Contour search-based molten pool width extraction method and system

Abstract

This invention provides a method and system for extracting molten pool width based on contour search. First, the original molten pool image is subjected to grayscale and binarization processing to obtain a binarized molten pool image; then, the image edge points are extracted, denoted as the contour point pixel set N(x... b ,y b Let M(x) be the set of pixels with a value of 0. a ,y a Then iterate through and calculate M(x); a ,y a Each point in the set N(x) is connected to the set N(x). b ,y b The sum of the Euclidean distances of all points in q is denoted as the set [q]. z ]; then find the set [q z The largest value in the [data] is denoted as q. k Record q at this time k The corresponding set M(x) a ,y a Given pixel K(x,y) in the set N(x,y), calculate the relationship between pixel K(x,y) and the set N(x,y). a ,y b The Euclidean distance of each point in the matrix is ​​calculated, and the minimum value is determined. Finally, the molten pool width D is calculated based on the minimum value of the Euclidean distance. Thus, the method of this invention can quickly and accurately extract the molten pool width in the metal laser directional energy deposition process, which is beneficial for controlling the molten pool morphology and improving the quality of the clad part.

Description

Technical Field

[0001] This invention relates to the field of metal laser additive manufacturing technology, and in particular to the extraction of molten pool width therein, specifically to a method and system for extracting molten pool width based on contour search. Background Technology

[0002] In laser directional energy deposition (DED), the stability of the molten pool represents the stability of the forming process, and controlling the molten pool morphology is a key factor in ensuring the quality of laser-clad parts. Accurate extraction of the molten pool edge is a prerequisite for extracting its geometric information during real-time monitoring and printing. The molten pool width information reflects its characteristics better than other information; therefore, monitoring the molten pool width during DED is crucial.

[0003] Existing technologies attempt to detect molten pool width using machine learning-based algorithms. For example, historical molten pool images are labeled and trained to create network models for detecting molten pool size, such as convolutional neural networks or segmentation networks. These models then perform real-time detection output on the input molten pool image, as exemplified by the detection methods proposed in CN115170545A and CN113554587A. However, we observe that using network models for molten pool size detection faces several challenges. First, the accuracy of the training data and the model itself is limited. The diverse and highly random features of molten pool morphology and size present significant obstacles in data labeling and feature analysis; preparing accurate labeling data remains a challenge. Second, interference from splashes and arcs during on-site molten pool detection leads to issues with image clarity, complexity, and angle, resulting in substantial errors in image recognition and size identification.

[0004] Another existing method for detecting the geometric dimensions of a molten pool is based on image processing. For example, patent application CN115018816A proposes a method for molten pool identification and extraction based on median filtering and threshold segmentation. Then, it uses an averaging method to process the grayscale image with clear edges to obtain the molten pool width data. Patent application CN115187567A proposes, after binarizing the molten pool image to obtain high-pixel-value and low-pixel-value region images, calculating the center points of the bright and dark regions of the molten pool. Based on these center points, it calculates the direction angle and direction line of the molten pool, then calculates the perpendicular line of the molten pool based on the direction line, and finally, combines this with the segmented molten pool image to calculate the width of the molten pool.

[0005] Existing technology:

[0006] Patent Document 1: CN115170545A A method for dynamic molten pool size detection and forming direction determination

[0007] Patent Document 2: CN113554587A A method and system for extracting geometric features from molten pool images based on deep learning

[0008] Patent Document 3: CN115018816A An image processing method, apparatus, device and storage medium for real-time detection of molten pool width

[0009] Patent Document 4: CN115187567A A method for detecting the forming direction and width of a molten pool in metal additive manufacturing Summary of the Invention

[0010] The purpose of this invention is to address the problem that interference such as spatter, noise, and arc light makes it difficult to accurately extract the width of the molten pool. This invention provides a method and system for extracting the width of the molten pool based on contour search, which eliminates the influence of noise such as spatter and burrs in the molten pool, and quickly and accurately extracts the width of the molten pool from the molten pool image. This provides accurate support for the shape control of the molten pool and ensures the quality of laser cladding processed parts.

[0011] According to a first aspect of the present invention, a method for extracting the melt pool width based on contour search is proposed, comprising the following steps:

[0012] The original molten pool image is converted to grayscale to generate a grayscale image of the molten pool;

[0013] The grayscale image of the molten pool is binarized to obtain a binarized molten pool image, and the grayscale value of each pixel in the single-channel image is determined.

[0014] An edge detection algorithm is used to traverse the binarized molten pool image and extract the image edge points, which are denoted as the contour point pixel set N(x). b ,y b );

[0015] For a binarized molten pool image, let M(x) be the set of pixels with a value of 0. a ,y a ), where M(x) a ,y a Each point in the image is within the contour of the binarized molten pool image;

[0016] Traverse and compute set M(x) a ,y a The set of pixels N(x) from each point within the contour points b ,y b The sum of the Euclidean distances of all points in q is denoted as the set [q]. z ], z = 1, 2, ..., n;

[0017] Find the set [q] zThe largest value in the [data] is denoted as q. k Record q at this time k The corresponding set M(x) a ,y a Given pixel K(x,y) in the contour point set N(x,y), calculate the relationship between pixel K(x,y) and the contour point set N(x,y). a ,y b The minimum Euclidean distance to each point in the array;

[0018] Based on the minimum value of the Euclidean distance calculated above, the molten pool width D is calculated.

[0019] As an optional embodiment, the calculation of pixel point K(x,y) and the set of contour point pixels N(x) a ,y b The minimum Euclidean distance of each point in the contour is denoted as the radius R of the largest inscribed circle within the boundary of the contour points, and the aforementioned pixel point K(x,y) is denoted as the center of the largest inscribed circle.

[0020] Therefore, the molten pool width D is calculated as follows:

[0021] D = 2R.

[0022] A second aspect of the present invention also proposes a melt pool width extraction system based on contour search, comprising:

[0023] One or more processors;

[0024] The memory stores operable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations, including the flow of the aforementioned contour-based melt pool width extraction method.

[0025] In a third aspect of the invention, a computer-readable medium for storing software is also provided, the software including instructions executable by one or more computers, the instructions causing the one or more computers to perform operations including the flow of the aforementioned contour-search-based melt pool width extraction method.

[0026] Therefore, the molten pool width extraction method based on contour search proposed in this invention can quickly and accurately extract the width set information of the molten pool in the metal laser directional energy deposition process, accurately acquire its geometric features in real time, which is beneficial for controlling the molten pool morphology and improving the quality of laser cladding parts.

[0027] It should be understood that all combinations of the foregoing concepts and the additional concepts described in more detail below may be considered part of the inventive subject matter of this disclosure, provided that such concepts do not contradict each other. Furthermore, all combinations of the claimed subject matter are considered part of the inventive subject matter of this disclosure.

[0028] The foregoing and other aspects, embodiments, and features of the teachings of the present invention will be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the invention, such as features and / or beneficial effects of exemplary embodiments, will become apparent from the following description or may be learned through practice of specific embodiments according to the teachings of the present invention. Attached Figure Description

[0029] The accompanying drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown in the various figures may be denoted by the same reference numeral. For clarity, not every component is labeled in each figure. Embodiments of various aspects of the invention will now be described by way of example and with reference to the accompanying drawings.

[0030] Figure 1 This is a flowchart of a melt pool width extraction method based on contour search according to an embodiment of the present invention.

[0031] Figures 2a-2d This is a schematic diagram illustrating the implementation process of the molten pool width extraction method based on contour search according to an embodiment of the present invention, wherein 3a-3d represent the processes of grayscale processing, binarization processing, contour extraction, and molten pool width extraction of the molten pool image, respectively.

[0032] Figure 3a , 3b This is a schematic diagram illustrating the implementation effect of the melt pool width extraction method based on contour search according to an embodiment of the present invention. Detailed Implementation

[0033] To better understand the technical content of the present invention, specific embodiments are described below in conjunction with the accompanying drawings.

[0034] Various aspects of the invention are described in this disclosure with reference to the accompanying drawings, which illustrate numerous illustrative embodiments. The embodiments of this disclosure are not necessarily intended to encompass all aspects of the invention. It should be understood that the various concepts and embodiments described above, as well as those described in more detail below, can be implemented in any of many ways, because the concepts and embodiments disclosed herein are not limited to any particular implementation. Furthermore, some aspects of the invention disclosed may be used alone or in any suitable combination with other aspects of the invention disclosed.

[0035] Combination Figure 1The melt pool width extraction method based on contour search in the illustrated embodiment includes the following steps:

[0036] S1: Perform grayscale processing on the original molten pool image to generate a grayscale image of the molten pool;

[0037] S2: Binarize the grayscale image of the molten pool to obtain a binarized molten pool image, and determine the grayscale value of each pixel in the single-channel image;

[0038] S3: An edge detection algorithm is used to traverse the binarized molten pool image and extract the image edge points, denoted as the contour point pixel set N(x b ,y b );

[0039] S4: For the binarized molten pool image, let M(x) be the set of pixels with a value of 0. a ,y a ), where M(x) a ,y a Each point in the image is within the contour of the binarized molten pool image;

[0040] S5: Traverse and compute set M(x) a ,y a The set of pixels N(x) from each point within the contour points b ,y b The sum of the Euclidean distances of all points in q is denoted as the set [q]. z ], z = 1, 2, ..., n;

[0041] S6: Find the set [q] z The largest value in the [data] is denoted as q. k Record q at this time k The corresponding set M(x) a ,y a Given pixel K(x,y) in the contour point set N(x,y), calculate the relationship between pixel K(x,y) and the contour point set N(x,y). a ,y b The minimum Euclidean distance to each point in the array; and

[0042] S7: Calculate the molten pool width D based on the minimum value of the Euclidean distance calculated above.

[0043] Therefore, the geometric information of the molten pool width can be quickly and accurately extracted from the monitoring video of the molten pool under online monitoring through image extraction and processing.

[0044] Below we will combine Figures 2a-2d The processing procedure shown further illustrates the implementation process of the aforementioned method.

[0045] As an optional embodiment, the original molten pool image is converted to grayscale to generate a grayscale image of the molten pool, such as... Figure 2a As shown, it performs grayscale processing based on the following grayscale processing function:

[0046] y gray = (R+G+B) / 3

[0047] In the formula, R, G, and B represent the grayscale values ​​of the corresponding pixels in the red, green, and blue channels of the original molten pool image, respectively, and y gray This represents the grayscale value of each pixel after grayscale processing.

[0048] In an embodiment of the present invention, the grayscale image of the molten pool is binarized to obtain a binarized molten pool image, such as... Figure 2b As shown.

[0049] As an optional example, image binarization is performed based on a preset threshold and according to the following binarization function:

[0050]

[0051] Specifically, when the gray value of a pixel is greater than the aforementioned preset threshold, the pixel value is set to 255; otherwise, the pixel value is set to 0, thereby obtaining a binarized melt pool image.

[0052] The preset threshold value 'a' for binarization processing ranges from 90 to 110.

[0053] like Figure 2b The binarized image processing result is shown, where the threshold value 'a' for binarization is 94.

[0054] In conjunction with embodiments of the present invention, the pixel point K(x,y) and the set of contour pixel points N(x) are calculated. a ,y b The minimum Euclidean distance of each point in the contour is denoted as the radius R of the largest inscribed circle within the boundary of the contour points, and the aforementioned pixel point K(x,y) is denoted as the center of the largest inscribed circle.

[0055] Therefore, the molten pool width D is calculated as follows:

[0056] D = 2R.

[0057] Therefore, combined Figures 2a-2dAs shown, this invention addresses the problem of inaccurate extraction of molten pool width caused by interference from spatter, noise, and arc light. It proposes a contour search-based molten pool width extraction method, employing edge-iterative calculation to extract molten pool width information. First, the molten pool image is extracted, then grayscale and binarized to obtain a binarized molten pool image. An edge extraction algorithm is used to extract the boundary of the binarized image, and edge iterative calculation is performed to calculate the maximum inscribed circle within the boundary range. The diameter of the inscribed circle is then determined as the molten pool width, thus achieving rapid and accurate extraction of the geometric features of the molten pool width. The image processing and extraction processes involve low computational load, high extraction speed and efficiency, and accurate and controllable results. This method can be applied to laser metal directional energy deposition additive manufacturing processes, facilitating control of the molten pool morphology and ensuring the quality of the clad parts.

[0058] In conjunction with the disclosed embodiments of the present invention, a melt pool width extraction system based on contour search is also proposed, including one or more processors and a memory.

[0059] The aforementioned memory is configured to store operable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations, including the flow of the melt pool width extraction method based on contour search as described in the foregoing embodiments.

[0060] According to embodiments disclosed in this invention, a computer-readable medium for storing software including instructions executable by one or more computers, which, when executed, cause the one or more computers to perform operations including a flow of a contour-search-based melt pool width extraction method as described in the foregoing embodiments.

[0061] While the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the invention. Those skilled in the art can make various modifications and refinements without departing from the spirit and scope of the invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. A method for extracting melt pool width based on contour search, characterized in that, Includes the following steps: The original molten pool image is converted to grayscale to generate a grayscale image of the molten pool; The grayscale image of the molten pool is binarized to obtain a binarized molten pool image, and the grayscale value of each pixel in the single-channel image is determined. An edge detection algorithm is used to traverse the binarized molten pool image and extract the image edge points, which are denoted as the contour point pixel set. N ( x b ,y b ); For a binarized molten pool image, let the set of pixels with a value of 0 be denoted as . M ( x a ,y a ),in M ( x a ,y a Each point in the image is within the contour of the binarized molten pool image; Traverse and compute the set M ( x a ,y a The set of pixels from each point within the contour point N ( x b ,y b The sum of the Euclidean distances of all points in a given set is denoted as set [ ]. q z ], z=1,2,..n; Find the set [ q z The largest value in the [ ] is denoted as q k Record this moment q k The corresponding set M ( x a ,y a Pixels in ) K ( x,y ), calculate pixel points K ( x,y ) and the set of outline pixels N ( x a ,y b The minimum Euclidean distance of each point in the contour is denoted as the radius R of the largest inscribed circle within the boundary of the contour points. (The aforementioned pixel points...) K ( x,y Let be the center of the largest inscribed circle; Calculate the molten pool width D: D=2R.

2. The method for extracting melt pool width based on contour search according to claim 1, characterized in that, The step of converting the original molten pool image to grayscale to generate a grayscale image of the molten pool includes: Grayscale conversion is performed based on the following grayscale processing function: y gray =(R+G+B) / 3 ; In the formula, R, G, and B represent the grayscale values ​​of the corresponding pixels in the red, green, and blue channels of the original molten pool image, respectively. y gray This represents the grayscale value of each pixel after grayscale processing.

3. The method for extracting melt pool width based on contour search according to claim 1, characterized in that, The step of binarizing the grayscale image of the molten pool to obtain a binarized molten pool image includes: Image binarization is performed based on a preset threshold and according to the following function: ； Specifically, when the gray value of a pixel is greater than the threshold, the value of that pixel is set to 255; otherwise, the value of the pixel is set to 0, thus obtaining a binarized melt pool image.

4. The method for extracting melt pool width based on contour search according to claim 3, characterized in that, The preset binarization threshold ranges from 90 to 110.

5. The method for extracting melt pool width based on contour search according to claim 3, characterized in that, The preset binarization threshold is 94.

6. A melt pool width extraction system based on contour search, characterized in that, include: One or more processors; The memory stores operable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations, including the flow of the contour-search-based melt pool width extraction method as described in any one of claims 1-5.

7. A computer-readable medium for storing software, characterized in that, The software includes instructions executable by one or more computers, which, when executed, cause the one or more computers to perform operations including the flow of the contour-search-based melt pool width extraction method as described in any one of claims 1-5.

Images