A surface quality detection method and system for metal zipper

Abstract

The invention discloses a surface quality detection system for metal zippers, which belongs to the technical field of quality detection. The system includes: an image acquisition module, which obtains zipper closing images and zipper opening images; an area division module, which divides the specified position grids of each mime tooth in the zipper closing images and zipper opening images; a mime tooth recognition module, which recognizes the actual mime tooth area of ​​each mime tooth in the zipper closing images and zipper opening images according to preset mime tooth shape data; a first analysis module, which analyzes and determines whether the number of zipper mime teeth and the position of each mime tooth are qualified according to the specified position grids and the actual mime tooth area of ​​each mime tooth. The present invention combines the detection of the zipper closing image and the zipper opening image to comprehensively detect the quality of all mime teeth of the zipper, improve the comprehensiveness and accuracy, and detects the mime tooth position by the divided specified position grid and the actual mime tooth area, thereby improving the detection efficiency.

Description

Technical Field

[0001] The invention relates to the technical field of quality detection, and in particular to a surface quality detection method and system for a metal zipper. Background Art

[0002] The metal zipper is a commonly used closure. Its main feature is that key components such as the teeth, sliders and limit codes are made of metal materials. The surface quality of the zipper directly affects the smoothness of opening and closing of the zipper, especially the surface quality of the teeth.

[0003] In the invention patent with application number CN202311117274.4, a surface quality detection method and system for metal zippers are disclosed. The global surface infrared image information is acquired by infrared imaging equipment, and the surface defect area of ​​interest is determined based on the infrared defect detection network to obtain the zipper surface defect image feature information. The invention detects surface cracks, defects, discoloration, and hole defects, but cannot detect the number and position of the metal zipper teeth, and also cannot detect the surface quality of the teeth blocked by the slider. Based on the above technical problems, the present invention provides a surface quality detection method and system for metal zippers. Summary of the invention

[0004] In order to solve the technical problems existing in the prior art, the present invention provides a surface quality detection system for a metal zipper, comprising:

[0005] An image acquisition module, used for acquiring zipper closing images and zipper opening images;

[0006] A micro-tooth digital detection module is used to detect whether the number and position of the micro-tooth of the zipper are qualified according to the zipper closing image and the zipper opening image;

[0007] The mimic teeth digital detection module includes a region division module, a mimic teeth recognition module and a first analysis module;

[0008] A region division module, used to divide the specified position grids of each miter tooth in the zipper closing image and the zipper opening image;

[0009] A mime tooth recognition module, used for recognizing the actual mime tooth area of ​​each mime tooth in the zipper closing image and the zipper opening image according to preset mime tooth shape data;

[0010] The first analysis module is used to analyze and determine whether the number of zipper mime teeth and the position of each mime tooth are qualified according to the specified position grid of each mime tooth and the actual area of ​​the mime teeth.

[0011] Furthermore, the region division module divides the specified position grids of each miter tooth in the zipper closed image and the zipper opened image, specifically:

[0012] Extracting a first zipper region in the zipper closed image and a second zipper region in the zipper open image by an edge detection algorithm;

[0013] According to the preset position data, a first predetermined position grid of each miter tooth is divided in the first zipper area and a second predetermined position grid of each miter tooth is divided in the second zipper area;

[0014] The first prescribed position grid of each miking tooth is matched with the corresponding second prescribed position grid.

[0015] Furthermore, the mimic tooth recognition module recognizes the actual mimic tooth area of ​​each mimic tooth in the zipper closed image and the zipper opened image according to the preset mimic tooth shape data, specifically:

[0016] Construct a miming tooth shape template according to the preset miming tooth shape data;

[0017] Extracting a plurality of first mimic tooth candidate regions from the first zipper region, and extracting a plurality of second mimic tooth candidate regions from the second zipper region;

[0018] Extracting local shape features of edge pixels in each of the first mimic tooth candidate regions and the second mimic tooth candidate regions;

[0019] Each local shape feature is compared with the mimic tooth shape template to evaluate the shape similarity. If it is greater than or equal to the preset similarity threshold, the corresponding first mimic tooth candidate area is used as the first mimic tooth actual area and the second mimic tooth candidate area is used as the second mimic tooth actual area.

[0020] Furthermore, the first analysis module determines whether the number of zipper mime teeth and the position of each mime tooth are qualified according to the specified position grid of each mime tooth and the actual area of ​​the mime teeth, specifically:

[0021] The first prescribed position grid is overlapped with the actual area of ​​the first mimic tooth and the second prescribed position grid is overlapped with the actual area of ​​the second mimic tooth;

[0022] If the number of pixels of the actual area of ​​the first mimic tooth that exceeds the corresponding first specified position grid is greater than the first preset point threshold, the position of the corresponding mimic tooth is unqualified; if the number of pixels of the actual area of ​​the second mimic tooth that exceeds the corresponding second specified position grid is greater than the first preset point threshold, the position of the corresponding mimic tooth is unqualified;

[0023] If the number of pixels of the actual area of ​​the first mimic tooth in the first specified position grid is less than the second preset point threshold, then the first specified position grid does not have a mimic tooth; if the number of pixels of the actual area of ​​the second mimic tooth in the second specified position grid is less than the second preset point threshold, then the second specified position grid does not have a mimic tooth;

[0024] If there is no miter tooth in the first specified position grid and the specified position grid matching it, it is determined that the number of miter teeth of the zipper is unqualified.

[0025] Furthermore, it also includes:

[0026] The tooth material detection module is used to detect whether the materials of the zipper teeth are consistent.

[0027] Furthermore, the tooth material detection module includes:

[0028] A visual feature analysis module, used for analyzing the visual features of each zipper through the zipper closed image and the zipper opened image;

[0029] Spectral analysis module, used to obtain metal information of each microphone tooth through spectral analysis technology;

[0030] The second analysis module is used to analyze whether the materials of each miking tooth are consistent according to the visual features and metal information of each miking tooth.

[0031] Furthermore, the visual features include color, texture and gloss; and the metal information includes metal composition and relative content.

[0032] Furthermore, the second analysis module analyzes whether the materials of the microphone teeth are consistent according to the visual features and metal information of the microphone teeth, specifically:

[0033] Through the trained multimodal deep neural network model, the material category of each miking tooth is analyzed according to the visual features and metal information of each miking tooth;

[0034] The material category with the largest number of miter teeth is taken as the main material, and it is determined whether there are miter teeth of a material category different from the main material. If so, the miter teeth of the zipper are of inconsistent material.

[0035] Furthermore, it also includes:

[0036] The meshing abnormality prediction module is used to predict the abnormal meshing position of the zipper teeth, specifically:

[0037] The number of zipper teeth and the position information of each zipper tooth obtained by the zipper tooth digital detection module and the material information of each zipper tooth obtained by the zipper tooth material detection module are input into the deep learning model that has been trained with historical data, and the estimated position of the zipper tooth meshing abnormality is obtained through analysis;

[0038] The historical data includes the abnormal position of the zipper where the engagement abnormality has occurred in the past, as well as the number of teeth of the corresponding zipper and the position information and material information of each tooth.

[0039] The present invention also provides a surface quality detection method for a metal zipper, comprising the following steps:

[0040] S1, obtaining a zipper closed image and a zipper opened image;

[0041] S2, detecting whether the number and position of the zipper teeth are qualified according to the zipper closed image and the zipper opened image, specifically:

[0042] S211, dividing a predetermined position grid of each miter tooth in the zipper closed image and the zipper opened image;

[0043] S212, identifying the actual area of ​​each mime tooth in the zipper closed image and the zipper opened image according to the preset mime tooth shape data;

[0044] S213, analyzing and judging whether the number of zipper mimic teeth and the position of each mimic tooth are qualified according to the specified position grid of each mimic tooth and the actual area of ​​the mimic tooth.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] By combining the closed and opened zipper images, the quality of all zipper teeth can be fully evaluated, avoiding the influence of slider occlusion and improving the comprehensiveness and accuracy of the detection.

[0047] The specified position grid divided by the preset position data is directly matched with the actual area of ​​the identified zipper teeth to detect the position of each zipper tooth and the number of zipper teeth, thereby improving the detection efficiency and accuracy;

[0048] The consistency of the zipper material is tested by XRF spectrum analysis technology, which greatly improves the detection efficiency. At the same time, non-destructive testing is used to ensure the integrity of the zipper to be tested.

[0049] Combined with the number, position and material of the zipper teeth, the abnormal engagement position of the zipper teeth can be estimated, further improving the comprehensiveness of metal zipper surface quality inspection and discovering potential production defects in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0052] Figure 1 It is a structural block diagram of the surface quality detection system for metal zippers of the present invention;

[0053] Figure 2 It is a structural block diagram of the micro tooth detection module of the system of the present invention;

[0054] Figure 3 is a structural block diagram of a tooth material detection module of a system in another embodiment of the present invention;

[0055] Figure 4 is a structural block diagram of a spectrum analysis module of a system in another embodiment of the present invention;

[0056] Figure 5 The present invention is a flow chart of a surface quality inspection method for a metal zipper. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0059] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their 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 at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0060] See also Figure 1 As shown, the present invention provides a surface quality detection system for a metal zipper, comprising: an image acquisition module, a micrometer tooth digital detection module and a central processing unit;

[0061] The central processing unit is in communication connection with the image acquisition module and the micro-tooth digital detection module.

[0062] The image acquisition module is used to obtain zipper closing images and zipper opening images.

[0063] The zipper closing image is an image of the metal zipper when the slider of the metal zipper is pulled to the rear limit code to close the teeth; the zipper opening image is an image of the metal zipper when the slider of the metal zipper is pulled to the front limit code to separate the teeth; the rear limit code is the limit code located at the closed end of the zipper, and when the zipper is fully closed, the rear limit code will prevent the slider from continuing to slide forward; the front limit code is the limit code located at the open end of the zipper, and when the zipper is fully opened, the front limit code will clamp the slider to prevent it from slipping out from the end of the chain belt.

[0064] The micro-tooth digital detection module is used to detect whether the number and position of the micro-tooth of the zipper are qualified according to the zipper closing image and the zipper opening image.

[0065] See also Figure 2 As shown, the mimic teeth detection module includes: a region division module, a mimic teeth recognition module and a first analysis module.

[0066] The area division module is used to divide the specified position grids of each miter tooth in the zipper closing image and the zipper opening image;

[0067] Specifically:

[0068] Preprocessing the zipper closed image and the zipper open image;

[0069] Extracting a first zipper region in the zipper closed image and a second zipper region in the zipper open image by an edge detection algorithm;

[0070] According to the preset position data, a first predetermined position grid of each miter tooth is divided in the first zipper area and a second predetermined position grid of each miter tooth is divided in the second zipper area;

[0071] The first prescribed position grid of each miking tooth is matched with the corresponding second prescribed position grid.

[0072] The specified position grid is the position where the miter teeth are installed in the metal zipper according to the preset position regulations.

[0073] The mimic tooth recognition module is used to recognize the actual mimic tooth area of ​​each mimic tooth in the zipper closing image and the zipper opening image according to the preset mimic tooth shape data;

[0074] Specifically:

[0075] Construct a miming tooth shape template according to the preset miming tooth shape data;

[0076] Extracting a plurality of first mimic tooth candidate regions from the first zipper region, and extracting a plurality of second mimic tooth candidate regions from the second zipper region;

[0077] Extracting local shape features of edge pixels in each of the first mimic tooth candidate regions and the second mimic tooth candidate regions;

[0078] Each local shape feature is compared with the mimic tooth shape template to evaluate the shape similarity. If it is greater than or equal to the preset similarity threshold, the corresponding first mimic tooth candidate area is used as the first mimic tooth actual area and the second mimic tooth candidate area is used as the second mimic tooth actual area.

[0079] The preprocessing may include grayscale processing, noise removal, contrast enhancement, etc.

[0080] The noise removal is performed by using a filter, and specifically, a Gaussian filter, a median filter, a bidirectional filter, or the like can be selected.

[0081] The edge detection algorithm may be selected to use Sobel operator or Canny edge detection, etc.

[0082] Sobel operator calculates the gradient of pixels in the image in the horizontal and vertical directions, and obtains the edge strength by synthesizing the gradients in these two directions;

[0083] Canny edge detection, a multi-level, multi-stage edge detection algorithm, includes Gaussian filtering, gradient calculation, non-maximum suppression and double threshold detection, which can effectively suppress noise and extract the most significant edges in the image.

[0084] The first analysis module is used to analyze and determine whether the number of zipper mi-tooths and the position of each mi-tooth are qualified according to the specified position grid of each mi-tooth and the actual area of ​​the mi-tooth;

[0085] Specifically:

[0086] The first prescribed position grid is overlapped with the actual area of ​​the first mimic tooth and the second prescribed position grid is overlapped with the actual area of ​​the second mimic tooth;

[0087] If the number of pixels of the actual area of ​​the first mimic tooth that exceeds the corresponding first specified position grid is greater than the first preset point threshold, the position of the corresponding mimic tooth is unqualified; if the number of pixels of the actual area of ​​the second mimic tooth that exceeds the corresponding second specified position grid is greater than the first preset point threshold, the position of the corresponding mimic tooth is unqualified;

[0088] If the number of pixels of the actual area of ​​the first mimic tooth in the first specified position grid is less than the second preset point threshold, then the first specified position grid does not have a mimic tooth; if the number of pixels of the actual area of ​​the second mimic tooth in the second specified position grid is less than the second preset point threshold, then the second specified position grid does not have a mimic tooth;

[0089] If there is no miter tooth in the first specified position grid and the specified position grid matching it, it is determined that the number of miter teeth of the zipper is unqualified.

[0090] Because the slider in the metal zipper will block part of the Mi-teeth, the solution of the present invention obtains the zipper closing image and the zipper opening image respectively, so as to identify the Mi-teeth actually contained in the metal zipper, and then detects the position and quantity of the Mi-teeth of the zipper, thereby improving the accuracy of the Mi-teeth quality detection.

[0091] In some embodiments, the surface quality detection system for a metal zipper provided by the present invention further includes a tooth material detection module for detecting whether the materials of the teeth of the zipper are consistent.

[0092] The tooth material detection module is in communication connection with the central processing unit.

[0093] Because the material of the micro-tooth directly affects the mechanical properties of the zipper, such as tensile strength, wear resistance, and corrosion resistance, if the material of the micro-tooth is inconsistent, the zipper may fail prematurely during use due to insufficient performance of individual micro-tooths, such as breakage, sticking, excessive wear, etc. At the same time, for some metal zippers for special purposes (such as anti-static, electromagnetic shielding, etc.), the consistency of the micro-tooth material directly affects its electrical performance. Inconsistent materials may weaken the electrical function of the zipper and fail to meet the needs of specific application scenarios. Therefore, the present invention adds a test of the consistency of the micro-tooth material in the surface quality test of the metal zipper.

[0094] See also Figure 3 As shown, the tooth material detection module includes: a visual feature analysis module, a spectral analysis module and a second analysis module.

[0095] The visual feature analysis module is used to analyze the visual features of each zipper through the zipper closed image and the zipper opened image;

[0096] The visual characteristics include color, texture, gloss, etc.

[0097] The spectrum analysis module is used to obtain the metal information of each microphone tooth through spectrum analysis technology;

[0098] The metal information includes metal composition and relative content.

[0099] See also Figure 4 As shown, the spectrum analysis module includes: a positioning module, a spectrometer and a material analysis module

[0100] The positioning module is used to obtain the actual spatial position of each mimic tooth according to the actual mimic tooth area of ​​each mimic tooth;

[0101] Specifically:

[0102] Establishing a mapping relationship between the image pixel coordinate system and the spectral device workbench coordinate system;

[0103] The pixel coordinates of the actual area of ​​each mimic tooth are converted into actual space coordinates to obtain the actual space position of each mimic tooth.

[0104] The mapping relationship between the image pixel coordinate system and the spectral device workbench coordinate system is established above. Specifically, the internal and external parameter matrices are solved through a calibration algorithm to achieve the conversion of image pixel coordinates to actual space coordinates.

[0105] The spectrometer is used to emit primary X-rays to each tooth according to the actual spatial position of each tooth and capture the corresponding fluorescent X-ray information;

[0106] Specifically:

[0107] According to the actual spatial position of each microphone tooth, the optimal moving path of the spectrometer probe is planned through the shortest path algorithm;

[0108] According to the optimal moving path, the three-axis platform on which the spectrometer is installed is driven by a stepper motor to adjust the position of the spectrometer probe, so that the spectrometer probe is aligned with the microphone tooth, primary X-rays are emitted and corresponding fluorescent X-ray information is captured.

[0109] The metal analysis module is used to analyze the material information of each tooth according to the fluorescent X-ray information;

[0110] Specifically:

[0111] Converting the fluorescent X-ray information into electrical signals;

[0112] Digitally processing the electrical signal;

[0113] Compare with the built-in standard spectrum database to identify the metal components and characteristic spectral lines contained in each tooth;

[0114] Based on the intensity of the characteristic spectral lines, the relative content of each metal component in each tooth is evaluated through a preset calculation model.

[0115] The spectrum analysis module specifically adopts XRF spectrum analysis technology. XRF uses an X-ray source (usually an X-ray tube) to emit primary X-rays to the sample. When the primary X-rays hit the sample, the inner electrons of the atoms in the sample are excited, causing the outer electrons to fill the vacancies and release fluorescent X-rays. The energy of the fluorescent X-rays is closely related to the atomic structure of the element. Therefore, by detecting the energy and intensity of these fluorescent X-rays, the metal components contained in each tooth and their relative content can be identified;

[0116] XRF spectral analysis technology does not require direct contact with the micro-tooth, will not cause obvious damage to it, and is less affected by environmental factors, thereby improving the accuracy of micro-tooth metal analysis.

[0117] The second analysis module is used to analyze whether the materials of the microphone teeth are consistent according to the visual characteristics and metal information of the microphone teeth;

[0118] Specifically:

[0119] Through the trained multimodal deep neural network model, the material category of each miking tooth is analyzed according to the visual features and metal information of each miking tooth;

[0120] The material category with the largest number of miter teeth is taken as the main material, and it is determined whether there are miter teeth of a material category different from the main material. If so, the miter teeth of the zipper are of inconsistent material.

[0121] In some embodiments, the surface quality detection system for metal zippers provided by the present invention further includes an engagement abnormality prediction module for predicting the engagement abnormality position of the zipper teeth;

[0122] Specifically:

[0123] The number of zipper teeth and the position information of each zipper tooth obtained by the zipper tooth digital detection module and the material information of each zipper tooth obtained by the zipper tooth material detection module are input into the deep learning model that has been trained with historical data, and the estimated position of the zipper tooth meshing abnormality is obtained through analysis;

[0124] The historical data includes the abnormal position of the zipper where the engagement abnormality has occurred in the past, as well as the number of teeth of the corresponding zipper and the position information and material information of each tooth.

[0125] The meshing abnormality prediction module is in communication connection with the central processing unit.

[0126] See also Figure 5 As shown, the present invention also provides a surface quality inspection method for a metal zipper, comprising the following steps:

[0127] S1, obtaining a zipper closed image and a zipper opened image;

[0128] S2. Detecting whether the number and positions of the zipper teeth are qualified according to the zipper closing image and the zipper opening image.

[0129] Step S2 is specifically:

[0130] S211, dividing a predetermined position grid of each miter tooth in the zipper closed image and the zipper opened image;

[0131] S212, identifying the actual area of ​​each mime tooth in the zipper closed image and the zipper opened image according to the preset mime tooth shape data;

[0132] S213, analyzing and judging whether the number of zipper mimic teeth and the position of each mimic tooth are qualified according to the specified position grid of each mimic tooth and the actual area of ​​the mimic tooth.

[0133] Step S211 is specifically as follows:

[0134] S2111, preprocessing the zipper closed image and the zipper opened image;

[0135] S2112, extracting a first zipper region from the zipper closed image and a second zipper region from the zipper open image by using an edge detection algorithm;

[0136] S2113, dividing the first zipper area into a first predetermined position grid for each miter tooth and dividing the second zipper area into a second predetermined position grid for each miter tooth according to the preset position data;

[0137] S2114, matching the first prescribed position grid of each miking tooth with the corresponding second prescribed position grid.

[0138] Step S212 is specifically:

[0139] S2121, constructing a miming tooth shape template according to preset miming tooth shape data;

[0140] S2122, extracting a plurality of first mimic tooth candidate regions from the first zipper region, and extracting a plurality of second mimic tooth candidate regions from the second zipper region;

[0141] S2123, extracting local shape features of edge pixels in each of the first mimic tooth candidate regions and the second mimic tooth candidate regions;

[0142] S2124, compare each local shape feature with the mimic tooth shape template to evaluate the shape similarity. If it is greater than or equal to a preset similarity threshold, the corresponding first mimic tooth candidate area is used as the first mimic tooth actual area and the second mimic tooth candidate area is used as the second mimic tooth actual area.

[0143] Step S213 is specifically:

[0144] S2131, overlap and identify the first prescribed position grid with the actual area of ​​the first miking tooth, and overlap and identify the second prescribed position grid with the actual area of ​​the second miking tooth;

[0145] S2132: If the number of pixels of the actual area of ​​the first mimic tooth that exceeds the corresponding first specified position grid is greater than the first preset point threshold, the position of the corresponding mimic tooth is unqualified; if the number of pixels of the actual area of ​​the second mimic tooth that exceeds the corresponding second specified position grid is greater than the first preset point threshold, the position of the corresponding mimic tooth is unqualified;

[0146] If the number of pixels of the actual area of ​​the first mimic tooth in the first specified position grid is less than the second preset point threshold, then the first specified position grid does not have a mimic tooth; if the number of pixels of the actual area of ​​the second mimic tooth in the second specified position grid is less than the second preset point threshold, then the second specified position grid does not have a mimic tooth;

[0147] If there is no miter tooth in the first specified position grid and the specified position grid matching it, it is determined that the number of miter teeth of the zipper is unqualified.

[0148] In some embodiments, step S2 further includes detecting whether the materials of the zipper teeth are consistent, specifically:

[0149] S221, analyzing the visual features of each miter through the zipper closed image and the zipper opened image;

[0150] S222, obtaining metal information of each microphone tooth through spectral analysis technology;

[0151] S223, analyzing whether the materials of the microphone teeth are consistent based on the visual features and metal information of the microphone teeth.

[0152] In some embodiments, after step S2, the following steps are further included:

[0153] S3. Estimation of abnormal engagement position of zipper teeth, specifically:

[0154] The number of zipper teeth and the position and material information of each tooth are input into a deep learning model that has been trained with historical data, and the estimated position of the zipper tooth engagement abnormality is analyzed to obtain.

[0155] The beneficial effects of the present invention are:

[0156] By combining the closed and opened zipper images, the quality of all zipper teeth can be fully evaluated, avoiding the influence of slider occlusion and improving the comprehensiveness and accuracy of the detection.

[0157] The specified position grid divided by the preset position data is directly matched with the actual area of ​​the identified zipper teeth to detect the position of each zipper tooth and the number of zipper teeth, thereby improving the detection efficiency and accuracy;

[0158] The consistency of the zipper material is tested by XRF spectrum analysis technology, which greatly improves the detection efficiency. At the same time, non-destructive testing is used to ensure the integrity of the zipper to be tested.

[0159] Combined with the number, position and material of the zipper teeth, the abnormal engagement position of the zipper teeth can be estimated, further improving the comprehensiveness of metal zipper surface quality inspection and discovering potential production defects in advance.

[0160] In the description of the specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0161] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A surface quality detection system for metal zippers, characterized in that: include: An image acquisition module is used to obtain a zipper closing image and a zipper opening image; the zipper closing image is an image of the metal zipper when the metal zipper slider is pulled to the rear limit code so that the zipper teeth are completely closed; the zipper opening image is an image of the metal zipper when the metal zipper slider is pulled to the front limit code so that the zipper teeth are completely separated; A Mi tooth digital detection module is used to detect whether the number and position of the Mi teeth of the zipper are qualified according to the zipper closing image and the zipper opening image; the Mi tooth digital detection module includes a region division module, a Mi tooth recognition module and a first analysis module; A region division module is used to divide the specified position grids of each Mi tooth in the zipper closed image and the zipper unzipped image, specifically: extracting a first zipper region in the zipper closed image and a second zipper region in the zipper unzipped image by an edge detection algorithm; dividing the first specified position grid of each Mi tooth in the first zipper region and the second specified position grid of each Mi tooth in the second zipper region according to preset position data; matching the first specified position grid of each Mi tooth with the corresponding second specified position grid; The mime tooth recognition module is used to identify the actual mime tooth area of ​​each mime tooth in the zipper closed image and the zipper opened image according to the preset mime tooth shape data, specifically: Construct a miming tooth shape template according to the preset miming tooth shape data; Extracting a plurality of first mimic tooth candidate regions from the first zipper region, and extracting a plurality of second mimic tooth candidate regions from the second zipper region; Extracting local shape features of edge pixels in each of the first mimic tooth candidate regions and the second mimic tooth candidate regions; Compare each local shape feature with the mimic tooth shape template to evaluate the shape similarity. If the shape similarity is greater than or equal to a preset similarity threshold, use the corresponding first mimic tooth candidate area as the first mimic tooth actual area and use the second mimic tooth candidate area as the second mimic tooth actual area; The first analysis module is used to analyze and determine whether the number of zipper miter teeth and the position of each miter tooth are qualified according to the specified position grid of each miter tooth and the actual area of ​​the miter tooth, specifically: The first prescribed position grid is overlapped with the actual area of ​​the first mimic tooth and the second prescribed position grid is overlapped with the actual area of ​​the second mimic tooth; If the number of pixels of the actual area of ​​the first mimic tooth that exceeds the corresponding first specified position grid is greater than the first preset point threshold, the position of the corresponding mimic tooth is unqualified; if the number of pixels of the actual area of ​​the second mimic tooth that exceeds the corresponding second specified position grid is greater than the first preset point threshold, the position of the corresponding mimic tooth is unqualified; If the number of pixels of the actual area of ​​the first mimic tooth in the first specified position grid is less than the second preset point threshold, then the first specified position grid does not have a mimic tooth; if the number of pixels of the actual area of ​​the second mimic tooth in the second specified position grid is less than the second preset point threshold, then the second specified position grid does not have a mimic tooth; If there is no miter tooth in the first specified position grid and the specified position grid matching it, it is determined that the number of miter teeth of the zipper is unqualified.

2. The surface quality detection system for metal zippers according to claim 1, characterized in that: The region division module divides the zipper closed image and the zipper unzipped image into specified position grids for each miter, and also includes: preprocessing the zipper closed image and the zipper unzipped image, wherein the preprocessing includes one or more of grayscale processing, noise removal and contrast enhancement.

3. The surface quality detection system for metal zippers according to claim 1, characterized in that: Also includes: The tooth material detection module is used to detect whether the materials of the zipper teeth are consistent.

4. The surface quality detection system for metal zippers according to claim 3, characterized in that: The tooth material detection module includes: A visual feature analysis module, used for analyzing the visual features of each zipper through the zipper closed image and the zipper opened image; Spectral analysis module, used to obtain metal information of each microphone tooth through spectral analysis technology; The second analysis module is used to analyze whether the materials of each miking tooth are consistent according to the visual features and metal information of each miking tooth.

5. The surface quality detection system for metal zippers according to claim 4, characterized in that: The visual characteristics include color, texture and gloss; the metal information includes metal composition and relative content.

6. The surface quality detection system for metal zippers according to claim 4, characterized in that: The second analysis module analyzes whether the materials of the microphone teeth are consistent according to the visual features and metal information of the microphone teeth, specifically: Through the trained multimodal deep neural network model, the material category of each miking tooth is analyzed according to the visual features and metal information of each miking tooth; The material category with the largest number of miter teeth is taken as the main material, and it is determined whether there are miter teeth of a material category different from the main material. If so, the miter teeth of the zipper are of inconsistent material.

7. The surface quality detection system for metal zippers according to claim 3, characterized in that: Also includes: The meshing abnormality prediction module is used to predict the abnormal meshing position of the zipper teeth, specifically: The number of zipper teeth and the position information of each zipper tooth obtained by the zipper tooth digital detection module and the material information of each zipper tooth obtained by the zipper tooth material detection module are input into the deep learning model that has been trained with historical data, and the estimated position of the zipper tooth meshing abnormality is obtained through analysis; The historical data includes the abnormal position of the zipper where the engagement abnormality has occurred in the past, as well as the number of teeth of the corresponding zipper and the position information and material information of each tooth.

8. A surface quality detection method for a metal zipper, applied to the surface quality detection system for a metal zipper according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1, obtaining a zipper closed image and a zipper opened image; S2, detecting whether the number and position of the zipper teeth are qualified according to the zipper closed image and the zipper opened image, specifically: S211, dividing the prescribed position grids of each Mi tooth in the zipper closed image and the zipper opened image, specifically: extracting a first zipper area in the zipper closed image and a second zipper area in the zipper opened image by an edge detection algorithm; dividing the first prescribed position grid of each Mi tooth in the first zipper area and the second prescribed position grid of each Mi tooth in the second zipper area according to preset position data; matching the first prescribed position grid of each Mi tooth with the corresponding second prescribed position grid; S212, identifying the actual area of ​​each mime tooth in the zipper closed image and the zipper opened image according to the preset mime tooth shape data, specifically: Construct a miming tooth shape template according to the preset miming tooth shape data; Extracting a plurality of first mimic tooth candidate regions from the first zipper region, and extracting a plurality of second mimic tooth candidate regions from the second zipper region; Extracting local shape features of edge pixels in each of the first mimic tooth candidate regions and the second mimic tooth candidate regions; Compare each local shape feature with the mimic tooth shape template to evaluate the shape similarity. If the shape similarity is greater than or equal to a preset similarity threshold, use the corresponding first mimic tooth candidate area as the first mimic tooth actual area and use the second mimic tooth candidate area as the second mimic tooth actual area; S213, analyzing and judging whether the number of zipper mime teeth and the position of each mime tooth are qualified according to the specified position grid of each mime tooth and the actual area of ​​the mime teeth, specifically: The first prescribed position grid is overlapped with the actual area of ​​the first mimic tooth and the second prescribed position grid is overlapped with the actual area of ​​the second mimic tooth; If the number of pixels of the actual area of ​​the first mimic tooth that exceeds the corresponding first specified position grid is greater than the first preset point threshold, the position of the corresponding mimic tooth is unqualified; if the number of pixels of the actual area of ​​the second mimic tooth that exceeds the corresponding second specified position grid is greater than the first preset point threshold, the position of the corresponding mimic tooth is unqualified; If the number of pixels of the actual area of ​​the first mimic tooth in the first specified position grid is less than the second preset point threshold, then the first specified position grid does not have a mimic tooth; if the number of pixels of the actual area of ​​the second mimic tooth in the second specified position grid is less than the second preset point threshold, then the second specified position grid does not have a mimic tooth; If there is no miter tooth in the first specified position grid and the specified position grid matching it, it is determined that the number of miter teeth of the zipper is unqualified.

Images