Flexible clamping device based on visual recognition and identification method thereof

Abstract

The application discloses a flexible clamping device based on visual identification and an identification method thereof, which comprises two mirror-imaged support plates, driving mechanisms are installed on the sides of the two support plates away from each other, mounting grooves are formed on the sides of the two support plates close to each other, and a plurality of flexible clamping mechanisms are installed in a circumferential array with the centers of the mounting grooves as base points. The visual identification system can efficiently and accurately correct the centers of the flexible clamping mechanisms in a circumferential array to coincide with the center of the circular pipe, so that the flexible clamping mechanisms uniformly bear the force of the circular pipe, the integrity of the shape of the circular pipe is ensured, the inner clamping assembly and the outer clamping assembly are matched, the inner and outer walls of the circular pipe can be simultaneously clamped, the conveying device of the circular pipe is loosened, the position of the circular pipe is prevented from being deviated due to static friction, and the center of the circular pipe can always coincide with the centers of the flexible clamping mechanisms in a circumferential array.

Description

Technical Field

[0001] This invention belongs to the field of pipe clamping technology, and particularly relates to a flexible clamping device and its recognition method based on visual recognition. Background Technology

[0002] High-power-density laser beams are used to irradiate the material being cut, quickly heating it to its vaporization temperature and causing it to evaporate and form holes. As the beam moves across the material, the holes continuously form narrow slits, completing the cutting of the material. In the processing of round tubes, lasers are used to cut the round tubes, and clamping devices are needed to limit and fix the round tubes during the cutting process.

[0003] Existing clamping devices directly clamp and fix circular pipes. Since the center of the circular pipe cannot be guaranteed to coincide with the center of the clamping device, the part of the circular pipe that first contacts the clamping device will deform due to uneven force. Summary of the Invention

[0004] To address the problems in the prior art, the present invention proposes the following technical solution:

[0005] The flexible clamping device based on vision recognition includes two mirror-mounted support plates. A drive mechanism is installed on the side of the two support plates that are far apart from each other, and a mounting groove is opened on the side that is close to each other. Multiple flexible clamping mechanisms are installed in a circumferential array with the center of the mounting groove as the base point. The drive mechanism is connected to the flexible clamping mechanism and is used to drive the multiple flexible clamping mechanisms to open and close simultaneously. A vision recognition system is provided on the side of the support plate that is close to the flexible clamping mechanism.

[0006] The visual recognition system includes an image acquisition module, an image processing module, a calculation module, and a feedback module. The image acquisition module includes a visual recognition camera, which is detachably mounted on the center of the mounting groove to capture images of the circular cross-section of the circular pipe and the opposing mounting groove. The image processing module uses the Hough transform algorithm for circular detection. The calculation module calculates the coordinates of the detected center and performs concentricity judgment. The feedback module adjusts the position of the circular pipe according to the calculation results to keep it aligned with the center of the mounting groove.

[0007] The driving mechanism includes a connecting component and a driving component. The connecting component includes a wire sleeve, and a plurality of hinge seats are fixedly connected to the outer wall of the wire sleeve. Each hinge seat is connected to a connecting rod. The driving component is detachably mounted on a support plate and is connected to the wire sleeve for driving the wire sleeve to move linearly.

[0008] The flexible clamping mechanism includes an inner clamping component, which includes a strip block that penetrates through a support plate. Both ends of the strip block are fixedly connected to a fixed seat. A hinge seat two is fixedly connected to one of the fixed seats on the side away from the strip block, and a clamping plate one is fixedly connected to the other fixed seat. The hinge seat two is connected to a connecting rod.

[0009] The flexible clamping mechanism further includes an external clamping assembly, which includes an L-shaped plate. The L-shaped plate is fixed on a fixed seat near the clamping plate one, and a lead screw two is threadedly connected to the side of the L-shaped plate away from the clamping plate one. A knob is fixedly connected to one end of the lead screw two, and the clamping plate two is rotatably connected to the other end. A guide slide rod is fixedly connected to the clamping plate two, and the guide slide rod passes through the L-shaped plate.

[0010] As a preferred embodiment of the above technical solution, the driving component includes:

[0011] The motor is mounted on the support plate by bolts, and the output end of the motor is fixedly connected to a lead screw, which is threaded into the motor.

[0012] As a preferred embodiment of the above technical solution, the support plate has multiple sliding openings arranged in a circular array with the center of the mounting groove as the base point. Each sliding opening has a sliding groove on both sides of the side wall of the support plate. The strip block passes through the sliding opening, and a slider is fixedly connected to the side of the two fixed seats that are close to each other. The slider is slidably connected to the sliding groove.

[0013] As a preferred embodiment of the above technical solution, flexible pad one and flexible pad two are fixedly connected to the sides of clamping plate one and clamping plate two that are close to each other.

[0014] A vision-based recognition method for identifying flexible clamping devices includes the following steps:

[0015] S1. Image Acquisition:

[0016] Images of the circular cross-section and mounting groove of the circular pipe to be clamped are captured by a visual recognition camera to ensure that the image quality is high enough for subsequent processing;

[0017] S2, Circular detection:

[0018] The image is transmitted to the image processing module, where the Hough transform algorithm is used to detect circular contours in the image.

[0019] S3. Calculation of the center coordinates:

[0020] The center coordinates and radius of each detected circle are calculated. The Hough transform provides the center coordinates and radius parameters of the circle. These parameters are extracted for further processing.

[0021] S4. Concentricity judgment:

[0022] Compare the coordinates of the circular cross-section of the circular pipe fitting to be clamped and the center coordinates of the mounting groove, calculate the distance between them, and if the distance between the two centers is less than the preset tolerance range, the two circles are considered to be concentric.

[0023] S5, Real-time Adjustment:

[0024] The calculation results are passed to the feedback module. If the two centers are not coincident, the position of the circular tube to be clamped is automatically adjusted until the centers are aligned, and then it is moved to the clamping area of ​​the flexible clamping mechanism.

[0025] As a preferred embodiment of the above technical solution, the calculation of the center coordinates includes the following steps:

[0026] S31. Circular detection:

[0027] The parameter space is a three-dimensional space where each point represents the center coordinates (x, y) and radius r of a circle;

[0028] S32, Edge Detection:

[0029] Edge detection algorithms are used to identify edge points in an image, which are potential candidates for circular edges;

[0030] S33, Parameter Space Voting:

[0031] For each edge point, assuming it is part of a circle, calculate the possible center positions of the point under different radius values, and accumulate these assumed center positions in the parameter space;

[0032] S34. Finding the peak value:

[0033] In parameter space, the combination of center coordinates and radius forms local peaks. By detecting the positions of these peaks, the most likely center coordinates (x, y) and radius r can be determined.

[0034] S35. Extraction of center and radius:

[0035] Based on the peak values ​​in the parameter space, the corresponding center coordinates and radius are extracted to determine the specific parameters of the circle.

[0036] The beneficial effects of this invention are as follows:

[0037] 1. The present invention can efficiently and accurately align the center of a circular tube with the center of a flexible clamping mechanism arranged in a circular array through a visual recognition system, so that the clamping force of the flexible clamping mechanism on the circular tube is uniform, thus ensuring the integrity of the shape of the circular tube.

[0038] 2. The present invention enables the inner and outer walls of the circular tube to be clamped simultaneously by the cooperation of the inner clamping component and the outer clamping component. When the conveying device of the circular tube is released, it avoids the displacement of the circular tube due to static friction, so that the circular tube can always keep its center coincident with the flexible clamping mechanism arranged in multiple circumferential arrays. Attached Figure Description

[0039] Figure 1 The diagram shown is a structural schematic of the flexible clamping device based on vision recognition in the embodiment.

[0040] Figure 2 The diagram shown is a structural schematic of the flexible clamping mechanism and the visual recognition system in the embodiment;

[0041] Figure 3 The diagram shown is a structural schematic of the flexible clamping mechanism in the embodiment;

[0042] Figure 4 The diagram shown is a structural schematic of the support plate in the embodiment;

[0043] Figure 5 The diagram shown is a structural schematic of the connection component and the driving component in the embodiment;

[0044] Figure 6 The diagram shown is a structural schematic of the inner clamping component and the outer clamping component in the embodiment;

[0045] Figure 7 The diagram shown is a structural block diagram of the visual recognition system in the embodiment.

[0046] Explanation of reference numerals in the attached figures:

[0047] 100. Support plate; 101. Mounting groove; 102. Slide opening; 103. Slide groove; 200. Drive mechanism; 210. Connecting assembly; 211. Screw sleeve; 212. Hinge seat one; 213. Connecting rod; 220. Drive assembly; 221. Motor; 222. Screw one; 300. Flexible clamping mechanism; 310. Inner clamping assembly; 311. Strip block; 312. Fixed seat; 313. Slider; 314. Hinge 315. Connector 2; 316. Clamping plate 1; 327. Flexible pad 1; 328. External clamping assembly; 329. L-shaped plate; 320. Lead screw 2; 321. Knob; 322. Clamping plate 2; 323. Flexible pad 2; 324. Guide slide rod; 400. Vision recognition system; 410. Image acquisition module; 411. Vision recognition camera; 420. Image processing module; 430. Calculation module; 440. Feedback module. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Example

[0049] A flexible clamping device and its recognition method based on vision recognition include two mirror-mounted support plates 100. A drive mechanism 200 is mounted on the side of each support plate 100 that is far apart from each other, and a mounting groove 101 is formed on the side that is close to each other. Multiple flexible clamping mechanisms 300 are arranged in a circular array around the center of the mounting groove 101. The drive mechanism 200 is connected to the flexible clamping mechanism 300 and is used to drive the multiple flexible clamping mechanisms 300 to open and close simultaneously. A vision recognition system 400 is provided on the side of each support plate 100 closest to the flexible clamping mechanism 300. The vision recognition system 400 includes the following components: (Figure 100) The system includes a data acquisition module 410, an image processing module 420, a calculation module 430, and a feedback module 440. The image acquisition module 410 includes a visual recognition camera 411, which is detachably mounted on the center of the mounting groove 101 to capture the circular cross-section of the circular pipe and the opposing mounting groove 101. The image processing module 420 uses the Hough transform algorithm to perform circular detection. The calculation module 430 calculates the coordinates of the detected center and performs concentricity judgment. The feedback module 440 adjusts the position of the circular pipe according to the calculation results to keep it aligned with the center of the mounting groove 101.

[0050] Specifically, the visual recognition system 400 can efficiently and accurately align the circular tube with the center of the multiple circumferentially arrayed flexible clamping mechanisms 300, ensuring uniform clamping force and maintaining the integrity of the tube's shape. The flexible clamping mechanisms 300 allow simultaneous clamping of the inner and outer walls of the circular tube. When the conveying device releases the tube, static friction prevents positional shifts, ensuring the tube always remains aligned with the center of the multiple circumferentially arrayed flexible clamping mechanisms 300.

[0051] like Figure 1 and Figure 5 As shown, the drive mechanism 200 includes a connecting component 210 and a drive component 220. The connecting component 210 includes a threaded sleeve 211. A plurality of hinge seats 212 are fixedly connected to the outer wall of the threaded sleeve 211. The hinge seats 212 are connected to a connecting rod 213. The drive component 220 is detachably mounted on the support plate 100 and is connected to the threaded sleeve 211 for driving the threaded sleeve 211 to move linearly.

[0052] It should be noted that the drive assembly 220 can drive the threaded sleeve 211 to move linearly, causing it to move closer to or further away from the support plate 100. At the same time, the connecting rod 213 can rotate with the hinge seat 212 as the base point. Multiple hinge seats 212 and connecting rods 213 are arranged in a circumferential array with the center of the threaded sleeve 211 as the base point. The center of the threaded sleeve 211 coincides with the center of the mounting groove 101.

[0053] like Figure 5 As shown, the drive assembly 220 includes a motor 221, which is mounted on the support plate 100 by bolts, and a lead screw 222 is fixedly connected to the output end of the motor 221, which is threadedly engaged with the motor 221.

[0054] It should be noted that the drive assembly 220 is used to drive the threaded sleeve 211 to move linearly, and can be, but is not limited to, using a cylinder.

[0055] Specifically, during the process of the motor 221 driving the lead screw 222 to rotate, since the connecting rod 213 is connected to the flexible clamping mechanism 300, the threaded sleeve 211 will move on the outer ring of the lead screw 222 through the threaded engagement between the threaded sleeve 211 and the lead screw 222. Compared with the lead screw 222, the cylinder occupies less space and has a simpler structure. The two methods can be freely selected.

[0056] Figure 1 , Figure 2 , Figure 3 and Figure 6 As shown, the flexible clamping mechanism 300 includes an inner clamping component 310, which includes a strip block 311 that passes through the support plate 100. Both ends of the strip block 311 are fixedly connected to a fixing seat 312. A hinge seat 2 314 is fixedly connected to the side of one fixing seat 312 away from the strip block 311. A clamping plate 315 is fixedly connected to the other fixing seat 312. The hinge seat 2 314 is connected to the connecting rod 213.

[0057] It should be noted that the combination of the strip block 311 and the two fixing seats 312 forms an "I" shape, and the two fixing seats 312 are respectively attached to the two sides of the support plate 100.

[0058] like Figure 3 , Figure 4 and Figure 6 As shown, the support plate 100 has multiple sliding openings 102 arranged in a circular array with the center of the mounting groove 101 as the base point. Each sliding opening 102 has a sliding groove 103 on both sides of the side wall of the support plate 100. The strip block 311 passes through the sliding opening 102. The two fixed seats 312 are fixedly connected to the side of each other with a slider 313. The slider 313 is slidably connected to the sliding groove 103.

[0059] Specifically, the number of sliding openings 102 is the same as the number of strip blocks 311, preferably an even number. Through the sliding cooperation between the slider 313 and the sliding groove 103, the stability of the strip blocks 311 sliding inside the sliding opening 102 can be increased. When the thread sleeve 211 moves towards the support plate 100, multiple strip blocks 311 will expand away from the center of the mounting groove 101. When the thread sleeve 211 moves away from the support plate 100, multiple strip blocks 311 will contract towards the center of the mounting groove 101. Before the strip blocks 311 expand, the visual recognition system 400 corrects and aligns the circular tube with the center of the mounting groove 101, and places the circular tube in the clamping area against the support plate 100. Afterward, the multiple strip blocks 311 expand simultaneously to fix and clamp the circular tube.

[0060] like Figure 6 As shown, the flexible clamping mechanism 300 also includes an external clamping assembly 320. The external clamping assembly 320 includes an L-shaped plate 321, which is fixed on a fixed seat 312 near the clamping plate 315. A lead screw 322 is threadedly connected to the side of the L-shaped plate 321 away from the clamping plate 315. A knob 323 is fixedly connected to one end of the lead screw 322, and a clamping plate 324 is rotatably connected to the other end. A guide slide rod 326 is fixedly connected to the clamping plate 324, and the guide slide rod 326 passes through the L-shaped plate 321.

[0061] It should be noted that flexible pad 316 and flexible pad 325 are fixedly connected to the sides of clamping plate 315 and clamping plate 324 respectively. The flexibility of flexible pad 316 and flexible pad 325 can prevent clamping plate 315 and clamping plate 324 from directly contacting the circular pipe and causing damage.

[0062] Specifically, after the inner clamping assembly 310 fixes the inner wall of the circular tube, rotating the knob 323 drives the lead screw 322 to rotate together. Through the threaded engagement with the L-shaped plate 321, and the engagement of the guide slide rod 326 fixed on the clamping plate 324 with the L-shaped plate 321, the clamping plate 324 and the flexible pad 325 can move towards the outer wall of the circular tube, thus clamping and fixing the inner and outer walls of the circular tube simultaneously. When the conveying device of the circular tube is released, it can prevent the position of the circular tube from shifting due to static friction, so that the circular tube can always keep its center aligned with the flexible clamping mechanism 300 arranged in multiple circumferential arrays. Example

[0063] like Figure 2 and Figure 7 As shown, the identification method for the flexible clamping device based on vision recognition, compared to Embodiment 1, further includes the following steps:

[0064] S1. Image Acquisition:

[0065] The visual recognition camera 411 captures images of the circular cross-section of the circular pipe to be clamped and the mounting groove 101 to ensure that the image quality is high enough for subsequent processing;

[0066] S2, Circular detection:

[0067] The image is transmitted to the image processing module 420, where the Hough transform algorithm is used to detect circular contours in the image.

[0068] S3. Calculation of the center coordinates:

[0069] The center coordinates and radius of each detected circle are calculated. The Hough transform provides the center coordinates and radius parameters of the circle. These parameters are extracted for further processing.

[0070] S4. Concentricity judgment:

[0071] Compare the coordinates of the circular cross-section of the circular pipe to be clamped and the center of the mounting groove 101, and calculate the distance between them. If the distance between the two centers is less than the preset tolerance range, the two circles are considered to be concentric.

[0072] S5, Real-time Adjustment:

[0073] The calculation results are passed to the feedback module 440. If the two centers are not coincident, the position of the circular tube to be clamped is automatically adjusted until the centers are aligned, and then it is moved to the clamping area of ​​the flexible clamping mechanism 300.

[0074] The calculation of the center coordinates includes the following steps:

[0075] S31. Circular detection:

[0076] The parameter space is a three-dimensional space where each point represents the center coordinates (x, y) and radius r of a circle;

[0077] S32, Edge Detection:

[0078] Edge detection algorithms are used to identify edge points in an image, which are potential candidates for circular edges;

[0079] S33, Parameter Space Voting:

[0080] For each edge point, assuming it is part of a circle, calculate the possible center positions of the point under different radius values, and accumulate these assumed center positions in the parameter space;

[0081] S34. Finding the peak value:

[0082] In parameter space, the combination of center coordinates and radius forms local peaks. By detecting the positions of these peaks, the most likely center coordinates (x, y) and radius r can be determined.

[0083] S35. Extraction of center and radius:

[0084] Based on the peak values ​​in the parameter space, the corresponding center coordinates and radius are extracted to determine the specific parameters of the circle.

[0085] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it.

Claims

1. A flexible gripping device based on visual recognition, characterized in that, The system includes two mirror-mounted support plates (100). A drive mechanism (200) is installed on the side of each support plate (100) that is far apart from each other, and a mounting groove (101) is opened on the side that is close to each other. Multiple flexible clamping mechanisms (300) are installed in a circular array with the center of the mounting groove (101) as the base point. The drive mechanism (200) is connected to the flexible clamping mechanism (300) and is used to drive the multiple flexible clamping mechanisms (300) to open and close simultaneously. A visual recognition system (400) is provided on the side of the support plate (100) that is close to the flexible clamping mechanism (300). The visual recognition system (400) includes an image acquisition module (410), an image processing module (420), a calculation module (430), and a feedback module (440). The image acquisition module (410) includes a visual recognition camera (411), which is detachably mounted on the center of the mounting slot (101) to capture the circular cross-section of the circular pipe and the opposing mounting slot (101). The image processing module (420) uses the Hough transform algorithm to perform circular detection. The calculation module (430) is used to calculate the coordinates of the detected center and perform concentricity judgment. The feedback module (440) adjusts the position of the circular pipe according to the calculation result so that it is aligned with the center of the mounting slot (101). The drive mechanism (200) includes a connecting component (210) and a drive component (220). The connecting component (210) includes a threaded sleeve (211). The outer wall of the threaded sleeve (211) is fixedly connected to a plurality of hinge seats (212). The hinge seats (212) are connected to a connecting rod (213). The drive component (220) is detachably mounted on the support plate (100) and is connected to the threaded sleeve (211) for driving the threaded sleeve (211) to move linearly. The flexible clamping mechanism (300) includes an inner clamping component (310), which includes a strip block (311) that penetrates the support plate (100). Both ends of the strip block (311) are fixedly connected to a fixing seat (312). One of the fixing seats (312) is fixedly connected to a hinge seat two (314) on the side away from the strip block (311). The other fixing seat (312) is fixedly connected to a clamping plate one (315). The hinge seat two (314) is connected to a connecting rod (213). The flexible clamping mechanism (300) further includes an external clamping assembly (320), which includes an L-shaped plate (321). The L-shaped plate (321) is fixed on a fixed seat (312) near the clamping plate (315), and a screw rod (322) is threadedly connected to the side of the L-shaped plate (321) away from the clamping plate (315). A knob (323) is fixedly connected to one end of the screw rod (322), and a clamping plate (324) is rotatably connected to the other end. A guide slide rod (326) is fixedly connected to the clamping plate (324), and the guide slide rod (326) passes through the L-shaped plate (321).

2. The vision recognition based flexible gripping device according to claim 1, wherein, The drive component (220) includes: The motor (221) is mounted on the support plate (100) by bolts, and the output end of the motor (221) is fixedly connected to a lead screw (222), which is threadedly engaged with the motor (221).

3. The vision recognition based flexible gripping device of claim 1, wherein, The support plate (100) has multiple sliding openings (102) arranged in a circular array with the center of the mounting groove (101) as the base point. Each sliding opening (102) has a sliding groove (103) on both sides of the side wall of the support plate (100). The strip block (311) passes through the sliding opening (102). The two fixed seats (312) are fixedly connected to a slider (313) on the side that is close to each other. The slider (313) is slidably connected to the sliding groove (103).

4. The vision recognition based flexible gripping device of claim 1, wherein, Flexible pad 1 (316) and flexible pad 2 (325) are respectively fixedly connected to the side of clamping plate 1 (315) and clamping plate 2 (324) that are close to each other.

5. The method according to any one of claims 1 to 4, wherein Includes the following steps, S1. Image Acquisition: The visual recognition camera (411) captures images of the circular cross-section of the circular pipe to be clamped and the mounting groove (101) to ensure that the image quality is high enough for subsequent processing; S2, Circular detection: The image is transmitted to the image processing module (420), where the Hough transform algorithm is used to detect circular contours in the image; S3. Calculation of the center coordinates: The center coordinates and radius of each detected circle are calculated. The Hough transform provides the center coordinates and radius parameters of the circle. These parameters are extracted for further processing. S4. Concentricity judgment: Compare the center coordinates of the circular cross section of the circular pipe to be clamped and the center coordinates of the mounting groove (101), calculate the distance between them, and if the distance between the two centers is less than the preset tolerance range, the two circles are considered to be concentric. S5, Real-time Adjustment: The calculation results are passed to the feedback module (440). If the two centers are not coincident, the position of the circular tube to be clamped is automatically adjusted until the centers are aligned, and then it is moved to the clamping area of ​​the flexible clamping mechanism (300).

6. The identification method of the flexible gripping device based on visual recognition according to claim 5, characterized in that, The calculation of the center coordinates includes the following steps: S31. Circular detection: The parameter space is a three-dimensional space where each point represents the center coordinates (x, y) and radius r of a circle; S32, Edge Detection: Edge detection algorithms are used to identify edge points in an image, which are potential candidates for circular edges; S33, Parameter Space Voting: For each edge point, assuming it is part of a circle, calculate the possible center positions of the point under different radius values, and accumulate these assumed center positions in the parameter space; S34. Finding the peak value: In parameter space, the combination of center coordinates and radius forms local peaks. By detecting the positions of these peaks, the most likely center coordinates (x, y) and radius r can be determined. S35. Extraction of center and radius: Based on the peak values ​​in the parameter space, the corresponding center coordinates and radius are extracted to determine the specific parameters of the circle.

Images