position (22) and the second imaging device position (32) to the distance from the metallurgical vessel (50) to its nearest imaging device position (22, 32, 42) of 0.2. The imaging devices (20, 30, 40) were protected by a heat shield (24, 34, 44), here by a metallic housing. The region of overlap relative to the total image content of the first optical image (21) and the second optical image (31) is 75%, of the second optical image (31) and the third optical image (41) is 75%, of the third optical image (41) and the first optical image (21) is 75%. The three imaging devices (20, 30, 40) are synchronized and capturing the first optical image (21), the second optical image (31) and the third optical image (41) is done within 150 milliseconds. FIG. 5 also shows a data processing device (60), which can receive a first optical image (21), a second optical image (31), a third optical image (41) of the metallurgical vessel (50) from data exchange device (61) of the imaging device (20) via cables connecting each of the imaging devices (21, 31, 41) with the data exchange device (61). In this example, the data processing device (60) and the data exchange device (61) are implemented in an industrial personal computer, programmed to preform the step of calculating (130) a 3D surface reconstruction (81) of an inner part (55) of the metallurgical vessel (50) from the first optical image (21), the second optical image (31) and the third optical image (41), and further to storing (140) a 3D information (90) of an inner part (55) of the metallurgical vessel (50), the 3D information (90) comprising the 3D point cloud (80). Alternatively, the data processing device (60) is programmed to calculating (230) a 3D surface reconstruction (81) of at least one inner part (55) of the metallurgical vessel (50) from at least the first optical image (21) and the second optical image (31); further to determining (240) the modification of the at least one inner part (55) of the metallurgical vessel (50) based on the comparison of the calculated 3D surface reconstruction (81) with a previously stored 3D surface reconstruction (81) of the metallurgical vessel (50); and also to sending (250) an output based on the determined modification of the at least one inner part (55) of a metallurgical vessel (50) to an imaging system (5), where the output can e.g. be displayed or further processed.
[0118] The data processing device (60) of the example in FIG. 5 is an industrial PC which is connected via a data exchange device (61) (being a part of the data processing device (60)) to the imaging devices (20, 30, 40), to trigger the capturing of the optical images (21, 31, 41) and to directly obtain the optical images (21, 31, 41) into the memory of the data processing device (60). Here, the data exchange device (61) uses the standard inputs of the data processing device (60) (here: an industrial PC) and is implemented as a software into the data processing device (60). The system (10) further comprises a data output device (70), designed to output a modification of the metallurgical vessel, the data output device (70) in this example is a monitor.
[0119] FIG. 6 shows an alternative setup to the one described in FIG. 5 with the following differences: Here the data processing device (60) is on a remote location, that is a different location than the data exchange device (61) and the imaging devices (20, 30, 40). The data processing device (60) is connected via a data exchange device (61) (being a local standalone device) to the imaging devices (20, 30, 40). The data exchange device is programmed to trigger the capturing of the optical images (21, 31, 41) and to send the optical images (21, 31, 41) to the data processing device (60). Here, the data exchange device (61) and the data processing device (61) are separate local standalone industrial PC’s. The connection from the data exchange device (61) to the data processing device (60) is via a network connection. The data processing device (60) will do the calculation steps as described above and send the output back to the data exchange device (61). The system (10) further comprises a data output device (70), designed to output a modification of the metallurgical vessel, the data output device (70) in this example is a monitor and the data output device (70) is connected to the data exchange device (61).
[0120] List of reference numerals and factors: [0121] 5 Imaging system for determination of a 3D information (90) [0122] 10 System for determination of a 3D information (90) [0123] 15 Frame [0124] 20 First imaging device [0125] 21 First optical image [0126] 22 First imaging device position [0127] 23 First optical axis [0128] 24 First heat shield [0129] 30 Second imaging device [0130] 31 Second optical image [0131] 32 Second imaging device position [0132] 33 Second optical axis [0133] 34 Second heat shield [0134] 40 Third imaging device [0135] 41 Third optical image [0136] 42 Third imaging device position [0137] 43 Third optical axis [0138] 44 Third heat shield [0139] 50 Metallurgical vessel [0140] 51 First inner part of the metallurgical vessel (50) [0141] 52 Second inner part of the metallurgical vessel (50) [0142] 53 Third inner part of the metallurgical vessel (50) [0143] 55 Inner part of the metallurgical vessel (50) [0144] 57 Refractory lining of the metallurgical vessel (50) [0145] 60 Data processing device [0146] 61 Data exchange device [0147] 65 Signal [0148] 70 Data output device [0149] 80 3D point cloud [0150] 81 3D surface reconstruction [0151] 82 3D object [0152] 90 3D information [0153] 91 Data structure [0154] 100 Providing a metallurgical vessel (50) [0155] 110 Capturing a first optical image (21) of at least one first inner part (51) of the metallurgical vessel (50) from a first imaging device position (22), with a first optical axis (23) by a first imaging device (20) [0156] 120 Capturing a second optical image (31) of at least one second inner part (52) of the metallurgical vessel (50) from a second imaging device position (32), with a second optical axis (33) by a second imaging device (30) [0157] 121 Capturing further optical images (41) of at least further inner parts (53) of the metallurgical vessel (50) from a further imaging device position (42), with a further optical axis (43) by a third imaging device (40) [0158] 130 Calculating (130) a 3D information (90), such as a point cloud (80) or a 3D surface reconstruction (81) or a 3D object (82), of at least one inner part (55) of the metallurgical vessel (50) from at least the first optical image (21) and the second optical image (31) [0159] 140 Storing (140) the 3D information (90) of the at least one inner part (55) of the metallurgical vessel (50) [0160] 200 Providing a metallurgical vessel (50) [0161] 210 Capturing a first optical image (21) of at least one first inner part (51) of the metallurgical vessel (50) from a first imaging device position (22), with a first optical axis (23) by a first imaging device (20) [0162] 220 Capturing a second optical image (31) of at least one second inner part (52) of the metallurgical vessel (50) from a second imaging device position (32), with a second optical axis (33) by a second optical imaging device (30) [0163] 230 Calculating (230) a 3D information (90), such as a point cloud (80) or a 3D surface reconstruction (81) or a 3D object (82), of at least one inner part (55) of the metallurgical vessel (50) from at least the first optical image (21) and the second optical image (31) [0164] 240 Determining (240) a modification of at least one inner part (55) of the metallurgical vessel (50) based on the comparison of the calculated 3D information (90) with a previously stored 3D information (90) of the metallurgical vessel (50) [0165] 250 Generating (250) an output based on the determined modification of the at least one inner part (55) of the metallurgical vessel (50) [0166] α angle between the directions defined by any pair of: first optical axis (23), second optical axis (33), third optical axis (43)
