Method and device for detecting broken filaments of filament spindle

Through the combination of multiple sets of cameras and lighting, multi-directional shooting and image splicing of fiber wire ingots is solved, and the problem of low detection accuracy of fiber wire ingots in the prior art is achieved, achieving more efficient defect recognition and detection effects.

CN118937344APending Publication Date: 2024-11-12HANGZHOU HUIZHILIAN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411100201.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

When existing visual recognition equipment detects fiber ingots, due to shooting angle and lighting problems, the image has dim or overexposed invalid areas, resulting in low accuracy of defect recognition.

Method used

Using multiple sets of cameras and lighting lamps, the wire ingots are photographed through multiple directions, and the captured images are cropped and spliced ​​to form a complete wire ingot image, identify defective parts, and improve detection accuracy.

Benefits of technology

Through the combination of multiple sets of lighting and cameras, the wire ingots on the detection station are surrounded, the image recognition rate and the accuracy of defect detection are improved, the interference of invalid areas is avoided, and the detection efficiency is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118937344A_ABST
    Figure CN118937344A_ABST
Patent Text Reader

Abstract

The invention discloses a filament ingot broken filament detection method and device, and the method comprises the five steps: S1, filament ingot input, S2, filament ingot shooting, S3, defect detection, S4, result output, and S5, filament ingot output. The device comprises a plurality of groups of illuminating lamps and a plurality of cameras, the number of the cameras is the same as that of the illuminating lamps, the center of a detection station serves as a base point, the illuminating lamps are evenly distributed in the circumferential direction of the base point, the number of each group of illuminating lamps is two, and the cameras are arranged between the two illuminating lamps. According to the invention, the filament spindle can be shot from multiple directions, the effective area of the image is cut and spliced, and the broken filament detection accuracy of the filament spindle is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of visual inspection, and in particular relates to a method and a device for detecting hair on a spindle. Background Art

[0002] During the production process of fiber ingots, the silk thread needs to be wound on the ingot paper tube. When the ingot is wound, its outer surface will be attached with defects such as hair. The defects on the surface of the ingot not only affect the appearance of the ingot, but also affect the grade of the ingot. Ingot detection generally relies on manual inspection or visual recognition equipment. However, existing visual recognition equipment generally uses a single camera for shooting and recognition. Due to problems such as shooting angle and lighting, the captured image will have some dark or overexposed invalid areas, resulting in low defect recognition accuracy. Summary of the invention

[0003] The purpose of the present invention is to solve the above-mentioned technical problems existing in the prior art, and to provide a method and a device for detecting silk spindle hair, which can photograph the silk spindle from multiple directions and crop and splice the effective area of ​​the image to improve the accuracy of silk spindle hair detection.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0005] The method for detecting hairy silk in a spindle is characterized by:

[0006] S1, ingot input

[0007] The ingot is arranged on a conveyor line and is input to the bottom of the ingot detection device through the conveyor line. When the ingot to be detected moves to the detection station, the movement of the conveyor line is stopped.

[0008] S2. Shooting of silk spindles

[0009] The silk ingot detection device is started and photographs are taken at various positions of the silk ingot in turn.

[0010] S3. Defect Detection

[0011] All the images taken are identified, and the invalid areas are cropped, and the valid areas are spliced ​​to form a complete ingot image. The spliced ​​images are then inspected to identify defective areas.

[0012] S4. Output results

[0013] Mark the defective areas on the ingot image.

[0014] S5, ingot output

[0015] The photographed silk ingots are marked and identified. When there are defective marks in the output results, the silk ingots are marked as defective products. The conveyor line is started after the silk ingots are inspected, and the silk ingots to be inspected at the rear are moved to the inspection location for the next round of inspection.

[0016] Furthermore, the detailed steps of step S2 include:

[0017] S21: The lighting lamps and cameras in the detection device are arranged in at least three groups according to a corresponding relationship, each group includes a camera and two lighting lamps, and the camera is arranged between the two lighting lamps;

[0018] S22: The cameras take pictures in sequence in a certain direction according to their positions on the silk ingot, with one of the cameras being the starting camera. When shooting, the lights of the corresponding group of cameras light up to fill in the light for the silk ingot. After shooting, the lights turn off.

[0019] Furthermore, in step S21, four groups of cameras and lighting lamps are used, and different cameras share some lighting lamps.

[0020] Furthermore, the detailed steps of step S22 include that when shooting, the lights of the corresponding light group are lit up in sequence, and the camera shoots multiple times to capture images when different lights are lit up and when all the corresponding light groups are lit up.

[0021] Furthermore, the detailed steps of step S3 are: identifying the photos taken by each group of cameras, cropping the effective areas therein, and then splicing them to form photos of the orientation, and then splicing the cropped and spliced ​​images of each orientation to form a photo of the complete silk ingot.

[0022] Furthermore, in step S3, the invalid area may be cropped first, and the valid area may be spliced ​​to form a complete silk ingot image, and then the spliced ​​image may be detected; or image detection may be performed first and then the invalid area may be cropped, and the valid area may be spliced ​​to form a complete silk ingot image.

[0023] The present invention also discloses a spindle hair detection device, which is characterized by comprising multiple groups of lighting lamps and multiple cameras, the number of cameras is the same as the lighting lamps, the center of the detection station is taken as the base point, the lighting lamps are evenly distributed in the circumferential direction of the base point, the number of lighting lamps in each group is two, and the camera is arranged between the two lighting lamps.

[0024] Furthermore, the number of groups of lighting lamps is four, and the two lighting lamps in each group are symmetrically arranged about the base point center.

[0025] Furthermore, the camera is arranged at the center of the boundary line of each group of lighting lamps.

[0026] Furthermore, the heights of the lighting lamp and the camera are both higher than the base point.

[0027] The present invention has the following beneficial effects due to the adoption of the above technical solution:

[0028] The present invention adopts five steps S1-S5, which can take pictures of the silk ingot from multiple directions, and can detect the effective area by lighting from different directions with corresponding lighting lamps, thereby improving the recognition rate of the image and the accuracy of defect detection. By cutting and splicing multiple groups of photographed images, it is possible to achieve clear and complete shooting of the entire silk ingot, avoid interference of invalid areas on defect detection, and improve detection efficiency.

[0029] The present invention, through the combination of multiple groups of lighting lamps and cameras, can surround the silk ingot on the detection station, photograph and image the silk ingot, detect the effective area of ​​the image, and improve the image recognition rate and the accuracy of defect detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present invention will be further described below in conjunction with the accompanying drawings:

[0031] Figure 1 This is a schematic diagram of the structure of the detection device of the present invention arranged on a conveyor line;

[0032] Figure 2 It is a schematic structural diagram of the spindle hair detection device of the present invention;

[0033] Figure 3 A top view of the present invention;

[0034] Figure 4 It is a structural schematic diagram of a left camera and an illumination lamp group matching the left camera in the present invention;

[0035] Figure 5 The image output by the left camera in the present invention;

[0036] Figure 6 It is a structural schematic diagram of a right camera and a lighting lamp group matching the right camera in the present invention;

[0037] Figure 7 The image output by the right camera in the present invention;

[0038] Figure 8 It is a structural schematic diagram of an upper camera and an illumination lamp group matching the upper camera in the present invention;

[0039] Fig. 9 The image output by the upper camera in the present invention;

[0040] Fig.10 It is a structural schematic diagram of the lower camera and the lighting lamp group matching with the lower camera in the present invention;

[0041] Fig.11 The image output by the lower camera in the present invention;

[0042] Fig.12 It is a complete silk ingot image formed after the effective areas in the present invention are spliced.

[0043] In the figure, 1- silk ingot; 2- camera; 3- lighting lamp; 4- invalid area; 5- valid area. DETAILED DESCRIPTION

[0044] like Figure 1 As shown, the method for detecting hairy silk in a spindle of the present invention comprises the following steps:

[0045] S1, ingot 1 input

[0046] The silk ingot 1 is arranged on a conveyor line and is input to the bottom of the silk ingot 1 detection device through the conveyor line. When the silk ingot 1 to be detected moves to the detection station, the movement of the conveyor line is stopped.

[0047] The silk ingots 1 are input in the form of a conveyor line, so that continuous detection of the silk ingots 1 can be achieved and detection efficiency can be improved.

[0048] S2, Spool 1 shooting

[0049] The ingot 1 detection device is started, and each position of the ingot 1 is photographed in sequence.

[0050] By shooting the silk ingot 1 at multiple positions through multiple groups of cameras 2, the effect of shooting images can be guaranteed.

[0051] S3. Defect Detection

[0052] All the captured images are identified, and the invalid area 4 is cropped, and the valid area 5 is spliced ​​to form a complete image of the silk ingot 1, and then the spliced ​​image is detected to identify the defective parts.

[0053] By identifying the image and cutting out the effective area 5 for splicing, the interference of the invalid area 4 on image detection can be reduced, and the accuracy of identifying defects such as hair can be improved.

[0054] S4. Output results

[0055] The defective portions of the ingot 1 image are marked.

[0056] Mark defective parts to facilitate the review of subsequent processes.

[0057] S5, ingot 1 output

[0058] The photographed silk ingot 1 is marked and identified. When a defect mark exists in the output result, the silk ingot 1 is marked as a defective product. The conveyor line is started after the silk ingot 1 is inspected, and the silk ingot 1 to be inspected behind is moved to the inspection location for the next round of inspection.

[0059] Mark and identify the silk ingot 1 to facilitate the removal of defective products. At the same time, when the silk ingot 1 is inspected, the next silk ingot 1 is sent for inspection, and the assembly line operation is carried out to improve the inspection efficiency.

[0060] The detailed steps of step S2 include:

[0061] S21 : The lighting lamps 3 and cameras 2 in the detection device are arranged into at least three groups according to a corresponding relationship, each group includes one camera 2 and two lighting lamps 3 , and the camera 2 is arranged between the two lighting lamps 3 .

[0062] S22: The cameras 2 take pictures in sequence in a certain direction according to their positions on the silk ingot 1, with one of the cameras 2 being the starting camera 2. When shooting, the lighting lamps 3 of the corresponding group of the cameras 2 light up to supplement the light for the silk ingot 1. After shooting, the lighting lamps 3 turn off.

[0063] In step S2 , the silk ingot 1 is photographed and imaged by using cameras 2 in multiple directions and the light groups corresponding to the cameras 2 , so that different images of the effective area 5 can be obtained. By cropping and splicing, the complete silk ingot 1 image can be spliced ​​from the effective area 5 .

[0064] In step S21 , four groups of cameras 2 and lighting lamps 3 are used, and different cameras 2 share some lighting lamps 3 .

[0065] According to the comparison between the number of cameras 2 and lighting lamps 3, four groups of cameras 2 and lighting lamps 3 are used, which can reduce the number of cameras 2 and lighting lamps 3 and save costs while ensuring the quality of captured images.

[0066] The detailed steps of step S22 include: when shooting, the lights of the corresponding light group are turned on in sequence, and the camera 2 shoots multiple times, shooting images when different lights are turned on and when all the corresponding light groups are turned on. When shooting, the camera 2 in a single orientation can increase the effective area 5 of the orientation image by shooting the images of the silk ingot 1 at different illumination angles, and can improve the image quality of the silk ingot 1 after splicing when the images are spliced, which is convenient for the subsequent identification of hair defects.

[0067] The detailed steps of step S3 are: identifying the photos taken by each group of cameras 2, cropping the effective area 5 therein, and then splicing them to form a photo of the orientation, and then splicing the cropped and spliced ​​images of each orientation to form a photo of the complete silk ingot 1.

[0068] In step S3, the invalid area 4 can be cropped first, and the valid area 5 can be spliced ​​to form a complete image of the silk ingot 1, and then the spliced ​​image can be detected; or image detection can be performed first and then the invalid area 4 can be cropped, and the valid area 5 can be spliced ​​to form a complete image of the silk ingot 1. Cropping the invalid area 4 first and then splicing can save computing power and improve recognition efficiency, while splicing first and then cropping can also ensure the final imaging, which will appropriately waste some computing power, but it will not make any difference to the final imaging of the silk ingot 1.

[0069] The present invention adopts five steps S1-S5, and can take pictures of the silk ingot 1 from multiple directions, and can detect the effective area 5 by illuminating from different directions with the corresponding lighting lamp 3, thereby improving the image recognition rate and the accuracy of defect detection. By cutting and splicing multiple groups of photographed images, it is possible to achieve clear and complete shooting of the entire silk ingot 1, avoid interference of the invalid area 4 on the detection of defects, and improve the detection efficiency.

[0070] like Figures 2 to 12 As shown, the present invention also discloses a device for detecting hairy wires in a silk ingot 1, comprising a plurality of groups of lighting lamps 3 and a plurality of cameras 2, wherein the number of cameras 2 is the same as that of lighting lamps 3, the center of the detection station is taken as the base point, the lighting lamps 3 are evenly distributed in the circumferential direction of the base point, the number of lighting lamps 3 in each group is two, and the camera 2 is arranged between the two lighting lamps 3. Through the above arrangement, it is possible to take photos and images of multiple positions of the silk ingot 1, detect the effective area 5 of the image, and improve the recognition rate of the image and the accuracy of defect detection.

[0071] As an implementation method, there are four groups of lighting lamps 3, and the two lighting lamps 3 in each group are symmetrically arranged about the base point center. Four groups of cameras 2 and lighting lamps 3 are used, which can reduce the number of cameras 2 and lighting lamps 3 while ensuring the quality of the captured image, thereby saving costs.

[0072] Specifically, the camera 2 is arranged at the center of the boundary line of each group of lighting lamps 3. Through the above arrangement, when the camera 2 is shooting, the illumination intensity on the left and right sides tends to be consistent, improving the imaging effect and facilitating the subsequent defect recognition of the image.

[0073] As an implementation method, the heights of the lighting lamp 3 and the camera 2 are both higher than the base point. Through the above arrangement, it is convenient to photograph the top surface of the silk ingot 1, ensure the imaging effect of the silk ingot 1 image, and improve the accuracy of defect recognition of the silk ingot 1.

[0074] The present invention, through the combination of multiple groups of lighting lamps 3 and cameras 2, can surround the silk ingot 1 on the detection station, photograph and image the silk ingot 1, detect the effective area 5 of the image, and improve the image recognition rate and the accuracy of defect detection.

[0075] The above are only specific embodiments of the present invention, but the technical features of the present invention are not limited thereto. Any simple changes, equivalent replacements or modifications made based on the present invention to solve basically the same technical problems and achieve basically the same technical effects are all included in the protection scope of the present invention.

Claims

1. A method for detecting hairy silk in a silk spindle, characterized in that The steps include: S1, ingot input The ingot is arranged on a conveyor line and is input to the bottom of the ingot detection device through the conveyor line. When the ingot to be detected moves to the detection station, the movement of the conveyor line is stopped; S2. Shooting of silk spindles The silk ingot detection device is started, and each position of the silk ingot is photographed in turn; S3. Defect Detection Identify all the images taken, crop the invalid areas, splice the valid areas to form a complete ingot image, and then detect the spliced ​​images to identify the defective parts; S4. Output results Marking defective parts on the ingot image; S5, ingot output The photographed silk ingots are marked and identified. When there are defective marks in the output results, the silk ingots are marked as defective products. The conveyor line is started after the silk ingots are inspected, and the silk ingots to be inspected at the rear are moved to the inspection location for the next round of inspection.

2. The method for detecting hairline in a spindle according to claim 1, characterized in that: The detailed steps of step S2 include: S21: The lighting lamps and cameras in the detection device are arranged in at least three groups according to a corresponding relationship, each group includes a camera and two lighting lamps, and the camera is arranged between the two lighting lamps; S22: The cameras take pictures in sequence in a certain direction according to their positions on the silk ingot, with one of the cameras being the starting camera. When shooting, the lights of the corresponding group of cameras light up to fill in the light for the silk ingot. After shooting, the lights turn off.

3. The method for detecting hairline in a spindle according to claim 2, characterized in that: In step S21, four groups of cameras and lighting lamps are used, and different cameras share some lighting lamps.

4. The method for detecting hairline in a spindle according to claim 2, characterized in that: The detailed steps of step S22 include: when shooting, the lights of the corresponding light group are lit up in sequence, and the camera shoots multiple times to capture images when different lights are lit up and when all the corresponding light groups are lit up.

5. The method for detecting hairline in a spindle according to claim 1, characterized in that: The detailed steps of step S3 are: identifying the photos taken by each group of cameras, cropping the effective areas therein, and then splicing them to form photos of the orientation, and then splicing the cropped and spliced ​​images of each orientation to form a photo of the complete silk ingot.

6. The method for detecting hairline hair in a spindle according to claim 5, characterized in that: In step S3, the invalid area may be cropped first, and the valid area may be spliced ​​to form a complete silk ingot image, and then the spliced ​​image may be detected; or image detection may be performed first and then the invalid area may be cropped, and the valid area may be spliced ​​to form a complete silk ingot image.

7. A device for detecting hairy silk in a spindle, characterized in that: It includes multiple groups of lighting lamps and multiple cameras, the number of the cameras is the same as the lighting lamps, the center of the detection station is used as the base point, the lighting lamps are evenly distributed in the circumferential direction of the base point, the number of the lighting lamps in each group is two, and the camera is arranged between the two lighting lamps.

8. The spindle hair detection device according to claim 7, characterized in that: The number of groups of the lighting lamps is four, and the two lighting lamps in each group are symmetrically arranged about the base point center.

9. The spindle hair detection device according to claim 8, characterized in that: The camera is arranged at the center of a line connecting the boundaries of each group of lighting lamps.

10. The spindle hair detection device according to claim 7, characterized in that: The heights of the lighting lamp and the camera are both higher than the base point.