JPH11264723A - Recognizing method for flaw in semi-finished steel product - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a recognizing method for flaws in a semi-finished steel product, capable of avoiding erroneous recognition of the shape of a marking, by making it easier to take a picture of the marking that expresses the position of a flaw. SOLUTION: In this recognizing method for flaws in a semi-finished steel product, a region of a defect produced in the semi-finished steel product is marked with a marking material including a fluorescent substance, then the marked portion is shifted into a shading hood 10, and the marking material is caused to emit light by irradiating it with ultraviolet rays from an ultraviolet- irradiating lamp 11 provided in the shading hood 10. A picture of the light- emitting part is taken by a CCD camera 9 provided in the shading hood 10, and the shape of the mark is discriminated by image-processing the picture taken.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for recognizing a flaw in a billet, which facilitates the imaging of a mark for indicating the position of the flaw on the billet and avoids erroneous recognition of the shape of the marking. .

[0002]

2. Description of the Related Art A flaw existing in a billet or the like is visually inspected by a person or a method such as a color check, and the surface of the flawed steel bill is marked. Place the piece on the transport roller and move it to the CCD camera installation position.
A method and an apparatus for reading a shape of the billet, reading the shape of the marking, and performing a flaw removing operation based on the result are proposed as follows.

Japanese Patent Application Laid-Open No. Hei 8-271210 discloses a technique in which, as shown in FIG. 6, an entire upper surface of a slab 2 on which a defect mark 24 is marked is imaged by a single wide-area camera 21 and a narrow-area image is formed. A method of recognizing the marking by imaging one side surface with a camera, then inverting the slab, imaging the back surface and the other side surface with the camera, and processing each image to recognize the marking (conventional method (1)) ). In addition, 23 is a fusing cart. Japanese Patent Application Laid-Open No. 8-271234 discloses that, as shown in FIG. 5, a plurality of cameras 20 are provided along the width direction of the slab 2 and an image is taken so that a marking line extends over a plurality of pixels of the cameras 20 and 20. Method (conventional method (2)). In addition, Japanese Patent Application Laid-Open No. 8-272969 discloses, as shown in FIG.
This is a method of providing a hood 25 with illumination 26 on the camera 21 and imaging the surface of the slab 2 using the hood and illumination (conventional method (3)).

[0004]

However, in the conventional method (1), since the front and back surfaces of the slab are imaged by one camera, the distance between the slab surface and the camera is reduced in order to focus on the mark written on the slab. Longer and smaller images,
Further, since chalk (black ink) is used as the marking material, the mark written on the slab having a rough surface is apt to be faint, so that there is a problem that the image is unclear. In addition, since the mark described by the chalk is reflected by the reflected light, especially in the case of a slab having a rough surface, a part of the surface which is partially smooth due to, for example, partial care of the surface, contact with a pinch roller, or a transport roll, or the like. In some cases, the reflected light at this portion was stronger than the reflected light from the mark. For this reason, the ground pattern around the mark is mixed, and an unnecessary image is taken.

The conventional method (2) eliminates the disadvantage that the image becomes small by taking an image with a plurality of cameras arranged in the width direction of the slab, but uses a chalk (black ink). In addition, the problem that the mark written on the slab having a rough surface is easily blurred and the reflected light is taken cannot be solved because of the same reason as described above.

In the conventional method (3), since the hood is provided, the reflected light of the illumination of the factory does not enter.
In order to image the mark written on the slab, the illumination provided in the hood illuminates the surface of the slab, so that the reflected light enters the camera, and the problem that the imaging becomes unclear for the same reason as above is solved. Can not.

[0007] The present invention solves the above problems,
An object of the present invention is to provide a recognition method that facilitates imaging of a mark by marking and avoids erroneous recognition of the marking.

[0008]

SUMMARY OF THE INVENTION The present invention provides the above object,
The steel sheet was defect-marked with a marking material containing a fluorescent substance, and then the marking was moved into a light-shielding hood and provided in the light-shielding hood. Irradiating ultraviolet rays from an ultraviolet irradiation lamp to cause the marking material to emit light,
An image of the light emitting portion is taken by a CCD camera provided in a light-shielding hood, and the image is processed by image processing to achieve a marking slab recognition method for identifying marking.

"Operation" A mark for indicating the defect is written using a marking material containing a fluorescent substance, and the mark is irradiated with ultraviolet light in a light-shielding hood, and the fluorescent light of the mark is emitted. Since the substance is made to emit light to make it clear and the image is taken by a camera, various reflected light from the slab surface can be eliminated, and erroneous recognition of the mark can be avoided. Therefore, accurate control information can be given to the flaw removing device in the next step.

[0010]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a plan view for explaining the method of the present invention, and FIG. 2 is a sectional view taken along line AA of FIG. 1 is a transport table, 1a is a transport roll, 2 is a slab which is an example of a steel piece on the transport roll 1a, and this slab 2 moves in the direction of arrow C. Numeral 3 is a marking material containing a fluorescent substance on the surface, and marks marked on the slab 2 where there are flaws: ○,
Δ, Δ, Δ, Δ, Δ, Δ, etc. As a marking material containing a fluorescent substance, a fluorescent crayon, a fluorescent choke,
An adhesive tape coated with a fluorescent substance is used. Also,
Blue, vermilion, and green are used as the types of colors of the marking material.

Numeral 15 comprises a girder member 6, a pair of leg members 5, and a traveling means 7 provided with traveling wheels, rockers and the like provided below the leg members 5, and is laid along the transport table 1. This is a gate-shaped frame that runs on the rail 4 in the direction of arrow A. Reference numeral 8 denotes a grinder device for removing flaws on the slab, and is provided on the beam member 6 of the portal frame 15 so as to be movable in a direction perpendicular to the rail 4 (arrow B, that is, in the left-right direction). .

A light-shielding hood 10 is provided to extend forward of the beam member 6 and to have one end fixed to a side surface of the beam member 6. Inside the light shielding hood 10, a CCD camera 9 and an ultraviolet irradiation lamp 11 are provided. The emission wavelength of the marking material is blue: 420 nm, vermilion: 600 nm, green:
Since the wavelength is 510 nm, the use of an ultraviolet lamp having an emission wavelength of 365 nm close to these wavelengths can effectively increase the emission intensity of each color.

When the type of flaw is classified according to the shape of the marking as the CCD camera 9, a monochrome C
When using a CD camera and classifying the types of flaws by the color of the marking or by a combination of the color and shape, a color CCD camera is used. One million pixels is sufficient for the number of pixels of the CCD camera. The reason is that if the field of view of the camera is set to 3000 mm with respect to the maximum width of the slab: 2100 mm, the number of pixels in the width direction of the camera is 1000.
mm, and the line drawn by the choke can be sufficiently identified.

Reference numeral 12 denotes an identification device, which is a color type CC.
A specific color signal is extracted from the image signal of the D camera based on the specific color signal input in advance from the image signal.

Numeral 13 denotes an image processing device, which is means for binarizing the image signal of the CCD camera or the signal from the discriminating device 12, and that the binarized image has blurred or missing lines as shown in FIG. There is a means for correcting the figure to an accurate figure as shown in FIG.

Reference numeral 14 denotes a display device for displaying or printing the processing results of the image processing device 13. The above identification device 12,
The image processing device 13 and the display device 14 are provided in the cab of the portal frame 15 as shown in FIG.

Next, the method of the present invention implemented in the above-described facility will be described. "In the case of a monochrome camera": (1) A mark 3 is drawn with a marking material of a predetermined color on the surface of the slab 2 where the flaw exists, based on the flaw inspection result at the inspection site. The mark shows a small flaw: ━, medium flaw:
Large flaws surrounded by △ or △: It is better to classify each corner as 囲 む, ┓, ┏, or 囲 む.

(2) The slab 2 on which the mark 3 is drawn with the marking material is placed on the transport roll 1a, and the transport roll is driven to rotate so that the slab passes under the light shielding hood 10. During this time, ultraviolet light is radiated from the ultraviolet irradiation lamp 11 to the slab surface to emit a mark and an image is taken by a CCD camera.

(3) The image pickup signal is sent from the image processing device 13 to the display device 14, and the shape and arrangement of the mark 3 on the steel plate are displayed or printed. The position is on the steel sheet between indicia 3 on the steel sheet - click has been Y-axis in the width direction of the steel sheet from the origin, the coordinates (X i the longitudinal direction by the image processing apparatus 13 as the _X-axis, Y
_i ) is calculated and the result is displayed. These coordinates are used for positioning by the grinder device 8 for removing a flaw in the next process.

"In the case of a color camera": In this case, the shape of the mark of the above-mentioned marking depends on the size of the flaw, as in the case of the monochrome camera, ○, △, ━, ┗, ┓, ┏. ,
┛ etc. are used properly. These marks are used for blue, vermilion, and green depending on the depth of the flaw.

The steps (1) and (2) of the flaw recognition process are the same as those in the case of the monochrome type camera, and will not be described.

(3) The imaging signal is sent to the identification device 12,
Here, blue, vermilion, and green are identified. Thereafter, the image is sent from the image processing device 13 to the display device 14, and the shape, the depth and arrangement of the marks 3 on the steel plate are displayed or printed.

[0024]

Since the present invention is configured as described above, erroneous recognition of a mark due to reflection of sunlight or the like can be avoided, and accurate control information can be given to a flaw removing device in the next step.

[Brief description of the drawings]

FIG. 1 is a plan view illustrating a method of the present invention.

FIG. 2 is a view as viewed in the direction of arrows AA in FIG. 1;

FIG. 3 is a diagram showing a state in which a mark subjected to the binarization processing is blurred and missing.

FIG. 4 is a diagram showing a state after a mark having a blur and a defect has been corrected.

FIG. 5 is an explanatory diagram of a conventional method (2).

FIG. 6 is an explanatory diagram of a conventional method (1).

FIG. 7 is an explanatory diagram of a conventional method (3).

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Conveyance table 1a Conveyance roll 2 Slab 3 Mark 4 Rail 7 Traveling means 8 Grinder 9 CCD camera 10 Shielding hood 11 Ultraviolet irradiation lamp 12 Identification device 13 Image processing device 14 Display device

Claims (3)

[Claims] 1. A marking material containing a fluorescent substance is marked on a portion of a steel slab where a defect has occurred, and then the marking portion is moved into a light shielding hood. Irradiating ultraviolet rays from an ultraviolet irradiating lamp provided in the illuminator to cause the marking material to emit light, taking an image of the light emitting section with a CCD camera provided in a light shielding hood, and performing image processing on the image pickup; A method for recognizing a billet flaw, characterized by identifying a marking. 2. The method of claim 1, wherein the CCD camera is a monochrome type. 3. A method according to claim 1, wherein the CCD camera has a color chip.