JPH09304035A - Waviness measuring apparatus for thread body - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable measuring of the shape and the amound of making of a thread body in a short time without containing errors attributed to snaking by detecting a distance of the thread body with respect to a reference line extending in the longitudinal direction of the thread body to be measured by a distance detection end separated by a specified value in the longitudinal direction to reduce a specified capacity of a memory. SOLUTION: As shown in a FIG. (a), distances X11 and X21 respectively from a position Q in the longitudinal direction Y0 of a side edge of a steel plate (thread material) 1 and from a parallel reference line Ref in the direction y at a position P of Y1 down by a specified distance Py therefrom are calculated from images obtained by simultaneous photography with line sensors arranged at the positions P and Q. A difference D1 =X11 -X21 between the distances X11 and X21 gives the amount of bending (amount of section bending) between points Q and P and does not contain the amount of snaking as turbulence of the steel plate 1. Likewise, the amounts of section bendings can be calculated between points P and N, N and M and M and L. A moving value in the direction X while the steel plate 1 moves by a specified distance Py , namely, a relative difference between the distance X21 indicated by the FIG (a) and the distance X12 indicated by the FIG. (b) gives the amount of snaking of the steel plate 1.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for measuring unevenness of a strip in a direction orthogonal to its longitudinal direction or displacement in the orthogonal direction, and particularly, although not intended to be limited to this, movement Apparatus for measuring the profile of the side edge (distribution of the x position of the side edge of the steel sheet in the longitudinal direction y) and / or the amount of meandering (deviation amount in the x direction while advancing a predetermined distance Py in the y direction) of the hot rolled steel sheet inside. Regarding

[0002]

2. Description of the Related Art For example, in a hot rolling process, in order to grasp a hot rolled state or a steel sheet conveying state, the width direction shape of a continuous hot rolled steel sheet is measured during rolling or during conveyance,
Alternatively, the meandering state is measured.

As a conventional technique for measuring the shape in the width direction of a hot rolled steel sheet, for example, in the method of Japanese Patent Publication No. 5-57524, three sets of distance sensors are arranged along the longitudinal direction of the hot rolled steel sheet (strip 1). . Each time the strip 1 is conveyed by an appropriate distance, the distances to the strip 1 are measured at three points, and the bending of the strip 1 is obtained by using these distances as n-th order polynomials of the longitudinal position of the strip 1.

Further, the apparatus according to Japanese Patent Laid-Open No. 6-201348 uses a two-dimensional two-camera system for photographing each side edge portion of the steel plate 1. The image processing device 5 compares the background image when the steel plate 1 is absent and the steel plate image when the steel plate 1 is present on a pixel-by-pixel basis to obtain a steel plate image from which roll reflected light noise has been removed, and to obtain a camber amount. There is.

Further, in the apparatus according to Japanese Patent Laid-Open No. 7-19830, the line sensor 2 having a visual field in the width direction of the hot steel plate.
The expressions are arranged in series, and the shape of the steel sheet in the plate width direction is measured by using an arithmetic unit that sequentially integrates the difference amount of each sensor signal at the same time.

[0006]

[Problems to be Solved by the Invention]
Since the method of Japanese Patent Laid-Open No. 57524 requires three sensors, the equipment cost is high, and the amount of calculation is enormous. Therefore, in order to continuously use the system online, it is possible to perform real-time calculation with extremely high calculation capability. Equipment is required.

Further, in the apparatus of Japanese Patent Laid-Open No. 6-201348, since the center line is calculated from both edges of the steel plate, it is difficult to receive the plate width variation element, but with respect to the steel plate meandering amount, the amount of meandering in the two-dimensional camera visual field range. You can only cancel. Naturally, if the longitudinal field of view is expanded, the widthwise field of view is also expanded.
Since the number of light receiving elements is constant, deterioration of the resolution in the width direction cannot be avoided.

In the apparatus disclosed in Japanese Patent Application Laid-Open No. 7-19830, the equipment cost is low because the two line sensors are used. However, for example, the photographed image data of the steel plate edge portion of the two line sensors separated by a predetermined distance Py is used. , The predetermined distance Py is subdivided and read into the memory at a short pitch, and by image processing technology,
The edge position (distance from the reference line) of the steel plate is calculated at the short pitch, and the captured image data of the line sensor 2 at the same time is calculated.
Difference amount (relative difference) of the distance at the same time point obtained based on
In order to withstand continuous online measurement (reading of image data) when measuring over the entire length of the steel sheet, a huge amount of memory for the number of measurement points is required for the image data memory, In addition, the amount of shape calculation for the number of measurement points becomes enormous, and it becomes difficult to complete the shape calculation within the hot rolling pitch (vacant transport time between the preceding and following hot rolled steel sheets), so extremely fast calculation is possible. An arithmetic unit with the ability is required. Every time the photographed image data of the steel plate edge portion of the two-type line sensor is read at the short pitch, the edge position at the two-type line sensor position is calculated by the image processing technique, and the difference amount between them is calculated. In the mode of performing the processing of adding to the integrated value up to that point, real-time processing within a short pitch is required, and therefore, an arithmetic unit having extremely high-speed computing capability is required.

Furthermore, none of the above-mentioned prior arts has a function capable of simultaneously measuring the amount of meandering.

The first object of the present invention is to reduce the required capacity of the data memory and to complete the shape calculation which does not substantially include the error due to the meandering of the strip in a relatively short time, and to measure the meandering amount. The second purpose is to calculate the shape and the meandering of the striations that do not substantially include the error due to the meandering of the striations in a relatively short time using a data memory having a relatively small capacity. Is the third purpose.

[0011]

[Means for Solving the Problems]

(1) The waviness measuring device of the first invention of the present application is a first distance detecting end (3) for detecting a distance (X1i) of a strip (1) to a reference line (Ref) extending in the y direction; Second distance detection for detecting a distance (X2i) of the strip (1) to a reference line (Ref) extending in the y direction, which is arranged at a predetermined distance Py from the first distance detection end (3) in the y direction. End (4); First and second distances for each relative movement in the y direction of the predetermined distance Py between the first and second distance detection ends (3, 4) and the strip (1). Detection end (3,
4) means (15) for extracting the first information and the second information at substantially the same time point; the first distance (X1i) of the article (1) based on the extracted first and second information, and Second distance (X2
distance calculating means (11) for calculating i); and a relative difference (Di = X1i-) between the first distance (X1i) and the second distance (X2i) at substantially the same time point.
X2i), an integrated value D (= ΣX1i−ΣX2i) of
1); In addition, in order to facilitate understanding, the reference numerals of the corresponding elements or corresponding matters of the embodiments shown in the drawings and described later are added in parentheses for reference.

The waviness measuring device of the second invention of the present application is
Distance (X1) of the strip (1) to the reference line (Ref) extending in the y direction
a first distance detecting end (3) for detecting i); a strip for a reference line (Ref) extending in the y direction, which is arranged at a predetermined distance Py from the first distance detecting end (3) in the y direction. Second distance detecting end (4) for detecting the distance (X2i) of 1); first and second
The first and second distance detecting ends (3, 4) are generated at every relative movement in the y direction by the predetermined distance Py between the distance detecting ends (3, 4) and the strip (1). Means (15) for extracting the first information and the second information at substantially the same time point; the extracted first and second information
Distance calculating means (11) for calculating the first distance (X1i) and the second distance (X2i) of the strip (1) based on the information; and the first distance (X1i- at substantially the same position on the strip) A meandering amount calculation means (11) for obtaining a relative difference (X2i-X1i-1) between 1) and the second distance (X2i) is provided.

(3) One embodiment common to the first and second inventions of the present application is to detect the distance (X1i) of the strip (1) to a reference line (Ref) extending in the y direction. First distance detecting end (3); Distance (X2i) of the strip (1) with respect to a reference line (Ref) extending in the y direction, arranged at a predetermined distance Py from the first distance detecting end (3) in the y direction. Second distance detecting end for detecting
(4); For each relative movement in the y direction of the predetermined distance Py between the first and second distance detecting ends (3, 4) and the strip (1),
Means (15) for extracting the first information and the second information at the substantially same time points generated by the first and second distance detecting ends (3, 4); the strip based on the extracted first and second information Distance calculating means (11) for calculating the first distance (X1i) and the second distance (X2i) in (1); the first distance (X1i) and the second distance (X2) at substantially the same time point.
i) relative difference (Di = X1i-X2i), integrated value D (= ΣX1i-ΣX2i)
And a relative difference (X2i) between the first distance (X1i-1) and the second distance (X2i) at substantially the same position on the strip.
-X1i-1) to obtain meandering amount calculation means (11);

Now, as shown in FIG. 2A, the position Q of the side edge of the steel plate 1 in the longitudinal direction Y0 (distance from the top) and the moving direction of the steel plate 1 (from left to right). ), Y1 (= Y0 + P) downstream of Py in the y direction
The distances of the position P of y) from the reference line Ref parallel to the y direction are X11 and X21, and these are the first and second positions.
If it is calculated based on the captured images of the line sensors 3 and 4 at the same time point, the difference between them is D1 = X11−X21.
Is the bending amount between points Q / P of the distance Py (hereinafter, section bending amount), and is calculated based on the captured images of the line sensors 3 and 4 at the same time point. It does not include a certain amount of meandering.

Similarly, as shown in (b), (c), and (d) of FIG. 2, the distance (X1
2, X22), (X13, X23) and (X14, X
24) relative difference D2 = X12−X22, D3 = X13−
X23 and D4 = X14-X24 are respectively P / N
Interval, N / M interval, and M / L interval bending amount, and these do not include the meandering amount, which is a disturbance of the steel plate 1. Integrated values D1, D1 + D2, D1 + of the bending amount of these sections
D2 + D3 and D1 + D2 + D3 + D4 are at position Y0
Each point Y1 (P), Y2 (N), Y based on (point Q)
The bending amount D is 3 (M) and Y4 (L).

Since D1 = X11-X21, D2 = X12-X22, D3 = X13-X23, D4 = X14-X24, the bending amount at the point Y1 (P) with the position Y0 (point Q) as the base point: D1 = X11−X21, Bending amount of point Y2 (N): D1 + D2 = (X11 + X12) − (X21 + X
22) Bending amount of point Y3 (M): D1 + D2 + D3 = (X11 + X12 + X1
3)-(X21 + X22 + X23) Bending amount at point Y4 (L): D1 + D2 + D3 + D4 = (X11 + X
12 + X13 + X14)-(X21 + X22 + X23 + X24). When this is generalized and shown, it can be expressed as the bending amount D = ΣX1i−ΣX2i (1) at the point Yi. Here, ΣX1i is the total sum of X11 to X1i, that is, the integrated value, and ΣX2i is X2.
It is a total sum from 1 to X2i, that is, an integrated value.

By the way, the relative difference X21-X12 between X21 shown in FIG. 2A and X12 shown in FIG.
The amount of movement in the x direction while the same point Y1 (P) moves by Py, that is, the point Y1 (P) of the steel plate 1 is the meandering amount 1 while moving from the second line sensor 4 to the first line sensor 3. . Similarly, the meandering amount 2 = X22−X13, the meandering amount 3
= X23-X14 is obtained. If this is generalized and shown, the point Yi + 1 can be expressed as the meandering amount i = X2i−X1i + 1 (2) while traveling between the sensors 4/3.

In the present invention, the bending calculation means (11) is the same as the above (1).
The amount of bending D of each point Yi represented by the formula (deviation of the x position of each point Yi from the x position of the reference point Y0 in the longitudinal direction of the strip) is calculated. These are accurate because they do not include the amount of meandering that is a disturbance. The first and second distance detecting ends (3, 4) are generated each time the arrangement interval Py segment (1) moves relative to the first and second distance detecting ends (3, 4). Since the first information and the second information at substantially the same time are extracted, the amount of extracted information is small, and in the mode in which the first information and the second information are stored in the shared memory, the required memory capacity is small.
Since the shape (bending amount D) can be calculated in a short time based on the data in the memory, the shape calculation must be completed within the conveyance pitch of the strips (the idle time between the preceding and following strips). Therefore, the calculation capability of the bend calculation means (11) can be adapted even at a relatively low speed. Each time the detection signals of the first and second distance detection ends (3, 4) are read, the edge position of the strip at the position of each detection end is calculated, the difference amount between them is calculated, and the integrated value up to that is calculated. Even in the mode in which the addition processing is performed, the bending operation means (11) having a relatively low processing speed can be applied. That is, real-time processing is possible.

Further, in the present invention, the meandering amount calculating means (11) is
The meandering amount represented by the equation (2) is calculated while the strip is moving between the first and second distance detecting ends (3, 4). As a result, the meandering amount of the strip at the detection end position can be obtained at the Py pitch. This meandering amount can be obtained by a simple calculation using the data collected for the above-mentioned shape calculation as it is, so that accurate shape data and meandering amount can be obtained at the same time in a short time. The memory capacity for that is small,
The computing means (11) does not require exceptionally high-speed processing capability.

Other objects and features of the present invention will become apparent from the following description of embodiments with reference to the drawings.

[0021]

BEST MODE FOR CARRYING OUT THE INVENTION

[0022]

【Example】

-First Example- An example of the present invention applied to a hot-rolled steel sheet 1 is shown in FIG.
The hot-rolled steel sheet 1 rolled by a rolling mill (not shown) is transported on the transport table 2 in the y direction. First line sensor 3
The second line sensor 4 has a photoelectric conversion element (CCD) having a linear distribution and has a one-dimensional (straight line) visual field.
The two line sensors 3 and 4 are arranged above the transport table 2 with a distance of Py in the y direction, and their field of view is y.
It is installed in the x direction (steel plate width direction) perpendicular to the direction and so that the steel plate edge (side edge: side edge) falls within the visual field. A rotary encoder 16 is connected to one roller of the transport table 2 and generates one electric pulse (transport synchronization pulse) for every minute rotation of the roller. Is given to the read / write controller 9 of the measuring device 15 which is mainly composed of a CPU.

The read / write controller 9 counts the carrier synchronization pulses, and when the count value corresponds to the distance Py, digitally outputs the image signals for one line of each of the line sensors 3 and 4. It is converted to data and written in the image memories 7 and 8, respectively. Then, the count value of the carrier synchronization pulse is initialized (set to 0), the carrier synchronization pulse is counted, and when the count value corresponds to the distance Py, each of the line sensors 3 and 4 for one line similarly. Read the image signal. In the measuring device 15, the second line sensor 4 detects the passage of the steel plate 1 (rear end: bottom) after the first line sensor 3 detects the arrival of the steel plate 1 (tip: top) in this way. Every time the steel plate 1 advances to Py, the image signals for one line of the first line sensor 3 and the second line sensor 4 are read. When the steel plate 1 passes directly under the second line sensor 4, the measuring device 15 (read / write controller 9 thereof) communicates the image data written in the image memories 7 and 8 as described above. Controller 10
To the computer (personal computer) 11 via.

The computer 11 receives the data of the image memories 7 and 8 from the measuring device 15, and the computer 11 receives the data.
Write to the internal memory. Then, by image processing technology, the data on the memory is subjected to noise removal and smoothing (edge smoothing), and a threshold value for clearly cutting out the steel plate edge is calculated, Image data is binarized using the threshold value (with steel plate: "0" (white) / without steel plate:
"1" (black)), the number of black ("1") pixels from the reference line Ref (Fig. 2; starting point of one line, for example) to the steel plate edge ("0": white) is counted, and a constant is added to this. (Scale factor) to multiply the distance from the reference line Ref to the steel sheet edge (X11, X12, X13, ..., X1i, ... X1n / X21, X22, X23,
..., X2i, ... X2n: Figures 2 and 5) are calculated.

FIG. 3 shows the contents of the image reading / writing process of the read / write controller 9 of the measuring device 15 shown in FIG. When the measuring device 15 is powered on and a predetermined voltage is applied to the read / write controller 9, the controller is turned on.
La9 is a register, counter,
The timer and the like are cleared, and the input and output ports are set to the signal level during standby (step 1). In the following, the word "step" is omitted in parentheses, and step N
o. Write only numbers.

Next, the controller 9 is a steel plate detection circuit MD.
The steel plate detection signal given by is from "without steel plate" to "with steel plate"
Wait for switching to (2). Since the measuring device 15 is powered on, the elements in the measuring device 15 and the first
And a predetermined operating voltage is applied to the second line sensors 3 and 4, and the first and second line sensors 3 and 4 have a predetermined cycle,
The image pickup signal (line unit: serial output of analog signal, that is, video signal in each line) is repeatedly output to the A / D converters 5 and 6.

The steel plate detection circuit MD simultaneously binarizes the image pickup signals of the first and second line sensors 3 and 4 and binarizes them.
There is a steel plate image (portion) in each line (a steel plate is present in the binarized data for one line: "0" (white) is checked, and the first and second line sensors 3 , 4 each determine whether or not the steel plate is detected, and when the second line sensor 4 detects the steel plate, the first line sensor 3
Is changed from non-detection of steel plate to detection of steel plate, the controller 9
The steel plate detection signal is switched from "without steel plate" to "with steel plate". When the second line sensor 4 switches from steel plate detection to steel plate non-detection, the steel plate detection signal to the controller 9 is switched from "with steel plate" to "without steel plate". Thereby, the steel plate detection signal to the controller 9 is transmitted to the steel plate 1
After the front end (top) of the steel plate 1 is detected by the first line sensor 3, the rear end (bottom) of the steel plate 1 is detected by the second line sensor 3
It will be "with steel plate" until the end of passing, and "without steel plate" otherwise.

When the steel plate detection signal is switched from "without steel plate" to "with steel plate", the controller 9 recognizes this in step 2, and detects the first and second line sensors 3, 3.
4, the video signal for one line is converted into an A / D converter 5,
Digital conversion (conversion into image data) is carried out at 6 and writing is carried out in the image memories 7 and 8 (3). Next, Rotarian Coder 16
Starts counting up the carrier synchronization pulse (4), and waits for the count value to reach the value equivalent to Py (5). When the count value reaches the value equivalent to Py, the video signals for one line of the first and second line sensors 3 and 4 are digitally converted (converted into image data) by the A / D converters 5 and 6. , Write to the image memories 7 and 8 (3). This processing (3 to 5) is repeated while the steel plate detection signal is “steel plate present”. As a result, the photographed image data of the line sensors 3 and 4 are written in the image memories 7 and 8 at the Py pitch line by line. That is, the photographed image data of the line sensors 3 and 4 at the same time are written in the image memory 7 and the image memory 8 corresponding to the address thereof.

When the steel plate detection signal is switched from "with steel plate" to "without steel plate", the controller 9 changes this to step 6
Then, the image data in the image memories 7 and 8 is transferred to the computer 11 via the communication controller 10 (7). Upon completion of this transfer, the read / write controller 9 waits for the steel plate detection signal to change from "no steel plate" to "with steel plate" (2).

The computer 11 transmits the image data in the image memories 7 and 8 to the communication controller 10 when the image data is transferred.
In response to the communication interrupt, the computer 11 responds to this by transmitting the transferred image data to the image data tables 1 and 2 in the computer 11 (computer 11). Allocate to an area of the memory inside). FIG. 4 shows the processing in the computer 11. When the power is turned on, the computer 11 executes initialization (21) according to its own system program, and waits for the user program to be turned on. The operator attaches the floppy disk in which the shape measuring program is written to the computer and gives an instruction to read it. In response to this,
The computer 11 reads the shape measuring program into the memory in the computer 11, and when the reading is completed, first, the initial input menu screen is displayed on the CRT display 13 according to the shape measuring program in the memory (22). ).
Then, when the operator inputs the bending and meandering measurement, the computer 11 permits the communication interrupt (24, 2).
5). Then, it is judged whether or not there is another input from the operator (26), and if there is no other input, it waits for the transfer of the image data from the measuring device 15.

When a communication interrupt is received from the communication controller 10, the computer 11 interrupts the transferred image data to the image data tables 1 and 2 in the computer 11. Write, write "1" (necessary for shape calculation) to the register RAR, and return to the main routine (FIG. 4).

Then, the amount of bending and the amount of meandering are calculated based on the image data written in the image data tables 1 and 2 (28), and the calculated data is displayed in a graph format (FIG. 7) on the CRT. The data is displayed on the display 13 (29), the calculated data is written (30) in the external memory (floppy disk) mounted in the read / write device 12, and "0" is written in the register RAR.
(Standby) is written (31), and the image data is read from the measuring device 15.
Wait for data transfer.

FIG. 6 shows the contents of the calculation (28) of the bending amount and the meandering amount in FIG. Here, the computer 11 first performs noise removal on the image data of the image data tables 1 and 2. In this case, isolated points on the image data distribution (two-dimensional) are removed. The isolated point is a low-brightness background area (for example, areas X11 and X2 in FIG. 2A).
1) High brightness points (small areas) and low brightness points in a high brightness steel plate area. High brightness points are converted into low brightness data, and low brightness points are converted into high brightness data. Next, smoothing is performed. In this case, the image data for one line is binarized (primary) with a reference threshold value (primary threshold value) determined by the program, and the steel plate edge position ("1": from black to " 0 ": position for switching to white) is detected, an average value of image data within a predetermined width centering on the position (boundary) is calculated, and this is used as a secondary threshold value for one line. Image data
The data is binarized (secondary) and written in the binary data table. This is executed for each line of the image data (noise processed) of the image data tables 1 and 2, and the obtained binary data is converted into binary data tables 1 and 2. Write. And
With respect to the binary data on the binary data tables 1 and 2, the steel plate edge position (“1”: black to “0”: white switching position) is smoothed in the Y direction. . That is, filter processing (processing to round fine irregularities in the y-direction distribution of edge positions) is performed. The above is the noise removal in step 41,
This is the content of smoothing.

Next, the computer 11 next moves the steel plate edge position ("1": black to "0": white) of each line on the binary data tables 1 and 2 which have been subjected to the above-mentioned processing. Switching position: "1" from the line start end (Ref) to the switching position from "0" to "0" at X11, X21 in FIG.
It is calculated by counting the number of consecutive times and written in the distance table (1, 2; FIG. 5). This process is performed for all lines on the binary data tables 1 and 2 (42).

From the above, the distance tables 1 and 2 are shown in FIG.
The distance data X11, X21, etc. are written as shown in FIG.

Further referring to FIG. 6, the computer 11 next reads the read line address i of the distance tables 1 and 2.
Is set to 0 (43), and the integrated value registers RΣX1, RΣX
2 is cleared (44), and the bending amount D at each line (line address i) position is calculated according to the equation (1) according to the distance data of the distance tables 1 and 2 (45 to 4).
7) Then, the meandering amount is calculated according to the equation (2) (48), and the calculated bending amount D and meandering amount are written into the line address i of the calculated value table (repetition of 45 to 50). Then, when the data read out from the distance tables 1 and 2 becomes END, the arithmetic processing is ended there, and the "display calculated data" of the main routine (FIG. 4) (29). Then, the calculation data is displayed on the CRT display 13 (29). As a result, as shown in FIG. 7, the calculation data is displayed in a graph format. The graph labeled "Side Edge Profile" on Fig. 7 shows the bending amount D on the vertical axis,
The position Y in the longitudinal direction of the steel sheet 1 is shown on the horizontal axis. A graph described as “meandering amount” in FIG. 7 shows the above-mentioned meandering amount on the vertical axis and the position Y in the longitudinal direction of the steel sheet 1 on the horizontal axis. The data of the calculated value table is registered in the floppy disk (30).

-Second Embodiment- FIG. 8 shows the configuration of the second embodiment. In the above-described first embodiment (FIG. 1), when the reading of the line images at the Py pitch for all one steel plate is completed, the image data of all the lines are read at once from the measuring device 15 to the computer 11. However, in this second embodiment, the video signals of the line sensors 3 and 4 are transferred to 2 in order to shorten the image data transfer time.
Image data 7, which is converted into binary data by the digitizing circuits 5 and 6,
In order to calculate the bending amount D and the meandering amount in real time (while the steel plate 1 is directly below the sensors 3 and 4) and display them on the CRT display 13,
The light controller 9 has one each from the line sensors 3 and 4.
Immediately after reading the line image signal, the computer 11
The read data (binary data for one line each) is transferred to the computer 11, and the computer 11 calculates the amount of bend D and the amount of meandering each time the transfer is received, and displays it on the CRT display 13. I chose

In FIG. 8, the binarization circuits 5 and 6 convert the video signals coming from the line sensors 3 and 4 into binary image data. Based on these binary image data, the steel plate detection circuit MD gives a steel plate detection signal indicating presence / absence of a steel plate to the read / write controller 9 as in the case of the first embodiment.

FIG. 9 shows a measurement device 15 of the second embodiment.
The contents of the image reading / writing process of the read / write controller 9 are shown. The same processing steps as those of the first embodiment shown in FIG. Is attached. This second
In the embodiment, when one line image of each of the line sensors 3 and 4 (image data converted into binary data by the binarization circuits 5 and 6) is read and written in the image memories 7 and 8, Immediately, the image data of each line of the memories 7 and 8 is transferred to the computer 11. The other processes of the read / write controller 9 are the same as in the case of the first embodiment.

FIG. 10 shows a computer 11 of the second embodiment.
The processing inside is shown. In the second embodiment, the computer 1
When the communication interrupt is permitted (25), 1 writes 1 (the No. of the steel plate to be processed is 1) in the steel plate register (2).
5a), a line address (line No.) register i for calculating the bending amount D, and an integration register RΣX
1, RΣX2 is initialized (43a). A timer for calculating the bending amount D and the meandering amount when receiving the binary image data of each one line (total two lines) of the sensors 3 and 4 from the measuring device 15 and monitoring the change of the steel plate. Tp (Tp
Is the time limit value, and the time required to move the steel sheet for Py <Tp
<Start time of the steel sheet moving by the free space between the front and rear steel sheets). While the image is being read at the Py pitch (while one steel plate is directly below the sensors 3 and 4), Py
Since the timer Tp is restarted at the pitch, the timer Tp
p does not time over. The timer Tp time-overs before one steel plate passes directly below the sensors 3 and 4 and the next steel plate arrives just below the sensors 3 and 4. This time
Since the bar means the completion of the passage of one steel plate, in response to this, the No. The value is incremented by 1 (27b), and the line address (line No.) register i for calculating the bending amount D and the integration registers RΣX1 and RΣX2 are initialized (43b) in preparation for the next steel sheet to arrive.

FIG. 11 shows the details of the calculation (28) of the bending amount D and the meandering amount in the second embodiment.

Since the calculation (28) of the second embodiment is executed only for the image data of each one line (total of two lines), the calculation (28) of the first embodiment shown in FIG. 6 is performed. , Steps 43, 44 and 51 are deleted. Further, since the binary image data is given to the computer 11, noise removal (1
Dimensional isolated point removal) is performed and smoothing (filtering in the y direction) is not performed.

The other processing of the computer 11 of the second embodiment is the same as that of the second embodiment shown in FIGS. 4 and 6. In addition, in the second embodiment, the display of the CRT 13 indicates the steel plate No. Is initialized when the display is switched (the display of the measurement data of the previous steel plate can be erased in preparation for the display of the next steel plate). Then, when one steel plate arrives just below the sensors 3 and 4 and the measured value is obtained, the measured value is displayed from the starting point of the horizontal axis,
While the steel sheet is present, as the measurement progresses, the newly obtained measurement data is sequentially added and displayed on the CRT 13.

The above-described embodiment of the present invention was applied to the rough rolling mill outlet side of a continuous hot rolling mill. The steel sheet 1 exiting from the rough rolling mill moves in the y direction at a transport speed of 4 m / sec. The steel plate (bar)
Has a surface temperature of about 800 ° C. to 1000 ° C. Therefore, the surface is emitting light with high brightness. The line sensor 3 and the line sensor 4 are CCDs having a resolution of 2048 pixels / line. Line sensors 3 and 4 are
The field of view is directed in the steel sheet width direction (x direction), and the steel sheet edge can be imaged in the field of view so that the steel sheet transport direction (y
(Direction), with an interval of Py = 1 m. Since the total length of the rolled steel plate 1 is 50 to 80 m, the number of sampling points is 50 to 80, and the image signal reading cycle of the line sensors 3 and 4 for each movement of Py of the steel plate 1 is 4H.
z (4 times / sec). It has become possible to monitor abnormal materials due to rolling defects in the hot rolling process or abnormal steel sheet passing during rolling.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a first exemplary embodiment of the present invention.

FIG. 2A is a plan view showing a state in which a steel plate moving in a meandering direction is measured by line sensors 3 and 4, and FIG. 2B is a measurement state when the strip-shaped steel plate advances by a line sensor interval Py. (C) shows the measurement state when progressing by 2Py, and (d) shows the measurement state when progressing by 3Py,
It is a top view which respectively shows.

3 is a flow chart showing the processing contents of the read / write controller 9 of the measuring device 15 shown in FIG.

FIG. 4 is a flowchart showing the processing contents of the computer 11 shown in FIG.

FIG. 5 is a distance data obtained by “bending and meandering calculation based on transferred image data” (28) shown in FIG.
6 is a chart showing data classification of data, bending amount D, and meandering amount.

FIG. 6 is a flowchart showing the contents of “Bending and meandering calculation based on transferred image data” (28) shown in FIG.

7 is a plan view showing a screen displayed on the CRT display 13 shown in FIG. 1. FIG.

FIG. 8 is a block diagram illustrating a configuration of a second exemplary embodiment of the present invention.

9 is a flowchart showing the processing contents of the read / write controller 9 of the measuring device 15 shown in FIG.

10 is a flowchart showing the processing contents of the computer 11 shown in FIG.

FIG. 11 is a flow chart showing the content of “Bending and meandering calculation based on transferred image data” (28) shown in FIG.
It is a chart.

[Explanation of symbols]

1: Steel plate 2: Transport table
Blue 3: Line sensor (downstream side) 4: Line sensor (upstream side) 5, 6: A / D converter / binarization circuit 7, 8: Image memory 9: Read /
Light controller 10: Communication controller 11: Computer (personal computer) 12: External memory reading / writing device 13: CRT
Display 14: Keyboard 15: Measuring device 16: Rotarian encoder MD: Steel plate detection circuit X11, X12, X13, X14: Detection distance X21, X22, X23, X24 based on the downstream line sensor 3 Detection distance based on upstream line sensor 4

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G06T 7/60 G06F 15/70 350Z

Claims (3)

[Claims] 1. A first distance detecting end for detecting a distance of a strip with respect to a reference line extending in the y direction; a reference extending in the y direction, which is arranged at a predetermined distance Py from the first distance detecting end in the y direction. A second distance detecting end for detecting the distance of the strip to the line; for each relative movement in the y direction of the predetermined distance Py between the first and second distance detecting ends and the strip, the first And means for extracting the first information and the second information at substantially the same time point generated by the second distance detecting end; calculating the first distance and the second distance of the striation based on the extracted first and second information A waviness measuring device for a strip body, comprising: a distance calculation means for performing a calculation; and a bending calculation means for obtaining an integrated value D of a relative difference between the first distance and the second distance at substantially the same time point. 2. A first distance detecting end for detecting a distance of a strip with respect to a reference line extending in the y direction; a reference extending in the y direction, which is arranged at a predetermined distance Py in the y direction from the first distance detecting end. A second distance detecting end for detecting the distance of the strip to the line; for each relative movement in the y direction of the predetermined distance Py between the first and second distance detecting ends and the strip, the first And means for extracting the first information and the second information at substantially the same time point generated by the second distance detecting end; calculating the first distance and the second distance of the striation based on the extracted first and second information And a meandering amount calculating means for obtaining the relative difference between the first distance and the second distance at substantially the same position on the strip, the strip waviness measuring device. 3. A first distance detecting end for detecting a distance of the strip with respect to a reference line extending in the y direction; a reference extending in the y direction, which is arranged at a predetermined distance Py in the y direction from the first distance detecting end. A second distance detecting end for detecting the distance of the strip to the line; for each relative movement in the y direction of the predetermined distance Py between the first and second distance detecting ends and the strip, the first And means for extracting the first information and the second information at substantially the same time point generated by the second distance detecting end; calculating the first distance and the second distance of the striation based on the extracted first and second information Bending calculation means for obtaining an integrated value D of the relative difference between the first distance and the second distance at substantially the same time point; and the first distance and the second distance at substantially the same position on the article. And a meandering amount calculating means for obtaining a relative difference with