KR102105503B1 - Method and apparatus for automatically evaluating weld quality - Google Patents

Abstract

A method for automatically evaluating welding quality is provided. The automatic welding quality evaluation method includes the steps of acquiring a gray image and a color image of a weld portion where spot welding is performed; Binarizing the gray image based on a predetermined threshold; Generating a circumscribed circle circumscribed with the welding portion; Generating an inscribed circle inscribed with the weld portion; And comparing the sizes of the inscribed circle and the circumscribed circle to evaluate the welding state of the spot welding.

Description

Welding quality automatic evaluation method and welding quality automatic evaluation device {METHOD AND APPARATUS FOR AUTOMATICALLY EVALUATING WELD QUALITY}

The present invention relates to a method for automatically evaluating a welding quality and an apparatus for automatically evaluating a welding quality, and more particularly, a method for automatically evaluating a welding quality using a C-scanned image and an apparatus for automatically evaluating a welding quality.

Resistance spot welding is a welding method in which an electrode is brought into contact with both ends of a plate to apply pressure and current, and melts the surfaces of the overlapped plate using resistance heat generated thereby. Such resistance spot welding is used in many automobile production processes because it has high productivity in the plate welding part because it has a high working speed and easy difficulty compared to other methods.

Process elements applied in the resistance spot welding process include pressure, current, operating frequency, and welding time, and a combination thereof determines welding quality. In particular, welding quality is very important because the quality of resistance spot welding in the automobile production process is directly related to the life of the vehicle driver and the occupant. The part in which a part of the joint surface is melted and welded in a cross-shaped cross-section is called a nugget. In order to evaluate the quality of resistance spot welding, a method of determining the quality of the weld by measuring the shape of the weld nugget is generally used. You can. Since the nugget generated in the resistance spot welding process exists inside the plate materials, a method of cutting or separating the welded part is generally used for quality evaluation.

Since such a fracture test is conducted through sampling, it is impossible to perform a full inspection, and there is a disadvantage in that the tested specimen or finished product must be discarded. Therefore, a non-destructive test (NDT) is generally used to evaluate the welding quality, and in the NDT field, a method using ultrasonic is becoming common.

The NDT using ultrasound is a method of predicting the state inside the specimen by analyzing the echo waveform collected by the transducer after the ultrasonic signal emitted from the ultrasonic transducer is reflected, and is largely divided into A-scan and C-scan. .

The A-scan method is a method of expressing the waveform of the echo signal with respect to the time axis using a one-dimensional ultrasonic probe. It is widely used in industrial fields because it is easy to carry and the equipment is relatively inexpensive. However, there is a problem in reliability since an experienced non-destructive inspection agent capable of analyzing the waveform of the ultrasonic signal is required and the welding quality is determined by their subjective judgment.

In order to compensate for the disadvantages of the A-scan method, the C-scan method was introduced, which is a method of displaying the welding area as a two-dimensional image by arranging the transducers used in the A-scan method in two-dimensional horizontal and vertical arrangements. . Since the C-scan type NDT equipment is expensive, it is not universalized in the production site and it is difficult to utilize the C-scan image. Therefore, it is required to develop a device capable of evaluating the welding quality without the help of a welding quality evaluation expert.

Korean Patent Publication No. 10-2013-0089353 (201.08.12)

The problem to be solved by the present invention is to provide an automatic welding quality evaluation method and an automatic welding quality evaluation device that can automatically evaluate welding quality without the help of a welding quality evaluation expert.

The problem to be solved by the present invention is to provide an objective method for automatically evaluating welding quality and an apparatus for automatically evaluating welding quality, in which the evaluation process is simple but the error is not large and the subject of the welding quality evaluator is excluded.

According to an embodiment of the present invention, a method for automatically evaluating welding quality is provided. The automatic welding quality evaluation method includes the steps of acquiring a gray image and a color image of a weld portion where spot welding is performed; Binarizing the gray image based on a predetermined threshold; Generating a circumscribed circle circumscribed with the welding portion; Generating an inscribed circle inscribed with the weld portion; And comparing the sizes of the inscribed circle and the circumscribed circle to evaluate the welding state of the spot welding.

In addition, according to another embodiment of the present invention, an apparatus for automatically evaluating welding quality is provided. The automatic welding quality evaluation apparatus includes a welding image processing unit that filters and interpolates an image of a spot welding portion received as an echo signal of an ultrasonic signal to generate a high resolution gray image and a color image; A gray image binarization unit which binarizes the gray image based on a predetermined threshold and detects a contour for a welded portion; A circumscribed circle and an inscribed circle generating unit for generating an inscribed circle inscribed with the circumscribed circle circumscribed with the welding portion; It may include a welding state evaluation unit for evaluating the welding state of the spot welding by comparing the size of the inscribed circle and the circumscribed circle.

According to an embodiment of the present invention, an automatic welding quality evaluation method and an automatic welding quality evaluation device are provided, which can automatically evaluate welding quality without the help of a welding quality evaluation expert.

According to an embodiment of the present invention, there is provided an objective method for automatically evaluating welding quality and an apparatus for automatically evaluating welding quality, in which the evaluation process is simple but the error is not large and the subject of the welding quality evaluator is excluded.

1 is a view for explaining a C-scan according to the present invention,
2 is a control block diagram of an apparatus for automatically evaluating welding quality according to an embodiment of the present invention,
3 is a welding image obtained from the welding image processing unit according to an embodiment of the present embodiment,
4A and 4B are diagrams for binarizing a welding gray image and detecting contours according to an embodiment of the present embodiment,
5A, 5B, and 5C are views illustrating a circumscribed circle and an inscribed circle contacting a welding part according to an embodiment of the present embodiment,
6A and 6B are views illustrating a weld part evaluated as a welding failure according to an embodiment of the present embodiment,
7A and 7B are views showing a welding part evaluated as good welding according to an embodiment of the present embodiment,
8A and 8B are views showing a welding part evaluated as excellent in welding according to an embodiment of the present embodiment,
9 is a control flowchart for explaining a method for automatically evaluating welding quality according to an embodiment of the present embodiment,
10 is a control flowchart for explaining a method for automatically evaluating welding quality according to another embodiment of the present embodiment,
11 is a diagram illustrating a computing device according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art to which the present invention pertains may easily practice. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. In addition, in order to clearly describe the present invention in the drawings, parts irrelevant to the description are omitted, and like reference numerals are assigned to similar parts throughout the specification.

In this specification, redundant description of the same components is omitted.

Also, in this specification, when a component is referred to as being 'connected' or 'connected' to another component, it may be directly connected to or connected to the other component, but other components in the middle It should be understood that may exist. On the other hand, in this specification, when a component is referred to as being 'directly connected' or 'directly connected' to another component, it should be understood that no other component exists in the middle.

In addition, the terms used in this specification are only used to describe specific embodiments, and are not intended to limit the present invention.

Also, in this specification, a singular expression may include a plural expression unless the context clearly indicates otherwise.

Also, in the present specification, terms such as 'include' or 'have' are only intended to designate the existence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, and one or more of them. It should be understood that the existence or addition possibilities of other features, numbers, steps, actions, components, parts or combinations thereof are not excluded in advance.

Also, in this specification, the term 'and / or' includes a combination of a plurality of listed items or any one of a plurality of listed items. In this specification, 'A or B' may include 'A', 'B', or 'both A and B'.

Also, in this specification, detailed descriptions of well-known functions and configurations that may obscure the subject matter of the present invention will be omitted.

An embodiment of the present invention performs a non-destructive test using ultrasound to evaluate the state of spot welding, and a C-scan method with an improved A-scan method is applied. The A-scan method is widely used in front-line industrial sites because it is easy to carry and use, but requires professional training to analyze the waveform of the ultrasonic signal, and requires skilled non-destructive inspection experts, and welding quality is determined by their subjective welding. Quality reliability problems can arise.

1 is a view for explaining a C-scan according to the present invention.

Complementing the disadvantages of A-scan non-destructive testing, C-scan non-destructive testing combines each one-dimensional waveform by placing the transducers used in the A-scan equipment, that is, ultrasonic transducers, in a horizontal and vertical array. As a result, the welding area can be expressed as a two-dimensional image. 1, the two-dimensional transducer propagates the ultrasonic signal in a wave form to the spot-welded portion while scanning the spot-welded plate (A) as shown in FIG. 1, and the propagated ultrasonic signal is inside the plate (A). Propagates.

The signal propagated into the plate A may be reflected at a defect site such as a crack inside, and then measured again by the transducer. By analyzing the detection time of the ultrasonic echo signal measured by the transducer, it is possible to predict the depth of the defect and the thickness of the object existing in the object. That is, if the time at which the ultrasound signal is propagated and the time at which the ultrasound signal is measured by measuring the transducer can be measured, the progress path of the ultrasound signal in the object may be calculated. In general, such short-wavelength ultrasonic signals are effective in measuring small-scale coupling in an object.

In other words, by analyzing the waveforms of the ultrasonic echo signals, the position of the defect existing in the object can be predicted, so that the position of the welding part can be grasped by the same method. If the defect waveform is measured from the echo signal measured around the weld, it can be judged that the welding was not properly performed at the corresponding part because the ultrasonic signals were reflected at the joint surfaces of the two unmelted plates. Conversely, if the defect waveform was not measured, it can be seen that the two sheets were completely melted by the welding process. Therefore, it is possible to accurately predict the position of the weld by analyzing the defect waveform existing between the initial pulse waveform and the back echo waveform.

As described above, the position of the welding part can be predicted through time waveform analysis of the ultrasonic echo signal, but a method for measuring the welding quality is required for ultrasonic non-destructive inspection. The welding quality after the welding process varies from no welding at all to no welding at all, and the two plates are completely burnt. This welding quality can be predicted by analyzing the magnitude component of the defect echo signal.

2 is a control block diagram of an apparatus for automatically evaluating welding quality according to an embodiment of the present invention.

The apparatus for automatically evaluating welding quality may be implemented as a computer that is connected to the ultrasonic transducer illustrated in FIG. 1 by wire or wirelessly to receive an echo signal, process an echo signal, and determine a welding state. The automatic welding quality evaluation apparatus may be implemented as a portable user terminal, and in this case, the user terminal may include a display unit capable of displaying a welding image and a user interface capable of receiving a user input signal.

As illustrated, the apparatus for automatically evaluating welding quality according to the present embodiment includes a welding image processing unit 210, a gray image binarization unit 220, a circumscribed and inscribed circle generating unit 230, and a welding state evaluation unit 240. You can.

The welding image processing unit 210 may acquire a gray image and a color image of the welding portion where spot welding is performed. To this end, the welding image processing unit 210 histogram stretches the gray image of the 5 level brightness level of the 8 * 8 spatial resolution obtained from the echo signal of the ultrasonic signal, through image interpolation using blurring using a Gaussian filter, and closing operation. A high-resolution image of the welding part may be acquired, and a color image may be acquired by mapping a color table suitable for a gray level.

3 is a welding image obtained from a welding image processing unit according to an embodiment of the present embodiment. The gray image of FIG. 3 is a high-resolution image having a 64 * 64 resolution, and the color image is a color map of a visible light region mapped to a gray level of the high-resolution gray image.

Two images of the welding part are necessary for automatic evaluation of the welding quality, and the welding quality inspection method using the gray image and the welding quality inspection method using the color image may be different. The welding quality inspection may be performed in one of two methods, or the welding quality may be finally evaluated by comprehensively applying the two quality inspections.

The gray image binarization unit 220 binarizes the gray image based on a predetermined threshold value and detects a contour of the welded portion.

Binarization of gray images and contour detection are used to distinguish the welded welded portion from the two-dimensional image and its periphery, and the size or shape inside the contour can be used to evaluate the welding quality.

The contour of the welding part can be detected through a canny edge operator, and a thin and sharp edge, that is, a contour, can be obtained using the canny edge detection method.

4A and 4B are diagrams of binarizing a welding gray image and detecting contours according to an embodiment of the present embodiment. The gray image having a plurality of levels in FIG. 3 was changed to an image having two gray levels as shown in FIG. 4A, and the boundary portion of the two gray levels was detected as the outline of the welding portion. The inside of the contour is considered to be a welded part, and the quality and weld quality can be inspected through a review of shape and size.

The circumscribed circle and the circumscribed circle generating unit 230 may generate a circumscribed circle circumscribed with the welded portion, and may generate an circumscribed circle circumscribed with the welded portion.

The circumscribed circle may be generated based on the length and center coordinates of the longest axis by deriving the length and center coordinates of the longest axis that the welding part abuts using the contour of FIG. 4B. 5A is a diagram illustrating a circumscribed circle generated according to the present embodiment, and a circumscribed circle is displayed on an image in which a contour is detected.

On the other hand, an optimal inscribed circle inscribed to the welded portion can be derived using the center pixel and the 8-directional pixel from the center pixel. For example, an arbitrary point inside the outline is used as the center pixel, pixels in the eight directions (horizontal, vertical and diagonal directions) are detected from the center pixel, and the center pixel of the inscribed circle is the center pixel in which eight pixels in contact with the contour are present. In this way, the inscribed circle can be detected. If it is difficult to find the center point corresponding to the eight directional pixels that come into contact with the contour, the inscribed circle may be found based on the center point for the seven directional pixels, or the inscribed circle may be detected centering on the pixel that inscribes the contour as much as possible.

5B is a diagram illustrating an inscribed circle generated according to the present embodiment, and an inscribed circle is displayed on the binarized image.

5C is a diagram showing the detected circumscribed and inscribed circles as shown in FIGS. 5A and 5B in color images and binarization images. By simultaneously displaying the circumscribed circle and the circumscribed circle, the shape and size of the welding can be observed at a glance. The size of the circumscribed circle and the circumscribed circle, that is, the area is similar, the center of the circumscribed circle and the circumscribed circle adjacent to each other may be determined to be a round and stable form of spot welding.

5A, 5B, and 5C, when a circumscribed circle and an inscribed circle are detected, the welding state evaluation unit 240 may compare the sizes of the inscribed circle and the circumscribed circle to evaluate the welding state of spot welding.

First, if the area of the inscribed circle is less than the first predetermined range of the gray image area, the welding state evaluation unit 240 may evaluate the welding defect.

6A and 6B are views showing a weld portion evaluated as a welding failure according to an embodiment of the present embodiment. As shown, it can be seen from the images of FIGS. 6A and 6B that the size of the welded portion is very small and the shape is not circular. This can be judged to mean that welding was not performed properly.

According to this embodiment, if the area of the inscribed circle is 15% or less of the total image area, that is, the first range is set to 15% of the total image area, and if the area of the inscribed circle is 15% or less of the total image area, the size is defective. It can be judged.

In addition, the welding state evaluation unit 240 may evaluate the welding as good if the area of the inscribed circle is greater than or equal to the first predetermined range of the gray image area, and the area of the inscribed circle is less than or equal to the preset second range of the circumscribed circle area.

If the area of the inscribed circle with respect to the shape of the weld is greater than or equal to the first range, i.e., not more than 15% of the total image area, and the welding is not defective, the weld is very well, i.e., it is judged as excellent welding. It can be difficult to do. In order to select such a case, the welding state evaluation unit 240 may compare the areas of the inscribed circle and the inscribed circle. For example, when the area of the inscribed circle is 15% or more of the total image area, and the area of the inscribed circle is 70% or less of the area of the circumscribed circle, the welding state evaluation unit 240 may determine the welding state as good. That is, the second range may be set to a constant range including 70%.

7A and 7B are views showing a welding part evaluated as good welding according to an embodiment of the present embodiment.

As shown, if the area of the inscribed circle is 70% or less of the circumscribed circle, the tail welding may have a tail shape or a crushed shape in one side, not a round shape. In this case, since the area of the inscribed circle is greater than or equal to the first range, it cannot be regarded as a welding defect, but it cannot be judged as excellent. In this case, the welding state is judged as good welding.

Finally, in the welding state evaluation unit 240, the centers of the inscribed circle and the circumscribed circle are within a predetermined range, the area of the inscribed circle is greater than or equal to a first predetermined range of the gray image area, and the area of the inscribed circle is a preset area of the circumscribed circle area If it exceeds the second range, the welding state can be judged as excellent welding.

For example, if the size of the welding portion is 15% or more of the entire image, and the area of the inscribed circle occupies more than 70% of the area of the circumscribed circle, and the center of the circumscribed circle and the inscribed circle is within a predetermined range, a tail is formed in a different direction from the weld center. It can be determined that the spot welding is stable with a round shape.

8A and 8B are views showing a weld part evaluated as excellent in welding according to an embodiment of the present embodiment, and as shown, it can be confirmed that the size of the weld is more than a certain level and the shape is close to the shape of the circle. have.

9 is a control flowchart for explaining a method for automatically evaluating welding quality according to an embodiment of the present embodiment. The automatic evaluation method according to the present embodiment is summarized as follows with reference to FIG. 9.

First, the apparatus for automatically evaluating welding quality may acquire a gray image and a color image of a welding portion where spot welding is performed (910).

The gray image and the color image correspond to a high-resolution image in which 8 × 8 low-resolution images are upsampled and noise is removed.

The gray image binarization unit 220 binarizes the gray image based on a predetermined threshold value and detects a contour using the canny edge detection method in the binarized image (920).

Then, the circumscribed circle and the circumscribed circle generating unit 230 may generate a circumscribed circle and an circumscribed circle by using a weld portion made clear with a contour (930, 940).

The circumscribed circle is a circle that circumscribes the contour of the welding portion, derives the length and center coordinates of the longest axis that the welding portion contacts, and can be generated based on the length and center coordinates of the longest axis.

The inscribed circle is a circle inscribed with the contour of the welded portion, and the circumscribed circle and inscribed circle generating unit 230 can derive an optimal inscribed circle inscribed to the welded portion using the center pixel and 8-directional pixels from the center pixel.

When the circumscribed circle and the circumscribed circle are generated, the welding state evaluation unit 240 may evaluate the welding state of the spot welding by comparing the sizes of the circumscribed circle and the circumscribed circle (950, 960, 970).

The welding state evaluation unit 240 may determine whether the area of the inscribed circle is less than or equal to 15% of a predetermined first range of the gray image area (950).

As a result of the determination, if the area of the inscribed circle is less than a predetermined first range of the gray image area, for example, 15%, the welding state evaluation unit 240 may evaluate the welding state of spot welding as a welding failure (951).

If the area of the inscribed circle is greater than or equal to the first range of the gray image area, and the preset second range of the area of the circumscribed circle is, for example, 70% or less (960), the welding state evaluation unit 240 determines the welding state of spot welding. It can be evaluated as good welding (961).

On the other hand, if the area of the inscribed circle contacting the welded portion exceeds the second range of the circumscribed circle area, and the center of the inscribed circle and the circumscribed circle is within a predetermined range (970), that is, if it is determined that the center of the inscribed circle and the circumscribed circle is located close, , The welding state evaluation unit 240 may evaluate the welding state of spot welding as excellent welding (971).

In order for the welding state to be excellent, the area difference between the inscribed circle and the circumscribed circle should not be large, and the shape of the inscribed circle and the circumscribed circle should have a similar center point, and the shape should be close to the circle.

10 is a control flowchart for explaining a method for automatically evaluating welding quality according to another embodiment of the present embodiment. Referring to FIG. 10, a method for automatically evaluating welding quality using a color image will be described as follows.

First, a gray image and a color image of a welding portion where spot welding is performed by the welding image processing unit 210 may be obtained (1010).

The gray image binarization unit 220 binarizes the gray image based on a predetermined threshold value and detects a contour using a canny edge detection method in the binarized image, and the contour area for the inside of the contour is calculated. You can (1020). The contour area can be regarded as the area of the welded part.

Also, the welding state evaluation unit 240 may calculate a color area having a color value smaller than a predetermined threshold level value in the color image of the welded portion (1030).

As described above, in the color image, the color of the visible light region is mapped to 256 gray images, and the welded round portion may be subdivided into a red system, and the other portion may be subdivided into a blue system through yellow.

According to the present embodiment, a specific color of the red portion representing the rounded portion welded may be set as a threshold level value. A portion having a color value smaller than the threshold value of the red system may be regarded as a portion formed with spot welding.

The welding state evaluation unit 240 may determine whether the color area is less than or equal to a predetermined first range of the outline area (1040).

As a result of the determination, if the color area is equal to or less than the first predetermined range of the contour area, the welding state evaluation unit 240 may evaluate the welding state as a welding failure (1041).

That is, when the area of the red portion corresponding to the main welding portion among the areas of the total spot welding is equal to or less than the reference value, it may be determined that the welding state is defective.

The first range may be set to 20 to 30% of the outline area, and may be set based on the experience value for spot welding, which is judged to be defective when evaluating the quality of the welding.

If the color area is not equal to or less than the preset first range of the contour area, the welding state evaluation unit 240 determines whether the color area is equal to or less than the preset second range (1050). That is, it is determined whether the color area exceeds the first range of the contour area and is the second range of the contour area.

The second range may be set to 50 to 60% of the outlined area. Of course, this range can also be adjusted according to the purpose of welding and the required welding accuracy level.

As a result of the determination, if the color area exceeds the first range of the outline area and is less than or equal to the second range, the welding state evaluation unit 240 may evaluate the welding state as good welding (1051).

Finally, if the color area of the spot welding is greater than or equal to the second range of the contour area (1060), the welding state evaluation unit 240 may evaluate the welding state as welding excellent (1061).

The automatic welding quality evaluation method with reference to FIG. 10 may be used independently and used for welding quality evaluation, and may be used together or auxiliary with the evaluation method of FIG. 9. For example, if the evaluation result is the same as that of the evaluation result of FIG. 9, that is, poor, good, and excellent, it may be a highly reliable evaluation. If the results of the two evaluation methods are different, a certain magnification may be applied to reflect which of the two evaluations is reliable, or the reliability of evaluation may be increased by adjusting a threshold range set in the area.

Welding technology is used in various industrial sites, including automobile manufacturing sites, and welding quality plays a very important role in product maturity. However, to date, in the field, non-destructive inspection experts who determine the welding quality are in the situation of determining the welding quality by analyzing the waveform of the one-dimensional A-scan signal. Therefore, there is a growing demand for an automatic welding quality determination algorithm because there may be a problem in welding quality determination due to human error.

In the present invention, a two-dimensional C-scan gray image is constructed using an ultrasound signal obtained from a C-scan probe in which the A-scan probe is arranged in two dimensions, and a high-resolution C-scan color image is obtained based on the ultrasound signal. Through this, the welding shape of the specimen can be expressed so that the worker or the manager can visually confirm the welding shape. In addition, the present invention proposes an algorithm for automatically determining the quality of the welding, not just expressing the welding shape, and through this, the results of the welding can be grasped in real time, thereby effectively preventing defective shipment. .

11 is a diagram illustrating a computing device according to an embodiment of the present invention. The computing device TN100 of FIG. 11 may be a device described in this specification (eg, an automatic welding quality evaluation device, etc.).

In the embodiment of FIG. 11, the computing device TN100 may include at least one processor TN110, a transmitting / receiving device TN120, and a memory TN130. Also, the computing device TN100 may further include a storage device TN140, an input interface device TN150, an output interface device TN160, and the like. Components included in the computing device TN100 may be connected by a bus TN170 to communicate with each other.

The processor TN110 may execute a program command stored in at least one of the memory TN130 and the storage device TN140. The processor TN110 may refer to a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor in which methods according to an embodiment of the present invention are performed. The processor TN110 may be configured to implement procedures, functions, and methods described in connection with embodiments of the present invention. The processor TN110 may control each component of the computing device TN100.

Each of the memory TN130 and the storage device TN140 may store various information related to the operation of the processor TN110. Each of the memory TN130 and the storage device TN140 may be configured as at least one of a volatile storage medium and a non-volatile storage medium. For example, the memory TN130 may be configured as at least one of read only memory (ROM) and random access memory (RAM).

The transmitting and receiving device TN120 may transmit or receive a wired signal or a wireless signal. The transmitting and receiving device TN120 may be connected to a network to perform communication.

Meanwhile, the embodiment of the present invention is not implemented only through the apparatus and / or method described so far, and may be implemented through a program that realizes a function corresponding to the configuration of the embodiment of the present invention or a recording medium in which the program is recorded. There is, such an implementation can be easily implemented by those skilled in the art from the description of the above-described embodiments.

Although the embodiments of the present invention have been described in detail above, the scope of rights of the present invention is not limited thereto, and various modifications and improvements of the skilled person using the basic concept of the present invention defined in the following claims are also provided. It belongs to the scope of the invention.

Claims (8)

As a method for automatically evaluating welding quality,
Acquiring a gray image and a color image of the welding portion where spot welding is performed;
Binarizing the gray image based on a predetermined threshold;
Generating a circumscribed circle circumscribed with the welding portion;
Generating an inscribed circle inscribed with the weld portion;
Comprising the step of evaluating the welding state of the spot welding by comparing the size of the inscribed circle and the circumscribed circle,
The step of generating the circumscribed circle derives the length and center coordinates of the longest axis that the welded part contacts using the contour of the welded part, and generates a circumscribed circle based on the length and the center coordinates of the longest axis,
The step of generating the inscribed circle sets an arbitrary point inside the contour as a center pixel, detects an 8-direction pixel contacting the contour from the center pixel, and the number of 8-direction pixels contacting the contour range is 8 ranges. The inscribed circle is detected by making the most central pixel in the center of the inscribed circle,
The step of evaluating the welding state of the spot welding is a welding for evaluating the welding state of the spot welding by using whether the center of the circumscribed circle and the center of the circumscribed circle are within a set range together with the size comparison of the inscribed circle and the circumscribed circle. Automatic quality evaluation method. delete delete According to claim 1,
Evaluating the welding state of the spot welding,
If the area of the inscribed circle is less than the predetermined first range of the gray image area, the welding quality automatic evaluation method characterized in that it is evaluated as a welding failure. According to claim 1,
Evaluating the welding state of the spot welding,
The area of the inscribed circle is greater than or equal to a first predetermined range of the gray image area,
If the area of the inscribed circle is less than the predetermined second range of the area of the circumscribed circle, the welding quality automatic evaluation method characterized in that it is evaluated as welding good. According to claim 1,
Evaluating the welding state of the spot welding,
The center of the inscribed circle and the circumscribed circle is within a predetermined range,
The area of the inscribed circle is greater than or equal to a first predetermined range of the gray image area,
Automatic welding quality evaluation method, characterized in that when the area of the inscribed circle exceeds a predetermined second range of the area of the circumscribed circle as excellent welding. According to claim 1,
Evaluating the welding state of the spot welding,
Detecting a contour on the weld portion in the binarized binarization image and calculating a contour area inside the contour;
Calculating a color area having a color value smaller than a predetermined threshold level value in the color image;
If the color area is equal to or less than the first predetermined range of the contour area, the welding state is evaluated as a welding failure,
If the color area exceeds the first range of the contour area and is equal to or less than a preset second range, the welding state is evaluated as good welding,
If the color area is greater than or equal to the second range of the contour area, the welding quality is automatically evaluated as the welding state. Automatic welding quality evaluation device,
A welding image processing unit that filters and interpolates an image of a spot welding portion received as an echo signal of an ultrasonic signal to generate a high resolution gray image and a color image;
A gray image binarization unit which binarizes the gray image based on a predetermined threshold and detects a contour for a welded portion;
A circumscribed circle and an inscribed circle generating unit for generating an inscribed circle inscribed with the circumscribed circle circumscribed with the welding portion;
It includes a welding state evaluation unit for evaluating the welding state of the spot welding by comparing the size of the inscribed circle and the circumscribed circle,
The generating unit derives the length and center coordinates of the longest axis that the welding part touches using the contour, and generates the circumscribed circle based on the length and the center coordinates of the longest axis,
The generation unit sets an arbitrary point inside the contour as a center pixel, detects an 8-direction pixel contacting the contour from the center pixel, and the number of 8-direction pixels contacting the contour is the largest within 8 ranges. The inscribed circle is detected with the center pixel as the center of the inscribed circle,
The welding state evaluation unit is a welding quality automatic evaluation device for evaluating the welding state of the spot welding by using whether the center of the circumscribed circle and the center of the circumscribed circle are within a set range together with the size comparison of the inscribed circle and the circumscribed circle.

Images