JPH10148622A - Method and device for automatically judging accept/ reject of welded part radiation transmission test image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To achieve an automatic judgment with high accuracy, reliability, stability, and efficiency by judging accept/reject only when all photographing conditions are acceptable and determining that the judgment is impossible when any photographing condition is not acceptable. SOLUTION: Defect information detected by performing a predetermined image processing to the image information of a radiation film image that is taken by a radiation film image pick-up device 78 is separated into shooting state information and pure defect information, and is stored at a shooting state information database(DB) and a pure defect information DB. When the preparation for the accept/reject automatic judgment is completed, a shooting condition criterion corresponding to the manufacturing standard of a lot and the shooting state information of a lot filed into the shooting state information DB out of the conformity criterion filed into the reference/specifications DB are compared, thus judging conformity for all shooting conditions. Then, only when all photographing conditions are satisfied, a defect accept/ reject judgment is performed by a specific procedure. On the other hand, when any photographing condition is not satisfied it is treated as automatic defect judgement- impossible condition.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for automatically determining the acceptability of a radiation transmission test image of a welded portion, and more particularly to a method and apparatus suitable for evaluating the quality of a welded portion of a metal material such as a steel plate or a steel pipe. , A radiographic test image of a welding portion on a photographic film or an image intensifier is captured by an image input means (imaging device), and the image information captured by the image input means is subjected to predetermined image processing to obtain a defect. The present invention relates to an automatic pass / fail determination method and apparatus for a welded part radiographic transmission test image that detects information, compares the defect information with a predetermined pass / fail determination criterion, and automatically determines whether the welded part is acceptable.

[0002]

2. Description of the Related Art Generally, in order to evaluate the quality of a welded portion of a metal material such as a steel plate or a steel pipe, a radiation transmission test is performed, and X-ray or γ
Photographic film obtained by radiation transmission test of wire
An inspector visually observes with a film observer called Schaukasten (German), examines the presence / absence, type, size, position, etc. of the defect, and determines the defect from the grayscale image appearing in the radiographic image. .

[0003] However, such a visual inspection by an inspector is usually performed in a dark room, and since a high-density film is observed with a bright bright Schaukasten, the inspector's eyes become tired due to repeated work. , Eyesight decreases during work. In addition, while the size and interval of the defects need to be measured in units of 0.1 mm, the visual measurement has poor measurement accuracy. In addition, the visual judgment work is performed while the inspector changes, but in order to judge a defect from a grayscale image, the inspector's experience and intuition are required. Has a problem of low reliability.

To solve such a problem, Japanese Patent Laid-Open No. 60-25023 discloses a method of automatically determining the presence, type, size, position, and the like of a defect from a radiographic image.
6 is proposed. In this method, a transmitted light transmitted through a radiographic photograph is taken by a television camera for image input, an image signal obtained by the television camera is input to an image processing device, and a predetermined image processing is performed to perform defect processing. The automatic determination is performed.

[0005] In addition, the reliability and stability of this automatic discrimination method,
As an elemental technology for the purpose of improving and improving the reliability and the like, Japanese Patent Application Laid-Open No. Hei 3-209583 discloses a three-dimensional defect image such as a blowhole or slag entrainment, and a relatively strong contrast.
Enhancement extraction of a first-type defect image that is thick in the direction of irradiation of radiation such as X-rays, and a surface that is long along the weld line direction and has relatively low contrast, poor fusion, etc. An automatic extraction processing method of a welding defect image has been proposed in which the extraction extraction of the second type of defect image is simultaneously performed in parallel. In Japanese Patent Application Laid-Open No. Hei 3-209582, a predetermined number of defect estimation rules are prepared as a knowledge base for each type of welding defect, and data of the defect feature amount of the welding defect is processed and collated with each defect estimation rule. There has been proposed a method of estimating the type of a welding defect in which a certainty factor corresponding to processing data satisfying the defect estimation rule is assigned, and the processing of the welding defect is stochastically estimated based on the certainty factor. Further, JP-A-3-2
In 10412, when reproducing the inspected portion on the display, data for canceling the density change based on the known thickness distribution of the inspection object is superimposed, thereby maintaining the density difference between the defective portion and the background, and A defect discrimination display method of a radiation transmission test in which the density is almost constant has been proposed.

[0006]

On the other hand, a UO which is manufactured by U-working an actual metal material weld, for example, a steel plate 10 as shown in FIG.
In a radiation transmission test using a radiographic film for a transmission imaging range 14 as shown in FIG. 2 of a tube end welded portion of the E-tube 12, as shown in Table 1, for example, as shown in Table 1, steel pipe production standards (JIS, API, ASTM) , Etc.), the pass / fail judgment criterion to be applied is different, and additional specifications of the customer are generally added. In FIG. 1, 12
M is a base material part, 12B is a bead part formed by welding, in FIG. 2, 12E is a pipe end, 12K is a pipe end groove processing part, and an image of an X-ray image corresponding to the transmission imaging range. Is shown in FIG.

[0007]

[Table 1]

[0008] In the pass / fail judgment criteria applied to each steel pipe manufacturing standard as described above, the items to be measured / judged are different for each standard. Items indicating the appropriateness of the shooting conditions of the shot image, such as the discrimination type, the discrimination, the necessity of the tone meter, the same density difference, the film density range, and the like, as shown in FIG.
2. Pure defect information such as the possibility of fusion failure 24, the possibility of large density defects, the distance between adjacent defects, the defect area, the defect area ratio, the maximum allowable size of individual spherical defects, the total length of elongated defects, and the maximum off-seam Are mixed.

In FIG. 4, numeral 20 indicates a crack, and numeral 22 indicates a penetration defect (I) which is a linear defect substantially parallel to the center line of the bead portion caused by a state where a groove remains.
P) and 24 are fusion defects (LF) which are linear defects which are also substantially parallel to the bead center line and which are caused by unwelding of the tack welded portion, etc., 26 are blowholes which are spherical defects, and 28 are blowholes which are spherical defects. Slag entrainment, which is a linear defect whose direction is indeterminate, 30 is an off-seam, which is a defect caused by displacement between an inner bead and an outer bead, and 32 is another imperfection, such as a pipe, tungsten entrainment, and undercut.

[0010] Further, the pure defect information includes individual defect information such as the type, size or length, position, and defect concentration of each individual defect, the interrelationship of a plurality of defects, that is, the total defect length, the defect interval. Group defect information such as the number of group defects, total area, total number of defects, defect area ratio, etc .; total constraining defects such as the distance from the pipe end of the defect closest to the pipe end in the longitudinal direction of the weld line, and the amount of weld seam off seam It contains information, etc., and is complex and diverse.

For this reason, in spite of the visual inspection decision made by the inspector as described above, for example, a method of automatically determining whether or not a radiographic transmission test image of a UOE steel pipe end is welded has not been put to practical use. That is the fact.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned conventional problems, and has a high inspection accuracy, reliability, and stability, and has a high inspection efficiency. The task is to provide

[0013]

According to the present invention, a radiation transmission test image of a welded portion is captured by image input means, and the captured image information is subjected to predetermined image processing to detect defect information. In the automatic pass / fail judgment method of the weld portion radiation transmission test image, which automatically determines the pass / fail of the weld by comparing the defect information with a predetermined pass / fail judgment criterion,
The pass / fail criterion is classified into a photographing condition criterion and a defect pass / fail criterion, and the defect information is classified into photographing state information and pure defect information, respectively.First, the photographing state information is compared with the photographing condition criterion. Only when all of the photographing conditions pass, the pure defect information is compared with the defect pass / fail decision criterion to make a defect pass / fail decision. If any of the photographing conditions is rejected, the defect pass / fail decision cannot be made. By doing
This has solved the above-mentioned problem.

Further, the photographing state information includes film density,
The transmissometer discrimination, including at least one of a tone meter density difference, wherein the shooting condition determination criterion is a film density rule, a transmittance meter rule, and a tone meter density difference rule. It is meant to be included.

[0015] In addition, the pass / fail judgment criterion determined in advance for each standard / criterion is stored in a storage device, and the standard / criterion applied to the lot is designated and read out through input means. On the other hand, in the case of a standard / standard not stored in the storage device, or when there is an additional specification, the pass / fail judgment criterion is manually input to the storage device via input means.

In addition, the defect pass / fail judgment is performed by determining individual defect information such as the type, size or length, position, and defect density of an individual defect by specifying a defect factor (type), an individual defect length, and a defect density. Individual defect acceptance / rejection judgment which determines pass / fail by comparing with individual defect provisions including, etc., and the interrelationship of a plurality of defects, that is, total defect length, defect interval, number of group defects, total area, total defect points, defects Group defect information such as area defect ratio, total defect length specification, defect interval specification, cluster defect specification, total defect point specification, defect area ratio specification, etc. , Determine the pass / fail by comparing the distance from the material edge of the defect closest to the material edge in the longitudinal direction of the weld line, the amount of welding seam off-seam, etc. to the overall constraint rules such as the material edge no defect definition, off-seam definition, etc. The overall constrained defect pass / fail At least one, and when at least one or more rejections are obtained by the above pass / fail judgment, a total rejection is caused by a rejection factor that is the innermost side from the end in the longitudinal direction of the weld line among these. In the case where a pass judgment is made in any of the above pass / fail judgments, a final pass / fail judgment is made to be a comprehensive pass.

Further, in the automatic pass / fail judgment apparatus for the radiographic transmission test image of the welded portion, the image input means for capturing the radiographic test image of the welded portion and the image information taken in by the image input means are determined in advance. Image processing means for performing image processing to detect defect information; criterion storage means for classifying and storing predetermined pass / fail criterion into photographing condition criterion rules and defect pass / fail criterion; Classifying into the state information and the pure defect information, first, comparing the photographing state information with the photographing condition criterion, and only when all the photographing conditions pass, comparing the pure defect information with the defect pass / fail criterion. This problem has been solved by providing a defect acceptance / rejection determination and determining means for failing the defect acceptance / rejection determination if any of the photographing conditions is unsuccessful.

[0018]

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

FIG. 5 schematically shows an example of an image of the developed radiographic film 40 in the pipe end radiation transmission test of the UOE steel pipe, which is photographed by the method shown in FIG. The reference numerals in FIG. 5 are the same as those in FIGS. 3 and 4, and are further used to determine the minimum recognizable wire diameter from the film number 42 and, for example, the number of recognizable wires.
An image 44 of a wire-type penetrometer in which a plurality of wires having different wire diameters are stretched is captured.

FIG. 6 shows that a radiation (eg, X-ray) transmission test is automatically performed on a pipe end welded portion of a UOE steel pipe being conveyed on a steel pipe conveying device, and a transmission photographic film is automatically developed. This is a system diagram showing a series of device configurations from capturing the radiographic film image with an image capturing device such as a television camera, taking the image signal into an image processing device, and automatically performing a pass / fail determination. is there.

In FIG. 6, reference numeral 50 denotes a transfer device for the UOE steel pipe 12; 52, a sequencer for controlling the steel pipe transfer device 50; 54, a radiographic photographing device; Sequencer to do. Reference numeral 60 denotes a radiation imaging room, which is a cassette automatic setting device 62 for automatically setting a cassette loaded with a film, and the sequencers 52, 5
6, the automatic cassette setting device 62
, A personal computer (personal computer) 66 for transmitting and receiving information, and a display terminal device 68 for displaying the order of photographing. Reference numeral 70 denotes an automatic film developing chamber, which includes an automatic film setting device 72, an automatic developing device 74, a post-developing film processing device 76, a radiation (photo) film imaging device 78 including, for example, an industrial television camera, and a determined film processing device. Device 80, the automatic film setting device 7
2 and a sequencer 82 for controlling the determined film processing apparatus 80, an image processing apparatus 84 for processing image signals output from the radiographic film imaging apparatus 78, for example, a film number reading apparatus 86 composed of a personal computer, for example, an engineering work A pass / fail determination device including a station (EWS) is provided. Reference numeral 90 denotes a process for general management, such as giving shooting instruction information to the personal computer 66 and displaying a tracking state on the display terminal device 68 in accordance with a tracking signal or the like input from the sequencer 52 for controlling the steel pipe conveyance. The computer (P / C) 92 is a criterion input terminal device for inputting a criterion from the office 94.

Of these, a cassette loaded with a film is transported from the dark room of the automatic film developing room 70 to the radiation imaging room 60, and the pipe number and the film number are automatically set by the automatic cassette setting device 62. It is sent to the radiographic photographing apparatus 54, and as shown in FIG. 2, the radiographic imaging is performed on the pipe end welding portion including the pipe end portion 12E and the pipe end groove portion 12K of the UOE steel pipe 12, and the photographing is completed. Until the film 40 is automatically returned to the automatic film developing chamber 70, it is already in practical use.

In the automatic film developing room 70, an unphotographed radiation film is taken out from the dark room by the automatic film setting device 72, the film is loaded in a cassette, and supplied to a cassette carrying device (not shown).
The cassette in which the photographed film is loaded and the collected film is opened, and the undeveloped film is inserted into the automatic developing machine 74. After the film is discharged from the automatic developing machine 74, the post-development film processing device 76 controls the radiographic film imaging device 78.
Supply the film to. After the radiographic film imaging device 78 completes capturing a film image, the determined film processing device 80 classifies the images and transfers them to a storage box (not shown).

As software processing, film image information photographed by the radiographic film imaging device 78 is transmitted to the image processing device 84, and defect information for determination is created, and then transmitted to the pass / fail determination device 88. Here, an automatic determination is performed by comparing with a corresponding determination criterion.
The input of standards and standards (specifications) for the determination is performed by an input terminal device 92 installed separately, and is sent to the pass / fail determination device 88 and stored.

Further, the image processing device 84 cuts out the image of the film number 42 imprinted on a predetermined position (for example, a marker portion) of the radiographic film 40 as shown in FIG. Send. This film number reading device 86 reads the film number 42 by character recognition processing.

Referring now to FIG. 7, the procedure of the pass / fail automatic determination process according to the present invention will be described.

In FIG. 7, reference numeral 100 denotes a reference / specification database (DB); 102, a photographing state information database (DB) for storing photographing state information among defect information;
Reference numeral 4 denotes a pure defect information database (DB) for storing pure defect information, and reference numeral 106 denotes a judgment result database (DB) for storing photographing condition judgment results and pass / fail judgment results.
These are all included in the pass / fail determination device 88.

Prior to the defect acceptance / rejection judgment, the acceptance / rejection judgment criteria predetermined for each manufacturing standard are stored in a storage device, and the specification / standard information applied to the lot is designated via the judgment standard input terminal device 92. Then, a pass / fail criterion corresponding to a manufacturing standard / criterion applied to the lot is manually input in advance through the criterion input terminal device 92 and stored in the criterion / specification DB 100.

In processing, first, the image processing device 84
In the image processing program, the radiographic film is divided and photographed together with the reference density piece by the radiographic film imaging device 78, the image is captured, and synthesized in the CPU of the image processing device 84. Information is created and transmitted to the pass / fail determination device 88.

More specifically, it is determined whether or not a defect is present from a photographed image. If the defect is identified, the type of the defect (spherical defect or linear defect) is estimated by a learned algorithm, and the defect is determined. Measure the dimensions of For example, regarding the position and size of each defect, as shown in FIGS. 8 and 9, two points of the rectangular coordinates circumscribing the defect 110 (PA in FIG. 8)
1, PA2) XY coordinates ((x1, y1) (x
2, y2)), a defect position shown in FIG.
The defect length (major axis) represented by a circumscribed length L parallel to the principal axis of inertia 110A, the defect width (minor axis) represented by a circumscribed length W orthogonal to the principal axis of inertia 110A, and the major axis and minor axis in parallel. Four points of contact coordinates with the rectangle circumscribing the defect (Fig. 8
(PB1, PB2, PB3, PB4) can be output as defect position information. As shown in FIG. 9, the coordinate system of the defect position can be represented by, for example, a coordinate system having the origin at the coordinates of the upper left vertex when arranged on the left side toward the tube end of the radiographic film 40.

As described above, the defect information detected by performing predetermined image processing on the image information of the radiographic film image captured by the radiation film imaging device 78 is as shown in FIG. , And photographing state information and pure defect information, which are stored in the photographing state information DB 102 and pure defect information DB 104, respectively.

When the preparation for the automatic pass / fail judgment is completed in this way, the process enters step 1010 of the automatic judgment flow 1000 shown in FIG. A photographing condition determination criterion corresponding to a manufacturing standard and the photographing state information DB;
By comparing the shooting status information of the production lot stored in the file 102 with the shooting status information, a pass / fail judgment is made for all shooting conditions.

More specifically, the density of the welded portion of the transmission photograph, the minimum wire diameter that can be identified by the wire type transmittance meter, or the transmittance meter discrimination calculated from the minimum wire diameter and the welding thickness (transmission thickness). (%), The number of holes that can be identified by the Hall-type transmissometer, and the ratio of the density difference between the density at the center of the tone meter and the density of the base material near the tone meter (density difference rate). Of these, items required by the standard are checked, and only when all the conditions are satisfied, the pass / fail judgment of step 1030 is performed. On the other hand, when any of the conditions is not satisfied In this case, the automatic defect acceptance / rejection determination is impossible, and the process ends.

If the result of the determination in step 1020 is OK and the film image satisfies all of the photographing conditions, the flow advances to step 1030, where the reference / specification D
Of the pass / fail judgment criteria filed in B100, the defect pass / fail judgment criterion corresponding to the production standard of the lot is compared with the pure defect information of the production lot filed in the pure defect information DB 104 to make the defect pass / fail judgment. Do. Specifically, the presence or absence of a defect within the target range is checked, the defect type output from the image processing information is recognized based on the determination conditions of the steel pipe standard, and the defect position / defect length is determined according to the determination criterion for each type. The pass / fail judgment is performed based on the result, and the judgment result is output.

FIG. 10 shows a schematic flow of the defect pass / fail judgment performed in step 1030.

The criterion is not limited to individual defects, but is determined by individual defects, group defects and the like. For this reason, as shown in FIG. 10, the judgment is classified into three types: individual defect, group defect, and overall constraint rule, and judgment is performed for each of them. Is determined. These rules are not specified for all lots, and vary from lot to lot. Therefore, the judgment is performed on the items specified by the radiation judgment specification.

As described above, the defect acceptance / rejection determination in the present invention is performed by determining the individual defect information such as the type, size or length, position, and defect density of an individual defect by using the defect factor (type) definition 232 and the individual defect length definition. 234, individual defect acceptance / rejection judgment (pass / fail judgment based on individual defect judgment) to determine pass / fail by comparing with individual defect regulation 230 including defect density regulation 236, etc., correlation of a plurality of defects, ie, total defect length Group defect information such as the number of group defects, the number of group defects, the total area, the total number of defect points, and the defect area ratio, are defined by a total defect length rule 242, a defect interval rule 244, a cluster defect rule 246, a total defect point rule 248, and a defect area. Group defect regulation 2 including rate regulation 250 etc.
A group defect pass / fail judgment (pass / fail decision is made by judging a plurality of judgments) in comparison with 40, the distance from the pipe end of the defect closest to the pipe end in the longitudinal direction of the weld line, and the off-seam amount of the weld line, End defect-free regulation 262,
Including at least one of a global constraint defect pass / fail decision (pass / fail decision based on global constraints) that determines pass / fail by comparing with a global constraint rule 260 having global constraints, such as an off-seam rule 264; Further, when at least one rejection is determined by the above pass / fail judgment, the rejection is determined as a total rejection due to a rejection factor which is the innermost side from the end in the longitudinal direction of the weld line among these, and the above pass / fail judgment is made. , And a final pass judgment 270, which is regarded as an overall pass when a pass judgment is made in any of the above cases.

The result of the photographing condition determination in step 1010 and the result of the pass / fail determination in step 1030 are as follows.
Both are stored in the determination result DB 106.

The criterion for determining whether a defect is acceptable or not depends on the shape of the defect (spherical defect or linear defect), and includes the type (factor) of the defect, the ratio of the length of the major axis to the minor axis, and the length of the major axis and the minor axis. Etc., and select the classification method required by the applicable steel pipe standard.

The defect factor specification 232 determines pass / fail of all steel pipe standards. If the defect type of the individual defect is any of crack, insufficient penetration (IP), and defective fusion (LF), , Regardless of the shape and size of the defect.

In the individual defect length specification 234, when the individual defect length (major axis) exceeds the upper limit, the defect is rejected.

In the defect concentration regulation 236, a defect having a higher concentration in the defect portion than the concentration in the base material portion is judged as having a small thickness or a cavity at that portion, and is rejected due to a defective concentration.

In the total defect length regulation 242, as shown in FIG. 11, the prescribed length L in the longitudinal direction of the steel pipe at an arbitrary position is determined.
If the sum of the lengths L1, L2, L3 of the defects included in S exceeds the upper limit, the test is rejected.

According to the defect interval regulation 244, as shown in FIG. 12, when there are two individual defects 112 and 114, for example, four points (Pa1
To Pa4) and four points of circumscribed rectangular contact coordinates of the defect 114 (Pb
Among the intervals of 1 to Pb4), the interval at the end is defined as the interval between the two defects, and the one having the minimum interval less than the specified value is rejected.

According to the cluster defect definition 246, when small defects having a size less than the reference length are collectively present at a distance less than the reference interval, they are caught as one collective defect, and the size of the cluster and the other defects are compared. Check the interval,
For example, using the sum of the defect lengths of the defects included in each cluster, the pass / fail judgment of the cluster defect length and the minimum cluster interval is performed in the same manner as the individual defect. Since the determination of the cluster defect definition involves searching for a combination of defects, in order to reduce the number of defects to be processed as much as possible, after the individual defect determination or the group defect determination, a defect other than a rejected defect to be cut is determined. It can be determined as a target.

In the total defect point specification 248, if the number of defects included in an arbitrary specified length range exceeds the upper limit value, the test is rejected.

In the defect area ratio regulation 250, as shown in FIG. 13, the sum of the defect areas of the spherical defects f (n), f (n + 2), and f (n + 4) over the reference length range in the longitudinal direction of the steel pipe is obtained. If the ratio (defect area ratio) to the reference area exceeds the upper limit, the test is rejected. The area of the spherical defect is represented by an ellipse L · W which is in contact with the major axis L / minor axis W.
-It can be calculated by π / 4. In FIG. 13, f
(N + 1) and f (n + 3) are not considered because they are linear defects.

In the pipe end defect-free specification 262, as shown in FIG. 14, when there is no rejection defect, the pipe end position is within a specified length range. When there is a rejection defect, the pipe is furthest from the pipe end. The defect is checked from the position immediately after the defect position to the specified length range. If a defect exists, the defect is also rejected. When a defect is detected, the position of the defect is defined as a pipe end, and the process is repeated until no defect is found.

In the off-seam regulation 264, as shown in FIG. 15, the off-seam amount is sequentially checked from the pipe end,
When the off-seam upper limit value is exceeded at the continuous n points, the off-seam farthest from the pipe end is determined to be off-seam and rejected.

In the above description, the present invention relates to a UO
In the pipe end radiation transmission test of the E steel pipe, an example in which a developed radiographic film image is captured by a radiographic film imaging apparatus has been described. However, the application target of the present invention is not limited to this. Automatic pass / fail judgment of radiographic transmission test images of welded parts and other metal material welds, and also when capturing images of fluorescent intensifier tubes obtained in a radiographic test of metal material welds via an imaging device It is possible.

[0051]

EXAMPLE A pass / fail judgment program was verified using a combination of five hypothetical standards and a simulated defect diagram, and it was confirmed that an accurate judgment was made for each specification.

[0052]

According to the present invention, when the photographing condition is not satisfied, the pass / fail judgment of the defect is omitted, and the inspection can be efficiently performed. The judgment result is output and the inspector is not confused.

In particular, according to the third aspect, the complicated work of inputting the pass / fail judgment criterion for each radiation transmission test image to be automatically judged is avoided.

Further, according to the present invention, the defect information and the pass / fail judgment criteria corresponding to the defect information are complicated and diversified, and it is difficult to automate the defect information. The pass / fail judgment can be almost completely automated, and the visual judgment of the inspector becomes unnecessary. Therefore, the determination error due to fatigue and the variation in the determination among a plurality of inspectors are eliminated, and the reliability of the inspection, along with the mechanical detection of the defect information, the detection by the mechanization of the measurement, and the improvement of the measurement accuracy, are improved. Stability is improved, and automatic pass / fail judgment with high inspection efficiency is possible.

[Brief description of the drawings]

FIG. 1 is a front view showing a method of manufacturing a UOE steel pipe to be measured according to an embodiment of the present invention.

FIG. 2 is a perspective view for explaining a radiographic imaging site.

FIG. 3 is a diagram showing an image of a radiograph.

FIG. 4 is a diagram for explaining major defects of a UOE steel pipe;

FIG. 5 is a diagram schematically showing a radiation film image;

FIG. 6 is a diagram showing an overall configuration of an automatic film determination system to which the present invention is applied.

FIG. 7 is a flowchart showing an automatic determination procedure according to the present invention.

FIG. 8 is a diagram for explaining defect position information;

FIG. 9 is a diagram for explaining a defect position coordinate system;

FIG. 10 is a flowchart showing a procedure for determining whether a defect is acceptable according to the present invention.

FIG. 11 is a diagram for explaining a total defect length regulation used in the defect pass / fail judgment.

FIG. 12 is a diagram showing a method of obtaining an interval between individual defects.

FIG. 13 is a diagram for explaining the definition of the defect area ratio.

FIG. 14 is a diagram for explaining a pipe end defect-free provision.

FIG. 15 is a diagram for explaining an off-seam rule.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 12 ... UOE steel pipe 12B ... Weld bead part 12M ... Base material part 12E ... Pipe end 12K ... Pipe end groove processing part 14 ... Transmission imaging range 20 ... Crack 22 ... Penetration defect (IP) 24 ... Poor fusion (LF) 26 ... Blowhole 28 ... Slag entrainment 30 ... Off-seam 40 ... Radiographic film 42 ... Film number 44 ... Permeability meter image 50 ... Steel pipe conveying device 52,56,64,82 ... Sequencer 54 ... Radiographic photographing device 60 ... Radiography Room 70: Automatic film developing room 74: Automatic developing machine 78: Radiographic film imaging device 84: Image processing device 88: Pass / fail judgment device 92: Judgment reference input terminal device 100: Reference / specification database (DB) 102: Shooting state information Database (DB) 104: Pure defect information database (DB) 106: Judgment result database (DB) 10, 112, 114 ... defect 230 ... individual defect specification 232 ... defect factor (type) specification 234 ... individual defect length specification 236 ... defect concentration specification 240 ... group defect specification 242 ... total defect length specification 244 ... defect interval specification 246 ... cluster Defect stipulation 248: Total defect point stipulation 250: Defect area ratio stipulation 260: Overall constraint stipulation 262: Tube end defect-free stipulation 264: Off-seam stipulation 270: Final pass judgment 1000: Automatic judgment flow

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Mitsuhiro Kusunoki 1 Kawasaki-cho, Chuo-ku, Chiba-shi, Chiba Chiba Works, Ltd. (72) Inventor Toshio Hasegawa 3-1-1 Higashi-Kawasaki-cho, Chuo-ku, Kobe-shi, Hyogo No. 1 Kawasaki Heavy Industries, Ltd.Kobe Plant (72) Inventor Seishiro Kawano 3-1-1, Higashi Kawasaki-cho, Chuo-ku, Kobe City, Hyogo Prefecture Kawasaki Heavy Industries, Ltd.Kobe Plant (72) Inventor Koji Sugimoto Kawasaki, Akashi, Hyogo Prefecture No. 1-1, Kawasaki Heavy Industries, Akashi Factory (72) Inventor Sadao Inuchi 1-1, Kawasaki-cho, Akashi City, Hyogo Prefecture, Kawasaki Heavy Industries, Ltd.

Claims (5)

[Claims] An image input means captures a radiation transmission test image of a welded part, performs predetermined image processing on the captured image information, detects defect information, and determines whether the defect information has a predetermined pass / fail status. In an automatic pass / fail judgment method of a welded part radiographic transmission test image that automatically judges pass / fail of the welded part by comparing with a judgment criterion, the pass / fail judgment criterion is a photographing condition judgment criterion and a defect pass / fail judgment criterion; Are classified into photographing state information and pure defect information, respectively. First, the photographing state information is compared with the photographing condition determination criteria, and only when all photographing conditions pass, the pure defect information is compared with the defect pass / fail. Automatic pass / fail judgment of the radiographic transmission test image of a weld, characterized in that a defect pass / fail judgment is made by comparing with a judgment criterion, and if any of the photographing conditions is unsuccessful, the defect pass / fail judgment cannot be made. Method. 2. The photographing state information according to claim 1, wherein
Film density, transparency meter discrimination, including at least one of the tone meter density difference, the shooting condition determination criteria, the film density regulation, transparency meter discrimination regulation, tone meter density difference regulation An automatic pass / fail judgment method for a radiographic examination image of a welded part, which includes a relevant rule. 3. The pass / fail judgment criterion according to claim 1, wherein the pass / fail judgment criterion is a file determined in advance for each standard / criterion, stored in a storage device, and a standard / criterion applied to the lot is input by input means. On the other hand, in the case of standards / standards not filed in the storage device, or when there is an additional specification of the customer, the pass / fail judgment criterion is stored in the storage device via input means. An automatic pass / fail determination method for a radiographic examination image of a welded part, which is manually input. 4. The defect acceptance / rejection determination according to claim 1, wherein the individual defect information such as the type, size or length, position, and density of the individual defect is determined by a defect factor (type). Individual defect pass / fail judgment, which determines pass / fail by comparing with individual defect provisions including regulations, individual defect length regulations, defect density regulations, etc., and the interrelationship of multiple defects, that is, total defect length, defect interval, group defect Group defect information such as quantity, total area, total number of defect points, defect area ratio, etc.
Group defect acceptance / rejection judgment which determines pass / fail by comparing with group defect regulations including cluster defect regulations, total defect point regulations, defect area ratio regulations, etc., and the material edge of the defect closest to the material edge in the longitudinal direction of the welding line The distance from the welding line off-seam amount, etc., including at least one of the overall constraint defect pass / fail determination to determine pass / fail by comparing with the overall constraint definition such as material end no defect definition, off-seam definition, etc. If at least one rejection is determined by the pass / fail judgment, a rejection is made as a total rejection due to a rejection factor that is the innermost side from the weld line longitudinal end portion among these, and it passes in any of the above pass / fail judgments An automatic pass / fail judgment method for a welded part radiographic transmission test image, which includes a final pass / fail judgment to determine a total pass when a judgment is made. 5. An image input means for taking in a radiation transmission test image of a weld, and an image processing means for performing predetermined image processing on the image information taken in by the image input means to detect defect information. A determination criteria storage unit that classifies and stores a predetermined pass / fail determination criterion into a shooting condition determination rule and a defect pass / fail determination criterion; and classifies the defect information into shooting state information and pure defect information. The photographing state information is compared with the photographing condition criterion, and only when all the photographing conditions pass, a defect pass / fail decision is made by comparing the pure defect information with the defect pass / fail decision criterion. An automatic pass / fail determination apparatus for a welded part radiographic transmission test image, comprising: a determination unit for determining a failure pass / fail determination in the case of rejection.