Porous structure surface layer detection device and method based on brightness change of rotating light source

By combining a rotating light source visual inspection module and an image processing module with AB axis rotation adjustment and deep learning algorithms, the problem of distinguishing the surface and underlying structures of porous spherical joint components was solved, improving detection accuracy and efficiency and ensuring the stability of the prosthesis.

CN121027112AInactive Publication Date: 2025-11-28ZHEJIANG WANLI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202511204068.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-08-27
Publication Date
2025-11-28
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing technologies struggle to effectively distinguish between the surface and underlying structures of porous spherical joint components, leading to difficulties in detecting tendon fractures and affecting the fixation effect between the prosthesis and bone tissue.

Method used

The system employs a rotating light source visual inspection module, an AB axis rotation adjustment module, and an image processing module. By varying the brightness of the rotating light source and processing images, it distinguishes the surface and bottom layers of porous structural components and uses deep learning algorithms for rib breakage detection.

Benefits of technology

It significantly improves the accuracy and efficiency of testing porous structural components, ensures the fixation effect between the prosthesis and bone tissue, and reduces the risk of prosthesis sinking or rotation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121027112A_ABST
    Figure CN121027112A_ABST
Patent Text Reader

Abstract

The invention discloses a porous structure surface layer detection device and method based on brightness change of a rotary light source, a rotary light source stepping motor in a visual detection module drives an LED lamp support frame to rotate around a detected object and drives an LED lamp to rotate and light in a dark field environment, and an industrial camera is arranged above the detected object to synchronously collect an image of the detected object; a rotating platform A and a rotating platform B in the AB-axis rotation adjusting module are driven by a stepping motor A and a stepping motor B to rotate and adjust a detected object, it is ensured that a target detection area right faces an industrial camera lens, and graying, binarization processing and multi-frame fusion are conducted on images collected by the industrial camera through the image processing module. The surface layer of the multi-frame fused porous structural part image is bright, the bottom layer is black, and it is ensured that the device can distinguish the surface layer structure and the bottom layer structure of the porous structural part. According to the invention, through function complementation and dynamic optimization, the detection precision, efficiency and reliability are significantly enhanced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of metal additive manufacturing quality inspection technology, and in particular to a device and method for inspecting porous surface layers based on the brightness variation of a rotating light source. Background Technology

[0002] After the production of porous spherical joints using metal additive manufacturing is completed, quality inspection is usually required to check for surface fractures. Currently, there are two common inspection methods: one relies on manual visual inspection, which is inefficient and has a high rate of missed detections; the other uses machine vision. This method typically uses an industrial camera to simultaneously capture images by directly illuminating the part under test with a light source. However, for porous spherical joints, this method makes it difficult to distinguish between the surface and underlying structures, causing many inconveniences for subsequent fracture detection. If fractures occur, the fixation between the prosthesis and bone tissue will be weakened. This may cause the prosthesis to gradually sink or rotate within the bone, affecting the normal anatomical structure and motor function of the joint.

[0003] Patent application number 2024114829076, entitled "A Multi-Angle Ring Uniform Illumination Device and Design Method," mentions adding an aspherical plano-convex lens outside the LED light source and symmetrically and uniformly installing multiple dimming light sources on a ring light source mounting base. A motor-controlled angle adjustment mechanism enables multi-angle adjustment of the light source, thereby reducing interference from strong reflected light. However, this invention cannot adjust the placement of the product being tested in multiple directions and angles. For products with irregular shapes, this device has difficulty detecting all structural features. Patent application number 2023111814144, entitled "A Surface Detection Device for Reflective Objects," mentions using diffuse reflection to reduce strong reflection. By staggering the imaging and illumination components, it avoids direct reflected light hitting the lens, thus reducing interference from reflected light on the detection. However, for multi-layered porous structures, this invention cannot distinguish between the surface and underlying structures, making it difficult to detect the surface. Summary of the Invention

[0004] To address the aforementioned shortcomings, this invention proposes a porous surface detection device and method based on the brightness variation of a rotating light source. Through functional complementarity and dynamic optimization, the detection accuracy, efficiency, and reliability are significantly enhanced.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a porous structure surface detection device based on the brightness variation of a rotating light source, comprising: The rotating light source visual inspection module includes a rotating light source stepper motor, an LED lamp support frame, an industrial camera, and LED lamps. The rotating light source stepper motor drives the LED lamp support frame to rotate around the porous structure being inspected, causing the LED lamps to rotate and illuminate in a dark environment. The industrial camera is set above the porous structure to synchronously acquire images of the porous structure, making its surface appear bright and its bottom layer appear as alternating bright and dark areas. The AB axis rotation adjustment module includes an A rotating platform, a B rotating platform, a rotating platform support frame, a chuck, an A stepper motor, and a B stepper motor. The chuck is used to fix the porous structure to be inspected. The A rotating platform and the B rotating platform are driven by the A stepper motor and the B stepper motor, respectively, to rotate the chuck, so as to realize the automatic rotation adjustment of the porous structure to be inspected in the A and B directions, ensuring that the target inspection area is facing the industrial camera lens. The image processing module performs grayscale and binarization processing on the images captured by the industrial camera, and then performs multi-frame fusion of the binarized images. The surface of the porous structure image after multi-frame fusion is bright and the bottom layer is black, ensuring that the device can distinguish between the surface structure and the bottom layer of the porous structure.

[0006] As an improvement, rotary platform A is rotatably connected to the rotary platform support frame via stepper motor A to adjust the tilt angle of the porous structure being tested in the horizontal direction. Meanwhile, rotary platform B is set inside rotary platform A and is rotatably connected to rotary platform A via stepper motor B to adjust the tilt angle of the porous structure being tested in the vertical direction.

[0007] As an improvement, the LED lamp support frame includes a horizontally arranged motor mounting part with mounting holes. The mounting holes are fixedly connected to the rotating light source stepper motor by sleeves mounted on the outer wall of the upper end of the rotating light source stepper motor output shaft. The rotating light source vision inspection module also includes a bushing sleeve sleeved on the outer wall of the lower end of the rotating light source stepper motor output shaft and fixedly connected to the rotating light source stepper motor output shaft. The industrial camera is connected to the rotating light source stepper motor output shaft through the bushing sleeve.

[0008] As an improvement, the LED lamp support frame also includes an LED lamp mounting part that extends outward from the motor mounting part in an arc shape, and the LED lamp is fixedly mounted at the end of the LED lamp mounting part.

[0009] As an improvement, The range of α is 0-255, and the range of T is 10-20. When the layer of the porous structure being tested is fine rib, the value of α is 0.6-1; when the layer of the porous structure being tested is coarse rib, the value of α is 1.1-1.3.

[0010] As an improvement, =255 and The time indicates that the position (x, y) in the merged image is determined to be white. =0 and The time indicates that the position (x, y) in the fused image is determined to be black.

[0011] As an improvement, the detection device also includes a screw drive module for adjusting the height of the rotating light source vision inspection module, and a mounting frame for mounting the screw drive module and the AB axis rotation adjustment module. The screw drive module includes a transmission column vertically arranged between the mounting frames, a guide column corresponding to the transmission column, a lifting motor connected to the upper end of the transmission column to drive the transmission column to rotate, a lifting support platform for mounting the rotating light source vision inspection module that can move up and down along the direction of the transmission column, and a bearing seat that forms a rotational support with the lower end of the transmission column through a thrust ball bearing assembly. One end of the lifting support platform is provided with a threaded transmission structure sleeved on the transmission column to make it rotatably connected to the transmission column, while the other end of the lifting support platform is provided with a linear bearing sleeved on the guide column to make it slidably connected to the guide column. The lifting motor and the bearing seat are both fixedly connected to the mounting frame.

[0012] As an improvement, the chuck surface is provided with several limiting blocks arranged in an orthogonal symmetrical manner to form a cross-shaped positioning reference, thereby forming a four-way constraint positioning structure for the porous structural component being inspected.

[0013] The detection method of any of the above-mentioned porous structure surface detection devices based on the brightness variation of a rotating light source includes the following steps: S1: Fix the porous structural component on the chuck, and the AB axis rotation adjustment module automatically adjusts the rotation angle to ensure that the target detection area of ​​the porous structural component is facing the industrial camera lens. S2: The screw drive module automatically adjusts the camera height to a distance h from the top of the workpiece being inspected, ensuring the camera's field of view clearly captures the top area of ​​the workpiece, and that the top field of view occupies a significant portion of the metal workpiece's area. to The final camera height is determined to be h + Δh using an autofocus algorithm. S3: In a dark environment, while the light source is rotated by the rotating light source vision inspection module, the industrial camera simultaneously captures images. The captured images are then processed by the image processing module to first convert them to grayscale and then to binarization. in, This represents the pixel value of the i-th input grayscale image at position (x, y); T is the threshold value, which is determined by the ambient light. It is an empirical threshold; α is an empirical coefficient; (x,y) represents the pixel value at position (x,y) of the i-th output binarized image, where 0 represents black and 255 represents white; Then, the binarized images are fused from multiple frames: in, This represents the pixel value at position (x, y) in the fused image; S4: Finally, anchor-based deep learning algorithms, including but not limited to YOLO, are used to detect rib breaks in the extracted surface structure.

[0014] As an improvement, in multi-frame fusion processing, the confidence threshold is dynamically adjusted according to the ambient light intensity, and the rotation speed of the rotating light source is matched with the shooting frame rate of the industrial camera to ensure that each frame corresponds to a different lighting angle.

[0015] Compared with the prior art, the advantages of the present invention are as follows: This patent constructs a complete detection system from mechanical adjustment to optical imaging and intelligent analysis through the deep collaboration of a rotating light source visual inspection module, an AB-axis rotation adjustment module, an image processing module, and a screw drive module. The AB-axis rotation adjustment module employs a nested dual-rotation platform with A-axis horizontal / B-axis vertical dual-degree-of-freedom adjustment and an orthogonal symmetrical limit chuck, achieving three-dimensional spatial positioning capability and four-way constraint stability. The screw drive module, through a threaded drive column and corresponding guide column structure, combined with a thrust ball bearing assembly, achieves load distribution and high-precision adjustment of the rotating light source visual inspection module's lifting and lowering. The rotating light source visual inspection module uses a ring... The shape-based light source layout and top coaxial camera design, combined with 360° continuous rotating illumination, create an optical feature of alternating bright and dark areas on the surface in dark environments. Multiple frames of images simultaneously acquired by the industrial camera are processed by the image processing module through grayscale conversion, binarization, and multi-frame fusion to clearly distinguish the surface and underlying structures. This, along with the AB axis rotation adjustment module and the screw drive module, forms a three-level collaborative process of "positioning-height adjustment-detection." Relying on a stepper motor encoder to monitor the rotation angle and lifting height in real time, the automatic focusing algorithm dynamically adjusts the imaging parameters. The fracture detection result triggers a secondary adjustment of the AB axis module, ultimately achieving intelligent detection through "perception-decision-execution." Simultaneously, the light shield design suppresses ambient light interference, ensuring the system's anti-interference capability. This structural innovation not only improves the adaptability of detecting complex porous structures but also significantly enhances detection accuracy, efficiency, and reliability through functional complementarity and dynamic optimization. Attached Figure Description

[0016] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: Figure 1A schematic diagram of a porous structure surface detection device based on the brightness change of a rotating light source, on which the porous structure to be detected is placed; Figure 2 This is a schematic diagram of a porous surface detection device based on the brightness variation of a rotating light source; Figure 3 This is a schematic diagram of the AB axis rotation adjustment module. Figure 4 This is a schematic diagram of the rotating light source visual inspection module. Figure 5 This is a schematic diagram of the screw drive module structure; Figure 6 This is a flowchart of a detection device for porous surface layers based on the brightness variation of a rotating light source.

[0017] The markings in the above diagrams are as follows: 1. Rotating light source vision inspection module; 1.1 Rotating light source stepper motor; 1.2 LED lamp support frame; 1.2.1 Motor mounting part; 1.2.2 LED lamp mounting part; 1.3 Industrial camera; 1.4 LED lamp; 1.5 Bushing; 2. AB axis rotation adjustment module; 2.1 A rotating platform; 2.2 B rotating platform; 2.3 Rotating platform support frame; 2.4 Chuck; 2.4.1 Limit block; 2.5 A stepper motor; 2.6 B stepper motor; 2.7 Motor support base; 2.8 Motor fixing plate; 3. Screw drive module; 3.1 Drive column; 3.2 Guide column; 3.3 Lifting motor; 3.4 Lifting support platform; 3.4.1 Threaded drive structure; 3.4.2 Linear bearing; 3.5 Bearing seat; 4. Mounting bracket; 4.1 Base; 4.2 Support plate; 4.3 Support column. Detailed Implementation

[0018] In this invention, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "planar direction", "circumferential", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention.

[0019] like Figures 1 to 5As shown, a porous structure surface inspection device based on the brightness variation of a rotating light source includes a rotating light source vision inspection module, an AB axis rotation adjustment module, and an image processing module. The rotating light source vision inspection module includes a rotating light source stepper motor, an LED lamp support frame, an industrial camera, and LED lamps. The rotating light source stepper motor drives the LED lamp support frame to rotate around the porous structure being inspected, causing the LED lamps to rotate and illuminate in a dark environment. The industrial camera is positioned above the porous structure to simultaneously acquire images of the porous structure, making its surface appear bright and its underlying surface exhibit alternating bright and dark characteristics. The AB axis rotation adjustment module includes an A rotating platform, a B rotating platform, and a rotating platform... The system includes a support frame, a chuck, stepper motor A, and stepper motor B. The chuck is used to fix the porous structure to be inspected. Rotary platforms A and B are driven by stepper motors A and B, respectively, to rotate the chuck, enabling automatic rotation adjustment of the porous structure in both directions (A and B) to ensure that the target inspection area is directly facing the industrial camera lens. The image processing module performs grayscale and binarization processing on the images acquired by the industrial camera, and then fuses multiple binarized images into a multi-frame image. The multi-frame fused image of the porous structure shows a bright surface and a black bottom layer, ensuring that the device can distinguish between the surface and bottom layers of the porous structure.

[0020] Rotary platform A is rotatably connected to the rotating platform support frame via stepper motor A to adjust the tilt angle of the porous structural component being inspected in the horizontal direction. Meanwhile, rotary platform B is set inside rotary platform A and is rotatably connected to rotary platform A via stepper motor B to adjust the tilt angle of the porous structural component being inspected in the vertical direction.

[0021] The rotating light source vision inspection module also includes a motor support base for mounting the housing of stepper motor A and a motor mounting plate for mounting the housing of stepper motor B.

[0022] The LED light support frame includes a horizontally arranged motor mounting part with mounting holes. The mounting holes are fixedly connected to the rotating light source stepper motor by sleeves installed on the outer wall of the upper end of the rotating light source stepper motor output shaft. The rotating light source vision inspection module also includes a bushing sleeve installed on the outer wall of the lower end of the rotating light source stepper motor output shaft and fixedly connected to the rotating light source stepper motor output shaft. The industrial camera is connected to the rotating light source stepper motor output shaft through the bushing sleeve.

[0023] The LED light support frame also includes an LED light mounting part that extends outward from the motor mounting part in an arc shape, and the LED light is fixedly mounted at the end of the LED light mounting part.

[0024] The detection device also includes a screw drive module for adjusting the height of the rotating light source visual inspection module, and a mounting frame for mounting the screw drive module and the AB axis rotation adjustment module. The screw drive module includes a transmission column vertically arranged between the mounting frames, a guide column corresponding to the transmission column, a lifting motor connected to the upper end of the transmission column to drive the transmission column to rotate, a lifting support platform for mounting the rotating light source visual inspection module that can move up and down along the direction of the transmission column, and a bearing seat that forms a rotational support with the lower end of the transmission column through a thrust ball bearing assembly. One end of the lifting support platform is provided with a threaded transmission structure sleeved on the transmission column to make it rotatably connected to the transmission column, while the other end of the lifting support platform is provided with a linear bearing sleeved on the guide column to make it slidably connected to the guide column. The lifting motor and the bearing seat are both fixedly connected to the mounting frame.

[0025] The mounting frame includes a base, a support plate corresponding to the base, four support columns symmetrically arranged at the four corners between the base and the support plate, a transmission column between the two support columns at the front end, a guide column between the two support columns at the rear end, a lifting motor housing fixedly connected to the upper end face of the support plate, and a bearing seat, a rotating platform support frame, a motor support base fixedly connected to the base.

[0026] The chuck surface is provided with several limiting blocks arranged in an orthogonal and symmetrical manner to form a cross-shaped positioning reference, thereby forming a four-way constraint positioning structure for the porous structural component being inspected.

[0027] Preferably, the above-mentioned fixed connection method adopts bolt connection.

[0028] A detection method for a porous structure surface detection device based on the brightness variation of a rotating light source includes the following steps: S1: Fix the porous structural component on the chuck, and the AB axis rotation adjustment module automatically adjusts the rotation angle to ensure that the target detection area of ​​the porous structural component is facing the industrial camera lens. S2: The screw drive module automatically adjusts the camera height to a distance h from the top of the workpiece being inspected, ensuring the camera's field of view clearly captures the top area of ​​the workpiece, and that the top field of view occupies a significant portion of the metal workpiece's area. to The final camera height is determined to be h + Δh using an autofocus algorithm. S3: In a dark environment, while the light source is rotated by the rotating light source vision inspection module, the industrial camera simultaneously captures images. The captured images are then processed by the image processing module to first convert them to grayscale and then to binarization. in, This represents the pixel value at position (x, y) of the i-th input grayscale image. The range is 0-255; T is the threshold, which ranges from 10-20, and the specific value is determined by the ambient light. α is an empirical threshold; α is an empirical coefficient. When the layer of the porous structure being tested is fine rib, the value of α ranges from 0.6 to 1; when the layer of the porous structure being tested is coarse rib, the value of α ranges from 1.1 to 1.3. (x,y) represents the pixel value at position (x,y) of the i-th output binarized image, where 0 represents black and 255 represents white; Then, the binarized images are fused from multiple frames: in, This represents the pixel value at position (x, y) in the fused image. =255 and The time indicates that the position (x, y) in the merged image is determined to be white. =0 and The time indicates that the position (x, y) in the fused image is determined to be black; S4: Finally, anchor-based deep learning algorithms such as YOLO are used to detect rib breaks in the extracted surface structure.

[0029] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0030] The units described in some embodiments of this disclosure can be implemented in software or in hardware. The described units can also be located in a processor, and the functions described above can be performed at least in part by one or more hardware logic components.

[0031] The above description only illustrates the preferred embodiments of the present invention and should not be construed as limiting the scope of the claims. The present invention is not limited to the above embodiments, and variations in its specific structure are permitted. All modifications made within the scope of the independent claims of this invention are also within the scope of protection of this invention.

Claims

1. A porous structure surface detection device based on the brightness variation of a rotating light source, characterized in that, include: The rotating light source visual inspection module includes a rotating light source stepper motor, an LED lamp support frame, an industrial camera, and an LED lamp. The rotating light source stepper motor drives the LED lamp support frame to rotate around the porous structure being inspected, causing the LED lamp to rotate and illuminate in a dark environment. The industrial camera is set above the porous structure to synchronously acquire images of the porous structure, making its surface appear bright and its bottom layer appear as alternating bright and dark areas. The AB axis rotation adjustment module includes an A rotation platform, a B rotation platform, a rotation platform support frame, a chuck, an A stepper motor, and a B stepper motor. The chuck is used to fix the porous structural component to be inspected. The A rotation platform and the B rotation platform are driven by the A stepper motor and the B stepper motor, respectively, to rotate the chuck, thereby realizing the automatic rotation adjustment of the porous structural component to be inspected in both directions A and B, ensuring that the target inspection area is facing the industrial camera lens. The image processing module performs grayscale and binarization processing on the images captured by the industrial camera, and then performs multi-frame fusion of the binarized images. The surface of the porous structure image after multi-frame fusion is bright and the bottom layer is black, ensuring that the device can distinguish between the surface structure and the bottom layer of the porous structure.

2. The porous structure surface detection device based on the brightness variation of a rotating light source according to claim 1, characterized in that, The A rotating platform is rotatably connected to the rotating platform support frame via an A stepper motor to adjust the tilt angle of the porous structure being tested in the horizontal direction. Meanwhile, the B rotating platform is located inside the A rotating platform and is rotatably connected to the A rotating platform via a B stepper motor to adjust the tilt angle of the porous structure being tested in the vertical direction.

3. The porous structure surface detection device based on the brightness variation of a rotating light source according to claim 1, characterized in that, The LED lamp support frame includes a horizontally arranged motor mounting part with a mounting hole. The mounting hole is fixedly connected to the rotating light source stepper motor by sleeved on the outer wall of the upper end of the output shaft of the rotating light source stepper motor. The rotating light source vision inspection module also includes a bushing sleeve sleeved on the outer wall of the lower end of the output shaft of the rotating light source stepper motor and fixedly connected to the output shaft of the rotating light source stepper motor. The industrial camera is connected to the output shaft of the rotating light source stepper motor through the bushing sleeve.

4. The porous structure surface detection device based on the brightness variation of a rotating light source according to claim 3, characterized in that, The LED lamp support frame also includes an LED lamp mounting part that extends outward from the motor mounting part in an arc shape, and the LED lamp is fixedly mounted at the end of the LED lamp mounting part.

5. The porous structure surface detection device based on the brightness variation of a rotating light source according to claim 1, characterized in that, The detection device also includes a helical transmission module for adjusting the height of the rotating light source visual inspection module, and a mounting frame for mounting the helical transmission module and the AB axis rotation adjustment module. The helical transmission module includes a transmission column vertically arranged between the mounting frames, a guide column corresponding to the transmission column, a lifting motor connected to the upper end of the transmission column to drive the transmission column to rotate, a lifting support platform for mounting the rotating light source visual inspection module that can move up and down along the direction of the transmission column, and a bearing seat that forms a rotational support with the lower end of the transmission column through a thrust ball bearing assembly. One end of the lifting support platform is provided with a threaded transmission structure sleeved on the transmission column to make it rotatably connected to the transmission column, while the other end of the lifting support platform is provided with a linear bearing sleeved on the guide column to make it slidably connected to the guide column. The lifting motor and the bearing seat are both fixedly connected to the mounting frame.

6. The porous structure surface detection device based on the brightness variation of a rotating light source according to claim 1, characterized in that, The chuck surface is provided with several limiting blocks arranged in an orthogonal and symmetrical manner to form a cross-shaped positioning reference, thereby forming a four-way constraint positioning structure for the porous structural component being inspected.

7. The detection method of a porous structure surface detection device based on the brightness change of a rotating light source according to any one of claims 1-6, characterized in that, Includes the following steps: S1: Fix the porous structural component on the chuck, and the AB axis rotation adjustment module automatically adjusts the rotation angle to ensure that the target detection area of ​​the porous structural component is facing the industrial camera lens. S2: The screw drive module automatically adjusts the camera height to a distance h from the top of the workpiece being inspected, ensuring the camera's field of view clearly captures the top area of ​​the workpiece, and that the top field of view occupies a significant portion of the metal workpiece's area. to The final camera height is determined to be h + Δh using an autofocus algorithm. S3: In a dark environment, while the light source is rotated by the rotating light source vision inspection module, the industrial camera simultaneously captures images. The captured images are then processed by the image processing module to first convert them to grayscale and then to binarization. in, This represents the pixel value of the i-th input grayscale image at position (x, y); T is the threshold value, which is determined by the ambient light. It is an empirical threshold; α is an empirical coefficient; (x,y) represents the pixel value at position (x,y) of the i-th output binarized image, where 0 represents black and 255 represents white; Then, the binarized images are fused from multiple frames: in, This represents the pixel value at position (x, y) in the fused image; S4: Finally, anchor-based deep learning algorithms, including but not limited to YOLO, are used to detect rib breaks in the extracted surface structure.

8. The method for detecting porous structure surfaces based on the brightness variation of a rotating light source according to claim 7, characterized in that, The range of α is 0-255, and the range of T is 10-20. When the layer of the porous structure being tested is fine rib, the value of α is 0.6-1; when the layer of the porous structure being tested is coarse rib, the value of α is 1.1-1.

3.

9. The porous structure surface detection device based on the brightness variation of a rotating light source according to claim 7, characterized in that, =255 and The time indicates that the position (x, y) in the merged image is determined to be white. =0 and The time indicates that the position (x, y) in the fused image is determined to be black.

10. The method for detecting porous structure surface layers based on the brightness variation of a rotating light source according to claim 9, characterized in that: In the multi-frame fusion process, the confidence threshold is dynamically adjusted according to the ambient light intensity, and the rotation speed of the rotating light source is matched with the shooting frame rate of the industrial camera to ensure that each frame corresponds to a different lighting angle.