JPH04369412A - Shape failure detecting apparatus - Google Patents

Abstract

PURPOSE:To detect a part of improper shape generated in a band-shaped body such as a steel plate or the like highly accurately during the travel of the band- shaped body. CONSTITUTION:This apparatus adapted to detect a failure in the shape of an object by casting a laser beam is constituted of two-dimensional cross photodetecting elements 10 for detecting the reflecting beam, a circuit 15 which detects the center of gravity of luminance of the beam from the signals detected by a longitudinal array of the photodetecting elements, and a circuit 13 which calculates the number of the photodetecting elements detecting the laser beam of not smaller than a predetermined value. The apparatus obtains the evaluating result corresponding to the kind of the shape failure and the degree of harm from the characteristic amount fed from the circuits. There are provided circuits 16, 16-1 to correct errors resulting from the curve of the scanning locus of the beam caused by an optical system, so that the detecting accuracy is improved. For a lateral array of photodetecting elements, there is provided a detecting circuit 20 to detect the maximum detecting amount of light of all the elements, and the shape failure is evaluated based on the detecting output.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for detecting with high precision a defective shape occurring in a strip, such as a steel plate, while the strip is running.

0002

[Prior Art] Shape defects that occur in strip-shaped objects such as steel plates include those that extend over a certain range, such as center stretch, side stretch, and side waves, but are localized such as unevenness, folds, etc. There are some things like that. When detecting defective shapes of these strips while traveling, conventional shape detectors for detecting defective shapes over a certain range have been used, for example, in Japanese Patent Application Laid-Open No. 57-124243.
A method is to use a television camera to take a virtual image of a bar-shaped light source reflected on a strip, and automatically detect image distortion caused by shape defects from the video signal, and a method is to directly detect changes in the height of the strip using a displacement meter. etc. are well known. However, with these methods, it is extremely difficult to thoroughly inspect the entire surface of the strip, and the sensitivity to minute shape defects typified by unevenness defects is also low.

[0003] Detection of minute shape defects has traditionally been the role of so-called optical flaw detectors, but since they basically measure changes in the amount of light reflected from the object, it is difficult to detect variations in color unevenness or gloss. However, it is effective against small shape defects such as shape defects where the shape changes gradually like stretches or waves, and unevenness defects where there is no difference in gloss from the surrounding area. It is powerless to detect.

Furthermore, Japanese Patent Laid-Open No. 61-254809 discloses a method in which a laser beam is irradiated onto a material to be inspected, the reflected beam is received by light receiving elements arranged in tandem, and shape defects are detected from vertical fluctuations in the light receiving position. It has a certain effect on detecting shape defects.

[0005]

However, there are the following problems in detecting shape defects over the entire width of a test material using a single shape detection device. In other words, in the production of strips such as cold-rolled steel sheets, surface-treated steel sheets, etc., although they are usually adequately controlled through visual observation by inspectors, there are various types and forms of shape defects that occur, as mentioned above. However, with conventional equipment, it has been difficult to detect and recognize these shape defects with sufficient accuracy within the on-site production line.

[0006] Furthermore, since scanning of the laser beam onto the material to be inspected must be performed using a combination of a rotating mirror and a parabolic mirror, the reflection position on the material to be inspected is not a straight line in the width direction of the material, but a curved line. Therefore, there is a limit to detection accuracy.

The present invention overcomes these drawbacks of the prior art, and provides a device that can detect various shape defects occurring in a strip with high precision during traveling using a single shape detection device. It is something.

[0008]

[Means for Solving the Problems] The present invention provides a parallel scanning device (5,
6), and the reflected beam reflected from the test material 1 is incident,
A scanning direction light collecting device 9 that compresses the spread in the width direction, a cross-shaped light receiving element group 10 that receives the reflected beam from the scanning direction light collecting device 9, and vertically arranged light receiving elements of the cross-shaped light receiving element group 10 receive light. Circuit 1 that calculates the center of gravity of the amount of received light in the height direction
5, and a circuit 1 that compares the signals received by the light receiving elements in the column with a predetermined threshold value and counts the number of light receiving elements that exceed the threshold value.
3, the signal from the counting circuit 13 and the center of gravity calculation circuit 1
First judgment circuits 17 and 18 compare the signals from 5 with respective predetermined threshold values and output shape defects based on the comparison results.
, 19, and a circuit 20 for detecting the maximum amount of reflected beams received by the horizontal rows of light receiving elements of the cross-shaped light receiving element group 10.
and a second determination circuit 21, 22 that compares the signal from the maximum received light amount detection circuit with a predetermined threshold value and outputs a shape defect based on the comparison result. It is.

[0009]

[Operation] Defects in the shape of the specimen in the two-dimensional direction are detected by decomposing them into vertical and horizontal displacements of the reflected beam. The threshold value for defect discrimination in each direction can be determined individually according to the type and form of the shape defect, and high detection accuracy can be obtained.

0010

DESCRIPTION OF THE PREFERRED EMBODIMENTS First, the principle of the present invention will be explained with reference to FIG. In FIG. 1, reference numeral 1 denotes a belt-shaped body which is a material to be inspected and runs in the X direction. A laser device 2 irradiates a rotating mirror 5 with a laser beam 3 through a beam expander 4. The laser beam 3 reflected by the rotating mirror 5 is reflected by the parabolic mirror 6, becomes a scanning beam that scans parallel to the plate width direction, and irradiates the surface of the specimen 1 at a certain angle with respect to the specimen 1. The scanning beam 7 reflected at the reflection point 8 of the material to be inspected 1 is turned back by a scanning direction focusing device, for example, a parabolic mirror 9, and further focused in the width direction, and a large number of beams are densely arranged in a cross shape. The light enters the cross-shaped light receiving element group 10.

Here, if the reflection point 8-1 is incident on the specimen 1 at the same angle θ in the width direction, the reflection spot on the cross-shaped light receiving element group 10 will be It changes as follows depending on the inclination of the reflection point of inspection material 1 in the running direction.

When the scanning beam 7 is reflected by the normal portion as shown in FIG. 2A, the scanning beam 7 is incident on the center point A of the cross-shaped light receiving element group 10. When the scanning beam 7 is reflected due to a change in the longitudinal direction of the defective shape portion, the position where the scanning beam 7 enters the cross-shaped light receiving element group 10 is determined by the inclination of the defective portion as shown in FIGS. 2(b) and 2(c). Point B may be higher than point A, or point C may be lower than point A.

Furthermore, the laser beam 3 has a finite thickness, and if the inclination of the defective shape part is not constant at the scanning location of the specimen 1, the scanning reflected beam 7 as shown in FIG. 3(a) may not be incident on a narrow area of the cross-shaped light receiving element 10, but may be incident over a wide range B to C. In particular, in the case of minute irregularities and flaws, two or more maximum points of the amount of received light may appear as shown in FIG. 3(b).

As described above, there are characteristics caused by the displacement of the test material 1 in the running direction due to the shape defect, but in the case of a shape defect that changes gradually in the width direction, the cross-shaped light receiving element is used as shown in FIG. The light comes to be incident on the width direction light receiving elements of group 10 at positions that are swung horizontally, vertically, and vertically. If the shape is even more severe, it will fall off from the light receiving element group 10.
These can be rejected from the product as serious defects.

In this way, the inclination of the surface of the reflection point 8 is detected, but if the specimen 1 is inclined in the running direction, the direction of the reflected scanning beam 7 is 2 times the inclination of the reflection point 8.
Furthermore, since the vertical movement distance in the cross-shaped light-receiving element group 10 is proportional to the distance L between the reflection point 8 and the light-receiving element group 10, if the distance L is sufficiently separated, the slope of the defective part will change extremely sensitively. It has the characteristic of capturing

On the other hand, when the laser beam 3 scanned by the rotating mirror 5 as described above is scanned in parallel by the parabolic mirror 6, for example, the distance between the rotating mirror 5 and the parabolic mirror 6 in the width direction is Because of this difference, the locus of positional scanning of the laser beam 3 on the test material 1 is not a straight line, but a curved line corresponding to the curvature of the parabolic mirror 6. As a result, even if the object to be inspected 1 is flat, the height at which the reflected laser beam 7 from the center of the object and the reflected laser beam 7 from the edges enter the cross-shaped light receiving element group 10 is increased. It will change.

Assuming that this amount of change is ΔH, ΔH increases as the parabolic mirror 6 for parallel scanning becomes wider, and as the above-mentioned distance L increases, ΔH increases. Increasing the distance L makes it possible to more sensitively detect changes in the inclination of the defective shape portion as described above, but on the other hand, there is a drawback that the amount of change in the incident height ΔH increases. Therefore, by calculating the correction amount in the width direction from the curved shape of the parabolic mirror 6 and correcting it by a signal processing section to be described later, the degree of shape defect can be determined with high accuracy. Note that if the scanning period is made sufficiently short and the scanning pitch is made sufficiently fine depending on the traveling speed of the specimen 1, it is possible to thoroughly inspect the surface of the specimen 1.

The cross-shaped light receiving element group 10 includes a plurality of light receiving elements 11 arranged in a cross shape, and detects horizontal and vertical displacements of the scanning reflected beam 7 from the scanning direction condenser 9. The reason why the light-receiving elements 11 are arranged in a cross shape is to realize the features described above in the detection principle section with the minimum number of light-receiving elements.

Next, signal processing will be explained using FIG. 5.

The cross-shaped light-receiving element group 10 is provided with M light-receiving elements 11 in the height direction and 6N light-receiving elements in the width direction, for a total of M+6N light-receiving elements 11, and 12 represents a preamplifier circuit. A circuit 13 coupled to each output of the preamplifier 12 has an M
This circuit calculates the number of blocks of light receiving elements 11 that exceed a predetermined variable threshold for each light receiving element 11 in the height direction. This circuit 13 is hereinafter referred to as “
The number of islands is called the number of times. Figure 6 shows its function.

The "number of islands" circuit 13 has M light receiving elements 11.
The input light is photoelectrically converted, and the comparator output of the light receiving element 11 that exceeds a certain threshold value 14 is set as H(1) by a comparator provided for each light receiving element 11, and the comparator output of the light receiving element 11 that does not exceed the threshold value 14 is set as H(1). output to L(0)
Then, as shown in FIG. 6(b), by comparing each comparator output with the adjacent comparator output, for example, the comparator output is divided into a non-inverted output and an inverted output, and The number of islands, n, is determined in real time by ANDing.

A circuit 15 coupled to each output of the preamplifier 12 calculates the center of gravity g of the amount of light received in the height direction by the light receiving elements 11 in the column of the cross-shaped light receiving element group 10. The circuit 15 includes a light receiving element 11 in the height direction.
, a circuit that weights and adds the amount of light received by each light receiving element 11 according to the position of the light receiving element, and a circuit that adds the amount of light received by each light receiving element 11 in the height direction.
1, it is composed of a circuit that calculates the sum of the amount of light received by each light receiving element 11, and has a function of dividing the amount calculated from the former adding circuit by the amount calculated from the latter summation circuit. That is, if the amount of light received by each light receiving element 11 is Vi, and the number of each light receiving element is i (i=1 to M), then g=Σ(i
・Calculate and output Vi)/ΣVi.

Reference numerals 16 and 16-1 are circuits that add the above-mentioned change amount ΔH to the output of the center of gravity calculation circuit 15. for example,
If the bending amount Ey of the laser beam in the length direction of the test material 1 on the test material 1 is calculated in advance by actual measurement or simulation, the amount of change △H is determined by the arrangement of the light emitting and receiving device as shown in Fig. 7. Determined by calculation.

[0024] H=L tanθ
・・・・・・(1) Hy = (L+△Ey)ta
nθ'・・・・・・・・・・・・(2) △H=L(t
anθ' - tanθ) + △Ey tanθ... (3) However, L: Distance between the scanning laser beam reflection position and the cross-shaped light receiving element group θ: Reflection angle of the scanning reflected laser beam at the plate end θ': Center of the plate Reflection angle H of the scanning reflected laser beam on the side of the plate: Height of incidence of the scanning reflected laser beam on the plate end to the cross-shaped light receiving element group Hy: Reflection angle of the scanning reflected laser beam on the plate center side to the cross-shaped light receiving element group Incident height △H: amount of change in incident height The amount of change in incident height △H is calculated in advance by this calculation, and the amount of change in the incident height △H is calculated from the center of gravity calculation circuit 15 via a voltage generation circuit (not shown) corresponding to this amount of change △H. There is a method of correcting the output center of gravity g, but to simplify this method, use Fig. 7 (b
), the amount of change only in the center of the test material △Hcente
r is calculated and approximated by a polygonal line in the board width direction, and a separately input board width direction scanning pulse signal is counted to generate a correction value ΔH corresponding to the board width position in the ΔH calculation circuit 16, and the addition circuit 1
In 6-1, it is added to the center of gravity g and corrected. Thereby, the shape of the entire width of the specimen 1 can be detected by one detection device.

Reference numerals 17 and 18 are determination circuits, which are compared with arbitrarily predetermined threshold values by comparators. The judgment circuit 1
For example, in the case of minute unevenness flaws, 7 and 18 indicate that the unevenness flaw is "present" under the conditions that the number n of islands is 2 or more and the corrected center of gravity g' is almost unchanged from the flat part (gmin≦g'≦gmax). ”, the AND gates connected to each circuit 17 and 18 output “1”.

[0026] For example, the part judged as "present" is inputted as "1" into the shift register 19, and the part judged as "not present" is inputted as "0". The connected components of 1 in the 6×6 square matrix are counted, and if the connected component is above a certain threshold of 14, it is considered a serious shape defect, and if it is below the threshold of 14, it is considered a minor shape defect. Note that by setting a plurality of threshold values 14, a more precise degree of harm can be determined.

Reference numeral 20 denotes a maximum value detection circuit, which detects the maximum value h of the amount of light received by the 6N light receiving elements 11 arranged in the horizontal rows, that is, in the width direction, of the cross-shaped light receiving element group 10, for example, by the preamplifier 12.
It is calculated using a diode OR circuit connected to each output of If the maximum value exceeds a certain threshold h0, it is determined by the determination circuit 21 that there is a defective shape portion.

For example, the part judged as "present" is inputted as "1" into the shift register 22, the part judged as "presence" is inputted as "0", and the longitudinal direction (i' )
Connected components with a value of 1 are extracted, and locations with more than 1 connected portion are treated as serious shape defects, and locations with less than 1 connected portion are treated as minor shape defects. Note that by setting a plurality of threshold values, detailed determination can be made.

This maximum amount of received light calculation circuit 20 is effective when the shape changes gradually in the width direction, such as bending or stretching, and the shape continues for a certain length in the longitudinal direction. Other feature quantities that are effective in determining shape defects include the maximum amount of light received by the light receiving element that is receiving light, the width of the light receiving element that is receiving light, and the like. It is easy to calculate feature quantities from these, and when combined with the above feature quantities, it is possible to determine shape defects with even higher accuracy.

[0030]

[Effects of the Invention] As described above, according to the present invention, it is possible to detect defective shapes of strip-shaped test materials such as cold-rolled steel sheets, surface-treated steel sheets, etc.
Defect discrimination can be performed accurately while driving by using defect discrimination thresholds that correspond to the type and form of shape defects.

[Brief explanation of drawings]

FIG. 1 is a diagram illustrating an embodiment and principle of the present invention.

FIG. 2 is a diagram explaining the principle of the present invention.

FIG. 3 is a diagram illustrating the principle of the present invention.

FIG. 4 is a diagram explaining the principle of the present invention.

FIG. 5 is a diagram showing a signal processing block according to an embodiment of the present invention.

FIG. 6 is a diagram for explaining signal processing according to an embodiment of the present invention.

FIG. 7 is a diagram for explaining a state in which the height of incidence of the scanning reflected beam on the cross-shaped light receiving element group changes in accordance with the position in the plate width direction in one embodiment of the present invention.

[Explanation of symbols]

1 Test material 2 Laser device 3 Laser beam 4 Beam expander 5 Rotating mirror 6 Parabolic mirror 7 Scanning beam 8 Reflection point 9 Parabolic mirror 10 Cross-shaped light receiving element group 11 Photo receiving element 12 Preamplifier 13 Island number circuit 15 Center of gravity calculation circuit 16 △H calculation circuit 17, 18 Judgment circuit 19 Shift register 20 Maximum value detection circuit 21 Judgment circuit 22 Shift register

Claims (4)

[Claims] Claim 1: a parallel scanning device that scans a laser beam in the width direction of a material to be inspected; a scanning direction focusing device that enters a reflected beam reflected from the material to be inspected and compresses the spread in the width direction; A cross-shaped light-receiving element group that receives the reflected beam from the condenser, a counting circuit that calculates the center of gravity of the amount of light received in the height direction received by the vertically-column light-receiving elements of the cross-shaped light-receiving element group, and the vertically-column light-receiving elements. A circuit that compares the signal received by the light with a predetermined threshold value and calculates the number of light receiving elements exceeding the threshold value, and compares the signal from the counting circuit and the signal from the center of gravity calculation circuit with the respective predetermined threshold values; a first determination circuit that outputs a shape defect based on the comparison result; a circuit that detects the maximum amount of reflected beams received by the horizontal light receiving elements of the cross-shaped light receiving element group; A shape defect detection device comprising a second determination circuit that compares the signal with a predetermined threshold value and outputs a shape defect based on the comparison result. 2. A circuit for calculating the amount of change in incident height in the height direction due to the optical system along the width direction of the specimen, a signal from the center of gravity calculation circuit and calculation of the amount of change in the incident height. 2. The shape defect detection apparatus according to claim 1, further comprising a circuit for correcting errors caused by bending of the scanning beam along the width direction of the test material. 3. The first determination circuit includes a first comparison circuit that compares the signal from the counting circuit with a predetermined threshold, and a first comparison circuit that compares the signal from the center of gravity calculation circuit with a predetermined threshold. 2 comparison circuits, an AND circuit that ANDs the outputs of the first and second comparison circuits, and a matrix-like shift register that stores the outputs of the AND circuit in correspondence with the surface of the material to be inspected. 2. The shape defect detection device according to claim 1, further comprising a circuit for determining the size of the defective portion based on the contents stored in the shift register. 4. The second determination circuit compares the signal from the maximum received light amount detection circuit with a predetermined threshold value.
Comprising a comparison circuit and a matrix-shaped shift register that stores the output of the third comparison circuit in correspondence with the surface of the material to be inspected, and determines the size of the defective portion based on the stored contents of the shift register. 2. The shape defect detection device according to claim 1, further comprising a circuit for detecting defects in shape.