Glove processing real-time detection method and system

Abstract

The invention relates to the technical field of image processing, and relates to a glove processing real-time detection method and system. Glove images and weight data of a plurality of target gloves are collected, the glove images are preprocessed to expand training samples, then glove detection standards are created in combination with the glove image and the weight data of each target glove, and then real-time images and real-time weights of real-time gloves in each production step are collected; and judging whether the real-time gloves have flaws or not based on the glove detection standards, if the real-time gloves have the flaws, stopping production, and recording the defective gloves. According to the invention, a plurality of glove detection standards are established by combining the weight data of the target gloves in each processing step, so that the flaw detection of the target gloves in different processing steps is completed; therefore, a user can quickly find and intercept the defective gloves in the processing process, and does not need to wait for the target gloves to complete detection after the target gloves are subjected to all processing steps, so that the material loss in the glove processing process is reduced.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and more specifically, to a real-time detection method and system for glove manufacturing. Background Technology

[0002] Gloves are used for hand warmth or labor protection, and some are also decorative. Gloves are a very special thing. They were not originally made for practical purposes. Only in modern times have they become an essential item for warmth in cold regions, or for medical antibacterial and industrial protective equipment.

[0003] Dipped gloves are protective gloves that have a layer of adhesive applied to their surface through a dipping process, enhancing their protective properties. The manufacturing process includes hand mold pretreatment, dipping in a base adhesive and a base adhesive (such as natural latex, nitrile, or neoprene rubber), multiple drying cycles, and post-treatment to ensure a uniform and firm adhesive layer. Common types include nitrile gloves (oil and acid / alkali resistant), latex gloves (elastic and slip-resistant), and neoprene gloves (aging and corrosion resistant). They are widely used in chemical, medical, and food processing industries, effectively protecting against chemicals, oil, sharp objects, and high temperatures, ensuring hand safety and hygiene.

[0004] Chinese Patent Publication No. CN117593500A discloses an image preprocessing method, apparatus, and medium for detecting defects in textile gloves. The method includes: acquiring a first image containing a front image of the textile glove to be inspected and a second image containing a back image of the textile glove to be inspected; stitching the first image and the second image together in a horizontal direction to obtain a stitched image; and compressing the stitched image to a specified size to obtain a compressed image, so as to use the compressed image for detecting defects in textile gloves. However, in the prior art, inspection is often carried out after the gloves have been processed, which makes it difficult to detect gloves that have problems during the processing in a timely manner, thus increasing the material cost of defective gloves. Summary of the Invention

[0005] The purpose of this invention is to provide a real-time detection method and system for glove manufacturing to solve the problems mentioned in the background art.

[0006] To address the aforementioned technical problems, one objective of this invention is to provide a real-time detection method for glove manufacturing, comprising the following steps: Glove images and weight data of multiple target gloves were collected, and the glove images were preprocessed to expand the training samples. Create glove inspection standards by combining glove images and weight data for each target glove; Real-time images and weights of gloves are collected at each production step, and the gloves are judged for defects in sequence according to the production steps based on glove inspection standards. If a live glove is defective at any step, production is stopped at the problematic step, and the live glove is marked as a defective glove.

[0007] Preferably, the acquisition of glove images and weight data of multiple target gloves, and the preprocessing of the glove images to expand the training samples, includes the following steps: Create a glove database; For each target glove, set the acquisition parameters separately; Based on the acquisition parameters, glove images and weight data of multiple target gloves are acquired, and all acquired data are put into the glove database. Randomly select multiple glove images from the glove database; Perform one or more operations, such as flipping or scaling, on each selected glove image to obtain multiple glove images.

[0008] Preferably, the creation of glove inspection criteria by combining glove images and weight data for each target glove includes the following steps: Randomly select an image and weight data of a target glove from the glove database; The glove image is converted to grayscale to obtain a grayscale image, and the glove region is extracted based on the grayscale image. Establish the coupling relationship between weight data and glove region, and calculate the unit weight of the target glove; Returns the glove image and weight data of a randomly selected target glove from the glove database, until all target gloves in the glove database have been selected, and obtains the unit weight of each target glove. Establish glove testing standards based on the unit weight of all target gloves.

[0009] Preferably, the step of converting the glove image to grayscale to obtain a grayscale image, and extracting the glove region based on the grayscale image, includes the following steps: Calculate the gradient of each pixel in the glove image; Sobel convolution based on gradient; For each pixel, the gradient is calculated using Sobel operator convolution to obtain the gradient magnitude and gradient direction of each pixel. Sort all pixels according to the gradient magnitude and select the top M pixels; the selected M pixels are recorded as edge pixels; Connect multiple edge pixels in sequence. The area surrounded by these edge pixels is the glove area, and the remaining area is the background area.

[0010] Preferably, establishing the coupling relationship between weight data and the glove region, and calculating the unit weight of the target glove, includes the following steps: Select a target glove; Calculate the number of pixels in the glove region of the target glove; Calculate the unit weight corresponding to a single pixel; this unit weight is the unit weight of the target glove. Return to select a target glove, and continue until all target gloves have been selected, obtaining the unit weight of each target glove.

[0011] Preferably, the real-time image and weight of the gloves at each production step are collected, and the gloves are judged for defects in sequence according to the production steps based on glove inspection standards, including the following steps: Collect real-time images and weights of the gloves at each production step. The target glove corresponding to each production step is obtained based on the real-time image of the real-time glove at that production step. Obtain the glove testing standards for the target gloves; By combining the glove inspection standards of the target glove with the real-time weight, it is determined whether the real-time glove has defects.

[0012] Preferably, the step of determining whether a real-time glove has defects by combining the glove inspection standard of the target glove with its real-time weight includes the following steps: Set the error threshold; Calculate the standard weight of the glove in real time based on the glove testing standards, and calculate the difference between the real-time weight and the standard weight. Determine whether the difference between the real-time weight and the standard weight is greater than or equal to the error threshold; If the difference between the real-time weight and the standard weight is greater than or equal to the error threshold, the real-time glove is defective.

[0013] Preferably, the real-time detection method for glove processing further includes the following steps: For each defective glove, obtain the weight data of the defective glove at each processing stage; Calculate the difference between the defective gloves and the glove inspection standard at each processing stage; The processing stage with the largest difference is identified and marked as a high-risk stage.

[0014] The second objective of this invention is to provide a real-time glove processing detection system, applicable to any of the aforementioned real-time glove processing detection methods, comprising an acquisition component and a detection component. The acquisition component acquires glove images and weight data of the target glove. The detection component is communicatively connected to the acquisition component. The glove images and weight data acquired by the acquisition component are input to the detection component. The detection component combines the glove images and weight data to establish glove detection standards for each target glove, and detects whether defects appear in the real-time glove based on the glove detection standards.

[0015] Preferably, the acquisition component includes an image acquisition module and a weight acquisition module. The image acquisition module acquires an image of the target glove, and the weight acquisition module acquires the weight data of the target glove.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: By collecting glove images and weight data from multiple target gloves and preprocessing the glove images to expand the training samples, a glove detection standard is created by combining the glove images and weight data of each target glove. Then, real-time images and weights of real-time gloves are collected, and the glove detection standard is used to determine whether the real-time glove has defects. If a real-time glove has defects, an alarm is issued and the defective glove is recorded. This application establishes multiple glove detection standards by combining the weight data of the target glove at each processing step to complete the defect detection of the target glove at different processing steps. This allows users to quickly discover and intercept defective gloves during the processing, without having to wait until the target glove has gone through all processing steps to complete the detection, thereby reducing material loss during the glove processing. Attached Figure Description

[0017] Figure 1 This is the overall flowchart of Example 1; Figure 2 This is a connection diagram for Example 2; Reference numerals: 100, Acquisition component; 101, Image acquisition module; 102, Weight acquisition module; 200. Detection components. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1 like Figure 1 As shown, one of the objectives of this invention is to provide a real-time detection method for glove manufacturing, comprising the following steps: S100 collects glove images and weight data of multiple target gloves and preprocesses the glove images to expand the training samples. S200 creates glove inspection standards by combining glove images and weight data for each target glove; The S300 collects real-time images and weights of the gloves at each production step, and judges whether the gloves are defective according to the order of the production steps based on the glove inspection standards. S400: If a live glove is defective at any step, production is stopped at the problematic step, and the live glove is recorded as a defective glove.

[0020] It should be noted that by collecting glove images and weight data of multiple target gloves and preprocessing the glove images to expand the training samples, a glove detection standard is created by combining the glove images and weight data of each target glove. Then, real-time images and weights of the real-time glove are collected at each production step. Based on the glove detection standard, it is determined whether the real-time glove has defects. If the real-time glove has defects, production is stopped and the defective glove is recorded. This application establishes multiple glove detection standards by combining the weight data of the target glove at each processing step to complete the defect detection of the target glove at different processing steps. This allows users to quickly discover and intercept defective gloves during the processing, without having to wait until the target glove has gone through all processing steps to complete the detection, thereby reducing material loss during the glove processing.

[0021] By acquiring target images corresponding to each processing step of the target glove during the manufacturing process, the glove region of the target glove obtained after each processing step is extracted from the target image. Then, based on the number of pixels contained in the glove region and the weight of the glove after each processing step, a glove inspection standard is established. Based on this glove inspection standard, inspection can be carried out after each processing step of the glove manufacturing process. On the one hand, this improves the inspection efficiency; on the other hand, since inspection is carried out after each processing step, defective gloves can be detected during the glove manufacturing process, thereby reducing material consumption.

[0022] Since the glove inspection standard is actually the unit weight corresponding to a single pixel, the standard weight of the glove can be inferred by combining the real-time image with the glove inspection standard. That is, the standard weight that the real-time glove should have after the processing steps. When the difference between the real-time weight and the standard weight is large, it can be determined that the real-time glove has defects in the processing steps.

[0023] In one embodiment of this application, S100 includes: S110, Create the glove database; S120, set the acquisition parameters separately for each target glove; S130: Collect glove images and weight data of multiple target gloves based on the acquisition parameters, and put all the acquired data into the glove database; Specifically, since the real-time target of this application is mainly a dipped glove, when acquiring glove images, it is necessary to acquire images of the palm and back of the hand of the target glove separately. S140, Randomly select multiple glove images from the glove database; S150, perform one or more operations such as flipping / scaling on each selected glove image to obtain multiple glove images.

[0024] It should be noted that, since there are multiple processing steps in the glove manufacturing process, a set of glove images and weight data need to be collected after each processing step is completed, so that the glove inspection standard established later can cover the complete production process of the target glove.

[0025] After the image data is collected, operations such as flipping / scaling are used to increase the size of the dataset, thereby making the glove detection standard more reliable.

[0026] In one embodiment of this application, S200 includes: S210, randomly select an image and weight data of a target glove from the glove database; S220, the glove image is converted to grayscale to obtain a grayscale image, and the glove region is extracted based on the grayscale image; S230, Establish the coupling relationship between weight data and glove region, and calculate the unit weight of the target glove; S240, return the glove image and weight data of a randomly selected target glove from the glove database, until all target gloves in the glove database have been selected, and obtain the unit weight of each target glove. S250 establishes glove testing standards based on the unit weight of all target gloves; Specifically, the glove testing standard is actually a set of standard data, namely the weight standard of different types of target gloves under normal processing conditions. The glove testing standard is shown in Table 1 - Glove Testing Standard Table. Glove type Testing standards for step 1 Step 2 testing standards Step 3 testing standards ... Glove A 0.03g / pixel 0.05g / pixel 0.07g / pixel ... Glove B 0.01g / pixel 0.03g / pixel 0.05g / pixel ... Gloves C 0.06g / pixel 0.08g / pixel 0.09g / pixel ... ... ... ... ... ... Table 1 Once the glove inspection standard is obtained, the standard weight of the target glove can be obtained simply by combining the number of pixels in the target glove.

[0027] It should be noted that, since the target glove and the background area are sampled simultaneously when acquiring glove images, the glove area needs to be separated from the background area first in order to ensure the accuracy of the unit weight. This ensures that the corresponding pixels of each target glove in the glove image can be used for the calculation of the unit weight, thereby improving the reliability of the glove inspection standard.

[0028] The glove inspection standard actually measures the weight of the target glove after each processing step. Although there are weight changes due to material loss during processing, the standardization of the processing steps ensures that the weight changes are always within a certain range. Therefore, if a large weight change occurs after a certain processing step, it can be considered that an abnormality has occurred in that processing step, and the target glove can be identified as defective.

[0029] In one embodiment of this application, S220 includes: S221, Calculate the gradient of each pixel in the glove image; S222, Sobel operator convolution based on gradient; S223, For each pixel, use the Sobel operator convolution to calculate the gradient, and obtain the gradient magnitude and gradient direction of each pixel; Specifically, through The formula calculates the gradient magnitude, where, It represents the gradient magnitude at a pixel, where x is the x-coordinate of the pixel and y is the y-coordinate of the pixel. It is the horizontal gradient of a pixel. It is the vertical gradient of a pixel; pass The formula calculates the gradient direction, where, It refers to the gradient direction of a pixel. It is the horizontal gradient of a pixel. It is the vertical gradient of a pixel; S224, sort all pixels according to the gradient magnitude and select the top M pixels; the selected M pixels are recorded as edge pixels; S225, connects multiple edge pixels in sequence, the area surrounded by multiple edge pixels is the glove area, and the remaining area is the background area; Specifically, since the target gloves in this application are mainly dipped gloves, the process of extracting the glove area is actually extracting the dipped area of ​​the glove. If some areas are not completely dipped, the area of ​​the glove area will be smaller than the standard area. Therefore, the standard weight obtained by combining the glove testing standards will also be less than the real-time weight. This allows this application to also detect the completeness of the dipping of the target glove, so as to promptly detect the target glove with incomplete dipping, thereby improving the applicability of this application.

[0030] It should be noted that this application uses the Sobel operator to perform edge detection on the glove region, thereby separating the glove region from the background region in the glove image. This ensures that only the number of pixels in the glove region needs to be considered when calculating the unit weight, thus improving the accuracy of the unit weight.

[0031] The Sobel operator is an edge detection method based on the first derivative. It identifies edges by calculating the gradient of image pixel intensity. It uses two 3*3 convolution kernels to detect brightness changes in the horizontal and vertical directions, respectively, and is suitable for extracting the boundary between objects and the background in an image.

[0032] In one embodiment of this application, S230 includes: S231, Select a target glove; S232, Calculate the number of pixels in the glove region of the target glove; S233, calculate the unit weight corresponding to a single pixel; this unit weight is the unit weight of the target glove; S234, return to select one target glove, until all target gloves have been selected, and obtain the unit weight of each target glove.

[0033] It should be noted that since the glove region of the target glove was obtained in the preceding steps, it is only necessary to determine the glove's weight based on the number of pixels within that region and the corresponding weight data of the glove image. Formula 1 calculates the weight corresponding to each pixel, where h is the unit weight of each pixel, H is the weight data corresponding to the target image, and N is the total number of pixels contained in the glove region.

[0034] In one embodiment of this application, S300 includes: S310 collects real-time images and real-time weights of gloves at each production step. S320: Based on the real-time image corresponding to each production step, obtain the target glove corresponding to the real-time glove in that production step. S330, Obtain the glove inspection standard for the target glove; S340 combines the glove inspection standards of the target glove with the real-time weight to determine whether the real-time glove has defects.

[0035] It should be noted that after obtaining the glove inspection standards, it is only necessary to collect real-time images and real-time weights corresponding to each production step in the real-time glove processing process, and then combine them with the glove inspection standards to complete the defect detection of the real-time gloves.

[0036] First, cluster analysis of the real-time gloves is performed using real-time images to determine the target gloves corresponding to the real-time gloves, so as to better obtain the glove detection standards corresponding to the real-time gloves and thus ensure the reliability of the glove detection standards.

[0037] In this application, K-means clustering can be used to cluster real-time gloves. K-means clustering is a widely used clustering algorithm, mainly used to divide data into predefined K categories. This method iteratively assigns data points to the nearest center (i.e., "centroid") and updates the positions of these centers until the stopping condition is met. In this application, the target gloves corresponding to real-time gloves are obtained based on K-means clustering. On the one hand, machine learning makes the classification of real-time gloves more accurate, and on the other hand, it also improves the efficiency of real-time glove classification.

[0038] In one embodiment of this application, S340 includes: S341, Set the error threshold; S342, calculate the standard weight of the glove in real time based on the glove inspection standard, and calculate the difference between the real-time weight and the standard weight; Specifically, after obtaining the glove inspection standard for real-time gloves, it is only necessary to obtain the pixel points of the real-time gloves based on the real-time images, and then multiply the pixel points by the unit weight to obtain the standard weight. The standard weight is actually the weight that a live glove should have if it is free of defects. Therefore, the live weight can be used as a reference standard for judging whether a live glove is defective. S343, determine whether the difference between the real-time weight and the standard weight is greater than or equal to the error threshold; Specifically, when performing step S343, the real-time weight of the real-time glove after each processing step is compared with the corresponding glove inspection standard. If the difference between the real-time weight after at least one processing step and the standard weight is greater than or equal to the error threshold, the real-time glove can be judged to be defective. S344. If the difference between the real-time weight and the glove inspection standard is greater than or equal to the error threshold, then the real-time glove is defective.

[0039] It should be noted that since the target glove itself will change weight in each processing step of production, the range of error threshold is actually larger than the range of normal weight change. That is, if the weight change of the target glove exceeds the error threshold, it means that the target glove has undergone an unexpected weight change. In this case, the processing step of the target glove is very likely to have been wrong, resulting in defects in the target glove.

[0040] In one embodiment of this application, the real-time detection method for glove processing further includes the following steps: S400: For each defective glove, obtain the weight data of the defective glove at each processing stage; S410, calculate the difference between the defective glove and the glove inspection standard at each processing stage; S420: Obtain the processing stage with the largest difference and mark that processing stage as a high-risk stage; Alternatively, in addition to designating the processing stage with the largest difference as a high-risk stage, the processing step that produces the most defects can also be designated as a high-risk stage.

[0041] It should be noted that after obtaining multiple defective gloves, the defective steps of the gloves can be traced back to check the steps in the processing that are prone to problems. This allows for timely improvement of high-risk stages in the processing, thereby increasing the efficiency of glove processing.

[0042] like Figure 2 As shown, a second objective of this invention is to provide a real-time glove processing detection system, applicable to any of the aforementioned real-time glove processing detection methods, comprising an acquisition component 100 and a detection component 200. The acquisition component 100 acquires glove images and weight data of the target glove. The detection component 200 is communicatively connected to the acquisition component 100. The glove images and weight data acquired by the acquisition component 100 are input to the detection component 200. The detection component 200, in conjunction with the glove images and weight data, establishes glove detection standards for each target glove and detects whether defects exist in the real-time glove based on the glove detection standards.

[0043] It should be noted that the acquisition component 100 collects the glove and weight data of the target glove at each processing step, and establishes a correlation between the glove image and weight data under the same processing step. Then, based on the correlation between the glove image and weight data, the glove inspection standard of the target glove is calculated, thereby completing the defect detection of the real-time glove based on the glove inspection standard.

[0044] In one embodiment of this application, the acquisition component 100 includes an image acquisition module 101 and a weight acquisition module 102. The image acquisition module 101 acquires an image of the target glove, and the weight acquisition module 102 acquires the weight data of the target glove.

[0045] It should be noted that when acquiring glove images, glove images need to be acquired at each step of the glove manufacturing process. Similarly, weight data needs to be acquired after each processing step, and glove images and weight data for the same processing step need to be correlated.

[0046] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for real-time detection of glove manufacturing, characterized in that, include: Glove images and weight data of multiple target gloves were collected, and the glove images were preprocessed to expand the training samples. Create glove inspection standards by combining glove images and weight data for each target glove; Real-time images and weights of gloves are collected at each production step, and the gloves are judged for defects in sequence according to the production steps based on glove inspection standards. If a live glove is defective at any step, production is stopped at the problematic step, and the live glove is marked as a defective glove.

2. The real-time detection method for glove processing according to claim 1, characterized in that: The process of acquiring glove images and weight data from multiple target gloves, and preprocessing the glove images to expand the training samples, includes the following steps: Create a glove database; For each target glove, set the acquisition parameters separately; Based on the acquisition parameters, glove images and weight data of multiple target gloves are acquired, and all acquired data are put into the glove database. Randomly select multiple glove images from the glove database; Perform one or more operations, such as flipping or scaling, on each selected glove image to obtain multiple glove images.

3. The real-time detection method for glove processing according to claim 2, characterized in that: The creation of glove inspection criteria by combining glove images and weight data for each target glove includes the following steps: Randomly select an image and weight data of a target glove from the glove database; The glove image is converted to grayscale to obtain a grayscale image, and the glove region is extracted based on the grayscale image. Establish the coupling relationship between weight data and glove region, and calculate the unit weight of the target glove; Returns the glove image and weight data of a randomly selected target glove from the glove database, until all target gloves in the glove database have been selected, and obtains the unit weight of each target glove. Establish glove testing standards based on the unit weight of all target gloves.

4. The real-time detection method for glove processing according to claim 3, characterized in that: The process of converting the glove image to grayscale to obtain a grayscale image, and then extracting the glove region based on the grayscale image, includes the following steps: Calculate the gradient of each pixel in the glove image; Sobel convolution based on gradient; For each pixel, the gradient is calculated using Sobel operator convolution to obtain the gradient magnitude and gradient direction of each pixel. Sort all pixels according to the gradient magnitude and select the top M pixels; the selected M pixels are recorded as edge pixels. Connect multiple edge pixels in sequence. The area surrounded by these edge pixels is the glove area, and the remaining area is the background area.

5. The real-time detection method for glove processing according to claim 4, characterized in that: The process of establishing the coupling relationship between weight data and the glove region, and calculating the unit weight of the target glove, includes the following steps: Select a target glove; Calculate the number of pixels in the glove region of the target glove; Calculate the unit weight corresponding to a single pixel; this unit weight is the unit weight of the target glove. Return to select a target glove, and continue until all target gloves have been selected, obtaining the unit weight of each target glove.

6. The real-time detection method for glove processing according to claim 5, characterized in that: The process involves collecting real-time images and weights of the gloves at each production step, and then, based on glove inspection standards, sequentially determining whether the gloves are defective according to the production steps. This includes the following steps: Collect real-time images and weights of the gloves at each production step. The target glove corresponding to each production step is obtained based on the real-time image of the real-time glove at that production step. Obtain the glove testing standards for the target gloves; By combining the glove inspection standards of the target glove with the real-time weight, it is determined whether the real-time glove has defects.

7. The real-time detection method for glove processing according to claim 6, characterized in that: The method of determining whether a real-time glove has defects by combining the glove inspection standards of the target glove with its real-time weight includes the following steps: Set the error threshold; Calculate the standard weight of the glove in real time based on the glove testing standards, and calculate the difference between the real-time weight and the standard weight. Determine whether the difference between the real-time weight and the standard weight is greater than or equal to the error threshold; If the difference between the real-time weight and the standard weight is greater than or equal to the error threshold, the real-time glove is defective.

8. The real-time detection method for glove processing according to claim 7, characterized in that: The real-time detection method for glove processing also includes the following steps: For each defective glove, obtain the weight data of the defective glove at each processing stage; Calculate the difference between the defective gloves and the glove inspection standard at each processing stage; The processing stage with the largest difference is identified and marked as a high-risk stage.

9. A real-time detection system for glove processing, applied to the real-time detection method for glove processing according to any one of claims 1-8, characterized in that: include: Data acquisition components; The acquisition component acquires images and weight data of the target glove. Detection components; The detection component is communicatively connected to the acquisition component. The glove image and weight data acquired by the acquisition component are input to the detection component. The detection component combines the glove image and weight data to establish a glove detection standard for each target glove, and detects whether the glove has defects in real time based on the glove detection standard.

10. The real-time detection system for glove processing according to claim 9, characterized in that, The acquisition component includes an image acquisition module and a weight acquisition module. The image acquisition module acquires an image of the target glove, and the weight acquisition module acquires the weight data of the target glove.

Images