JP2004125686A - Method and apparatus for detecting flaw of steel sheet, computer program, and computer-readable recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately detect a continuous flaw which has occurred over the entire length of a steel sheet. <P>SOLUTION: The surface of the steel sheet is imaged (step S11). Flaws are detected from imaged data regardless of their types or sizes at the same location in the width direction of the steel sheet (step S12). The number of the detected flaws is integrated in a longitudinal direction to create a histogram, and a location in a width direction in which the number of the flaws exceeds a prescribed threshold value is detected (step S13). By recognizing the presence or absence of the occurrence of a continuous flaw by comparing the detected locations in the width direction of a plurality of steel plates with one another (step S14), it is possible to highly accurately detect even a minor flaw as a continuous flaw, which is easily overlooked in the case of detecting continuous flaws in a single steel sheet. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method for detecting flaws on a steel sheet, a flaw detection apparatus for a steel sheet, a computer program, and a computer-readable recording medium, and is particularly suitable for use in preventing occurrence of a large number of flaws on the surface of a steel sheet in a rolling line. Things.
[0002]
[Prior art]
In rolling lines that manufacture strips such as steel sheets, the surface of the rolled steel sheets was inspected to assure the quality of the products, and the types and grades of surface flaws that occurred on the steel sheet surfaces were determined. . Conventionally, as a surface flaw inspection apparatus for inspecting the surface flaw of such a steel sheet, the presence or absence of a flaw is determined based on a characteristic amount such as a luminance difference of a single flaw.
[0003]
[Problems to be solved by the invention]
Since the conventional steel sheet flaw detection device determines the presence or absence of a flaw based on a characteristic amount such as a brightness difference (contrast) of the flaw alone and the area of the flaw, the flaw generated on the surface of the steel sheet occurs in a single shot. Could not be determined.
[0004]
Further, since flaws such as oyster flaws and roll flaws are formed in a thin linear shape, flaws having a small contrast and area are often overlooked as minor flaws (harmless flaws) by themselves. However, such oyster flaws and roll flaws are continuous flaws that occur over the entire length of the steel sheet, and therefore become serious flaws in terms of products.
[0005]
Therefore, if all of the above-mentioned minor flaws are detected, there is a problem that a flaw alarm is always output or an erroneous detection is performed.
SUMMARY OF THE INVENTION In view of the above problems, an object of the present invention is to make it possible to accurately detect continuous flaws generated over the entire length of a steel sheet.
[0006]
[Means for Solving the Problems]
The flaw detection method for a steel sheet according to the present invention is a flaw detection method for a steel sheet that detects continuous flaws of the steel sheet based on image data obtained by imaging the surface of the steel sheet. In the flaw detection step to detect, the surface flaws of the steel sheet detected in the flaw detection step, the flaw accumulation step of integrating the number of flaws along the longitudinal direction for each position in the width direction of the steel sheet, A continuous flaw candidate detection step of detecting the integrated flaw number as a continuous flaw candidate if the integrated flaw number value exceeds a predetermined value, and a plurality of width direction positions of the steel sheet where the continuous flaw candidates detected in the continuous flaw candidate detection step exist. A continuous flaw determination step of collating steel sheets to determine whether or not the continuous flaw candidate is a continuous flaw.
Another feature of the present invention is that the surface flaw detection of the steel sheet in the flaw detection step is performed irrespective of the type and size of the flaw.
Further, other features of the present invention include a roll flaw determination step of determining whether the flaw detected in the continuous flaw candidate detection step is a roll flaw, and a roll flaw determined by the roll flaw determination step. And a roll flaw position specifying step of specifying a flaw occurrence position.
Another feature of the present invention is that, in the roll flaw specifying step, the roll flaw generation position is specified by rolling back the rolling rate of the steel sheet.
[0007]
The steel sheet flaw detection device of the present invention is a steel plate flaw detection device that detects continuous flaws of the steel sheet based on image data obtained by imaging the surface of the steel sheet, and detects a surface flaw of the steel sheet from the image data. Flaw detecting means for detecting, for the surface flaw of the steel sheet detected by the flaw detecting means, flaw accumulating means for integrating the number of flaws along the longitudinal direction for each position in the width direction of the steel sheet, in the flaw accumulating means A continuous flaw candidate detection means for detecting as a continuous flaw candidate if the integrated flaw number value exceeds a predetermined value, and a plurality of width direction positions of the steel sheet where the continuous flaw candidates detected by the continuous flaw candidate detection means exist. A continuous flaw determining means for collating the steel sheet to determine whether or not the continuous flaw candidate is a continuous flaw.
Another feature of the present invention is that the surface flaw detection of the steel sheet by the flaw detection means is performed irrespective of the type and size of the flaw.
Further, other features of the present invention include a roll flaw determining means for determining whether the flaw detected by the continuous flaw candidate detecting means is a roll flaw, and a roll flaw determined by the roll flaw determining means. A roll flaw position specifying means for specifying a flaw occurrence position.
Further, another feature of the present invention is that the roll flaw specifying means specifies the position where the roll flaw occurs by rebating with the rolling rate of the steel sheet.
[0008]
The computer program of the present invention is a program that causes a computer to execute a steel sheet flaw detection method for detecting continuous flaws of the steel sheet based on image data obtained by imaging the surface of the steel sheet, and from the image data, A flaw detection step of detecting surface flaws, a flaw accumulation step of integrating the number of flaws along the longitudinal direction at each position in the width direction of the steel sheet with respect to the surface flaw of the steel sheet detected in the flaw detection step, A continuous flaw candidate detection step of detecting as a continuous flaw candidate if the number of flaws integrated in the integration step exceeds a predetermined value; and a width direction position of a steel sheet where the continuous flaw candidate detected in the continuous flaw candidate detection step exists. A continuous flaw determination step of determining whether or not the continuous flaw candidate is a continuous flaw by comparing the plurality of steel sheets with each other. To have.
[0009]
A computer-readable recording medium according to the present invention is characterized by recording the computer program described above.
[0010]
Since the present invention has the technical means described above, the number of flaws is integrated over the longitudinal direction at the same position in the width direction of the steel sheet regardless of the type and size of the flaw, so that the peak of the number of flaws is obtained. It is possible to reliably detect the width direction position where the portion exists. Then, the detected width direction position is compared for a plurality of steel plates, and when a peak portion of the number of flaws exists at the same width direction position for a plurality of steel plates, it is determined that there is a continuous flaw. If the disturbance elements such as are removed and even a small flaw that can be overlooked by a single steel sheet is continuously generated at the same width direction position, it can be detected as a continuous flaw with high accuracy .
[0011]
BEST MODE FOR CARRYING OUT THE INVENTION
Next, an embodiment of a method for detecting a flaw of a steel sheet, a flaw detection apparatus for a steel sheet, a computer program, and a computer-readable recording medium of the present invention will be described with reference to the accompanying drawings.
[0012]
FIG. 1 is a configuration diagram showing an embodiment of a hot flaw inspection device according to the present invention. As shown in FIG. 1, a hot flaw inspection device 2 of the present embodiment is disposed between a finish rolling device 1 and a winding device 3, and the upper and lower surfaces of a finish-rolled steel sheet S. The flaws on the surface of the steel sheet S are detected from both sides.
[0013]
The hot flaw inspection device 2 includes an upper steel plate flaw detection device 20a and a lower steel plate flaw detection device 20b. Both the flaw detection device 20a for the upper steel plate and the flaw detection device 20b for the lower steel plate have the same configuration, and are constituted by the steel plate imaging camera 21 and the lighting device 22. In the present embodiment, six steel plate imaging cameras 21 are arranged along the width direction of the steel plate S, and three lighting devices 22 are arranged.
[0014]
The flaw detection device 20a for the upper steel plate is disposed on the upper surface side of the steel plate S, and the steel plate imaging camera 21 and the illuminating device 22 are disposed so that the first distance 11 is equal to 2925 mm. On the other hand, the flaw detection device 20b for the lower steel plate is provided such that the steel plate imaging camera 21 and the lighting device 22 are located at the second distance l2 = 2500 mm.
[0015]
The lower steel plate flaw detection device 20b captures an image of the lower surface of the steel sheet S from the gap between the inner cooling rolls 24 that transports the steel sheet S. For this reason, at the position where the flaw detection device 20b for the lower steel sheet is provided, the interval between the inner cooling rolls 24 is slightly widened, and the movable apron 23 is provided there, so that the tip of the steel sheet S passes. In addition, it does not enter between the widened inner cooling rolls 24. Then, the movable apron 23 is moved in the lateral direction after the tip of the steel plate S has passed, so that the steel plate imaging camera 21 can capture an image.
[0016]
FIG. 2 is a block diagram illustrating a schematic configuration of the flaw detection devices 20a and 20b for detecting flaws on the steel sheet S. As shown in FIG. 2, in the present embodiment, the image processing unit 131, the flaw detection unit 132, the flaw classification unit 133, the flaw feature storage unit 134, the flaw determination unit 135, the display unit 136, and the like are configured. .
[0017]
The image processing unit 131 is for performing predetermined image processing on image data output from each steel plate imaging camera 21. In the present embodiment, the image processing unit 131 is about 50 to 60 mm in the width direction of the steel plate S. Blocks are formed as an aggregate of pixels of a size.
[0018]
The flaw detection unit 132 detects a surface flaw of the steel sheet S from image data on which predetermined image processing has been performed by the image processing unit 31 based on a luminance difference or the like. The flaw classification unit 133 classifies the type of flaw detected by the flaw detection unit 132 based on the features stored in the flaw feature storage unit 134. In the present embodiment, as shown in FIG. 4, classification is made into a single flaw 31, a water drop 32, a formation 33, an oyster flaw, a roll flaw 35, and the like.
[0019]
Of the above-mentioned flaws, the formation 33 is caused by a change in the roughness of the steel sheet S, a red scale, and the like, and the water droplet 32 is caused by water used in hot rolling. The red scale is a red scale remaining on the surface of the steel sheet S, and is generated from a band shape to a linear shape depending on circumstances.
[0020]
The flaw determination unit 135 determines whether or not the continuous flaw 34 has occurred on the surface of the steel sheet S based on the type of the flaw classified by the flaw classification unit 133, and displays the determination result on the display unit 136. They are output and displayed, or output to a flaw alarm device (not shown).
[0021]
Next, the operation of the hot flaw inspection apparatus 2 of the present embodiment will be described with reference to the flowchart of FIG.
First, in the first step S11, the front surface (back surface) of the steel sheet S is imaged. This imaging is performed by the flaw detection device 20a for the upper steel plate and the flaw detection device 20b for the lower steel plate described above. In the present embodiment, in order to inspect substantially all width directions of the steel sheet S, the six steel sheet imaging cameras 21 are used as described above, and in each steel sheet imaging camera 21, the width direction is 300 mm and the length direction is 200 mm. Each time, the surface of the steel sheet S is imaged.
[0022]
Next, in step S12, a flaw is detected from the image data. As described above, this is performed by the flaw detection unit 132, and is performed for each preset score (to be described later), and for each score, a single flaw 31, a water droplet 32, a flaw classification unit 33, and an oyster flaw or a roll flaw. Classify flaws such as 35.
[0023]
Next, in step S13, the distribution state of the flaws in the width direction of the steel sheet S is checked, and a surface flaw profile formed of a histogram is created. Thereby, the distribution state of the position where the flaw is generated in the width direction of the steel sheet S can be clearly known.
[0024]
Next, the process proceeds to step S14, and it is determined whether or not a continuous flaw has occurred. This determination is made by examining the flaw accumulation data as shown in FIG. 4 and determining whether or not the value exceeds a predetermined threshold value in the width direction of the steel sheet S. In the example shown in FIG. 4, since the aggregate of the pixels denoted by reference numeral 30 exceeds the predetermined threshold, the continuous flaw 34 is located at the position in the width direction of the steel sheet S corresponding to the aggregate 30 of the pixels. It is determined that it has occurred.
[0025]
In the present embodiment, when the continuous flaws 34 are generated for a plurality of steel sheets S, it is determined that the continuous flaws 34 based on the oyster flaws and the roll flaws 35 are generated, and a flaw alarm is output. I am trying to do it.
[0026]
This is to prevent a malfunction such as determining that the continuous flaw 34 is occurring when there are many disturbances such as the formation 33 and the water drop 32 depending on the state of hot rolling. Yes, by confirming that the continuous flaws 34 are generated at the same width direction position for a plurality of steel sheets S, the influence of disturbances such as the formation 33 and the water droplets 32 is removed and the continuous flaw determination with high accuracy is performed. Like that.
[0027]
If the result of determination in step S14 is that no continuous flaw has occurred, the process is terminated. However, if it is determined that continuous flaw has occurred, the process proceeds to step S15, where the generated continuous flaw is determined as a roll flaw. Check whether or not. This is performed by detecting the periodicity of the generated flaws, and if it is not a roll flaw, the process directly proceeds to step S17. If it is a roll flaw, the process proceeds to step S15, and processing for specifying the roll on which the flaw occurs is performed.
[0028]
This is because the pitch of each roll in the hot rolling line is different from place to place, and the roll that causes the roll flaws is identified by determining the pitch at which the flaw is generated by recalculating by the rolling rate. To do.
[0029]
In the present embodiment, by performing the above-described processing, the roll generating the continuous flaws 34 can be identified early, so that the cause of the continuous flaws 34 can be removed early. It is possible to reliably prevent a large number of continuous flaws 34 from being generated on the plurality of steel sheets S.
[0030]
Next, in step S17, a flaw alarm process for notifying that the continuous flaw 34 has been detected is performed. This is performed by displaying on the display unit 136 as described above, or by outputting a flaw detection signal to an external alarm device (not shown).
[0031]
FIG. 5 is a diagram illustrating an example of the number of flaws and the flaw positions of the front and rear steel sheets S. As is clear from FIG. 5, many flaws are generated at the same position in the width direction of the front and rear steel sheets S. In the example of the figure, 16 flaws are detected in the front material and 20 flaws are detected in the rear material, and it can be seen that continuous flaws 34 are generated.
[0032]
FIG. 6 is a diagram showing the positions of the flaws generated on the surface of the steel sheet S in further detail. It can be seen that the continuous flaws 34 are concentrated in a narrow range in the width direction. Note that this case shows a case where the flaw detection is performed at “rating 2”.
[0033]
FIG. 7 is a diagram for explaining that the number of flaws that can be detected is different due to a difference in rating, and is a diagram illustrating an example in which flaws are detected on the lower surface of the steel sheet S. As is clear from FIG. 7, when the flaw detection is performed at “score 3”, not many flaws are detected.
[0034]
On the other hand, when flaw detection is performed at "score 1", a very large number of flaws are detected. For this reason, even if even a small flaw is detected, even if many flaws are detected indiscriminately, it has not been conventionally known which of the flaws is to be subjected to a flaw alarm. However, in the present embodiment, in which the flaw distribution for each width direction position is collated with a plurality of steel plates, a large number of flaws are detected by “rating 1”, and the type and size of the flaws detected are determined. Irrespective of this, it is possible to distinguish a flaw such as a single flaw 31, a water droplet 32 and a formation 33 from a flaw that occurs continuously. Then, when the continuous flaw 34 is generated for a plurality of steel sheets S, the continuous flaw 34 is determined and the flaw warning is output, so that the flaw warning can be output with high accuracy.
[0035]
In the present embodiment, since the continuous flaw 34 is detected as described above, for example, when a roll flaw caused by dive occurs, the roll that causes the damage is specified before the damage is expanded. It becomes possible.
[0036]
For example, in the case of the conventional detection method, the number of roll flaw damage caused by plunge was about 6 to 7, but as a result of performing the flaw detection method of the steel sheet of the present embodiment, it can be reduced to about 3 flaws. As a result, roll damage was minimized.
[0037]
(Another embodiment of the present invention)
The method for detecting a flaw on a steel sheet, the apparatus for detecting a flaw on a steel sheet, the computer program, and the computer-readable recording medium according to the embodiment described above are configured by a computer CPU or MPU, RAM, ROM, or the like. Can be realized by operating a program stored in a RAM or a ROM.
[0038]
Therefore, the present invention can be realized by recording a program that causes a computer to perform the above functions on a recording medium such as a CD-ROM, and reading the program into the computer. As a recording medium for recording the above program, a flexible disk, a hard disk, a magnetic tape, a magneto-optical disk, a nonvolatile memory card, or the like can be used in addition to the CD-ROM.
[0039]
Further, the functions of the above-described embodiments are realized not only by the computer executing the supplied program, but also the program cooperates with an operating system (OS) or other application software running on the computer. When the functions of the above-described embodiments are realized, or when all or a part of the processing of the supplied program is performed by a function expansion board or a function expansion unit of a computer to realize the functions of the above-described embodiments. Such a program is also included in the embodiment of the present invention.
[0040]
Further, in order to use the present invention in a network environment, all or some of the programs may be executed on another computer. For example, the screen input processing may be performed by a remote terminal computer, and various determinations and log recording may be performed by another center computer or the like.
[0041]
【The invention's effect】
As described above, according to the present invention, for the same position in the width direction of the steel sheet, the number of flaws detected regardless of the type and size of the flaw is integrated over the longitudinal direction. The width direction position where the number of flaws exceeds a predetermined threshold is detected, the detected width direction position is compared for a plurality of steel plates, and a peak portion of the number of flaws exists at the same width direction position of the plurality of steel plates. In such a case, since it is determined that there is a continuous flaw, it is possible to detect a continuous flaw by removing a disturbance element such as a single flaw or water. Thereby, even if a minor flaw that is overlooked when a continuous flaw is detected with respect to a single steel sheet is detected as a continuous flaw with high accuracy when it continuously occurs at the same width direction position, it is detected. be able to.
[0042]
In addition, according to another feature of the present invention, since the roll generating the continuous flaw can be identified early, it is possible to remove the cause of the continuous flaw early, and the continuous flaw is formed on a plurality of steel sheets. It is possible to reliably prevent the occurrence of a large amount.
[Brief description of the drawings]
FIG. 1 is a view showing an embodiment of the present invention and is a view for explaining an arrangement state of a hot flaw detection device.
FIG. 2 is a block diagram illustrating a configuration example of a flaw detection device for a steel plate.
FIG. 3 is a flowchart illustrating a processing procedure of a method for detecting a flaw on a steel sheet.
FIG. 4 is a diagram showing an example of the types of flaws generated on the surface of a steel sheet and the results of integrating the generated flaws in the width direction of the steel sheet.
FIG. 5 is a diagram showing an example in which many flaws are generated at the same width direction position of a plurality of steel plates.
FIG. 6 is a diagram showing a detailed example of the number of flaws when a plurality of flaws are generated at a position in the width direction.
FIG. 7 is a diagram showing an example in which the number of flaws detected differs depending on the difference in evaluation scores.
[Explanation of symbols]
REFERENCE SIGNS LIST 1 finish rolling device 2 hot flaw inspection device 3 winding device 20a flaw detection device for upper steel plate 20b flaw detection device for lower steel plate 21 steel plate imaging camera 22 lighting device 23 movable apron S steel plate

Claims (10)

A flaw detection method for a steel sheet that detects continuous flaws of the steel sheet based on image data obtained by imaging the surface of the steel sheet,
A flaw detection step of detecting a surface flaw of the steel sheet from the image data,
For a surface flaw of the steel sheet detected in the flaw detection step, a flaw integration step of integrating the number of flaws along the longitudinal direction for each position in the width direction of the steel sheet,
A continuous flaw candidate detection step of detecting as a continuous flaw candidate if the number of flaws integrated in the flaw integration step exceeds a predetermined value,
A continuous flaw determining step of collating the width direction position of the steel sheet having the continuous flaw candidate detected in the continuous flaw candidate detection step with respect to a plurality of steel sheets to determine whether the continuous flaw candidate is a continuous flaw. A method for detecting flaws on a steel sheet, comprising: 2. The method according to claim 1, wherein the surface flaw detection of the steel sheet in the flaw detection step is performed irrespective of the type and size of the flaw. Roll flaw determination step of determining whether the flaw detected in the continuous flaw candidate detection step is a roll flaw,
3. The method for detecting flaws on a steel sheet according to claim 1, further comprising a roll flaw position specifying step of specifying a roll flaw occurrence position determined in the roll flaw determination step. 4. The method according to claim 3, wherein in the roll flaw identification step, the position at which the roll flaw is generated is determined by rebate with a rolling rate of the steel sheet. A flaw detection device for a steel sheet that detects continuous flaws of the steel sheet based on image data obtained by imaging the surface of the steel sheet,
Flaw detection means for detecting surface flaws of the steel sheet from the image data,
For the surface flaw of the steel sheet detected by the flaw detection means, flaw accumulation means for integrating the number of flaws along the longitudinal direction for each position in the width direction of the steel sheet,
A continuous flaw candidate detecting means for detecting as a continuous flaw candidate if the number of flaws integrated in the flaw integrating means exceeds a predetermined value,
A continuous flaw determining means for comparing the width direction position of the steel sheet having the continuous flaw candidate detected by the continuous flaw candidate detection means with respect to a plurality of steel sheets and determining whether or not the continuous flaw candidate is a continuous flaw; An apparatus for detecting flaws on a steel sheet, comprising: The steel sheet flaw detection device according to claim 5, wherein the surface flaw detection of the steel sheet by the flaw detection means is performed irrespective of the type and size of the flaw. Roll flaw determining means for determining whether the flaw detected by the continuous flaw candidate detection means is a roll flaw,
7. The steel sheet flaw detecting device according to claim 5, further comprising a roll flaw position specifying means for specifying a roll flaw occurrence position determined by the roll flaw determining means. The steel sheet flaw detection device according to claim 7, wherein the roll flaw identification means specifies the position at which the roll flaw is generated by rebating the roll flaw with a rolling rate of the steel sheet. A program for causing a computer to execute a steel sheet flaw detection method for detecting continuous flaws of the steel sheet based on image data obtained by imaging the surface of the steel sheet,
A flaw detection step of detecting a surface flaw of the steel sheet from the image data,
For a surface flaw of the steel sheet detected in the flaw detection step, a flaw integration step of integrating the number of flaws along the longitudinal direction for each position in the width direction of the steel sheet,
A continuous flaw candidate detection step of detecting as a continuous flaw candidate if the number of flaws integrated in the flaw integration step exceeds a predetermined value,
A continuous flaw determining step of collating the width direction position of the steel sheet having the continuous flaw candidate detected in the continuous flaw candidate detection step with respect to a plurality of steel sheets to determine whether the continuous flaw candidate is a continuous flaw. A computer program that is executed by a computer program. A computer-readable recording medium on which the computer program according to claim 9 is recorded.

Images