Metal accessory defect monitoring device based on machine vision

Abstract

The invention discloses a metal fitting defect monitoring device based on machine vision, which is applied to the technical field of metal fitting detection, and is characterized in that a main frame is arranged to be matched with a mounting plate, so that structures can be conveniently connected, and a back frame is arranged to be matched with a top frame, so that a back plate and a front cover can be mounted; the front cover and the back plate are used for shading an image acquisition position, so that the influence of external ambient light on the image acquisition effect is avoided, and when the transmission shaft is matched with the synchronous pulley and the external motor to drive the transmission shaft to rotate, the connecting belt drives the transparent plate to move; then the detected metal fitting is placed on the transparent plate, all-around image acquisition and monitoring can be carried out on the metal fitting, an arranged vision auxiliary system can conveniently assist a machine vision system to work, and a fault diagnosis module can conveniently carry out continuous fault detection on the system; the image acquisition module is used for acquiring an image of the detected metal fitting.

Description

Technical Field

[0001] The present invention belongs to the technical field of metal fitting detection, and particularly relates to a metal fitting defect monitoring device based on machine vision. Background Art

[0002] Machine vision is a rapidly developing branch of artificial intelligence. Briefly speaking, machine vision uses machines to replace human eyes for measurement and judgment. A machine vision system converts the captured target into an image signal through machine vision products, transmits it to a dedicated image processing system to obtain the morphological information of the captured target, and converts it into a digital signal according to pixel distribution, brightness, color, and other information; the image system performs various operations on these signals to extract the features of the target, and then controls the actions of on-site equipment according to the discrimination results. When monitoring a batch of metal fittings, the labor intensity of personnel can be reduced and the accuracy can be improved through machine vision.

[0003] When the existing metal fitting defect monitoring device based on machine vision monitors metal fittings, it is limited by its placement position, and it is difficult to collect images and conduct reasonable monitoring at the bottom. Moreover, there is still room for improvement in the processing of images by machine vision. It is difficult to record and alarm defects in real time, and it can only simply exclude defective parts, and it is difficult to adjust the metal fitting processing technology according to the actual situation.

[0004] Based on the above-mentioned situation, we found that it is difficult for the existing metal fitting defect monitoring device based on machine vision to avoid the above problems at the same time. Therefore, we propose a metal fitting defect monitoring device based on machine vision that can collect and monitor images of metal fittings in multiple directions, optimize the detection and maintenance efficiency at the same time, and can mark and record defects. Summary of the Invention

[0005] The purpose of the present invention is to provide a metal fitting defect monitoring device based on machine vision, and its advantage is that it can collect and monitor images of metal fittings in multiple directions, optimize the detection and maintenance efficiency at the same time, and can mark and record defects.

[0006] The above technical purpose of the present invention is achieved through the following technical solutions: A metal fitting defect monitoring device based on machine vision includes a main frame and a vision assistance system. An installation plate is bolted inside the main frame, a back frame is bolted to the bottom of the main frame, a top frame is bolted to the top of the back frame, a back plate is bolted to the front side of the back frame, a front cover is rotatably connected to the top of the top frame, a transmission shaft is rotatably connected inside the main frame, a synchronous belt pulley is bolted to the outside of the transmission shaft, the transmission shaft is externally connected to a control motor, a connecting belt is movably connected to the inside of the synchronous belt pulley, and a transparent plate is movably connected to the top of the connecting belt; The vision assistance system includes a central control module. The input end of the central control module is bidirectionally electrically connected to a fault diagnosis module. The input end of the central control module is unidirectionally electrically connected to an image acquisition module. The output end of the central control module is bidirectionally electrically connected to a defect detection module. The output end of the central control module is bidirectionally electrically connected to a log recording module. The output end of the central control module is bidirectionally electrically connected to a user interface module. The output end of the central control module is bidirectionally electrically connected to an image preprocessing module. The output end of the central control module is bidirectionally electrically connected to a feature extraction module. The output end of the central control module is bidirectionally electrically connected to a database module. The output end of the central control module is bidirectionally electrically connected to a maintenance management module.

[0007] With the above technical solutions, by setting the main frame and the mounting plate, it is convenient to connect the structures. The back frame and the top frame are set to install the back plate and the front cover. The front cover and the back plate are used to shade the image acquisition position to avoid the influence of external ambient light on the image acquisition effect. The set transmission shaft and synchronous belt pulley are configured such that when the external motor drives the transmission shaft to rotate, the connecting belt drives the transparent plate to move. Then, the metal fitting to be detected is placed on the transparent plate, and thus all-round image acquisition and monitoring of the metal fitting can be performed. The set vision assistance system can facilitate the operation of the machine vision system. Its fault diagnosis module can continuously detect the faults of the system itself. The image acquisition module is used to acquire the images of the metal fittings to be detected. The set defect detection module can facilitate the detection of defects of the metal fittings based on the acquired images. The set log recording module can facilitate the recording of the operations of the equipment. The user interface module is convenient for personnel to control the equipment. The image preprocessing module can preprocess the acquired images, facilitating direct calling in the later stage and reducing the false alarm probability. The set database module can facilitate the storage and later calling of data. The maintenance management module can facilitate equipment maintenance and remote management.

[0008] The present invention is further configured as: a connection frame is bolted to the outside of the transparent plate, and the bottom of the connection frame is clamped with the connecting belt.

[0009] With the above technical solutions, by setting the connection frame, it is convenient to connect the transparent plate and the connecting belt and drive the structure to move.

[0010] The present invention is further configured as: a supplementary light is bolted to the top of the mounting plate, and image acquisition cameras are bolted to the top of the mounting plate and the bottom of the top frame respectively.

[0011] With the above technical solution, by setting up fill lights, it is convenient to supplement light for the structure, making the image acquisition environment consistent, reducing errors. The multiple image acquisition cameras set up are used to comprehensively perform image acquisition actions on metal fittings.

[0012] The present invention is further configured as: the defect detection module includes a defect detection module, a flaw classification module, and a flaw localization module.

[0013] With the above technical solution, by setting up a defect detection module, it is convenient to detect defects based on the acquired image data. The set flaw classification module and flaw localization module cooperate to facilitate the classification processing of the type and location of flaws.

[0014] The present invention is further configured as: the log recording module includes an abnormal alarm recording module, an operation recording module, and a system performance evaluation module.

[0015] With the above technical solution, by setting up an abnormal alarm recording module, it is convenient to give timely alarm notifications for abnormal situations. The set operation recording module can facilitate the recording of various operation types, and the set system performance evaluation module can facilitate the evaluation of the recorded content.

[0016] The present invention is further configured as: the user interface module includes a parameter setting interface module, a real-time monitoring interface module, and an alarm information display module.

[0017] With the above technical solution, by setting up the parameter setting interface module in cooperation with the real-time monitoring interface module, it is convenient for personnel to set parameters and observe the monitoring content. The set alarm information display module can facilitate the recording and notification of alarm information, making it convenient for personnel to observe.

[0018] The present invention is further configured as: the image preprocessing module includes an image denoising processing module, an image smoothing processing module, and an image enhancement processing module.

[0019] With the above technical solution, by setting up the image denoising module in cooperation with the image smoothing processing module, it is convenient to clarify and smooth the image, reducing false alarms of flaws caused by image errors. The set image enhancement processing module can facilitate the enhancement processing of image features, making them more accurate.

[0020] The present invention is further configured as: the feature extraction module includes an edge detection module, a texture extraction module, and a shape extraction module.

[0021] With the above technical solution, by setting up an edge detection module, it is convenient to detect the edge of the metal fitting to be detected. The set texture extraction module can facilitate the extraction of the texture features of the metal fitting, and the set shape extraction module can facilitate the extraction of the metal fitting.

[0022] The present invention is further configured as: the database module includes a historical data storage module, a detection result recording module, and an image storage module.

[0023] With the above technical solution, by setting up a historical data storage module in cooperation with a detection result recording module and an image storage module, it is convenient to store and call past data and image content.

[0024] The present invention is further configured as: the maintenance management module includes a remote maintenance module, a firmware upgrade module, and a system diagnosis module.

[0025] With the above technical solution, by setting up a remote maintenance module, it is convenient for personnel to remotely maintain the system. The set firmware upgrade module can facilitate continuous update and upgrade, and the set system diagnosis module can facilitate self-checking of the system.

[0026] In summary, the present invention has the following beneficial effects: By setting up a main frame in cooperation with a mounting plate, it is convenient to connect the structure. The set back frame in cooperation with the top frame can install the back plate and the front cover. The front cover and the back plate are used to shield the image acquisition position to avoid the influence of external ambient light on the image acquisition effect. By setting up a transmission shaft in cooperation with a synchronous pulley, when an external motor drives the transmission shaft to rotate, the connecting belt will drive the transparent plate to move, and then the metal fitting to be detected can be placed on the transparent plate to perform omnidirectional image acquisition and monitoring of the metal fitting. The set visual assistance system can facilitate the work of the machine vision system. Its fault diagnosis module can facilitate continuous fault detection of the system itself. The image acquisition module is used to acquire the image of the metal fitting to be detected. The set defect detection module can facilitate the detection of defects of the metal fitting based on the acquired image. The set log recording module can facilitate the recording of the operation of the device. The interface module is convenient for personnel to control the device. The image preprocessing module can preprocess the acquired image, which is convenient for later direct call and reduces the false alarm probability. The set database module can facilitate the storage and later call of data. The maintenance management module can facilitate device maintenance and remote management. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is the overall structural schematic diagram of the present invention; Figure 2 is the internal structural schematic diagram of the present invention; Figure 3 It is a schematic diagram of the visual assistance system of the present invention.

[0028] Reference numerals: 1, main frame; 2, visual assistance system; 201, central control module; 202, fault diagnosis module; 203, image acquisition module; 204, defect detection module; 205, log recording module; 206, user interface module; 207, image preprocessing module; 208, feature extraction module; 209, database module; 210, maintenance management module; 3, mounting plate; 4, back frame; 5, top frame; 6, back panel; 7, front cover; 8, drive shaft; 9, synchronous pulley; 10, system diagnosis module; 11, connecting belt; 12, transparent plate; 13, connecting frame; 14, supplementary light; 15, image acquisition camera; 16, defect detection module; 17, defect classification module; 18, defect location module; 19, abnormal alarm recording module; 20, operation record module; 21, system performance evaluation module; 22, parameter setting interface module; 23, real-time monitoring interface module; 24, alarm information display module; 25, image denoising processing module; 26, image smoothing processing module; 27, image enhancement processing module; 28, edge detection module; 29, texture extraction module; 30, shape extraction module; 31, historical data storage module; 32, detection result record module; 33, image storage module; 34, remote maintenance module; 35, firmware upgrade module. Detailed implementation manners

[0029] The present invention will be further described in detail below with reference to the accompanying drawings. Embodiment

[0030] Referring to Figure 1-2 , a metal fitting defect monitoring device based on machine vision includes a main frame 1 and a visual assistance system 2. A mounting plate 3 is bolted to the inner side of the main frame 1, a back frame 4 is bolted to the bottom of the main frame 1, a top frame 5 is bolted to the top of the back frame 4, a back panel 6 is bolted to the front side of the back frame 4, the top of the top frame 5 is rotatably connected to a front cover 7, a drive shaft 8 is rotatably connected to the inner side of the main frame 1, a synchronous pulley 9 is bolted to the outer side of the drive shaft 8, the drive shaft 8 is externally connected to a control motor, and the inner side of the synchronous pulley 9 is movably connected to a connecting belt 11. The top of the connecting belt 11 is movably connected to a transparent plate 12. By providing the main frame 1 in cooperation with the mounting plate 3, it is convenient to connect the structures. By providing the back frame 4 in cooperation with the top frame 5, the back panel 6 and the front cover 7 can be installed. The front cover 7 and the back panel 6 are used to shield the image acquisition position to avoid the influence of external ambient light on the image acquisition effect. By providing the drive shaft 8 in cooperation with the synchronous pulley 9, when the externally connected motor drives the drive shaft 8 to rotate, the connecting belt 11 will drive the transparent plate 12 to move, and then the metal fitting to be detected can be placed on the transparent plate 12 to perform omnidirectional image acquisition and monitoring of the metal fitting.

[0031] As Figure 2 shown, a connection frame 13 is bolted to the outside of the transparent plate 12. The bottom of the connection frame 13 is snap-connected to the connection belt 11. By providing the connection frame 13, it is convenient to connect the transparent plate 12 and the connection belt 11 and drive the structure to move.

[0032] As Figure 2 shown, a supplementary light 14 is bolted to the top of the mounting plate 3, and image acquisition cameras 15 are bolted to both the top of the mounting plate 3 and the bottom of the top frame 5. By providing the supplementary light 14, it is convenient to supplement light to the structure, making the image acquisition environment consistent and reducing errors. The multiple image acquisition cameras 15 provided are used to comprehensively perform image acquisition actions on the metal fittings.

[0033] Brief description of the usage process: First, the structure is connected through the main frame 1 and the mounting plate 3, and the back frame 4 is coordinated with the top frame 5 to install the back plate 6 and the front cover 7. The front cover 7 and the back plate 6 are used to shield the image acquisition position to avoid the influence of external ambient light on the image acquisition effect. When the externally connected motor drives the transmission shaft 8 to rotate with the provided transmission shaft 8 and the synchronous belt pulley 9, the connection belt 11 will drive the transparent plate 12 to move. Then, the metal fitting to be detected is placed on the transparent plate 12, and the image acquisition cameras 15 can perform omnidirectional image acquisition and monitoring on the metal fitting. Embodiment

[0034] Reference Figure 3, A metal fitting flaw monitoring device based on machine vision, including a vision assistance system 2. The vision assistance system 2 includes a central control module 201. The input end of the central control module 201 is bidirectionally electrically connected to a fault diagnosis module 202. The input end of the central control module 201 is unidirectionally electrically connected to an image acquisition module 203. The output end of the central control module 201 is bidirectionally electrically connected to a flaw detection module 204. The output end of the central control module 201 is bidirectionally electrically connected to a log recording module 205. The output end of the central control module 201 is bidirectionally electrically connected to a user interface module 206. The output end of the central control module 201 is bidirectionally electrically connected to an image preprocessing module 207. The output end of the central control module 201 is bidirectionally electrically connected to a feature extraction module 208. The output end of the central control module 201 is bidirectionally electrically connected to a database module 209. The output end of the central control module 201 is bidirectionally electrically connected to a maintenance management module 210. The set vision assistance system 2 can facilitate the work of the machine vision system. Its fault diagnosis module 202 can facilitate continuous fault detection of the system itself. The image acquisition module 203 is used to acquire images of the metal fittings to be detected. The set flaw detection module 204 can facilitate the detection of flaws in the metal fittings based on the acquired images. The set log recording module 205 can facilitate the recording of the operations of the device. The user interface module is convenient for personnel to control the device. The image preprocessing module 207 can preprocess the acquired images, facilitating direct later calls and reducing the false alarm probability. The set database module 209 can facilitate data storage and later calls. The maintenance management module 210 can facilitate device maintenance and remote management.

[0035] The flaw detection module 204 includes a defect detection module 16, a flaw classification module 17, and a flaw localization module 18. By setting the defect detection module 16, it can facilitate the detection of defects based on the acquired image data. The set flaw classification module 17, in cooperation with the flaw localization module 18, can facilitate the classification of the types and positions of flaws.

[0036] The log recording module 205 includes an abnormal alarm recording module 19, an operation recording module 20, and a system performance evaluation module 21. By setting the abnormal alarm recording module 19, it can facilitate the timely alarm notification of abnormal situations. The set operation recording module 20 can facilitate the recording of various operation types. The set system performance evaluation module 21 can facilitate the evaluation of the recorded content.

[0037] The user interface module 206 includes a parameter setting interface module 22, a real-time monitoring interface module 23, and an alarm information display module 24. By setting the parameter setting interface module 22 in cooperation with the real-time monitoring interface module 23, it is convenient for personnel to set parameters and observe the monitoring content. The set alarm information display module 24 can facilitate the recording and notification of alarm information, making it convenient for personnel to observe.

[0038] The image preprocessing module 207 includes an image denoising processing module 25, an image smoothing processing module 26, and an image enhancement processing module 27. The image preprocessing module 207 includes an image denoising processing module 25, an image smoothing processing module 26, and an image enhancement processing module 27.

[0039] The feature extraction module 208 includes an edge detection module 28, a texture extraction module 29, and a shape extraction module 30. By setting the edge detection module 28, it is convenient to detect the edges of the metal fittings to be detected. The set texture extraction module 29 can facilitate the extraction of the texture features of the metal fittings, and the set shape extraction module 30 can facilitate the extraction of the metal fittings.

[0040] The database module 209 includes a historical data storage module 31, a detection result recording module 32, and an image storage module 33. By setting the historical data storage module 31 in cooperation with the detection result recording module 32 and the image storage module 33, it is convenient to store and call past data and image content.

[0041] The maintenance and management module 210 includes a remote maintenance module 34, a firmware upgrade module 35, and a system diagnosis module 10. By setting the remote maintenance module 34, it is convenient for personnel to remotely maintain the system. The set firmware upgrade module 35 can facilitate continuous update and upgrade, and the set system diagnosis module 10 can facilitate self-checking of the system.

[0042] Brief description of the usage process: The fault diagnosis module 202 of the visual assistance system 2 can facilitate continuous fault detection of the system itself. The image acquisition module 203 is used to acquire images of the metal fittings to be detected. The set flaw detection module 204 can facilitate the detection of flaws in the metal fittings based on the acquired images. The set log recording module 205 can facilitate the recording of the operations of the device. The interface module is used to facilitate personnel to control the device. The image preprocessing module 207 can preprocess the acquired images, which is convenient for direct later call and reduces the false alarm probability. The set database module 209 can facilitate data storage and later call. The maintenance management module 210 can facilitate device maintenance and remote management. It has intelligent functions such as automatic flaw annotation and fault diagnosis algorithms, which can reduce the manual burden and improve the detection and maintenance efficiency. The system designs functional modules such as remote maintenance, firmware upgrade, and automatic troubleshooting, improving the reliability and stability of the system. It provides a comprehensive user interface module 206, including functions such as parameter setting, real-time monitoring, and alarm information display, making the operation more convenient.

[0043] This specific embodiment is only an interpretation of the present invention and is not a limitation thereof. After reading this specification, those skilled in the art can make modifications to this embodiment without creative contributions as needed, but as long as it is within the scope of the claims of the present invention, it is protected by the patent law.

Claims

1. A metal parts defect monitoring device based on machine vision, comprising a main frame (1) and a visual assistance system (2), characterized in that: The main frame (1) is bolted with a mounting plate (3) on the inner side, the main frame (1) is bolted with a back frame (4) on the bottom, the back frame (4) is bolted with a top frame (5) on the top, the back frame (4) is bolted with a back plate (6) on the front side, the top of the top frame (5) is rotatably connected to a front cover (7), the main frame (1) is rotatably connected with a transmission shaft (8) on the inner side, a synchronous pulley (9) is bolted with the outer side of the transmission shaft (8), the transmission shaft (8) is externally connected to a control motor, the inner side of the synchronous pulley (9) is movably connected to a connecting belt (11), and the top of the connecting belt (11) is movably connected to a transparent plate (12); The visual assistance system (2) comprises a central control module (201), wherein an input end of the central control module (201) is bidirectionally electrically connected to a fault diagnosis module (202), an input end of the central control module (201) is unidirectionally electrically connected to an image acquisition module (203), an output end of the central control module (201) is bidirectionally electrically connected to a defect detection module (204), an output end of the central control module (201) is bidirectionally electrically connected to a log recording module (205), an output end of the central control module (201) is bidirectionally electrically connected to a user interface module (206), an output end of the central control module (201) is bidirectionally electrically connected to an image preprocessing module (207), an output end of the central control module (201) is bidirectionally electrically connected to a feature extraction module (208), an output end of the central control module (201) is bidirectionally electrically connected to a database module (209), and an output end of the central control module (201) is bidirectionally electrically connected to a maintenance management module (210).

2. The metal parts defect monitoring device based on machine vision according to claim 1, characterized in that: A connection frame (13) is bolted to the outer side of the transparent plate (12), and the bottom of the connection frame (13) is snap-connected to the connection belt (11).

3. The metal parts defect monitoring device based on machine vision according to claim 1, characterized in that: A fill light (14) is bolted to the top of the mounting plate (3), and an image acquisition camera (15) is bolted to the top of the mounting plate (3) and the bottom of the top frame (5).

4. The metal parts defect monitoring device based on machine vision according to claim 1, characterized in that: The defect detection module (204) comprises a defect detection module (16), a defect classification module (17) and a defect positioning module (18).

5. The metal parts defect monitoring device based on machine vision according to claim 1, characterized in that: The log recording module (205) comprises an abnormal alarm recording module (19), an operation recording module (20) and a system performance evaluation module (21).

6. The metal parts defect monitoring device based on machine vision according to claim 1, characterized in that: The user interface module (206) comprises a parameter setting interface module (22), a real-time monitoring interface module (23) and an alarm information display module (24).

7. The metal parts defect monitoring device based on machine vision according to claim 1, characterized in that: The image preprocessing module (207) comprises an image denoising processing module (25), an image smoothing processing module (26) and an image enhancement processing module (27).

8. The metal parts defect monitoring device based on machine vision according to claim 1, characterized in that: The feature extraction module (208) includes an edge detection module (28), a texture extraction module (29) and a shape extraction module (30).

9. The metal parts defect monitoring device based on machine vision according to claim 1, characterized in that: The database module (209) comprises a historical data storage module (31), a test result recording module (32) and an image storage module (33).

10. The metal parts defect monitoring device based on machine vision according to claim 1, characterized in that: The maintenance management module (210) comprises a remote maintenance module (34), a firmware upgrade module (35) and a system diagnosis module (10).

Images