Intelligent safety detection method for clothing wearable product

Abstract

The application discloses a kind of smart safety detection methods of clothing wearable product, which comprises the following steps: S1, X-ray scanning is carried out to clothing wearable product unfolding flat;S2, X-ray image is preprocessed and segmented using improved watershed segmentation algorithm;S3, using edge detection algorithm divides clothing and electronic device contour line and calculates area;S4, obtain the ratio between electronic device area and clothing area, obtain comfort ratio;S5, obtain the parameter information of clothing, predict the durability of clothing through big data analysis;S6, circuit safety detection is carried out to electronic device;S7, combined with each detection index, draw comprehensive safety performance evaluation.The classification independent detection of clothing and electronic product in clothing wearable product, and the calculation of the relationship between the two, realize multi-aspect deep-level detection, so as to ensure the accurate analysis of the performance of wearable product in all aspects.

Description

Technical Field

[0001] This invention relates to the field of wearable product testing technology, and more specifically, to an intelligent safety testing method for clothing-type wearable products. Background Technology

[0002] Wearable clothing products are devices that incorporate functional electronic circuits into products with established testing standards to enable natural interaction between humans and the external environment. Therefore, the safety testing of wearable clothing products consists of two main parts: the safety testing of the electronic circuits and the safety testing of the products with existing comprehensive testing standards.

[0003] Currently, to conduct safety testing on wearable clothing products, the first step is to classify them, for example, into clothing, textiles and leather, plastics and rubber, environmental health and safety, precision instruments and jewelry, or glass and ceramics, or into electronic circuit devices, and then test them according to the corresponding safety testing standards.

[0004] However, wearable clothing products require prolonged close contact with the human body, and existing comprehensive safety testing standards cannot adequately test their safety. Furthermore, current testing methods cannot address the comfort aspects of wearable clothing products. Factors such as the fabric of the clothing, the placement and size of electronic devices, and many other factors can affect the product's usability and, in severe cases, pose safety hazards. Therefore, more in-depth and multifaceted testing is necessary.

[0005] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0006] In response to the problems in related technologies, this invention proposes an intelligent security detection method for wearable clothing products to overcome the aforementioned technical problems existing in the prior art.

[0007] Therefore, the specific technical solution adopted by the present invention is as follows:

[0008] A smart security detection method for wearable clothing products, the method comprising the following steps:

[0009] S1. Lay out the wearable clothing product flat and perform an X-ray scan to obtain an X-ray image;

[0010] S2. Preprocess the X-ray image and segment it using an improved watershed segmentation algorithm;

[0011] S3. Use edge detection algorithms to divide the outlines of clothing and electronic devices and calculate their areas;

[0012] S4. Obtain the ratio between the area of ​​electronic devices and the area of ​​clothing to obtain the comfort ratio;

[0013] S5. Obtain the parameter information of the clothing and predict its durability through big data analysis;

[0014] S6. Perform circuit safety testing on electronic equipment;

[0015] S7. A comprehensive safety performance evaluation is derived by combining various test indicators.

[0016] Furthermore, the preprocessing of the X-ray image and the segmentation using the improved watershed segmentation algorithm include the following steps:

[0017] S21. Eliminate noise from the X-ray image and perform morphological opening and closing reconstruction filtering to obtain an initial image;

[0018] S22. Perform morphological gradient operation on the initial image to obtain a gradient image;

[0019] S23. The initial image is labeled using the maximum entropy thresholding method to obtain a labeled image;

[0020] S24. The gradient image is corrected using the labeled image by the minimum value calibration method;

[0021] S25. Perform watershed segmentation on the corrected image to obtain a segmented image.

[0022] Furthermore, the step of eliminating noise from the X-ray image and performing morphological opening and closing reconstruction filtering to obtain the initial image includes the following steps:

[0023] S211. Perform open reconstruction on the X-ray image, expressed as:

[0024]

[0025] in, This represents the open reconstruction of X-ray images using geodesic erosion as the initial image for dilation reconstruction, where f represents the initial image and ε represents the open reconstruction. (n) (f) represents geodesic corrosion. This represents the dilation reconstruction of the mask image by the initial image;

[0026] S212. Perform closed-loop reconstruction on the X-ray image, expressed as:

[0027]

[0028] in, This represents the closed reconstruction of X-ray image using geodesic dilatation as the initial image, where f represents the initial image and δ represents the initial image. (n)(f) represents geodesic dilatation. This represents the erosion reconstruction of the mask image by the initial image.

[0029] Furthermore, the process of labeling the reconstructed image using the maximum entropy thresholding method to obtain a labeled image includes the following steps:

[0030] S231. The maximum entropy thresholding method is used to select the optimal threshold based on the statistical characteristics of the initial image, and the formula for calculating the optimal threshold is:

[0031]

[0032] Where L represents the number of gray levels in the initial image, and the gray levels are divided into background class A and target class B. A represents the background class and takes the value A = (0, 1, ..., t), B represents the target class and takes the value B = (t+1, t+2, ..., L-1), ωA represents the gray mean of class A, ωB represents the gray mean of class B, and ω0 represents the total gray mean of the image.

[0033] S232. Remove the noise and pseudo-minimum values ​​formed by image texture in the initial image after gradient operation to obtain the labeled image.

[0034] Furthermore, the wearable products in the clothing category include apparel and electronic devices.

[0035] Furthermore, the step of using an edge detection algorithm to segment the outlines of clothing and electronic devices and calculate their areas includes the following steps:

[0036] S31. Remove discontinuous point interference from the segmented image after dilation and erosion processing;

[0037] S32. Determine the pixel coordinates of the clothing boundary in the segmented image and obtain the outline of the corresponding clothing boundary coordinate points;

[0038] S33, Calculate the pixel area of ​​the clothing;

[0039] S34. Perform digital binarization on the segmented image and topological analysis on the digitally binarized image;

[0040] S35. Determine the pixel coordinates of the electronic device boundary and obtain the outline of the corresponding electronic device boundary coordinate points.

[0041] S36. Calculate the pixel area of ​​the electronic device.

[0042] Furthermore, both the pixel area of ​​the calculated clothing and the pixel area of ​​the calculated electronic device are calculated using Green's formula to determine the pixel area enclosed by the center point.

[0043] Furthermore, the acquisition of clothing parameters and the prediction of clothing durability through big data analysis include the following steps:

[0044] S51. Obtain the parameter information of the clothing, including the material and weaving process;

[0045] S52. Input the clothing parameter information into the big data platform;

[0046] S53. The big data platform compares the parameter information with known clothing parameters in the database to find N sets of closest lifespans.

[0047] S54. Calculate the durability index of the clothing by averaging the N groups of closest parameters.

[0048] Furthermore, the circuit safety testing of the electronic device includes the following steps:

[0049] S61. Detect the operating temperature of the electronic device;

[0050] S62. Detect the operating power of the electronic device;

[0051] S63. Conduct safety inspections on transformers using transformer safety inspection standards;

[0052] S64. Conduct information security testing using information security testing standards;

[0053] S65. Conduct safety testing on overload protection circuits using overload protection circuit safety testing standards;

[0054] S66. Conduct safety testing on the battery using battery safety testing standards;

[0055] S67. Conduct radiation safety testing using radiation safety testing standards.

[0056] Furthermore, the comprehensive safety performance evaluation derived by combining various test indicators includes the following steps:

[0057] S71. If the number of failures in the circuit safety test of an electronic device is greater than zero, the wearable clothing product shall be deemed a non-compliant product.

[0058] S71. If the number of failures in the circuit safety test of electronic devices is zero, then a comprehensive evaluation should be conducted by combining the durability and comfort ratio of wearable clothing products.

[0059] The beneficial effects of this invention are as follows: by classifying and independently testing clothing and electronic products in wearable clothing products, and calculating the relationship between the two, it is possible to achieve multi-faceted and in-depth testing, thereby ensuring accurate analysis of the performance of wearable products in all aspects and solving the safety hazards caused by heat radiation from clothing wearable products due to their close contact with the human body.

[0060] By introducing X-ray scanning to examine wearable clothing products, clear and intuitive images of clothing and electronic devices are generated. Preprocessing and an improved watershed segmentation algorithm are used to denoise and enhance the X-ray images, enabling precise segmentation of clothing and electronic devices and avoiding oversegmentation. Simultaneously, an edge detection algorithm is introduced to calculate the area of ​​clothing and electronic devices within the image, determining their proportions. Based on these proportions, a preliminary analysis of the comfort and safety of the wearable clothing product can be conducted. Furthermore, by acquiring and analyzing parameters such as the material and weaving process of the clothing, the comprehensiveness of the testing for wearable clothing products is increased. Finally, by testing the safety of various aspects of the electronic devices, a comprehensive testing and evaluation of wearable clothing products is achieved, ensuring the safe and stable use of these products. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a flowchart of an intelligent security detection method for wearable clothing products according to an embodiment of the present invention. Detailed Implementation

[0063] According to an embodiment of the present invention, an intelligent security detection method for wearable clothing products is provided.

[0064] Safety testing for wearable clothing products includes two parts: products with existing comprehensive testing standards and electronic circuits. For mature products, classification can be done by referring to the national standards search published by the China National Standardization Administration, and relevant safety items can be tested according to the national standards in the classification catalog. However, the electronic circuits, due to their application scenarios involving contact with or close contact with human skin, differ from conventional electronic products and require more rigorous and comprehensive safety monitoring to ensure that no potential circuit malfunctions occur during daily use.

[0065] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, the intelligent security detection method for wearable clothing products according to an embodiment of the present invention includes the following steps:

[0066] S1. Lay out the wearable clothing product flat and perform an X-ray scan to obtain an X-ray image;

[0067] The wearable products mentioned above include clothing and electronic devices.

[0068] S2. Preprocess the X-ray image and segment it using an improved watershed segmentation algorithm;

[0069] The watershed segmentation algorithm is a region-based mathematical morphology segmentation method that is highly effective in image segmentation applications. Its fast, efficient, and accurate segmentation characteristics have gradually gained attention. The watershed algorithm utilizes the concept of "topography." Its basic idea is to treat the image as a terrain surface, with gray values ​​representing terrain elevation. It assumes that a hole is made in the surface at each local minimum region, and water will slowly enter the surface through these holes, gradually flooding basins starting from the lowest local minimum. Furthermore, at a certain point, when the water levels from two different local minima converge, a dam is built to flood the entire terrain surface, thus segmenting many regions. These regions are called water-bearing basins, and the dam built at the edge of the regions is called the watershed. Because images are easily affected by noise and dark texture details, many pseudo-local minima exist in the image. They generate corresponding pseudo-water basins in the image. Therefore, pseudo-local minima, along with true local minima, are treated as independent regions and segmented using the traditional watershed algorithm, ultimately leading to severe oversegmentation. To solve this problem, improvements to the watershed segmentation algorithm are needed. Currently, there are two main approaches: one is image post-processing, where similar regions are merged using a certain criterion after applying the watershed segmentation algorithm; however, this results in a large number of segmented regions after image transformation, making the merging process quite complex. The other approach is to preprocess the image before using the watershed segmentation algorithm to effectively suppress pseudo-boundaries caused by noise and texture. This invention adopts the second approach.

[0070] Step S2 includes the following steps:

[0071] S21. Eliminate noise from the X-ray image and perform morphological opening and closing reconstruction filtering to obtain an initial image;

[0072] Morphological opening and closing reconstruction operations can simultaneously eliminate light and dark details in gradient images, correct regional minima and maxima, and mitigate the oversegmentation problem of watersheds caused by details and noise interference. Meanwhile, the salient contours of objects are restored during reconstruction, allowing the image to be simplified while preserving the shape information of the main objects. This smoothing process does not produce boundary shifting.

[0073] The step of eliminating noise from the X-ray image and performing morphological opening and closing reconstruction filtering to obtain the initial image includes the following steps:

[0074] S211. Perform open reconstruction on the X-ray image, expressed as:

[0075]

[0076] in, This represents the open reconstruction of X-ray images using geodesic erosion as the initial image for dilation reconstruction, where f represents the initial image and ε represents the open reconstruction. (n) (f) represents geodesic corrosion. This represents the dilation reconstruction of the mask image by the initial image;

[0077] S212. Perform closed-loop reconstruction on the X-ray image, expressed as:

[0078]

[0079] in, This represents the closed reconstruction of X-ray image using geodesic dilatation as the initial image, where f represents the initial image and δ represents the initial image. (n) (f) represents geodesic dilatation. This represents the erosion reconstruction of the mask image by the initial image.

[0080] S22. Perform morphological gradient operation on the initial image to obtain a gradient image;

[0081] Gradient images can effectively reflect the changing trends of the original image. Watershed segmentation performed on gradient images yields better results and more accurate segmentation. Since the selection of different gradient values ​​significantly impacts the final segmentation effect, this paper employs morphological gradients to perform gradient transformation on the original image. Morphological gradients allow for greater variation in the grayscale levels of the input image while exhibiting less dependence on edge directionality. The morphological expression for a grayscale image is:

[0082] ρ B =δ B -ε B ;

[0083] In the formula, ρ represents the morphological gradient, δ and ε represent morphological dilation and erosion, respectively, and B is the selected structural element. Because the disk-shaped structural element is isotropic, the morphological gradient has less dependence on the directionality of the edges.

[0084] S23. The initial image is labeled using the maximum entropy thresholding method to obtain a labeled image;

[0085] The step of labeling the reconstructed image using the maximum entropy thresholding method to obtain a labeled image includes the following steps:

[0086] S231. The maximum entropy thresholding method is used to select the optimal threshold based on the statistical characteristics of the initial image, and the formula for calculating the optimal threshold is:

[0087]

[0088] Where L represents the number of gray levels in the initial image, and the gray levels are divided into background class A and target class B. A represents the background class and takes the value A = (0, 1, ..., t), B represents the target class and takes the value B = (t+1, t+2, ..., L-1), ωA represents the gray mean of class A, ωB represents the gray mean of class B, and ω0 represents the total gray mean of the image.

[0089] After morphological gradient and opening / closing reconstruction filtering, some dark noise and irregular details still exist in the gradient image, causing the image to be segmented into many meaningless regions. Label extraction involves marking the target points of interest. Traditional H-minimal label extraction methods are somewhat arbitrary in their threshold selection. This invention uses the maximum entropy thresholding method to select a threshold based on the statistical characteristics of the image, removing pseudo-minimum values ​​in the gradient image caused by noise and image texture details.

[0090] S232. Remove the noise and pseudo-minimum values ​​formed by image texture in the initial image after gradient operation to obtain the labeled image.

[0091] S24. The gradient image is corrected using the labeled image by the minimum value calibration method;

[0092] S25. Perform watershed segmentation on the corrected image to obtain a segmented image.

[0093] S3. Use edge detection algorithms to divide the outlines of clothing and electronic devices and calculate their areas;

[0094] Step S3 includes the following steps:

[0095] S31. Remove discontinuous point interference from the segmented image after dilation and erosion processing;

[0096] S32. Determine the pixel coordinates of the clothing boundary in the segmented image and obtain the outline of the corresponding clothing boundary coordinate points;

[0097] S33, Calculate the pixel area of ​​the clothing;

[0098] S34. Perform digital binarization on the segmented image and topological analysis on the digitally binarized image;

[0099] S35. Determine the pixel coordinates of the electronic device boundary and obtain the outline of the corresponding electronic device boundary coordinate points.

[0100] S36. Calculate the pixel area of ​​the electronic device.

[0101] The pixel area of ​​both the computational clothing and the computational electronic device is calculated using Green's formula, which measures the pixel area enclosed by the center point.

[0102] Specifically, the pixel area calculation method is based on Green's formula, transforming the area integral of each region into a line integral along the boundary of the integration region, and representing the boundary of the integration region by the vertex coordinates of a polygon. This leads to the method for calculating the dielectric constant Fourier expansion coefficients of a periodic grating represented by a polygon. Green's formula is as follows:

[0103]

[0104] S4. Obtain the ratio between the area of ​​electronic devices and the area of ​​clothing to obtain the comfort ratio;

[0105] S5. Obtain the parameter information of the clothing and predict its durability through big data analysis;

[0106] Step S5 includes the following steps:

[0107] S51. Obtain the parameter information of the clothing, including the material and weaving process;

[0108] S52. Input the clothing parameter information into the big data platform;

[0109] S53. The big data platform compares the parameter information with known clothing parameters in the database to find N sets of closest lifespans.

[0110] S54. Calculate the durability index of the clothing by averaging the N groups of closest parameters.

[0111] S6. Perform circuit safety testing on electronic equipment;

[0112] Step S6 includes the following steps:

[0113] S61. Detect the operating temperature of the electronic device;

[0114] S62. Detect the operating power of the electronic device;

[0115] S63. Conduct safety inspections on transformers using transformer safety inspection standards;

[0116] S64. Conduct information security testing using information security testing standards;

[0117] S65. Conduct safety testing on overload protection circuits using overload protection circuit safety testing standards;

[0118] S66. Conduct safety testing on the battery using battery safety testing standards;

[0119] S67. Conduct radiation safety testing using radiation safety testing standards.

[0120] S7. A comprehensive safety performance evaluation is derived by combining various test indicators.

[0121] During safety testing, it is necessary to check whether the rated power of the heating element exceeds the power threshold and / or whether the maximum temperature exceeds the temperature threshold. Thermal safety testing of clothing-type wearable products can be conducted with reference to the safety testing methods in GB 4706.8-2008. For example, the clothing-type wearable product can be placed in a constant temperature and humidity environment to test the maximum temperature rise and maximum temperature to determine whether it meets safety standards.

[0122] Radiation safety testing of wearable clothing products requires measuring their radiation intensity. The testing methods in GB / T 6113.101-2016, GB9159-2008, GB / T9254-2008, GB / T12572-2008, and GB / T 28446.1-2012 can be referenced. The testing method is as follows: Place the wearable clothing product in an environment with a temperature of 15-35℃ and humidity not exceeding 75%. Use an EMF tester to test at distances of 0cm, 3cm, 6cm, and 9cm from the product. Measure each point five times consecutively, with each measurement lasting at least 15 seconds. Record the maximum value at a stable state. For normal operation, the wearable clothing product requires wireless communication between the mobile terminal and the product. The communication status on the mobile terminal determines whether the product is functioning correctly. The distance between location A and location B should be at least 25m. This test obtains the radiation intensity of wearable clothing products and determines whether the radiation safety test results of the wearable clothing products meet a preset safety test result.

[0123] Step S7 includes the following steps:

[0124] S71. If the number of failures in the circuit safety test of an electronic device is greater than zero, the wearable clothing product shall be deemed a non-compliant product.

[0125] S71. If the number of failures in the circuit safety test of electronic devices is zero, then a comprehensive evaluation should be conducted by combining the durability and comfort ratio of wearable clothing products.

[0126] In summary, by utilizing the above-mentioned technical solution of the present invention, through the independent testing of clothing and electronic products in wearable clothing products, and the calculation of the relationship between the two, multi-faceted and in-depth testing can be achieved, thereby ensuring accurate analysis of the performance of various aspects of wearable products and solving the safety hazards caused by heat radiation from clothing wearable products due to their close contact with the human body.

[0127] By introducing X-ray scanning to examine wearable clothing products, clear and intuitive images of clothing and electronic devices are generated. Preprocessing and an improved watershed segmentation algorithm are used to denoise and enhance the X-ray images, enabling precise segmentation of clothing and electronic devices and avoiding oversegmentation. Simultaneously, an edge detection algorithm is introduced to calculate the area of ​​clothing and electronic devices within the image, determining their proportions. Based on these proportions, a preliminary analysis of the comfort and safety of the wearable clothing product can be conducted. Furthermore, by acquiring and analyzing parameters such as the material and weaving process of the clothing, the comprehensiveness of the testing for wearable clothing products is increased. Finally, by testing the safety of various aspects of the electronic devices, a comprehensive testing and evaluation of wearable clothing products is achieved, ensuring the safe and stable use of these products.

[0128] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart security detection method for wearable clothing products, characterized in that, The method includes the following steps: S1. Lay out the wearable clothing product flat and perform an X-ray scan to obtain an X-ray image; The wearable products in the clothing category include apparel and electronic devices; S2. The X-ray image is preprocessed and segmented using an improved watershed segmentation algorithm to obtain a segmented image; S3. Use edge detection algorithms to divide the outlines of clothing and electronic devices and calculate their areas; S3 includes the following steps: S31. Remove discontinuous point interference from the segmented image after dilation and erosion processing; S32. Determine the pixel coordinates of the clothing boundary in the segmented image and obtain the outline of the corresponding clothing boundary coordinate points; S33, Calculate the pixel area of ​​the clothing; S34. Perform digital binarization on the segmented image and topological analysis on the digitally binarized image; S35. Determine the pixel coordinates of the electronic device boundary and obtain the outline of the corresponding electronic device boundary coordinate points. S36. Calculate the pixel area of ​​the electronic device; The pixel area of ​​the calculated clothing and the pixel area of ​​the calculated electronic device are both calculated using Green's formula to calculate the pixel area enclosed by the center point. S4. Obtain the ratio between the area of ​​electronic devices and the area of ​​clothing to obtain the comfort ratio; S5. Obtain the parameter information of the clothing and predict its durability through big data analysis; S6. Perform circuit safety testing on electronic equipment; S7. A comprehensive safety performance evaluation is derived by combining various test indicators.

2. The intelligent security detection method for wearable clothing products according to claim 1, characterized in that, The preprocessing of the X-ray image and segmentation using the improved watershed segmentation algorithm includes the following steps: S21. Eliminate noise from the X-ray image and perform morphological opening and closing reconstruction filtering to obtain an initial image; S22. Perform morphological gradient operation on the initial image to obtain a gradient image; S23. The initial image is labeled using the maximum entropy thresholding method to obtain a labeled image; S24. The gradient image is corrected using the labeled image by the minimum value calibration method; S25. Perform watershed segmentation on the corrected image to obtain a segmented image.

3. The intelligent security detection method for wearable clothing products according to claim 2, characterized in that, The process of eliminating noise from the X-ray image and performing morphological opening and closing reconstruction filtering to obtain the initial image includes the following steps: S211. Perform open reconstruction on the X-ray image, expressed as: ； in, This represents the opening reconstruction of an X-ray image with geodesic erosion as the initial image for dilatation reconstruction. Indicates the initial image. Indicates geodesic corrosion. This represents the dilation reconstruction of the mask image by the initial image; S212. Perform closed-loop reconstruction on the X-ray image, expressed as: ； in, This represents the closed reconstruction of X-ray images using geodesic dilatation as the initial image for dilatation reconstruction. Indicates the initial image. Indicates geodesic dilatation. This represents the erosion reconstruction of the mask image by the initial image.

4. The intelligent security detection method for wearable clothing products according to claim 2, characterized in that, The reconstructed image is labeled using the maximum entropy thresholding method to obtain a labeled image, including the following steps: S231. The maximum entropy thresholding method is used to select the optimal threshold based on the statistical characteristics of the initial image, and the formula for calculating the optimal threshold is: ； Where L represents the number of gray levels in the initial image, and the gray levels are divided into background class A and target class B. A represents the background class and takes the value A = (0, 1, ..., t), and B represents the target class and takes the value B = (t+1, t+2, ..., L-1). This represents the average gray value of class A. This represents the average gray value of class B. This represents the total average gray level of the image; S232. Remove the noise and pseudo-minimum values ​​formed by image texture in the initial image after gradient operation to obtain the labeled image.

5. The intelligent security detection method for wearable clothing products according to claim 1, characterized in that, The process of obtaining clothing parameters and predicting clothing durability through big data analysis includes the following steps: S51. Obtain the parameter information of the clothing, including the material and weaving process; S52. Input the clothing parameter information into the big data platform; S53. The big data platform compares the parameter information with known clothing parameters in the database to find N sets of closest lifespans. S54. Calculate the durability index of the clothing by averaging the N groups of closest parameters.

6. The intelligent security detection method for wearable clothing products according to claim 1, characterized in that, The circuit safety test of the electronic device includes the following steps: S61. Detect the operating temperature of the electronic device; S62. Detect the operating power of the electronic device; S63. Conduct safety inspections on transformers using transformer safety inspection standards; S64. Conduct information security testing using information security testing standards; S65. Conduct safety testing on overload protection circuits using overload protection circuit safety testing standards; S66. Conduct safety testing on the battery using battery safety testing standards; S67. Conduct radiation safety testing using radiation safety testing standards.

7. The intelligent security detection method for wearable clothing products according to claim 1, characterized in that, The comprehensive safety performance evaluation, derived by combining various testing indicators, includes the following steps: S71. If the number of failures in the circuit safety test of an electronic device is greater than zero, the wearable clothing product shall be deemed a non-compliant product. S71. If the number of failures in the circuit safety test of electronic devices is zero, then a comprehensive evaluation should be conducted by combining the durability and comfort ratio of wearable clothing products.

Images