Defect detection device based on computer vision

Abstract

The utility model provides a visual defect detection device based on a computer, which belongs to the technical field of appearance detection, and comprises a base, a six-axis robot, a computer, a computer and a display screen, the guiding structure comprises a second sliding rail fixed to the bottom end of the fourth rotating arm, a gear ring fixed in the first sliding rail and the second sliding rail, and a moving assembly arranged on the outer side of the second sliding rail. According to the utility model, the driving motor drives the gear to rotate, so that the moving frame drives the shell to move, and the visual angle direction of the CCD camera is adjusted, thereby realizing the multi-angle adjustment function of the device, and obtaining the multi-dimensional information of an object at different visual angles through multi-angle visual detection, such as shape, size, texture and the like. The fusion of the information can provide more comprehensive object description, and is helpful to more accurately judge the quality and state of the object.

Description

Technical Field

[0001] The utility model relates to the technical field of appearance detection, and particularly relates to a computer vision-based defect detection device. Background Art

[0002] In the process of product production, in order to prevent defective products from flowing into the market, it is usually necessary to set up staff to detect the appearance of products. Since manual appearance detection is affected by the emotional state and fatigue degree of the inspectors, a computer vision-based defect detection device is set up to automatically detect and classify the appearance, size, shape and other characteristics of products during industrial production by using image processing technology and machine learning algorithms, so as to identify and eliminate defective products and ensure product quality.

[0003] After retrieval, a patent with the Chinese patent application number 202122295941.0 discloses a computer vision-based defect detection device, including a base. A motor is fixedly connected to the top of the base. A storage box is fixedly connected to one side of the base close to the motor, and a display screen is fixedly connected to the other side. An industrial lens is installed on one side of the display screen. A discharge hole is opened at the bottom of the storage box. The output end of the motor is rotationally connected to a turntable. A placement cavity is opened on the turntable. An electric push rod is fixedly connected to the bottom of the turntable. A moving plate is installed at the telescopic end of the electric push rod. An anti-slip pad is installed on the top of the moving plate. In the utility model, the article slides into the placement cavity through the discharge hole, and the turntable is driven by the motor to rotate, so that the turntable drives the article to rotate under the industrial lens. When the device detects the article, there is no need to place the article one by one for detection, which greatly reduces the detection difficulty of the article and improves the detection efficiency of the article;

[0004] The above patent still has the following deficiencies: the defect of a single detection angle. When the device detects an object with an irregular volume, the viewing angle range of the camera is limited, resulting in partial detection areas being blocked. This blockage will directly affect the accuracy and comprehensiveness of the detection results and increase the risk of false detection or missed detection.

[0005] Therefore, there is an urgent need for a computer vision-based defect detection device to solve the above problems. Summary of the Utility Model

[0006] The purpose of the utility model is to address the current problem of a single detection angle, which is prone to the risk of false detection or missed detection.

[0007] In order to achieve the above-mentioned invention purpose, the utility model provides the following technical solutions:

[0008] A computer vision-based defect detection device includes a base, and further includes,

[0009] A six-axis robot is installed at the top of the base. Among them, the six-axis robot includes a first rotating base, a first rotating arm, a second rotating arm, a third rotating arm, a second rotating base, and a fourth rotating arm;

[0010] The first rotating base is rotatably connected to the top of the base. Inside the top of the first rotating base, a first rotating arm is rotatably connected. The top of the first rotating arm is rotatably connected to a second rotating arm. One side of the second rotating arm is rotatably connected to a third rotating arm. The bottom of the third rotating arm is rotatably connected to a second rotating base. The bottom of the second rotating base is rotatably connected to a fourth rotating arm;

[0011] A guiding structure is arranged at the bottom of the fourth rotating arm. Among them, the guiding structure includes a second slide rail fixed to the bottom of the fourth rotating arm, first slide rails installed on both sides of the second slide rail, a gear ring fixed inside the first slide rail and the second slide rail, and a moving component arranged outside the second slide rail;

[0012] A casing is arranged on one side of the guiding structure. On both sides of the bottom of the casing, CCD cameras are installed, and a fill light is installed at the bottom of the casing;

[0013] A folding structure is arranged at the tops of both sides of the guiding structure for folding and storing the guiding structure.

[0014] As a preferred technical solution of the present application, the moving component includes a moving frame sleeved outside the second slide rail, moving wheels installed on both sides inside the moving frame and sliding inside the first slide rail and the second slide rail, a gear installed at the bottom inside the moving frame and meshing with the gear ring, a reducer installed at the bottom of the moving frame and with its rotating end connected to the gear, and a driving motor installed on one side of the reducer.

[0015] As a preferred technical solution of the present application, the first slide rail and the second slide rail cooperate with each other, and the moving frame forms a sliding structure with the first slide rail and the second slide rail through the moving wheels.

[0016] As a preferred technical solution of the present application, the casing is installed on one side of the moving frame through bolts, and a rotating structure is formed between the moving frame and the gear.

[0017] As a preferred technical solution of the present application, the folding structure includes a first extension frame installed at the top of the first slide rail, a connecting wheel rotatably connected inside the first extension frame, a connecting rope fixed to the outside of the connecting wheel, a second extension frame installed at the top of the second slide rail, a winding wheel installed inside one side of the second extension frame and connected to the connecting rope, a servo motor installed on one side of the second extension frame and with its rotating end connected to the winding wheel, a guiding wheel rotatably connected inside the top of the second extension frame, a rotating seat fixed between the second slide rail and the first slide rail, and a torsion spring installed inside the rotating seat.

[0018] As a preferred technical solution of the present application, the first slide rail and the second slide rail form a rotating structure through a rotating seat, and the winding wheel is connected to the connecting wheel through a connecting rope.

[0019] Compared with the prior art, the beneficial effects of the present utility model are as follows:

[0020] In the solution of the present application: the driving motor drives the gear to rotate, so that the moving frame drives the machine shell to move, adjusting the viewing angle direction of the CCD camera, thereby realizing the multi-angle adjustment function of this device. Through multi-angle visual detection, multi-dimensional information of an object under different viewing angles can be obtained, such as shape, size, texture, etc. The fusion of this information can provide a more comprehensive description of the object, which helps to more accurately judge the quality and state of the object; the servo motor drives the winding wheel to rotate, so that the connecting rope passes through the connecting wheel and pulls the first slide rail around the rotating seat to flip through the first extension frame, folding the first slide rail and the second slide rail, thereby realizing the folding and storage function of this device, facilitating the folding and storage of the guiding structure when it is idle, reducing the volume of the device, and facilitating the transportation and storage of the device. Description of the Drawings

[0021] Figure 1 It is one of the structural schematic diagrams of the present utility model;

[0022] Figure 2 It is the second of the structural schematic diagrams of the present utility model;

[0023] Figure 3 It is one of the three-dimensional structural schematic diagrams of the guiding structure provided by the present utility model;

[0024] Figure 4 Provided by the present utility model Figure 3 The enlarged partial sectional structural schematic diagram at position A in the figure;

[0025] Figure 5 It is the second of the three-dimensional structural schematic diagrams of the guiding structure provided by the present utility model.

[0026] Reference numerals in the figure: 1, base; 2, first rotating seat; 3, first rotating arm; 4, second rotating arm; 5, third rotating arm; 6, second rotating seat; 7, fourth rotating arm; 8, guiding structure; 801, first slide rail; 802, second slide rail; 803, gear ring; 804, moving frame; 805, driving motor; 806, reducer; 807, gear; 808, moving wheel; 9, folding structure; 901, first extension frame; 902, connecting wheel; 903, rotating seat; 904, torsion spring; 905, connecting rope; 906, winding wheel; 907, guiding wheel; 908, servo motor; 909, second extension frame; 10, machine shell; 11, CCD camera; 12, supplementary light. Detailed Embodiment

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present utility model clearer, the technical solutions in the embodiments of the present utility model will be clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are some, but not all, of the embodiments of the present utility model.

[0028] As Figures 1-5 shown, a computer vision-based defect detection device proposed in this embodiment includes a base 1, and further includes

[0029] a six-axis robot installed at the top of the base 1. Among them, the six-axis robot includes a first rotating base 2, a first rotating arm 3, a second rotating arm 4, a third rotating arm 5, a second rotating base 6, and a fourth rotating arm 7;

[0030] The first rotating base 2 is rotatably connected to the top of the base 1, and the inside of the top of the first rotating base 2 is rotatably connected to the first rotating arm 3. The top of the first rotating arm 3 is rotatably connected to the second rotating arm 4. One side of the second rotating arm 4 is rotatably connected to the third rotating arm 5. The bottom of the third rotating arm 5 is rotatably connected to the second rotating base 6. The bottom of the second rotating base 6 is rotatably connected to the fourth rotating arm 7;

[0031] A guiding structure 8 is provided at the bottom of the fourth rotating arm 7. Among them, the guiding structure 8 includes a second slide rail 802 fixed to the bottom of the fourth rotating arm 7, first slide rails 801 installed on both sides of the second slide rail 802, a gear ring 803 fixed inside the first slide rails 801 and the second slide rail 802, and a moving component provided outside the second slide rail 802;

[0032] A housing 10 is provided on one side of the guiding structure 8, and CCD cameras 11 are installed on both sides of the bottom of the housing 10. A supplementary light 12 is installed at the bottom of the housing 10;

[0033] A folding structure 9 is provided at the tops of both sides of the guiding structure 8 for folding and storing the guiding structure 8.

[0034] As Figure 3 、 Figure 4 and Figure 5As shown, as a preferred embodiment, on the basis of the above method, further, the moving component includes a moving frame 804 sleeved outside the second slide rail 802, moving wheels 808 installed on both sides inside the moving frame 804 and sliding inside the first slide rail 801 and the second slide rail 802, a gear 807 installed at the bottom end inside the moving frame 804 and meshing with the gear ring 803, a speed reducer 806 installed at the bottom end of the moving frame 804 and with its rotating end connected to the gear 807, and a driving motor 805 installed on one side of the speed reducer 806. The first slide rail 801 and the second slide rail 802 cooperate with each other. The moving frame 804 forms a sliding structure with the first slide rail 801 and the second slide rail 802 through the moving wheels 808. The machine shell 10 is installed on one side of the moving frame 804 by bolts. A rotating structure is formed between the moving frame 804 and the gear 807. By starting the driving motor 805, the driving motor 805 drives the gear 807 to rotate through the speed reducer 806. Immediately afterwards, the gear 807 moves on the surface of the gear ring 803 through the teeth, so that the moving frame 804 slides at the bottom of the first slide rail 801 and the second slide rail 802, thereby driving the machine shell 10 to move by the moving frame 804 and adjusting the viewing direction of the CCD camera 11. The resistance of the movement of the moving frame 804 is reduced by multiple groups of moving wheels 808, making the movement of the moving frame 804 smooth.

[0035] As Figure 5As shown, as a preferred embodiment, on the basis of the above method, further, the folding structure 9 includes a first extension frame 901 installed at the top of the first slide rail 801, a connecting wheel 902 rotatably connected inside the first extension frame 901, a connecting rope 905 fixed to the outside of the connecting wheel 902, a second extension frame 909 installed at the top of the second slide rail 802, a winding wheel 906 installed inside one side of the second extension frame 909 and connected to the connecting rope 905, a servo motor 908 installed on one side of the second extension frame 909 and with its rotating end connected to the winding wheel 906, a guide wheel 907 rotatably connected inside the top of the second extension frame 909, a rotating seat 903 fixed between the second slide rail 802 and the first slide rail 801, and a torsion spring 904 installed inside the rotating seat 903. The first slide rail 801 and the second slide rail 802 form a rotating structure through the rotating seat 903. The winding wheel 906 is connected to the connecting wheel 902 through the connecting rope 905. The first slide rail 801 and the second slide rail 802 are connected through the rotating seat 903. When the servo motor 908 is started to rotate, it drives the winding wheel 906 to rotate. Immediately, the winding wheel 906 winds the connecting rope 905, so that the connecting rope 905 passes through the connecting wheel 902 and pulls the first slide rail 801 to flip around the rotating seat 903, folding the first slide rail 801 and the second slide rail 802, reducing the space occupied by the first slide rail 801 and the second slide rail 802. At the same time, the rotating rotating seat 903 drives the torsion spring 904 to deform. The movement of the connecting rope 905 is guided by the guide wheel 907. By reversely rotating the servo motor 908 to release the connecting rope 905, at the same time, the torsion spring 904 that is twisted and deformed pushes the rotating seat 903 to reset, so that the first slide rail 801 and the second slide rail 802 are unfolded and aligned.

[0036] Specifically, when this computer vision defect detection device is in use: by installing the base 1 on one side of the product conveyor belt and connecting the machine shell 10 to the computer in the computer room, the computer can detect the appearance image of the product through the CCD camera 11. Start the six-axis robot to drive the machine shell 10 to move synchronously with the product, so that there is no need to stop the conveyor belt during the product detection process, improving the production efficiency of the product. Drive the machine shell 10 to rotate through the guiding structure 8. At the same time, the second rotating seat 6 rotates horizontally at the bottom of the third rotating arm 5, so that the machine shell 10 can detect the appearance of the product from multiple angles, improving the detection accuracy, and at the same time facilitating the collection of a large amount of data for the computer image algorithm to learn and supplement. The guiding structure 8 can be folded and stored through the folding structure 9, so as to facilitate the transportation and storage of the device.

[0037] By starting the drive motor 805, the drive motor 805 drives the gear 807 to rotate through the speed reducer 806. Immediately afterwards, the gear 807 moves on the surface of the gear ring 803 through its teeth, causing the moving frame 804 to slide at the bottom of the first slide rail 801 and the second slide rail 802. Thus, the moving frame 804 drives the machine housing 10 to move, adjusting the viewing direction of the CCD camera 11. The resistance of the movement of the moving frame 804 is reduced by multiple sets of moving wheels 808, making the movement of the moving frame 804 smooth;

[0038] The first slide rail 801 and the second slide rail 802 are connected by a rotating seat 903. The servo motor 908 is started to rotate, driving the winding wheel 906 to rotate. Immediately afterwards, the winding wheel 906 winds up the connecting rope 905, causing the connecting rope 905 to pull the first slide rail 801 around the rotating seat 903 through the connecting wheel 902 and the first extension frame 901, folding the first slide rail 801 and the second slide rail 802, reducing the space occupied by the first slide rail 801 and the second slide rail 802. At the same time, the rotating rotating seat 903 drives the torsion spring 904 to deform. The movement of the connecting rope 905 is guided by the guide wheel 907. The connecting rope 905 is loosened by reversely rotating the servo motor 908. At the same time, the torsion spring 904 with distorted deformation pushes the rotating seat 903 to reset, causing the first slide rail 801 and the second slide rail 802 to unfold and align.

[0039] The above embodiments are only used to illustrate the present invention rather than to limit the technical solutions described in the present invention. Although this specification has described the present invention in detail with reference to the above respective embodiments, the present invention is not limited to the above specific embodiments. Therefore, any modification or equivalent replacement of the present invention; and all technical solutions and their improvements that do not depart from the spirit and scope of the invention are covered by the scope of the claims of the present invention.

Claims

1. A computer vision-based defect detection device, comprising a base (1), characterized in that: Also includes, A six-axis robot is mounted on the top of the base (1), wherein the six-axis robot comprises a first rotating base (2), a first rotating arm (3), a second rotating arm (4), a third rotating arm (5), a second rotating base (6) and a fourth rotating arm (7); A first rotating seat (2) is rotatably connected to the top of the base (1), and the top of the first rotating seat (2) is internally rotatably connected to a first rotating arm (3), the top of the first rotating arm (3) is rotatably connected to a second rotating arm (4), one side of the second rotating arm (4) is rotatably connected to a third rotating arm (5), the bottom end of the third rotating arm (5) is rotatably connected to a second rotating seat (6), and the bottom end of the second rotating seat (6) is rotatably connected to a fourth rotating arm (7); A guide structure (8) is arranged at the bottom end of the fourth rotating arm (7), wherein the guide structure (8) comprises a second slide rail (802) fixed at the bottom end of the fourth rotating arm (7), a first slide rail (801) installed on both sides of the second slide rail (802), a gear ring (803) fixed inside the first slide rail (801) and the second slide rail (802), and a moving component arranged outside the second slide rail (802); A housing (10) is arranged on one side of the guide structure (8), and CCD cameras (11) are installed on both sides of the bottom of the housing (10), and a fill light (12) is installed at the bottom of the housing (10); The folding structure (9) is arranged at the top ends of both sides of the guide structure (8) and is used to fold and store the guide structure (8).

2. The computer vision-based defect detection device according to claim 1, characterized in that: The moving assembly comprises a moving frame (804) sleeved on the outside of the second slide rail (802), moving wheels (808) installed on both sides of the moving frame (804) and sliding inside the first slide rail (801) and the second slide rail (802), a gear (807) installed at the bottom end of the moving frame (804) and meshing with the gear ring (803), a reducer (806) installed at the bottom end of the moving frame (804) and the rotating end of which is connected to the gear (807), and a driving motor (805) installed on one side of the reducer (806).

3. The computer vision-based defect detection device according to claim 2, characterized in that: The first slide rail (801) and the second slide rail (802) cooperate with each other, and the movable frame (804) forms a sliding structure with the first slide rail (801) and the second slide rail (802) through the movable wheel (808).

4. The computer vision-based defect detection device according to claim 2, characterized in that: The housing (10) is mounted on one side of a moving frame (804) by means of bolts, and a rotating structure is formed between the moving frame (804) and the gear (807).

5. The computer vision-based defect detection device according to claim 1, characterized in that: The folding structure (9) comprises a first extension frame (901) mounted on the top of the first slide rail (801), a connecting wheel (902) rotatably connected to the inside of the first extension frame (901), a connecting rope (905) fixed to the outside of the connecting wheel (902), a second extension frame (909) mounted on the top of the second slide rail (802), a reeling wheel (906) mounted on one side of the second extension frame (909) and connected to the connecting rope (905), a servo motor (908) mounted on one side of the second extension frame (909) and having a rotating end connected to the reeling wheel (906), a guide wheel (907) rotatably connected to the inside of the top of the second extension frame (909), a rotating seat (903) fixed between the second slide rail (802) and the first slide rail (801), and a torsion spring (904) mounted on the inside of the rotating seat (903).

6. The computer vision-based defect detection device according to claim 5, characterized in that: The first slide rail (801) and the second slide rail (802) form a rotating structure via a rotating seat (903), and the winding wheel (906) is connected to the connecting wheel (902) via a connecting rope (905).

Images