A deep learning-based detection method for wearing cut-resistant gloves

Abstract

The embodiment of the present invention discloses a method for detecting the wearing of anti-cut gloves based on deep learning, which relates to the field of image recognition technology. The method comprises: collecting images of the hand area of ​​a worker when operating a machine; inputting the images of the hand area of ​​the worker when operating the machine into a pre-trained neural network model, and obtaining the detection result of the worker wearing anti-cut gloves output by the neural network model; wherein the neural network model is pre-trained by a large number of hand image samples wearing and not wearing anti-cut gloves. The present invention can detect the wearing of anti-cut gloves by workers in an intelligent, automatic, accurate and all-weather manner through a neural network model, has the advantage of high detection accuracy, and effectively saves human resource costs.

Description

Technical Field

[0001] The present invention belongs to the technical field of image recognition, and in particular relates to a method for detecting the wearing of cut-resistant gloves based on deep learning. Background Art

[0002] Anti-cut gloves are a type of metal gloves that are woven from countless small rings. They are used to protect hands from being cut when using cutting machinery in industrial scenarios. However, in actual operations, wearing anti-cut gloves for a long time can cause discomfort to the hands and even extra fatigue. Therefore, some workers violate the rules and do not wear anti-cut gloves. Once a safety accident occurs, the consequences are generally serious, ranging from injuries to disability or death.

[0003] At present, the wearing of anti-cut gloves mainly depends on the staff's self-consciousness. Through some safety accident training, the staff can fully understand the hazards of not wearing anti-cut gloves, so that the staff can consciously wear anti-cut gloves at work. However, due to the lack of supervision at all times, there are still many staff who do not wear gloves. To solve this problem, the factory arranges manual supervision to promptly discover and urge staff who are not wearing gloves to wear gloves as soon as possible. However, manual supervision requires additional human resource costs, and it is difficult to supervise an area for a long time, which poses certain safety hazards. Summary of the invention

[0004] In view of this, the embodiment of the present invention provides a method for detecting the wearing of anti-cutting gloves based on deep learning, which is used to solve the problem that the existing supervision of the wearing of anti-cutting gloves mainly relies on manual work and the supervision is not in place, resulting in potential safety hazards. The present invention can detect the wearing of anti-cutting gloves by workers in an intelligent, automatic, accurate and all-weather manner through a neural network model, and has the advantages of high detection efficiency and accuracy, effectively saving human resource costs.

[0005] The embodiment of the present invention provides a method for detecting the wearing of cut-resistant gloves based on deep learning, comprising:

[0006] Collect images of the hand area of ​​workers when they operate machines;

[0007] The hand area image of the worker when operating the machine is input into a pre-trained neural network model, and the detection result of the worker wearing anti-cut gloves is obtained by the output of the neural network model; wherein the neural network model is pre-trained by a large number of hand image samples wearing and not wearing anti-cut gloves.

[0008] In an optional embodiment, the collecting of hand area images of a worker when operating a machine includes:

[0009] Collect hand area images of a worker when operating a machine from multiple preset angles to obtain a hand area image set;

[0010] The hand area image of the worker when operating the machine is input into a pre-trained neural network model to obtain the detection result of the worker wearing anti-cut gloves output by the neural network model, including:

[0011] Inputting all the hand area images in the hand area image set into a plurality of pre-trained neural network models respectively, and obtaining detection results of the staff wearing anti-cut gloves corresponding to each preset angle output by each neural network model;

[0012] For each preset angle, the comprehensive detection result of the staff wearing anti-cut gloves corresponding to the preset angle is determined according to the detection result of the staff wearing anti-cut gloves corresponding to the preset angle output by each neural network model.

[0013] In an optional embodiment, for each preset angle, according to the detection result of the worker wearing the anti-cutting gloves corresponding to the preset angle output by each neural network model, determining the comprehensive detection result of the worker wearing the anti-cutting gloves corresponding to the preset angle includes:

[0014] Calculate the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to each preset angle according to the first formula;

[0015] For each preset angle, determine whether the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to the current preset angle is equal to 0; if so, it is determined that the worker corresponding to the current preset angle is wearing the anti-cutting gloves correctly; otherwise, it is determined that the worker corresponding to the current preset angle is wearing the anti-cutting gloves incorrectly;

[0016] Among them, the first formula is:

[0017]

[0018] In the first formula, H(i) represents the comprehensive detection value of the staff wearing anti-cut gloves corresponding to the i-th preset angle; C(i_a) represents the detection result of the staff wearing anti-cut gloves corresponding to the i-th preset angle output by the a-th neural network model, if the detection result is that the anti-cut gloves are worn correctly, C(i_a)=1, if the detection result is that the anti-cut gloves are worn incorrectly, C(i_a)=0; a=1,2,…,m; m represents the total number of pre-trained neural network models; i=1,2,…,n; n represents the total number of preset angles.

[0019] In an optional embodiment, after calculating the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to each preset angle according to the first formula, it also includes:

[0020] Calculate the incorrect statistical values ​​of the current staff wearing anti-cutting gloves corresponding to all preset angles;

[0021] Determine whether the statistical value of the current staff wearing the anti-cutting gloves incorrectly corresponding to all preset angles is greater than a preset threshold value, if so, determine that the final detection result of the current staff wearing the anti-cutting gloves is incorrect wearing, otherwise, determine that the final detection result of the current staff wearing the anti-cutting gloves is correct wearing;

[0022] Output the final test results of the current staff wearing anti-cut gloves.

[0023] In an optional embodiment, after calculating the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to each preset angle according to the first formula, it also includes:

[0024] Generate an alarm control value for the staff wearing anti-cutting gloves corresponding to each preset angle according to the comprehensive detection value of the staff wearing anti-cutting gloves corresponding to each preset angle;

[0025] The alarm situation is controlled according to the alarm control value of the staff wearing anti-cutting gloves corresponding to each preset angle.

[0026] In an optional embodiment, the alarm condition includes: a preset light display color and brightness status of the alarm light.

[0027] In an optional embodiment, the generating of the alarm control value of the worker wearing the anti-cutting gloves corresponding to each preset angle according to the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to each preset angle includes:

[0028] Identify the hand region in each hand region image in the hand region image set by a preset image recognition method, and determine the total number of pixels occupied by the identified hand region;

[0029] According to the comprehensive detection value of the staff wearing anti-cutting gloves corresponding to each preset angle and the total number of pixels occupied by the hand area in the hand area image corresponding to each preset angle, an alarm control value of the staff wearing anti-cutting gloves corresponding to each preset angle is generated based on the second formula;

[0030] Wherein, the second formula is:

[0031]

[0032] In the second formula, G(i) represents the alarm control value of the staff wearing anti-cutting gloves corresponding to the i-th preset angle; S(i) represents the total number of pixels occupied by the hand area in the hand area image corresponding to the i-th preset angle; S0 represents the total number of all pixels in the hand area image corresponding to the i-th preset angle;

[0033] The alarm control value of the worker wearing anti-cutting gloves corresponding to each preset angle controls the alarm situation, including:

[0034] Calculate the current alarm light color control value and the alarm light brightness control value according to the third formula;

[0035] According to the correspondence between the preset light color control value and the preset color, and the correspondence between the preset light brightness control value and the preset brightness, the color and brightness of the alarm light are determined and the alarm light is controlled accordingly;

[0036] Wherein, the third formula is:

[0037]

[0038] In the third formula, J represents the alarm light color control value; E represents the alarm light brightness control value; E M Indicates the maximum brightness value of the alarm system light; It means to substitute the value of i from 1 to n into the brackets to get the maximum value in the brackets; Indicates rounding up; It means that the value of i is substituted from 1 to n into the brackets to obtain the maximum value of i in the brackets. If there are multiple i values, the i value corresponding to the maximum value in S(i) is taken; k is the preset number of controllable light colors.

[0039] In an alternative embodiment, k=7.

[0040] The present invention provides a method for detecting the wearing of anti-cutting gloves based on deep learning. First, the hand area image of the worker when operating the machine is collected, and then the hand area image is input into a trained neural network model to obtain the detection result of the worker wearing the anti-cutting gloves. The present invention can detect the behavior of the worker not wearing anti-cutting gloves intelligently, automatically, accurately and all-weather through the neural network model, and has the advantages of high detection efficiency and accuracy, effectively saving human resource costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] 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.

[0042] Figure 1 A flow chart of a method for detecting the wearing of cut-resistant gloves based on deep learning provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0044] It should be clear that the described embodiments are only some embodiments of the present invention, not all 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.

[0045] Figure 1 A flow chart of a method for detecting the wearing of cut-resistant gloves based on deep learning provided by an embodiment of the present invention. Figure 1 The method comprises the following steps S101-S102:

[0046] S101: Collecting images of the hand area of ​​a worker when operating a machine.

[0047] S102: Inputting the hand area image of the worker when operating the machine into a pre-trained neural network model, and obtaining the detection result of the worker wearing anti-cut gloves output by the neural network model.

[0048] The neural network model is pre-trained by a large number of hand image samples with and without anti-cut gloves.

[0049] The beneficial effects of the above technical solution are as follows: the embodiment of the present invention provides a method for detecting the wearing of anti-cut gloves based on deep learning, which first collects the hand area image of the staff when operating the machine, and then inputs the hand area image into the trained neural network model to obtain the detection result of the staff wearing anti-cut gloves. The present invention can use a neural network model based on deep learning, that is, a detection method based on deep learning, to intelligently, automatically, accurately, and all-weather detect the wearing of anti-cut gloves by the staff, allowing the system to replace manual labor, saving human resource costs, and also has the advantages of high detection efficiency and accuracy.

[0050] As an optional embodiment, the step S101 may include: collecting hand area images of a worker when operating a machine from a plurality of different preset angles to obtain a hand area image set.

[0051] The step S102 may include the following steps S201-S202:

[0052] S201: inputting all the hand area images in the hand area image set into a plurality of pre-trained neural network models respectively, and obtaining detection results of the staff wearing anti-cut gloves corresponding to each preset angle output by each neural network model;

[0053] S202: For each preset angle, according to the detection result of the staff wearing anti-cut gloves corresponding to the preset angle output by each neural network model, determine the comprehensive detection result of the staff wearing anti-cut gloves corresponding to the preset angle.

[0054] The beneficial effects of the above technical solution are: collecting images of the hand area of ​​the staff operating the machine from various preset angles, and then sending the images to each trained neural network model for detection and recognition, thereby preventing the hand area images collected from certain angles from being blocked, resulting in inaccurate detection results. It can also analyze the situation of the staff wearing anti-cut gloves from various angles, further improving the accuracy of anti-cut gloves wearing detection.

[0055] As an optional embodiment, the step S202 may include the following steps S2021-S2024:

[0056] S2021: Calculate the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to each preset angle according to the first formula;

[0057] Among them, the first formula is:

[0058]

[0059] In the first formula, H(i) represents the comprehensive detection value of the staff wearing anti-cut gloves corresponding to the i-th preset angle; C(i_a) represents the detection result of the staff wearing anti-cut gloves corresponding to the i-th preset angle output by the a-th neural network model, if the detection result is that the anti-cut gloves are worn correctly, C(i_a)=1, if the detection result is that the anti-cut gloves are worn incorrectly, C(i_a)=0; a=1,2,…,m; m represents the total number of pre-trained neural network models; i=1,2,…,n; n represents the total number of preset angles.

[0060] S2022: For each preset angle, determine whether the comprehensive detection value of the staff wearing anti-cut gloves corresponding to the current preset angle is equal to 0; if so, execute S2023, otherwise, execute S2024.

[0061] S2023: Determine whether the worker corresponding to the current preset angle is wearing the cut-resistant gloves correctly.

[0062] S2024: determining that the worker corresponding to the current preset angle is wearing cut-resistant gloves incorrectly;

[0063] The beneficial effect of the above technical solution is: using the first formula (1) to calculate the comprehensive model detection value of each hand area image of the staff member, so as to quickly screen out the hand areas with problems, which is convenient for subsequent control.

[0064] As an optional embodiment, after calculating the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to each preset angle according to the first formula, the following steps S301-S305 may also be included:

[0065] S301: Calculating the statistical values ​​of incorrect wearing of cut-resistant gloves by the current staff corresponding to all preset angles;

[0066] In this step, obtain The value of is used as the statistical value of incorrect wearing of cut-resistant gloves by the current staff corresponding to all preset angles.

[0067] S302: Determine whether the statistical value of incorrect wearing of the anti-cutting gloves by the current staff corresponding to all preset angles is greater than a preset threshold, if so, execute step S303, otherwise execute step S304;

[0068] Preferably, the preset threshold is greater than 0 and does not exceed n; specifically, the smaller the preset threshold is, the higher the accuracy of the detection result of the present invention is.

[0069] S303: Determine that the final test result of the current worker wearing the cut-resistant gloves is incorrect wearing, and then execute step S305;

[0070] S304: Determine that the final test result of the current worker wearing the cut-resistant gloves is that the gloves are worn correctly, and then execute step S305;

[0071] S305: Output the final detection result of the current staff wearing the anti-cut gloves.

[0072] The beneficial effect of the above technical solution is: according to the hand area images taken at each preset angle, the pre-trained neural network models are respectively input to obtain the comprehensive detection value of the staff wearing anti-cut gloves corresponding to each preset angle, and then the final detection result of the current staff wearing anti-cut gloves is determined by multiple comprehensive detection values, thereby effectively improving the accuracy of detection.

[0073] As an optional embodiment, after step S102, the following steps S401-S402 may also be included:

[0074] S401: generating an alarm control value for the staff wearing anti-cutting gloves corresponding to each preset angle according to the comprehensive detection value of the staff wearing anti-cutting gloves corresponding to each preset angle;

[0075] S402: Control the current alarm situation according to the alarm control value of the staff wearing anti-cutting gloves corresponding to each preset angle.

[0076] Preferably, the alarm condition includes: a preset light display color and brightness status of the alarm light.

[0077] The beneficial effect of the above technical solution is: after obtaining the detection results of each staff member wearing anti-cut gloves, the light display color and brightness status of the pre-set alarm light are controlled according to the alarm control value of the staff wearing anti-cut gloves corresponding to each preset angle, thereby achieving the purpose of visual automatic alarm, improving the intelligence level of the system, and further improving the level of anti-cut gloves wearing detection management.

[0078] As an optional embodiment, the step S401 may be performed as follows: steps S501-S502:

[0079] S501: identifying a hand region in each hand region image in the hand region image set by a preset image recognition method, and determining the total number of pixels occupied by the identified hand region;

[0080] S502: generating an alarm control value for the worker wearing anti-cutting gloves corresponding to each preset angle based on a second formula according to the comprehensive detection value of the worker wearing anti-cutting gloves corresponding to each preset angle and the total number of pixels occupied by the hand area in the hand area image corresponding to each preset angle;

[0081] Wherein, the second formula is:

[0082]

[0083] In the second formula, G(i) represents the alarm control value of the staff wearing anti-cutting gloves corresponding to the i-th preset angle; S(i) represents the total number of pixels occupied by the hand area in the hand area image corresponding to the i-th preset angle; S0 represents the total number of all pixels in the hand area image corresponding to the i-th preset angle;

[0084] The step S402 may include the following steps S601-S602:

[0085] S601: Calculate the current alarm light color control value and the alarm light brightness control value according to the third formula;

[0086] Wherein, the third formula is:

[0087]

[0088] In the third formula, J represents the alarm light color control value; E represents the alarm light brightness control value; EM represents the maximum brightness value of the alarm system light; It means to substitute the value of i from 1 to n into the brackets to get the maximum value in the brackets; Indicates rounding up; It means that the value of i is substituted from 1 to n into the brackets to obtain the maximum value of i in the brackets. If there are multiple i values, the i value corresponding to the maximum value in S(i) is taken; k is the preset number of controllable light colors.

[0089] S602: According to the correspondence between the preset light color control value and the preset color, and the correspondence between the preset light brightness control value and the preset brightness, the color and brightness of the alarm light are determined and the alarm light is controlled accordingly.

[0090] Preferably, the value of k is 7, and the alarm light color control values ​​J are 1, 2, 3, 4, 5, 6, and 7, which correspond to 7 preset colors: red, orange, yellow, green, cyan, blue, and purple, respectively.

[0091] The beneficial effect of the above technical solution is: using the second formula (2) to generate the alarm control value of each hand area image of the staff member according to the comprehensive model detection value of each hand area image of the staff member, and combining with the third formula (3) to control the light color display and brightness display state of the alarm system according to the detected alarm control value, and then according to the light state and the alarm control value, the position where the anti-cutting gloves are not worn correctly can be located, thereby improving the analysis efficiency, reducing the analysis errors, and effectively ensuring that the staff members wear the gloves correctly, thereby ensuring the safety of the staff members.

[0092] It can be seen from the contents of the above embodiments that the hand area image of the staff is obtained; a pre-trained neural network model is obtained, and the hand area image is input into the neural network model; and finally an alarm is issued according to the model detection result; wherein, obtaining the hand area image of the staff includes photographing at multiple angles to obtain multiple hand area images of the staff, and then inputting the multiple hand area images of the staff into multiple neural network models, and issuing alarms of different degrees according to the detection results of multiple models, thereby effectively improving the efficiency of the anti-cutting gloves wearing detection, and being able to issue automatic alarms all day and all the time according to the detection results.

[0093] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 An apparatus for the method specified in one or more blocks.

[0094] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 The method specified in one or more boxes.

[0095] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the method are specified in one or more boxes.

[0096] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the claims and their

[0097] The present invention also intends to include such changes and modifications if they are within the scope of equivalent technologies. The above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A deep learning-based method for detecting the wearing of cut-resistant gloves, characterized in that: include: Collect images of the hand area of ​​workers when they operate machines; Inputting the hand area image of the worker when operating the machine into a pre-trained neural network model, and obtaining the detection result of the worker wearing anti-cut gloves output by the neural network model; wherein the neural network model is pre-trained by a large number of hand image samples wearing and not wearing anti-cut gloves; The collecting of hand area images of the staff when operating the machine includes: Collect hand area images of a worker when operating a machine from multiple preset angles to obtain a hand area image set; The hand area image of the worker when operating the machine is input into a pre-trained neural network model to obtain the detection result of the worker wearing anti-cut gloves output by the neural network model, including: Inputting all the hand area images in the hand area image set into a plurality of pre-trained neural network models respectively, and obtaining detection results of the staff wearing anti-cut gloves corresponding to each preset angle output by each neural network model; For each preset angle, according to the detection result of the worker wearing the anti-cutting gloves corresponding to the preset angle output by each neural network model, a comprehensive detection result of the worker wearing the anti-cutting gloves corresponding to the preset angle is determined; Wherein, for each preset angle, according to the detection result of the worker wearing the anti-cutting gloves corresponding to the preset angle output by each neural network model, the comprehensive detection result of the worker wearing the anti-cutting gloves corresponding to the preset angle is determined, including: Calculate the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to each preset angle according to the first formula; For each preset angle, determine whether the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to the current preset angle is equal to 0; if so, it is determined that the worker corresponding to the current preset angle is wearing the anti-cutting gloves correctly; otherwise, it is determined that the worker corresponding to the current preset angle is wearing the anti-cutting gloves incorrectly; Among them, the first formula is: In the first formula, H(i) represents the comprehensive detection value of the staff wearing anti-cutting gloves corresponding to the i-th preset angle; C(i_a) represents the detection result of the staff wearing anti-cutting gloves corresponding to the i-th preset angle output by the a-th neural network model, if the detection result is that the anti-cutting gloves are worn correctly, C(i_a)=1, if the detection result is that the anti-cutting gloves are worn incorrectly, C(i_a)=0; a=1,2,…,m; m represents the total number of pre-trained neural network models; i=1,2,…,n; n represents the total number of preset angles; Wherein, after calculating the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to each preset angle according to the first formula, it also includes: Generate an alarm control value for the staff wearing anti-cutting gloves corresponding to each preset angle according to the comprehensive detection value of the staff wearing anti-cutting gloves corresponding to each preset angle; According to the alarm control value of the staff wearing anti-cutting gloves corresponding to each preset angle, control the alarm situation; Wherein, the alarm condition includes: the light display color and brightness status of the preset alarm light; The method of generating the alarm control value of the worker wearing the anti-cutting gloves corresponding to each preset angle according to the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to each preset angle includes: Identify the hand region in each hand region image in the hand region image set by a preset image recognition method, and determine the total number of pixels occupied by the identified hand region; According to the comprehensive detection value of the staff wearing anti-cutting gloves corresponding to each preset angle and the total number of pixels occupied by the hand area in the hand area image corresponding to each preset angle, an alarm control value of the staff wearing anti-cutting gloves corresponding to each preset angle is generated based on the second formula; Wherein, the second formula is: In the second formula, G(i) represents the alarm control value of the staff wearing anti-cutting gloves corresponding to the i-th preset angle; S(i) represents the total number of pixels occupied by the hand area in the hand area image corresponding to the i-th preset angle; S0 represents the total number of all pixels in the hand area image corresponding to the i-th preset angle; The alarm control value of the worker wearing anti-cutting gloves corresponding to each preset angle controls the alarm situation, including: Calculate the current alarm light color control value and the alarm light brightness control value according to the third formula; According to the correspondence between the preset light color control value and the preset color, and the correspondence between the preset light brightness control value and the preset brightness, the color and brightness of the alarm light are determined and the alarm light is controlled accordingly; Wherein, the third formula is: In the third formula, J represents the alarm light color control value; E represents the alarm light brightness control value; E M Indicates the maximum brightness value of the alarm system light; It means to substitute the value of i from 1 to n into the brackets to get the maximum value in the brackets; Indicates rounding up; It means that the value of i is substituted from 1 to n into the brackets to obtain the maximum value of i in the brackets. If there are multiple i values, the i value corresponding to the maximum value in S(i) is taken; k is the preset number of controllable light colors.

2. The deep learning-based cut-resistant glove wearing detection method according to claim 1, characterized in that: After calculating the comprehensive detection value of the worker wearing the anti-cutting gloves corresponding to each preset angle according to the first formula, the method further includes: Calculate the incorrect statistical values ​​of the current staff wearing anti-cutting gloves corresponding to all preset angles; Determine whether the statistical value of the current staff wearing the anti-cutting gloves incorrectly corresponding to all preset angles is greater than a preset threshold value, if so, determine that the final detection result of the current staff wearing the anti-cutting gloves is incorrect wearing, otherwise, determine that the final detection result of the current staff wearing the anti-cutting gloves is correct wearing; Output the final test results of the current staff wearing anti-cut gloves.

3. The deep learning-based cut-resistant glove wearing detection method according to claim 1, characterized in that: k＝7。

Images