JP5454404B2 - Edge detection method and detection system, strip running condition measuring method and measurement system, strip running control method and control system, and strip manufacturing method and manufacturing system - Google Patents

Description

  The present invention relates to a method and a detection system for appropriately detecting the edge of a member (particularly, a strip) traveling in a production line filled with mist-like water droplets or fumes, and when the strip is conveyed in a loop shape. The present invention relates to a method and a measurement system for appropriately measuring the traveling state of the strip, a method and a control system for appropriately controlling the traveling of the strip, and a method or a manufacturing system for manufacturing the strip using these methods and systems.

(Edge detection method for strips in production lines filled with water droplets and fumes)
In steel strip production lines such as steelworks, we measure the running status of steel strips from the viewpoints of stable operation and improvement of operation rate by preventing troubles and ensuring product quality, and feed back the measurement results to the control equipment of the production line. Therefore, it is desired to control the traveling state of the steel strip. Examples of the running control of the steel strip include a meandering control in a heating furnace and a continuous rolling line, and a slack control of the steel strip loosened in a loop shape in a continuous pickling treatment line. In addition, as a method of measuring the running condition of the steel strip in a non-contact manner, edge detection by image processing is applied to the image of the running steel strip imaged by a two-dimensional camera to measure the position and angle of the edge line. Can also be considered. However, in the production line as described above, there are many cases where the environmental temperature is high or water or the like is used, and a large amount of mist-like water droplets or fumes that hinder image processing often occur. In such a case, the edge of the steel strip to be measured is hidden by the mist-like water droplets or fumes, so that stable edge detection by image processing becomes difficult.

  Patent Document 1 discloses a meandering amount between rolling stands where fume is generated in order to perform meandering control on a plate material rolled by a rolling mill having a plurality of rolling stands such as a hot continuous finish rolling mill. A method of measuring the image by image processing is disclosed. In Patent Document 1, the edge of a plate material that travels in red is imaged with a two-dimensional camera, the plate material edge position is obtained from the differential strength for each scanning line, and is hidden by a fume by the weighted least square method based on the differential strength. It is possible to obtain a straight line related to the edge portion that is not. However, this method cannot be applied when the entire edge area on the imaging screen is frequently hidden by mist-like water droplets or fumes, or when a background portion other than the plate material is reflected on the imaging screen.

(Measurement of slackness in continuous pickling line)
The scale generated in the hot rolling process is attached to the surface of the hot rolled steel strip. Therefore, in the continuous pickling treatment line, the scale on the steel strip surface is removed by continuously passing through the pickling tank storing the acid solution while loosening the steel strip in a loop shape. Yes. At that time, the amount of slack in the steel strip affects the descaling ability and quality. For example, if the slack of the steel strip in the pickling tank increases, the length of the steel strip immersed in the acid in the pickling tank becomes longer and the scale can be removed efficiently. On the other hand, the steel strip and the bottom of the pickling bath are in contact with each other. There is a risk that. Conversely, if the slack of the steel strip is reduced, there is no risk of contact between the steel strip and the bottom of the pickling tank, but the length of immersion of the steel strip in the pickling tank is shortened and the scale is efficiently removed. It becomes impossible to do. That is, it is preferable to operate while maintaining a slack amount in which a sufficiently long steel strip is immersed in acid and the steel strip does not contact the bottom of the pickling tank. Also, not only in the pickling line, but also in the line that conveys the steel strip in a loop shape, if the amount of looseness changes, the roll and steel strip may slip, and there is a concern that the steel strip may generate wrinkles. The amount of slack is preferably maintained constant.

  Regarding the control of the slack amount of the steel strip, a method of measuring the maximum slack amount itself, a method of measuring the height of the steel strip from the tank bottom, and the like have been tried. For example, in Patent Document 2, a camera for observing the edge of the strip material is provided on the entrance side of the annealing furnace in order to control the slack amount of the strip material loosened in a loop shape in the catenary type annealing furnace, and the edge of the strip material A method for measuring the slope is shown. However, this method cannot be applied when the imaging target is hidden by mist-like water droplets or fumes.

  In addition, as a general method for measuring the height of the steel strip from the bottom of the pickling tank, an excitation coil and a detection coil that are excited by a constant current power source are provided at the bottom of the pickling tank, and the impedance of the detection coil is changed. An electromagnetic sensor method for measuring the height of the steel strip can be mentioned. However, since the electromagnetic sensor method decreases the detection sensitivity when the coil and the steel strip are separated, manual intervention by an operator is required when the measurement range exceeds about 300 mm, and depending on the material of the steel strip, Since the characteristics are different, there is a problem in measurement accuracy and efficiency, such as the need for a calibration curve for each steel strip. Alternatively, since it is necessary to install the measuring head at the bottom of the pickling tank or in the vicinity thereof, there is a possibility that the measuring head frequently fails due to acid corrosion, and there is a problem with durability. In addition, since the electromagnetic sensor method is limited to a magnetic material, it cannot be applied to anything other than a magnetic material such as steel.

  On the other hand, in Patent Document 3, a pulsed ultrasonic wave is transmitted to the surface of the metal strip by an ultrasonic sensor installed at the bottom of the pickling tank, and the time until the ultrasonic wave is reflected and returned is shown. A method for measuring the metal strip height from the tank bottom by measuring is disclosed. Alternatively, Patent Document 4 discloses a method in which a contact roll is brought into contact with a band material in a tank from above, and a slack amount of the band material is directly measured from a mechanical displacement amount of the contact roll. Although these methods have solved the problems of the measurement range and material effect of the electromagnetic sensor, it is necessary to place the measurement head at the bottom of the pickling tank or in the vicinity of it as with the electromagnetic sensor. Remains.

JP 2004-141956 A JP 61-177328 A JP-A-7-110229 JP-A-8-192216

  The present invention has been made in view of the above problems, and edge detection capable of stably detecting the edge of a traveling member (especially a strip) even in a production line filled with mist-like water droplets or fume. Method, running that can measure the running condition of the strip in a wide measurement range with sufficient durability, even when the strip is run in the pickling tank in the continuous pickling line Provided are a situation measurement method, a travel control method for appropriately controlling the travel of the strip using these methods, a manufacturing method for manufacturing the strip using the travel control method for the strip, and these methods can be executed. The problem is to provide a system.

In order to solve the above problems, the present invention has the following configuration. That is,
According to a first aspect of the present invention, a plurality of regions including edge lines of a traveling member are imaged by an imaging unit, and each of a plurality of temporally continuous images obtained by the imaging step A differential image generation step of generating a differential image by obtaining a differential intensity, and a composite differential image generation step of generating a composite differential image by combining a plurality of temporally continuous differential images obtained by the differential image generation step If, in the synthetic differential image obtained by combining the differential image producing step, the differential intensity sum of pixels existing on a straight line to identify the straight line having the maximum and a straight line identifying step, an edge detection method.

In the present invention, the “traveling member” is a member that travels continuously in a row, and is not particularly limited as long as it has a substantially straight edge, and is not limited to a strip, plate, sheet (film). ) Materials, square bars, wire rods and the like. “Edge” refers to an end portion of a member, and particularly an end portion extending in the traveling direction of the member, in other words, an end portion in the width direction perpendicular to the traveling direction of the member. When the “traveling member” is a wire, the wire itself is regarded as an “edge”. The “edge line” refers to a substantially straight line generated by an edge when imaging the “traveling member”. For example, when a boundary between a bright part and a dark part occurs due to a brightness difference in the edge part, the boundary can be used as an edge line. “Area including an edge line” means an area to be imaged and may include a background or the like in addition to the edge line. That is, it is sufficient that the region includes at least a portion where the edge line is located. The “differential intensity of pixels in an image” refers to a value obtained by quantifying the amount of change in luminance between pixels for each pixel of a captured image. In particular, the absolute value of the luminance change amount may be the differential intensity. A “differential image” refers to an image having differential intensity information obtained for each pixel. Specifically, it can be said that the portion where the luminance change is large in the captured image is extracted as the high differential intensity, and the portion where the luminance change is small is extracted as the low differential intensity. A “synthesized differential image” is obtained by organizing differential intensity information by combining a plurality of temporally continuous differential images. For example, as will be described later, each of a plurality of differential images that are temporally continuous is coordinate-transformed, and each of the plurality of differential images that have undergone coordinate transformation is compared with the differential intensity of a pixel existing on a specific coordinate. By specifying this maximum value and performing this for predetermined coordinates, preferably all coordinates, of the differential image, it is possible to generate a composite differential image as an image in which differential intensities related to the maximum value are aggregated. Alternatively, by integrating a plurality of differential images that are temporally continuous (adding the differential intensity cumulatively, it may be the sum of squares of the differential intensity.) Alternatively, a portion where the luminance change is always small may be extracted as a low differential intensity to obtain a composite differential image. The sum of differential intensities of pixels on the composite differential image specified in the “straight line specifying step” can be specified as follows, for example. That is, by changing the intercept and slope, which are linear coefficients, and calculating the differential intensity sum of the pixels in all combinations, a desired straight line is obtained by selecting the combination that maximizes the differential intensity sum be able to. In this case, the calculation time can be shortened by using the Hough transform logic.

  In the first aspect of the present invention, the combined differential image generation step performs coordinate conversion on each of the plurality of temporally continuous differential images, and on each of the plurality of differential images that have undergone coordinate conversion, And comparing the differential intensities of the pixels existing in the image, specifying the maximum value of the differential intensity at the specific coordinates, and a maximum value specifying step, and performing the maximum value specifying step on the predetermined coordinates of the differential image It is preferable to generate a differential image. “Perform the maximum value specifying step on the predetermined coordinates of the differential image” means that the maximum value specifying step is performed on all the coordinates of the differential image, and at least on the pixel portion that is estimated to have an edge. For such coordinates, it means that the maximum value specifying step may be performed. However, from the viewpoint of more reliably performing edge detection, it is preferable to perform the maximum value specifying step for all coordinates of the differential image. In this case, a synthetic differential image obtained by collecting differential intensity information related to the maximum value is obtained.

In the first aspect of the present invention, the differential intensity sum of pixels, it determines greater or not than the threshold value, further Ru comprising a determination step. This is because it becomes possible to prevent erroneous detection of edges more appropriately.

  In 1st this invention, it is preferable to irradiate light to the edge part of the member to drive | work. This is because the edge of the member can be detected with higher accuracy.

  In the first aspect of the present invention, the member may be a band material. In this case, the strip can be a steel strip.

According to a second aspect of the present invention, for each of a plurality of temporally continuous images obtained by an imaging unit and an imaging unit that captures an area including an edge line of a traveling member, a differential intensity of a pixel in the image is obtained. generates a differential image Te, to produce a synthesized and synthesized differential image differential image obtained, in the synthetic differential image, identifies a straight line differential intensity sum of pixels existing on a straight line is maximum, the An edge detection system comprising image processing means for determining whether the sum of differential intensities of pixels is greater than a threshold value .

  In 2nd this invention, it is preferable to further provide the illumination means which irradiates light to the edge part of the member to drive | work. This is because the edge of the member can be detected with higher accuracy.

  In the second aspect of the present invention, the member is preferably a strip, and in this case, the strip can be a steel strip.

  The third aspect of the present invention is a method for measuring the traveling state of the strip material when the continuously traveling strip material is immersed in the treatment liquid in the tank, and includes the edge detection method according to the first aspect of the present invention. Detecting the edge of the strip material and identifying the angle of the edge of the strip material, from the angle specified step and the angle specified by the angle specifying step, the amount of slack of the strip material or the height of the strip material from the tank bottom A running condition measurement method for a strip, comprising a calculating step.

  In the third aspect of the present invention, in the imaging step, a plurality of areas including the edge line of the traveling strip and the angle reference means provided outside the tank are imaged, and the edge and angle of the strip detected in the angle identification step It is preferable to specify the angle of the band from the positional relationship with the reference means. This is because it becomes easier and more appropriate to specify the angle of the strip, and it is possible to more appropriately measure the traveling state of the strip.

  4th this invention is a system which measures the driving | running | working condition of the said strip, when the strip which runs continuously is immersed in the process liquid in a tank, Comprising: The edge detection system which concerns on 2nd this invention, The angle of the band material detected by the edge detection system is specified, and the slack amount of the band material or the height of the band material from the tank bottom is calculated from the angle specifying means and the angle specified by the angle specifying means. And a running condition measurement system for the strip, comprising a calculation means.

  In 4th this invention, it is preferable to further provide the angle reference | standard means used as the reference | standard of the angle of a strip | belt material outside a tank. This is because it becomes easier and more appropriate to specify the angle of the strip, and it is possible to more appropriately measure the traveling state of the strip.

  In the fourth aspect of the present invention, it is preferable that the imaging means is provided at a position not facing the processing liquid. For example, when applied to a tank in which the processing liquid is volatilized / evaporated and the volatile matter / evaporated component is rising, the volatile matter / evaporated component can be prevented from reaching the imaging unit. This is because deterioration and the like can be more appropriately prevented.

  The fifth aspect of the present invention is based on the amount of slack of the strip obtained by the running condition measuring method according to the third aspect of the present invention or the height of the strip from the tank bottom, This is a strip travel control method for controlling either the tank exit side speed or the strip tension.

  The sixth aspect of the present invention is based on the traveling state measurement system according to the fourth aspect of the present invention and the amount of slack of the strip obtained from the traveling state measurement system or the height of the strip from the tank bottom. A strip running control system comprising a control means for controlling any one of a tank entry speed, a tank exit speed, or a tension of the strip.

  7th this invention is a manufacturing method of a strip | belt material provided with the process of controlling driving | running | working of a strip | belt material using the travel control method which concerns on 5th this invention.

  The eighth aspect of the present invention is a strip manufacturing system that includes the travel control system according to the sixth aspect of the present invention, and that manufactures the strip while controlling the travel of the strip using the travel control system.

  ADVANTAGE OF THE INVENTION According to this invention, the edge detection method which can detect stably the edge of the member (especially strip | belt) to drive | work in the production line which is filled with mist-like water droplets or a fume, pickling in a continuous pickling process line Even when the belt material traveling in the tank is a target, the traveling condition measuring method capable of measuring the traveling condition of the belt material with sufficient durability and a wide measurement range, and the band using these methods It is possible to provide a travel control method for appropriately controlling the travel of the material, and a manufacturing method for manufacturing the band material using the travel control method, and a system capable of executing these methods.

It is a figure which shows roughly the form of the pickling process line of the conventional steel strip. It is a figure which shows the form of a catenary curve schematically. It is a figure which shows the relationship between strip material angle (alpha) and slack amount d, and the relationship between strip material angle (alpha) and S / 2C. It is a figure showing roughly the form of the pickling processing line concerning one embodiment of the present invention. It is a figure which expands and shows a part of pickling processing line of FIG. It is a figure which shows an example of the edge detection method which concerns on this invention. It is a figure which shows the imaged image (FIG.7 (A)), the differential image obtained from the said image (FIG.7 (B)), and the synthetic | combination differential image (FIG.7 (C)) obtained by synthesize | combining a differential image. . It is a figure which shows an example of synthetic | combination differential image generation process S3. It is the schematic for demonstrating an example of synthetic | combination differential image generation process S3. It is a figure for demonstrating the Hough transformation in straight line specific process S4. It is a figure which shows an example of the driving | running | working condition measuring method which concerns on this invention. The figure which shows the edge detection accuracy result (FIG.12 (a)) when an electromagnetic sensor is used (conventional example), and the edge detection accuracy result (FIG.12 (b)) when an imaging means is used (example). It is. When edge detection is performed using only one differential image (comparative example), an edge detection result (FIG. 13A) and a composite differential image are generated using 10 differential images, and the combined differential image is generated. It is a figure which shows the edge detection result (FIG.13 (b)) at the time of performing edge detection using an image (Example).

  Hereinafter, although the form in which the present invention is applied to the pickling line for steel strip is described, the present invention is not limited to the form. As long as the member having an edge such as a belt material, a plate material, a sheet material, or a square material conveyance line runs continuously, the present invention is not particularly limited. In particular, the above-mentioned steel strip and steel plate manufacturing and processing lines, other metal strips other than steel strip and steel plates, metal plate manufacturing and processing lines, resin film and resin sheet manufacturing lines, processing lines (embossing, laminating, printing) Etc.), paper manufacturing / processing line, non-woven fabric manufacturing / processing line, fiber / thread manufacturing / processing line, wire rod (metal wire, etc.) manufacturing / processing line, etc. Can be suitably used. The present invention is characterized in that the edge of a member can be detected appropriately even when an obstacle intermittently passes between an imaging means and an imaging target, such as a production line filled with mist-like water droplets or fumes. Have

  FIG. 1 schematically shows the form of a conventional pickling treatment line. As shown in FIG. 1, a conventional pickling treatment line 200 includes a bridle roll 202 that travels a steel strip 201, a support roll 203 that supports the steel strip 201, and a pickling tank 205 ( 205a, 205b,...) And an electromagnetic sensor 210 provided near the bottom of the pickling tank 205. In FIG. 1, the steel strip 201 is running from the right side to the left side of the paper surface by the rotation of the bridle roll 202, and the scale of the surface is gradually submerged through the pickling tanks 205a, 205b while maintaining a predetermined amount of slack. Removed. Here, the slack amount of the steel strip 201 can be specified by the steel strip angle α or the steel strip height H. In the conventional pickling line 200, the height H of the steel strip is measured by the electromagnetic sensor 210, and the rotational speed of the bridle roll 202 and the tension of the steel strip are controlled so that the height H falls within a predetermined range. Thus, the amount of slack is controlled within a predetermined range.

  However, the detection sensitivity of the electromagnetic sensor 210 decreases when the coil and the steel strip 201 are separated from each other. Therefore, when the measurement range exceeds about 300 mm, manual intervention by an operator is required. There are problems in measurement accuracy and efficiency, such as the need for a calibration curve for each steel strip because the characteristics differ depending on the material. Alternatively, since it is necessary to install the measurement head at the bottom of the pickling tank 205 or in the vicinity thereof, there is a possibility that the measurement head frequently fails due to acid corrosion, and there is a problem with respect to durability.

  In order to solve the above problems, the present inventors have conducted intensive research on a form in which the amount of slack is specified and controlled by measuring the steel strip angle α instead of the steel strip height H. The specification of the amount of slack according to the steel strip angle α will be described with reference to FIG.

  Considering the case where a steel strip having a weight w [kg / m] is stretched with a tension T [kg], the slack shape of the steel strip is a catenary curve as shown in FIG. Therefore, the shape and inclination distribution can be expressed by the following formula (1).

  Further, when the distance between the support points of the steel strip is S [m] and the catenary coefficient C = T / w [m], the slack amount d [m] and the steel strip angle α at the support point (x = S / 2). [Rad] can be expressed by the following formula (2).

  Here, in the case of S / 2C << 1, the following approximate expression (3) can be applied.

  Therefore, the steel strip angle α [rad] and the slack amount d [m] can be obtained as in the following formula (4).

  Thus, when the distance S between the support points is fixed, when the catenary coefficient C (ratio of the tension and the weight per unit length of the steel strip) is determined, the steel strip angle α and the slack amount d are uniquely determined. Can be determined. That is, the slack amount d can be obtained by measuring the steel strip angle α in the vicinity of the support point regardless of the tension and the weight per unit length of the steel strip, that is, the thickness and width of the steel strip. FIG. 3 shows the results of a trial calculation of the relationship between the steel strip angle α, S / 2C, and the amount of slack d at a distance between support points of 20 m (corresponding to the length of one pickling tank). In the case of 2000 mm or less, which is an assumed range of slackness, since S / 2C << 1, the above equation (4) is approximately established. The slack amount d and the steel strip angle α are substantially proportional to each other, and the angle variation is about 0.11 ° per 10 mm height variation. Thus, according to the embodiment in which the slack amount d is determined by measuring the steel strip angle α, there is no need to place a measuring head at the bottom of the pickling tank or in the vicinity thereof, and the measuring device breaks down due to acid corrosion. It is considered that the durability of the measuring head can be secured for a long time.

  When measuring the steel strip angle α, the inventors of the present invention preferably captures the edge of the steel strip using an imaging means to form a two-dimensional image, and specifies the steel strip angle α based on the edge position in the image. I thought. According to the image processing using the imaging means, it is considered that the steel strip angle α can be easily specified without contact, and because the steel strip angle α can be specified by remote control, it is excellent in safety. Because it was thought. However, when the edge of the steel strip is detected by image processing for the pickling line 200, the gap between the imaging means and the edge of the steel strip is intermittently caused by mist-like water droplets or fumes generated from the pickling tank 205. There were frequent cases where the desired edge detection could not be performed due to being blocked.

In view of such a problem, the present inventors have taken the image pickup means even when the image pickup means and the edge of the steel strip are intermittently blocked by an obstruction such as a mist-like water droplet or a fume. Further research was carried out on the method that can detect the edge of the steel strip properly using the. As a result of earnest research, the following findings were obtained.
(1) In order to detect the edge of a steel strip by image processing using an image picked up by the image pickup means, the image is subjected to a differentiation process in a direction substantially perpendicular to the edge line, and the differential intensity between pixels is determined. It is preferable to obtain a differential image.
(2) Focusing on the image in which fume and mist-like water droplets are generated, the edge line of the steel strip can be partially confirmed at a certain timing due to the movement of fume and mist-like water droplets with time. (In other words, a portion having a high differential strength can be confirmed at a certain timing and can be specified as an edge portion of the steel strip), and the portion having more fume and mist-like water droplets has a lower differential strength at the steel strip edge portion.
(3) By generating a plurality of differential images from a plurality of temporally continuous images and combining the plurality of differential images by a predetermined method, the steel strip edge portion is intermittently blocked by fume or mist-like water droplets. Even if it is a case, the synthetic | combination differential image extracted by making the differential intensity | strength of a steel strip edge part large can be obtained.
(4) In the obtained synthetic differential image, by specifying a straight line by a predetermined method, the straight line can be appropriately detected as an edge of the steel strip.

  The present invention has been made based on the above findings. The present invention will be described below.

1. Pickling Line FIG. 4 and FIG. 5 schematically show a pickling line 100 according to an embodiment of the present invention. FIG. 4 is a diagram schematically showing an overall image of the pickling treatment line 100, and FIG. 5 is a diagram showing an enlarged portion of the pickling treatment line 100, particularly, relating to the edge detection system. As shown in FIG. 4, the pickling treatment line 100 includes a bridle roll 102 that travels the steel strip 101, a support roll 103 that supports the steel strip 101, and a pickling tank 105 (105a, 105) having a treatment liquid 104 inside. 105b,...) And imaging means 110 provided on the steel strip entrance side of the pickling tank 105a. The imaging means 110 is connected to an image processing means (image processing PC) 111, and the image processing means 111 is connected to a control means (pickling line controller PLC) 112. The control unit 112 controls the operation of the driving device 106 such as a motor based on information from the image processing unit 111, and controls the rotation of the bridle roll 102 to control the steel strip travel. In addition, as shown in FIG. 5, in the imaging region of the imaging unit 110 (region X in FIGS. 4 and 5), an illumination unit 113 that irradiates light to the edge portion (region Z in FIG. 5) is provided. This is preferable because the edge detection accuracy is improved. Further, when it is assumed that the steel strip angle α is specified after the edge detection, the steel strip angle α can be specified with high accuracy if the angle reference means 114 serving as a reference for the angle is provided. preferable.

  In the pickling treatment line 100, the steel strip 101, the bridle roll 102, the support roll 103, the treatment liquid 104 for pickling treatment, the pickling tank 105, or the drive device 106 are not particularly limited, and are conventionally known. From what has been used for the pickling treatment line of steel strip. For example, the driving device 106 is operated to rotate the bridle roll 102, and the steel strip 101 is caused to travel while being loosened in a loop shape by the rotation, and enters the pickling tank 105. Here, the scale adhering to the surface of the steel strip 101 can be appropriately removed by using hydrochloric acid as the treatment liquid 104 for the pickling treatment.

  The imaging means 110 images an end portion that extends along the traveling direction of the steel strip 101. The obtained image may be a color image or a black and white image. From the viewpoint of facilitating processing, a black and white image is preferable. As the imaging unit 110, a known camera or the like can be applied. About the installation location of the imaging means 110, if it is a location which can image the area | region containing the said edge part of the steel strip 101, it will not specifically limit. However, it is preferable to install in the location which does not oppose the process liquid 104 of the pickling tank 105a. In this way, it is possible to suppress contact between the imaging device 110 and the volatile vapor of the processing liquid 104 generated from the inside of the pickling tank 105a, and deterioration of the imaging device 110 can be prevented. In addition, in order to prevent deterioration or failure of the imaging unit 110, the imaging unit 110 may be installed inside the housing. As the casing, for example, a dustproof BOX made of stainless steel can be applied.

  Further, in the pickling treatment line 100, the pickling tank 105 is usually arranged in series, and the treatment liquid 104 is supplied from the downstream pickling tank 105 in the steel strip conveyance direction, and the upstream pickling process is performed. It sets so that the density | concentration of the process liquid 104 may become low as the tank (the pickling tank 105a side). A cover for preventing fume scattering is provided at the upper portion of the pickling tank 105, and further, the pickling tank 105 is set to a negative pressure to suppress fume scattering. And when confirming the traveling condition of the steel strip 101, the clearance of the cover of the entrance side or exit side of the pickling tank 105 with which the steel strip 101 enters / exits is considered as a candidate of a measurement location. Here, when the entrance side and exit side of the pickling tank 105 are compared, since the steel strip 101 is pulled up from the treatment liquid 104 on the exit side, generation of fumes is severe. Further, the upstream pickling tank has less volatilization of the processing liquid 104, and the adverse effect on the imaging means 110 is also reduced. Therefore, it is preferable that the imaging means 110 is installed further on the most upstream side (pickling tank entrance side) of the pickling tank 105a that is on the most upstream side with respect to the traveling direction of the steel strip 101. In this case, the imaging means 110 measures the steel strip angle α by imaging the region X indicated by the dotted line in FIG. However, the steel strip angle α can be measured even at other locations. For example, the steel strip angle α can also be measured by imaging the region Y indicated by the dotted line in FIG.

  Furthermore, it is preferable that the imaging direction of the imaging unit 110 is set so as to be an angle larger than 0 ° with respect to the plane of the steel strip 101 (the front surface or the back surface of the steel strip). In particular, when setting the angle of the imaging means 110, in order to stably detect the edge of the steel strip that fluctuates up and down, the surface of the steel strip is reflected by the steel strip height within the steel strip height fluctuation range. It is preferable that the setting is made so that there is no mixing with when the back surface is reflected. In other words, after installing the camera above the maximum height in the fluctuation range of the steel strip height or below the minimum height, the edge part should be within the field of view in the entire fluctuation range of the steel strip height. The angle of the image pickup means 110 (camera lens angle of view or camera angle) may be set. By setting the angle of the image pickup means 110 in this way, it is possible to take an image of the end portion together with one surface of the steel strip 101, and the image pickup means 110 is installed directly beside the steel strip 101 to pick up only the end portion of the steel strip. Rather than doing this, the edge detection accuracy is improved.

  The image processing unit 111 is connected to the imaging unit 110, identifies and detects the edge of the steel strip 101 using a plurality of image data acquired by the imaging unit 110, and detects the steel from the edge position information of the steel strip 101. The band angle α is specified. Then, the slack amount d (see FIG. 2) or the steel strip height H (see FIG. 1) of the steel strip 101 is calculated from the steel strip angle α. The image processing unit 111 is not particularly limited as long as it can perform image processing on a captured image. For example, a known personal computer or the like can be applied. The installation location of the image processing unit 111 is not particularly limited as long as image data can be appropriately acquired from the imaging unit 110. Details of the image processing in the image processing unit 111 will be described later.

  Based on the slack amount d or the steel strip height H specified by the image processing means 111, the control unit 112 performs traveling control of the steel strip 101 so that the slack amount d or the height H falls within a predetermined range. . Specifically, the steel strip 101 is driven so as to have an appropriate slack amount d or height H that does not come into contact with the bottom of the pickling tank 105 while running the steel strip 101 and submerging it in the pickling solution 104. A control signal is sent to the device 106 to control the rotational speed of the bridle roll 102 and the tension of the steel strip 101. For example, if the slack amount d is too large, the steel strip 101 may come into contact with the bottom of the pickling tank 105, so the rotational speed of the bridle roll 102 is decreased and the slack amount d is decreased. On the other hand, if the slack amount d is too small, the pickling time of the steel strip 101 may be too short, so the rotational speed of the bridle roll 102 is increased and the slack amount d is increased.

  The illumination means 113 is not particularly limited as long as it can irradiate light to the edge portion of the steel strip 101. For example, a halogen illumination device, a metal halide illumination device, or an LED illumination device can be applied. About the installation location of the illumination means 113, the position which does not oppose the pickling process liquid 104 similarly to the imaging means 110 from a viewpoint of preventing deterioration is preferable. Moreover, it is preferable to install so that the illumination direction of the illumination means 113 may become an angle larger than 0 degree with respect to the plane (the surface or the back surface of the steel strip) of the steel strip 101. As for the setting of the illumination angle, similarly to the angle of the imaging means 110, the steel strip surface illumination and the back illumination surface are not mixed due to the steel strip height fluctuation. It is good to set. Thereby, a brightness difference arises between the edge part of the steel strip 101, and the one surface of the steel strip 101, Based on the said brightness difference, the boundary part of the surface and edge of the steel strip 101 can be specified easily. it can.

  The angle reference means 114 is not particularly limited as long as it is an angle reference of the steel strip angle α. For example, as shown in FIG. 5, the angle reference means 114 is provided below the steel strip 101 so as to be included in the imaging region X. It can be a provided reference piece. The angle reference means 114 is preferably fixed at a location where there is little influence of vibration such as existing equipment in order to reduce the influence of vibration caused by the traveling steel strip 101 or the like.

  Thus, in the pickling processing line 100, the image processing unit 111 performs image processing using the image acquired by the imaging unit 110, thereby detecting the edge of the steel strip 101 and detecting the steel strip from the position of the edge. The angle α is obtained, and the slack amount d of the steel strip 101 or the steel strip height H is obtained based on the angle α. Then, the steel strip 101 can be pickled appropriately by running the steel strip 101 while controlling the slack amount d or the steel strip height H to be within a predetermined range. The attached scale can be removed efficiently.

2. Edge detection method and detection system In the pickling treatment line 100, even in the treatment liquid in the most upstream pickling tank, the acid concentration (hydrochloric acid concentration) is about 3% by mass, and the temperature reaches 80 ° C. As described above, edge detection may be an obstacle. This fume countermeasure is necessary for edge detection. In this regard, in the pickling processing line 100, the imaging unit 110 and the image processing unit 111 can appropriately function as the edge detection system according to the present invention. That is, the edge detection system according to the present invention captures an image of an area including the edge line of the steel strip 101 that travels, and a plurality of temporally continuous images obtained by the imaging means 110. A differential image is generated by obtaining the differential intensity of the pixel at, and the resultant differential image is combined to generate a combined differential image. In the combined differential image, the straight line that maximizes the differential intensity sum of the pixels existing on the straight line Or an image processing unit 111 that identifies one of the pixels existing on the straight line and the straight line having the maximum number of pixels having a differential intensity equal to or greater than the threshold value. Details of the edge detection method executed in the edge detection system will be described below.

  FIG. 6 shows an example of the edge detection method (edge detection method S10) according to the present invention. As shown in FIG. 6, the edge detection method S <b> 10 is continuous in time with the imaging step S <b> 1 and the imaging step S <b> 1 obtained by imaging a plurality of regions including the edge line of the traveling steel strip 101 by the imaging unit 110. For each of the plurality of images, a differential image generation step S2 that obtains a differential image of the pixel in the image and generates a differential image, and a plurality of temporally continuous differential images obtained by the differential image generation step S2 are synthesized. In the synthetic differential image generation step S3 for generating the synthetic differential image and the synthetic differential image obtained by the synthetic differential image generation step, the straight line on which the sum of the differential intensities of the pixels existing on the straight line is maximized, or on the straight line A straight line specifying step S4 for specifying any one of the pixels having a differential intensity greater than or equal to a threshold value among the pixels existing in Differential intensity sum of pixels existing on a straight line identified in step S4 or the number of pixels, it is determined whether the larger or not than the threshold value, and a determination step S5 arbitrarily. In the edge detection method S10, in the determination step S5, the sum of the differential intensities of the pixels existing on the straight line, or the straight line in which the number of pixels having the differential intensity equal to or greater than the threshold is detected as the edge of the steel strip. To do.

2.1. Imaging step S1
The imaging step S1 is a step in which a region including the edge line of the traveling steel strip 101 is imaged by the imaging unit 110 to obtain a plurality of temporally continuous images. Here, as shown in FIG. 5, the illumination means 113 irradiates the edge portion of the steel strip 101 with light, and the imaging step S1 is performed so as to include the light irradiation portion. It is clear and preferable. Moreover, by including the angle reference means 114 in the imaging region, the steel strip angle α can be specified with higher accuracy. The plurality of images obtained in the imaging step S1 are, for example, N images shown in FIG. As can be seen from FIG. 7A, in the image obtained by the imaging step S1, when the steel strip edge line can be confirmed (images 1 and 2), the edge line cannot be confirmed due to the presence of mist-like water droplets or fumes. A case (image N). The image picked up by the image pickup means 110 is taken into the memory of the image processing means 111.

2.2. Differential image generation step S2
The differential image generation step S2 is a step of generating a differential image by obtaining the differential intensity of pixels in the image for each of a plurality of temporally continuous images obtained in the imaging step S1. Specifically, for each pixel of the captured image, the absolute value of the luminance change amount between the pixels is converted into a numerical value as a differential intensity by differentiation. The direction of the differentiation process is preferably a direction that is perpendicular to the edge line. That is, a luminance change amount in a direction perpendicular to the edge line is specified as a differential intensity, and a differential image is generated. For example, as shown in FIG. 7B, for each of the N captured images, a portion having a large luminance change in a direction perpendicular to the edge line is extracted as a high differential intensity. A portion having a small change is extracted as a low differential intensity, and N differential images are generated. The generated differential image is stored in, for example, a FIFO buffer.

  Here, the edge of the steel strip 101 is included in the imaging region in the imaging step S1 as described above. There is a luminance difference between the edge portion and the other portion (background). This is particularly noticeable when the edge portion is illuminated by the illumination means 113. That is, normally, pixels related to the edge line tend to have high luminance, and the background portion tends to have low luminance. Therefore, the luminance change amount tends to increase at the edge boundary portion in the direction perpendicular to the edge line. Therefore, in the differential image, the differential intensity in the vicinity of the edge line is large, and the differential intensity is small in the other portions (differential images 1 and 2 in FIG. 7B). However, if mist-like water droplets or fumes are present between the imaging means 110 and the edge portion of the steel strip 101 that is the imaging target, the edge line is not captured in the captured image, so the differential intensity near the edge line is also small in the differential image. (Differential image N in FIG. 7B).

  As described above, when the image obtained in the imaging step S1 is converted into the differential image, when there are mist-like water droplets or fumes, the edge can be detected (for example, differential images 1 and 2). There is a case (for example, differential image N). In the present invention, a composite differential image is generated using a plurality of differential images that are temporally continuous, and edge detection is performed using the composite differential image, whereby mist-like water droplets, fumes, and the like are present. Also, the detection of the edge of the steel strip was ensured.

2.3. Synthetic differential image generation step S3
The combined differential image generation step S3 is a step of generating a combined differential image by combining a plurality of temporally continuous differential images obtained in the differential image generation step S2. FIG. 8 shows an example of the synthetic differential image generation step S3. As shown in FIG. 8, the combined differential image generation step S3 specifies the coordinate conversion step S3-1 that performs coordinate conversion of each of a plurality of temporally continuous differential images, and the plurality of coordinate-converted differential images. A maximum value specifying step S3-2 for comparing the differential intensities of the pixels existing on the coordinates and specifying the maximum value of the differential intensities. And a synthetic | combination differential image is generable by performing the said maximum value specific process 3-2 about the predetermined coordinate of a differential image.

2.3.1. Coordinate transformation process S3-1
The coordinate conversion step S3-1 is a step of performing coordinate conversion for each differential image. For example, as shown in FIG. 9A, one differential image is converted into xy two-dimensional coordinates corresponding to the number of pixels. When the number of pixels is n × m, there are n × m coordinates (specifically, (0,0), (0,1),... (N−1, m−2), (n −1, m−1)).

2.3.2. Maximum value identification step S3-2
The maximum value specifying step S3-2 is a step of comparing the differential intensities of pixels existing on specific coordinates for each of the plurality of differential images subjected to coordinate conversion, and specifying the maximum value of the differential intensity at the specific coordinates. . For example, as shown in FIG. 9B, in 1 to N differential images, the differential intensities of pixels existing at coordinates (o, p) are compared, and the maximum value of the differential intensities can be specified. . Then, the maximum value of the differential intensity related to the coordinates (o, p) in the differential image is extracted as the differential intensity at the coordinates (o, p) of the synthesized differential image.

  In the synthetic differential image generation step S3, the maximum value specifying step S3-2 is performed for a predetermined number of coordinates or more to generate a synthetic differential image in which the maximum values of the differential intensities of a predetermined number or more are collected. Can do. Here, the “predetermined number of coordinates” means that the maximum value specifying step S3-2 is performed on coordinates included in an area that is predicted to include at least an edge line. However, from the viewpoint of more reliably detecting the edge, it is preferable to perform the maximum value specifying step S3-2 for all coordinates in the differential image.

  The composite differential image obtained in this way includes edge line information with as high a differential strength as is currently available. On the other hand, since the pixel having the highest differential intensity is extracted, there may be a case where the entire image is noisy compared to the original differential image.

  Further, in the synthetic differential image generation step S3, a synthetic differential image may be generated as follows instead of the maximum value specifying step S3-2. That is, the differential intensities of pixels at specific coordinates are integrated for a plurality of temporally continuous differential images obtained in the differential image generation step S2 (in this case, the sum of squares of the differential intensities may be used). By performing this for predetermined coordinates (preferably all coordinates), it is possible to obtain a single composite differential image in which the difference between the portion where the differential strength is always high and the portion where the differential strength is always low is increased.

2.4. Straight line identification step S4
When an edge is detected using the composite differential image obtained in the composite differential image generation step S3, (1) the edge line is a straight line even if there is a lot of noise in the composite differential image, (2) the edge By using the two points that the line portion has a higher differential intensity than the surrounding area, a stable edge line can be detected. Specifically, in the straight line specifying step S4, the straight line included in the position range and the angle range of the steel strip edge line assumed in the image is searched in order, and the differential intensity sum of the pixels located on the straight line is calculated. To go. Then, the differential intensity sum is calculated for all of the assumed straight lines, and the one straight line having the maximum differential intensity sum is specified. It can be said that this straight line is most likely to be the edge line of the steel strip 101 stochastically. That is, it is possible to appropriately detect an edge line by specifying one straight line in the straight line specifying step S4.

  Further, in the straight line specifying step S4, the straight line included in the position range and the angle range of the steel strip edge line assumed in the image is searched in order, and the differential intensity is greater than or equal to the threshold among the pixels located on the straight line. The number of pixels may be calculated. Then, the number of pixels having differential intensities equal to or greater than the threshold is calculated for all the assumed straight lines, and one straight line having the maximum number of pixels is specified. Even such a straight line can be said to be most likely the edge line of the steel strip 101. That is, it is possible to appropriately detect an edge line by specifying one straight line in the straight line specifying step S4.

  In the straight line specifying step S4, one straight line can be specified by using an image processing method such as Hough transform. In the Hough transform, it is possible to find a portion where a plurality of points are linearly arranged in an image and apply a linear expression. A feature of the Hough transform is that a straight line can be appropriately specified even when a straight line in the image is cut off in the middle or when noise exists.

The Hough transform will be further described in detail. Consider one straight line in the xy coordinate system as shown in FIG. If a perpendicular line is drawn from the origin to this straight line, its length is ρ ′, and the angle between the x axis and θ ′ is θ ′, the formula of this straight line is ρ ′ = x cos θ ′ + ysin θ ′.
It can be expressed as. That is, in the polar coordinate system, if one point (ρ ′, θ ′) is known, one straight line is determined. Here, the point (ρ ′, θ ′) is called a Hough transform of a straight line y = ax + b. Further, a collection of straight lines having different inclinations passing through the point Pi (x _i , y _i ) of the xy coordinate system is
ρ = x _i cosθ + y _i sinθ
Can be expressed as For example, when a locus of a straight line group passing through each of the three points shown in FIG. 10A is drawn on the ρ-θ plane, the result is as shown in FIG. Since these three points are on the same straight line on the xy plane, the locus of the three points drawn on the ρ-θ plane intersects at the point (ρ ′, θ ′).

  In linear fitting using ordinary Hough transform, a two-dimensional array representing the ρ-θ plane is prepared, and the measurement angle θ of the straight line to be obtained is scanned for each target point on the xy plane to obtain ρ And the count of the two-dimensional array corresponding to the coordinates (ρ, θ) is incremented by +1. After all the target points have been counted, if a place where the value in the two-dimensional array representing the ρ-θ plane is maximum is searched, (ρ, θ) corresponding to the straight line to be obtained and the straight line pass. The number of target points on the xy plane (= the maximum value obtained) can be obtained.

  Here, for example, instead of increasing the count of the two-dimensional array by +1, when the array value is increased by the differential intensity value of the target point on the xy plane and the maximum value is searched, Thus, (ρ, θ) corresponding to the straight line that maximizes the differential intensity sum of the points on the straight line can be obtained. Alternatively, as a condition for increasing the count of the two-dimensional array by +1, by setting the differential intensity to be equal to or greater than the threshold, the number of points having the differential intensity equal to or greater than the threshold among the points on the straight line within the measurement range is maximized. It is also possible to obtain (ρ, θ) corresponding to a straight line (that is, a straight line that maximizes the number of pixels having differential intensities equal to or greater than a threshold value).

2.5. Determination process S5
In the edge detection method S10 according to the present invention, the sum of the differential intensities of the pixels existing on the straight line specified in the straight line specifying step S4 or the number of pixels having the differential intensity equal to or greater than the threshold is greater than a predetermined threshold. It is preferable to include a determination step S5 for determining whether or not. That is, if the sum of differential intensities of pixels existing on the specified straight line or the number of pixels having a differential intensity equal to or greater than a threshold is smaller than a predetermined threshold (in the case of negative determination), a straight line that is not an edge line is defined as an edge line. In such a case, the edge detection is performed again (step S6b → step S7 → step S1), or the previous detection data is maintained and the edge of the steel strip 101 is located at the previous detection position. It is assumed that it exists (step S6b → step S7 → END). When performing edge detection again, step S6b may be skipped and the process may return from step S5 (negative determination) to step S1. On the other hand, if the sum of the differential intensities of the pixels existing on the specified straight line or the number of pixels having the differential intensity equal to or higher than the threshold is equal to or higher than a predetermined threshold (in the case of a positive determination), the specified straight line is an edge line. In such a case, the edge detection is performed again as necessary (step S6a → step S7 → step S1), or the specified straight line is used as an edge line, and an edge is detected at the position of the edge line. Edge detection is performed assuming that it exists (step S6a → step S7 → END). Thus, by setting the threshold value when detecting the edge of the steel strip 101, it is possible to more appropriately prevent erroneous edge detection, and it is possible to perform edge detection with higher reliability.

  As described above, according to the edge detection method S10 including the steps S1 to S4 (preferably the steps S1 to S5), the traveling member (especially a strip) of the traveling line is filled with the mist-like water droplets or the fume. Edges can be detected stably.

3. Method and system for measuring the running state of the strip material In the pickling treatment line 100, the steel strip 101 that runs continuously is used as the treatment liquid 104 in the pickling tank 105 by using the imaging means 110 and the image processing means 111. In the case of dipping, it can be suitably functioned as a system for measuring the traveling state of the steel strip 101. In particular, when a known arithmetic device such as a personal computer is used as the image processing unit 111, the image processing unit 111 is provided with an angle specifying unit for specifying the detected angle α of the steel strip 101, and the angle specifying unit. From the angle α specified by the means, the slack amount d of the steel strip or the steel strip height H from the bottom of the pickling tank 105 can be functioned as calculation means.

  In FIG. 11, an example (traveling condition measuring method S20) of the traveling condition measuring method of the strip which concerns on this invention is shown. As shown in FIG. 11, the traveling condition measuring method S20 according to the present invention specifies the edge detection method S10 and the angle of the edge of the steel strip 101 detected by the edge detection method S10, and the angle is determined as the steel strip angle. The calculation step of calculating the slack amount d of the steel strip or the steel strip height H from the bottom of the pickling tank 105 from the angle specifying step S21 and the steel strip angle α specified by the angle specifying step S21. S22. As already explained, in the pickling treatment line 100, the loop shape of the traveling steel strip 101 draws a catenary curve. That is, by detecting the edge of the steel strip 101 through the steps S1 to S4 and specifying the angle α of the steel strip 101 from the positional relationship of the edge (step S21), the slack amount d of the steel strip 101 and pickling The steel strip height H from the bottom of the tank 105 can be uniquely determined (step S22).

  In particular, as shown in FIG. 5, for example, among the locations included in the imaging region, an angle that serves as a reference for the angle of the steel strip 101 at a location outside the pickling tank 105 and below the steel strip 101. By providing the reference means 114, the angle α of the steel strip 101 can be specified with higher accuracy. As already described, in the slack range assumed in the steel strip 101 in the pickling line 100, the angle α and the slack amount d or the height H are in a linear relationship, and about 0 per 10 mm in height fluctuation. Angle variation of .11 °. On the other hand, considering that the depth of the general pickling tank 105 is 500 mm to 1000 mm, a measurement accuracy of about ± 50 mm is required, which corresponds to a minute angle variation of ± 0.5 °. . That is, if the optical axis of the imaging unit 110 is slightly deviated, the required accuracy may not be ensured. In addition, it is not easy to hold the optical axis of the imaging unit 110 without deviating over a long period at a manufacturing site with a lot of vibration. Therefore, the angle reference means 114 including a straight line that serves as an angle reference is provided in the imaging region, and the angle of the angle reference means 114 is detected by the same processing as the edge detection of the steel strip 101, whereby the optical axis of the imaging means 110 is detected. In addition, the angle measurement accuracy of the steel strip 101 can be maintained over a long period of time by correcting the angle measurement value of the steel strip 101 using the change amount.

  Thus, according to the method of calculating the running state (slack amount d or height H) of the steel strip 101 in the continuous pickling treatment line 100 using the edge detection method S10 or the edge detection system according to the present invention, Compared to the case where the height H is directly measured using a conventional electromagnetic sensor, it is possible to measure the running condition of the steel strip with sufficient durability and a wide measurement range.

4). Strip material traveling control method and control system In the pickling line 100, the traveling state (slack amount d or height H) of the steel strip 101 is measured and calculated using the imaging unit 110 and the image processing unit 111. A control signal is transmitted from the control means 112 to the driving device 106 based on the traveling state, and the rotation speed of the bridle roll 102 or the tension of the steel strip 101 is controlled by controlling the operation of the driving device 106. . That is, based on the slack amount d or height H specified by the image processing means 111, a control signal is sent from the control means 112 to the driving device 106 so that the slack amount d or height H falls within a predetermined range. The traveling control of the steel strip 101 is performed by controlling the rotational speed of the bridle roll 102. Specifically, when the slack amount d obtained by measuring the traveling state is too large (when the height H is too small), the steel strip 101 may come into contact with the bottom of the pickling tank 105. A control signal is issued from the control means 112 to the drive device 106 so as to reduce the rotational speed of the 102 and reduce the slack amount d. On the other hand, when the slack amount d is too small (when the height H is too large), the pickling treatment time of the steel strip 101 may be too short. Therefore, the rotational speed of the bridle roll 102 is increased to reduce the slack amount d. A control signal is issued from the control means 112 to the driving device 106 so as to increase.

  In the above description, the control means 112 controls the rotational speed of the bridle roll 102 via the driving device 106, but the present invention is not limited to this embodiment. The control means 112 may control the rotational speed of a bridle roll (not shown) on the exit side of the pickling tank 105. Other than that, there is no particular limitation as long as the tension of the steel strip 101 is controlled.

  Thus, by performing traveling control using the traveling state measuring method and measuring system according to the present invention, pickling treatment can be performed while appropriately controlling the traveling of the steel strip 101, and the surface of the steel strip 101 is controlled. The attached scale can be efficiently removed.

5. Manufacturing method and manufacturing system of strip material By using the traveling control method or the control system according to the present invention, pickling treatment of the strip material can be performed while appropriately controlling the traveling of the strip material (steel strip 101). It is possible to produce a high-quality strip whose scale has been removed appropriately. The production method or production system of the strip according to the present invention is not particularly limited as long as the above-described travel control method or travel control system is incorporated, and about lines other than the pickling treatment line A known line (such as a rolling line) can be applied.

(Device configuration)
As an example, as shown in FIGS. 4 and 5, the edge detection system according to the present invention was applied to a continuous pickling treatment line. Specifically, a CCD camera 110 was installed beside the entry line of the pickling tank 105, and the angle of the steel strip 101 on the exit side of the support roll 103 was measured. At that time, the steel strip edge was illuminated from the side with the halogen illumination 113 so that the steel strip edge could be clearly observed. Since the camera 110 is installed at a site where a large amount of dust and mist-like water droplets are scattered, the camera 110 is housed in a stainless steel dustproof BOX. The CCD camera 110 for imaging is a general NTSC output camera, and outputs an interlaced output of VGA size (horizontal 640 pixels, vertical 480 pixels) images at 30 frames per second.
The image processing personal computer 111 is a standard Windows (registered trademark) personal computer that uses a Core 2 Duo processor with a clock frequency of 2.4 GHz as a CPU, and images from the camera are displayed in 256 gradations (8 bits) using a built-in image capture board. Collected in memory. Image data The image data captured in the memory of the personal computer 110 is converted into the steel strip height from the tank bottom after measuring the angle of the edge line by the slack amount analysis software, and the amount of slack on the screen and above The output is made to the controller (control means) 112.

(Processing procedure)
a. Capture differential image into FIFO (First In First Out) buffer Extract image from camera 110 captured by image capture board, extract only processing area image, and perform differential process in vertical direction to generate differential image After that, N images were accumulated in a FIFO buffer capable of accumulation. Here, N is a natural number of 1 or more, but N = 20 in this embodiment. After accumulating 20 differential images, the following differential composite image generation and edge line detection processing were repeated each time a new image was accumulated in the FIFO buffer.

b. Synthetic differential image generation The differential image is coordinate-converted, and for each pixel (coordinate), the differential intensity is compared among the 20 differential images stored in the FIFO buffer, and the maximum intensity value is extracted to create a new image. Assigned to buffer. When the extraction of the maximum values of all the pixels is finished, a new differential image containing the edge line information is generated in the new image buffer with the high differential intensity as long as it can be obtained with 20 images continuously.

c. Edge Line Detection The coordinates on the straight line included in the position range and the angle range of the edge line assumed in the screen were calculated, and the differential intensity at the coordinates on the straight line was added to calculate the differential intensity sum. After calculating the differential intensity sum for all possible straight line groups, the position and angle for the straight line with the maximum differential intensity sum were determined. In this embodiment, the edge line detection is speeded up by applying the Hough transform logic. The details of the Hough transform are as described above. In this example, since the angle α of the steel strip changes in the range of 5 to 20 °, θ (= α + 90 °) is set in the range of 95 to 110 ° and scanned at a pitch of 0.25 °.

(Test results)
A dummy steel strip is attached to the pickling tub when the operation is stopped, the steel strip height H is changed by changing the tension, the steel strip height measured with a ruler from above the basin, the angle measurement result by the method of the present invention, and FIG. 12 shows the relationship between the output values of the conventional electromagnetic sensor system (two coils having different materials and dimensions). In the conventional electromagnetic sensor system, when the steel strip height exceeds 300 mm, the output is saturated and cannot be detected (FIG. 12A). On the other hand, in the image processing system of the present invention, the measurement angle and the steel strip height are in a proportional relationship up to 1000 mm, and an angle variation of 0.1 ° corresponds to a height variation of approximately 10 mm (FIG. 12B). ). This is almost the same as the above-described calculation result. Moreover, it becomes the same straight line irrespective of the width of the band or the thickness of the band.

  FIG. 13 shows a steel strip height measurement result in which a camera image during operation in which a large amount of mist-like water droplets is generated is recorded by VTR and measured by the apparatus of the present invention. The analysis was performed with N = 1 (edge detection by Hough transform of a normal differential image) and N = 20. In the case of N = 20, the measurement value could be output 10 times per second. When N = 1, many measurement errors occur due to the influence of mist-like water droplets (FIG. 13A), but when N = 20, no large measurement errors occur (FIG. 13B). ). The measurement variation is about ± 50 mm due to the end-stretched shape and vibration of the steel strip, but when applied to actual catenary control, the measurement variation can be sufficiently suppressed by performing control output with about 100 moving averages.

  From the above results, by applying the edge detection method of the present invention to the slack amount control of the continuous pickling treatment line, there is no manual intervention by the operator due to the measurement range deviation that occurred in the conventional electromagnetic sensor system, and It was found that stable running control was always possible.

  The present invention can be suitably used for a member manufacturing line and a member processing line in which members travel continuously. In particular, even in an environment where mist-like water droplets, fumes, and the like are present as in a continuous pickling treatment line, the feature is that the edge of a traveling strip is detected with high accuracy. In the continuous pickling line, by detecting the edge of the running strip with high accuracy, the running situation of the strip can be measured appropriately, and the running control of the strip can be performed stably. Become. Thereby, a pickling process can be performed appropriately and a high quality strip can be manufactured.

100 Pickling line 101 Steel strip (traveling strips and members)
102 Bridle roll 103 Support roll 104 Pickling solution 105 Pickling tank 106 Driving device 110 Imaging means 111 Image processing means (angle specifying means, calculating means)
112 Control means 113 Illumination means 114 Angle reference means

Claims (18)

An imaging step of imaging a plurality of regions including edge lines of a traveling member by an imaging means;
For each of a plurality of temporally continuous images obtained by the imaging step, a differential image generation step of obtaining a differential image by obtaining a differential intensity of a pixel in the image; and
Combining a plurality of temporally continuous differential images obtained by the differential image generation step to generate a composite differential image, and a synthetic differential image generation step;
In the composite differential image generating step the composite differential image obtained by the differential intensity sum of pixels existing on a straight line to identify the straight line having the maximum and the line identifying step,
A determination step of determining whether or not the differential intensity sum of the pixels is greater than a threshold;
An edge detection method comprising: The synthetic differential image generation step includes
A coordinate transformation step of coordinate transformation of each of the plurality of differential images continuous in time;
For each of the plurality of differential images that have undergone coordinate conversion, a differential value of a pixel existing on a specific coordinate is compared, and a maximum value specifying step of specifying a maximum value of the differential intensity at the specific coordinate;
With
The edge detection method according to claim 1, wherein the combined differential image is generated by performing the maximum value specifying step on predetermined coordinates of the differential image. Irradiating light to an edge portion of the member to travel, the edge detection method according to claim 1 or 2. The member is a strip, the edge detection method according to any one of claims 1-3. The edge detection method according to claim 4 , wherein the strip is a steel strip. Imaging means for imaging an area including an edge line of a traveling member;
For each of a plurality of temporally continuous images obtained by the imaging means, a differential image is generated by obtaining a differential intensity of a pixel in the image, and a synthesized differential image is generated by synthesizing the obtained differential images, in the synthetic differential image, identifies a straight line differential intensity sum of pixels existing on a straight line is maximum, it is determined whether the differential intensity sum of the pixel is greater than a threshold, an image processing unit,
An edge detection system comprising: The edge detection system according to claim 6 , further comprising illumination means for irradiating light to an edge portion of the traveling member. The edge detection system according to claim 6 or 7 , wherein the member is a strip. The edge detection system according to claim 8 , wherein the strip is a steel strip. A method of measuring the running status of the strip when the strip running continuously is immersed in the treatment liquid in the tank,
Detecting the edge of the strip using the edge detection method according to claim 4 or 5 , and identifying an angle of the edge of the strip;
From the angle specified by the angle specifying step, the amount of slack of the strip, or the height of the strip from the tank bottom, a calculation step;
A method for measuring a traveling state of a strip. In the imaging step, a plurality of areas including the edge line of the traveling strip and the angle reference means provided outside the tank are imaged,
The travel condition measuring method according to claim 10 , wherein in the angle specifying step, the angle of the strip is specified from a positional relationship between the detected edge of the strip and the angle reference means. A system for measuring the running status of the strip when the strip running continuously is immersed in the treatment liquid in the tank,
The edge detection system according to claim 8 or 9 ,
An angle specifying means for specifying the angle of the edge of the strip detected by the edge detection system;
From the angle specified by the angle specifying means, calculating the amount of slack of the band or the height of the band from the tank bottom;
A system for measuring the running condition of a strip. The travel condition measuring system according to claim 12 , further comprising angle reference means that serves as a reference for the angle of the strip material outside the tank. The travel condition measuring system according to claim 12 or 13 , wherein the imaging means is provided at a position not facing the processing liquid. Based on the amount of slack of the strip obtained by the traveling state measuring method according to claim 10 or 11 or the height of the strip from the tank bottom, the tank entry side speed of the strip, the tank exit speed of the strip, Or the running control method of a strip which controls either the tension of a strip. The traveling state measurement system according to any one of claims 12 to 14 ,
Based on the amount of slack of the strip obtained from the traveling state measuring system or the height of the strip from the tank bottom, either the speed of the strip entering the tank, the speed of the strip exiting the strip, or the tension of the strip Control means for controlling
A belt material traveling control system. The manufacturing method of a strip | belt material provided with the process of controlling driving | running | working of a strip | belt material using the travel control method of Claim 15 . A travel control system according to claim 16 , comprising:
A strip manufacturing system that manufactures a strip while controlling the travel of the strip using the traveling control system.