A multifunctional bag production defect detection method and system

Abstract

The invention discloses a method and system for detecting production defects of multifunctional luggage, belonging to the technical field of production defect detection of multifunctional luggage. The method comprises: establishing a display model of the multifunctional luggage; performing a benchmark detection on the produced multifunctional luggage to obtain a corresponding benchmark detection result; performing a sewing detection on the multifunctional luggage to determine the sewing area of ​​the multifunctional luggage; collecting a sewing image of the sewing area; performing grayscale processing on the sewing image to obtain a sewing grayscale image; obtaining a corresponding sewing background image in the sewing image; performing grayscale adjustment on the sewing grayscale image according to the sewing background image; determining each hole area and stitching area through the adjusted sewing grayscale image; respectively analyzing each hole area and stitching area to obtain a hole detection result and a stitching detection result, integrating the hole detection result and the stitching detection result into a sewing detection result; and marking the sewing detection result and the benchmark detection result correspondingly in the display model.

Description

Technical Field

[0001] The present invention belongs to the technical field of multifunctional luggage production defect detection, and specifically relates to a multifunctional luggage production defect detection method and system. Background Art

[0002] With the continuous expansion of the global leather goods and luggage market, especially the booming high-end leather luggage market, the requirements for luggage production quality are getting higher and higher. Leather luggage, as a traditional luxury product, has a complex production process and expensive materials, so quality control during the production process is particularly critical. However, traditional luggage production defect detection mainly relies on manual visual inspection, which is not only inefficient, but also easily affected by human factors, resulting in missed detection or false detection.

[0003] In order to solve the above problems, in recent years, with the rapid development of computer vision and artificial intelligence technology, more and more automated inspection equipment has been applied to the field of luggage production. These devices realize automatic detection of luggage defects through image acquisition, feature extraction, classification and recognition, greatly improving the detection efficiency and accuracy.

[0004] However, the existing bag production defect detection system still has some shortcomings. First, most systems mainly detect the overall appearance of bags, and it is difficult to accurately identify the texture, color, scratches and other details that are unique to leather bags. Secondly, due to the particularity of leather materials, such as irregular texture and color differences, defect detection becomes more complicated. In addition, existing systems can often only realize single-function detection, such as scratch detection or size detection, and cannot achieve multi-functional integration.

[0005] Based on this, the present invention provides a multifunctional bag production defect detection method and system. Summary of the invention

[0006] In order to solve the problems existing in the above-mentioned solutions, the present invention provides a multifunctional bag production defect detection method and system.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A multifunctional bag production defect detection system, comprising a display module, a reference detection module and a sewing detection module;

[0009] The display module is used to establish a display model of the multifunctional luggage.

[0010] The benchmark detection module is used to perform benchmark detection on the produced multifunctional luggage, obtain corresponding benchmark detection results, and make corresponding marks in the display model according to the benchmark detection results.

[0011] The sewing detection module is used to perform sewing detection on the multifunctional luggage, determine the sewing area of ​​the multifunctional luggage; collect sewing images of the sewing area; perform grayscale processing on the sewing images to obtain corresponding sewing grayscale images; obtain corresponding sewing background images in the sewing images;

[0012] Adjusting the grayscale of the sewing grayscale image according to the sewing background image; determining each hole area and stitching area through the adjusted sewing grayscale image;

[0013] Analyze each hole area and stitch area separately to obtain corresponding hole detection results and stitch detection results, integrate the hole detection results and stitch detection results into sewing detection results; and mark the sewing detection results accordingly in the display model.

[0014] Furthermore, the method for collecting sewing images includes:

[0015] Acquire a regional reference image corresponding to the sewing area; perform grayscale processing on the regional reference image to obtain a reference grayscale image, identify the grayscale value of each pixel in the reference grayscale image, merge regions according to the grayscale value of each pixel, and obtain a number of unit regions; determine the acquisition conditions of each unit region;

[0016] Merge adjacent unit areas with the same acquisition conditions to obtain conditional acquisition areas, and mark the acquisition conditions corresponding to each conditional acquisition area;

[0017] Marking each condition acquisition area in the sewing area accordingly, acquiring images of the sewing area according to the acquisition conditions corresponding to each condition acquisition area, obtaining initial acquisition images corresponding to each condition acquisition area, and cutting the image corresponding to the condition acquisition area in the initial acquisition image to obtain a single acquisition image;

[0018] The single collected images are merged to obtain the sewing image.

[0019] Furthermore, the method of merging regions according to the grayscale value of each pixel includes:

[0020] Step SA1: Set a reference point; mark the gray value of the reference point as h(x, y), where (x, y) represents the position of the pixel corresponding to the reference point;

[0021] Mark the pixels adjacent to the reference point as the candidate point, and mark the gray value of the candidate point as h(x´, y´), where (x´, y´) represents the position of the pixel corresponding to the candidate point;

[0022] The grayscale difference between the reference point and the point to be selected is calculated according to the formula HC=|h(x, y)-h(x´, y´)|; where: HC is the grayscale difference;

[0023] Perform a merge evaluation between the reference point and the candidate point based on the grayscale difference;

[0024] When the assessment does not meet the merger requirements, no corresponding merger will be carried out;

[0025] When the evaluation meets the merging requirements, the reference point and the candidate point are merged to obtain the merged area;

[0026] Step SA2: Mark the pixels adjacent to the merged area as the candidate points, regard each pixel in the merged area as a reference point, calculate the grayscale difference between each reference point and the candidate point, mark the largest grayscale difference as the grayscale representative difference, and perform a merge evaluation between the merged area and the candidate point according to the grayscale point difference;

[0027] When the assessment does not meet the merger requirements, no corresponding merger will be carried out;

[0028] When the evaluation meets the merging requirements, the merging area is merged with the point to be selected to obtain a new merging area;

[0029] Step SA3: looping step SA2 until there are no points to be selected that meet the merging requirements in each merged area, marking each merged area as a unit area;

[0030] Step SA4: Return to step SA1 until there is no reference point.

[0031] Furthermore, the method for determining the acquisition conditions of each unit area includes:

[0032] Performing illumination simulation on each unit area under different simulation conditions, obtaining collected images of sewing holes and sutures on each unit area under different simulation conditions, and performing recognition effect analysis on the collected images of sewing holes and sutures, respectively, to obtain corresponding first effect values ​​and second effect values;

[0033] Calculate the simulation value corresponding to each simulation condition according to the formula PL=b3×XL1+b4×XL2;

[0034] The simulation condition with the largest simulation value is selected as the acquisition condition for the corresponding unit area.

[0035] Further, the method for analyzing each hole area includes:

[0036] Identify the area of ​​each hole region, establish a hole anomaly model, analyze the area of ​​each hole region through the hole anomaly model, and obtain the corresponding anomaly value;

[0037] Identify the interval distances between adjacent hole areas, and calculate corresponding distribution values ​​according to each interval distance;

[0038] According to the formula Calculate the corresponding hole detection value;

[0039] Where: KPR is the hole detection value; b1 and b2 are both proportionality coefficients, and the value range is 0 < b1 ≤ 1, 0 < b2 ≤ 1; FB is the distribution value; K(si) represents the hole anomaly model; si represents the area of the corresponding hole region, i = 1, 2, ……, n, and n is a positive integer;

[0040] When the hole detection value is greater than the threshold X2, the hole detection is evaluated as unqualified;

[0041] When the hole detection value is not greater than the threshold X2, the hole detection is evaluated as qualified.

[0042] Furthermore, the expression of the hole anomaly model is ;

[0043] Where: si represents the area of the corresponding hole region, i = 1, 2, ……, n, and n is a positive integer; the output data is the anomaly value 1 or 0.

[0044] Furthermore, the method for analyzing the suture area includes:

[0045] Real-time identify the width at each suture in the suture area; obtain the suture color and the background color corresponding to each suture, and integrate the width, suture color and background color into the suture characteristics of the suture at the corresponding position;

[0046] Establish a visual evaluation model, analyze each suture characteristic through the visual evaluation model to obtain the visual value of each suture position; generate a corresponding visual curve according to the visual value of each position, fit the visual curve to obtain the corresponding curve function, and mark the curve function as QP(t); t is the independent variable of the curve function, and the independent variable is the suture position;

[0047] According to the formula Calculate the corresponding suture evaluation value;

[0048] Where: QAU is the suture evaluation value;

[0049] When the suture evaluation value is greater than the threshold X3, the suture detection is evaluated as unqualified;

[0050] When the suture evaluation value is not greater than the threshold X3, the suture detection is evaluated as qualified.

[0051] Furthermore, the expression of the visual evaluation model is ;

[0052] Where: rj represents the corresponding suture characteristic; the output data is the visual value.

[0053] A method for detecting production defects of multifunctional luggage, the method includes:

[0054] Establish a display model of multifunctional luggage; conduct benchmark tests on the produced multifunctional luggage to obtain corresponding benchmark test results;

[0055] Perform sewing detection on the multifunctional luggage to determine the sewing area of ​​the multifunctional luggage; collect sewing images of the sewing area; perform grayscale processing on the sewing images to obtain corresponding sewing grayscale images; obtain corresponding sewing background images in the sewing images;

[0056] Adjusting the grayscale of the sewing grayscale image according to the sewing background image; determining each hole area and stitching area through the adjusted sewing grayscale image;

[0057] Analyze each hole area and stitch area respectively to obtain corresponding hole detection results and stitch detection results, and integrate the hole detection results and stitch detection results into sewing detection results;

[0058] The sewing test results and the benchmark test results are marked correspondingly in the display model.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] Through the cooperation between the display module, the benchmark detection module and the sewing detection module, the intelligent defect monitoring of the leather multifunctional bags is realized, the sewing detection and benchmark detection of the leather multifunctional bags are solved, and the defect monitoring accuracy of the multifunctional bags is improved. The various defects of the leather bags in the production process can be efficiently and accurately detected, which greatly improves the production efficiency and product quality. At the same time, the system is also scalable and customizable, and can be flexibly configured and upgraded according to different production needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] 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, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0063] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. 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.

[0064] like Figure 1 As shown, a multifunctional bag production defect detection system includes a display module, a reference detection module, and a sewing detection module;

[0065] The display module is used to establish a display model of the multifunctional luggage based on the existing visualization technology, and is used to subsequently display the inspection results of the multifunctional luggage. For example, if the size is unqualified, the corresponding unqualified position in the display model is marked.

[0066] The benchmark detection module is used to perform benchmark detection on the produced multifunctional bags, and the benchmark detection includes applying various settings of existing technologies such as weight detection, size detection, scratch detection, appearance detection, etc.; presetting corresponding detection methods and detection standards; and obtaining corresponding benchmark detection results.

[0067] The sewing detection module is used to perform sewing detection on the multifunctional luggage, determine the sewing area for sewing according to the production design data of the multifunctional luggage; perform positioning adjustment on the multifunctional luggage, move the sewing area of ​​the multifunctional luggage to the collection area, and collect the sewing image of the sewing area;

[0068] Performing grayscale processing on the sewing image to obtain a corresponding sewing grayscale image; identifying the corresponding conditional acquisition areas in the sewing image, obtaining the image corresponding to each conditional acquisition area in the reference grayscale image, marking it as a standard image of the acquisition area, and combining the standard images of each acquisition area into a sewing background image;

[0069] Grayscale adjustment is performed on the sewing grayscale image according to the sewing background image, that is, the grayscale value in the sewing grayscale image is adjusted according to the difference between the grayscale value of each pixel in the sewing grayscale image and the grayscale value of the corresponding pixel in the sewing background image, so as to obtain more accurate grayscale values ​​of sewing holes and stitches; the grayscale adjustment of the sewing grayscale image can be realized by using intelligent technologies such as neural networks;

[0070] According to the sewing background image and the adjusted sewing grayscale image, the hole area corresponding to each sewing hole and the stitching area corresponding to the stitching line can be determined; one hole area corresponds to one thread hole; because of the continuity of the stitching line, one sewing area corresponds to one stitching line area;

[0071] Analyze each hole area and stitch area separately to obtain corresponding hole detection results and stitch detection results, integrate the hole detection results and stitch detection results into sewing detection results; and mark the sewing detection results accordingly in the display model.

[0072] In one embodiment, the method for collecting the sewing image of the sewing area is to collect images according to the prior art.

[0073] In another embodiment, the method for collecting the sewing image of the sewing area is as follows:

[0074] Obtain the bag pattern of the designed sewing area and mark it as the regional reference image; perform grayscale processing on the regional reference image to obtain a reference grayscale image, identify the grayscale value of each pixel in the reference grayscale image, merge the regions according to the grayscale value of each pixel, and obtain several unit regions; determine the acquisition conditions of each unit region;

[0075] Merge adjacent unit areas with the same acquisition conditions to obtain conditional acquisition areas. Unit areas that have not been merged are also marked as conditional acquisition areas; and the acquisition conditions corresponding to each conditional acquisition area are marked;

[0076] Marking each condition acquisition area in the sewing area accordingly, acquiring images of the sewing area according to the acquisition conditions corresponding to each condition acquisition area, obtaining an initial acquisition image corresponding to each condition acquisition area, and cutting the image corresponding to the condition acquisition area in the initial acquisition image to obtain a single acquisition image;

[0077] The single collected images are merged to obtain the sewing image.

[0078] Methods for merging regions based on the grayscale value of each pixel include:

[0079] Step SA1: Setting reference points, that is, randomly selecting a number of non-adjacent pixel points in the reference grayscale image as reference points, but pixels within the unit area will not be set as reference points;

[0080] The gray value of the reference point is marked as h(x, y), where (x, y) represents the position of the pixel corresponding to the reference point;

[0081] Mark the pixels adjacent to the reference point as the candidate point, and mark the gray value of the candidate point as h(x´, y´), where (x´, y´) represents the position of the pixel corresponding to the candidate point;

[0082] The grayscale difference between the reference point and the point to be selected is calculated according to the formula HC=|h(x, y)-h(x´, y´)|; where: HC is the grayscale difference;

[0083] The merging evaluation between the reference point and the candidate point is performed according to the grayscale difference, that is, the corresponding preset value is set according to whether the grayscale difference can be regarded as the same degree of color, and the grayscale difference is compared with the preset value. If the grayscale difference is greater than the preset value, it is considered that the merging requirement is not met, otherwise it is considered that the merging requirement is met;

[0084] When the assessment does not meet the merger requirements, no corresponding merger will be carried out;

[0085] When the evaluation meets the merging requirements, the reference point and the candidate point are merged to obtain the merged area;

[0086] Step SA2: Mark the pixels adjacent to the merged area as the candidate points, regard each pixel in the merged area as the reference point, calculate the grayscale difference between each reference point and the candidate point, mark the largest grayscale difference as the grayscale representative difference, and perform a merge evaluation between the merged area and the candidate point according to the grayscale point difference. That is, if the grayscale representative difference is greater than the preset value, it is considered that the merge requirement is not met, otherwise it is considered that the merge requirement is met;

[0087] When the assessment does not meet the merger requirements, no corresponding merger will be carried out;

[0088] When the evaluation meets the merging requirements, the merging area is merged with the point to be selected to obtain a new merging area;

[0089] Step SA3: looping step SA2 until there are no points to be selected that meet the merging requirements in each merged area, marking each merged area as a unit area;

[0090] Step SA4: Return to step SA1 until there is no reference point.

[0091] Methods for determining the acquisition conditions of each unit area include:

[0092] Perform illumination simulation on each unit area, simulate the image recognition of sewing holes and sutures of each unit area under different colors, light intensities and other illuminations, mark different colors, light intensities and other illuminations as simulation conditions, obtain collected images of sewing holes and sutures of each unit area under different simulation conditions, recognize the collected images, recognize according to the recognition method of the present invention, obtain the recognition effect of sewing holes and sutures, obtain various recognition effects in the current image recognition field, sort the recognition effects in descending order, set the effect value of the first-ranked recognition effect to 10, and set the effect value of the lowest-ranked recognition effect to 1; set the effect values ​​of different recognition effects according to the degree of difference of each recognition effect; set the corresponding first effect value and second effect value according to the recognition effect of sewing holes and sutures;

[0093] Calculate the simulation value corresponding to each simulation condition according to the formula PL=b3×XL1+b4×XL2;

[0094] The simulation condition with the largest simulation value is selected as the acquisition condition for the unit area.

[0095] Methods for analyzing each hole area include:

[0096] Identify the area of ​​each hole region, compare the area of ​​each hole region obtained with the threshold X1, and when the area is greater than the threshold X1, it is considered a hole anomaly; establish the corresponding hole anomaly model, the expression is ; where: si represents the area of the corresponding hole region, i = 1, 2, ……, n, and n is a positive integer; the output data is the outlier value 1 or 0;

[0097] Analyze the areas of the respective hole regions through the hole anomaly model to obtain the corresponding outlier values;

[0098] Identify the interval distances between adjacent hole regions, calculate the corresponding distribution values based on the respective interval distances, that is, calculate the corresponding standard deviation using the standard deviation formula, and label it as the distribution value;

[0099] According to the formula Calculate the corresponding hole detection value; where: KPR is the hole detection value; b1 and b2 are both proportionality coefficients, and the value range is 0 < b1 ≤ 1, 0 < b2 ≤ 1; FB is the distribution value;

[0100] When the hole detection value is greater than the threshold X2, evaluate that the hole detection is unqualified;

[0101] When the hole detection value is not greater than the threshold X2, evaluate that the hole detection is qualified.

[0102] The method for analyzing the sewing thread area includes:

[0103] Real-time identify the widths at each sewing thread in the sewing thread area; obtain the corresponding sewing thread color and background color at each sewing thread, and the background color is the corresponding luggage color; integrate the width, sewing thread color and background color into the sewing thread characteristics of this position; establish a visual evaluation model for evaluating whether the sewing thread at this position meets the visibility requirements according to the sewing thread characteristics of this position, that is, it cannot be too visible and prominent; specifically, obtain a large number of historical sewing thread characteristics, mark whether they meet the visibility requirements of the corresponding visibility level, integrate them into training data, and establish a visual evaluation model according to the training data for evaluating whether the sewing thread characteristics meet the visibility level required by users; the expression is ; where: rj represents the sewing thread characteristics of the sewing thread at the corresponding position; the output data is the visual value;

[0104] Analyze each sewing thread characteristic through the visual evaluation model to obtain the visual value of the sewing thread at each position; generate a corresponding visual curve according to the visual values at each position, with the horizontal axis being the sewing thread position and the vertical axis being the visual value; fit the visual curve to obtain the corresponding curve function, and label the curve function as QP(t), where t represents the corresponding horizontal axis; according to the formula Calculate the corresponding sewing thread evaluation value; where: QAU is the sewing thread evaluation value;

[0105] When the sewing thread evaluation value is greater than the threshold X3, evaluate that the sewing thread detection is unqualified;

[0106] When the sewing thread evaluation value is not greater than the threshold X3, evaluate that the sewing thread detection is qualified.

[0107] A multifunctional bag production defect detection method, the method comprising:

[0108] Establish a display model of multifunctional luggage; conduct benchmark tests on the produced multifunctional luggage to obtain corresponding benchmark test results;

[0109] Perform sewing detection on the multifunctional luggage to determine the sewing area of ​​the multifunctional luggage; collect sewing images of the sewing area; perform grayscale processing on the sewing images to obtain corresponding sewing grayscale images; obtain corresponding sewing background images in the sewing images;

[0110] Adjusting the grayscale of the sewing grayscale image according to the sewing background image; determining each hole area and stitching area through the adjusted sewing grayscale image;

[0111] Analyze each hole area and stitch area respectively to obtain corresponding hole detection results and stitch detection results, and integrate the hole detection results and stitch detection results into sewing detection results;

[0112] The sewing test results and the benchmark test results are marked correspondingly in the display model.

[0113] The above formulas are all calculated by removing dimensions and taking numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data.

[0114] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A multifunctional bag production defect detection system, characterized in that: It includes a display module, a benchmark detection module and a sewing detection module; The display module is used to establish a display model of the multifunctional luggage; The benchmark detection module is used to perform benchmark detection on the produced multifunctional luggage, obtain corresponding benchmark detection results, and make corresponding marks in the display model according to the benchmark detection results; The sewing detection module is used to perform sewing detection on the multifunctional luggage and determine the sewing area of ​​the multifunctional luggage; Collecting sewing images of the sewing area; Performing grayscale processing on the sewing image to obtain a corresponding sewing grayscale image; Obtaining a sewing background image corresponding to the sewing image; grayscale adjustment is performed on the sewing grayscale image according to the sewing background image; Determine each hole area and stitch area through the adjusted sewing grayscale image; Analyze each hole area and stitch area respectively to obtain corresponding hole detection results and stitch detection results, and integrate the hole detection results and stitch detection results into sewing detection results; Mark the sewing test results accordingly in the display model; The acquisition methods of sewing images include: Acquire a regional reference image corresponding to the sewing area; perform grayscale processing on the regional reference image to obtain a reference grayscale image, identify the grayscale value of each pixel in the reference grayscale image, merge regions according to the grayscale value of each pixel, and obtain a number of unit regions; determine the acquisition conditions of each unit region; Merge adjacent unit areas with the same acquisition conditions to obtain conditional acquisition areas, and mark the acquisition conditions corresponding to each conditional acquisition area; Marking each condition acquisition area in the sewing area accordingly, acquiring images of the sewing area according to the acquisition conditions corresponding to each condition acquisition area, obtaining initial acquisition images corresponding to each condition acquisition area, and cutting the image corresponding to the condition acquisition area in the initial acquisition image to obtain a single acquisition image; Merge each single collected image to obtain a sewing image; Methods for determining the acquisition conditions of each unit area include: Performing illumination simulation on each unit area under different simulation conditions, obtaining collected images of sewing holes and sutures on each unit area under different simulation conditions, and performing recognition effect analysis on the collected images of sewing holes and sutures, respectively, to obtain corresponding first effect values ​​and second effect values; Calculate the simulation value corresponding to each simulation condition according to the formula PL=b3×XL1+b4×XL2; The simulation condition with the largest simulation value is selected as the acquisition condition for the corresponding unit area.

2. A multifunctional bag production defect detection system according to claim 1, characterized in that: Methods for merging regions based on the grayscale value of each pixel include: Step SA1: Set a reference point; mark the gray value of the reference point as h(x, y), where (x, y) represents the position of the pixel corresponding to the reference point; Mark the pixels adjacent to the reference point as the candidate point, and mark the gray value of the candidate point as h(x´, y´), where (x´, y´) represents the position of the pixel corresponding to the candidate point; The grayscale difference between the reference point and the point to be selected is calculated according to the formula HC=|h(x, y)-h(x´, y´)|; where: HC is the grayscale difference; Perform a merge evaluation between the reference point and the candidate point based on the grayscale difference; When the assessment does not meet the merger requirements, no corresponding merger will be carried out; When the evaluation meets the merging requirements, the reference points and the candidate points are merged to obtain a merged area; Step SA2: Mark the pixels adjacent to the merged area as candidate points, regard each pixel in the merged area as a reference point, calculate the gray - level difference between each reference point and the candidate points, mark the maximum gray - level difference as the representative gray - level difference, and conduct a merging evaluation between the merged area and the candidate points according to the gray - level point difference; When the evaluation does not meet the merging requirements, no corresponding merging is performed; When the evaluation meets the merging requirements, the merged area and the candidate points are merged to obtain a new merged area; Step SA3: Loop step SA2 until there are no candidate points that meet the merging requirements in each merged area, and mark each merged area as a unit area; Step SA4: Return to step SA1 until there are no reference points.

3. The multifunctional bag production defect detection system according to claim 1 is characterized in that: The methods for analyzing each hole area include: Identify the area of each hole area, establish a hole anomaly model, analyze the area of each hole area through the hole anomaly model to obtain the corresponding anomaly value; Identify the interval distance between each adjacent hole area, and calculate the corresponding distribution value according to each interval distance; According to the formula Calculate the corresponding hole detection value; In the formula: KPR is the hole detection value; b1 and b2 are both proportionality coefficients, and the value range is 0 < b1 ≤ 1, 0 < b2 ≤ 1; FB is the distribution value; K(si) represents the hole anomaly model; si represents the area of the corresponding hole area, i = 1, 2, ……, n, and n is a positive integer; When the hole detection value is greater than the threshold X2, the hole detection is evaluated as unqualified; When the hole detection value is not greater than the threshold X2, the hole detection is evaluated as qualified.

4. A multifunctional bag production defect detection system according to claim 3, characterized in that: The expression of the hole anomaly model is: ; In the formula: si represents the area of the corresponding hole area, i = 1, 2, ……, n, and n is a positive integer; the output data is the anomaly value 1 or 0; X1 is the threshold.

5. A multifunctional bag production defect detection system according to claim 4, characterized in that: The methods for analyzing the sewing line area include: Real - time identify the width at each sewing line in the sewing line area; obtain the corresponding sewing line color and background color at each sewing line, and integrate the width, sewing line color and background color into the sewing line features of the corresponding position; Establish a visual evaluation model, analyze each sewing line feature through the visual evaluation model to obtain the visual value of each position sewing line; generate a corresponding visual curve according to the visual values of each position, fit the visual curve to obtain the corresponding curve function, and mark the curve function as QP(t), where t is the independent variable of the curve function, and the independent variable is the sewing line position; According to the formula Calculate the corresponding suture evaluation value; In the formula: QAU is the sewing line evaluation value; When the sewing line evaluation value is greater than the threshold X3, the sewing line detection is evaluated as unqualified; When the sewing line evaluation value is not greater than the threshold X3, the sewing line detection is evaluated as qualified.

6. A multifunctional bag production defect detection system according to claim 5, characterized in that: The expression of the visual assessment model is ; In the formula: rj represents the corresponding sewing line feature; the output data is the visual value.

7. A multifunctional bag production defect detection method, characterized in that: Applied to a multifunctional luggage production defect detection system as described in any one of claims 1 to 6, the method includes: Establish a display model of the multifunctional luggage; conduct a reference detection on the produced multifunctional luggage to obtain the corresponding reference detection result; Conduct a sewing detection on the multifunctional luggage to determine the sewing area of the multifunctional luggage; collect the sewing image of the sewing area; perform gray - level processing on the sewing image to obtain the corresponding sewing gray - level image; obtain the corresponding sewing background image in the sewing image; Adjusting the grayscale of the sewing grayscale image according to the sewing background image; determining each hole area and stitching area through the adjusted sewing grayscale image; Analyze each hole area and stitch area respectively to obtain corresponding hole detection results and stitch detection results, and integrate the hole detection results and stitch detection results into sewing detection results; The sewing test results and the benchmark test results are marked correspondingly in the display model.

Images