CN111275659A

CN111275659A - Weld image processing method and device, terminal device and storage medium

Info

Publication number: CN111275659A
Application number: CN201811466376.6A
Authority: CN
Inventors: 邓景煜; 李�昊; 孙小峰
Original assignee: Commercial Aircraft Corp of China Ltd; Shanghai Aircraft Manufacturing Co Ltd
Current assignee: Commercial Aircraft Corp of China Ltd; Shanghai Aircraft Manufacturing Co Ltd
Priority date: 2018-12-03
Filing date: 2018-12-03
Publication date: 2020-06-12
Anticipated expiration: 2038-12-03
Also published as: CN111275659B

Abstract

The invention discloses a welding seam image processing method and device, terminal equipment and a storage medium. The method comprises the following steps: acquiring an inclination angle of a welding seam region in a welding seam image; correcting the weld image according to the inclination angle; and intercepting the corrected welding seam image according to the rectangular area to obtain the corrected welding seam image, wherein the parameters of the rectangular area correspond to the texture features of the corrected welding seam image. By using the method, the accuracy of nondestructive testing of the welding seam image can be improved.

Description

Weld image processing method and device, terminal device and storage medium

Technical Field

The embodiment of the invention relates to the technical field of digital image processing, in particular to a welding seam image processing method and device, terminal equipment and a storage medium.

Background

With the rapid development of modern industrial technologies, welding technology, which is one of the important means for machine manufacturing, has been widely used in various sectors of manufacturing industry. The quality of welding quality seriously affects the use safety of welding products, so that nondestructive testing of welding structural parts is essential.

In the technical field of nondestructive testing, methods such as ray detection, acoustic emission detection, ultrasonic detection and the like can be used for detecting the welding quality. After acquiring the digital image of the object to be detected, the main work content is to perform detection, such as defect detection, by manual or digital image processing techniques.

When manual processing is adopted, the problems of low efficiency and high false detection rate exist. The defect detection of digital images has become an important means for quality evaluation with the development of image processing technology. However, when the evaluation is performed by the image processing technology, some characteristics similar to the defect are taken as the defect to be misjudged, and the accuracy of the nondestructive testing is reduced.

Disclosure of Invention

The embodiment of the invention provides a welding seam image processing method and device, terminal equipment and a storage medium, which are used for improving the accuracy of nondestructive testing on a welding seam image.

In a first aspect, an embodiment of the present invention provides a weld image processing method, including:

acquiring an inclination angle of a welding seam region in a welding seam image;

correcting the weld image according to the inclination angle;

and intercepting the corrected welding seam image according to the rectangular area to obtain the corrected welding seam image, wherein the parameters of the rectangular area correspond to the texture features of the corrected welding seam image.

In a second aspect, an embodiment of the present invention further provides a weld image processing apparatus, including:

the acquisition module is used for acquiring the inclination angle of a welding seam area in a welding seam image;

the correction module is used for correcting the welding seam image according to the inclination angle;

and the intercepting module is used for intercepting the corrected welding seam image according to the rectangular area to obtain the corrected welding seam area image, and the parameters of the rectangular area correspond to the texture features of the corrected welding seam image.

In a third aspect, an embodiment of the present invention further provides a terminal device, including:

one or more processors;

storage means for storing one or more programs;

the one or more programs are executed by the one or more processors, so that the one or more processors implement the weld image processing method provided by the embodiment of the invention.

In a fourth aspect, the embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the weld image processing method provided by the embodiment of the present invention.

The embodiment of the invention provides a welding seam image processing method, a welding seam image processing device, terminal equipment and a storage medium, and by using the technical scheme, a welding seam image can be corrected based on the inclination angle of a welding seam area in the welding seam image; then, the corrected welding seam image is intercepted according to the rectangular area corresponding to the texture features of the corrected welding seam image, so that the corrected welding seam area image is obtained, the non-welding seam area in the welding seam image is effectively excluded, and the accuracy of subsequently acquiring the defect features of the welding seam area is improved.

Drawings

Fig. 1 is a schematic flowchart of a weld image processing method according to an embodiment of the present invention;

fig. 2a is a schematic flowchart of a weld image processing method according to a second embodiment of the present invention;

FIG. 2b is a schematic view of a weld image provided by the second embodiment of the present invention;

FIG. 2c is a schematic diagram of a small-sized image provided by the second embodiment of the present invention;

FIG. 2d is a schematic diagram of an enhanced image provided by the second embodiment of the present invention;

FIG. 2e is a schematic diagram of a first binarized image according to a second embodiment of the present invention;

FIG. 2f is a schematic diagram of a centerline image provided by the second embodiment of the present invention;

FIG. 2g is a schematic diagram of an image after rotation according to the second embodiment of the present invention;

FIG. 2h is a schematic diagram of a corrected weld image according to the second embodiment of the present invention;

FIG. 2i is a schematic diagram of a texture image provided by a second embodiment of the present invention;

fig. 2j shows a schematic diagram of a second binarized image according to a second embodiment of the present invention;

FIG. 2k is a schematic diagram of a filtered image according to a second embodiment of the present invention;

FIG. 2l is a schematic diagram of a minimum rectangle provided by the second embodiment of the present invention;

FIG. 2m is a schematic diagram of a corrected weld region image provided by the second embodiment of the present invention;

fig. 2n is a schematic diagram illustrating a principle of removing a black border according to a second embodiment of the present invention;

fig. 3 is a schematic structural diagram of a weld image processing apparatus according to a third embodiment of the present invention;

fig. 4 is a schematic structural diagram of a terminal device according to a fourth embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.

Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the operations (or steps) as a sequential process, many of the operations can be performed in parallel, concurrently or simultaneously. In addition, the order of the operations may be re-arranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like.

Example one

Fig. 1 is a schematic flow chart of a method for processing a weld image according to an embodiment of the present invention, where the method is applicable to a situation of processing a weld image, and in particular, the method is applicable to a situation of processing an X-ray image of a weld to improve accuracy of nondestructive testing. The method may be executed by a weld image processing apparatus, wherein the apparatus may be implemented by software and/or hardware, and is generally integrated on a terminal device provided in the embodiment of the present invention, where the terminal device includes but is not limited to: servers, notebook computers, desktop computers, and the like.

At present, an X-ray technology is used as a common welding nondestructive testing technology and is very commonly applied to the detection of welding defects of welded structural parts with high quality requirements. After the X-ray image of the welding seam is obtained, the main working content is to detect the defects of the welding seam through manual work or digital image processing technology. The manual processing has the problems of low efficiency and high false detection rate. With the development of image processing technology, the defect detection of digital welding images becomes an important means for quality judgment of welding products.

In the X-ray image of the welding seam, the background area outside the welding seam area has some characteristics like welding seam defects and the like because the welding part can be caused by welding spatter, groove bulge of parts and the like. These characteristics may be determined empirically during manual processing. However, for the adaptive weld X-ray digital image algorithm, the characteristics may be regarded as weld defects and misjudged, and the accuracy of nondestructive testing is reduced. Therefore, the embodiment provides a weld image processing method, which can effectively solve the technical problem of low detection accuracy when performing nondestructive detection on a weld image.

As shown in fig. 1, a method for processing a weld image according to a first embodiment of the present invention includes the following steps:

s101, acquiring the inclination angle of the welding seam area in the welding seam image.

In the present embodiment, the weld image may be understood as an image containing the weld, such as an X-ray image of the weld, acquired by the image acquisition device. The weld seam region can be understood as the region in the weld seam image in which the weld seam is located. The angle of inclination is understood to be the angle at which the weld seam region is inclined.

It should be noted that the weld region in the weld image processing method in this embodiment needs to satisfy two characteristics: the welding seam area is approximately strip-shaped; the difference of the texture features of the weld seam region relative to the background is larger than a certain threshold value, the threshold value is not limited, and the texture features can be set by a person skilled in the art according to actual conditions.

The specific means for obtaining the inclination angle of the welding seam area of the welding seam image is not limited, for example, the welding seam image can be analyzed in the step to obtain the center line of the welding seam area, and then the center line is subjected to linear fitting to obtain the inclination angle of the fitted straight line; in the step, a rectangular coordinate system can be established by taking the upper left corner of the weld image as the origin of coordinates, the horizontal direction as the abscissa and the vertical direction as the ordinate. The measured inclination angle of the weld zone diagonal is then determined based on the coordinate system. And determining the actual inclination angle of the welding seam region according to the length and the width of the welding seam region. And finally, determining the inclination angle of the welding seam area according to the actually measured inclination angle and the actual inclination angle.

It is understood that the inclination angle may be an angle relative to the horizontal direction or an angle relative to the vertical direction, and is not limited herein. The inclination angle can be used for identifying the inclination angle of the welding seam area, and the welding seam image can be adjusted based on the inclination angle to obtain the welding seam area in a horizontal state or a vertical state, so that subsequent processing and analysis are facilitated. Specifically, the positive and negative values of the inclination angle can identify the inclination direction of the weld region in the weld image, and the value of the inclination angle can identify the inclination degree of the weld region in the weld image.

Illustratively, taking the upper left corner of the weld image as the origin of coordinates, the right direction of the horizontal direction as the positive direction of the horizontal axis, and the downward direction of the vertical axis as the positive direction of the vertical axis, a rectangular coordinate system is established, and taking the example of obtaining the inclination angle of the weld area relative to the horizontal direction: if the weld region in the weld image is in a horizontal state, the acquired inclination angle may be zero; if the weld region in the weld image is above the horizontal, the inclination angle may be negative; the inclination angle may be positive if the weld region in the weld image is below the horizontal.

And S102, correcting the welding seam image according to the inclination angle.

After acquiring the inclination angle of the weld region, this step may correct the weld image based on the inclination angle to obtain the weld region in a horizontal or vertical state, thereby facilitating subsequent processing analysis.

In this step, it is not limited to adjust the weld area in the weld image to be in a horizontal state or adjust the weld area in the weld image to be in a vertical state, and a person skilled in the art can select a corresponding means to process according to an actual scene.

Specifically, in the step of correcting the weld image, the weld image may be rotated according to the acquired inclination angle, so that the weld region in the weld image is in a horizontal state or a vertical state. It is understood that, when the welding seam image is rotated, in order to ensure the integrity of the image, blank areas inevitably appear on the edge portions of the rotated welding seam image, and this step may further remove the blank areas to obtain corrected welding seam areas. It is understood that this step may rotate only the weld image when correcting the weld region according to the inclination angle; the blank area formed by rotation may be removed after the rotation of the weld image, which is not limited herein.

S103, intercepting the corrected welding seam image according to a rectangular area to obtain the corrected welding seam area image, wherein the parameters of the rectangular area correspond to the texture features of the corrected welding seam image.

In this embodiment, the rectangular region may be understood as a minimum circumscribed rectangular region of the weld region determined by performing minimum rectangle analysis on the texture features of the corrected weld image. The corrected weld region image may be understood as including only the corrected weld image.

After correcting the weld image according to the inclination angle, the step may intercept the corrected weld image according to a rectangular region corresponding to the texture feature of the corrected weld image. Specifically, the area corresponding to the rectangular area obtained after the truncation may be taken as the corrected weld image, where the area corresponds to the rectangular area, and the corresponding position of the corrected weld image may be truncated according to a parameter of the rectangular area, such as a coordinate value. The corrected weld region image may be an image obtained by performing rotation correction on the weld region and removing the background. Nondestructive detection is carried out based on the corrected welding seam area image, and misjudgment of welding seam defects can be effectively reduced.

The parameters of the rectangular area can be obtained based on the texture features of the corrected welding seam image, the obtaining means of the parameters of the rectangular area is not limited, and for example, the parameters corresponding to the minimum circumscribed rectangle of the welding seam area can be directly determined based on the texture features of the corrected welding seam image; the texture features of the corrected weld image may also be further analyzed to determine parameters corresponding to the minimum circumscribed rectangle of the weld region.

The embodiment of the invention provides a welding seam image processing method, which can be used for correcting a welding seam image based on the inclination angle of a welding seam area in the welding seam image; and then intercepting the corrected welding seam image according to the rectangular area corresponding to the texture features of the corrected welding seam image to obtain the corrected welding seam area image, so that the non-welding seam area in the welding seam image is effectively excluded, and the accuracy of subsequently acquiring the defect features of the welding seam area is improved.

Example two

Fig. 2a is a schematic flow chart of a weld image processing method according to a second embodiment of the present invention, and the second embodiment is optimized based on the foregoing embodiments. In this embodiment, the obtaining of the inclination angle of the weld region in the weld image is further embodied as: acquiring a welding seam image; reducing the resolution of the welding seam image to obtain a small-size image; selecting a corresponding Gaussian laplacian operator template to perform image enhancement on the small-size image according to the pixel width of the welding seam region in the small-size image to obtain an enhanced image; carrying out self-adaptive binarization processing on the enhanced image to obtain a first binarized image; obtaining a central line of a welding seam area in the first binary image; and performing linear fitting on the central line to obtain the inclination angle of the central line.

Further, the present embodiment will also correct the weld image according to the inclination angle, further optimizing as: acquiring a welding seam image; rotating the welding seam image according to the inclination angle to obtain a rotated image; and removing the blank area of the rotated image to obtain a corrected welding seam image.

On the basis of the optimization, the determination of the parameters of the rectangular area is specifically optimized as follows: processing the corrected welding seam image by adopting an edge detection algorithm to obtain a texture image; and determining the coordinate value of the rectangular area according to the texture image. Please refer to the first embodiment for a detailed description of the present embodiment.

As shown in fig. 2, a weld image processing method provided by the second embodiment of the present invention includes the following steps:

s201, obtaining a weld image.

In acquiring the inclination angle of the bead region in the bead image, the present embodiment may acquire the bead image first. The means for acquiring the weld image is not limited, and may be input by the user or acquired by the image acquisition device.

S202, reducing the resolution of the welding seam image to obtain a small-size image.

In the present embodiment, the small-size image may be understood as an image formed after the resolution of the bead image is reduced, and the small-size image may be regarded as an image of a small size with respect to the bead image. And the size of the small-size image is reduced compared with the size of the welding seam image, and the small-size image is not limited and can be set according to actual requirements.

After the weld image is acquired, this step may reduce the resolution of the weld image to reduce the size of the weld image. The means for reducing the weld image is not limited, for example, the weld image may be sampled, for example, at equal intervals or by taking the average value of the pixels in the preset area as the pixel value of the preset area.

In the step, the resolution ratio of the welding seam image is reduced, so that the pixel width of the welding seam area is reduced to a reasonable range, and meanwhile, the reduction of the resolution ratio can remove much detail interference in the image, thereby facilitating the subsequent processing.

Further, before the resolution of the weld image is reduced to obtain a small-size image, the method further comprises the following steps:

preprocessing the welding seam image to obtain a first de-noising image; correspondingly, the reducing the resolution of the weld image to obtain a small-size image comprises the following steps:

and reducing the resolution of the first denoised image to obtain a small-size image.

The first de-noising image can be understood as an image obtained after noise in the weld image is filtered in the stage of obtaining the inclination angle of the weld area.

In this embodiment, the technical means of preprocessing used to obtain the first denoised image include, but are not limited to: median filtering, smoothing filtering, etc. The purpose of preprocessing the welding seam image is to eliminate various noise and other interference information in the welding seam image, improve the signal-to-noise ratio of the image and improve the processing efficiency of the welding seam image.

S203, selecting a corresponding Gaussian operator template to perform image enhancement on the small-size image according to the pixel width of the welding seam region in the small-size image to obtain an enhanced image.

After the small-size image is obtained, the Gaussian laplacian operator template matched with the pixel width is selected for image enhancement processing based on the pixel width of the welding seam region in the small-size image. The size of the selected laplacian of gaussian template is not limited herein, and those skilled in the art can select the size of the corresponding laplacian of gaussian template according to actual requirements, for example, the size may be twice the pixel width of the weld region.

In the step, the pixel width of a welding seam area in the small-size image can be firstly obtained, and then the matched Gaussian Laplacian operator template is selected according to the obtained pixel width to carry out image enhancement on the small-size image. The specific means for obtaining the pixel width of the weld region in the small-size image is not limited herein. Such as by a pixel width capture tool, or by a corresponding image processing algorithm.

Because the noise point has a certain influence on the edge detection, the image enhancement can be performed on the small-size image by adopting a Gaussian Laplacian operator, namely a LoG operator. LOG operator: the method is a double combination of Gauss and Laplace, namely an operator model integrating smoothness and edges, can improve the contrast of a welding seam area and a background, and is convenient for subsequent processing.

And S204, carrying out self-adaptive binarization processing on the enhanced image to obtain a first binarized image.

In this embodiment, the first binarized image may be understood as a binarized image obtained by performing adaptive binarization processing on the enhanced image.

After obtaining the enhanced image, the step may perform binarization processing on the enhanced image to segment a main region of the weld from the enhanced image.

Specific means of the adaptive binarization processing performed on the enhanced image in the step include but are not limited to: a cluster binarization algorithm, a maximum entropy threshold method and the like. The specific adaptive binarization method is not limited herein, and a person skilled in the art can select the method according to the specific situation of the weld image.

S205, obtaining the center line of the welding seam area in the first binary image.

In this embodiment, the centerline may be understood as a bone line that can represent the characteristics of the weld region. This embodiment can use the inclination angle of the center line as the inclination angle of the bead region.

After the first binarized image is obtained, the center line of the weld area of the first binarized image can be obtained in this step, that is, the weld area with a certain pixel width is refined into the center line with a single pixel width. And analyzing the central line of the welding seam area to determine the inclination angle of the welding seam area.

Specifically, a specific means for obtaining the center line of the weld seam region is not limited, and the center line may be obtained by selecting a corresponding method according to a specific shape of the weld seam region in the art, such as a morphology-based skeleton algorithm, a centroid method (i.e., taking a median value from the upper edge and the lower edge), and the like.

S206, performing straight line fitting on the central line to obtain the inclination angle of the central line.

After obtaining the center line of the weld region, this step may perform a straight line fitting on the center line to determine the inclination angle of the center line by analyzing the fitted straight line.

The specific means of the line fitting is not limited herein, and those skilled in the art can select a corresponding means according to the shape of the center line to perform the line fitting, for example, performing the line fitting by using the least square method.

After the straight line fitting is performed on the central line, the tangent value of the fitted straight line can be obtained in the step as the inclination angle of the central line. After acquiring the inclination angle of the center line, the present embodiment may correct the bead image based on the inclination angle.

And S207, acquiring a weld image.

When the weld image is corrected according to the inclination angle, the step may first acquire the weld image to correct the weld image. The weld image may be an image of the weld image obtained when the inclination angle of the weld region is obtained. The means for acquiring the weld image is not limited herein, such as the weld image acquired when the inclination angle of the weld region is acquired may be acquired; it is also possible to acquire a weld image reintroduced by the user.

And S208, rotating the welding seam image according to the inclination angle to obtain a rotated image.

After the weld image is obtained, the step may rotate the weld image based on the obtained inclination angle to obtain a rotated image. The rotation direction of the welding seam image can be determined according to the positive and negative of the inclination angle, and the rotation size of the welding seam image can be determined according to the absolute value of the inclination angle.

Further, before rotating the weld image according to the inclination angle to obtain a rotated image, the method further includes:

preprocessing the welding seam image to obtain a second de-noising image; correspondingly, the rotating the weld image according to the inclination angle to obtain a rotated image includes:

and rotating the second denoising image according to the inclination angle to obtain a rotated image.

The second denoised image can be understood as an image obtained after noise in the weld image is filtered in a weld image correction stage.

In this embodiment, the preprocessing means used to obtain the second denoised image includes, but is not limited to: median filtering, smoothing filtering, etc. The purpose of preprocessing the welding seam image is to eliminate various noise and other interference information in the welding seam image, improve the signal-to-noise ratio of the image and improve the processing efficiency of the welding seam image after subsequent interception and correction.

S209, removing the blank area of the rotated image to obtain a corrected welding seam image.

It is understood that, when the welding image is rotated, in order to ensure the integrity of the weld image, a black edge, i.e., a blank region, is inevitably present in the edge portion of the image after the rotation. The blank area may be considered as a non-significant area in the weld image, which may not be non-destructively detected. Therefore, the blank area in the rotated image can be removed in the step, and the corrected welding seam image is obtained.

The means for removing the blank area in the rotated image is not limited, and the position of the cut-out may be determined according to four corner points of the weld image in the rotated image, so as to remove the blank area in the rotated image.

And S210, processing the corrected welding seam image by adopting an edge detection algorithm to obtain a texture image.

After the corrected weld image is obtained, the embodiment may determine the minimum circumscribed rectangle of the weld region based on the corrected weld image, and then intercept the corrected weld image based on the determined minimum circumscribed rectangle.

First, in order to obtain the minimum circumscribed rectangle of the corrected weld image, this step may first obtain a texture image of the corrected weld image. Because the welding seam has defects and large surface roughness, a large amount of texture characteristics exist in the welding seam area. The texture of the welding seam area is completely different from the texture of the background area, and the texture characteristics of the welding seam area can be highlighted through an edge detection algorithm in the step, so that a texture image is obtained. The edge detection algorithm in this step includes but is not limited to: laplacian of Gaussian (LoG) and Sobel edge detection operators.

And S211, determining the coordinate value of the rectangular area according to the texture image.

After the texture image is obtained, the coordinate value of the rectangular area can be determined by performing minimum external rectangle analysis on the texture features in the texture image; the texture image can be further processed, and then the minimum circumscribed rectangle of the welding seam area is determined to obtain the coordinate value of the rectangular area. The rectangular area may encompass all of the features of the weld area. The specific means for performing the minimum rectangle analysis is not limited herein.

It is understood that, since the weld region is approximately strip-shaped, the coordinate values of the rectangular region obtained in this step may include only the coordinate values of the upper edge and the lower edge of the rectangle.

Further, according to the texture image, determining coordinate values of a rectangular area includes:

carrying out self-adaptive binarization processing on the texture image to obtain a second binarization image;

removing interference information of the second binary image to obtain a filtered image;

and determining the coordinate value of the rectangular region according to the welding seam region in the filtered image.

In order to improve the accuracy of acquiring the parameters of the rectangular region, when the coordinate values of the rectangular region are determined according to the texture image, the texture image may be subjected to adaptive binarization processing to obtain a second binarized image. The second binarized image may be understood as a binarized image obtained by performing adaptive binarization processing on the texture image. Adaptive binarization includes, but is not limited to: a cluster binarization algorithm, a maximum entropy threshold method and the like. The purpose of carrying out adaptive binarization processing on the texture image is as follows: the main area of the weld zone is more accurately segmented.

After the second binary image is obtained, the interference information in the second binary image can be removed. The interference information can be understood as interference particles in the second binarized image. The means for removing the interference information in the second binarized image is not limited herein. For example, the step can adopt morphology to eliminate interfering particles and particle filtering to filter out the interfering information.

The morphological interference elimination particles mainly adopt structural elements with proper sizes, and various interference particles existing in the texture image are removed through a series of morphological algorithms such as open operation and close operation. The main content of the particle filtering is to use an adaptive algorithm, take the pixel area of the largest particle in the texture image as a basic value, and use a proper coefficient to obtain the threshold value of the particle filtering. Particles with pixel areas below this threshold are all filtered out. When the interference information is removed, whether the particles are filtered or not can be judged according to a series of characteristics such as the size, the dimension, the shape and/or the inner holes of the particles.

S212, intercepting the corrected welding seam image according to the rectangular area to obtain the corrected welding seam area image.

The present embodiment is directed to a weld image processing method for obtaining a corrected weld region image, i.e., a region of interest (ROI) image, from a weld image, where the corrected weld region image only includes a weld region, and the method mainly includes two steps:

(1) rotation corrected weld X-ray image (i.e. weld image): the method mainly comprises the steps of preprocessing a welding seam image, reducing the size of the image, enhancing an LoG operator image, performing self-adaptive binarization, obtaining a central line of a welding seam area, obtaining an inclination angle theta of the image through linear fitting, and rotating the welding seam image by an angle theta.

(2) Obtaining an ROI (region of interest) by using texture features of a welding seam region: the method mainly comprises the steps of preprocessing a welding seam image, obtaining texture features of the image after rotation correction, self-adaptive binarization, eliminating interference particles in morphology, filtering the particles and intercepting an ROI (region of interest).

The purpose of the rotation correction of the X-ray image of the welding seam is to perform rotation correction on the oblique X-ray image of the welding seam so as to facilitate the extraction of a subsequent ROI (region of interest).

The main purpose of image preprocessing is to eliminate various kinds of interference information such as noise in the X-ray image of the weld joint and improve the signal-to-noise ratio of the image. The preprocessing algorithm mainly comprises median filtering, smooth filtering and the like.

The size of the image is reduced, the main purpose is to reduce the resolution of the image, so that the pixel width of a welding seam area is reduced to a reasonable range, meanwhile, the reduction of the resolution can remove much detail interference information in the image, and the subsequent LoG operator can conveniently and effectively perform image enhancement processing.

The LoG operator adopts a one-dimensional operator template, and the size of the one-dimensional operator template is matched with the pixel width of the welding line region of the low-size image obtained in the last step. After the algorithm template processing, the contrast of a welding seam area and a background is obviously improved, and subsequent binaryzation and image segmentation are facilitated.

The self-adaptive binarization method comprises a common clustering binarization algorithm, a maximum entropy threshold method and other self-adaptive binarization algorithms, and the selection of the algorithm is selected according to the specific condition of the weld image. The main purpose of binarization is to segment the main area of the weld from the image.

The main contents for obtaining the center line of the welding seam area are as follows: and thinning the welding seam area with a certain pixel width into a bone line with a single pixel width. The main selectable methods include a morphology-based skeleton algorithm, a gravity center method (taking the median value of the upper edge and the lower edge), and the like.

The main purpose of the straight line fitting to obtain the inclination angle theta of the image is: the inclination angle θ of the bone line of a single pixel width is obtained. The inclination angle theta is the inclination angle of the original image weld zone.

The above-described flow is performed based on the image after size reduction, and the image subjected to the rotation correction processing is directed to the original image. And after the original image is subjected to rotation correction according to the obtained inclination angle theta, deleting blank areas in the rotated image.

The method is characterized in that the ROI is obtained by using the texture features of the weld joint region based on the corrected weld joint image, and when the ROI is obtained, the image is preprocessed mainly to eliminate various noise and other interference information in the X-ray image of the weld joint and improve the signal-to-noise ratio of the image. The preprocessing algorithm mainly comprises median filtering, smooth filtering and the like.

The main reason for obtaining the texture features of the corrected weld image is that the texture of the weld region is completely different from that of the background region, and the X-ray image of the weld region has a large number of texture features due to the defects and the large surface roughness of the weld. These textures can be visualized by edge operator algorithms. The commonly used edge detection algorithm includes a LoG operator and a Sobel edge detection operator.

Morphological elimination of interfering particles and particle filtering, there are more interfering particles in the image that need to be further removed. The morphological interference elimination particles mainly adopt structural elements with proper sizes, and various interference particles existing in the second binary image corresponding to the texture image are removed through a series of morphological algorithms such as open operation and close operation. The main content of the particle filtering is to use an adaptive algorithm, take the pixel area of the largest particle in the image as a basic value, and use a proper coefficient to obtain the threshold value of the particle filtering. Particles with pixel areas below this threshold are all filtered out.

And (3) cutting an ROI (region of interest): after some of the above processing, essentially only the weld area remains in the image with a large "grain". By acquiring the upper and lower edges of the smallest rectangle in which the "grain" is located, the ROI region is obtained by cutting out the region between the upper and lower edges in the rotation-corrected image.

The ROI of the X-ray image of the welding seam is obtained by the technical means, the technical problems that the welding seam image is inclined and has more interference information can be solved, the non-welding seam area in the X-ray image of the welding seam is effectively excluded, and the accuracy of obtaining defect characteristics by algorithms such as follow-up image segmentation and characteristic identification is effectively improved.

The following describes an exemplary embodiment of the present invention:

taking an X-ray image of a weld of an aluminum-lithium alloy laser welding part as an example for explanation, an original image, i.e., an X-ray image of the original weld, is obtained first, and the image is an image of the weld in this embodiment. Fig. 2b is a schematic diagram of a weld image provided by the second embodiment of the present invention, and as shown in fig. 2b, the weld image is in a tilted state, and the original image size is 1024 × 160.

The weld image is subjected to reduction processing, that is, the size of the weld image is reduced, and a low-size image, that is, a small-size image in this embodiment, is obtained. Fig. 2c is a schematic diagram of a small-size image provided by the second embodiment of the present invention, as shown in fig. 2c, the size of the small-size image is 114 × 18, and the pixel width of the weld region in the small-size image is reduced by about 7 pixels compared to that in fig. 2 b.

Selecting a template of a Log operator with a proper size, and performing image enhancement processing on the small-size image in fig. 2c, wherein fig. 2d shows an enhanced image schematic diagram provided by the second embodiment of the invention, the enhanced image is the Log operator enhanced image, as shown in fig. 2d, the contrast between a white area where a welding seam is located and the background is greatly improved, the background is basically black, and the size of the enhanced image is 114 × 18.

And (3) performing binarization processing on the enhanced image in fig. 2d by using an adaptive binarization algorithm, wherein fig. 2e shows a schematic diagram of a first binarized image provided by the second embodiment of the present invention, and as shown in fig. 2e, the size of the first binarized image is 114 × 18.

Selecting a morphological bone algorithm, processing the graph 2e to obtain a bone center line of a region where the weld is located, and as shown in fig. 2f, showing a center line image schematic diagram provided by the second embodiment of the present invention, as shown in fig. 2f, the size of the center line image is 114 × 18, the center line image can represent the characteristic information of the weld region, and the inclination angle of the center line can be used as the inclination angle of the weld region.

And (3) performing straight line fitting of a least square method on the image 2f to obtain a straight line equation (note: the origin of the image coordinate system is in the upper left corner, the horizontal right direction is the positive direction of the X axis, and the vertical downward direction is the square of the Y axis):

0.0525*X+0.9986Y＝10.808

according to the equation of the straight line, the inclination angle is 3 degrees, the original image in fig. 2b is rotated by 3 degrees, and a rotated image is obtained, and fig. 2g shows a schematic diagram of the rotated image provided by the second embodiment of the invention. As shown in fig. 2g, the size of the rotated image is 1031 × 214. Because the original image is rectangular, after the rotation processing, in order to ensure the integrity of the image, the rotated image inevitably has black edges.

In order to only reserve the weld joint region and remove the black edge in the rotated image after the rotated image is obtained, fig. 2h shows a schematic diagram of the corrected weld joint image provided by the second embodiment of the present invention, as shown in fig. 2h, the size of the corrected weld joint image is 1014 × 106, and the corrected weld joint image with the black edge removed, i.e., the blank region, contains less background region than the weld joint image in fig. 2 b.

To further remove the background area in fig. 2h, an analysis may be performed based on the texture features of the weld area. For example, the Sobel operator can be used to process the corrected weld image in fig. 2h, so as to obtain the texture features of the corrected weld image. Fig. 2i shows a schematic diagram of a texture image according to a second embodiment of the present invention, where the texture image is obtained by processing the corrected weld image in fig. 2h with a Sobel operator and then performing histogram equalization processing. Since the image processed by the Sobel operator has low gray value and is not dark, the image processed by the Sobel operator is highlighted for display by histogram equalization, and the size of the texture image is 1014 × 106.

And carrying out self-adaptive binarization processing on the texture image to obtain a second binarization image. Fig. 2j shows a schematic diagram of a second binarized image according to a second embodiment of the present invention, and the size of the second binarized image is 1014 × 106, as shown in fig. 2 j.

And performing morphology degranulation (morphology on operation and off operation) and grain filtering on the texture image to obtain a filtered image, and fig. 2k shows a schematic diagram of the filtered image provided by the second embodiment of the present invention, and as shown in fig. 2k, the size of the filtered image is 1014 × 106.

And carrying out minimum rectangle analysis on the filtered image to obtain a minimum rectangle of the filtered image. Fig. 2l shows a schematic diagram of a minimum rectangle provided by the second embodiment of the present invention, as shown in fig. 2l, the minimum rectangle is a gray line area in two areas of fig. 2, an upper edge of the minimum rectangle is at a coordinate value of 32, and a lower edge of the minimum rectangle is at a coordinate value of 96. And (5) intercepting images in the vertical direction between 32 and 96 in the corrected weld image shown in FIG. 2h to obtain a corrected weld area image. Since the bead region is elongated, only the upper and lower edges of the minimum rectangle, i.e., the coordinate values in the vertical direction, may be considered when performing the minimum rectangle analysis.

Fig. 2m shows a schematic diagram of a corrected weld region image provided by the second embodiment of the present invention, and as shown in fig. 2m, the size of the corrected weld region image is 1014 × 60, which only includes the weld region, and the weld region is in a horizontal state.

Fig. 2n is a schematic diagram illustrating a principle of removing black borders according to a second embodiment of the present invention, where as shown in fig. 2n, an oblique white area is an original image, i.e., a weld image, and a black area is a black border generated after the original image is rotated. The range enclosed by the dotted line is the final intercepted image, i.e. the corrected weld image. The first dotted line side 1, the second dotted line side 2, the third dotted line side 3 and the fourth dotted line side 4 are four sides of a dotted line frame, and the first corner point 5, the second corner point 6, the third corner point 7 and the fourth corner point 8 are four corner points of a white area in the figure. The principle of removing the black border (i.e. the blank area) is: the first dotted line edge 1 and the point first corner point 5 of the dotted line frame are on the same line, the second dotted line edge 2 and the second corner point 6 are on the same line, the third dotted line edge 3 and the third corner point 7 are on the same line, and the fourth dotted line edge 4 and the fourth corner point 8 are on the same line, so that the specific position of the dotted line frame is obtained.

The welding seam image processing method provided by the second embodiment of the invention embodies the operation of obtaining the inclination angle, the operation of correcting the welding seam image and the rectangular area parameter determining operation. The method comprises the steps of firstly reducing the resolution of a welding seam image to obtain a small-size image, and secondly selecting a corresponding LOG operator module to perform image enhancement on the small-size image according to the pixel width of a welding seam region in the small-size image to obtain an enhanced image; then, carrying out self-adaptive binarization processing on the enhanced image to obtain a first binarized image so as to obtain a central line of a welding line area in the first binarized image; and performing straight line fitting on the central line to obtain the inclination angle of the central line. And rotating the welding seam image according to the inclination angle of the central line to obtain a rotated image, and then removing a blank area of the rotated image to obtain a corrected welding seam area. And determining coordinate values of the rectangular area according to the texture features of the corrected welding seam area to intercept the corrected welding seam image and obtain the corrected welding seam area image, so that the non-welding seam area in the welding seam image is more accurately removed, and the obtained welding seam area is the corrected welding seam area, thereby providing convenience for performing defect analysis on the subsequent welding seam area.

EXAMPLE III

Fig. 3 is a schematic structural diagram of a weld image processing apparatus according to a third embodiment of the present invention, where the apparatus is applicable to a situation of processing a weld image, and in particular, the apparatus is applicable to a situation of processing an X-ray image of a weld to improve accuracy of nondestructive testing, where the apparatus may be implemented by software and/or hardware and is generally integrated on a terminal device.

As shown in fig. 3, the bead image processing apparatus includes: an acquisition module 31, a correction module 32 and an interception module 33;

the acquiring module 31 is used for acquiring an inclination angle of a weld region in a weld image;

a correction module 32 for correcting the weld image according to the inclination angle;

and the intercepting module 33 is configured to intercept the corrected weld image according to a rectangular region, so as to obtain a corrected weld image, where parameters of the rectangular region correspond to texture features of the corrected weld image.

In the present embodiment, the weld image processing apparatus first acquires the inclination angle of the weld region in the weld image by the acquisition module 31; secondly, correcting the welding seam image according to the inclination angle through a correction module 32; and finally, intercepting the corrected welding seam image according to a rectangular area through an intercepting module 33 to obtain the corrected welding seam area image, wherein the parameters of the rectangular area correspond to the texture features of the corrected welding seam image.

The embodiment provides a weld image processing apparatus capable of correcting a weld image based on an inclination angle of a weld region in the weld image; and then intercepting the corrected welding seam image according to the rectangular area corresponding to the texture features of the corrected welding seam image to obtain the corrected welding seam area image, so that the non-welding seam area in the welding seam image is effectively excluded, and the accuracy of subsequently acquiring the defect features of the welding seam area is improved.

Further, the obtaining module 31 is specifically configured to: acquiring a welding seam image; reducing the resolution of the welding seam image to obtain a small-size image; selecting a corresponding Gaussian laplacian operator template to perform image enhancement on the small-size image according to the pixel width of the welding seam region in the small-size image to obtain an enhanced image; carrying out self-adaptive binarization processing on the enhanced image to obtain a first binarized image; obtaining a central line of a welding seam area in the first binary image; and performing linear fitting on the central line to obtain the inclination angle of the central line.

On the basis of the optimization, the welding seam image processing device further comprises: the first denoising module is used for preprocessing the welding seam image to obtain a first denoising image before reducing the resolution of the welding seam image to obtain a small-size image; correspondingly, when the resolution of the weld image is reduced and a small-size image is obtained, the obtaining module 31 is specifically configured to: and reducing the resolution of the first denoised image to obtain a small-size image.

Based on the above technical solution, the correction module 32 is specifically configured to: acquiring a welding seam image; rotating the welding seam image according to the inclination angle to obtain a rotated image; and removing the blank area of the rotated image to obtain a corrected welding seam image.

Further, the weld image processing apparatus further includes: the parameter determining module is used for processing the corrected welding seam image by adopting an edge detection algorithm to obtain a texture image; and determining the coordinate value of the rectangular area according to the texture image.

Further, the parameter determination module is specifically configured to: processing the corrected welding seam image by adopting an edge detection algorithm to obtain a texture image; carrying out self-adaptive binarization processing on the texture image to obtain a second binarization image; removing interference information of the second binary image to obtain a filtered image; and determining the coordinate value of the rectangular region according to the welding seam region in the filtered image.

Based on above-mentioned technical scheme, welding seam image processing apparatus still includes: the second denoising module is used for preprocessing the welding seam image to obtain a second denoising image before rotating the welding seam image according to the inclination angle to obtain a rotated image; correspondingly, when the correction module 32 rotates the weld image according to the inclination angle to obtain a rotated image, the correction module is specifically configured to: and rotating the second denoising image according to the inclination angle to obtain a rotated image.

The welding seam image processing device can execute the welding seam image processing method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.

Example four

Fig. 4 is a schematic structural diagram of a terminal device according to a fourth embodiment of the present invention. As shown in fig. 4, a terminal device provided in the fourth embodiment of the present invention includes: one or more processors 41 and storage 42; the processor 41 in the terminal device may be one or more, and one processor 41 is taken as an example in fig. 4; storage 42 is used to store one or more programs; the one or more programs are executed by the one or more processors 41, so that the one or more processors 41 implement the weld image processing method according to any one of the embodiments of the present invention.

The terminal device may further include: an input device 43 and an output device 44.

The processor 41, the storage device 42, the input device 43 and the output device 44 in the terminal equipment may be connected by a bus or other means, and the connection by the bus is exemplified in fig. 4.

The storage device 42 in the terminal device is used as a computer-readable storage medium for storing one or more programs, which may be software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the weld image processing method provided in one or two embodiments of the present invention (for example, the modules in the weld image processing device shown in fig. 3, including the acquiring module 31, the correcting module 32, and the intercepting module 33). The processor 41 executes various functional applications and data processing of the terminal device by executing software programs, instructions and modules stored in the storage device 42, that is, implements the weld image processing method in the above-described method embodiment.

The storage device 42 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal device, and the like. Further, the storage 42 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, storage 42 may further include memory located remotely from processor 41, which may be connected to the device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The input means 43 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function control of the terminal device. The output device 44 may include a display device such as a display screen.

And, when the one or more programs included in the above-mentioned terminal device are executed by the one or more processors 41, the programs perform the following operations:

acquiring an inclination angle of a welding seam region in a welding seam image; correcting the weld image according to the inclination angle; and intercepting the corrected welding seam image according to the rectangular area to obtain the corrected welding seam image, wherein the parameters of the rectangular area correspond to the texture features of the corrected welding seam image.

EXAMPLE five

An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program is used, when executed by a processor, to execute a weld image processing method, where the method includes:

Optionally, the program may be further configured to perform a weld image processing method according to any embodiment of the present invention when executed by the processor.

Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a Read Only Memory (ROM), an Erasable Programmable Read Only Memory (EPROM), a flash Memory, an optical fiber, a portable CD-ROM, an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. A computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take a variety of forms, including, but not limited to: an electromagnetic signal, an optical signal, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, Radio Frequency (RF), etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A weld image processing method is characterized by comprising the following steps:

correcting the weld image according to the inclination angle;

2. The method of claim 1, wherein the obtaining the inclination angle of the weld region in the weld image comprises:

acquiring a welding seam image;

reducing the resolution of the welding seam image to obtain a small-size image;

selecting a corresponding Gaussian laplacian operator template to perform image enhancement on the small-size image according to the pixel width of the welding seam region in the small-size image to obtain an enhanced image;

carrying out self-adaptive binarization processing on the enhanced image to obtain a first binarized image;

obtaining a central line of a welding seam area in the first binary image;

and performing linear fitting on the central line to obtain the inclination angle of the central line.

3. The method of claim 2, further comprising, before reducing the resolution of the weld image to obtain a small-size image:

4. The method of claim 1, wherein the correcting the weld image according to the inclination angle comprises:

acquiring a welding seam image;

rotating the welding seam image according to the inclination angle to obtain a rotated image;

and removing the blank area of the rotated image to obtain a corrected welding seam image.

5. The method of claim 1, wherein the determining of the parameters of the rectangular region comprises:

processing the corrected welding seam image by adopting an edge detection algorithm to obtain a texture image;

and determining the coordinate value of the rectangular area according to the texture image.

6. The method according to claim 5, wherein the determining coordinate values of the rectangular region according to the texture image comprises:

7. The method of claim 4, further comprising, prior to rotating the weld image according to the tilt angle to obtain a rotated image:

8. A weld image processing apparatus, characterized by comprising:

9. A terminal device, comprising:

one or more processors;

storage means for storing one or more programs;

when executed by the one or more processors, cause the one or more processors to implement the weld image processing method of any one of claims 1-7.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the weld image processing method according to any one of claims 1 to 7.