Method for detecting surface defects in wire materials

Abstract

The present invention provides a wire rod surface defect detection method that can detect non-opening linear defects in wire rods, caused by the material or rolling process, online on a wire rod rolling line without detecting harmless defects. [Solution] The wire rod surface defect detection method involves installing a surface inspection device 4 on the exit side of the rolling mill 2 in the wire rod rolling line 1, which detects surface defects in the wire rod S2 based on the difference in brightness value due to the difference in spectral emissivity caused by the difference in scale thickness between the defective part and the sound part, thereby detecting non-opening linear defects D formed on the wire rod S2, which are material-derived or rolling-derived.

Description

Technical Field

[0001] The present invention relates to a method for detecting surface defects of a wire rod, which detects non-opening linear defects caused by a material formed on the wire rod or by rolling.

Background Art

[0002] A wire rod, which is a kind of steel product, is a material used for springs of automobiles and the like. This wire rod is manufactured by a wire rod rolling line including a heating furnace for heating the material to a predetermined temperature, a rolling mill for rolling the material heated in the heating furnace, a cooling zone for cooling the wire rod rolled by the rolling mill, and a winding device for winding the wire rod cooled in the cooling zone.

[0003] Here, on the surface of the wire rod, defects called non-opening linear defects may occur in the manufacturing process by the wire rod rolling line. The non-opening linear defect is a defect that extends along the longitudinal direction of the wire rod with an extremely small opening width of about several μm on the surface of the wire rod and cannot be visually observed in a cross section cut across the wire rod. This non-opening linear defect is formed either due to the material or due to rolling. The non-opening linear defect formed due to the material is generated by rolling the defect in the material with a rolling mill. Further, the non-opening linear defect due to rolling is generated by finally rolling the defect extending along the longitudinal direction of the wire rod generated in the rolling mill or the conveying equipment with the rolling mill.

[0004] On the other hand, in the wire rod rolling line, due to the characteristics of the manufacturing process such as a conveying speed exceeding 100 m / s at maximum and a process of winding hot, it is difficult to ensure the quality of all wire rods. The above-mentioned non-opening linear defect of the wire rod has a great influence on the quality of the wire rod, but it is difficult to detect it with a generally used through-type eddy current flaw detector suitable for online full inspection. Further, the non-opening linear defect can be detected by magnetic particle inspection of the terminal sample of the wire rod, but it is difficult to detect defects outside the terminal sample range. The non-opening linear defect outside the terminal sample range often becomes apparent by cold forging processing by a secondary processing manufacturer and is often detected after the secondary processing manufacturer.

[0005] Furthermore, since it is difficult to perform a full-length, full-circumference inspection of the wire after it has been wound, it is desirable to detect non-opening linear defects in the wire, whether material-derived or rolling-derived, online at high speed and with high sensitivity on the wire rolling line. Herein, conventionally, a method for detecting surface defects in workpieces such as rolled and drawn metal rods is known, as shown in Patent Document 1, which involves capturing the entire circumference of the surface of a rod material using multiple high-speed visible cameras and detecting surface defects in the rod material by image processing the captured image data. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Patent No. 4642474 [Overview of the project] [Problems that the invention aims to solve]

[0007] However, the conventional method for detecting surface defects in workpieces such as rolled and drawn metal rods, as shown in Patent Document 1, had the following problems. In other words, the surface defect detection method for workpieces such as rolled and drawn metal rods shown in Patent Document 1 involves using multiple high-speed visible cameras to photograph the entire circumference of the surface of a rod as it moves at high speed, and then processing the captured image data to detect defects extending along the longitudinal direction with a width of 0.025 mm. It is believed that non-opening linear defects formed on wires can also be detected. However, there is a problem in that it often produces false positives, including shallow scratches that would otherwise be judged as harmless.

[0008] Therefore, the present invention has been made to solve this conventional problem, and its objective is to provide a wire surface defect detection method that can detect non-opening linear defects caused by the material or rolling process on the wire without detecting harmless defects, in an online wire rolling line. [Means for solving the problem]

[0009] To solve the above problems, a wire surface defect detection method according to one aspect of the present invention involves installing a surface inspection device on the exit side of the rolling mill in a wire rolling line, which detects surface defects in the wire based on the difference in brightness value due to the difference in spectral emissivity caused by the difference in scale thickness between the defective part and the sound part, thereby detecting non-opening linear defects in the wire caused by the material or rolling process. [Effects of the Invention]

[0010] According to the wire rod surface defect detection method of the present invention, non-opening linear defects caused by the material or rolling process, formed on the wire rod, can be detected online on the wire rod rolling line without detecting harmless defects. [Brief explanation of the drawing]

[0011] [Figure 1] This is a schematic diagram of a part of a wire rolling line to which a wire surface defect detection method according to one embodiment of the present invention is applied. [Figure 2] Figure 1 is a diagram illustrating the surface inspection device in the wire rolling line shown. [Figure 3] Figure 2 shows the imaging equipment of the surface inspection apparatus as viewed from the direction in which the wire material is transported. [Figure 4] This diagram illustrates the guide range for wires in a wire conveying system. [Figure 5] This graph shows the calculated values ​​of the spectral emissivity of iron against the oxide film thickness on the iron surface and the measurement wavelength. [Figure 6] This figure shows the results of magnetic particle testing when non-aperture linear defects were intentionally created in a wire. [Figure 7] This figure shows images taken within the shooting range shown in Figure 4 when a non-opening linear defect was intentionally created in the wire material. [Figure 8]This explains the masking process using a specified noise width on an image extracted from the captured image shown in Figure 7, where the wire material being inspected is visible. (a) is an image extracted from the captured image shown in Figure 7, where the wire material being inspected is visible, and (b) is an image taken after masking with a specified noise width has been applied to the captured image shown in (a). [Modes for carrying out the invention]

[0012] Embodiments of the present invention will be described below with reference to the drawings. The embodiments shown below are illustrative examples of devices and methods for realizing the technical concept of the present invention, and the technical concept of the present invention is not limited to the following embodiments in terms of the material, shape, structure, arrangement, etc. of the components. Furthermore, drawings are schematic representations. Therefore, it should be noted that the relationship and ratios between thickness and planar dimensions may differ from those in reality, and there may be differences in dimensional relationships and ratios between drawings themselves.

[0013] Figure 1 shows a schematic configuration of a part of a wire rolling line to which a wire surface defect detection method according to one embodiment of the present invention is applied. In the wire rolling line 1, a material S1 such as a billet made of steel is heated to a predetermined temperature in a heating furnace (not shown), and then rolled by a rolling mill 2 to become wire S2. The rolling mill 2 is equipped with a roughing mill 21, an intermediate rolling mill 22, and a finishing mill 23. The wire S2 is then cooled in a water cooling zone 3 and then wound up by a winding device (not shown). In this way, the wire S2 is manufactured.

[0014] Here, defects called non-opening linear defects D (see Figures 6 to 8(a), (b)) may occur on the surface of the wire rod S2 during the manufacturing process by the wire rod rolling line 1. Non-opening linear defects D are defects that extend along the longitudinal direction of the wire rod S2 and are invisible to the naked eye when the wire rod S2 is cut across, with an opening width of only a few micrometers on the surface of the wire rod S2. These non-opening linear defects D can be formed due to material defects or due to rolling. Non-opening linear defects D formed due to material defects occur when the rolling mill 2 rolls defects that extend along the longitudinal direction of the wire rod S2 and were generated in the rolling mill 2 or conveying equipment. This non-opening linear defect D becomes apparent during cold forging at secondary processing manufacturers, and is therefore often detected at secondary processing manufacturers and beyond.

[0015] In order to detect these non-opening linear defects D online on the wire rolling line 1, the wire rolling line 1 of this embodiment has a surface inspection device 4 installed on the exit side of the rolling mill 2 that detects surface defects in the wire S2 based on the difference in brightness value due to the difference in spectral emissivity caused by the difference in scale thickness between the defective part and the sound part.

[0016] Here, FIG. 5 shows the calculated values of the spectral emissivity of iron with respect to the oxide film thickness on the surface of iron and the measurement wavelength. According to the calculation, when the oxide film thickness is greater than 0 μm (in FIG. 5, when d is greater than 0 μm), the spectral emissivity is 0.8 in the short wavelength region of the measurement wavelength, and the spectral emissivity is about 0.8 up to a certain wavelength. When exceeding a certain wavelength, the spectral emissivity rapidly decreases and drops to about 0.1 at a sufficiently large wavelength. Also, the magnitude of the wavelength at which the spectral emissivity starts to rapidly decrease is roughly proportional to the oxide film thickness. In FIG. 5, when d is 0.04 μm, the spectral emissivity starts to rapidly decrease near a measurement wavelength of 0.8 μm, when d is 0.2 μm, the spectral emissivity starts to rapidly decrease near a measurement wavelength of 3.2 μm, and when d is 0.52 μm, the spectral emissivity starts to rapidly decrease near a measurement wavelength of 7.3 μm. Therefore, it can be seen that by appropriately selecting the imaging wavelength, when there are differences in the oxide film thickness on the surface of iron, there is a possibility that it can be captured as a large spectral emissivity difference.

[0017] In the manufacturing process of hot rolling and processing the wire rod S2, black scale due to high-temperature acid is generated on the surface of the wire rod S2. In such a manufacturing process, when there are surface defect flaws such as non-opening linear flaws D on the surface of the wire rod S2, scale is generated in the part that is in contact with air even inside the defect. The scale generated along the shape of the defect in this way appears as a region with a substantially larger scale thickness than the healthy part when viewed from the imaging device. Therefore, when scale formation progresses on the surface of the wire rod S2 where there are harmful defects such as non-opening linear flaws D, a difference in scale thickness occurs between the defective part and the healthy part, a difference in spectral emissivity occurs in a specific wavelength band, and there is a possibility of detecting the defective part by capturing the difference in luminance values associated with the difference in spectral emissivity. For this reason, in the present embodiment, a surface inspection device 4 for detecting surface flaws of the wire rod S2 is used based on the difference in luminance values associated with the difference in spectral emissivity caused by the difference in scale thickness occurring between the defective part and the healthy part, and by installing the surface inspection device 4 on the outlet side of the rolling mill 2, non-opening linear flaws D caused by the material or rolling are detected.

[0018] The configuration of the surface inspection device 4 will be described. As shown in FIG. 2, the surface inspection device 4 includes an imaging device 41 that captures the surface of the wire S2, and an image processing device 42 that processes the captured image captured by the imaging device 41 to detect surface defects of the wire S2. As shown in FIG. 3, a plurality of (four in this embodiment) imaging devices 41 are installed at appropriate intervals (90° intervals in this embodiment) along the outer circumference of the wire S2, enabling the entire circumference of the wire S2 to be imaged. It is not always necessary to install four imaging devices 41 at 90° intervals along the outer circumference of the wire S2. One imaging device may be installed according to the wire S2 to be imaged, or a plurality of imaging devices other than four may be installed.

[0019] Each imaging device 41 has a light reception sensitivity in the mid-infrared wavelength band (3 - 5 μm). This is because a spectral emissivity difference occurs between the defective part and the healthy part on the surface of the wire S2 in this mid-infrared wavelength band (3 - 5 μm). However, the imaging device 41 may have a light reception sensitivity in the far-infrared wavelength band (5 μm - 1 mm). The imaging device 41 has a function of continuously imaging mid-infrared light or far-infrared light. Specifically, it is a camera having a two-dimensional cooled element sensor or a two-dimensional bolometer, or a one-dimensional line camera, etc. As described above, the imaging device 41 has a light reception sensitivity in the mid-infrared or far-infrared wavelength band and images the hot material in which scale generation on the surface of the wire S2 is progressing. Each imaging device 41 preferably has a cooled element sensor that is relatively fast and has a fast sensitivity response when receiving light in order to image the entire length of the wire S2 being conveyed at high speed. In addition, each imaging device 41 is equipped with an optical system such as an appropriate lens and aperture, and has sufficient resolution and depth of field for detecting surface defects of the wire S2. The resolution may be about 0.1 mm / pix when targeting a non-opening linear defect D. Also, the depth of field may be designed such that the focus of the imaging range is adjusted according to the conveyance vibration and curvature of the wire S2.

[0020] Furthermore, as shown in Figure 4, in the conveying equipment 5 at the position where the wire S2 is photographed by the imaging device 41 (a pair of conveying equipment 5 in front of and behind the conveying direction of the wire S2 relative to the imaging range PR of the wire S2 by the imaging device 41), it is preferable to set the guidance range IR of the wire S2 to within +5 mm of the product diameter d of the wire S2. In the rolling of the wire S2, high-speed rolling (rolling at a maximum of 100 m / s) is required, so it is necessary to increase the imaging speed of the wire S2 and the processing speed of the image processing device 42. Therefore, it is required to increase the processing speed of the image processing by making the imaging range PR of the wire S2 by the imaging device 41 as small as possible. In the conveying equipment 5 at the position where the wire S2 is photographed by the imaging device 41, by setting the guidance range IR of the wire S2 to within +5 mm of the product diameter d of the wire S2, the vibration of the wire S2 during conveying can be suppressed. This allows the imaging range PR of the wire S2 by the imaging device 41 to be made as small as possible (the imaging range PR of the wire S2 by the imaging device 41 can be set to a maximum of 30 mm x 30 mm), and the processing speed of the image can be increased. For wires S2 with a thin product diameter d, the transport speed is fast, and high-speed imaging and high-speed image processing are required, so it is particularly effective to set the guidance range IR of the wire S2 to within +5 mm of the product diameter d of the wire S2. The transport equipment 5 consists of pinch rolls, troughs for material transport, pipes for material transport, etc.

[0021] Furthermore, the image processing device 42 processes the captured images G (see Figure 7) taken by each imaging device 41 to detect surface defects in the wire S2, and is composed of a computer with calculation processing capabilities. In processing the image, the image processing device 42 first extracts the region in which the wire S2 is visible from the captured image G taken by each imaging device 41.

[0022] Next, the image processing device 42 performs a masking process on the captured image G1 (see Figure 8(a)), which is an extracted region from the captured image G in which the wire S2 is visible, by specifying a noise width. An example of the captured image G2 after the masking process is shown in Figure 8(b). In the example shown in Figures 8(a) and (b), a noise width W1 of noise N is specified on one side (upper side in Figures 8(a) and (b)) of the wire S2 in the circumferential direction (up and down direction in Figures 8(a) and (b)), and masking M is applied. Also, a noise width W2 of noise N is specified on the other side (lower side in Figures 8(a) and (b)) of the wire S2 in the width direction (up and down direction in Figures 8(a) and (b)), and masking M is applied. The noise widths W1 and W2 are predetermined by the wire S2 being rolled. Due to rolling, high-luminance band-shaped noise N may occur in the wire S2. In this case, the band-shaped noise N significantly exceeds the threshold of the luminance integral value described later, leading to false detection. By performing masking on the captured image G1 with specified noise widths W1 and W2, the noise N is eliminated in the captured image G2 after masking, reducing false detections associated with the noise N and improving the detection rate of surface defects (non-aperture linear defects D).

[0023] Then, in the masked captured image G2, the longitudinal luminance integral value is taken for each pixel in the circumferential direction of the wire S2. If the longitudinal luminance integral value is greater than or equal to a predetermined threshold, that area is determined to be a defect, and if the longitudinal luminance integral value is less than the predetermined threshold, that area is determined to be sound. Areas determined to be defects by the image processing device 42 are designated as surface defects (non-opening linear defects D).

[0024] As mentioned above, non-opening linear defects D can be formed by material defects or by rolling. Non-opening linear defects D formed by material defects occur when the rolling mill 2 rolls over defects in the material S1. Non-opening linear defects D caused by rolling occur when defects extending along the longitudinal direction of the wire S2, which were generated in the rolling mill 2 or conveying equipment, are finally rolled by the rolling mill 2. In order to enable detection of non-opening linear defects D in either case, a surface inspection device 4 is installed on the exit side of the rolling mill 2 to detect surface defects of the wire S2 based on the difference in brightness value due to the difference in spectral emissivity caused by the difference in scale thickness between the defective part and the sound part. In particular, each imaging device 41 of the surface inspection device 4 needs to be installed on the exit side of the rolling mill 2. By having each imaging device 41 photograph the surface of the wire S2 being transported on the exit side of the rolling mill 2, non-opening linear defects D caused by the material or rolling can be detected.

[0025] As described above, the surface inspection device 4 of this embodiment detects surface defects, or non-opening linear defects D, of the wire S2 based on the difference in brightness values ​​resulting from the difference in spectral emissivity caused by the difference in scale thickness between the defective area and the sound area during hot rolling. Since scale is also generated around the defective area in the depth direction relative to the wire S2 in the non-opening linear defects D, the scale thickness becomes thicker than that of the sound area. Therefore, it is necessary to perform hot imaging on the wire rolling line 1 and generate scale appropriately so that the difference in brightness values ​​between the defective area and the sound area becomes more pronounced. Scale is generated more deeply at the defect site due to cooling at the exit of the water-cooled zone 3 than immediately after the exit of the rolling mill 2. For this reason, in this embodiment, as shown in Figure 1, the surface inspection device 4 is installed at the exit of the water-cooled zone 3, which is located at the exit of the rolling mill 2. As a result, since scale is generated more deeply at the defect site due to cooling at the exit of the water-cooled zone 3 than immediately after the exit of the rolling mill 2, non-opening linear defects D can be detected more appropriately than if the surface inspection device 4 were installed immediately after the exit of the rolling mill 2.

[0026] On the other hand, since water splashing may cause field defects in the imaging equipment 41 immediately after exiting the water cooling zone 3, it is preferable to install the surface inspection device 4 at a certain distance from the water cooling zone 3 (experiments have shown that 10m or more is necessary), or to suppress water splashing by performing air purging or air blowing. Furthermore, the imaging by each imaging device 41 and the image processing by the image processing device 42 are repeated at a speed sufficient to inspect the entire length of the surface of the wire S2, and finally, surface defects (non-opening linear defects D) are automatically detected along the entire length of the wire S2.

[0027] As described above, according to the wire surface defect detection method of this embodiment, a surface inspection device 4 that detects surface defects in the wire S2 based on the difference in brightness value due to the difference in spectral emissivity caused by the difference in scale thickness between the defective part and the sound part is installed on the exit side of the rolling mill 2 in the wire rolling line 1, thereby detecting non-opening linear defects D formed on the wire S2, which are material-derived or rolling-derived. This allows for the online detection of non-opening linear defects D, either material-derived or rolling-derived, formed on the wire rod S2 without detecting harmless defects, on the wire rod rolling line 1.

[0028] Furthermore, according to the wire surface defect detection method of this embodiment, the surface inspection device 4 is installed on the outlet side of the water-cooling zone 3 located on the outlet side of the rolling mill 2. As a result, the scale is generated more deeply on the defective area at the exit of the water-cooled zone 3 due to cooling than immediately after the exit of the rolling mill 2. Therefore, non-opening linear defects D can be detected online more effectively than if the surface inspection device 4 were installed immediately after the exit of the rolling mill 2.

[0029] Furthermore, according to the wire surface defect detection method of this embodiment, the surface inspection device 4 includes an imaging device 41 having light-receiving sensitivity in the mid-infrared or far-infrared wavelength band for photographing the surface of the wire S2, and an image processing device 42 that processes the image captured by the imaging device 41 to detect surface defects in the wire S2. This enables the realization of a surface inspection device 4 with an accurate configuration that detects surface defects in the wire S2 based on the difference in brightness value resulting from the difference in spectral emissivity caused by the difference in scale thickness between the defective and sound areas, and allows for the appropriate online detection of non-opening linear defects D formed on the wire S2.

[0030] Furthermore, according to the wire surface defect detection method of this embodiment, in the transport equipment 5 at the position where the wire S2 is photographed by the imaging device 41, the guidance range IR of the wire S2 is set to within +5 mm of the product diameter d of the wire S2. This suppresses vibration of the wire S2 during transport, minimizes the imaging range PR of the wire S2 by the imaging device 41, and speeds up the image processing by the image processing device 42. As a result, non-aperture linear defects D formed on the high-speed moving wire S2 can be appropriately detected online.

[0031] Furthermore, according to the wire surface defect detection method of this embodiment, when image processing is performed by the image processing device 42, masking is performed on the captured image G1 by specifying noise widths W1 and W2. As a result, noise N is eliminated in the captured image G2 after masking, reducing false detections associated with noise N and improving the detection rate of surface defects (non-aperture linear defects D).

[0032] Although embodiments of the present invention have been described above, the present invention is not limited thereto and can be modified and improved in various ways. For example, the surface inspection device 4 only needs to be installed on the exit side of the rolling mill 2, and does not necessarily need to be installed on the exit side of the water-cooling zone 3. Furthermore, in the transport equipment 5 at the position where the wire S2 is photographed by the imaging device 41, it is not necessarily required that the guidance range IR of the wire S2 be within +5 mm of the product diameter d of the wire S2. Furthermore, when performing image processing by the image processing device 42, it is not always necessary to perform masking on the captured image G1 by specifying noise widths W1 and W2. [Examples]

[0033] To verify the effects of the present invention, a surface inspection device 4 was installed at a distance of 12.5 m or more from the water-cooling zone 3, located at the exit side of the water-cooling zone 3, which is installed at the exit side of the rolling mill 2, in the wire rolling line 1 shown in Figure 1. Meanwhile, non-opening linear defects D were intentionally formed in the wire S2, and the wire S2 was photographed. Figure 6 shows the results of magnetic particle testing when non-opening linear defects D were intentionally formed in the wire S2. In Figure 6, which shows the results of magnetic particle testing, two non-opening linear defects D (shown as two bright lines extending in the longitudinal direction, with a distance of 2 mm between them) have occurred in the wire S2. The surface of the wire S2 in which these two non-opening linear defects D have occurred was photographed using the imaging device 41 of the aforementioned surface inspection device 4 within the imaging range shown in Figure 4. The resulting image G is shown in Figure 7. In Figure 7, two non-aperture linear defects D, indicated by bright lines, can be seen in the center of the circumferential direction (vertical direction) of the captured image G, and the non-aperture linear defects D were captured. Furthermore, these two non-aperture linear defects D were also detected in the middle portion of the entire length of the wire S2, indicating that non-aperture linear defects D could be detected even outside the range of magnetic particle testing performed in a partial manner.

[0034] Furthermore, the image processing device 42 of the surface inspection device 4 performed a masking process on the captured image G1, which was extracted from the captured image G shown in Figure 7, in which the wire S2 was visible, by specifying noise widths W1 and W2, and detected two non-aperture linear defects D (see Figures 8(a) and (b)). As a result of this masking process, the noise N was eliminated in the captured image G2 after the masking process, reducing false detections associated with the noise N and improving the detection rate of non-aperture linear defects D. [Explanation of Symbols]

[0035] 1. Wire Rolling Line 2 Rolling mill 3. Water-cooled zone 4. Surface inspection device 5. Conveying equipment 21 Roughing mill 22 Intermediate Rolling Mill 23 Finishing Rolling Mill 41. Imaging equipment 42 Image Processing Devices D Non-opening linear defect G Photographed image G1 Photo G2 shooting images M Masking N Noise S1 Material S2 wire rod

Claims

1. A method for detecting surface defects in a wire rod, characterized by installing a surface inspection device on the exit side of the rolling mill in a wire rod rolling line, which detects surface defects in the wire rod based on the difference in brightness value resulting from the difference in spectral emissivity caused by the difference in scale thickness between the defective part and the sound part, thereby detecting non-opening linear defects in the wire rod that are material-derived or rolling-derived.

2. The method for detecting surface defects in a wire rod according to claim 1, characterized in that the surface inspection device is installed on the outlet side of a water-cooling zone located on the outlet side of the rolling mill.

3. The method for detecting surface defects in a wire according to claim 1 or 2, characterized in that the surface inspection apparatus comprises an imaging device having light-receiving sensitivity in the mid-infrared or far-infrared wavelength band for imaging the surface of the wire, and an image processing device for detecting surface defects in the wire by image processing the image captured by the imaging device.

4. The method for detecting surface defects in a wire according to claim 3, characterized in that, in the conveying equipment at the position where the wire is photographed by the photographing device, the guidance range of the wire is set to within +5 mm of the product diameter of the wire.

5. The method for detecting surface defects in a wire according to claim 3, characterized in that, when performing image processing by the image processing device, a masking process is performed on the captured image with a specified noise width.

Images