JPH01216237A - Automatic flaw detecting method for screw part - Google Patents

Abstract

PURPOSE:To easily detect flaws of the screw part by kinds with high accuracy by picking up and image of the screw part by a telecamera, processing its image signal of one picture plane as a fractionized picture element unit, and comparing it with a predetemrined flaw detection level value. CONSTITUTION:If the screw part 2 has a flaw, illumination light is scattered at the part of the flaw or forms a shadow due to the recessed part of the flaw, so the light decreases in intensity. The image signal of one picture plane from the telecamera 4 is converted by an A/D converter 5 and fractionized picture element units are stored in the storage part 7 of a storage arithmetic unit 6. The arithmetic part 8 compares the image signal of one picture plane which is stored in picture element units with the predetermined flaw detection level value to decide that the signal is 1 when its value is the flaw detection level value or less or 0 when its value is said flaw detection level value or more.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an automatic flaw inspection method for threaded parts of steel pipes, bolts, etc.

[Prior Art] For example, when the pipes are connected and used, flaws that occur in the threaded part of pipe materials such as steel pipes are caused by the presence of such flaws, which may cause problems such as oil, water, or gas passing through the pipe from the joint. This may cause leakage of circulating materials.

In addition to thread thinning and uneven thickness, the threads carved into the pipe material may have chattering defects, which are short, linear flaws at intervals on the thread surface in the axial direction of the pipe material, and chaser defects. There are also flaws that appear linearly in the circumferential direction of the screw, and there are also flaws that form a line in the circumferential direction of the screw, and there are also flaws that are called slug flaws, where the crests and bottoms of the screws are more concave than normal.

Incidentally, since flaws in the threaded portion are a cause of the above-mentioned problems, it is necessary to inspect for their presence or absence, and conventionally this has been done by visual or tactile inspection. However, individual differences among inspectors are unavoidable, and inspection errors may occur due to fatigue.

For this reason, automatic inspection of threaded parts for flaws has recently been proposed. For example, attach a strip of light to the screw shaft (
The beam is irradiated from a direction perpendicular to the axis of the tube material, and the reflection pattern at the screw part is imaged by a television camera arranged so that the direction of the horizontal scanning line matches the direction of the screw axis, and the image signal is processed.
There is a device that inspects screws by detecting changes in autocorrelation image signals while rotating screws (Japanese Patent Application Laid-Open No. 1G-250235).
issue).

Furthermore, in JP-A-61-10761, a light beam is irradiated from an acute angle direction that does not irradiate the trough of the screw, and a light beam is irradiated from an acute angle direction that does not irradiate the trough of the screw, and is perpendicular to the screw axis in a straight line area of a predetermined width along the screw axis on the illuminated surface of the screw part. It has been proposed to detect the brightness distribution in the direction and compare it with information on a predetermined brightness distribution when the threaded portion is normal to determine whether it is acceptable or not.

[Problems to be Solved by the Invention] These have certain effects, and automatic inspection of threaded portions is performed.

However, it is difficult to accurately detect the chattering defects, chaser defects, peeling defects, etc. that occur in the form of minute lines on the threaded portion, and it is necessary to consider automatic inspection methods.

The object of the present invention is to automatically inspect threaded parts of pipes, bolts, etc. for defects with high accuracy.

[Means for Solving the Problems] The present invention advantageously solves the above-mentioned problems, and the gist thereof is to irradiate a threaded portion to be inspected with a light beam, image the threaded portion with a television camera, and capture one screen of the threaded portion. The image signal is subdivided and stored in pixel units, and the signal stored in the pixel unit is compared with a predetermined flaw detection level value, and the signal of one screen divided into subdivisions and shown in pixel units after comparison is compared with the signal of one screen shown in pixel units. A method for automatically detecting flaws on a threaded portion is characterized in that flaws are detected by comparing them with a flaw pattern detection signal determined for each type of flaw.

[Operations and Examples] Hereinafter, the present invention will be described in detail with reference to the drawings and examples.

In FIG. 1, numeral 1 is a tube material, and a threaded portion 2 is provided at the end thereof. 3 is an illumination device that irradiates the threaded portion 2 with light. Further, the tube material 1 is rotated by a rotating device (not shown),
The entire circumference of the threaded portion 2 is irradiated with light.

A television camera 4 takes an image of the screw portion 2 and outputs an image signal. By the way, if a flaw occurs in the threaded portion 2, the intensity of the light will decrease because the irradiated light will be scattered at the flaw, or a shadow will be created by the concave part of the flaw. Therefore, in the image signal output from the television camera 4, the image signal shown as an electrical signal is reduced at the flawed portion. This is detected by subsequent signal processing to detect a flaw, and in the present invention, it is performed as follows.

The image signal of one screen from the television camera 4 is converted by the A/D converter 5, and is subdivided into pixel units, for example, 258×256 or 512×512 vertically and horizontally, respectively. 6, the second storage unit 7
It is stored as shown in the figure.

The calculation unit 8 compares the image signal of one screen stored in pixel units with a predetermined flaw detection level value, and in this embodiment, the flaw detection level value or less is set as 1, and the above is set as 0. do. A part of the image signal of one screen after this arithmetic processing is shown in FIG.

Next, in the calculation unit 8, in order to detect the type of flaw, the flaw pattern detection signal is compared with a flaw pattern detection signal predetermined for each type of flaw, such as chatter flaw, chaser flaw, flaw, and blistering flaw.

This will be discussed further. In the image signal of one screen after the arithmetic processing, a predetermined number of pixels marked with 1 are arranged in the X direction, for example, 3 in the example of FIG. 3, and pixels marked with 1 are arranged at intervals in the X direction. If there is, this is a chatter flaw that exists linearly in the threaded portion 2 in the axial direction of the tube material. In addition, a chaser defect is a flaw that exists linearly in the circumferential direction of the threaded part 20, but this is a flaw that is marked with a predetermined number of consecutive 1's in the X direction in the image image of one screen after the calculation processing. This is marked with 1 at intervals in the X direction. Furthermore, if a swelling defect exists, a predetermined number of consecutive pixels marked with 1 in both the X and y directions will appear in the image signal of one screen after the arithmetic processing.

As described above, each image signal of one screen exhibits a characteristic pattern depending on the type of flaw, so when the flaw pattern detection signal determined in advance for each type of flaw is compared with the image signal of one screen after calculation processing, the calculation unit 8 Defects are detected by type.

In addition, in the image signal of one screen,
The size of the flaw is determined from the number of consecutive pixels in the direction and the number of consecutive pixels in the X direction. Further, if the pixels marked with 1 exist continuously between the X directions, it is determined that there is a diagonal flaw in the threaded portion 2.

[Effects of the Invention] As described above, the flaw inspection of the threaded portion according to the present invention can easily detect each type of flaw with high accuracy.

Further, instead of setting one flaw detection level value as in the above-described embodiment, by setting a plurality of flaw detection levels, for example two, remarkable effects such as being able to detect the size of flaws can be obtained.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an embodiment of the present invention, FIG. 2 is a diagram showing a part of an image signal of one screen in an embodiment of the present invention, and FIG. 3 is a diagram showing arithmetic processing in an embodiment of the present invention. It is a figure which shows a part of image signal of one subsequent screen.

Claims (1)

[Claims] 1. A light beam is irradiated onto the threaded part to be inspected, the threaded part is imaged with a television camera, and the image signal of one screen is recorded in units of subdivided pixels, and the signals recorded in each pixel are compared with the predetermined flaws. A method for automatically detecting flaws on a threaded part, characterized in that a flaw is detected by comparing a detection level and comparing an image signal of one screen subdivided after comparison calculation with a flaw pattern signal predetermined for each type of flaw.