TWI454327B - Detecting method of steel strap terminal-cutting - Google Patents

Description

Steel strip cutting detection method

The invention relates to a detection method, and in particular to a method for detecting a steel strip tail.

In the steelmaking process, an end cutting machine is used to remove the warped two ends of the steel strip during the hot rolling process. The cut end of the strip should, under normal conditions, slip directly into the groove below the side of the end cutter. However, if the tail end of the cut-off is left on the end-cutting machine, it is very likely that the end of the stranded steel strip will be cut to the end cut before the next strip enters the end-cutting machine and the end-cutting machine is ready for the returning knife. The ramp guide on the front of the machine. In this way, when the steel strip is passed through the guide plate in front of the end cutter, the steel strip and the tail end of the strand may be rolled into the hot rolling mill together. This will not only affect the quality of the rolling of the steel strip, but even cause the whole coil of steel strip to be crushed, which will greatly increase the loss of the hot rolling line.

In order to avoid the stagnation of the tail end of the stripped steel strip and affect the rolling quality, a sound sensing method is proposed to judge the falling position of the cut end. In this method, an acoustic sensor is disposed under the groove of the end-cutting machine to receive the impact sound when the tail of the steel strip falls, for the operator to judge. However, due to the many sound disturbances at the hot rolling mill site, the sensing effect of the sound sensor is seriously affected. As a result, the correctness of the detection will be poor. In addition, since the sound sensor is disposed under the groove of the end-cutting machine, it is subjected to the impact of the falling of the steel strip, and the sound sensor has a high frequency of failure and is inferior in practicability.

Therefore, one aspect of the present invention is to provide a method for detecting the cutting of a steel strip by using a non-contact method of photographic image taking to detect the residual tail of the steel strip. In this way, the detecting device can avoid contact with the production equipment, and can effectively avoid the interference of the scene noise, vibration and dust, thereby reducing the failure rate of the detecting device.

Another aspect of the present invention provides a method for detecting a tail of a steel strip, which can effectively detect the condition of the steel strip in the end-cutting machine, which is beneficial to the on-site staff to immediately eliminate the abnormal condition. Therefore, in addition to reducing the burden on the field staff, the use of this method can avoid the influence of the cutting of the steel strip on the rolling quality of the steel strip, thereby improving the efficiency of the hot rolling operation and reducing the process cost.

In accordance with the above object of the present invention, a method of detecting a tail of a steel strip is provided, which comprises the following steps. The first sensing step is performed on the one end cutting machine to obtain a first time point of the steel strip entering the end cutting machine. A first photographic taking step is performed inside the end cutting machine to obtain a first image. An image processing unit of one controller performs image processing on the first image to obtain a plurality of first feature values of the first image. Based on these first characteristic values, a Region Of Interest (ROI) is determined. A second sensing step is performed on the end cutting machine to obtain a second time point for the end cutting machine to perform all tail treatment on the steel strip. A second photographic taking step is performed inside the end cutting machine to obtain a second image. The image processing unit performs another image processing on the second image in the image processing range to obtain a plurality of second feature values of the second image. The controller performs a determining step according to the second characteristic values to determine whether the second image in the image processing range has an image of the tail portion of the steel strip. When the result of the determining step is YES, the second photographic image capturing step, the other image processing step, and the determining step are repeated a predetermined number of times. When the result of the judging step in the preset number of times is YES, it means that the tail-cut portion remains in the end-cutting machine.

According to an embodiment of the invention, the step of performing the first photographic image capturing step and the second photographic image capturing step comprises using a color camera or an infrared camera, and each of the first image and the second image is a color image.

According to another embodiment of the present invention, the following steps are performed when performing each image processing step and another image processing step. Convert a color image into a grayscale image. A noise reduction process is performed on the grayscale image. A grayscale image is converted to a binarized image using a predetermined threshold (Threshold). The white block in the binarized image is divided into a plurality of blocks by using the segmentation method in the image processing technology. A eigenvalue calculation process is performed on the blocks to obtain a first eigenvalue and a second eigenvalue.

According to still another embodiment of the present invention, before the step of dividing the white block into the plurality of blocks, each of the image processing steps and the another image processing step further comprise utilizing a Morphology method in the image processing technology. , smoothing the edges of white blocks.

According to still another embodiment of the present invention, the performing the noise filtering process includes using a Gaussian Smooth Filter.

According to still another embodiment of the present invention, the predetermined threshold value is 149.

According to still another embodiment of the present invention, the first characteristic value includes a rectangularity and an area of an image of the steel strip.

According to still another embodiment of the present invention, the second characteristic value includes an area of an image of the tail-cut portion and an X coordinate of a center of the position.

According to still another embodiment of the present invention, in the determining step, when the area of the image of the tail-cutting portion is within the first predetermined range, and the X coordinate of the center of the position of the image of the tail-cutting portion is in the second pre-predetermined When the range is set, the result of these judgment steps is yes.

According to still another embodiment of the present invention, when the result of the determining step is YES, the method for detecting the tail of the steel strip further comprises issuing a warning.

Please refer to FIG. 1 , which is a schematic diagram of a device for detecting a steel strip tail in accordance with an embodiment of the present invention. In the present embodiment, the strip cutting detection system 100 mainly includes a camera 102, a controller 104, a sensor 112, and an alarm device 114. The camera 102 is disposed beside the opening of the end cutting machine and can perform photographic imaging against the inside of the end cutting machine. The camera 102 can be an infrared camera or a color camera, such as an industrial color camera.

The sensor 112 can be mounted in an end cutting machine. The sensor 112 can be used not only to sense the point in time at which the steel strip enters the end-cutting machine, but also to sense the point in time at which the end-cutting machine performs the tail-cutting process on the steel strip.

Both camera 102 and sensor 112 are coupled to controller 104 to communicate the captured image and sensed information to controller 104. In an embodiment, the controller 104 mainly includes an image capture card 106, an image processing unit 108, and a control unit 110. Controller 104 can be, for example, a computer device.

In the controller 104, the image capture card 106 is in communication with the camera 102 to capture images captured by the camera 102. The image capture card 106 is also in communication with the image processing unit 108, and images captured from the camera 102 can be transferred to the image processing unit 108. The image processing unit 108 can perform image processing on the image captured by the image capture card 106.

The image processing unit 108 is connected to the control unit 110, and the image information obtained after the image processing can be transmitted to the control unit 110. The control unit 110 can determine whether the strip tail is left in the end-cutting machine according to the image information from the image processing unit 108. The control unit 110 can include a digital input/output module (DI/DO Module). Therefore, the image information processed by the image processing unit 108 can be digital information. On the other hand, the sensor 112 is also in communication with the control unit 110, and information about the point in time at which the steel strip enters the end-cutting machine and the end-cutting machine performs the tail-cutting process on the steel strip is transmitted to the control unit 110. Similarly, the time information obtained by the sensor 112 can be digital information.

The alarm device 114 is also connected to the control unit 110. When the control unit 110 determines that the tail portion of the steel strip is retained in the end-cutting machine, an activation signal can be issued to the alarm device 114 to drive the alarm device 114 to issue an alarm. In some embodiments, the alarm device 114 can be an alarm, a display that can display alert information, or a warning light.

Please refer to FIG. 1 and FIG. 2 together. FIG. 2 is a flow chart showing a method for detecting a steel strip tail according to an embodiment of the present invention. In the present embodiment, when the method for detecting the tail of the strap cutting is performed, the end cutting machine is sensed by the sensor 112 as described in step 202 of FIG. 2 to sense the tail cutting processing. Whether the steel strip has entered the end cutting machine and the time point at which the steel strip enters the end cutting machine.

When the sensor 112 senses that the steel strip enters the end-cutting machine, as described in step 204, the interior of the end-cutting machine is imaged by the camera 102 to obtain an image of the steel strip inside the end-cutting machine. In one embodiment, the image obtained by the camera 102 for taking an image is a color image. Next, the image captured by the camera 102 is captured by the image capture card 106 of the controller 104. The image captured by the image capture card 106 is transmitted to the image processing unit 108 of the controller 104. Then, as described in step 206, the image processing unit 108 performs image processing on the image transmitted by the image capturing card 106.

Please refer to FIG. 3, which is a flow chart of an image processing method according to an embodiment of the present invention. In one embodiment, since the obtained color image is composed of three channels of red, green and blue (RGB) images, and the steel strip is in a red hot state, the color difference of the red channel of the steel strip image is more obvious. Therefore, as shown in FIG. 3, when the image processing 300 is performed, the image grayscale value of the red channel in the color image is separated and analyzed as described in step 302, thereby converting the color image into a grayscale image. Among them, the image grayscale value ranges from 0 to 255.

Next, as described in step 304, the grayscale image is subjected to noise filtering processing using, for example, an image filter to filter out the noise of the grayscale image. The grayscale value fraction of the grayscale image filtered out of the noise is analyzed. In one embodiment, referring to FIG. 4, the grayscale values of the steel strips are all greater than 149. Thus, in this embodiment, as described in step 306 of image processing 300, the grayscale image is converted to a binarized image with 149 as the preset threshold. That is, this embodiment marks pixels having a gray value of more than 149 as 255 (white), and pixels having a gray value of less than 149 are indicated as 0 (black). In this way, the grayscale image can be converted into a binarized image composed only of white and black pixels. In other embodiments, the preset threshold may be adjusted according to the grayscale value distribution of each grayscale image, and is not limited to 149 of the above embodiment.

Next, as described in step 308, according to the degree of smoothness of the obtained binarized image, the Dilation processing method such as the pattern method in the image processing technology can be selectively utilized to make the binarized image The jagged edges of the white blocks are smoother. Next, as described in step 310, the white block in the binarized image is segmented into a plurality of blocks using a segmentation method in the image processing technique. These blocks correspond to images of objects attached to the end-cutting machine, such as steel strip images.

Then, as described in step 312, the image feature values of the blocks are calculated, such as the block area, the similarity of the block to the rectangle, the Bounding Rectangle surrounding the block corresponding to the steel strip image, and the block. The center's X-axis coordinates. The similarity between the block and the rectangle is also called the Rectangularity of the block, and the X-axis coordinate of the center of the block can be defined as the X coordinate of the center of the smallest rectangle. In one embodiment, the area of the block corresponding to the steel strip image may be 12817 pixels.

After the image processing is completed, please refer to FIG. 2 again. As described in step 208, according to the feature values obtained after the image processing, the image processing range of the entire image to be further processed is determined, thereby greatly reducing the subsequent processing. Calculation and detection time. In an embodiment, after actual observation on the spot, it is found that when the steel strip is left behind, the tail cutting portion will be located on the platform of the end cutting machine. Therefore, this embodiment sets the image processing range to the area on the platform of the end-cutting machine, and estimates the range of image processing by enclosing the smallest rectangle corresponding to the block of the steel strip image in the image feature value.

For example, referring to FIG. 5, in the binarized image 400, let A be the X coordinate value of the left edge 406 of the smallest rectangle 402 surrounding the block corresponding to the steel strip image, and the image processing range 404 The X coordinate value of the left edge 412 is defined as the width 410 of A + 0.21 x the minimum rectangle 402, the width 414 of the image processing range 404 is defined as the width 410 of the minimum rectangle 402 of 0.54 x, and the height 416 of the image processing range 404 is defined as the same. The height 408 of the smallest rectangle 402.

Next, referring again to FIG. 1 and FIG. 2, as described in step 210, the end-cutting machine is again sensed by the sensor 112 to sense whether the end-cutting machine has finished cutting the steel strip. And get the time point of the steel strip cutting. After the sensor 112 senses that the end cutting machine has finished cutting the steel strip, as described in step 212, the camera 102 is again used to perform photographic imaging on the inside of the end cutting machine to obtain a steel strip cutting tail end cut. The image inside the machine. Similarly, the image obtained by the camera 102 for capturing images may be a color image. Then, the image captured by the camera 102 is captured by the image capture card 106 of the controller 104. The image captured by the image capture card 106 is transmitted to the image processing unit 108 of the controller 104. Then, as described in step 214, the image processing unit 108 performs image processing on the image transmitted by the image capturing card 106.

In step 214 of this embodiment, when the image processing is performed, the image processing program used may be the same as the image processing program of step 206. For example, the image processing programs of steps 214 and 206 can be performed in accordance with the program of image processing 300 shown in FIG. Steps 214 and 206 differ in that the image processing of step 214 only processes the portion of the image that is in the image processing range. After the image processing of step 214 is completed, the feature value of the image after the tail of the steel strip is obtained, and includes, for example, the area of the block corresponding to the image of the tail portion and the X-axis coordinate of the center of the position of the block.

Next, as described in step 216, the control unit 110 of the controller 104 of FIG. 1 is used to filter the area of the image falling within the preset area, and the X coordinate of the center of the position in the image falls within the preset coordinate range. Block (ie the image of the tail cut). When a block is selected, the area falls within a preset area, and the X coordinate of the center of the position also falls within the preset coordinate range, and the control unit 110 can determine the second image located in the image processing range accordingly. There is an image of the tail section of the steel strip. If there is no block selected, the area falls within the preset area and the X coordinate of the center of the position also falls within the preset coordinate range, the control unit 110 can determine the first in the image processing range. The second image does not have an image of the tail portion of the steel strip. At this point, the detection method 200 can be re-executed to detect the tail-tailing condition of the next steel strip entering the end-cutting machine for cutting.

When the result of the determination in step 216 is YES, the steps 212, 214 and 216 are repeated a predetermined number of times, for example ten times, as described in step 218. When the result of the judgment step 216 obtained by the preset number of times is YES, it means that the cut end portion of the steel strip does remain in the end slitter. At this time, referring again to FIG. 1, the control unit 110 of the controller 104 can issue an activation signal to the alarm device 114 to drive the alarm device 114 to issue an alarm. After the alarm device 114 of the detection system 100 issues an alert, the camera capture and image processing will be suspended. At this point, the field staff can check whether there is an abnormal condition and further eliminate the abnormal condition.

In some embodiments, after the steel strip is tail-cut, the first step of the determining step 216 is to find the image of the tail-cut portion, and the subsequent determining step may only be for the position center of the block corresponding to the image of the tail-cut portion. Whether the range of variation of the X coordinate falls within a certain range. In an example, as shown in FIG. 6, after the image of the tail-cut portion is found in the first determining step 216, the position center of the block corresponding to the image of the tail-cut portion is determined in the subsequent ten determinations. When the X coordinate is oscillated between the X coordinate of the left edge of the image processing range and the X coordinate of the right edge, it means that the cut end portion of the steel strip does remain in the end cutting machine.

It can be seen from the above embodiments that one of the advantages of the present invention is that the detection method of the steel strip tail cutting of the present invention utilizes the non-contact method of photographic image taking to detect the residual tail of the steel strip. Therefore, the detecting device can avoid contact with the production equipment, and can effectively avoid the interference of the scene noise, vibration and dust, thereby reducing the failure rate of the detecting device.

It can be seen from the above embodiments that the method for detecting the tail of the steel strip of the present invention can effectively detect the condition of the steel strip in the end-cutting machine, which is beneficial to the on-site staff to eliminate the abnormal condition immediately. Therefore, in addition to reducing the burden on the field staff, the use of this method can avoid the influence of the cutting of the steel strip on the rolling quality of the steel strip, thereby improving the efficiency of the hot rolling operation and reducing the process cost.

While the present invention has been described above by way of example, it is not intended to be construed as a limitation of the scope of the invention. Therefore, the scope of the invention is defined by the scope of the appended claims.

100. . . Detection system

102. . . camera

104. . . Controller

106. . . Image capture card

108. . . Image processing unit

110. . . control unit

112. . . Sensor

114. . . Alarm device

200. . . Detection method

202. . . step

204. . . step

206. . . step

208. . . step

210. . . step

212. . . step

214. . . step

216. . . step

218. . . step

300. . . Image processing

302. . . step

304. . . step

306. . . step

308. . . step

310. . . step

312. . . step

400. . . Binarized image

402. . . Minimum rectangle

404. . . Image processing range

406. . . Left edge

408. . . height

410. . . width

412. . . Left edge

414. . . width

416. . . height

The above and other objects, features, advantages and embodiments of the present invention will become more apparent and understood.

1 is a schematic view of a device for detecting a steel strip tail in accordance with an embodiment of the present invention.

2 is a flow chart showing a method for detecting a steel strip tail according to an embodiment of the present invention.

3 is a flow chart showing an image processing method according to an embodiment of the present invention.

FIG. 4 is a diagram showing a gray scale value distribution diagram of a gray scale image according to an embodiment of the present invention.

FIG. 5 is a schematic diagram showing the relative relationship between the minimum rectangle surrounding the block corresponding to the image of the steel strip and the image processing range according to an embodiment of the present invention.

Fig. 6 is a graph showing the variation of the position center X coordinate of the block corresponding to the image of the tail portion of the steel strip on the time axis.

200. . . Detection method

202. . . step

204. . . step

206. . . step

208. . . step

210. . . step

212. . . step

214. . . step

216. . . step

218. . . step

Claims (10)

A method for detecting a tail of a steel strip comprises: performing a first sensing step on one end cutting machine to obtain a first time point of a steel strip entering the end cutting machine; and performing an internal portion of the end cutting machine a photographic imaging step to obtain a first image; performing image processing on the first image by using an image processing unit of a controller to obtain a plurality of first eigenvalues of the first image; a characteristic value, determining an image processing range; performing a second sensing step on the end cutting machine to obtain a second time point of the end cutting machine performing all tail processing on the steel strip; the end of the end cutting machine Performing a second photographic image capturing step to obtain a second image; performing, by the image processing unit, another image processing on the second image in the image processing range to obtain a plurality of second features of the second image Using the controller to perform a determining step according to the second characteristic values to determine whether the second image in the image processing range has an image of all tail portions of the steel strip; and when the result of the determining step is Yes And repeating the second photographic image capturing step, the another image processing step, and the determining step a predetermined number of times, wherein when the result of the determining steps in the preset number of times is YES, indicating The tail cut portion remains in the end cutter. The method for detecting a tail of a steel strip according to claim 1, wherein the step of performing the first photographing and the step of taking the second photographing comprises using a color camera or an infrared camera, and each of the first The image and the second image are a color image. The method for detecting a steel strip tail according to claim 2, wherein each of the image processing steps and the another image processing step comprises: converting the color image into a grayscale image; and performing the grayscale image on the grayscale image a noise filtering process; converting the gray scale image into a binarized image by using a preset threshold; using one of the image processing techniques to divide the white block into the binarized image Dividing into a plurality of blocks; and performing an eigenvalue calculation process on the blocks to obtain the first feature values and the second feature values. The method for detecting a tail of a steel strip according to claim 3, wherein before the step of dividing the white block into the blocks, each of the image processing steps and the another image processing step further comprise utilizing the One of the image processing techniques is to smooth the edges of the white block. The method for detecting a steel strip tail according to claim 3, wherein the noise filtering processing comprises using an image filter. The method for detecting a steel strip tail according to claim 3, wherein the preset threshold is 149. The method for detecting a steel strip tail according to claim 3, wherein the first characteristic values comprise a rectangularity and an area of an image of the steel strip. The method for detecting a steel strip tail according to claim 3, wherein the second characteristic values include an area of an image of the tail portion and an X coordinate of a center of the position. The method for detecting a tail of a steel strip according to claim 8, wherein in the determining step, when the area of the image of the tail portion is within a first predetermined range, and the image of the tail portion is When the X coordinates of the position center are both within a second predetermined range, the result of the determining steps is yes. The method for detecting the tail of the steel strip as described in claim 1 further includes issuing a warning when the result of the determining step is YES.