CN115018816A - Image processing method, device, equipment and storage medium for real-time detection of molten pool width - Google Patents

Abstract

The invention is applicable to the technical field of on-line detection of welding quality, and the invention provides an image processing method, device, equipment and storage medium for real-time detection of molten pool width. The processing method includes image acquisition, median filtering, threshold segmentation, edge smoothing and calculation The width of the molten pool, through the method provided by the present invention, the information of the width of the molten pool can be obtained clearly and accurately, and it is convenient to provide a reliable basis for the subsequent detection of the welding quality.

Description

An image processing method, device, equipment and storage medium for real-time detection of molten pool width

technical field

The invention belongs to the technical field of on-line detection of welding quality, and in particular relates to an image processing method, device, equipment and storage medium for real-time detection of the width of a molten pool.

Background technique

In order to realize the online control of welding quality status, the problem that must be solved is the real-time detection of welding pool information. With the in-depth study of GMAW (Gas Shielded Welding) welding, many researchers have extracted rich molten pool information from different angles. By processing and analyzing the information, it can be used to characterize the quality state of the weld to a certain extent. According to According to the different acquisition methods of arc and molten pool information, the online detection methods of welding quality mainly include arc acoustic sensing detection method, temperature field sensing detection method, oscillation frequency detection sensing method, and visual sensing detection method.

From the research status of welding pool detection technology at home and abroad, it can be seen that the visual sensing detection method does not contact the welding circuit, and there is no mutual influence with the welding process, and the processing of the obtained molten pool image can be intuitive. It has gradually become the mainstream trend in molten pool monitoring technology.

During GMAW (Gas Shielded Welding) welding, the arc light interferes with the molten pool area, and how to reliably extract the width characteristic parameters of the frontal molten pool has become an urgent problem to be solved.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide an image processing method, device, equipment and storage medium for detecting the width of the molten pool in real time, aiming to solve the problem of how to solve the problem of how to solve the problem that the arc light interferes with the molten pool area during the GMAW (gas shielded welding) welding process. The problem of reliably extracting the width characteristic parameter of the frontal melt pool.

In one aspect, the present invention provides an image processing method capable of detecting the width of a molten pool in real time, comprising the following steps:

Image collection: Based on the visual sensing system, collect the frontal molten pool image during the welding process;

Median filter: use median filter to denoise the original molten pool image;

Threshold segmentation: grayscale the molten pool image, and use the threshold segmentation method in the grayscale image to extract the region where the molten pool is located in the molten pool image:

Edge smoothing: perform edge smoothing processing on grayscale images to obtain grayscale images with clear edges;

Calculation of molten pool width: The grayscale image with clear edges is processed by the mean value calculation method, and the data of molten pool width is obtained.

Further, in the step threshold segmentation, the molten pool area is divided into an upper arc area, an arc area and a lower arc area.

Further, the image threshold of the arc region is selected to segment all the images.

Further, the method for calculating the average value in the calculation of the molten pool width in the step includes:

Select the integer row pixel value of the molten pool part in the grayscale molten pool image with clear edges;

Calculate the number of pixels in each row of pixels that are greater than the threshold selected in the step threshold segmentation to obtain the calculation result;

Take the average of the calculation results to obtain the molten pool width data. The specific calculation formula is:

In formula (1), d is the width of the molten pool, n is the number of pixels in each line of image pixels greater than the threshold, and B=r2-r1 lines are counted, and γ is the camera calibration data, that is, the world coordinate length value of the unit pixel.

Further, in the step of edge smoothing, it specifically includes: using an open operation to smooth the edge of the grayscale image.

On the other hand, the present invention also provides an image processing device capable of detecting the width of the molten pool in real time, comprising:

The image acquisition module is used to collect the frontal molten pool image during the welding process based on the visual sensing system;

The median filter module is used to denoise the original molten pool image by the median filter method;

The threshold segmentation module is used to grayscale the molten pool image, and the threshold segmentation method is used in the grayscale image to extract the region where the molten pool is located in the molten pool image:

The edge smoothing module is used to perform edge smoothing processing on grayscale images to obtain grayscale images with clear edges;

The molten pool width calculation module is used to process the grayscale images with clear edges by using the mean calculation method to obtain the molten pool width data.

In another aspect, the present invention also provides an image processing device capable of detecting the width of a molten pool in real time, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, the processing The steps of the method according to claims 1-5 are implemented when the computer executes the computer program.

In another aspect, the present invention also provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, implements the steps of the method according to claims 1-5 .

Beneficial effects of the present invention: Aiming at the problem that the arc light interferes with the molten pool area during GMAW (gas shielded welding) welding, the method provided by the present invention can clearly and accurately obtain the width data of the front molten pool, which is convenient for subsequent butt welding. The detection of seam quality status provides a reliable basis.

Description of drawings

1 is a flowchart of an image processing method for detecting the width of a molten pool in an embodiment of the present invention;

FIG. 2 is a structural diagram of an image processing device for detecting the width of a molten pool according to an embodiment of the present invention.

FIG. 3 is a structural diagram of an image processing device for detecting the width of a molten pool according to an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

The specific implementation of the present invention is described in detail below in conjunction with specific embodiments:

Example 1:

As shown in FIG. 1 , Embodiment 1 of the present invention provides an image processing method for detecting the width of a molten pool in real time, including the following steps:

Step S101 , collecting an image: based on a visual sensing system, collect a frontal molten pool image during the welding process.

Step S102, median filtering: using median filtering method to denoise the original molten pool image.

In the welding process, the existence of noise will have a great impact on the extraction of molten pool features, which will increase the difficulty of image processing and reduce the accuracy of processing results. Median filtering is a nonlinear smoothing method to reduce edge blur. The median value of brightness in the neighborhood is not affected by individual noise burrs, and it eliminates impact noise quite well. denoising processing.

The specific process of median filtering: For a finite set of real numbers, the middle value after sorting is its median. The size of the set is usually odd to guarantee the uniqueness of the median.

In the melt pool image processing of the embodiment of the present invention, a 3*3 matrix is taken, and there are 9 pixels in it, the 9 pixels are sorted, and finally the center point of the matrix is assigned as the median value of the nine pixels, The median is calculated as follows:

g=median[f(x-1,y-1)+f(x,y-1)+f(x+1,y-1)+f(x-1,y)+f(x,y) +f(x+1,y)+f(x-1,y+1)+f(x,y+1)+f(x+1,y+1)] (2)

In formula (2), g is the median of 9 pixels in the 3*3 matrix.

Step S103, threshold segmentation: grayscale the molten pool image, and use the threshold segmentation method in the grayscale image to extract the region where the molten pool is located in the molten pool image:

During the welding process, due to the existence of arc light, the gray value of the molten pool area in the image is not consistent. The brightest part of the original image of the molten pool is the arc area, and the area covered by the arc and the areas with weaker brightness above and below the arc belong to the molten pool area. From the grayscale histogram, it can be concluded that the closer to the arc, the higher the grayscale value of the image.

When using the threshold segmentation method to process the image, because the arc area is wider and brighter than the other two areas, the pixel value of the image is larger, because the final melt pool width is calculated as the width of the melt pool at the widest point. value, where the image threshold of the arc region is used to segment all images.

Threshold segmentation formula:

In formula (3), f(x, y) is all pixels of the original image, g(x, y) is the image after threshold segmentation, and T is the threshold.

Step S104 , edge smoothing: use the open operation to perform edge smoothing processing on the grayscale image to obtain a grayscale image with clear edges.

The commonly used edge smoothing methods are erosion and expansion algorithms, which belong to morphological methods. The basic principle is to use structural elements to erode and expand the molten pool area, reduce or enlarge the original image, and then pass the Taking the difference, the contour image of the molten pool can be obtained. Erosion followed by dilation is an important morphological transformation called opening.

The opening operation of image X with respect to structuring element B is denoted as X·B, then

为腐蚀运算，⊕为膨胀运算，结构元素B选择3*3的矩阵。In formula (4),

For the erosion operation, ⊕ for the expansion operation, and the structuring element B selects a 3*3 matrix.

Step S105, calculation of the width of the molten pool: adopt the mean value calculation method to process the grayscale image with clear edges, and obtain the width of the molten pool. The specific steps are as follows:

Step S1051, selecting integer row pixel values of the molten pool part in the grayscale molten pool image with clear edges;

Step S1052, calculating the number of pixels in each row of pixels that is greater than the threshold selected in the step threshold segmentation, to obtain a calculation result;

Step S1053, taking the average value of the calculation results to obtain molten pool width data, and the specific calculation formula is:

In formula (1), d is the width of the molten pool, n is the number of pixels in each line of image pixels greater than the threshold, and B=r2-r1 lines are counted, and γ is the camera calibration data, that is, the world coordinate length value of the unit pixel.

Embodiment 2:

Fig. 2 shows the structure of the image processing device for real-time detection of molten pool width provided by the second embodiment of the present invention:

The image acquisition module 201 is used to collect the frontal molten pool image in the welding process based on the visual sensing system;

The median filtering module 202 is used for denoising the original molten pool image by the median filtering method;

The threshold segmentation module 203 is used to grayscale the molten pool image, and use the threshold segmentation method in the grayscale image to extract the region where the molten pool is located in the molten pool image:

an edge smoothing module 204, configured to perform edge smoothing processing on the grayscale image to obtain a grayscale image with clear edges;

The molten pool width calculation module 205 is used for processing the grayscale image with clear edge by adopting the mean value calculation method to obtain the molten pool width data.

In the embodiment of the present invention, each module in the image processing device for real-time detection of the width of the molten pool can be implemented by corresponding hardware or software units, and each unit can be an independent software and hardware unit, or can be integrated into a software and hardware unit, This is not intended to limit the present invention.

Embodiment three:

FIG. 3 shows the structure of the image processing device for real-time detection of the width of the molten pool provided by the third embodiment of the present invention. For the convenience of description, only the parts related to the embodiment of the present invention are shown, including:

The image processing apparatus 3 for detecting the width of the molten pool in real time according to the embodiment of the present invention includes a processor 30 , a memory 31 , and a computer program 32 stored in the memory 31 and running on the processor 30 . When the processor 30 executes the computer program 32 , the steps in the above-mentioned embodiment of the image processing method for detecting the width of the molten pool in real time are implemented, for example, steps S101 to S105 shown in FIG. 1 . Alternatively, when the processor 30 executes the computer program 32, the functions of each module in the above-mentioned embodiment of the image processing apparatus for real-time detection of the width of the molten pool are implemented, for example, the functions of the modules 201 to 205 shown in FIG. 2 .

Embodiment 4:

In an embodiment of the present invention, a computer-readable storage medium is provided, and the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned embodiment of the image processing method for detecting the width of a molten pool in real time is implemented steps, for example, steps S101 to S105 shown in FIG. 1 . Alternatively, when the computer program is executed by the processor, the functions of the modules in the above device embodiments, for example, the functions of the modules 201 to 205 shown in FIG. 2 , are implemented.

The computer-readable storage medium of the embodiments of the present invention may include any entity or device capable of carrying computer program codes, recording medium, for example, memory such as ROM/RAM, magnetic disk, optical disk, flash memory, and the like.

The above descriptions are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention shall be included in the protection of the present invention. within the range.

Claims (8)

1. an image processing method of real-time detection of molten pool width, is characterized in that, described method comprises the following steps: Image collection: Based on the visual sensing system, collect the frontal molten pool image during the welding process; Median filter: use median filter to denoise the original molten pool image; Threshold segmentation: grayscale the molten pool image, and use the threshold segmentation method in the grayscale image to extract the region where the molten pool is located in the molten pool image: Edge smoothing: perform edge smoothing processing on grayscale images to obtain grayscale images with clear edges; Calculation of molten pool width: The grayscale image with clear edges is processed by the mean value calculation method, and the data of molten pool width is obtained. 2 . The image processing method for real-time detection of molten pool width according to claim 1 , wherein, in the step threshold segmentation, the molten pool region is divided into an arc upper region, an arc region and an arc lower region. 3 . 3 . The image processing method for real-time detection of molten pool width according to claim 2 , wherein the image threshold value of the arc region is selected to divide all the images. 4 . 4. the image processing method of real-time detection of molten pool width according to claim 3, is characterized in that, the method for taking average value in the calculation of molten pool width of described step comprises: Select the integer row pixel value of the molten pool part in the grayscale molten pool image with clear edges; Calculate the number of pixels in each row of pixels that are greater than the threshold selected in the step threshold segmentation to obtain the calculation result; Take the average of the calculation results to obtain the molten pool width data. The specific calculation formula is:

In the above formula, d is the width of the molten pool, n is the number of pixels in each line of image pixels greater than the threshold, and a total of B=r2-r1 lines are counted, and γ is the camera calibration data, that is, the world coordinate length value of the unit pixel. 5 . The image processing method for real-time detection of molten pool width according to claim 1 , wherein in the step of smoothing the edge, it specifically comprises: using an open operation to smooth the edge of the grayscale image. 6 . 6. An image processing device for real-time detection of molten pool width, wherein the device comprises: The image acquisition module is used to collect the frontal molten pool image during the welding process based on the visual sensing system; The median filter module is used to denoise the original molten pool image by the median filter method; The threshold segmentation module is used to grayscale the molten pool image, and the threshold segmentation method is used in the grayscale image to extract the region where the molten pool is located in the molten pool image: The edge smoothing module is used to perform edge smoothing processing on grayscale images to obtain grayscale images with clear edges; The molten pool width calculation module is used to process the grayscale images with clear edges by using the mean calculation method to obtain the molten pool width data. 7. an image processing device for detecting the width of molten pool in real time, comprising a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor executes the A computer program implementing the steps of the method as claimed in any one of claims 1 to 5. 8. A computer-readable storage medium storing a computer program, wherein the computer program implements the steps of the method according to any one of claims 1 to 5 when the computer program is executed by a processor .

Images