Human body static electricity elimination detection method and system based on attention mechanism and fully convolutional neural network

Abstract

The present invention discloses a method and system for detecting human static electricity elimination based on an attention mechanism and a fully convolutional neural network. The method includes: constructing a model based on an attention mechanism and a fully convolutional neural network; constructing a clothing material recognition model and a static electricity elimination detection model based on the model; collecting images of people to be identified, and identifying the material of the clothing worn by the people based on the clothing material recognition model; setting the detection time according to different clothing materials; removing static electricity through an electrostatic ball, and detecting whether the static electricity has been eliminated based on the static electricity elimination detection model and the detection time. The present invention can detect whether there is residual static electricity in the body of the people entering and leaving the venue, and can remind people to fully remove the static electricity to prevent the residual static electricity from causing harm to the human body, and improve the detection efficiency when there are many people to be detected.

Description

Technical Field

[0001] The present invention belongs to the technical field of static electricity detection, and in particular relates to a method and system for eliminating static electricity from the human body based on an attention mechanism and a fully convolutional neural network. Background Art

[0002] Static electricity does not refer to static electricity, but refers to the formation of static electricity when charges gather on an object or surface. When an object with static electricity touches an object with zero potential (grounded object) or an object with a potential difference with it, charge transfer will occur, which is the spark discharge phenomenon we see in daily life. For example, in winter when the weather is dry, your hair becomes more and more messy when you comb it, you feel a tingling sensation of electric shock when you touch others or metal conductors, and there is a crackling sound and flash when you take off your clothes at night. These are all phenomena of static electricity on the human body.

[0003] Static electricity on the human body is static electricity attached to the human body due to the friction between clothes and other things on the human body. Static electricity is generated because the nucleus of an atom does not have enough attraction to the outer electrons, so it loses electrons under the action of friction or other factors, causing the friction object to be negatively charged (i.e., the one that gains electrons is negatively charged, and the one that loses electrons is positively charged). In the case of good insulation performance of the friction object, these charges cannot be lost and will accumulate. And because the capacitance of the insulating material is extremely poor, the voltage is very high even though the charge is not large. Wearing clothes made of chemical fibers is more likely to generate static electricity, while cotton clothes generate less. And because a dry environment is more conducive to the transfer and accumulation of charges, people will feel that the static electricity on their bodies is larger in winter. Under different humidity conditions, the static electricity potential generated by human activities is different. In dry seasons, human static electricity can reach several thousand volts or even tens of thousands of volts. Experiments have shown that the human body does not feel uncomfortable when the static electricity voltage is 50,000 volts, and there is no life-threatening situation when carrying 120,000 volts of high-voltage static electricity. However, static discharge will also generate an electromagnetic field around it, although the duration is short, the intensity is very high. Therefore, in order to eliminate the potential dangers that static electricity may cause to the human body, people have developed a variety of methods to eliminate static electricity.

[0004] Moreover, in some occasions with a large number of people, the elimination of static electricity mostly requires people to touch the static electricity elimination equipment with their hands, and use the equipment to conduct the static electricity carried by the human body into the ground to achieve the effect of eliminating static electricity. However, if people do not contact the static electricity elimination equipment for a long enough time or do not contact it completely, the static electricity carried by the human body cannot be completely eliminated. Therefore, it is necessary to detect it while removing static electricity to determine whether static electricity still exists in the human body. Moreover, the residual static electricity can easily cause harm to the human body in certain specific occasions. Summary of the invention

[0005] The purpose of the present invention is to provide a human body static electricity elimination detection method based on attention mechanism and fully convolutional neural network. Based on the attention mechanism and fully convolutional neural network, the camera can detect whether the entrant has a touching action and whether the hand of the entrant is in full contact with the electrostatic ball. If it continues for a period of time, it will display the OK state, indicating that the static electricity is eliminated. Otherwise, it will display NO, indicating that the static electricity of the person has not been eliminated, so as to solve the problem in the prior art that it is impossible to determine whether there is still static electricity in the body of the entrant due to the lack of static electricity detection.

[0006] The technical solution to achieve the purpose of the present invention is: a method for detecting human body static electricity elimination based on an attention mechanism and a fully convolutional neural network, the method comprising the following steps:

[0007] Step 1: Build a model based on attention mechanism and fully convolutional neural network;

[0008] Step 2, constructing a clothing material recognition model and a static electricity elimination detection model based on the model;

[0009] Step 3, collecting an image of the person to be identified, and identifying the material of the clothing worn by the person based on the clothing material recognition model;

[0010] Step 4: Set the detection time according to different clothing materials;

[0011] Step 5, removing static electricity by using an electrostatic ball, and detecting whether the static electricity has been completely eliminated based on the static electricity elimination detection model and the detection time.

[0012] Furthermore, the model based on the attention mechanism and the fully convolutional neural network described in step 1 includes: a 224×224×3 input layer, and a spatial attention mechanism channel and a temporal attention mechanism channel connected in parallel thereafter;

[0013] The spatial attention mechanism channel includes a 55×55×48 convolutional layer, a 27×27×48 pooling layer, a 27×27×128 convolutional layer, a 13×13×128 pooling layer, a 13×13×192 convolutional layer, a 13×13×192 convolutional layer, a 13×13×128 convolutional layer, a 6×6×128 pooling layer, a spatial attention module, a 13×13×128 convolutional layer and a 1×1×N convolutional layer, wherein N is the preset number of types to be identified;

[0014] The temporal attention mechanism channel includes a 55×55×48 convolutional layer, a 27×27×48 pooling layer, a 27×27×128 convolutional layer, a 13×13×128 pooling layer, a 13×13×192 convolutional layer, a 13×13×192 convolutional layer, a 13×13×128 convolutional layer, a 6×6×128 pooling layer, a temporal attention module, and a 1×1×N convolutional layer.

[0015] Each value in the feature map output by the two channels is processed by softmax to obtain the candidate probability distribution P of each category based on the temporal attention module or the spatial attention module. t or P s , where the category corresponding to the maximum probability is taken as the final recognized category.

[0016] Furthermore, the spatial attention module includes a long short-term memory network LSTM, a 6×6×128 convolutional layer with a step size of 1, a first fully convolutional network unit, and a Normal layer arranged in sequence; wherein the first fully convolutional network unit includes a 1×1×N fully convolutional layer, using tanh as an activation function.

[0017] Furthermore, the temporal attention module includes a long short-term memory network LSTM, a 13×13×128 convolutional layer with a step size of 1, a second full convolutional network unit, and a Normal layer arranged in sequence; wherein the second full convolutional network unit includes a 7×7×64 full convolutional layer with a step size of 2, a 5×5×128 full convolutional layer with a step size of 2, a 3×3×128 full convolutional layer with a step size of 1, and a 1×1×N full convolutional layer arranged in sequence, using ReLU as the activation function.

[0018] Furthermore, step 2 constructs a clothing material recognition model and a static electricity elimination detection model based on the model, specifically including:

[0019] Step 2-1, collect images of people wearing different clothing materials to form a data set, and divide it into a training set, a test set, and a validation set;

[0020] Step 2-2, using the training set to train a model based on an attention mechanism and a fully convolutional neural network to obtain a clothing material recognition model;

[0021] Step 2-3, collecting several images of hands touching the electrostatic ball to form a data set, and dividing it into a training set, a test set, and a validation set;

[0022] Step 2-4, using the training set to train a model based on an attention mechanism and a fully convolutional neural network to obtain a static electricity elimination detection model.

[0023] Furthermore, step 4 sets the detection time according to different clothing materials, specifically including:

[0024] Step 4-1, classify clothing materials into two categories, classify materials with chemical fiber content below a preset threshold into the first category, and the others into the second category;

[0025] Step 4-2, setting the detection time for different types of clothing materials, wherein the detection time corresponding to the first type of clothing material is shorter than that corresponding to the second type of clothing material.

[0026] Furthermore, in step 5, removing static electricity by using an electrostatic ball and detecting whether the static electricity is completely eliminated based on the model and the detection time specifically include:

[0027] The static electricity ball image is continuously collected in real time, and the static electricity elimination detection model is used to identify whether the person's hand touches the static electricity ball, and whether the contact time meets the corresponding detection time. If so, it is determined that the static electricity elimination is completed, otherwise it is determined that the static electricity elimination is not completed.

[0028] A human body static electricity elimination detection system based on attention mechanism and full convolutional neural network, the system comprising:

[0029] A person identification module is used to identify the material of the clothes worn by a person based on a clothing material identification model;

[0030] A time value setting module, used to set different detection time lengths according to the identified clothing material;

[0031] A static electricity elimination detection module is used to eliminate static electricity and detect whether the static electricity has been eliminated based on a static electricity elimination detection model and a detection time;

[0032] The detection prompt module is used to display the detection results of the static elimination detection module.

[0033] Furthermore, information is wirelessly transmitted between the personnel identification module and the time value setting module, between the time value setting module and the static electricity detection module, and between the static electricity detection module and the detection prompt module through a wireless mobile communication system.

[0034] Compared with the prior art, the present invention has the following significant advantages:

[0035] 1. The present invention enables the camera to accurately identify the hands of the entrants based on the attention mechanism, and uses the full convolutional neural network technology to enable the camera to detect whether the entrants have any touching actions and whether the hands of the entrants are in full contact with the electrostatic ball. If it continues for a period of time and displays an OK state, it means that the static electricity is eliminated. Otherwise, it will display NO, indicating that the static electricity of the person has not been eliminated. Compared with the prior art, it can detect whether there is residual static electricity in the body of the entrants, and can remind people to fully remove the static electricity to prevent the residual static electricity from causing harm to the human body.

[0036] 2. The present invention uses an attention mechanism to enable the camera to accurately identify the clothes worn by the entrants. When it is identified that the material of the clothes is the second category (chemical fiber), a static electricity detection time of 30 seconds will be set to ensure that the static electricity carried by the entrants can be fully removed. When the identified material of the clothes is the first category (pure cotton), a static electricity detection time of 15 seconds will be set. The static electricity detection time can be flexibly set according to the different amounts of static electricity carried by different entrants, so that the present invention can improve the detection efficiency when there are more people to be detected.

[0037] 3. This paper proposes a new model based on attention mechanism and fully convolutional neural network, which can significantly improve the recognition accuracy.

[0038] The present invention is further described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of the structure of a human body static electricity elimination detection system based on an attention mechanism and a fully convolutional neural network in one embodiment.

[0040] Figure 2 Schematic diagram of a model based on an attention mechanism and a fully convolutional neural network in one embodiment.

[0041] Figure 3 Schematic diagram of the spatial attention module structure in one embodiment.

[0042] Figure 4 Schematic diagram of the structure of the temporal attention module in one embodiment.

[0043] Figure 5 The figure is a flow chart of human body static electricity elimination detection in one embodiment. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0045] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0046] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0047] A method for detecting human body static electricity elimination based on an attention mechanism and a fully convolutional neural network, the method comprising the following steps:

[0048] Step 1: Build a model based on attention mechanism and fully convolutional neural network;

[0049] Step 2, constructing a clothing material recognition model and a static electricity elimination detection model based on the model;

[0050] Step 3, collecting an image of the person to be identified, and identifying the material of the clothing worn by the person based on the clothing material recognition model;

[0051] Step 4: Set the detection time according to different clothing materials;

[0052] Step 5, removing static electricity by using an electrostatic ball, and detecting whether the static electricity has been completely eliminated based on the static electricity elimination detection model and the detection time.

[0053] Further, in one embodiment, in combination Figure 2 , the model based on the attention mechanism and the fully convolutional neural network described in step 1 includes: a 224×224×3 input layer, and a spatial attention mechanism channel and a temporal attention mechanism channel connected in parallel;

[0054] The spatial attention mechanism channel includes a 55×55×48 convolutional layer, a 27×27×48 pooling layer, a 27×27×128 convolutional layer, a 13×13×128 pooling layer, a 13×13×192 convolutional layer, a 13×13×192 convolutional layer, a 13×13×128 convolutional layer, a 6×6×128 pooling layer, a spatial attention module, a 13×13×128 convolutional layer and a 1×1×N convolutional layer, wherein N is the preset number of types to be identified;

[0055] The temporal attention mechanism channel includes a 55×55×48 convolutional layer, a 27×27×48 pooling layer, a 27×27×128 convolutional layer, a 13×13×128 pooling layer, a 13×13×192 convolutional layer, a 13×13×192 convolutional layer, a 13×13×128 convolutional layer, a 6×6×128 pooling layer, a temporal attention module, and a 1×1×N convolutional layer.

[0056] Each value in the feature map output by the two channels is processed by softmax to obtain the candidate probability distribution P of each category based on the temporal attention module or the spatial attention module. t or P s , where the category corresponding to the maximum probability is taken as the final recognized category.

[0057] Further, in one embodiment, in combination Figure 3 The spatial attention module includes a long short-term memory network LSTM, a 13×13×128 convolutional layer with a step size of 1, a first full convolutional network unit, and a Normal layer, which are arranged in sequence; wherein the first full convolutional network unit includes a 1×1×N full convolutional layer, using tanh as the activation function.

[0058] Further, in one embodiment, in combination Figure 4 The temporal attention module includes a long short-term memory network LSTM, a 6×6×128 convolutional layer with a step size of 1, a second full convolutional network unit, and a Normal layer arranged in sequence; wherein the second full convolutional network unit includes a 7×7×64 full convolutional layer with a step size of 2, a 5×5×128 full convolutional layer with a step size of 2, a 3×3×128 full convolutional layer with a step size of 1, and a 1×1×N full convolutional layer arranged in sequence, and ReLU is used as the activation function.

[0059] Furthermore, in one embodiment, step 2 of constructing a clothing material recognition model and a static electricity elimination detection model based on the model specifically includes:

[0060] Step 2-1, collect images of people wearing different clothing materials to form a data set, and divide it into a training set, a test set, and a validation set;

[0061] Step 2-2, using the training set to train a model based on an attention mechanism and a fully convolutional neural network to obtain a clothing material recognition model;

[0062] Step 2-3, collecting several images of hands touching the electrostatic ball to form a data set, and dividing it into a training set, a test set, and a validation set;

[0063] Step 2-4, using the training set to train a model based on an attention mechanism and a fully convolutional neural network to obtain a static electricity elimination detection model.

[0064] Further, in one embodiment, the step 4 of setting the detection time according to different clothing materials specifically includes:

[0065] Step 4-1, classify clothing materials into two categories, classify materials with chemical fiber content below a preset threshold into the first category, and the others into the second category;

[0066] Step 4-2, setting the detection time for different types of clothing materials, wherein the detection time corresponding to the first type of clothing material is shorter than that corresponding to the second type of clothing material.

[0067] Further, in one embodiment, the step 5 of removing static electricity by using an electrostatic ball and detecting whether the static electricity is completely eliminated based on the model and the detection time specifically includes:

[0068] The static electricity ball image is continuously collected in real time, and the static electricity elimination detection model is used to identify whether the person's hand touches the static electricity ball, and whether the contact time meets the corresponding detection time. If so, it is determined that the static electricity elimination is completed, otherwise it is determined that the static electricity elimination is not completed.

[0069] In one embodiment, in combination Figure 1 , a human body static electricity elimination detection system based on attention mechanism and full convolutional neural network, the system comprising:

[0070] A person identification module is used to identify the material of the clothes worn by a person based on a clothing material identification model;

[0071] A time value setting module, used to set different detection time lengths according to the identified clothing material;

[0072] A static electricity elimination detection module is used to eliminate static electricity and detect whether the static electricity has been eliminated based on a static electricity elimination detection model and a detection time;

[0073] The detection prompt module is used to display the detection results of the static elimination detection module.

[0074] Furthermore, in one embodiment, the wireless transmission of information between the personnel identification module and the time value setting module, between the time value setting module and the static electricity detection module, and between the static electricity detection module and the detection prompt module is performed through a wireless mobile communication system. The mobile communication system can be a cellular system, a cluster system, an AdHoc network system, a satellite communication system, a packet wireless network, a cordless telephone system, or a radio paging system.

[0075] Furthermore, in one of the embodiments, the wireless mobile communication system is a cellular system.

[0076] As a specific example, in one of the embodiments, the present invention is further described, based on the attention mechanism and the full convolutional neural network, the human body static electricity elimination detection method is combined with Figure 5 , including the following steps:

[0077] Step 1: Use the personnel identification module to identify the material of the clothes worn by the entrants: install a group of cameras at the entrance of the place where static electricity needs to be removed (the number of cameras is 5-10, and the specific number can be designed according to the flow of people at the entrance). This group of cameras can accurately capture the clothes worn by the entrants and identify the material of the clothes based on the attention mechanism and the fully convolutional neural network model. For some clothes with a relatively smooth and shiny surface, it can be determined that the material composition contains more chemical fibers. For some clothes with uneven and dull surfaces, it can be determined that the material does not contain chemical fibers or contains less chemical fibers. The attention mechanism used by the camera refers to the need to select a specific part in the visual area and then focus on it. It mainly includes two aspects: deciding which part of the input needs to be focused on and allocating limited information processing resources to the important part; the camera will select the clothes in the visual area for identification, and use the identification information stored in itself to identify the material of the clothes;

[0078] Step 2: Use the time value setting module to set the detection time: set different detection times according to the clothing material information of the entrants transmitted by the personnel identification module. For people wearing clothes with more chemical fiber components, the time value setting module will set a detection time of 30s, and for people wearing clothes with less chemical fiber components, the time value setting module will set a detection time of 15s;

[0079] Step 3. During the entry process, the entrants will place one hand in the electrostatic ball according to the prompt to use the electrostatic ball to remove the static electricity carried by the person's body. In addition, another set of cameras is installed on the top of the electrostatic ball (the number of cameras is 10-20, and the specific number can be designed according to the number of electrostatic balls, so that each electrostatic ball is installed with a camera on the top). This set of cameras can accurately capture the hands of the personnel based on the attention mechanism and the fully convolutional neural network and can identify whether the hands of the detection personnel are in contact with the electrostatic ball. At the same time, the time value setting module will pass the set electrostatic detection time to the electrostatic detection module. Once the hand of the person touching the electrostatic ball leaves the electrostatic ball within the electrostatic detection time, it will be recognized by the camera on the top, and the word NO will be displayed through the detection prompt module to remind people that the static electricity needs to be completely eliminated. The hand needs to be placed on the electrostatic ball again. When the camera recognizes that the hand of the person has been in contact with the electrostatic ball and has continued in this state for a period of time, the word OK will be displayed through the detection prompt module to remind people that the static electricity elimination is complete and they can leave. Among them, the attention mechanism used by the camera refers to the need to select a specific part in the visual area and then focus on it, which mainly includes two aspects: deciding which part of the input needs to be paid attention to and allocating limited information processing resources to the important part; the camera will select the hand of the person entering the visual area that touches the electrostatic ball for identification, and identify whether it is in contact with the electrostatic ball.

[0080] The fully convolutional neural network in step 1 and step 3 classifies images at the pixel level, thereby solving the problem of semantic-level image segmentation. Unlike the classic convolutional neural network, which uses a fully connected layer after the convolutional layer to obtain a fixed-length feature vector for classification, the fully convolutional neural network can accept input images of any size and use a deconvolution layer to upsample the feature map of the last convolutional layer to restore it to the same size as the input image, thereby generating a prediction for each pixel while retaining the spatial information in the original input image. Finally, pixel-by-pixel classification is performed on the upsampled feature map.

[0081] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A human body static electricity elimination detection method based on attention mechanism and fully convolutional neural network, characterized in that: The method comprises the following steps: Step 1: Build a model based on attention mechanism and fully convolutional neural network; Step 2, constructing a clothing material recognition model and a static electricity elimination detection model based on the model; Step 3, collecting an image of the person to be identified, and identifying the material of the clothing worn by the person based on the clothing material recognition model; Step 4: Set the detection time according to different clothing materials; specifically including: Step 4-1, classify clothing materials into two categories, classify materials with chemical fiber content below a preset threshold into the first category, and the others into the second category; Step 4-2, setting the detection time for different types of clothing materials, wherein the detection time corresponding to the first type of clothing material is shorter than that corresponding to the second type of clothing material; Step 5, removing static electricity by using an electrostatic ball, and detecting whether the static electricity has been completely eliminated based on the static electricity elimination detection model and the detection duration.

2. According to claim 1, the method for eliminating static electricity in human body based on attention mechanism and full convolutional neural network is characterized in that: The model based on the attention mechanism and fully convolutional neural network described in step 1 includes: a 224×224×3 input layer, and a spatial attention mechanism channel and a temporal attention mechanism channel in parallel; The spatial attention mechanism channel includes a 55×55×48 convolutional layer, a 27×27×48 pooling layer, a 27×27×128 convolutional layer, a 13×13×128 pooling layer, a 13×13×192 convolutional layer, a 13×13×192 convolutional layer, a 13×13×128 convolutional layer, a 6×6×128 pooling layer, a spatial attention module, a 13×13×128 convolutional layer and a 1×1×N convolutional layer, wherein N is the preset number of types to be identified; The temporal attention mechanism channel includes a 55×55×48 convolutional layer, a 27×27×48 pooling layer, a 27×27×128 convolutional layer, a 13×13×128 pooling layer, a 13×13×192 convolutional layer, a 13×13×192 convolutional layer, a 13×13×128 convolutional layer, a 6×6×128 pooling layer, a temporal attention module, and a 1×1×N convolutional layer. Each value in the feature map output by the two channels is processed by softmax to obtain the candidate probability distribution P of each category based on the temporal attention module or the spatial attention module. t or P s , where the category corresponding to the maximum probability is taken as the final recognized category.

3. The method for detecting human body static electricity elimination based on attention mechanism and full convolutional neural network according to claim 2, characterized in that: The spatial attention module includes a long short-term memory network LSTM, a 6×6×128 convolutional layer with a step size of 1, a first full convolutional network unit, and a Normal layer arranged in sequence; wherein the first full convolutional network unit includes a 1×1×N full convolutional layer, using tanh as an activation function.

4. The method for detecting human body static electricity elimination based on attention mechanism and full convolutional neural network according to claim 2, characterized in that: The temporal attention module includes a long short-term memory network LSTM, a 13×13×128 convolutional layer with a step size of 1, a second full convolutional network unit, and a Normal layer arranged in sequence; wherein the second full convolutional network unit includes a 7×7×64 full convolutional layer with a step size of 2, a 5×5×128 full convolutional layer with a step size of 2, a 3×3×128 full convolutional layer with a step size of 1, and a 1×1×N full convolutional layer arranged in sequence, and ReLU is used as the activation function.

5. The method for detecting human body static electricity elimination based on attention mechanism and full convolutional neural network according to claim 2, characterized in that: Step 2 constructs a clothing material recognition model and a static elimination detection model based on the model, specifically including: Step 2-1, collect images of people wearing different clothing materials to form a data set, and divide it into a training set, a test set, and a validation set; Step 2-2, using the training set to train a model based on an attention mechanism and a fully convolutional neural network to obtain a clothing material recognition model; Step 2-3, collecting several images of hands touching the electrostatic ball to form a data set, and dividing it into a training set, a test set, and a validation set; Step 2-4, using the training set to train the model based on the attention mechanism and the fully convolutional neural network to obtain a static electricity elimination detection model.

6. The method for detecting human body static electricity elimination based on attention mechanism and fully convolutional neural network according to claim 1, characterized in that: Step 5, removing static electricity by using an electrostatic ball, and detecting whether the static electricity is completely eliminated based on the model and the detection time, specifically includes: The static electricity ball image is continuously collected in real time, and the static electricity elimination detection model is used to identify whether the person's hand touches the static electricity ball, and whether the contact time meets the corresponding detection time. If so, it is determined that the static electricity elimination is completed, otherwise it is determined that the static electricity elimination is not completed.

7. A human body static electricity elimination detection system based on an attention mechanism and a fully convolutional neural network based on the method according to any one of claims 1 to 6, characterized in that: The system comprises: A person identification module is used to identify the material of the clothes worn by a person based on a clothing material identification model; A time value setting module is used to set different detection time lengths according to the identified clothing material; A static electricity elimination detection module is used to eliminate static electricity and detect whether the static electricity has been eliminated based on a static electricity elimination detection model and a detection time; The detection prompt module is used to display the detection results of the static elimination detection module.

8. The human body static electricity elimination detection system based on the attention mechanism and the fully convolutional neural network according to claim 7 is characterized in that: The wireless transmission of information between the personnel identification module and the time value setting module, between the time value setting module and the static electricity detection module, and between the static electricity detection module and the detection prompt module is performed through the wireless mobile communication system.

9. The human body static electricity elimination detection system based on the attention mechanism and the fully convolutional neural network according to claim 8, characterized in that: The wireless mobile communication system is a cellular system.

Images