Part sorting error detection device and system
The component identification equipment using a neural network model has solved the problem of identifying errors in component sorting, enabling automated detection and correction, and improving production efficiency and product quality.
Patent Information
- Application Number
- CN202422888452.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-26
- Publication Date
- 2025-11-18
- Estimated Expiration
- 2034-11-26
AI Technical Summary
In the automobile manufacturing process, errors are prone to occur after parts are sorted due to their high similarity, which can affect subsequent production processes. Existing technologies cannot effectively and automatically identify and correct sorting errors.
The component identification device, which employs a component identification model composed of neural networks, combined with image acquisition equipment and machine vision controller, along with placement equipment and identification result display equipment, enables automatic detection and correction of component sorting errors.
It improved the accuracy of parts sorting, reduced errors in the production process, and improved production efficiency and product quality.
Smart Images
Figure CN223556588U_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The utility model relates to the field of spare parts sorting, especially to a spare parts sorting error detection device and system. BACKGROUND
[0002] In the automobile production manufacturing process, the spare parts are transported to the production line after sorting, and due to the high similarity of some spare parts, errors are prone to occur in the sorting process, which affects the subsequent production and manufacturing process, and how to automatically identify the sorting error is a problem to be solved. SUMMARY
[0003] In order to solve the above problems, the utility model provides a spare parts sorting error detection device, comprising:
[0004] The spare part recognition device comprises an image acquisition device and a machine vision controller, the image acquisition device is arranged in a shell, the machine vision controller is arranged in the shell and connected with the image acquisition device, the machine vision controller comprises a spare part recognition model composed of a neural network, and the spare part recognition model is used to identify whether the spare part has sorting error according to the spare part image collected by the image acquisition device.
[0005] The spare part placing device comprises a placing rack and a spare part fixing assembly, the placing rack is used to place spare parts of multiple sizes, and the fixing assembly is used to fix spare parts of multiple sizes.
[0006] The recognition result display device is used to display the spare part sorting recognition result.
[0007] In an embodiment of the utility model spare parts sorting error detection device, the placing rack has a plurality of placing areas for placing the spare parts, and the spare part fixing assembly is used to fix the spare parts in the placing area and the spare parts located on the side of the placing rack.
[0008] In an embodiment of the utility model spare parts sorting error detection device, the placing area is provided with an area label, and the spare part recognition model is used to identify whether the spare parts in the placing area correspond to the area label.
[0009] In an embodiment of the utility model spare parts sorting error detection device, the spare part fixing assembly further comprises a first limiting fixing part located in the placing area and a second limiting fixing part located on the side of the placing rack.
[0010] In an embodiment of the part sorting error detection device, the part placing device further comprises an actuating controller and a driving device, the driving device is arranged at the bottom of the placing rack, and the actuating controller is used to receive instructions of the machine vision controller to control the driving device to displace the placing rack.
[0011] In an embodiment of the part sorting error detection device, a displacement unit is further included, the displacement unit is connected with the image acquisition device of the part recognition device and the machine vision controller, and is used to displace the image acquisition device and / or the machine vision controller according to instructions of the machine vision controller to acquire the part image from different positions.
[0012] In an embodiment of the part sorting error detection device, an alarm unit is further included, and the alarm unit alarms according to instructions of the machine vision recognition controller when the part recognition model identifies that the part sorting error occurs.
[0013] In an embodiment of the part sorting error detection device, a storage unit is further included, and is used to store error information when the part recognition model identifies that the part sorting error occurs.
[0014] In an embodiment of the part sorting error detection device, the recognition result display device is one or more of a smart phone, a wearable smart device, a personal computer and a workstation.
[0015] The utility model also provides a kind of part sorting error detection system, including automatic sorting device and part transport equipment, the automatic sorting device is used to sort the part according to vehicle model, the part transport equipment is used to transport the part to detection position, further include at least one as described in any one of the part sorting error detection device.
[0016] Through the part sorting error detection device and system provided by the utility model, the part recognition device including the part recognition model composed of neural network is used to detect the error of part sorting in cooperation with part placing device and recognition result display device, so as to solve the problem that current part sorting error cannot be effectively and automatically recognized. BRIEF DESCRIPTION OF DRAWINGS
[0017] Figure 1 The schematic diagram of the part sorting error detection device in an embodiment of the utility model is shown.
[0018] Figure 2 The schematic diagram of the part placing device in an embodiment of the utility model is shown.
[0019] Figure 3 Fig. 1 illustrates a schematic diagram of a part placing device in an embodiment of the present application.
[0020] Figure 4 Fig. 2 illustrates a schematic diagram of a part sorting error detection device in another embodiment of the present application.
[0021] Figure 5 Fig. 3 illustrates a schematic diagram of a part sorting error detection system in an embodiment of the present application.
[0022] In the drawings, reference numerals:
[0023] 1… part sorting error detection device
[0024] 2… automatic sorting device
[0025] 3, 3a, 3b… part
[0026] 4… part transport device
[0027] 10… part identification device
[0028] 11… image acquisition device
[0029] 12… machine vision controller
[0030] 13… displacement unit
[0031] 14… alarm unit
[0032] 15… storage unit
[0033] 20… housing
[0034] 30… part placing device
[0035] 31… placing rack
[0036] 32… part fixing assembly
[0037] 321… first limiting fixing member
[0038] 322… second limiting fixing member
[0039] 33… placing area
[0040] 40… identification result display device
[0041] 100… part sorting error detection system DETAILED DESCRIPTION
[0042] The technical scheme of the utility model will be described in detail below with reference to the drawings and specific embodiments, so as to further understand the purpose, scheme and beneficial technical effects of the utility model. Obviously, the specific embodiments described in the utility model are only a part of the embodiments of the utility model, rather than all the embodiments. Based on the embodiments disclosed in the utility model, all other embodiments obtained by those skilled in the art without creative labor also belong to the scope disclosed by the technical scheme of the utility model.
[0043] It should be noted that in the specification, relational terms such as first and second and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying that there is any such actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof are intended to cover non-exclusive inclusions, so that a process, article, or apparatus including a series of elements includes not only those elements but also other elements not explicitly listed or inherent to such process, article, or apparatus. Without more limitations, the element defined by the phrase "comprising a" does not exclude the presence of additional identical elements in the process, method, article, or apparatus including the element.
[0044] In the specification and the claims appended thereto, certain words are used to refer to particular components or parts, and it will be understood by those skilled in the art that the same component or part can be referred to by different names or terms by the user or manufacturer. The specification and the claims appended thereto do not distinguish components or parts by name, but by functional differences.
[0045] In the utility model, the orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", "front", "rear", "top", "bottom", "inner", "outer", "middle", "vertical", "horizontal", "transverse", "longitudinal" and the like is based on the orientation or positional relationship shown in the drawings. These terms are mainly used to better describe the utility model and its embodiments, and are not used to limit the indicated devices, elements or components to have a specific orientation, or to be constructed and operated in a specific orientation.
[0046] In addition, the terms "mount", "set", "provided with", "connect", "connect" should be broadly understood. For example, it can be fixedly connected, detachably connected, or integrally constructed; it can be mechanically connected, or electrically connected; it can be directly connected, or indirectly connected through an intermediate medium, or internal communication between two devices, elements or components. For those skilled in the art, the specific meaning of the above terms in the utility model can be understood according to the specific circumstances.
[0047] In order to better understand the technical scheme of the utility model, reference can be made to Figure 1 , in order to solve the problem that sorting error is prone to occur in the process of sorting parts, the utility model provides a kind of sorting error detection device 1 of parts, comprising: parts identification equipment 10, parts placement equipment 30 and identification result display equipment 40;Parts identification equipment 10 includes image acquisition equipment 11 and machine vision controller 12, image acquisition equipment 11 is set to a shell 20;Machine vision controller 12 is set to shell 20 and is connected with image acquisition equipment 11, machine vision controller 12 includes the parts identification model of neural network composition, and the parts identification model is used to identify whether there is sorting error according to the parts image collected by image acquisition equipment 11 Whether the parts 3 exist sorting error;Parts placement equipment 30 includes placing rack 31 and parts fixing assembly 32, placing rack 31 is used to place parts 3 of multiple sizes, and fixing assembly is used to fix parts 3 of multiple sizes;Identification result display equipment 40 is used to show parts sorting identification result.In an embodiment, image acquisition equipment 11 can be set to the outside or inside of shell 20, as long as the image of parts 3 can be collected, the utility model is not limited to this.In an embodiment, the parts identification model of neural network composition can be yolo5 model, those skilled in the art should know that the parts identification model can also be other machine vision identification model, the utility model is not limited to this.When sorting parts (including manual sorting or machine automatic sorting), for example, parts A1 and B1 belonging to two models of vehicles are similar, when sorting correctly, parts A1, A2 and A3 of A vehicle should be placed in placing rack 31, and when sorting error, parts B1, A2 and A3 of A vehicle are placed in placing rack 31.Because the parts identification model is trained, the images of all different angles of parts of one model of vehicle are associated with the model of vehicle, so when sorting error, the sorting error can be identified.
[0048] Please refer to Figures 2-3 , in an embodiment, placing rack 31 has a plurality of placing areas 33 for placing parts, and parts fixing assembly 32 is used to fix parts 3 in placing area 33 and parts 3 located on the side of placing rack 31.Parts 3b are placed in a placing area 33, and parts 3a are placed on the side of placing rack 31, which facilitates parts identification equipment 10 to sort and identify them.Further, the plurality of placing areas 33 in placing rack 31 can be adjusted according to the size of parts.The placing mode can utilize the space inside and on the side of placing rack 31 at the same time, which facilitates the identification of different faces of parts of different sizes.
[0049] In an embodiment, the placement area 33 is provided with an area label, and the spare part recognition model is used to identify whether the spare part in the placement area 33 corresponds to the area label. Through training of the spare part recognition model, a relationship between a spare part image and an area number is established, and it is identified that the spare part placement position is incorrect, which can prevent the delivery of an incorrect type of spare part to a production area, and improve production efficiency.
[0050] In an embodiment, the spare part fixing assembly 32 further includes a first limiting fixing member 321 located at the placement area 33 and a second limiting fixing member 322 located at the side of the placement rack 31. The first limiting fixing member 321 of the placement area 33 can be an elastic band, a buckle connection fixing band, a pocket type fixing member, a grid type fixing member, and the like, and one or more of them can be provided on the placement rack 31 according to requirements. The second limiting fixing member 322 of the side of the placement rack 31 can be a hook type fixing member, a clamping type fixing member, a pasting type fixing member, and the like, and one or more of them can be provided on the side of the placement rack 31 according to requirements.
[0051] In an embodiment, the spare part placement device 30 further includes an actuating controller (not shown in the figure) and a driving device, the driving device is arranged at the bottom of the placement rack 31, and the actuating controller is used to receive instructions of the machine vision controller 12 to control the driving device to displace the placement rack 31. The driving device can be a conventional motor, a stepping motor, a servo motor, a hydraulic driving motor, a pneumatic driving motor, and the like, but is not limited thereto. In an embodiment, when the machine vision controller 12 detects the spare part on the placement rack 31 at the current position, the actuating controller is notified to rotate the placement rack 31 to detect other surfaces of the spare part. Through displacement of the driving device, the placement rack 31 is rotated to sort and detect errors of multiple surfaces of the spare part, and identification of sorting errors of the spare part is more accurate.
[0052] Please refer to Figure 4 , in Figure 4In the figure, the shell 20 is omitted for clarity to show the relationship between the devices. In an embodiment, the displacement unit 13 is further included, which is connected with the image acquisition device 11 and the machine vision controller 12 of the spare part recognition device 10, and is used to displace the image acquisition device 11 and / or the machine vision controller 12 according to the instruction of the machine vision controller 12 to acquire spare part images from different positions. In an embodiment, the displacement unit 13 can be a track displacement unit 13, an electromagnetic position device, a gear position device, etc., without being limited thereto. In addition, the displacement unit 13 can directly drive the shell 20 to move, or can be separately located outside the shell 20 to move the image acquisition device 11, as long as it can make the image acquisition device 11 acquire spare part images from different positions, without being limited thereto. Through the displacement unit 13, spare part images can be acquired from different positions, and the recognition of spare part sorting errors is more accurate.
[0053] In an embodiment, the alarm unit 14 is further included, which is used to alarm according to the instruction of the machine vision recognition controller when the spare part recognition model recognizes that the spare part has sorting errors. In an embodiment, the alarm unit 14 can use a loudspeaker, an indicator light, send a notification on an electronic device, etc., without being limited thereto.
[0054] In an embodiment, the storage unit 15 is further included, which is used to store error information when the spare part recognition model recognizes that the spare part has sorting errors. The storage unit 15 can be arranged in the spare part recognition device 10, or can be separately arranged outside the spare part recognition device 10.
[0055] In an embodiment, the recognition result display device 40 is one or more of a smart phone, a wearable smart device, a personal computer, and a workstation.
[0056] The working process of the part sorting error detection device 1 in an embodiment is described below. First, a part recognition model based on yolo5 is trained. The training process is as follows: a plurality of images of parts are collected, the vehicle model corresponding to the part images is labeled using a labeling tool, and then the labeled data set is used to train the part recognition model. Next, the trained part recognition model is used for part sorting error detection. The detection process is as follows: after automatic sorting or manual sorting, the parts are placed in a part placing device, the part conveying device transports the part placing device to a detection position, the part recognition device collects images of the parts on the part conveying device, and the part recognition model in the part recognition device recognizes the collected part images. If it is found that the parts on the part conveying device do not all belong to the same vehicle, an alarm is triggered, and the parts with sorting errors are displayed. In an embodiment, when the sorting error rate of a part exceeds a threshold value within a period of time, an escalation alarm is triggered, and then personnel training or inspection is performed on the sorting of the part, and the automatic sorting device is adjusted for the sorting of the part to reduce the sorting error rate of the part.
[0057] The utility model also provides a part sorting error detection system 100, including automatic sorting device 2 and part conveying device 4, automatic sorting device 2 is used to sort parts according to vehicle model, and part conveying device 4 is used to transport parts to a detection position, and further comprising at least one part sorting error detection device 1 as any one of the above. In an embodiment, a part recognition device 10 can also be provided at the automatic sorting device 2 to perform a first detection of whether the sorting is correct at a first time, and then perform a second detection at the detection position, further improving the accuracy of the sorting error and improving the production efficiency.
[0058] The part sorting error detection device and system provided by the utility model solve the problem that the current part sorting error cannot be effectively and automatically recognized.
[0059] The above disclosure is only a preferred and feasible embodiment of the utility model, and does not limit the patent application range of the utility model, so any equivalent technical changes made by applying the contents of the utility model specification and drawings fall within the patent application range of the utility model.
Claims
1. A parts mis-sorting error detection apparatus, characterized by, The application relates to a part sorting error detection device. The part sorting error detection device comprises a part recognition device, a part placing device, an alarm unit, a storage unit, a displacement unit and an identification result display device. The part recognition device comprises an image acquisition device and a machine vision controller. The machine vision controller is arranged in the shell and connected with the image acquisition device. The machine vision controller comprises a part recognition model composed of a neural network.
2. The component mis-sort detection apparatus of claim 1, wherein The part recognition model is used to identify whether a sorting error exists in a part according to a part image acquired by the image acquisition device.
3. The part mis-sort detection apparatus of claim 2, wherein, The part placing device comprises a placing rack and a part fixing assembly.
4. The component mis-sort detection apparatus of claim 2 or 3, wherein The placing rack is used to place parts of multiple sizes.
5. The part mis-sort detection apparatus of claim 4, wherein, The part fixing assembly is used to fix the parts in the placing rack and on the side of the placing rack.
6. The apparatus of claim 5, wherein The placing rack has multiple placing areas for placing the parts.
7. The part mis-sort detection apparatus of claim 6, wherein, The part recognition model is used to identify whether the parts in the placing areas correspond to area labels.
8. The part mis-sort detection apparatus of claim 7, wherein, The part fixing assembly further comprises a first limiting fixing member in the placing areas and a second limiting fixing member on the side of the placing rack.
9. The part mis-sort detection apparatus of claim 8, wherein, The part placing device further comprises an actuating controller and a driving device.
10. A parts sorting error detection system comprising an automatic sorting device for sorting parts according to vehicle model and a parts transport device for transporting the parts to a detection position, characterized in that, The driving device is arranged at the bottom of the placing rack. The actuating controller is used to receive instructions from the machine vision controller to control the driving device to displace the placing rack. The displacement unit is connected with the image acquisition device and the machine vision controller of the part recognition device. The displacement unit is used to displace the image acquisition device and / or the machine vision controller according to instructions from the machine vision controller to acquire part images from different positions. The alarm unit is used to alarm according to instructions from the machine vision controller when the part recognition model identifies that a sorting error exists in a part. The storage unit is used to store error information when the part recognition model identifies that a sorting error exists in a part. The identification result display device is one or more of a smartphone, a wearable smart device, a personal computer and a workstation. The part sorting error detection device comprises at least one part sorting error detection device as claimed in any one of claims 1 to 9.