A screw defect detection device and method based on image processing

Abstract

The present application relates to the technical field of screw defect detection, and discloses a screw defect detection device and method based on image processing, which aims at the problems of low efficiency, weak adaptability, easy to cause secondary damage of screws and the like of the existing detection technology; the device comprises a feeding vibration disc and an image detection mechanism, the image detection mechanism comprises a supporting base, a screw conveying assembly, a selection assembly, a driving assembly, a power assembly, an adjustable support, an image acquisition assembly and a PLC controller; the detection method realizes screw defect detection through the steps of equipment debugging, automatic feeding, image acquisition, defect identification, automatic sorting and cycle detection. The present application adopts a non-contact image acquisition mode, matches a deep learning defect detection algorithm, can accurately identify various defects such as screw head cracking and thread defects, and is compatible with detection of screws of different specifications through an adjustable structure, without frequent replacement of tooling.

Description

Technical Field

[0001] This invention relates to the field of screw defect detection technology, and in particular to a screw defect detection device and method based on image processing. Background Technology

[0002] As a core component in mechanical connections, screws directly affect the assembly reliability and service life of the entire machine. In large-scale production scenarios, screws are prone to various defects during processing due to factors such as mold wear, material impurities, and fluctuations in processing parameters. If defective screws flow into subsequent assembly stages, they may lead to loose connections, equipment failure, or even safety hazards.

[0003] Currently, screw defect detection in the industry mainly employs two methods: manual visual inspection and traditional mechanical inspection. Manual visual inspection relies on operator experience and suffers from low inspection efficiency (the inspection volume per shift is usually no more than 5,000 pieces), unstable inspection accuracy (susceptible to fatigue and subjective judgment), and high missed detection rate (difficult to identify minute defects), making it unsuitable for the high-efficiency production demands of modern assembly lines. Traditional mechanical inspection equipment mostly uses contact inspection methods, measuring dimensions after positioning with tooling fixtures. This not only has poor adaptability (requiring the replacement of special fixtures for different screw sizes) but also cannot detect non-dimensional defects such as surface scratches and micro-cracks, limiting the inspection range. Furthermore, contact inspection can easily cause secondary damage to the screw surface, affecting the product's appearance quality. Therefore, this paper proposes to develop a screw defect detection device and method based on image processing. Summary of the Invention

[0004] In view of the above-mentioned problems existing in the prior art, the purpose of the present invention is to provide a screw defect detection device and method based on image processing, so as to realize non-contact and automated detection of screw defects and improve detection efficiency.

[0005] The present invention is achieved by the following technical solution: a screw defect detection device and method based on image processing.

[0006] As a further improvement to the above solution, a screw defect detection device based on image processing includes a feeding vibratory feeder and an image detection mechanism; the feeding vibratory feeder is used for automatic feeding and orderly conveying of screws; the image detection mechanism is located directly below the conveying port of the feeding vibratory feeder. The image detection mechanism includes a support base, a screw conveying assembly, a selection assembly, a drive assembly, a power assembly, two sets of support frames, an adjustment bracket, an image acquisition assembly, and a PLC controller. The support base is fixed to the front side of the output port of the feeding vibratory feeder, providing support for the image detection mechanism; the screw conveying assembly is installed on the top of the support base and docks with the feeding vibratory feeder to receive and convey screws. The selection component is rotatably mounted inside the front side of the screw conveying component and is used to sort screws according to the detection results; the drive component is installed between the selection component and the support base and is used to transmit power; the power component is installed on one side of the support base and its output end is connected to the drive component to provide driving force for the selection component to flip. The two sets of support frames are symmetrically installed on both sides of the screw conveying assembly; the adjusting bracket is installed between the two sets of support frames and can be slidably adjusted along the height direction of the support frames. The image acquisition component is installed at the bottom of the adjustment bracket to acquire images of the screw from all directions; the PLC controller is installed on one side of the support base to receive image signals, analyze and process them, and send control commands to realize automated control of the device and interface with the production line control system.

[0007] As a further improvement to the above solution, the support base includes a support frame and a support bracket; the support frame is a frame structure welded from rectangular steel pipes; the support bracket is fixed inside the support frame, and its top is fully in contact with the bottom of the screw conveying assembly.

[0008] As a further improvement to the above solution, the screw conveying assembly includes a conveying slide; a U-shaped groove is provided in the middle of the bottom of the conveying slide, and a stepped groove is integrally formed at the bottom of the U-shaped groove; multiple evenly distributed fixing screw holes are provided on the side walls of both sides of the conveying slide.

[0009] As a further improvement to the above solution, the selection component includes a V-shaped selection plate adapted to the U-shaped slot; the bottom of the V-shaped selection plate is symmetrically welded with support ears, and the support ears are provided with mounting holes with wear-resistant bushings.

[0010] As a further improvement to the above solution, the drive assembly includes a driven gear, a drive gear, a driven shaft, and a drive shaft. The transmission ratio between the driven gear and the driving gear is 1:1; the driven gear is fixed to the outer wall of the driven shaft by a flat key, and the driven shaft passes through the support lug. The drive gear is mounted on one end of the drive shaft via a flat key, and the drive gear meshes with the driven gear for transmission; wear-resistant bearings are installed at the connection points between the driven shaft, the drive shaft and the support base.

[0011] As a further improvement to the above solution, the adjusting bracket includes an I-shaped plate, and the four ends of the I-shaped plate are fixed with limiting sleeves; the limiting sleeves are sleeved on the outer wall of the support frame; a nut is fixed on the outer wall of the limiting sleeve, and the nut is internally threaded with a threaded rod with an anti-slip knob, and the threaded end of the threaded rod passes through the limiting sleeve and abuts against the side wall of the support frame.

[0012] As a further improvement to the above solution, the image acquisition component includes a U-shaped bracket, two fixing clips (first type), two fixing clips (second type), two rectangular frames, and an industrial camera; The first fixing clip is symmetrically fixed at both ends of the U-shaped bracket, and the second fixing clip is fixed to the top of the first fixing clip by bolts. The two together form a clamp structure that is clamped in the middle of the I-shaped plate; the two rectangular frames are symmetrically fixed to the bottom of the U-shaped bracket by bolts, and the industrial camera is embedded in the rectangular frame.

[0013] A screw defect detection method based on image processing, implemented using the aforementioned detection device, includes the following steps: Step 1: Equipment debugging, adjust the height of the support bracket and the vibration frequency of the feeding vibratory feeder, import standard image parameters and defect judgment thresholds, and complete the initial debugging; Step 2: Automatic feeding. The feeding vibratory feeder orients and aligns the screws and transports them to the screw conveying assembly. Step 3: Image acquisition. The industrial camera captures real-time images of the screw that has rolled into the acquisition area, collecting omnidirectional image data and transmitting it to the PLC controller. Step 4: Defect identification. The PLC controller uses a built-in deep learning algorithm to compare and analyze image data to determine whether the screw has defects and the type of defects. Step 5: Automatic sorting. Qualified screws are conveyed along the forward track to the receiving mechanism, while defective screws are discharged by flipping through the selection component, which then resets. Step Six: Cyclic inspection, repeating steps two through five to achieve continuous and automated screw inspection and sorting.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention employs a combination of an industrial CCD camera and a deep learning defect detection algorithm to achieve non-contact, automated detection of screw defects. It can accurately identify minute defects and solves the problems of unstable accuracy and high missed detection rate of manual inspection. This invention, through its adjustable bracket and V-shaped selection plate, is compatible with the inspection of various screws within different specification ranges, eliminating the need for frequent changes of tooling fixtures. It adapts to the inspection needs of different batches and specifications of screws, and its versatility is significantly better than traditional mechanical inspection equipment. This invention employs a non-contact image acquisition method. During the screw conveying process, a wear-resistant engineering plastic conveying slide structure is used to avoid hard contact between the screw surface and equipment components, effectively preventing secondary damage to the screw surface and ensuring the product's appearance quality. Attached Figure Description

[0015] Figure 1 This is a three-dimensional stereoscopic view of the screw defect detection device based on image processing according to the present invention; Figure 2 This is a top-view three-dimensional perspective view of the image detection mechanism of the screw defect detection device based on image processing according to the present invention. Figure 3 This is a three-dimensional perspective view of the image detection mechanism of the screw defect detection device based on image processing according to the present invention, viewed from below. Figure 4 This is a three-dimensional perspective view of the support base of the screw defect detection device based on image processing according to the present invention; Figure 5 This is a three-dimensional stereoscopic view of the screw conveying component of the screw defect detection device based on image processing of the present invention; Figure 6 This is a three-dimensional stereoscopic view of the adjustment bracket of the screw defect detection device based on image processing of the present invention; Figure 7 This is a three-dimensional stereoscopic view of the image acquisition component of the screw defect detection device based on image processing of the present invention; Figure 8 This is a three-dimensional stereoscopic view of the selected component of the screw defect detection device based on image processing of the present invention in the state of turning from forward to reverse. Figure 9 This is a flowchart of the screw defect detection method based on image processing according to the present invention.

[0016] Explanation of key symbols: 1. Feeding vibratory feeder; 2. Image detection mechanism; 21. Support base; 211. Support frame; 212. Support bracket; 22. Screw conveying assembly; 221. Conveyor slide; 222. U-shaped groove; 223. Stepped groove; 224. Fixing screw hole; 23. Selection assembly; 231. V-shaped selection plate; 232. Support ear plate; 24. Drive assembly; 241. Driven gear; 242. Drive gear; 243. Driven shaft 244. Drive shaft rod; 25. Power assembly; 251. Servo motor; 252. Reducer; 26. Support frame; 27. Adjusting bracket; 271. I-beam plate; 272. Limiting sleeve; 273. Nut; 274. Threaded rod; 28. Image acquisition assembly; 281. U-shaped bracket; 282. Fixing clip one; 283. Fixing clip two; 284. Rectangular frame; 285. Industrial camera; 29. ​​PLC controller. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0018] Obviously, the described embodiments are only a part of the embodiments of this disclosure, not all of them. All other embodiments obtained by those skilled in the art based on the described embodiments of this disclosure without inventive effort are within the scope of protection of this disclosure.

[0019] Unless otherwise defined, the technical or scientific terms used in this disclosure shall have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms "comprising" or "including," and similar terms used in this disclosure, mean that an element or object preceding the term encompasses the elements or objects listed following the term and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but may also include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; these relative positional relationships may change accordingly when the absolute position of the described objects changes.

[0020] To keep the following description of the embodiments of this disclosure clear and concise, detailed descriptions of known functions and known components are omitted.

[0021] Please combine Figure 1-8 As shown in the figure, the screw defect detection device based on image processing provided by this invention can effectively identify common defects such as screw head cracks, thread defects, dimensional deviations, and surface scratches. It possesses core advantages such as high detection accuracy, fast response speed, and strong adaptability. It includes a feeding vibratory feeder 1 and an image detection mechanism 2. The feeding vibratory feeder 1 is used to automatically feed and orderly convey screws, arranging them along the feeder track through vibration excitation before conveying them to the output port. The image detection mechanism 2 is located directly below the conveying port of the feeding vibratory feeder 1 and is responsible for real-time image acquisition and defect analysis of the screws during the conveying process. Figure 2-3 As shown, the image detection mechanism 2 further includes a support base 21, a screw conveying assembly 22, a selection assembly 23, a drive assembly 24, a power assembly 25, two sets of support frames 26, an adjustment bracket 27, an image acquisition assembly 28, and a PLC controller 29. The support base 21 is bolted to the front of the output port of the feeding vibratory feeder 1, providing a stable and reliable support foundation for the entire image detection mechanism 2. The screw conveying assembly 22 is fastened to the top of the support base 21 with high-strength internal hex bolts, which are aligned with the conveying port of the feeding vibratory feeder 1 to ensure that the screws fall smoothly into the conveying channel. The receiving end is located directly below the conveying port of the feeding vibratory feeder 1 for accurate reception and conveying of the screws. This allows the screws to roll smoothly along the preset track of the screw conveying assembly 22, avoiding deviation, jamming, or other issues, so that the image acquisition assembly 28 can accurately receive and convey the screws. During the rolling process, clear, all-around image capture is completed. The selection component 23 is rotatably mounted inside the front side of the screw conveying component 22, and its rotation angle can be precisely controlled by the PLC controller 29. Based on the image acquisition component 28's detection results, it can achieve forward and backward flipping under the coordinated drive of the power component 25 and the drive component 24. When a defect is detected in the screw, the PLC controller 29 immediately sends a control signal, driving the selection component 23 to switch from forward rotation to reverse rotation, causing the defective screw to be discharged from the rear discharge port of the screw conveying component 22 (e.g., ...). Figure 8 As shown), qualified screws are conveyed along the forward track to the subsequent receiving mechanism; the drive assembly 24 is assembled between the selection assembly 23 and the support base 21, and as the core component of power transmission, it is responsible for smoothly transmitting the torque output by the power assembly 25 to the selection assembly 23; the power assembly 25 is fixed to one side of the support base 21, and the output end of the power assembly 25 is connected to the drive assembly 24 through a coupling, providing a stable driving force to drive the drive assembly 24 to rotate, and then the drive assembly 24 drives the selection assembly 23 to complete a precise flipping action; two sets of support frames 26 are symmetrically fastened to both sides of the screw conveying assembly 22 with bolts, which are used to provide stable installation support for the adjusting bracket 27; the adjusting bracket 27 7 is assembled between two sets of support frames 26, and the adjusting bracket 27 can slide up and down along the height direction of the support frame 26; the image acquisition component 28 is fixed to the bottom of the adjusting bracket 27 by bolts. By adjusting the installation height of the adjusting bracket 27 on the support frame 26, the image acquisition range of the image acquisition component 28 can be precisely adjusted to adapt to the detection requirements of screws of different specifications; the PLC controller 29 is fixed to one side of the support base 21 by bolts. It has a built-in dedicated defect detection algorithm program, which can realize functions such as image signal reception, analysis and processing, and control command transmission. It is used to realize the automated control of the entire device. At the same time, a communication interface is reserved to realize the docking with the production line control system, which facilitates the realization of integrated assembly line operation. like Figure 2-5As shown, the support base 21 includes a support frame 211 and a support bracket 212. The support frame 211 adopts a frame structure design and is welded from rectangular steel pipes, which not only ensures structural strength but also effectively reduces the overall weight of the base, facilitating the handling and installation of the equipment. The support bracket 212 is made of steel plate bent and welded to the inside of the support frame 211. The top of the support bracket 212 fully fits against the bottom of the screw conveying assembly 22, preventing displacement of the screw conveying assembly 22 during equipment operation and further improving the installation stability and operational reliability of the screw conveying assembly 22. The screw conveying assembly 22 includes a conveying slide 221, which is made of wear-resistant engineering plastic to prevent the screw surface from being scratched and has good wear resistance. A U-shaped groove 222 is provided in the middle of the bottom of the conveying slide 221, and a stepped groove 223 is integrally formed at the bottom of the U-shaped groove 222 to provide an installation position and rotation space for the assembly 23. Multiple fixing screw holes 224 are provided on both side walls of the conveying slide 221. The fixing screw holes 224 are evenly distributed along the length of the side wall to realize detachable connection with components such as the support frame 26, which facilitates the maintenance and replacement of the assembly.

[0022] The selection component 23 includes a V-shaped selection plate 231 adapted to the U-shaped slot 222. The V-shaped selection plate 231 is made of stainless steel and has an anti-corrosion treatment, giving it good corrosion resistance and structural strength. The bottom of the V-shaped selection plate 231 is symmetrically welded with support ear plates 232. The support ear plates 232 have mounting holes adapted to the driven shaft 243. The inner wall of the mounting holes is provided with wear-resistant bushings for precise connection with the driven shaft 243 of the drive component 24, ensuring that the selection component 23 can rotate synchronously with the driven shaft 243 and achieve stable flipping action.

[0023] The drive assembly 24 includes a driven gear 241, a drive gear 242, a driven shaft 243, and a drive shaft 244. Both the driven gear 241 and the drive gear 242 are involute gears with a 1:1 transmission ratio, enabling smooth power transmission and speed regulation, ensuring the precision of the rotational action of the drive assembly 23. The driven gear 241 is fitted onto the outer wall of the driven shaft 243 via a key; this key connection ensures reliable torque transmission and prevents slippage. The driven shaft 243 is laterally mounted on the front side of the support base 211, and it passes through two support lugs 232. The driven gear 241 and the driven shaft 243 are connected to each other via the support lugs 232. The lug plate 232 and the driven shaft 243 are circumferentially fixed by metal keys to ensure synchronous rotation of all components. The drive gear 242 is sleeved and fixed to one end of the drive shaft 244 by a flat key. The two are reliably connected by metal keys, and the drive gear 242 and the driven gear 241 mesh with each other for transmission. The other end of the drive shaft 244 is mounted on the front side of the support base 211 through a bearing seat, and the drive shaft 244 is located below the driven shaft 243, forming a reasonable transmission layout. In addition, deep groove ball wear-resistant bearings are installed at the connection points of the driven shaft 243, the drive shaft 244 and the support base 211, which can effectively reduce rotational wear and improve transmission stability and service life. The power assembly 25 includes a servo motor 251 and a reducer 252. The servo motor 251 is a high-precision servo motor with a rated power of 0.5-1.5kW and a rated speed of 3000r / min. It has advantages such as fast response speed, high positioning accuracy, and stable operation, and can achieve precise control of speed and torque to meet the real-time requirements of the rotational action of the selected assembly 23. The servo motor 251 is fixed to a special motor mounting bracket on one side of the support base 211 by bolts. The output end of the servo motor 251 is connected to the reducer 252 via a flexible coupling. The input end of 2 is connected to the transmission; the reducer 252 is a planetary gear reducer, and the reduction ratio can be adjusted to 10:1-30:1 according to actual needs, which can realize the reduction and torque increase of power and ensure that the output torque meets the requirements of the selection component 23 to rotate; the reducer 252 is fixed to the support base 211 by bolts, and its output end is connected to the end of the drive shaft 244 away from the drive gear 242 through a rigid coupling to realize the stable transmission of power and ensure that the drive component 24 can drive the selection component 23 to complete the precise and fast rotation action.

[0024] like Figure 6As shown, the adjusting bracket 27 includes an I-beam plate 271, which is made of aluminum alloy and has advantages such as high structural strength and light weight, effectively reducing the overall weight of the adjusting bracket. Limiting sleeves 272 are welded to each of the four ends of the I-beam plate 271. The inner diameter of the limiting sleeves 272 matches the outer diameter of the support frame 26. The clearance fit design ensures that the limiting sleeves 272 can slide smoothly along the support frame 26. At the same time, the inner wall of the limiting sleeves 272 is provided with guide grooves to prevent rotational deviation during the sliding of the adjusting bracket 27. The limiting sleeves 272 are fitted onto the outer wall of the support frame 26, realizing the adjustment of the bracket. 27 slides into the support frame 26; a high-strength nut 273 is welded to the outer wall of the limiting sleeve 272, the nut 273 and the limiting sleeve 272 are coaxially set, the internal thread of the nut 273 is connected to a threaded rod 274 with a knob, the surface of the knob is provided with anti-slip texture, which is convenient for manual adjustment; the threaded end of the threaded rod 274 passes through the limiting sleeve 272 and abuts against the side wall of the support frame 26; by tightening the threaded rod 274, the tightening force of the threaded rod is used to firmly position the adjusting bracket 27 at the preset height position of the support frame 26, the positioning is reliable and not easy to loosen, the adjustment process is simple and convenient, and no special tools are required.

[0025] like Figure 7 As shown, the image acquisition component 28 includes a U-shaped bracket 281, two fixing clips 282, two fixing clips 283, two rectangular frames 284, and an industrial camera 285. The U-shaped bracket 281 is made of stainless steel and has good structural strength and stability, providing installation support for the industrial camera 285. The two fixing clips 282 are symmetrically welded to both ends of the U-shaped bracket 281, and the inner diameter of the fixing clips 282 matches the outer diameter of the middle crossbeam of the I-beam plate 271. The two fixing clips 283 are respectively fixed to the top of the two fixing clips 282 by bolts. The structure of the fixing clips 283 is the same as that of the fixing clips 281, and together with the fixing clips 282, they form a complete clamp structure, which is fixed in the middle of the I-beam plate 271 to realize the U-shaped bracket 285. The U-shaped bracket 281 and the adjusting bracket 27 are detachably connected, ensuring a secure connection and facilitating the installation, disassembly, and position adjustment of the U-shaped bracket 281. Two rectangular frames 284 are fixed to the bottom of the U-shaped bracket 281 with bolts. The inner dimensions of the rectangular frames 284 are adapted to the outer dimensions of the industrial camera 285, and the side walls of the rectangular frames 284 have fastening bolt holes. The industrial camera 285 is an industrial CCD camera with a resolution of no less than 20 million pixels and a frame rate of no less than 30fps, equipped with a macro lens, which can clearly capture minute defects on the surface of screws. The two ends of the industrial camera 285 are embedded in the inner cavities of the two rectangular frames 284, and the positioning and fixation of the industrial camera 285 are achieved by tightening the fastening bolts, which also facilitates the angle adjustment, maintenance, and replacement of the industrial camera 285.

[0026] like Figure 9 As shown, based on the above-described apparatus, the present invention also provides a screw defect detection method based on image processing, the method comprising the following steps: 1. Equipment debugging: According to the specifications of the screw to be tested, adjust the height of the adjusting bracket 27 so that the acquisition range of the industrial camera 285 covers the entire surface of the screw; adjust the vibration frequency of the feeding vibratory plate 1 to ensure that the screw can be conveyed in an orderly manner along the track without jamming or deviation; import the standard image parameters and defect judgment threshold of the screw to be tested through the PLC controller 29 to complete the equipment initialization debugging. 2. Automatic feeding: Start the feeding vibratory feeder 1. The screws to be tested are oriented along the track of the feeder body under the action of vibration excitation and are conveyed to the output port of the feeding vibratory feeder 1 in sequence. Then they fall smoothly into the conveying slide 221 of the screw conveying assembly 22. 3. Image Acquisition: The screw rolls smoothly along the conveyor slide 22. When it rolls to the acquisition area of ​​the image acquisition component 28, the industrial camera 285 takes real-time pictures of the screw, acquiring all-round image data of the screw head, thread, surface and end face, and transmitting the acquired image signal to the PLC controller 29 in real time. 4. Defect Identification: The PLC controller 29 uses a built-in deep learning defect detection algorithm to compare and analyze the collected screw images with standard image parameters to determine whether the screw has defects and the type of defects (head cracking, thread defects, etc.). 5. Automatic Sorting: If the PLC controller 29 determines that the screw is qualified, the selection component 23 remains in the forward initial state, and the qualified screw continues to roll along the conveyor track and is eventually conveyed to the subsequent receiving mechanism; if the PLC controller 29 determines that the screw is defective, it immediately sends a control signal to the power component 25, the servo motor 251 starts, and after the speed reduction and torque increase through the reducer 252, it drives the drive shaft 244 to rotate, the drive gear 242 drives the driven gear 241 to rotate, and then drives the V-shaped selection plate 231 to flip in the opposite direction through the driven shaft 243, so that the defective screw is discharged from the rear discharge port, completing the sorting of defective screws; after the sorting is completed, the servo motor 251 rotates in the opposite direction, driving the selection component 23 to reset to the initial state, ready for the next sorting; 6. Cyclic Inspection: Steps 2-5 above are continuously repeated to achieve continuous and automated defect inspection and sorting of screws.

[0027] The above embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of protection of the present invention. Any non-substantial changes and substitutions made by those skilled in the art based on the present invention shall fall within the scope of protection claimed by the present invention.

Claims

1. A screw defect detection device based on image processing, comprising a feeding vibratory feeder (1) and an image detection mechanism (2); the feeding vibratory feeder (1) is used for automatic feeding and orderly conveying of screws; the image detection mechanism (2) is located directly below the conveying port of the feeding vibratory feeder (1); The image detection mechanism (2) includes a support base (21), a screw conveying assembly (22), a selection assembly (23), a drive assembly (24), a power assembly (25), two sets of support frames (26), an adjustment bracket (27), an image acquisition assembly (28), and a PLC controller (29). The support base (21) is fixedly mounted on the front side of the output port of the feeding vibratory plate (1) to provide support for the image detection mechanism (2); the screw conveying assembly (22) is installed on the top of the support base (21) and docks with the conveying port of the feeding vibratory plate (1) to receive and convey screws; The selection component (23) is rotatably mounted inside the front side of the screw conveying component (22) and is used to sort screws according to the detection results; the drive component (24) is installed between the selection component (23) and the support base (21) and is used to transmit power; the power component (25) is installed on one side of the support base (21) and its output end is connected to the drive component (24) to provide driving force for the selection component (23) to flip. Two sets of support frames (26) are symmetrically installed on both sides of the screw conveying assembly (22); the adjusting bracket (27) is installed between the two sets of support frames (26) and can be slidably adjusted along the height direction of the support frames (26); The image acquisition component (28) is installed at the bottom of the adjustment bracket (27) to acquire images of the screw from all directions; the PLC controller (29) is installed on one side of the support base (21) to receive image signals, analyze and process them and send control commands to realize automated control of the device and docking with the production line control system.

2. The screw defect detection device based on image processing as described in claim 1, characterized in that, The support base (21) includes a support frame (211) and a support bracket (212); the support frame (211) is a frame structure welded from rectangular steel pipes; the support bracket (212) is fixed inside the support frame (211), and its top is fully in contact with the bottom of the screw conveying assembly (22).

3. The screw defect detection device based on image processing as described in claim 1, characterized in that, The screw conveying assembly (22) includes a conveying slide (221); a U-shaped groove (222) is provided in the middle of the bottom of the conveying slide (221), and a stepped groove (223) is integrally formed at the bottom of the U-shaped groove (222); a plurality of evenly distributed fixing screw holes (224) are provided on the side walls of both sides of the conveying slide (221).

4. The screw defect detection device based on image processing as described in claim 1, characterized in that, The selection component (23) includes a V-shaped selection plate (231) adapted to the U-shaped slot (222); the bottom of the V-shaped selection plate (231) is symmetrically welded with a support ear plate (232), and the support ear plate (232) is provided with a mounting hole with a wear-resistant bushing.

5. The screw defect detection device based on image processing as described in claim 1, characterized in that, The drive assembly (24) includes a driven gear (241), a drive gear (242), a driven shaft (243), and a drive shaft (244). The transmission ratio between the driven gear (241) and the driving gear (242) is 1:1; the driven gear (241) is fixed to the outer wall of the driven shaft (243) by a flat key, and the driven shaft (243) passes through the support ear plate (232); The drive gear (242) is mounted on one end of the drive shaft (244) via a flat key, and the drive gear (242) meshes with the driven gear (241) for transmission; wear-resistant bearings are installed at the connection points of the driven shaft (243), the drive shaft (244) and the support base (211).

6. The screw defect detection device based on image processing as described in claim 1, characterized in that, The adjusting bracket (27) includes an I-shaped plate (271), and the four ends of the I-shaped plate (271) are fixed with limiting sleeves (272); the limiting sleeves (272) are sleeved on the outer wall of the support frame (26); a nut (273) is fixed on the outer wall of the limiting sleeve (272), and the nut (273) is internally threaded with a threaded rod (274) with an anti-slip knob, and the threaded end of the threaded rod (274) passes through the limiting sleeve (272) and abuts against the side wall of the support frame (26).

7. The screw defect detection device based on image processing as described in claim 1, characterized in that, The image acquisition component (28) includes a U-shaped bracket (281), two fixing clips (282), two fixing clips (283), two rectangular frames (284), and an industrial camera (285). The first fixing clip (282) is symmetrically fixed at both ends of the U-shaped bracket (281), and the second fixing clip (283) is fixed to the top of the first fixing clip (282) by bolts. The two together form a clamp structure and are clamped in the middle of the I-shaped plate (271). The two rectangular frames (284) are symmetrically fixed to the bottom of the U-shaped bracket (281) by bolts, and the industrial camera (285) is embedded in the rectangular frame (284).

8. A screw defect detection method based on image processing, characterized in that, Based on the detection device according to any one of claims 1-7, the method includes the following steps: Step 1: Equipment debugging, adjust the height of the adjustment bracket (27) and the vibration frequency of the feeding vibratory plate (1), import the standard image parameters and defect judgment threshold, and complete the initial debugging; Step 2: Automatic feeding, the feeding vibratory plate (1) orients and transports the screws to the screw conveying assembly (22). Step 3: Image acquisition. The industrial camera (285) takes real-time pictures of the screw that rolls into the acquisition area, acquires all-round image data and transmits it to the PLC controller (29). Step 4: Defect identification. The PLC controller (29) uses a built-in deep learning algorithm to compare and analyze image data to determine whether the screw has defects and the type of defects. Step 5: Automatic sorting. Qualified screws are conveyed to the receiving mechanism along the forward track, while defective screws are discharged by flipping through the selection component (23), and then the selection component (23) is reset. Step Six: Cyclic inspection, repeating steps two through five to achieve continuous and automated screw inspection and sorting.

Images