A method for detecting the forming direction and width of a molten pool in metal additive manufacturing

Abstract

The present invention discloses a method for detecting the forming direction and width of a metal additive manufacturing melt pool, comprising: collecting a laser additive manufacturing melt pool image, segmenting and pre-processing the melt pool image to obtain a melt pool segmentation image; binarizing the intercepted image to extract high-pixel value area images and low-pixel value area images of the melt pool; calculating the center point A of the bright area of ​​the melt pool based on the high-pixel value area image; calculating the center point B of the dark area of ​​the melt pool based on the low-pixel value area image; calculating the direction angle and direction line of the melt pool based on the center points A and B; calculating the direction perpendicular to the melt pool based on the direction line of the melt pool, and calculating the width of the melt pool in combination with the melt pool segmentation image. The present invention can accurately determine the direction of the melt pool and calculate the width of the melt pool for melt pool parameter analysis. This method has the advantages of low computational complexity, good real-time performance, and ease of implementation, and can be extended to the calculation of the direction and area of ​​melt pools of other additive manufacturing types.

Description

Technical Field

[0001] The present invention relates to the technical fields of metal additive manufacturing, image processing and machine vision, and in particular to a method for detecting the forming direction and width of a molten pool in metal additive manufacturing. Background Art

[0002] Laser additive manufacturing (AM) technology, based on digital manufacturing and design, has been widely applied in aerospace, industrial production, biomedicine, and military applications. Image recognition is a current hot topic in the field of artificial intelligence. Capturing images of the AM melt pool and using various image processing techniques to identify, segment, and detect it has become a key research direction. For example, detection networks such as Yolo and EfficientDent are used to detect the melt pool area, while segmentation networks such as Mask-RCNN and PSPNet are used to extract the pixel area of ​​the melt pool. While significant progress has been made in the extraction of contours and morphology recognition for laser powder feeding AM melt pools, relatively little research has been conducted on extracting parameters such as width, area, center point, and variation trends from the identified melt pool images. Summary of the Invention

[0003] Purpose of the invention: In order to solve the problem in the prior art that it is difficult to analyze the melt pool image parameters after the additive manufacturing melt pool is detected, a method for detecting the forming direction and width of the metal additive manufacturing melt pool is provided, which can accurately determine the direction of the melt pool and calculate the width of the melt pool for parameter analysis of the melt pool. This method has the advantages of small computational complexity, good real-time performance, and easy implementation, and can be extended to the calculation of the melt pool direction and area of ​​other additive manufacturing types.

[0004] Technical Solution: To achieve the above-mentioned purpose, the present invention provides a method for detecting the forming direction and width of a metal additive manufacturing molten pool, comprising the following steps:

[0005] S1: Collect the laser additive manufacturing melt pool image, and perform segmentation preprocessing on the melt pool image to obtain the melt pool segmentation image;

[0006] S2: Binarize the captured image to extract the high pixel value area image and the low pixel value area image of the molten pool respectively;

[0007] S3: Calculate the center point A of the bright area of ​​the molten pool based on the image of the high pixel value area; calculate the center point B of the dark area of ​​the molten pool based on the image of the low pixel value area;

[0008] S4: Calculate the direction angle and direction line of the molten pool based on the center points A and B;

[0009] S5: Calculate the direction perpendicular line of the melt pool according to the direction straight line of the melt pool, and calculate the width of the melt pool by combining the melt pool segmentation image.

[0010] Furthermore, the method for collecting the laser additive manufacturing molten pool image in step S1 is:

[0011] A1: Use an industrial CCD camera to capture video of the molten pool during laser powder feeding. The camera is fixed to the laser and moves with the laser. The laser is reflected from the laser channel into the camera lens. Filters and attenuators are used to adjust the lens focal length to obtain a clear image of the molten pool.

[0012] A2: Videos of the melt pool were captured at different laser powers, powder feed rates, and laser travel speeds. The camera parameters, such as exposure rate and gain, remained fixed, and the video size was 200 × 200 pixels.

[0013] A3: After shooting, extract each frame of the video according to the frame rate of the video.

[0014] Furthermore, the method of segmenting and preprocessing the melt pool image in step S1 is: sending the extracted image to the melt pool morphology detection method based on the edge angle additional loss function for segmentation to obtain the pixel area of ​​the melt pool; the output melt pool segmentation image has only the pixel values ​​of the melt pool area that are not 0, and the rest are 0, which facilitates the calculation of the melt pool parameters.

[0015] Furthermore, the step S2 is specifically as follows:

[0016] B1: After reading the original image of the molten pool, the image is filtered using a filter of size 5 to remove some noise and splashing powder;

[0017] B2: Binarize the processed image. First, change the pixel values ​​of all pixels between 175 and 255 to 255, and change the remaining pixels to 0. This will produce an image with high pixel values, that is, an image of the high pixel value area of ​​the melt pool. Then, change the pixel values ​​of all pixels between 20 and 255 to 255, and change the remaining pixels to 0. This will produce an image with both high and low pixel values.

[0018] B3: Subtract the image containing high pixel values ​​and low pixel values ​​obtained in step B2 from the image with high pixel values ​​to obtain an image with low pixel values, i.e., an image of the low pixel value area of ​​the molten pool.

[0019] Furthermore, the step S3 is specifically as follows:

[0020] C1: Use OpenCV's contour detection operator to process the two images of different areas of the melt pool to obtain the outer contours of the high-pixel value area and the low-pixel value area;

[0021] C2: Use the grab_contours function of the imutils graphics processing toolkit to calculate the two images obtained in step C1, and calculate the center points A and B of the bright area and dark area respectively. The direction of the melt pool is from point B to point A. The upper left corner of the image is the coordinate origin, the horizontal and vertical coordinates are both 0, the distance between the pixel point and the leftmost pixel is the horizontal coordinate of the pixel point, and the distance from the pixel point to the top pixel of the image is the vertical coordinate.

[0022] Furthermore, the step S4 is specifically as follows:

[0023] D1: The direction angle of the molten pool is the arc tangent value of A minus B. The calculation algorithm of the direction angle is shown in formula (1):

[0024] D pool_rad =arctan((B y -A y ) / (B x -A x )) (1)

[0025] In formula (1), D poolrad is the value of the radian system of the molten pool direction angle, B y 、A y is the vertical coordinate of point B and point A, B x 、A x is the horizontal coordinate of point B and point A;

[0026] Then convert the calculated result into angle expression, as shown in formula (2):

[0027]

[0028] In formula (2), D pool is the value of the angle of the molten pool direction, and the other parameters are the same as those in formula (1);

[0029] D2: Calculate the direction straight line formula of the molten pool based on the direction angle between point A and the molten pool, as shown in formula (3):

[0030]

[0031] In formula (3), x image is any horizontal coordinate on the direction line, y image It is the ordinate corresponding to the abscissa on the direction line. The direction line is used to fit the molten pool area to calculate the width of the molten pool.

[0032] Furthermore, the step S5 is specifically as follows:

[0033] E1: The slope of the vertical line can be obtained according to the direction of the molten pool. The equation of the vertical line of the molten pool is:

[0034]

[0035] In formula (4), C x 、C y The horizontal and vertical coordinates of any point on the direction line of the melt pool. The vertical line of the melt pool remains perpendicular to the direction line and can be translated on the direction line to fit the melt pool.

[0036] E2: Starting from the center point A of the bright area, the vertical line is translated to both sides. During each translation, the number of pixels on the vertical line belonging to the melt pool area is calculated based on the melt pool segmentation image.

[0037] When the translation reaches the point where the number of pixels in the melt pool is 0, the maximum number of pixels in the melt pool is the width of the melt pool. The calculation formula for the number of pixels W belonging to the melt pool area is shown in formula (5):

[0038]

[0039] In formula (5), x i 、y i is the horizontal and vertical coordinates of any point on the vertical line, n is the width of the melt pool image, f judge Determine whether this coordinate area in the melt pool segmentation image belongs to the melt pool. If it does, the result is 1, otherwise it is 0. Use formula (5) to calculate the number of pixels belonging to the melt pool area based on the horizontal and vertical coordinates of the melt pool segmentation image, and then use the Pythagorean theorem to calculate the melt pool width. At this point, the melt pool width calculation is completed.

[0040] The present invention provides a method for detecting the forming direction and width of a metal additive manufacturing melt pool in terms of calculating the direction and width of the melt pool image. The method obtains relevant parameters of the melt pool based on the distribution of pixels in the melt pool image and is used to analyze the melt pool. The method achieves high-speed and high-precision calculation of the direction and width of the melt pool, and the program is easy to design and transplant.

[0041] The method developed in this paper independently designs an algorithm for determining the direction of the melt pool image and uses it in conjunction with a melt pool segmentation image to calculate the melt pool width. This method analyzes parameters of both the original melt pool image and the melt pool segmentation image, accurately calculating the melt pool direction and width. This is of great significance for melt pool status monitoring and quality analysis of additive manufacturing components, and lays the foundation for implementing melt pool parameter feedback to control laser power.

[0042] Beneficial effects: Compared with the existing technology, for application scenarios such as the original image and melt pool segmentation image of the laser additive manufacturing melt pool, the present invention adopts an independently designed melt pool image processing algorithm, combined with the melt pool segmentation image, to measure the pixel parameters of the melt pool image, thereby realizing fast and efficient calculation of the melt pool direction and width, and being able to provide the requirements of real-time parameter feedback in the manufacturing process, which is of great significance for accurately analyzing the melt pool parameters, providing feedback on the melt pool parameters, and controlling the laser power of the manufacturing equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of the workflow of the method of the present invention;

[0044] Figure 2 The image is obtained by installing the CCD camera on the laser provided by the embodiment of the present invention;

[0045] Figure 3 This is the original image of the molten pool provided by the embodiment of the present invention;

[0046] Figure 4 The extracted images of bright and dark areas of the molten pool provided by the embodiment of the present invention are as follows;

[0047] Figure 5 This is an image showing the effect of the melt pool direction calculation according to an embodiment of the present invention;

[0048] Figure 6 This is a directional distribution image of the molten pool taken continuously for 1000 times according to an embodiment of the present invention;

[0049] Figure 7 Schematic diagram of calculation of molten pool width according to an embodiment of the present invention;

[0050] Figure 8 These are 1000 consecutive images of the width distribution of the molten pool according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The present invention is further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, modifications of various equivalent forms of the present invention made by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0052] The present invention provides a method for detecting the forming direction and width of a molten pool in metal additive manufacturing, such as Figure 1 As shown, it includes the following steps:

[0053] Step 1: Collect the laser powder feeding molten pool video [see Jiang Houfeng, Wurikaixi Ayiti, An Pengfang. A review of the research on molten pool morphology and molten pool temperature in laser cladding forming [J]. Hot Working Technology, 2019, 48(10): 10-14.], intercept the molten pool image according to the frame rate, and pre-process the image;

[0054] Step 2: Binarize the captured image [see Shi Pengfei, Liu Hui. Polaroid detection method based on OpenCV [J]. Mechanical Management and Development, 2020, 35(10): 162-164+202], and extract the high pixel value area image and low pixel value area image of the molten pool respectively;

[0055] Step 3: Calculate the outlines of the bright and dark areas of the molten pool based on the high pixel value area image and the low pixel value area image, and extract the center points A and B of the bright area and the dark area;

[0056] Step 4: Calculate the direction angle and direction line of the molten pool based on the center points A and B;

[0057] Step 5: Calculate the direction perpendicular line of the melt pool according to the direction straight line of the melt pool. Combined with the melt pool segmentation image output by the melt pool morphology detection method based on the edge angle additional loss function, fit the area of ​​the melt pool and calculate the width of the melt pool.

[0058] In this embodiment, the above method is applied as follows:

[0059] Step 1 includes:

[0060] Step 1.1: Use an industrial CCD camera to capture the video of the molten pool during laser powder feeding. The camera is fixed on the laser and moves with the laser. Figure 2 As shown in the figure, the laser is reflected from the laser channel into the camera lens. A filter and attenuator are used to adjust the lens focus to achieve a clear image of the molten pool. The laser additive manufacturing equipment is an LDM4030, the manufacturing material is 50µs iron powder, and the laser diameter is 1mm. The video capture frame rate is 25 frames / s, and the filter transmittance is 0.5%.

[0061] Step 1.2: Collect videos of the melt pool at different laser powers, powder feed speeds, and laser travel speeds. The camera parameters such as exposure rate and gain remain fixed, and the video size is 200 × 200 pixels.

[0062] Step 1.3: After the shooting is completed, extract each frame of the video according to the frame rate of the video, such as Figure 3 As shown. The melt pool image is sent to the melt pool morphology detection method based on the edge angle additional loss function for segmentation, and the pixel area of ​​the melt pool is obtained. The output melt pool segmentation image has only the pixel values ​​in the melt pool area that are not 0, while the rest are all 0, which makes it easier to calculate the melt pool parameters;

[0063] Step 2 includes:

[0064] Step 2.1: After reading the original image of the melt pool, filter the image using a filter of size 5 to remove some noise and splashing powder to reduce interference;

[0065] Step 2.2: Binarize the processed image. First, change the pixel values ​​of all pixels between 175 and 255 to 255, and change the rest of the pixels to 0. This will produce an image with high pixel values, i.e., an image of the high pixel value area of ​​the melt pool. Then, change the pixel values ​​of all pixels between 20 and 255 to 255, and change the rest of the pixels to 0. This will produce an image with both high and low pixel values, as shown in the following example: Figure 4 As shown;

[0066] Step 2.3: Subtract the image containing high and low pixel values ​​obtained in the previous step from the image with high pixel values ​​to obtain an image with low pixel values, i.e., an image of the low pixel value area of ​​the melt pool;

[0067] Step 3 includes:

[0068] Step 3.1: Use the contour detection operator of OpenCV (a computer vision and machine learning software library) to process the two images of different areas of the melt pool to obtain the outer contours of the two areas;

[0069] Step 3.2: Use the grab_contours function of imutils (a graphics processing toolkit that encapsulates OpenCV) to calculate the two images obtained in step 3.1, and calculate the center points A and B of the bright area and the dark area. The direction of the molten pool is from point B to point A, as shown in the following example: Figure 5 As shown. The upper left corner of the image is the coordinate origin, the horizontal and vertical coordinates are both 0, the distance between the pixel point and the leftmost pixel is the horizontal coordinate of the pixel point, and the distance from the pixel point to the top of the image is the vertical coordinate;

[0070] Step 4 includes:

[0071] Step 4.1: The direction angle of the molten pool is the arc tangent value of A minus B. The calculation algorithm of the direction angle is shown in formula (1):

[0072] D pool_rad =arctan((B y -A y ) / (B x -A x )) (1)

[0073] In formula (1), D poolrad is the value of the radian system of the molten pool direction angle, B y 、Ay is the vertical coordinate of point B and point A, B x 、A x The horizontal coordinates of point B and point A are shown in the figure below. Figure 6 The calculated result is then converted into an angle expression, as shown in formula (2):

[0074]

[0075] In formula (2), D pool is the value of the angle of the direction of the molten pool, and the rest of the values ​​are the same as in formula (1). The direction of the molten pool includes 360°, while the value of the inverse tangent can only represent -90° to 90°, and does not include the upper and lower bounds, so it must be calculated separately. When B x -A x When <0, calculate directly, and when B x -A x >0, it is necessary to add 180° to the value obtained, and when B x =A x When the tangent value cannot be calculated, directly according to B y and A y The size relationship is used to determine whether it is 90° or 270°.

[0076] Step 4.2: The formula for calculating the direction line of the molten pool based on the direction angle between point A and the molten pool is shown in formula (3):

[0077]

[0078] In formula (3), x is an arbitrary horizontal coordinate on the directional line, and y is the vertical coordinate corresponding to the horizontal coordinate on the directional line. The directional line is used to fit the melt pool area to calculate the width of the melt pool.

[0079] Step 5 includes:

[0080] Step 5.1: The slope of the perpendicular line can be obtained according to the direction of the molten pool. The equation of the perpendicular line of the molten pool is:

[0081]

[0082] In formula (4), C x 、C y is the horizontal and vertical coordinates of any point on the molten pool direction line. The vertical line of the molten pool remains perpendicular to the direction line. Transform C x 、C y The value of is used to translate on the straight line to fit the molten pool;

[0083] Step 5.2: Starting from the center point A of the bright area, translate the vertical line to both sides. Each time you translate, calculate the number of pixels on the vertical line that belong to the melt pool area based on the melt pool segmentation image. Stop when the number of melt pool pixels reaches 0. The maximum number of melt pool pixels is the width of the melt pool. The calculation formula for the number of pixels W belonging to the melt pool area is shown in Equation (5):

[0084]

[0085] In formula (5), x i 、y i is the horizontal and vertical coordinates of any point on the vertical line, n is the width of the melt pool image, f judge Determine whether the coordinate area in the melt pool segmentation image belongs to the melt pool. If it does, the result is 1, otherwise it is 0. Use formula (5) to calculate the number of pixels belonging to the melt pool area based on the horizontal and vertical coordinates of the melt pool segmentation image, and then use the Pythagorean theorem to calculate the melt pool width. The calculation diagram of the melt pool width is as follows: Figure 7 The width distribution of the molten pool taken in 1000 consecutive shots is shown in Figure 8 As shown, the straight line in the figure is the average value of the width.

[0086] This embodiment also provides a metal additive manufacturing molten pool forming direction and width detection system, which includes a network interface, a memory and a processor; wherein the network interface is used to realize signal reception and transmission in the process of sending and receiving information between other external network elements; the memory is used to store computer program instructions that can be run on the processor; the processor is used to execute the steps of the above-mentioned consensus method when running the computer program instructions.

[0087] This embodiment also provides a computer storage medium that stores a computer program that can implement the method described above when a processor executes the computer program. The computer-readable medium can be considered to be tangible and non-transitory. Non-limiting examples of non-transitory tangible computer-readable media include non-volatile memory circuits (such as flash memory circuits, erasable programmable read-only memory circuits, or mask read-only memory circuits), volatile memory circuits (such as static random access memory circuits or dynamic random access memory circuits), magnetic storage media (such as analog or digital tapes or hard drives), and optical storage media (such as CDs, DVDs, or Blu-ray discs). The computer program includes processor-executable instructions stored on at least one non-transitory tangible computer-readable medium. The computer program may also include or rely on stored data. The computer program may include a basic input / output system (BIOS) that interacts with the hardware of a special-purpose computer, device drivers that interact with specific devices of the special-purpose computer, one or more operating systems, user applications, background services, background applications, etc.

[0088] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0089] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0090] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0091] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

Claims

1. A method for detecting the forming direction and width of a metal additive manufacturing molten pool, characterized in that: The steps include: S1: Collect the laser additive manufacturing melt pool image, and perform segmentation preprocessing on the melt pool image to obtain the melt pool segmentation image; S2: Binarize the captured image to extract the high pixel value area image and the low pixel value area image of the molten pool respectively; S3: Calculate the center point A of the bright area of ​​the molten pool based on the image of the high pixel value area; calculate the center point B of the dark area of ​​the molten pool based on the image of the low pixel value area; S4: Calculate the direction angle and direction line of the molten pool based on the center points A and B; S5: Calculate the direction perpendicular line of the melt pool according to the direction straight line of the melt pool, and calculate the width of the melt pool by combining the melt pool segmentation image; Step S4 is specifically as follows: D1: The direction angle of the molten pool is the arc tangent value of A minus B. The calculation algorithm of the direction angle is shown in formula (1): D pool_rad =arctan((B y -A y ) / (B x -A x )) (1) In formula (1), D pool_rad is the value of the radian system of the molten pool direction angle, B y 、A y is the vertical coordinate of point B and point A, B x 、A x is the horizontal coordinate of point B and point A; Then convert the calculated result into angle expression, as shown in formula (2): In formula (2), D pool is the value of the angle of the molten pool direction, and the rest of the parameters are the same as those in formula (1); D2: Calculate the direction straight line formula of the molten pool based on the direction angle between point A and the molten pool, as shown in formula (3): In formula (3), x image is any horizontal coordinate on the direction line, y image is the ordinate corresponding to the abscissa on the direction line. The direction line is used to fit the molten pool area to calculate the width of the molten pool; Step S5 is specifically as follows: E1: The slope of the vertical line can be obtained according to the direction of the molten pool. The equation of the vertical line of the molten pool is: In formula (4), C x 、C y The horizontal and vertical coordinates of any point on the direction line of the melt pool. The vertical line of the melt pool remains perpendicular to the direction line and can be translated on the direction line to fit the melt pool. E2: Starting from the center point A of the bright area, the vertical line is translated to both sides. During each translation, the number of pixels on the vertical line belonging to the melt pool area is calculated based on the melt pool segmentation image. When the translation reaches the point where the number of pixels in the melt pool is 0, the maximum number of pixels in the melt pool is the width of the melt pool. The calculation formula for the number of pixels W belonging to the melt pool area is shown in formula (5): In formula (5), x i 、y i is the horizontal and vertical coordinates of any point on the vertical line, n is the width of the melt pool image, f judge Determine whether this coordinate area in the melt pool segmentation image belongs to the melt pool. If it does, the result is 1, otherwise it is 0. Use formula (5) to calculate the number of pixels belonging to the melt pool area based on the horizontal and vertical coordinates of the melt pool segmentation image, and then use the Pythagorean theorem to calculate the melt pool width. At this point, the melt pool width calculation is completed.

2. A method for detecting the forming direction and width of a metal additive manufacturing molten pool according to claim 1, characterized in that: The method for collecting the laser additive manufacturing molten pool image in step S1 is: A1: Use an industrial CCD camera to capture video of the molten pool during laser powder feeding. The camera is fixed to the laser and moves with the laser. The laser is reflected from the laser channel into the camera lens. Filters and attenuators are used to adjust the lens focal length to obtain a clear image of the molten pool. A2: Capture melt pool videos at different laser powers, powder feed rates, and laser travel speeds, with the camera parameters kept fixed. A3: After shooting, extract each frame of the video according to the frame rate of the video.

3. The method for detecting the forming direction and width of a metal additive manufacturing molten pool according to claim 1, characterized in that: The method of segmenting and preprocessing the molten pool image in step S1 is: sending the extracted image to the molten pool morphology detection method based on the edge angle additional loss function for segmentation, and obtaining the pixel area of ​​the molten pool.

4. The method for detecting the forming direction and width of a metal additive manufacturing molten pool according to claim 1, characterized in that: The step S2 is specifically as follows: B1: After reading the original image of the molten pool, the image is filtered using a filter of size 5 to remove some noise and splashing powder; B2: Binarize the processed image. First, change the pixel values ​​of all pixels between 175 and 255 to 255, and change the remaining pixels to 0. This will produce an image with high pixel values, that is, an image of the high pixel value area of ​​the melt pool. Then, change the pixel values ​​of all pixels between 20 and 255 to 255, and change the remaining pixels to 0. This will produce an image with both high and low pixel values. B3: Subtract the image containing high pixel values ​​and low pixel values ​​obtained in step B2 from the image with high pixel values ​​to obtain an image with low pixel values, i.e., an image of the low pixel value area of ​​the molten pool.

5. The method for detecting the forming direction and width of a metal additive manufacturing molten pool according to claim 1, characterized in that: The step S3 is specifically as follows: C1: Use OpenCV's contour detection operator to process the two images of different areas of the melt pool to obtain the outer contours of the high-pixel value area and the low-pixel value area; C2: Use the grab_contours function of the imutils graphics processing toolkit to calculate the two images obtained in step C1, and calculate the center points A and B of the bright area and dark area respectively. The direction of the melt pool is from point B to point A.

Images