Powder spreading quality detection and stage treatment method, device and equipment

Abstract

The invention relates to a powder laying quality detection and grading processing method, device and equipment, and the powder laying quality detection and grading processing method comprises the steps: obtaining a powder laying image through a CCD (Charge Coupled Device) camera of which the shooting breadth fully covers a powder bed; performing defect identification on the powder spreading image by using a convolutional neural network, and determining a defect position and a defect type; and determining a position relationship between the defect position and a scanning path, and performing grading treatment on the powder spreading defect according to the position relationship and the defect type. The powder spreading defect can be detected in real time and corrected in time, the forming efficiency is improved, and resource waste caused by the powder spreading defect is reduced.

Description

Technical Field

[0001] The present application relates to the field of powder bed additive manufacturing technology, and in particular to a method, device and equipment for powder bed quality detection and grading. Background Art

[0002] With the rapid development of my country's aerospace weapon equipment manufacturing technology, many key structures use a large number of complex cavity components. The traditional processing technology has a long manufacturing cycle and a low yield rate. What's worse, some complex structures cannot be developed, so they cannot meet the structural, performance and functional requirements of aircraft. Selective melting forming technology is an advanced manufacturing technology based on discrete stacking forming ideas. This technology can realize the rapid manufacturing of complex cavities, spatial lattices and brittle materials, greatly shorten the production cycle, quickly verify the design ideas, and realize the integrated design and manufacturing of materials, structures and functions. It has broad application prospects in the field of aerospace. The powder bed formed after powder spreading is the basis for electron beam or laser beam scanning to melt powder. If the powder bed is not flat, it will cause the instability of the molten pool during the scanning process, produce abnormal protrusions or pores and other defects, and affect subsequent forming. The selective melting forming process often takes several days or even dozens of days. Various powder spreading defects are very likely to occur during the printing process. Powder spreading defects will directly affect the process quality of the formed parts. Therefore, it is necessary to detect powder spreading defects in real time and correct them in time.

[0003] Common powder spreading defects include vibration of the powder spreading mechanism, defective scraper, insufficient powder, protrusion of molded parts, powder impurities, etc. Different types of powder spreading defects have different corrective measures. However, no matter which corrective measures are taken, the forming efficiency will be affected, and even the printing needs to be interrupted to open the door for hardware repair, which will cause the print to be scrapped. Therefore, how to quickly and accurately detect powder spreading defects and give appropriate grading treatment strategies according to the defect location and type is very necessary to improve the forming quality and efficiency and avoid the waste of resources caused by excessive intervention in powder spreading defects. Summary of the invention

[0004] (1) Technical issues to be solved

[0005] The present application provides a method, device and equipment for powder spreading quality detection and grading processing. The method and equipment can detect the location and type of powder spreading defects, and by detecting the defect location and the relative position relationship of the printing path, quickly and accurately provide grading processing measures for the powder spreading defects, thereby solving the problems in the above-mentioned background technology.

[0006] (2) Technical solution

[0007] In a first aspect, the present application provides a method for powder spreading quality detection and grading, comprising:

[0008] The powder spreading image is obtained by shooting a CCD camera that fully covers the powder bed;

[0009] Using a convolutional neural network to identify defects in the powder spreading image, and determine the defect location and defect type;

[0010] The positional relationship between the defect position and the scanning path is determined, and the powder spreading defects are graded according to the positional relationship and the defect type.

[0011] Furthermore, the use of a convolutional neural network to perform defect recognition on the powder spreading image to determine the defect location and defect type includes:

[0012] The size of the powder spreading interconnection defect is identified by using a convolutional neural network on the powder spreading image, and the defect position is framed by a minimum rectangular frame that can enclose the interconnection defect, and the rectangular frame area is used to replace the actual powder spreading defect position to participate in the position detection operation;

[0013] According to the image features of the powder spreading image, a convolutional neural network is used to determine the defect type.

[0014] Furthermore, the determining of the positional relationship between the defect position and the scanning path, and performing graded processing on the powder spreading defect according to the positional relationship and the defect type, includes:

[0015] Determining whether the defect position overlaps with the scanning path;

[0016] If yes, the powder spreading defect is graded according to the defect type; otherwise, no graded processing is performed.

[0017] Further, the determining whether the defect position overlaps with the scanning path includes:

[0018] Slice the three-dimensional digital model of the part to be formed to obtain the scanning path image of the forming layer;

[0019] It is determined whether the defect position overlaps with the scanning path according to the scanning path image and the powder spreading image.

[0020] Further, judging whether the defect position overlaps with the scanning path according to the scanning path image and the powder spreading image includes:

[0021] Filling the grayscale value of the scanned area in the scan path image with [0, 0, 0], and converting the powder spreading image in the pixel coordinate system into the machine tool coordinate system;

[0022] In the machine tool coordinate system, the scanning path image is used as a foreground image, and the powder spreading image is used as a background image, and the scanning path image and the powder spreading image are synthesized;

[0023] It is determined whether there is a connected area with a gray value of [0,0,0] in the rectangular frame. If so, the defect position overlaps with the scanning path, otherwise, there is no overlap.

[0024] In a second aspect, the present application provides a powder spreading quality detection and grading processing device, comprising:

[0025] An image acquisition module is used to acquire a powder spreading image by shooting a CCD camera that fully covers the powder bed;

[0026] A defect recognition module is used to use a convolutional neural network to perform defect recognition on the powder spreading image to determine the defect location and defect type;

[0027] The grading processing module is used to determine the positional relationship between the defect position and the scanning path, and to perform grading processing on the powder spreading defects according to the positional relationship and the defect type.

[0028] In a third aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the powder quality detection and grading processing method as described above is implemented.

[0029] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the powder quality detection and grading processing method as described above.

[0030] (3) Beneficial effects

[0031] The above technical solution of the present application has the following advantages:

[0032] The first aspect of the present application provides a method for powder spreading quality detection and grading processing. By using a CCD camera to collect powder spreading images, a convolutional neural network is used to identify and classify defects in the collected images, and the powder spreading defects are scientifically and effectively graded according to the defect category and its positional relationship with the processing path. It is capable of accurately detecting powder spreading defects in real time and performing corrective processing in a timely manner, thereby improving forming quality and efficiency and reducing waste of resources caused by excessive intervention in powder spreading defects.

[0033] In this application, the defect position is calibrated using the smallest rectangular frame that can enclose the connected defect area, and the calibrated rectangular frame position is used to participate in the calculation of the relative position relationship of the scanning path area, which simplifies the algorithm of using various types of special-shaped defect graphics to participate in relative position detection operations, greatly improves the detection efficiency, and realizes real-time detection and precise intervention of powder laying quality.

[0034] It can be understood that the beneficial effects of the second, third and fourth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 Flow chart of the powder quality detection and grading method provided in this application;

[0037] Figure 2 A schematic diagram of powder spreading image acquisition provided for this application;

[0038] Figure 3 A schematic diagram of powder spreading defect coordinates in the pixel coordinate system provided in this application;

[0039] Figure 4 A schematic diagram of powder spreading defects and scanning path coordinates in the machine tool coordinate system provided in this application;

[0040] Figure 5 The structure diagram of the powder quality detection and grading processing device provided in this application;

[0041] Figure 6 A schematic diagram of the structure of the electronic device provided in this application. DETAILED DESCRIPTION

[0042] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0043] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0044] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0045] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways. "Multiple" means "two or more".

[0046] There are many reasons for the unevenness of the powder bed, such as: ① Wear and damage of the powder brush, resulting in grooves or ridges distributed along the powder spreading direction; ② The molten layer is convex, and the scraper is forced to jump at the convexity, resulting in grooves or ridges perpendicular to the powder spreading direction; ③ Insufficient powder, no powder is filled at the end of the powder bed. The more mature technical solution for powder bed morphology monitoring is to use an optical camera to capture the morphology of the powder layer, and perform grayscale texture analysis on the image to determine the powder spreading quality.

[0047] Common powder spreading defects include vibration of the powder spreading mechanism, defective scraper, insufficient powder, protrusion of molded parts, powder impurities, etc. Different types of powder spreading defects require different corrective measures. However, no matter which corrective measures are taken, the forming efficiency will be affected, and even the printing needs to be interrupted to open the door for hardware repair, which will cause the print to be completely scrapped. Therefore, it is very necessary to quickly and accurately detect powder spreading defects and give appropriate graded treatment measures according to different defect types.

[0048] In response to the above problems, this application proposes a powder spreading quality detection and grading method, which uses a CCD camera to collect powder spreading images, uses a neural network algorithm to accurately identify and classify powder spreading defects, and designs a grading treatment strategy based on the causes of various powder spreading defects and the corrective measures for the defects in the processing path of the current layer. This method can effectively avoid the reduction of forming efficiency and waste of resources caused by unnecessary correction of powder spreading quality.

[0049] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0050] like Figure 1As shown, the powder spreading quality detection and grading processing method provided in this embodiment specifically includes the following steps:

[0051] S100, obtain the powder spreading image by shooting the CCD camera with the whole coverage of the powder bed.

[0052] In applications, such as Figure 2 As shown in the figure, the CCD camera is installed on the top of the chamber, and the angle between the installation angle and the horizontal plane of the chamber is 45°. The CCD shooting format can achieve full coverage of the powder bed.

[0053] S200, using a convolutional neural network to perform defect recognition on the powder spreading image to determine the defect location and defect type.

[0054] In some embodiments, the use of a convolutional neural network to perform defect recognition on the powder laying image to determine the defect location and defect type includes: using a convolutional neural network to identify the size of the powder laying interconnected defect on the powder laying image, and framing the defect location with a minimum rectangular frame that can enclose the interconnected defect, and using the rectangular frame area to replace the actual powder laying defect location to participate in position detection operations; based on the image features of the powder laying image, using a convolutional neural network to determine the defect type.

[0055] In the application, the neural network algorithm can be used to identify and classify the defects of the collected powder bed images. The algorithm can accurately identify the size of the powder bed interconnection defect and use the minimum rectangular frame to frame the specific location of the defect, and use the rectangular frame position instead of the actual defect position to participate in the calculation.

[0056] like Figure 3 As shown in the figure, two defects are identified in the powder spreading image in the pixel coordinate system. The positions of the four vertices of the rectangular box of defect 1 are Q 1 1(U1 V1),Q 1 2(U2 V2),Q 1 3(U3 V3),Q 1 4(U4 V4), the positions of the four vertices of the defect 2 rectangle are Q 2 1(U1 V1),Q 2 2(U2 V2),Q 2 3(U3 V3),Q 2 4 (U4 V4), and so on, the positions of the four vertices of the Nth defect rectangle are Q N 1(U1 V1),Q N 2(U2 V2),Q N 3(U3 V3),Q N 4 (U4 V4).

[0057] The neural network algorithm classifies defects into the following categories according to different image features: insufficient powder spreading defect, powder impurity defect, scraper defect, scraper vibration defect, and forming layer protrusion defect.

[0058] S300, determining the positional relationship between the defect position and the scanning path, and performing classification processing on the powder spreading defects according to the positional relationship and the defect type.

[0059] In some embodiments, determining the positional relationship between the defect position and the scanning path, and grading the powder spreading defect according to the positional relationship and the defect type, includes: determining whether there is an overlap between the defect position and the scanning path; if so, grading the powder spreading defect according to the defect type, otherwise no grading is performed.

[0060] In some embodiments, determining whether the defect position overlaps with the scanning path includes: slicing the three-dimensional digital model of the part to be formed to obtain a scanning path image of the forming layer; and determining whether the defect position overlaps with the scanning path based on the scanning path image and the powder laying image.

[0061] In some embodiments, judging whether there is overlap between the defect position and the scanning path based on the scanning path image and the powder laying image includes: filling the grayscale value of the scanning area in the scanning path image with [0,0,0], and converting the powder laying image in the pixel coordinate system to the machine tool coordinate system; in the machine tool coordinate system, taking the scanning path image as the foreground image and the powder laying image as the background image, synthesizing the scanning path image and the powder laying image; judging whether there is a connected area with a grayscale value of [0,0,0] in the rectangular frame, if so, the defect position overlaps with the scanning path, otherwise, there is no overlap.

[0062] In the application, the 3D digital model of the part to be formed can be sliced ​​to obtain the slice image of the scanning path of the forming layer, and the grayscale value of the scanning area can be filled with [0,0,0], such as Figure 4 The powder spreading image in the pixel coordinate system is converted into the machine tool coordinate system. After the conversion, the coordinate system of the defect 1 rectangular box is converted into P 1 1(X1 Y1 Z1), P 1 2(X2 Y2 Z2),P 1 3(X3Y3 Z3),P 1 4(X4 Y4 Z4), the coordinate system of the defect 2 rectangular box is transformed into P 2 1(X1 Y1 Z1), P 2 2(X2Y2 Z2),P 2 3(X3Y3 Z3),P 24(X4 Y4 Z4), and so on, such as Figure 4 shown.

[0063] In the machine tool coordinate system, the scanning path image and the powder-laying defect image are synthesized into one image. During the synthesis process, the scanning path image is used as the foreground image, and the powder-laying defect image is used as the background image. The coordinate traversal method is used to detect the pixel values ​​in each defective rectangular frame. When a connected area with a grayscale value of [0,0,0] is detected in the defective rectangular frame, it means that the powder-laying defect overlaps with the scanning path image. There is a powder-laying defect on the scanning path, and classification processing is required. When it is detected that there is no connected area with a grayscale value of [0,0,0] in the defective rectangular frame, it means that the powder-laying defect does not overlap with the scanning path image. The powder-laying defect is not within the range of the scanning path and will not affect the forming quality. The processing can be completed normally without classification processing.

[0064] There is no overlap between the scanning path image and the powder spreading defect image, and the forming process is carried out normally, and no intervention is required for the powder spreading defect. There is overlap between the scanning path image and the powder spreading defect image. According to the type of defect identified by the neural network algorithm, the following graded processing interventions are made: insufficient powder spreading defect, adjust the rising height of the powder spreading cylinder, and re-powder the powder bed; powder impurity defect, re-powder the powder bed; scraper defect, stop processing and replace the scraper; scraper vibration defect, adjust the transmission system or motion control system parameters; forming layer convex defect, rescan the current layer, manually lower the forming cylinder to a certain height (usually about 10 microns) and adjust the corresponding laser power, and scan and process the current layer after spreading the powder.

[0065] The following is an explanation through specific embodiments.

[0066] Example

[0067] The detection accuracy of the CCD stereo camera used in this embodiment is 0.001mm, and the detection field of view is 280*260mm. The size of the powder spreading area of ​​the forming equipment is 250*250mm. AlSi10Mg powder is selected as the forming material. The specific steps are as follows:

[0068] Set the process parameters, where the system powder spreading height is set to 30μm;

[0069] CCD camera collects powder spreading images;

[0070] The neural network algorithm detects the powder spreading defect and determines the type. When the detection result is no defect, the forming process is continued, and the next layer of powder spreading defect is re-detected. When the detection result shows that there is a powder spreading defect, the next step is executed. In this embodiment, the type of powder spreading defect detected by the neural network algorithm is insufficient powder spreading;

[0071] The neural network algorithm identifies the size of the powder spreading defect and uses a rectangular frame to mark the specific location of the defect; two defects are identified in the powder spreading image, and the positions of the four vertices of the rectangular frame of defect 1 are Q 1 1(U1 V1),Q 1 2(U2 V2),Q 1 3(U3 V3),Q 1 4(U4 V4), the positions of the four vertices of the defect 2 rectangle are Q 2 1(U1 V1),Q 2 2(U2 V2),Q 2 3(U3 V3),Q 2 4 (U4 V4);

[0072] Slice the 3D digital model of the part to be formed, obtain the slice image of the scanning path of the forming layer, and fill the grayscale value of the scanning area with [0,0,0];

[0073] Perform coordinate conversion to convert the powder spreading defect image in the pixel coordinate system into an image in the machine tool coordinate system;

[0074] In the machine tool coordinate system, the scanning path image and the powder-spreading defect image are synthesized into one image. During the synthesis process, the scanning path image is used as the foreground image and the powder-spreading defect image is used as the background image. The grayscale value within the scanning path range is directly overlapped on the powder-spreading defect image.

[0075] The pixel value in each defect rectangular frame is detected by using the coordinate traversal method. In this embodiment, there is an area with a gray value of [0,0,0] in the rectangular frame of defect 1;

[0076] Defects of insufficient powdering are graded and processing is continued after re-powdering.

[0077] The powder spreading quality detection and grading processing method provided in the embodiment of the present application relates to the phenomenon that powder spreading defects in the selective melting forming process lead to poor quality and precision of formed parts. A grading processing method that can effectively improve the quality and efficiency of formed parts is designed according to different powder spreading defects, thereby avoiding printing interruptions and waste of resources due to the inability to repair powder spreading defects online.

[0078] By using a CCD camera to capture powder spreading images, a convolutional neural network is used to identify and classify defects in the captured images, and a coordinate calibration algorithm is used to convert the coordinates of the defect image and the forming processing path. The powder spreading defects are scientifically and effectively graded according to the defect category and its positional relationship with the processing path, thereby improving the forming efficiency and reducing the waste of resources caused by powder spreading defects.

[0079] This method can adopt different grading treatment methods according to the type of powder spreading defects, avoiding the waste of resources caused by excessive intervention in defects. The method of overlapping and synthesizing the scanning path and the defect image is used to detect the positional relationship between the powder spreading defect and the scanning path. When the powder spreading defect is not on the scanning path, normal processing can be continued without any processing, thus avoiding the forming efficiency problems caused by terminating the processing or repeatedly spreading the powder. A rectangular frame is used to represent the defect, and the grayscale value traversal is performed using the vertices of the rectangular frame, thereby simplifying the irregular defect image recognition algorithm and improving the efficiency of online powder spreading defect detection.

[0080] Corresponding to the powder quality detection and grading method described in the above embodiment, as Figure 5 As shown, this embodiment provides a powder spreading quality detection and grading processing device, and the powder spreading quality detection and grading processing device 500 includes:

[0081] An image acquisition module 501 is used to acquire a powder spreading image by using a CCD camera that takes a picture of the powder bed in full coverage;

[0082] A defect recognition module 502 is used to perform defect recognition on the powder spreading image using a convolutional neural network to determine the defect location and defect type;

[0083] The classification processing module 503 is used to determine the positional relationship between the defect position and the scanning path, and to perform classification processing on the powder spreading defects according to the positional relationship and the defect type.

[0084] It should be noted that the information interaction, execution process, etc. between the above-mentioned modules / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0085] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0086] The present application embodiment also provides an electronic device 600, such as Figure 6 As shown, it includes a memory 601, a processor 602, and a computer program 603 stored in the memory 601 and executable on the processor 602. When the processor 602 executes the computer program 603, the steps of the powder quality detection and grading processing method provided in the first aspect are implemented.

[0087] In applications, the electronic device may include, but is not limited to, a processor and a memory. Figure 6 This is only an example of an electronic device and does not constitute a limitation on the electronic device. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input and output devices, network access devices, etc. Input and output devices may include cameras, audio acquisition / playback devices, display screens, etc. Network access devices may include a network module for wirelessly communicating with external devices.

[0088] In applications, the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0089] In application, the memory may be an internal storage unit of an electronic device in some embodiments, such as a hard disk or memory of the electronic device. The memory may also be an external storage device of the electronic device in other embodiments, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. The memory may also include both an internal storage unit and an external storage device of the electronic device. The memory is used to store operating systems, applications, boot loaders (BootLoader), data, and other programs, such as program codes of computer programs, etc. The memory may also be used to temporarily store data that has been output or is to be output.

[0090] The embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0091] The present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to an electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk.

[0092] Those of ordinary skill in the art will appreciate that the devices and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0093] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of devices, which can be electrical, mechanical or other forms.

[0094] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for powder quality detection and grading, It is characterized in that include: The powder spreading image is obtained by shooting a CCD camera that fully covers the powder bed; Using a convolutional neural network to identify defects in the powder spreading image, and determine the defect location and defect type; The positional relationship between the defect position and the scanning path is determined, and the powder spreading defects are graded according to the positional relationship and the defect type.

2. The powder spreading quality detection and grading method according to claim 1, It is characterized in that The method of using a convolutional neural network to identify defects in the powder spreading image and determine the defect location and defect type includes: The size of the powder spreading interconnection defect is identified by using a convolutional neural network on the powder spreading image, and the defect position is framed by a minimum rectangular frame that can enclose the interconnection defect, and the rectangular frame area is used to replace the actual powder spreading defect position to participate in the position detection operation; According to the image features of the powder spreading image, a convolutional neural network is used to determine the defect type.

3. The powder spreading quality detection and grading method according to claim 1, It is characterized in that Determining the positional relationship between the defect position and the scanning path, and performing graded processing on the powder spreading defect according to the positional relationship and the defect type, includes: Determining whether the defect position overlaps with the scanning path; If yes, the powder spreading defect is graded according to the defect type; otherwise, no graded processing is performed.

4. The powder spreading quality detection and grading method according to claim 3, It is characterized in that The determining whether the defect position overlaps with the scanning path includes: Slice the three-dimensional digital model of the part to be formed to obtain the scanning path image of the forming layer; It is determined whether the defect position overlaps with the scanning path according to the scanning path image and the powder spreading image.

5. The powder spreading quality detection and grading method according to claim 4, It is characterized in that The determining, based on the scanning path image and the powder spreading image, whether the defect position overlaps with the scanning path includes: Filling the grayscale value of the scanned area in the scan path image with [0, 0, 0], and converting the powder spreading image in the pixel coordinate system into the machine tool coordinate system; In the machine tool coordinate system, the scanning path image is used as a foreground image, and the powder spreading image is used as a background image, and the scanning path image and the powder spreading image are synthesized; It is determined whether there is a connected area with a gray value of [0,0,0] in the rectangular frame. If so, the defect position overlaps with the scanning path, otherwise, there is no overlap.

6. A powder quality detection and grading device, It is characterized in that include: An image acquisition module is used to acquire a powder spreading image by shooting a CCD camera that fully covers the powder bed; A defect recognition module is used to use a convolutional neural network to perform defect recognition on the powder spreading image to determine the defect location and defect type; The grading processing module is used to determine the positional relationship between the defect position and the scanning path, and to perform grading processing on the powder spreading defects according to the positional relationship and the defect type.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the powder spreading quality detection and grading method as described in any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium storing a computer program. It is characterized in that When the computer program is executed by a processor, the powder spreading quality detection and grading method according to any one of claims 1 to 5 is implemented.

Images