Comb production quality monitoring system based on Internet big data

Abstract

The invention relates to the technical field of comb production, in particular to a comb production quality monitoring system based on internet big data, which comprises a big data module, an image acquisition module, a defect analysis module and a traceability module. The method comprises the following steps: binding a comb through a comb code, acquiring a detection image of a plane of the comb through an image acquisition module, identifying a defect position in a comb area in the detection image through a defect analysis module according to the detection image, and tracing a defective comb by utilizing a tracing module. According to the method, source information of corresponding comb production lines is found out, errors and omissions of all production links of the production lines are monitored and judged by collecting the production lines of all comb sources according to the source information, then the production lines are reversely monitored by monitoring comb quality, and therefore the quality of follow-up shipment products is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of comb production, and more specifically, to a comb production quality monitoring system based on Internet big data. Background Art

[0002] The air cushion comb is a new type of comb for combing hair. It is very popular in the market due to the comfort of its air cushion.

[0003] The air cushion comb mainly consists of a comb handle, an air cushion and comb teeth. The general production process of the air cushion comb includes making the comb handle through injection molding, assembling the comb teeth into the air cushion, and installing the semi-assembled parts completed by the air cushion and comb teeth into the manufactured comb handle, thereby completing the entire air cushion comb production and assembly.

[0004] In the entire production line of air cushion combs, the quality of the air cushion combs produced is difficult to control due to the control differences of production equipment and the long-term use of production equipment. At present, most of the air cushion combs produced still use offline testing to screen their quality manually or by equipment to ensure the quality of shipped products. However, the problems of errors and omissions in production equipment still cannot be truly improved and solved. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide a comb production quality monitoring system based on Internet big data, which has the advantages of...

[0006] The above technical objectives of the present invention are achieved through the following technical solutions: a comb production quality monitoring system based on Internet big data, comprising:

[0007] The big data module obtains the source information of each comb produced in each production line, and performs comb coding on each comb, and obtains the source information of the corresponding comb through the comb coding;

[0008] An image acquisition module is used to acquire detection images of several planes of the comb, identify the comb area through edge positioning based on the detection images of each plane of the comb, and acquire the comb code;

[0009] The defect analysis module performs image processing according to the comb area of ​​each plane, identifies the defect position in the comb area, and determines the quality grade of the comb based on the area size and degree of the defect position. The quality grade is: excellent, good, and poor;

[0010] The traceability module is based on the quality grade classification of combs. For combs with good and poor quality grades, the module obtains the source information of the comb production line based on the comb coding information obtained by the image acquisition module. Based on the source information, the module collects the production lines from which each comb comes, thereby monitoring and judging the errors and omissions in each production link of the production line.

[0011] Preferably, the source information includes production process information of injection molding of the comb handle, assembly production process information of the air cushion and comb teeth, and assembly production process information of the air cushion and the comb handle. By binding each production equipment and assembly equipment with the corresponding production process information, a data information network is constructed, stored through a big data module, and bundled based on the comb code identification.

[0012] Preferably, the image acquisition module is based on the detection images of each plane of the comb, and pre-processes the detection images to distinguish the handle area, air cushion area and comb tooth area in the comb area according to the comb handle edge data information, air cushion edge data information and comb tooth edge data information.

[0013] Preferably, based on the detection image acquired by the image acquisition module, the step of implementing detection image preprocessing includes the following methods:

[0014] S1: Convert the acquired detection image into a grayscale detection image in advance;

[0015] S2: Perform image noise reduction processing on the grayscale detection image;

[0016] S3: Obtain the grayscale value of each pixel of the grayscale detection image, construct a grayscale histogram, and perform binarization processing, perform tilt correction on the grayscale detection image after binarization processing, and complete the detection image preprocessing.

[0017] Preferably, the method for the defect analysis module to identify the defect location in the comb area is:

[0018] For the comb handle area and the air cushion area, the defect edge information in the comb handle area and the air cushion area is extracted through secondary edge detection;

[0019] For the comb tooth area, the defect shape information is extracted through shape features;

[0020] Based on the standard comparison template image constructed by the big data module, the acquired defect edge information and defect shape information are compared with the local features corresponding to the standard comparison template image to determine the correctness of the acquired defect edge information and defect shape information.

[0021] Preferably, after obtaining the verified defect edge information and defect shape information, the specific position and shape of the defect on the defect edge information and defect shape information are re-extracted through three edge detections;

[0022] According to the specific location and shape of the defect, the area size of the defect is calculated, and the actual number of defects is counted. By setting the defect number threshold and the defect area threshold, and in the set range of the defect number threshold and the defect area threshold, the quality grade of the comb is divided into three types: excellent, good and poor.

[0023] Preferably, the combs are divided into three types: excellent, good and poor based on their quality grade, and the good and poor types of combs are counted. They are identified and traced through a traceability module, and the specific number of good and poor types of combs produced on each production line is counted.

[0024] Preferably, based on statistics on the specific numbers of good and poor combs produced on each production line, the defect positions on each good and poor comb are matched, and based on the defect positions, specific processing errors and omissions in the production links on each production line can be quickly identified.

[0025] Preferably, the image acquisition module is also provided with a correction unit, and among the two types of good and bad combs statistically, the combs that are incorrectly identified are re-recorded, and the image acquisition module re-identifies and acquires the combs that are incorrectly identified, and based on the information encoded by the comb, the combs that are incorrectly identified are defaulted to be defect-free, and the detection images obtained by the image acquisition module twice before and after are stored, and the detection images are analyzed by the defect analysis module again, and determined to be suspected defects based on the defect position analysis of the defect analysis module, and the analysis data information of the suspected defects are retransmitted to the big data module to build a deep learning network.

[0026] In summary, the present invention has the beneficial effects: the comb production quality monitoring system based on Internet big data includes a big data module, an image acquisition module, a defect analysis module and a traceability module. First, the source information of each comb produced in each production line is obtained through the big data module, and each comb is comb-coded. The source information of the comb is bound through the comb code, and then the plane of the comb is acquired through the image acquisition module. The detection image is used to perform an edge positioning to identify the comb area and obtain the comb code. After that, the defect analysis module performs image processing according to the comb area of ​​each plane, and identifies the defect position in the comb area. Based on the area size and degree of the defect position, the quality grade of the comb is judged. Finally, the traceability module is used to trace the combs with good and poor quality grades to find the source information of the corresponding comb production line. According to the source information, the production lines from which each comb comes are aggregated to monitor and judge the errors and omissions of each production link of the production line, and then the production line is reversely monitored by monitoring the quality of the comb, so as to ensure the quality of subsequent shipped products. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a schematic diagram of electrical connections between modules in an embodiment of the present invention;

[0028] Figure 2 is a schematic diagram of edge positioning of a detection image according to an embodiment of the present invention;

[0029] Figure 3 It is a schematic diagram of the coding mark of the production line of an embodiment of the present invention.

[0030] Figure numerals: 1. Big data module; 11. Image acquisition module; 12. Defect analysis module; 13. Tracing module. DETAILED DESCRIPTION

[0031] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0032] It should be noted that when a component is referred to as being "fixed to" or "disposed on" another component, it can be directly on the other component or indirectly on the other component. When a component is referred to as being "connected to" another component, it can be directly or indirectly connected to the other component.

[0033] It should be understood that the orientation or position relationship indicated by terms such as "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0034] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0035] Comb production quality monitoring system based on Internet big data, see Figure 1 ,include

[0036] Big data module 1 obtains the source information of each comb produced in each production line, and performs comb coding on each comb. The source information of the corresponding comb can be obtained through the comb coding;

[0037] An image acquisition module 11 is used to acquire detection images of several planes of the comb, identify the comb area through edge positioning based on the detection images of each plane of the comb, and acquire the comb code;

[0038] The defect analysis module 12 performs image processing according to the comb area of ​​each plane, identifies the defect position in the comb area, and determines the quality level of the comb based on the area size and degree of the defect position, wherein the quality level is: excellent, good, and poor;

[0039] The traceability module 13 is based on the quality grade classification of the combs. For combs with good and poor quality grades, the source information of the comb production line is obtained according to the comb coding information obtained by the image acquisition module 11. Based on the source information, the production lines where each comb comes from are aggregated to monitor and determine the errors and omissions in each production link of the production line.

[0040] The comb production quality monitoring system based on Internet big data in this embodiment includes a big data module 1, an image acquisition module 11, a defect analysis module 12 and a traceability module 13. First, the source information of each comb produced in each production line is obtained through the big data module 1, and each comb is comb-coded. The source information of the comb is bound through the comb code, and then the image acquisition module 11 is used to obtain a detection image of the plane of the comb. The detection image is used to perform an edge positioning to identify the comb area and obtain the comb code. After that, the defect analysis module 12 performs image processing according to the comb area of ​​each plane, and identifies the defect position in the comb area. Based on the area size and degree of the defect position, the quality grade of the comb is judged. Finally, the traceability module 13 is used to trace the combs with good and poor quality grades to find the source information of the corresponding comb production line. According to the source information, the production lines from which each comb comes are aggregated to monitor and judge the errors and omissions of each production link of the production line, and then the comb production line is reversely monitored by monitoring the comb quality, so as to ensure the quality of subsequent shipped products.

[0041] Among them, the air cushion comb is a common comb, which mainly achieves a comfortable and smooth combing effect through the air cushion design at the bottom of the comb teeth. The production process of the air cushion comb is relatively complicated and involves multiple links, mainly including the following steps:

[0042] 1. Design and mold making:

[0043] The production of air cushion combs begins with the design and mold making stage. First, the designer will design the shape, size and structure of the air cushion comb according to the functional requirements and appearance requirements, and select the appropriate material. After the design is completed, a mold will be made to produce the main structural parts of the comb, including the comb handle, air cushion, comb teeth, etc.

[0044] 2. Raw materials preparation:

[0045] The main materials of air cushion combs usually include plastic, rubber, silicone, etc. The specific material selection will be based on the functional requirements of the comb. Common raw materials include:

[0046] Plastic: Commonly used for the main body of the comb, such as polypropylene (PP) or polyester (PET).

[0047] Rubber or silicone: used for the air cushion part to ensure good elasticity and air tightness.

[0048] Nylon or stainless steel: used for the comb teeth part, with better durability and resistance to breakage.

[0049] 3. Comb handle injection molding machine:

[0050] The main structure of the air cushion comb is mostly produced by injection molding. The injection molding process usually includes:

[0051] Heat and melt the plastic raw materials.

[0052] The molten plastic is injected into the mold and cooled to form the comb body.

[0053] Trim the comb handle to remove any excess plastic.

[0054] 4. Air cushion and air hole production of air cushion production machine:

[0055] The key part of the air cushion comb is the design of the air cushion and air holes. The air cushion is usually made of soft rubber or silicone, and the production process may include:

[0056] Silicone molding: Silicone is injected into the mold and heated and cured to form a soft and elastic air cushion.

[0057] Pore ​​design: The air cushion needs to be perforated so that air can pass through these pores, which can better reduce the pressure on the scalp when combing and provide a comfortable massage effect. The pores are generally achieved through mold design.

[0058] 5. Installation of comb assembly machine:

[0059] The comb teeth are usually made of nylon, plastic or metal. The comb teeth are individually molded and installed on the air cushion, and then fixed to the surface of the air cushion by heat pressing.

[0060] 6. Assembly of comb handle assembly machine:

[0061] After the air cushion and comb teeth are assembled, they are assembled together through the comb handle to form a complete air cushion comb. The assembly process needs to ensure that the various parts are tightly connected, the air holes of the air cushion are not blocked, and the comb teeth remain stable.

[0062] The above production process steps constitute the production line of the comb, and each production link is recorded by the big data module 1. Figure 3 For example, the comb handle injection molding machine in the production line is represented by numbers A1-An; the air cushion production machine is represented by numbers B1-Bn; the comb tooth assembly machine is represented by numbers C1-Cn; and the comb handle assembly machine is represented by numbers D1-Dn.

[0063] Based on the identification of each production link, the production source of each comb is recorded, and the comb code is used for identity recognition. The comb code is bound to the source information of the comb, and the source information of the comb can be obtained by identifying the comb code.

[0064] Based on the source information, by comparing it with the production line of the comb source information, the numbers of the comb handle injection molding machine, air cushion production machine, comb tooth assembly machine, comb handle assembly machine and other production equipment in the production line can be known.

[0065] Based on this, since the produced combs need to undergo quality inspection before they can further enter the market, they need to be quality inspected and classified to find out the defective combs.

[0066] First, the image acquisition module 11 is used to acquire the detection image of the plane of each produced comb and photograph it through the CDD camera.

[0067] The acquired inspection image is passed through an edge positioning to identify the comb area, where the comb code must be found.

[0068] A detection image of the comb area is obtained, and the detection image is first pre-processed to distinguish the comb handle area, the air cushion area and the comb tooth area in the comb area according to the comb handle edge data information, the air cushion edge data information and the comb tooth edge data information.

[0069] See also Figure 2 , the comb handle area in the comb area is distinguished by the comb handle edge data information.

[0070] The steps of implementing the detection image preprocessing include the following methods:

[0071] S1: Convert the acquired detection image into a grayscale detection image in advance;

[0072] S2: Perform image noise reduction processing on the grayscale detection image;

[0073] S3: Obtain the grayscale value of each pixel of the grayscale detection image, construct a grayscale histogram, and perform binarization processing, perform tilt correction on the grayscale detection image after binarization processing, and complete the detection image preprocessing.

[0074] Afterwards, the defect analysis module 12 performs image processing based on the acquired comb area, identifies the defect position in the comb area, and determines the quality grade of the comb based on the area size and degree of the defect position. The quality grades are: excellent, good, and poor.

[0075] The method of the defect analysis module 12 identifying the defect position in the comb area:

[0076] For the comb handle area and the air cushion area, the defect edge information in the comb handle area and the air cushion area is extracted through secondary edge detection;

[0077] For the comb tooth area, the defect shape information is extracted through shape features;

[0078] Based on the standard comparison template image constructed by the big data module 1, the acquired defect edge information and defect shape information are compared with the local features corresponding to the standard comparison template image to determine the correctness of the acquired defect edge information and defect shape information.

[0079] Based on the division of the comb handle area, the air cushion area and the comb tooth area, and the confirmation of the defect edge information and the defect shape information, the specific location of the defect in the comb area can be known.

[0080] On this basis, the specific position and shape of the defect on the defect edge information and defect shape information are re-extracted through three edge detections;

[0081] According to the specific location and shape of the defect, the area size of the defect is calculated, and the actual number of defects is counted. By setting the defect number threshold and the defect area threshold, and in the set range of the defect number threshold and the defect area threshold, the quality grade of the comb is divided into three types: excellent, good and poor.

[0082] After the quality level of the combs is divided into three types, namely, excellent, good and poor, the good and poor combs are counted, and the traceability module 13 is used to identify and trace the combs, and the specific number of good and poor combs produced on each production line is counted. The good and poor combs are collectively referred to as defective products.

[0083] For combs with excellent quality, production will be completed.

[0084] However, for defective products, the problem lies in the equipment of the production line, which leads to a decline in the production quality of the combs. Therefore, the specific number of defective products is counted and the traceability module 13 is used to identify and trace the source, so as to reversely monitor the production equipment on the production line.

[0085] To avoid accidental incidents, avoid accidental emergencies caused by defective products.

[0086] By counting the specific number of defective products produced on each production line and matching the defect locations on each defective product, the specific processing errors and omissions in the production links on each production line can be quickly identified based on the defect locations.

[0087] For example, the defective product is identified on the production line through comb code identification, and the production is completed by comb handle injection molding machine A1, air cushion production machine B1, comb tooth assembly machine C1, and comb handle assembly machine D1. The defect analysis module 12 determines that the defect position of the defective product is in the comb handle area. It can be known that the production fault may be in the comb handle injection molding machine A1. For this reason, the comb handle injection molding machine can be improved by overhauling it, so as to adjust the quality of subsequent production.

[0088] Based on the detection of defective products, since there may be errors in defect analysis, a correction unit is provided. After the comb code is identified twice by the image acquisition module 11, the quality grade of the comb is assumed to be excellent.

[0089] The combs that are incorrectly identified are re-recorded, and the image acquisition module 11 re-identifies and acquires the combs that are incorrectly identified. Based on the information encoded by the comb, the combs that are incorrectly identified are assumed to be defect-free, and the detection images acquired by the image acquisition module 11 twice before and after are stored, and the detection images are analyzed by the defect analysis module 12 again, and the defect position analysis of the defect analysis module 12 determines that it is a suspected defect, and the analysis data information of the suspected defect is retransmitted to the big data module 1 to construct a deep learning network. The deep learning network uses a convolutional neural network.

[0090] The above embodiments are merely explanations of the present invention and are not limitations of the present invention. After reading this specification, those skilled in the art may make modifications to the embodiments without any creative contribution as needed. However, such modifications are protected by the patent law as long as they are within the scope of the claims of the present invention.

Claims

1. The comb production quality monitoring system based on Internet big data is characterized by: include The big data module obtains the source information of each comb produced in each production line, and performs comb coding on each comb, and obtains the source information of the corresponding comb through the comb coding; An image acquisition module is used to acquire detection images of several planes of the comb, identify the comb area through edge positioning based on the detection images of each plane of the comb, and acquire the comb code; The defect analysis module performs image processing according to the comb area of ​​each plane, identifies the defect position in the comb area, and determines the quality grade of the comb based on the area size and degree of the defect position. The quality grade is: excellent, good, and poor; The traceability module is based on the quality grade classification of combs. For combs with good and poor quality grades, the module obtains the source information of the comb production line based on the comb coding information obtained by the image acquisition module. Based on the source information, the module collects the production lines from which each comb comes, thereby monitoring and judging the errors and omissions in each production link of the production line.

2. The comb production quality monitoring system based on Internet big data according to claim 1 is characterized in that: The source information includes the production process information of the injection molding of the comb handle, the assembly production process information of the air cushion and the comb teeth, and the assembly production process information of the air cushion and the comb handle. By binding each production equipment and assembly equipment with the corresponding production process information, a data information network is constructed, stored through a big data module, and bundled based on the comb code identification.

3. The comb production quality monitoring system based on Internet big data according to claim 1 is characterized in that: The image acquisition module is based on the detection images of each plane of the comb. By preprocessing the detection images, the comb handle area, the air cushion area and the comb tooth area in the comb area are distinguished according to the comb handle edge data information, the air cushion edge data information and the comb tooth edge data information.

4. The comb production quality monitoring system based on Internet big data according to claim 3 is characterized in that: Based on the detection image acquired by the image acquisition module, the steps of implementing detection image preprocessing include the following methods: S1: Convert the acquired detection image into a grayscale detection image in advance; S2: Perform image noise reduction processing on the grayscale detection image; S3: Obtain the grayscale value of each pixel of the grayscale detection image, construct a grayscale histogram, and perform binarization processing, perform tilt correction on the grayscale detection image after binarization processing, and complete the detection image preprocessing.

5. The comb production quality monitoring system based on Internet big data according to claim 4 is characterized in that: The method for the defect analysis module to identify the defect location in the comb area: For the comb handle area and the air cushion area, the defect edge information in the comb handle area and the air cushion area is extracted through secondary edge detection; For the comb tooth area, the defect shape information is extracted through shape features; Based on the standard comparison template image constructed by the big data module, the acquired defect edge information and defect shape information are compared with the local features corresponding to the standard comparison template image to determine the correctness of the acquired defect edge information and defect shape information.

6. The comb production quality monitoring system based on Internet big data according to claim 5 is characterized in that: After obtaining the verified defect edge information and defect shape information, the specific position and shape of the defect on the defect edge information and defect shape information are re-extracted through three edge detections; According to the specific location and shape of the defect, the area size of the defect is calculated, and the actual number of defects is counted. By setting the defect number threshold and the defect area threshold, and in the set range of the defect number threshold and the defect area threshold, the quality grade of the comb is divided into three types: excellent, good and poor.

7. The comb production quality monitoring system based on Internet big data according to claim 6 is characterized in that: Based on the quality grade of combs, they are divided into three types: excellent, good and poor. The good and poor types of combs are counted, and the traceability module is used to identify and trace them, and the specific number of good and poor types of combs produced on each production line is counted.

8. The comb production quality monitoring system based on Internet big data according to claim 7 is characterized in that: By counting the specific number of good and poor combs produced on each production line, and matching the defect locations on each good and poor comb, the specific processing errors and omissions in the production links on each production line can be quickly identified based on the defect locations.

9. The comb production quality monitoring system based on Internet big data according to claim 1 is characterized in that: The image acquisition module is also provided with a correction unit. Among the two types of combs, good and bad, the combs with incorrect identification are re-recorded, and the image acquisition module re-identifies and acquires the combs with incorrect identification. Based on the information encoded by the comb, the combs with incorrect identification are defaulted as defect-free, and the detection images obtained by the image acquisition module twice before and after are stored. The detection images are then analyzed by the defect analysis module, and determined to be suspected defects based on the defect position analysis of the defect analysis module. The analysis data information of the suspected defects is retransmitted to the big data module to build a deep learning network.

Images