Visual intelligent inspection device and inspection method for defects of overhead rail busbar of hoisting trolley

Abstract

The present invention discloses a visual intelligent inspection device and an inspection method for defects in a ceiling rail busbar of a lifting trolley, wherein the visual intelligent inspection device comprises a visual intelligent inspection terminal and a server, the visual intelligent inspection terminal is installed on the lifting trolley, the visual intelligent inspection terminal comprises an industrial computer, the industrial computer comprises an industrial camera, a power supply, a camera light source, an auxiliary light source and an operation control mainboard, the operation control mainboard and the server are located in the same local area network, inspection operations can be performed during the operation of the equipment, and the inspection cycle is flexible and maneuverable; the picture information of the ceiling rail busbar is collected by shooting with a set high-speed camera, the barcode picture on the ceiling rail busbar is shot with a digital camera, and the defect information and position information of the ceiling rail busbar is determined and fed back through the operation control mainboard and the server, without manual participation, thereby improving the inspection efficiency; the server continuously improves the inspection precision and accuracy of the defects of the ceiling rail busbar by establishing a detection network model.

Description

Technical Field

[0001] The present invention belongs to the field of bottled product conveying production equipment, and in particular relates to a visual intelligent inspection device and an inspection method for defects of a ceiling rail busbar of a lifting trolley. Background Art

[0002] At present, the lifting trolley is used for the transportation tasks such as the delivery and in-and-out of finished beer products and the distribution of production materials between the stereoscopic warehouse and the production workshop of the manufacturer. The lifting trolley travels on the overhead rail. The busbar in the middle of the overhead rail is mainly used to provide power supply and communication data transmission for the lifting trolley. When the busbar is damaged or defective, it will cause the lifting trolley to malfunction or stop during driving, which has a great impact on normal production operations. The current defects include the following situations:

[0003] 1. The copper rail inside the busbar is damaged and has a gap;

[0004] 2. Irregular mottled patterns appear on the surface of the copper rail inside the busbar;

[0005] 3. Breakpoints appear on the surface of the copper rail inside the sliding conductor;

[0006] 4. The plastic cover of the busbar is damaged.

[0007] Therefore, regular daily inspections are the most effective means of eliminating defects. Current inspections generally have the following problems:

[0008] 1. Inspections must be carried out when all equipment is stopped and power is off. Production is affected and relevant work plans cannot be formulated. The inspection frequency cannot be determined, which is not conducive to troubleshooting defects and establishing maintenance files;

[0009] 2. When the defective position of the busbar is found during the inspection, it is impossible to determine the ID address of the position from the number of the barcode (the barcode is pasted on the ceiling rail and arranged parallel to the busbar, which represents an information code of an absolute value) in time, and the barcode needs to be converted manually to know;

[0010] 3. Manual inspection cannot guarantee the precision and accuracy of the inspection, which depends on the personal ability and work experience of the inspector. Since it relies on human visual observation, the inspection results may have loopholes, and the inspection methods and problem identification capabilities of maintenance personnel may be biased, resulting in omissions and hidden dangers in equipment operation;

[0011] 4. Manual inspection consumes a lot of manpower and takes a long time, resulting in low work efficiency. Summary of the invention

[0012] In order to overcome the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide a visual intelligent inspection device for defects in the overhead rail busbar of a lifting trolley, aiming to solve the technical problems that the current inspection work affects production, the location information of busbar defects cannot be obtained in time, the manual inspection method has low work efficiency, and the precision and accuracy of the inspection cannot be guaranteed.

[0013] In order to achieve the purpose, the technical solution adopted by the present invention is as follows:

[0014] A visual intelligent inspection device for defects of a ceiling rail busbar of a lifting trolley comprises a visual intelligent inspection terminal and a server, wherein the visual intelligent inspection terminal comprises an industrial computer based on an Ubuntu system, the industrial computer comprises an industrial camera, a power supply, a camera light source, an auxiliary light source and an operation control mainboard, a rotating device and a fixing device are also provided on the industrial computer, one end of the rotating device is connected to the lifting trolley, one end of the industrial computer rotates relative to the lifting trolley through the rotating device, and the other end of the industrial computer is connected to the lifting trolley through the fixing device, the power supply is respectively connected to the industrial camera, the camera light source, the auxiliary light source and the operation control mainboard, the operation control mainboard is respectively connected to the industrial camera, the camera light source and the auxiliary light source, the operation control mainboard and the server are located in the same local area network, and the industrial camera comprises a high-speed camera and a digital camera.

[0015] Furthermore, the industrial computer includes a shell and a mounting plate, the power supply and operation control mainboard are arranged on one mounting surface of the mounting plate, the industrial camera, camera light source, auxiliary light source, rotating device and fixing device are arranged on another mounting surface, and the shell covers the power supply and operation control mainboard.

[0016] Furthermore, the rotating device includes a rotating frame, on which a rotating shaft is provided, and the upper and lower ends of the rotating shaft are respectively rotatably connected to the lifting trolley. The fixing device includes a fixing frame, on which a fixing fastener is provided, and on the lifting trolley is a fixing seat that cooperates with the fixing fastener.

[0017] Furthermore, the operation control mainboard is provided with a touch screen, the shell is provided with a first avoidance hole at a position corresponding to the touch screen, a power display device is provided on the power supply, and the shell is provided with a second avoidance hole at a position corresponding to the power display device.

[0018] Furthermore, a plurality of ventilation holes are provided on the side of the shell.

[0019] Furthermore, when the other end of the industrial computer is connected to the lifting trolley, the industrial camera, camera light source and auxiliary light source are oriented toward the direction of the overhead rail busbar.

[0020] Furthermore, the server includes a detection module, a request processing module and an interaction module. The request processing module is used to receive information collected by the visual intelligent inspection terminal and send information processed by the detection module, the detection module is used to process the information collected by the visual intelligent inspection terminal, and the interaction module is used to control the operation control mainboard.

[0021] The present invention also proposes an inspection method for a visual intelligent inspection device for defects of a ceiling rail busbar of a hoisting trolley, comprising the following steps:

[0022] S1. Install the visual intelligent inspection terminal on the lifting trolley and start the lifting trolley;

[0023] S2. The high-speed camera and digital camera of the visual intelligent inspection terminal collect image data of the overhead rail busbar, and the camera light source and auxiliary light source are used for illumination. The high-speed camera captures and collects image information of the overhead rail busbar, and the digital camera captures the barcode image on the overhead rail busbar, and sends the collected image data to the operation control mainboard;

[0024] S3, the computing control mainboard processes the barcode image information and the overhead rail busbar image information respectively, identifies the ID address of the overhead rail busbar, analyzes the defects of the overhead rail busbar, and sends the defect information of the overhead rail busbar to the server for secondary analysis;

[0025] S4. The operation control mainboard integrates the analyzed defect information and ID address information of the overhead rail busbar and sends them back to the server. The server classifies the integrated information, and then records the determined defect information of the overhead rail busbar and issues feedback.

[0026] Further, in step S3, the operation control mainboard analyzes the defects of the overhead rail busbar, including the following steps:

[0027] s31. Determine the connection of the overhead rail busbar: collect the connection and non-connection images in the overhead rail busbar image information as a data set, establish a classification model for the connection and non-connection, and make a judgment in the classification model;

[0028] s32. Establish a detection network model in the server and train the detection network model. The training steps include: first, the detection network model iteratively learns the defective pictures of the overhead rail busbar, fits the learned data set to the weight parameter dictionary, performs the prediction task through the weight parameter dictionary, and performs multi-category defect recognition and classification on the defective pictures of the overhead rail busbar; secondly, the scores of the defects of the rail busbar under the defect category are estimated; thirdly, the defects of the rail busbar are multi-scale fused, including the fusion of shallow features and deep features; thirdly, the defective pictures of the overhead rail busbar in the data set are classified into difficulty; finally, the defective pictures of the overhead rail busbar are processed before and after, including traditional algorithms such as white balance and image alignment; after the training is completed, the detection network model performs secondary analysis on the defective pictures of the overhead rail busbar;

[0029] Furthermore, in step s32, the detection network model adopts a neural network architecture with MobileNet as the main deep neural network backbone framework, followed by an improved classification network layer; the detection network model is connected to a small network composed mainly of convolutional layers and Softmax layers to estimate the scores of the defects of the rail busbar under the defect category; the difficulty classification of the defective images of the overhead rail busbar in the data set includes offline and online difficulty classification, wherein the offline difficulty classification is to add annotations to the static labels of the data set before training, and add one-dimensional labels for training during the training process, or it can be to classify the data by the difference between the score value after the first training and the score of the label, and the difficulty score will no longer be updated in the subsequent training stage; the online difficulty classification is to classify the training sample results of the current stage, and use them for training in the next training stage, and the processing of the samples after difficulty classification: the classification result scores of simple samples have higher weights for loss value optimization, and the loss values ​​of difficult samples have lower weights for loss value optimization.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] The present invention proposes a visual intelligent inspection device for defects of the overhead rail busbar of a lifting trolley. The visual intelligent inspection terminal of the device is installed on the lifting trolley. Inspection operations can be carried out during the operation of the equipment, and the inspection cycle is flexible and maneuverable. The image information of the overhead rail busbar is collected by a set high-speed camera, and the barcode image on the overhead rail busbar is photographed by a digital camera. The defect information and position information of the overhead rail busbar are determined and fed back through the operation control mainboard and server, without manual participation, thereby improving the inspection efficiency. The server continuously improves the inspection precision and accuracy of the overhead rail busbar defects by establishing a detection network model. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0033] Figure 1 A front view of an industrial computer according to an embodiment of the present invention;

[0034] Figure 2 A side view of an industrial computer according to an embodiment of the present invention;

[0035] Figure 3 A top view of an industrial computer according to an embodiment of the present invention;

[0036] Figure 4 It is a front view of an industrial computer without a housing in one embodiment of the present invention;

[0037] Figure 5 It is a rear view of an industrial computer without a housing in one embodiment of the present invention;

[0038] Figure 6 The present invention is a flow chart of the inspection method.

[0039] Description of reference numerals:

[0040] 1-housing, 2-mounting plate, 3-power supply, 4-operation control mainboard, 5-rotating frame, 6-fixing frame, 7-fixing fastener, 8-touch screen, 9-power display device, 10-ventilation exhaust hole, 11-high-speed camera, 12-digital camera. DETAILED DESCRIPTION

[0041] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. In the following description, many specific details are set forth in order to fully understand the present invention, and the embodiments described are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present invention.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0043] Reference Figures 1 to 6 The embodiment of the present invention provides a visual intelligent inspection device for defects of a ceiling rail busbar of a lifting trolley, including a visual intelligent inspection terminal and a server. The visual intelligent inspection terminal includes an industrial computer based on the Ubuntu system. The industrial computer includes a shell 1 and a mounting plate 2. A power supply 3 and a calculation control mainboard 4 are provided on one mounting surface of the mounting plate 2. An industrial camera, a camera light source, an auxiliary light source, a rotating device and a fixing device are provided on the other mounting surface of the mounting plate 2. The shell 1 covers the power supply 3 and the calculation control mainboard 4. The rotating device includes a rotating frame 5. A rotating shaft is provided on the rotating frame 5. The upper and lower ends of the rotating shaft are respectively connected to the lifting trolley for rotation. The fixing device includes a fixing frame 6. A fixing fastener 7 is provided on the fixing frame 6. A fixing seat matching the fixing fastener 7 is provided on the lifting trolley. The fixing fastener 7 When connected with the fixing seat, the industrial camera, camera light source, and auxiliary light source face the direction of the overhead rail busbar, the operation control mainboard 4 is provided with a touch screen 8, the housing 1 is provided with a first avoidance hole at the position corresponding to the touch screen 8, a power display device 9 is provided on the power supply 3, the housing 1 is provided with a second avoidance hole at the position corresponding to the power display device 9, and the side of the housing 1 is provided with a plurality of ventilation holes 10, the power supply 3 is respectively connected to the industrial camera, camera light source, auxiliary light source, and operation control mainboard 4, the operation control mainboard 4 is respectively connected to the industrial camera, camera light source, and auxiliary light source, the operation control mainboard 4 and the server are located in the same local area network, the industrial camera includes a high-speed camera 11 and a digital camera 12, and the server includes a detection module, a request processing module, and an interaction module. The request processing module is used to receive information collected by the visual intelligent inspection terminal and send information processed by the detection module, the detection module is used to process information collected by the visual intelligent inspection terminal, and the interaction module is used to control the operation control mainboard 4.

[0044] The visual intelligent inspection device for defects of the overhead rail busbar of the lifting trolley in the present embodiment has a visual intelligent inspection terminal installed on the lifting trolley. Inspection operations can be carried out during the operation of the equipment, and the inspection cycle is flexible and maneuverable. The high-speed camera is set to capture the image information of the overhead rail busbar, and the digital camera captures the barcode image on the overhead rail busbar. The mainboard and server are controlled by calculation to determine the defect information and position information of the overhead rail busbar, and feedback is given without manual participation, thereby improving the inspection efficiency. The server continuously improves the inspection precision and accuracy of overhead rail busbar defects by establishing a detection network model.

[0045] The embodiment of the present invention further provides an inspection method of a visual intelligent inspection device for defects of a lifting trolley ceiling rail busbar, which is applied to the above-mentioned visual intelligent inspection device for defects of a lifting trolley ceiling rail busbar, and comprises the following steps:

[0046] S1. Install the visual intelligent inspection terminal on the lifting trolley and start the lifting trolley;

[0047] S2. The high-speed camera and digital camera of the visual intelligent inspection terminal collect image data of the overhead rail busbar, and the camera light source and auxiliary light source are used for illumination. The high-speed camera captures and collects image information of the overhead rail busbar, and the digital camera captures the barcode image on the overhead rail busbar, and sends the collected image data to the operation control mainboard;

[0048] S3, the computing control mainboard processes the barcode image information and the overhead rail busbar image information respectively, identifies the ID address of the overhead rail busbar and analyzes the defects of the overhead rail busbar. The defect analysis of the overhead rail busbar includes the following steps:

[0049] s31. Determine the connection of the overhead rail busbar: collect the connection and non-connection images in the overhead rail busbar image information as a data set, establish a classification model for the connection and non-connection, and make a judgment in the classification model;

[0050] s32. Establish a detection network model in the server. The detection network model adopts a neural network architecture with MobileNet as the main deep neural network backbone framework and an improved classification network layer. The detection network model is trained. The training steps include:

[0051] Firstly, the detection network model iteratively learns the defective images of the overhead rail busbar, fits the learned data set to the weight parameter dictionary, performs the prediction task through the weight parameter dictionary, and performs multi-category defect recognition and classification on the defective images of the overhead rail busbar.

[0052] Secondly, the detection network model connects a small network mainly composed of convolutional layers and Softmax layers to estimate the scores of the defects of the rail busbar under the defect category;

[0053] Secondly, multi-scale fusion is performed on the defects of the rail busbar, including fusion of shallow features and deep features;

[0054] Thirdly, the defective images of the overhead rail busbar in the data set are classified into difficulty and ease, including offline and online difficulty classification. The difficulty and ease classification of the offline method is to add annotations to the static labels of the data set before training, and add one-dimensional labels for training during the training process. It can also be to classify the data into difficulty and ease using the difference between the score value after the first training and the score of the label, and no longer update the difficulty score in the subsequent training stage; the difficulty and ease classification of the online method is to classify the training sample results of the current stage and use them for training in the next training stage. The processing of the samples after difficulty and ease classification: the classification result scores of simple samples have higher weights for loss value optimization, and the loss value optimization is given lower weights for difficult samples;

[0055] Finally, the defective images of the overhead rail busbar are processed before and after, including traditional algorithms such as white balance and image alignment. After the training is completed, the defective information of the overhead rail busbar is sent to the server, and the detection network model performs secondary analysis on the defective images of the overhead rail busbar.

[0056] S4. The operation control mainboard integrates the analyzed defect information and ID address information of the overhead rail busbar and sends them back to the server. The server classifies the integrated information, and then records the determined defect information of the overhead rail busbar and issues feedback.

[0057] In summary, compared with the prior art, the present application has the following advantages:

[0058] 1. The terminal is flexible to use and can be installed at any time. The inspection of overhead rail busbar defects is no longer affected by the operation of the equipment, and online inspection is realized, which greatly improves the convenience of maintenance personnel in carrying out inspection work;

[0059] 2. The visual intelligent inspection terminal can significantly improve the inspection precision and accuracy of overhead rail busbar defects. In addition, the intelligent terminal can also conduct autonomous deep learning through the server, covering various defect types, sizes, background changes, etc. that may occur in the future;

[0060] For example, in the server's detection network model training, the detection network model connection is mainly designed with convolutional layers and Softmax layers, which assists in verifying the accuracy of the front-end detection network. On the basis of simple discrete classification, the accuracy of the algorithm is improved by using score estimation. In the training stage, the value output by the score estimation network is used for back propagation and loss optimization, which can help improve the accuracy by 3-5%.

[0061] Multi-scale fusion uses multi-scale fusion at the feature level. This multi-scale process is implemented in the network framework and can be accelerated by batch processing by the graphics processor. At the same time, it does not increase redundant information, but integrates the feature information at multiple scales. This method of multi-scale fusion effectively improves the accuracy of the final training result without increasing the amount of calculation. The accuracy can be increased by 5-10%;

[0062] Online difficulty classification can smooth the training process, smooth the iteration of loss values, and smooth the update of training weights, effectively avoiding the risk of local optimal traps, gradient explosion, and gradient loss.

[0063] The purpose of the pre- and post-image processing algorithms is to regularize the images at the low-dimensional stage. The unified data enables the neural network to better fit the distribution of the data set, making the trained weights have higher confidence, robustness, and generalization capabilities.

[0064] 3. The inspection operation is highly efficient. No human intervention is required during the entire process. The lifting trolley carries the visual intelligent inspection terminal along the entire length of the overhead rail layout to complete the inspection. The time consumption is greatly reduced from the original 8 hours to only about 10 minutes.

[0065] 4. Change the traditional method of visual inspection to rely entirely on intelligent inspection terminals to automatically complete inspections, locate defects, issue alarms, record and save defect images, thereby ensuring the reliability of inspections to the greatest extent.

[0066] In addition, it should be noted that the other contents of the visual intelligent inspection device and inspection method for defects of the overhead rail busbar of the lifting trolley disclosed in the present invention can be referred to the prior art and will not be described in detail here.

[0067] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Therefore, any modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A method for visual inspection of defects in overhead rail busbars of a trolley, applied to a visual intelligent inspection device for defects in overhead rail busbars of a trolley, characterized in that: The visual intelligent inspection device for defects of the overhead rail busbar of the lifting trolley comprises a visual intelligent inspection terminal and a server, the visual intelligent inspection terminal comprises an industrial computer based on the Ubuntu system, the industrial computer comprises an industrial camera, a power supply, a camera light source, an auxiliary light source and an operation control mainboard, a rotating device and a fixing device are also provided on the industrial computer, one end of the rotating device is connected to the lifting trolley, one end of the industrial computer rotates relative to the lifting trolley through the rotating device, and the other end of the industrial computer is connected to the lifting trolley through the fixing device, the power supply is respectively connected to the industrial camera, the camera light source, the auxiliary light source, and the operation control mainboard, the operation control mainboard is respectively connected to the industrial camera, the camera light source, and the auxiliary light source, the operation control mainboard and the server are located in the same local area network, and the industrial camera comprises a high-speed camera and a digital camera; The inspection method comprises the following steps: S1. Install the visual intelligent inspection terminal on the lifting trolley and start the lifting trolley; S2, the high-speed camera and digital camera of the visual intelligent inspection terminal collect image data of the overhead rail busbar, and the camera light source and auxiliary light source are used for illumination. The high-speed camera captures and collects image information of the overhead rail busbar, and the digital camera captures the barcode image on the overhead rail busbar, and sends the collected image data to the operation control mainboard; S3, the computing control mainboard processes the barcode image information and the overhead rail busbar image information respectively, identifies the ID address of the overhead rail busbar, analyzes the defects of the overhead rail busbar, and sends the defect information of the overhead rail busbar to the server for secondary analysis; In step S3, the operation control mainboard analyzes the defects of the overhead rail busbar, including the following steps: s31. Determine the connection of the overhead rail busbar: collect the connection and non-connection images in the overhead rail busbar image information as a data set, establish a classification model for the connection and non-connection, and make a judgment in the classification model; s32. Establish a detection network model in the server and train the detection network model. The training steps include: first, the detection network model iteratively learns the defective pictures of the overhead rail busbar, fits the learned data set to the weight parameter dictionary, performs the prediction task through the weight parameter dictionary, and performs multi-category defect recognition and classification on the defective pictures of the overhead rail busbar; secondly, the scores of the defects of the rail busbar under the defect category are estimated; thirdly, the defects of the rail busbar are multi-scale fused, including the fusion of shallow features and deep features; thirdly, the defective pictures of the overhead rail busbar in the data set are classified into difficulty; finally, the defective pictures of the overhead rail busbar are processed before and after, and the processing includes white balance and image alignment algorithms; after the training is completed, the detection network model performs secondary analysis on the defective pictures of the overhead rail busbar; In step s32, the detection network model adopts a neural network architecture with MobileNet as the main deep neural network backbone framework, followed by an improved classification network layer; the detection network model is connected to a small network mainly composed of convolutional layers and Softmax layers to estimate the scores of the defects of the rail busbar under the defect category; the difficulty classification of the defect images of the overhead rail busbar in the data set includes offline and online difficulty classification, wherein the difficulty classification of the offline method is to add annotations to the static labels of the data set before training, add one-dimensional labels for training during the training process, or use the difference between the score value after the first training and the label score to classify the data, and no longer update the difficulty score in the subsequent training stage; the difficulty classification of the online method is to classify the training sample results of the current stage, and use them for training in the next training stage, and process the samples after difficulty classification: the classification result scores of simple samples have higher weights for loss value optimization, and the loss value optimization of difficult samples has lower weights; S4. The operation control mainboard integrates the analyzed defect information and ID address information of the overhead rail busbar and sends them back to the server. The server classifies the integrated information, and then records the determined defect information of the overhead rail busbar and issues feedback.

2. The visual inspection method for defects of the overhead rail busbar of the hoisting trolley according to claim 1 is characterized in that: The industrial computer includes a shell and a mounting plate, the power supply and operation control mainboard are arranged on one mounting surface of the mounting plate, the industrial camera, camera light source, auxiliary light source, rotating device and fixing device are arranged on another mounting surface, and the shell covers the power supply and operation control mainboard.

3. The method for visual inspection of defects of overhead rail busbars of a trolley according to claim 2 is characterized in that: The rotating device includes a rotating frame, on which a rotating shaft is provided, and the upper and lower ends of the rotating shaft are respectively rotatably connected to the lifting trolley. The fixing device includes a fixing frame, on which a fixing fastener is provided, and the lifting trolley is provided with a fixing seat that matches the fixing fastener.

4. The visual inspection method for defects of overhead rail busbars of a trolley according to claim 2 is characterized in that: The operation control mainboard is provided with a touch screen, the shell is provided with a first avoidance hole at a position corresponding to the touch screen, a power display device is provided on the power supply, and the shell is provided with a second avoidance hole at a position corresponding to the power display device.

5. The method for visual inspection of defects of overhead rail busbars of a trolley according to claim 2 is characterized in that: A plurality of ventilation holes are arranged on the side of the shell.

6. The method for visual inspection of defects of overhead rail busbars of a trolley according to any one of claims 1 to 5, characterized in that: When the other end of the industrial computer is connected to the lifting trolley, the industrial camera, camera light source and auxiliary light source are oriented toward the direction of the overhead rail busbar.

7. The method for visual inspection of defects of overhead rail busbars of a trolley according to any one of claims 1 to 5, characterized in that: The server includes a detection module, a request processing module and an interaction module. The request processing module is used to receive information collected by the visual intelligent inspection terminal and send information processed by the detection module. The detection module is used to process the information collected by the visual intelligent inspection terminal. The interaction module is used to control the operation control mainboard.

Images