Vehicle frame surface defect identification system based on machine vision

By adopting a machine vision system in the frame quality inspection of electric bicycles, using image acquisition modules and fill lights in the light-proof room to detect frame defects, the problem of low manual detection accuracy is solved, and efficient and accurate defect identification is achieved. Marking, suitable for assembly line production.

CN118980693APending Publication Date: 2024-11-19WUXI YINGTELI INTELLIGENT TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411088976.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-09
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

The prior art has poor manual observation accuracy and low working efficiency in the quality inspection of electric bicycle frames, which affects product quality.

Method used

Using a frame surface defect recognition system based on machine vision, the frame hoisting and transporting is carried out by installing interlaced image acquisition modules and fill lights in the light-proof room, combining the driving module to carry out frame hoisting and transportation, using the image comparison module to detect defects, and through the grid module Label defect points to improve detection accuracy and efficiency.

Benefits of technology

It realizes the efficiency and accuracy of frame defect detection, improves detection efficiency and accuracy, reduces manual errors, and is suitable for assembly line production needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118980693A_ABST
    Figure CN118980693A_ABST
Patent Text Reader

Abstract

A frame surface defect identification system based on machine vision disclosed by the present invention comprises a travelling crane module, a light shielding module, a detection module and a light supplementing module, the travelling crane module comprises a travelling crane rail, the travelling crane rail is slidably provided with a lifting hook for lifting a frame, the light shielding module comprises a light shielding chamber, and the two ends of the light shielding module are both provided with light shielding curtains. The detection module comprises a control module and at least two groups of image acquisition modules mounted on the inner side of the light shielding chamber, an image comparison module is arranged in the control module, the light supplementing module comprises a light supplementing lamp arranged in the light shielding chamber, and the light supplementing module is in signal connection with the control module; and a background plate is arranged on one side opposite to the image acquisition module. Defects of the vehicle frame are detected through the image acquisition module and the image comparison module, and the vehicle frame is transported through the traveling module for continuous detection, so that the detection efficiency and the detection precision are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of vehicle frame detection, and relates to a vehicle frame surface defect recognition system based on machine vision. Background Art

[0002] Electric bicycles are a common means of transportation. The frame of some electric bicycles is an irregular frame structure made of steel pipes or aluminum materials. During production, the frame needs to go through machining, painting, spray painting, labeling and other processes. During the production process, especially during painting and other processes, some frames may have defects, affecting product quality.

[0003] When performing quality inspection on the vehicle frame, manual observation is required, which has poor accuracy and low work efficiency, and is not conducive to the assembly line production of the vehicle frame. Summary of the invention

[0004] In view of the above problems, the present invention proposes a frame surface defect recognition system based on machine vision, which well solves the problems in the prior art.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A vehicle frame surface defect recognition system based on machine vision, comprising:

[0007] A crane module, the crane module comprising a crane track arranged on the top of the workshop, a plurality of lifting assemblies being slidably mounted on the crane track, and the lifting assemblies being provided with at least two hooks for lifting the frame;

[0008] A light-proof module, the light-proof module comprising a light-proof chamber, a through slot for the hook to pass through is provided at the upper end of the light-proof chamber, and light-shielding curtains are provided at both ends of the light-proof module;

[0009] A detection module, the detection module comprising a control module and an image acquisition module installed inside the light-proof chamber, the control module being provided with an image comparison module;

[0010] A fill light module, the fill light module comprising a fill light disposed in the light-proof chamber, the fill light module being signal-connected to the control module;

[0011] Among them, at least two groups of image acquisition modules are provided, and the two groups of image acquisition modules are staggeredly arranged on both sides of the light-proof room. A background board is provided on the side facing the image acquisition module, and the fill light module and the driving module are both connected to the control module signal.

[0012] Optionally, a marking module is further included, the marking module includes a planar driving component, an output end of the planar driving component is fixedly connected to a marking spray gun, the marking module is arranged at the exit end of the light-proof chamber, and two marking modules are arranged in mirror symmetry.

[0013] Optionally, a grid module is provided in the control module, and the grid module divides the image into grid coordinates according to the image acquired by the image acquisition module, and the marking module is connected to the control module by signal.

[0014] Optionally, the planar driving member includes a Z-axis frame slidably mounted on the light-proof chamber, the Z-axis frame is connected to the driving member, the Z-axis frame is slidably mounted with a Y-axis frame, the Y-axis frame is also connected to the driving member, and the marking spray gun is mounted on the Y-axis frame.

[0015] Optionally, a positioning module is further included, the positioning module includes at least two positioning lasers, the positioning lasers are installed on the background plate, and the positioning module is connected to the control module by signal.

[0016] Optionally, an auxiliary transmission module is installed at the through slot position, and the auxiliary transmission module includes two transmission components arranged on both sides of the through slot, the hook is cylindrical, and the conveyor belt surface of the transmission component is provided with a semi-arc ring, and the semi-arc rings of the two transmission components are mirror-symmetrically fitted on the sides, and the auxiliary transmission module is connected to the control module signal.

[0017] Optionally, sponge strips are bonded to both side walls of the through groove, and the sides of the two sponge strips are fitted together.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] 1. A relatively stable light detection environment is set up through a light-proof room and a blackout curtain. Two sets of image acquisition modules and fill-in lights are installed in the light-proof room, and the two sets of image acquisition modules and fill-in lights are arranged to be staggered and opposite to each other. The frame is hoisted and transported to the light-proof room through the driving module. The two sets of image acquisition modules and fill-in lights respectively acquire images of both sides of the frame, and compare them with the standard picture model through the image comparison module. The defects of the frame are detected through the image acquisition module and the image comparison module, and the frame is transported by the driving module for continuous detection to improve the detection efficiency and detection accuracy;

[0020] 2. The grid module is a program for dividing the image into grids. It divides the image collected by the image acquisition module into grid coordinates, marks the defective points, and outputs the coordinate information of the point on the image. The control module controls the plane drive component to move the marking spray gun to the corresponding position through the coordinate information to spray the mark on the defective point;

[0021] 3. When the driving module transports the frame to the position of the image acquisition module, the image acquisition module acquires images of it and calibrates the position of the frame by setting a positioning laser, so as to facilitate image acquisition at a more precise position, so as to facilitate subsequent image comparison and improve the accuracy of defect comparison. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of an overall flow chart of an embodiment of the present invention;

[0023] Figure 2 is a schematic diagram of the overall structure of an embodiment of the present invention;

[0024] Figure 3 is a partial structural schematic diagram of a light-shielding module according to an embodiment of the present invention;

[0025] Figure 4 It is a schematic diagram of the structure of the driving track part of an embodiment of the present invention;

[0026] Figure 5 It is a partial structural diagram of an auxiliary transmission module according to an embodiment of the present invention.

[0027] Reference numerals: 1, driving module; 11, driving track; 12, lifting assembly; 121, lifting hook;

[0028] 2. Light-proof module; 21. Light-proof room;

[0029] 3. Detection module; 31. Control module; 32. Image acquisition module; 33. Image comparison module; 34. Grid module;

[0030] 4. Fill light module; 41. Fill light; 42. Background board;

[0031] 5. Marking module; 51. Planar driving member; 511. Z-axis frame; 512. Y-axis frame; 52. Marking spray gun;

[0032] 6. Positioning module; 61. Positioning laser;

[0033] 7. Auxiliary transmission module; 71. Transmission assembly; 72. Semi-arc ring. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] See also Figure 1-Figure 3, a vehicle frame surface defect recognition system based on machine vision disclosed in an embodiment of the present invention, including a driving module 1, a light-shielding module 2, a detection module 3 and a fill-light module 4, the driving module 1 includes a driving track 11 arranged on the top of a workshop, the driving track 11 is slidably mounted with a plurality of hoisting components 12, the hoisting components 12 are provided with at least two hooks 121 for hoisting the vehicle frame, the light-shielding module 2 includes a light-shielding chamber 21, the upper end of the light-shielding chamber 21 is provided with a through slot for the hook 121 to pass through, and both ends of the light-shielding module 2 are provided with a blackout curtain, The detection module 3 includes a control module 31 and an image acquisition module 32 installed on the inner side of the light-proof chamber 21. The control module 31 is provided with an image contrast module 33. The fill light module 4 includes a fill light 41 arranged in the light-proof chamber 21. The fill light module 4 is signal-connected to the control module 31. The image acquisition module 32 is provided with at least two groups. The two groups of image acquisition modules 32 are staggered on both sides of the light-proof chamber 21. A background board 42 is provided on the side directly opposite to the image acquisition module 32. The fill light module 4 and the driving module 1 are both signal-connected to the control module 31.

[0036] Specifically, a detection environment with relatively stable light is set up through the light-proof room 21 and the blackout curtain, two groups of image acquisition modules 32 and fill-in lights 41 are installed in the light-proof room 21, and the two groups of image acquisition modules 32 and fill-in lights 41 are arranged to be staggered and opposite to each other, and the frame is hoisted and transported to the light-proof room 21 through the driving module 1. The two groups of image acquisition modules 32 and fill-in lights 41 respectively capture images of both sides of the frame, and compare them with the standard picture model through the image comparison module 33.

[0037] In this way, the defects of the frame are detected through the image acquisition module 32 and the image comparison module 33, and the frame is transported by the driving module 1 for continuous detection, thereby improving the detection efficiency and detection accuracy.

[0038] For some possible approaches, see Figure 4 The driving module 1 includes a driving track 11, which includes a circular track with a C-shaped cross section. A chain is installed in the circular track, and the chain is driven by a motor and a sprocket. The C-shaped cross section opens downward, and a connecting rod passing through the C-shaped cross section opening is fixedly installed under some chain links of the chain. Two hooks 121 are fixedly installed at the bottom of the connecting rod, and the frame is hoisted on the hooks 121;

[0039] The light-proof room 21 is in the shape of a rectangular passage, and a through slot is provided in the middle of the upper side of the light-proof room 21. When the driving module 1 runs through the light-proof room 21, the hook 121 moves along the through slot. The image acquisition module 32 is a camera, and the control module 31 is a computer-based control program for controlling the operation of the driving module 1, the image acquisition module 32, the fill light 41, etc. The image comparison module 33 is an image comparison program set in the computer, and the image collected by the image acquisition module 32 is compared with the standard picture. The blackout curtain can be set as a rectangular rubber strip set in a linear array, and can also be used as a control program for controlling the operation of the driving module 1, the image acquisition module 32, the fill light 41, etc. A door that can be set to be automatically closed is provided with double doors at both ends of the light-proof chamber 21, and a return spring is installed at the door axis. The door is pushed open by the frame lifted by the driving module 1, and is reset by the return spring after the frame enters, so that the fill light module 4 can become the only light source in the light-proof chamber 21 as much as possible, providing a relatively stable light environment for the image acquisition module 32, reducing the influence of light changes on the imaging effect, and improving the accuracy of contrast. The background board 42 is a rectangular plate. By setting the background board 42 as the background for image acquisition, the influence of changes in the imaging environment on the contrast result can be further reduced.

[0040] As a specific implementation of the machine vision-based vehicle frame surface defect recognition system provided in the application, please refer to Figure 3 , and also includes a marking module 5, the marking module 5 includes a planar driving member 51, the output end of the planar driving member 51 is fixedly connected to a marking spray gun 52, the marking module 5 is arranged at the outlet end of the light-proof chamber 21, and two marking modules 5 are arranged in mirror symmetry.

[0041] In general, some of the detected defects can be eliminated by repairing or the like. A marking module 5 is provided at the exit of the light-proof chamber 21. After the defects are detected, the defect points are marked by a marking spray gun 52 to facilitate the staff to carry out repairs.

[0042] For further information, see Figure 1 A grid module 34 is provided in the control module 31 . The grid module 34 divides the image into grid coordinates according to the image collected by the image acquisition module 32 . The marking module 5 is connected to the control module 31 by signal.

[0043] It should be understood that the grid module 34 is a program for gridding an image. It divides the image captured by the image acquisition module 32 into grid coordinates, marks the defective points, and outputs the coordinate information of the point on the image. The control module 31 controls the plane drive 51 through the coordinate information to move the marking spray gun 52 to the corresponding position and sprays the mark on the defective point.

[0044] For some possible approaches, see Figure 3The plane driving member 51 includes a Z-axis frame 511 slidably mounted on the light-shielding chamber 21, the Z-axis frame 511 is connected to a driving member, the Z-axis frame 511 is slidably mounted with a Y-axis frame 512, the Y-axis frame 512 is also connected to a driving member, and the marking spray gun 52 is mounted on the Y-axis frame 512. The driving member of the Z-axis frame 511 can be selected as a hydraulic rod, the Y-axis frame 512 can be selected as a ball screw mechanism, the marking spray gun 52 is connected to a spray pump and a dye barrel, and the dye should be easy to clean for subsequent processing.

[0045] As another specific implementation of the vehicle frame surface defect recognition system based on machine vision provided by the application, please refer to Figure 1 and Figure 3 , and also includes a positioning module 6, the positioning module 6 includes at least two positioning lasers 61, the positioning lasers 61 are installed on the background plate 42, and the positioning module 6 is connected to the control module 31 by signal.

[0046] In combination with specific usage scenarios, when the driving module 1 transports the frame to the position of the image acquisition module 32, the image acquisition module 32 acquires images of it. By setting the positioning laser 61, the position of the frame is calibrated to facilitate image acquisition at a more precise position, so as to facilitate subsequent image comparison and improve the accuracy of defect comparison.

[0047] In some feasible embodiments, the positioning laser 61 includes a transmitting end and a receiving end. The positioning laser 61 is fixedly installed at the farthest points at both ends of the frame. When the frame reaches the appropriate position, both positioning lasers 61 are blocked, and then the receiving end is used to determine whether the frame has reached the appropriate position.

[0048] As a specific implementation of the machine vision-based vehicle frame surface defect recognition system provided in the application, please refer to Figure 2 and Figure 5 An auxiliary transmission module 7 is installed at the through slot position, and the auxiliary transmission module 7 includes two transmission components 71 arranged on both sides of the through slot. The hook 121 is cylindrical, and a semi-arc ring 72 is arranged on the conveyor belt surface of the transmission component 71. The semi-arc rings 72 of the two transmission components 71 are mirror-symmetrically fitted on the sides, and the auxiliary transmission module 7 is connected to the control module 31 by signal.

[0049] It should be understood that after the crane module 1 hoists the frame into the through slot, the two semi-arc rings 72 are clamped with the hooks 121. By clamping the two hooks 121, the frame is fixed during transportation, thereby improving the stability of the frame when running in the light-proof room 21, thereby improving the stability during image acquisition and improving the accuracy of defect comparison.

[0050] In some feasible embodiments, the transmission component 71 can be selected as a chain transmission mechanism or a toothed belt mechanism, and the semi-arc ring 72 includes a rectangular mounting plate, a semicircular arc ring is arranged on the side of the rectangular mounting plate, and the other side of the rectangular mounting plate can be rotatably mounted on the link of the chain.

[0051] Furthermore, sponge strips are bonded to the two side walls of the through groove, and the sides of the two sponge strips are fitted together.

[0052] It should be understood that the sponge strip is used to close the through slot, thereby reducing the influence of external light on the light-proof chamber 21, making the light environment in the light-proof chamber 21 more stable. At the same time, the sponge strip is a soft material, which facilitates the passage of the hook 121.

[0053] In some feasible embodiments, the rod-shaped portion of the hook 121 can be set as a telescopic rod and fixed by a top screw. The frame is an irregular structure. By setting the telescopic rod, the hook 121 can adjust the drooping length according to different hanging positions to adapt to different frames.

[0054] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A frame surface defect recognition system based on machine vision, characterized in that: include: A crane module, the crane module comprising a crane track arranged on the top of the workshop, a plurality of lifting assemblies being slidably mounted on the crane track, and the lifting assemblies being provided with at least two hooks for lifting the frame; A light-proof module, the light-proof module comprising a light-proof chamber, a through slot for the hook to pass through is provided at the upper end of the light-proof chamber, and light-shielding curtains are provided at both ends of the light-proof module; A detection module, the detection module comprising a control module and an image acquisition module installed inside the light-proof chamber, the control module being provided with an image comparison module; A fill light module, the fill light module comprising a fill light disposed in the light-proof chamber, the fill light module being signal-connected to the control module; Among them, at least two groups of image acquisition modules are provided, and the two groups of image acquisition modules are staggeredly arranged on both sides of the light-proof room. A background board is provided on the side facing the image acquisition module, and the fill light module and the driving module are both connected to the control module signal.

2. The frame surface defect recognition system based on machine vision according to claim 1, characterized in that: It also includes a marking module, which includes a planar driving component, an output end of which is fixedly connected to a marking spray gun, and the marking module is arranged at the exit end of the light-proof chamber, and two marking modules are arranged in mirror symmetry.

3. The frame surface defect recognition system based on machine vision according to claim 2, characterized in that: The control module is provided with a grid module, and the grid module divides the image into grid coordinates according to the image collected by the image collection module. The marking module is connected with the control module by signal.

4. The frame surface defect recognition system based on machine vision according to claim 3 is characterized in that: The planar driving member comprises a Z-axis frame slidably mounted on the light-proof chamber, the Z-axis frame is connected to a driving member, the Z-axis frame is slidably mounted with a Y-axis frame, the Y-axis frame is also connected to a driving member, and the marking spray gun is mounted on the Y-axis frame.

5. The frame surface defect recognition system based on machine vision according to claim 1, characterized in that: It also includes a positioning module, which includes at least two positioning lasers. The positioning lasers are installed on the background plate, and the positioning module is connected to the control module by signal.

6. The frame surface defect recognition system based on machine vision according to claim 1, characterized in that: An auxiliary transmission module is installed at the through slot position, and the auxiliary transmission module includes two transmission components arranged on both sides of the through slot. The hook is cylindrical, and the conveyor belt surface of the transmission component is provided with a semi-arc ring. The semi-arc rings of the two transmission components are mirror-symmetrically fitted on the side, and the auxiliary transmission module is connected to the control module signal.

7. The frame surface defect recognition system based on machine vision according to claim 1, characterized in that: Sponge strips are bonded to the two side walls of the through slot, and the side surfaces of the two sponge strips are fitted together.