Electrostatic clothes wearing detection method and device, electronic equipment and storage medium

Abstract

The invention discloses an electrostatic clothes wearing detection method and device, electronic equipment and a storage medium, video data of a production area is collected through a camera, the video data is input into a pre-trained human body detection tracking model to obtain a human body detection frame, and the area of the human body detection frame is intercepted from the video data to obtain a human body image; and inputting the human body image into an electrostatic clothes wearing detection model to obtain an electrostatic clothes wearing detection result, when the electrostatic clothes wearing detection result is that the electrostatic clothes are not worn in a compliance manner, performing face recognition on the human body image to obtain identity information of the target person, and generating alarm information including the identity information. No personnel need to be arranged to supervise the wearing condition of the electrostatic clothes, manpower is saved, the electrostatic clothes do not need to be improved and provided with sensors, the cost of the electrostatic clothes is reduced, the electrostatic clothes wearing is detected through image recognition, the efficiency is high, and the non-compliant electrostatic clothes wearing behavior can be accurately detected.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and in particular, to a method, device, electronic device, and storage medium for detecting the wearing of static electricity protection clothing. Background Art

[0002] In the production and manufacturing workshop of electronic products, since various electronic devices need to be produced, in order to prevent the high voltage released by the static electricity generated by the human body from damaging the electronic devices in production, the personnel entering the production and manufacturing workshop need to wear static electricity protection clothing.

[0003] Currently, the supervision of wearing static electricity protection clothing mainly arranges personnel to conduct inspections at the entrance of the workshop, which is inefficient and wastes manpower. Or sensors are installed on the static electricity protection clothing to detect the static electricity protection clothing. For example, an electronic tag is installed on the static electricity protection clothing, and the electronic tag is read by a card reader to detect whether there is a static electricity protection clothing. Although this method can detect the static electricity protection clothing, it is possible that the personnel entering the workshop carry but do not wear (such as holding the static electricity protection clothing but not wearing it on the body) the static electricity protection clothing into the workshop. This not only requires improving the static electricity protection clothing and increasing the cost of the static electricity protection clothing, but also fails to ensure that the static electricity protection clothing is worn on the personnel according to the regulations. Summary of the Invention

[0004] The present invention provides a method, device, electronic device, and storage medium for detecting the wearing of static electricity protection clothing to solve the problems of low efficiency and waste of manpower in manually supervising the wearing of static electricity protection clothing, and the inability to ensure that the static electricity protection clothing is worn according to the specifications by detecting the static electricity protection clothing through sensors.

[0005] In a first aspect, the present invention provides a method for detecting the wearing of static electricity protection clothing, including:

[0006] Collecting video data of a production area through a camera;

[0007] Inputting the video data into a pre-trained human detection and tracking model to obtain a human detection frame;

[0008] Cropping the area of the human detection frame from the video data to obtain a human image;

[0009] Inputting the human image into a static electricity protection clothing wearing detection model to obtain a static electricity protection clothing wearing detection result;

[0010] When the static electricity protection clothing wearing detection result is that the static electricity protection clothing is not worn in compliance, performing face recognition on the human image to obtain the identity information of the target person;

[0011] Generating an alarm message including the identity information.

[0012] Optionally, inputting the video data into a pre-trained human detection and tracking model to obtain a human detection frame includes:

[0013] Extract multiple frames of first video images from the video data;

[0014] Intercept a preset close-up area from the first video image to obtain a second video image;

[0015] Input the second video image into a human detection and tracking model to obtain a human detection frame.

[0016] Optionally, the electrostatic clothing wearing detection model includes a human key point detection sub-model, an electrostatic clothing detection sub-model, an electrostatic hat detection sub-model, and a wearing detection sub-model. Input the human image into the electrostatic clothing wearing detection model to obtain an electrostatic clothing wearing detection result, including:

[0017] Input the human image into the human key point detection sub-model, the electrostatic clothing detection sub-model, and the electrostatic hat detection sub-model respectively;

[0018] Extract human key points from the human image in the human key point detection sub-model;

[0019] Extract the electrostatic clothing area from the human image in the electrostatic clothing detection sub-model;

[0020] Extract the electrostatic hat area from the human image in the electrostatic hat detection sub-model;

[0021] Perform wearing detection based on the human key points, the electrostatic clothing area, and the electrostatic hat area in the wearing detection sub-model to obtain an electrostatic clothing wearing detection result.

[0022] Optionally, performing wearing detection based on the human key points, the electrostatic clothing area, and the electrostatic hat area in the wearing detection sub-model to obtain an electrostatic clothing wearing detection result includes:

[0023] Determine the head key points in the wearing detection sub-model;

[0024] Judge whether the electrostatic hat area covers the head key points;

[0025] If so, generate a detection result of compliant electrostatic hat wearing;

[0026] If not, generate a detection result of non-compliant electrostatic hat wearing;

[0027] Judge whether the electrostatic clothing area covers the preset target key points, where the target key points are the preset key points among the human key points excluding the head key points;

[0028] If so, generate a detection result of compliant electrostatic clothing wearing;

[0029] If not, generate a detection result of non-compliant electrostatic clothing wearing.

[0030] Optionally, before intercepting the area of the human detection box from the video data to obtain a human body image, it further includes:

[0031] When there are more than two human detection boxes, calculate the coincidence degree of adjacent two human detection boxes;

[0032] Determine whether the coincidence degree is less than a preset coincidence degree threshold;

[0033] If so, perform the step of intercepting the area of the human detection box from the video data to obtain a human body image;

[0034] If not, determine that the person corresponding to the human detection is blocked, mark the person corresponding to the human detection box as a person to be investigated, and return to the step of inputting the video data into a pre-trained human detection and tracking model to obtain a human detection box.

[0035] Optionally, after inputting the human body image into an electrostatic clothing wearing detection model to obtain an electrostatic clothing wearing detection result, it further includes:

[0036] If the electrostatic clothing wearing detection result of the person marked as to be investigated is wearing the electrostatic clothing in compliance, mark the person to be investigated as a person wearing the electrostatic clothing in compliance.

[0037] Optionally, generating an alarm message including the identity information includes:

[0038] Intercept a video segment or video image including the target person not wearing the electrostatic clothing in compliance from the video data;

[0039] Send the video segment or video image and the identity information of the target person to the server;

[0040] Generate a voice alarm message for broadcasting the identity information of the target person and a preset prompt message.

[0041] In a second aspect, the present invention provides an electrostatic clothing wearing detection device, including:

[0042] A video data acquisition module, configured to acquire video data of a production area through a camera;

[0043] A human detection and tracking module, configured to input the video data into a pre-trained human detection and tracking model to obtain a human detection box;

[0044] A human body image interception module, configured to intercept the area of the human detection box from the video data to obtain a human body image;

[0045] An electrostatic clothing wearing detection module, configured to input the human body image into an electrostatic clothing wearing detection model to obtain an electrostatic clothing wearing detection result;

[0046] An identity information determination module, configured to perform face recognition on the human body image to obtain the identity information of the target person when the electrostatic clothing wearing detection result indicates non-compliant wearing of the electrostatic clothing;

[0047] An alarm module, configured to generate an alarm message including the identity information.

[0048] In a third aspect, the present invention provides an electronic device, which includes:

[0049] At least one processor; and

[0050] A memory communicatively connected to the at least one processor; wherein,

[0051] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the electrostatic clothing wearing detection method described in the first aspect of the present invention.

[0052] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer instructions for causing a processor to implement the electrostatic clothing wearing detection method described in the first aspect of the present invention when executed.

[0053] In the embodiments of the present invention, video data is collected from the production area through a camera, the video data is input into a human body detection and tracking model to obtain a human body detection frame, the area of the human body detection frame is intercepted from the video data to obtain a human body image, and further the human body image is input into an electrostatic clothing wearing detection model to obtain an electrostatic clothing wearing detection result. When the electrostatic clothing wearing detection result indicates non-compliant wearing of the electrostatic clothing, face recognition is performed on the human body image to obtain the identity information of the target person, and an alarm message including the identity information is generated. There is no need to arrange personnel to supervise the wearing situation of the electrostatic clothing, which saves manpower, and there is no need to improve the electrostatic clothing by installing sensors, which reduces the cost of the electrostatic clothing. The detection of the wearing of the electrostatic clothing by image recognition has high efficiency and can accurately detect the behavior of non-compliant wearing of the electrostatic clothing.

[0054] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0056] Figure 1 It is a flowchart of a method for detecting the wearing of an electrostatic suit provided in the first embodiment of the present invention;

[0057] Figure 2 It is a flowchart of a method for detecting the wearing of an electrostatic suit provided in the second embodiment of the present invention;

[0058] Figure 3 It is a block diagram of the model structure of the electrostatic suit wearing detection mode;

[0059] Figure 4 It is a schematic diagram of the wearing of human key points, electrostatic clothes, and electrostatic caps;

[0060] Figure 5 It is a schematic structural diagram of an electrostatic suit wearing detection device provided in the third embodiment of the present invention;

[0061] Figure 6 It is a schematic structural diagram of an electronic device provided in the fourth embodiment of the present invention. Detailed implementation manners

[0062] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0063] Embodiment 1

[0064] Figure 1 It is a flowchart of a method for detecting the wearing of an electrostatic suit provided in the first embodiment of the present invention. This embodiment is applicable to the situation of detecting whether personnel in the production area wear electrostatic suits. This method can be executed by an electrostatic suit wearing detection device, which can be implemented in the form of hardware and / or software and can be configured in an electronic device, such as configured in an edge terminal or a server. As Figure 1 shown, the method for detecting the wearing of an electrostatic suit includes:

[0065] S101. Collect video data of the production area through a camera.

[0066] In this embodiment, the production area can be an area where electrostatic clothing wearing detection is required. Exemplarily, it can be a workshop for manufacturing electronic products in a manufacturing enterprise. For example, in an elevator manufacturing enterprise, the production area can be the internal area of the production workshop for manufacturing circuit boards such as the main control board and door machine board of the elevator. One or more cameras can be installed in the production area to collect video data of the production area from different angles, so as to avoid large-area occlusion of personnel in the production area. Among them, the cameras can collect video data at a fixed or dynamic frame rate and send the video data to the edge terminal or the server for electrostatic clothing wearing detection through the video data at the edge terminal or the server.

[0067] S102. Input the video data into a pre-trained human detection and tracking model to obtain a human detection box.

[0068] In this embodiment, a human detection and tracking model can be pre-trained. The human detection and tracking model is used to perform human detection on an image to output a human detection box and continuously perform human tracking. Exemplarily, each frame of video image in the video data can be input into the human detection and tracking model, or the video data can be frame-sampled according to a preset frame rate to obtain video images and input them into the human detection and tracking model, or key frames (I frames) can be extracted from the video data and input into the human detection and tracking model to obtain the human detection results of each frame of video image input, that is, if there is a human in the input video image, a human detection box is output, and if there is no human, no human detection box is output.

[0069] S103. Intercept the area of the human detection box from the video data to obtain a human image.

[0070] If the human detection and tracking model outputs a human detection box, indicating that a human is detected in the production area, the area of the human detection box can be intercepted from the video image of the input video data to obtain a human image including the human.

[0071] S104. Input the human image into an electrostatic clothing wearing detection model to obtain an electrostatic clothing wearing detection result.

[0072] In this embodiment, an electrostatic clothing wearing detection model can be pre-trained. Exemplarily, training images of a person not wearing electrostatic clothing, wearing electrostatic clothing in compliance, wearing electrostatic clothing not in compliance, and wearing ordinary clothing (non-electrostatic clothing) can be collected, and the training images can be labeled. Then, the training images are used to train the electrostatic clothing wearing detection model, so that the trained electrostatic clothing wearing detection model has the ability to distinguish electrostatic clothing from ordinary clothing and to identify whether electrostatic clothing is worn in compliance or not. After inputting a human body image into the electrostatic clothing wearing detection model, an electrostatic clothing wearing detection result of the person in the human body image can be obtained. The electrostatic clothing wearing detection result can be wearing in compliance or not wearing in compliance. Here, wearing in compliance means that the person wears electrostatic clothing according to the wearing specifications, and not wearing in compliance means that the person does not wear electrostatic clothing or does not wear it according to the specifications.

[0073] S105. When the electrostatic clothing wearing detection result is not wearing electrostatic clothing in compliance, perform face recognition on the human body image to obtain the identity information of the target person.

[0074] Exemplarily, when it is detected that electrostatic clothing is not worn in compliance, face recognition can be performed on the human body image input into the electrostatic clothing wearing detection model to obtain the identity information of the target person who is not wearing electrostatic clothing in compliance. The identity information can be information such as the name and work number of the target person.

[0075] S106. Generate an alarm message including the identity information.

[0076] Specifically, a video clip or video image including the target person not wearing electrostatic clothing in compliance can be intercepted from the video data, and the video clip or video image and the identity information of the target person can be sent to a server or an edge terminal to record the behavior information of the target person not wearing electrostatic clothing in compliance on the server or the edge terminal, so as to evaluate and train the target person, and generate a voice alarm message for broadcasting the identity information of the target person and a preset prompt message. For example, the voice alarm message is broadcast through a broadcasting device in the production area to remind and supervise the target person to wear electrostatic clothing in a standardized manner.

[0077] In the embodiment of the present invention, video data is collected from the production area through a camera, the video data is input into a human body detection and tracking model to obtain a human body detection frame, the area of the human body detection frame is intercepted from the video data to obtain a human body image, and further the human body image is input into an electrostatic clothing wearing detection model to obtain an electrostatic clothing wearing detection result. When the electrostatic clothing wearing detection result is not wearing electrostatic clothing in compliance, face recognition is performed on the human body image to obtain the identity information of the target person, and an alarm message including the identity information is generated. There is no need to arrange personnel to supervise the wearing situation of electrostatic clothing, which saves manpower. There is also no need to install sensors on the electrostatic clothing, which reduces the cost of electrostatic clothing. The detection of electrostatic clothing wearing through image recognition is highly efficient and can accurately detect the behavior of not wearing electrostatic clothing in compliance.

[0078] Example 2

[0079] Figure 2 The figure is a flowchart of a method for detecting the wearing of an electrostatic suit provided in Example 2 of the present invention. This embodiment of the present invention is optimized on the basis of the above Example 1. As Figure 2 shown, the method for detecting the wearing of an electrostatic suit includes:

[0080] S201. Collect video data of the production area through a camera.

[0081] In this embodiment, the production area may be an area where the wearing of an electrostatic suit needs to be detected. One or more cameras may be installed in the production area to collect video data of the production area from different angles, so as to avoid large-area occlusion of personnel in the production area. The camera may collect video data at a fixed or dynamic frame rate and send the video data to the edge terminal.

[0082] S202. Extract multiple frames of first video images from the video data.

[0083] The first video image is an original image extracted from the video data. Multiple frames of first video images may be extracted by frame extraction from the video data according to a preset frame rate. In one example, key frames (I frames) may be extracted from the video data as the first video images. The multiple frames of first video images may be arranged in the order of the time stamps of the images in the video data to obtain a first video image sequence.

[0084] S203. Intercept a preset close-up area from the first video image to obtain a second video image.

[0085] When the area of the production area is large, the imaging area of the human body in the area far from the camera is small, which is not conducive to accurately detecting the wearing of the electrostatic suit. When the camera is fixed, a reference image may be collected, and a close-up area may be divided from the reference image. The close-up area may be an image area of an area within a preset range from the camera. Further determine the boundary of the image area to obtain the boundary of the close-up area. For each frame of the first video image, the close-up area may be intercepted through the boundary of the preset close-up area to obtain a second video image, avoiding the situation that the target in the far-view area is too small to be accurately detected.

[0086] S204. Input the second video image into a human detection and tracking model to obtain a human detection frame.

[0087] In this embodiment, the human detection and tracking model may be an existing target detection and tracking model, or a human detection and tracking model trained with training images with the production area as the background, so as to improve the ability of the human detection and tracking model to identify humans in the production area.

[0088] In an optional embodiment, the second video image is input into the human body detection and tracking model. If there are more than two human body detection frames output, the overlap of two adjacent human body detection frames can be calculated. For example, the intersection over union (IOU) of the two human body detection frames can be calculated. The overlap of the two human body detection frames is measured by the intersection over union ratio, and it is further determined whether the overlap is less than a preset overlap threshold. If so, it is determined that the people in the production area do not occlude each other, and S205 can be executed; if not, it is determined that the people corresponding to the human body detection are occluded by each other, and the people corresponding to the human body detection frames are marked as people to be checked, and the process returns to S202 to continue detecting and tracking the occluded people.

[0089] S205: intercepting an area of ​​a human body detection frame from the second video image to obtain a human body image.

[0090] When it is determined that the human bodies in the production area do not block each other, the area of ​​the human body detection frame of the human body that is not blocked can be intercepted from the input second video image to obtain a human body image including the human body.

[0091] Of course, semantic segmentation may also be performed on the second video image, for example, the image may be input into a semantic segmentation model to obtain a human body region, and the human body region may be cut out from the second video image by cutting out the image to obtain a human body image.

[0092] S206, inputting the human body image into the human body key point detection sub-model, the electrostatic clothing detection sub-model, and the electrostatic hat detection sub-model respectively.

[0093] In this embodiment, Figure 3 As shown, the electrostatic clothing wearing detection model can include a human body key point detection sub-model, an electrostatic clothing detection sub-model, an electrostatic hat detection sub-model and a wearing detection sub-model, wherein the human body key point detection sub-model, the electrostatic clothing detection sub-model and the electrostatic hat detection sub-model are respectively connected to the wearing detection sub-model.

[0094] Among them, the human key point detection sub-model is used to detect the key points of the human body, such as Figure 4 The figure shows a schematic diagram of the key points of the human body. This embodiment adopts 15 key points. The electrostatic clothing detection sub-model is used to detect the electrostatic clothing and output the electrostatic clothing area. The electrostatic cap detection sub-model is used to detect the electrostatic cap and output the electrostatic cap area. The wearing detection sub-model is used to predict the wearing detection results of the electrostatic clothing and the electrostatic cap based on the key points of the human body, the electrostatic clothing area and the electrostatic cap area.

[0095] When training an electrostatic clothing wearing detection model, human body images including those with compliant wearing of electrostatic clothing and suits, non-compliant wearing of electrostatic clothing and suits, and non-wearing of electrostatic clothing and suits can be obtained as training images, and the training images are labeled with human body key points, electrostatic clothing regions, electrostatic cap regions, compliant wearing and non-compliant wearing labels. Then, the training images are input into the human body key point detection sub-model, the electrostatic clothing detection sub-model, and the electrostatic cap detection sub-model to obtain detection results from the wearing detection sub-model. The key point loss value is calculated using the human body key points detected by the human body key point detection sub-model, the electrostatic clothing detection loss value is calculated using the electrostatic clothing region output by the electrostatic clothing detection sub-model and the labeled electrostatic clothing region, the electrostatic cap detection loss value is calculated using the electrostatic cap region output by the electrostatic cap detection sub-model and the labeled electrostatic cap region, and the wearing detection loss value is calculated using the detection results output by the wearing detection sub-model and the labeled results. Furthermore, the weighted sum of the key point loss value, the electrostatic clothing detection loss value, the electrostatic cap detection loss value, and the wearing detection loss value is calculated to obtain the total loss value. After adjusting the model parameters with the total loss value, the model is continuously trained until the total loss value is less than the preset value after each round of training, and the trained electrostatic clothing wearing detection model is obtained.

[0096] S207. Extract human body key points from the human body image in the human body key point detection sub-model.

[0097] As Figure 4 shown, after the human body image is input into the human body key point detection sub-model, the human body key points in the human body image can be obtained, including head key point P0, torso key points P1, P2, and limb key points P3, P6, P9, P10, P11, P12, P13, P14, P4, P5, P7, P8.

[0098] S208. Extract the electrostatic clothing region from the human body image in the electrostatic clothing detection sub-model.

[0099] In this embodiment, after the human body image is input into the electrostatic clothing detection sub-model, the electrostatic clothing region can be obtained. The electrostatic clothing region can refer to the region formed by the pixels belonging to the electrostatic clothing in the human body image, and the electrostatic clothing region can be a continuous or discontinuous closed region.

[0100] S209. Extract the electrostatic cap region from the human body image in the electrostatic cap detection sub-model.

[0101] After the human body image is input into the electrostatic cap detection sub-model, the electrostatic cap region can be obtained. The electrostatic clothing region can refer to the region formed by the pixels belonging to the electrostatic cap in the human body image.

[0102] S210. Perform wearing detection based on the human body key points, the electrostatic clothing region, and the electrostatic cap region in the wearing detection sub-model to obtain the electrostatic clothing wearing detection result.

[0103] Specifically, first, the head key points can be determined in the wearable detection sub-model, and it is judged whether the static electricity cap area covers the head key points. If so, a detection result of wearing the static electricity cap in compliance is generated; if not, a detection result of not wearing the static electricity cap in compliance is generated. It is judged whether the static electricity clothing area covers the preset target key points, and the target key points are the preset key points among the human key points excluding the head key points. If so, a detection result of wearing the static electricity clothing in compliance is generated; if not, a detection result of not wearing the static electricity clothing in compliance is generated.

[0104] Exemplarily, as Figure 4 shown, the head key point P0 is covered by the static electricity cap area, and it is determined that the static electricity cap is worn qualified. If the target person carries the static electricity cap in the hand, even if the static electricity cap area is detected in the hand area, since the head key point P0 is not covered, it is still judged that the static electricity cap is not worn in compliance.

[0105] Further, as Figure 4 shown, those skilled in the art can determine the target key points according to the style of the static electricity clothing. Exemplarily, if the static electricity clothing is a long-sleeved one with long trousers, the target key points are the key points P1 - P14 except the head key points. If the static electricity clothing has no long trousers, the target key points are the key points except the head key point P0 and the lower limb key points P4, P5, P7, P8. When the static electricity clothing area covers the target key points, it indicates that the person wears the static electricity clothing in compliance, otherwise it indicates that the person does not wear the static electricity clothing or does not wear the static electricity clothing in compliance. As Figure 4 shown, if the static electricity clothing is a long-sleeved static electricity clothing and the static electricity clothing area does not cover the key points P10 and P11, it can be determined that it is detected that the person has rolled up the right sleeve of the static electricity clothing, which belongs to not wearing the static electricity clothing according to the specification.

[0106] In this embodiment, by detecting the human key points, the static electricity clothing area and the static electricity cap area, and whether the static electricity clothing area and the static electricity cap area cover the human key points, it can be accurately determined whether the static electricity clothing and the static electricity cap are worn in compliance, avoiding the use of electronic labels to detect whether the personnel wear static electricity clothing to enter the production area, and being unable to detect that the personnel do not wear but hold the static electricity clothing or do not wear the static electricity clothing in compliance, and being able to accurately detect that the static electricity clothing is not worn in compliance.

[0107] S211. When the detection result of wearing the static electricity clothing is not wearing the static electricity clothing in compliance, perform face recognition on the human body image to obtain the identity information of the target person.

[0108] Exemplarily, when it is detected that the static electricity clothing is not worn in compliance, face recognition can be performed on the human body image input into the static electricity clothing wearing detection model to obtain the identity information of the target person who does not wear the static electricity clothing in compliance, and this identity information can be information such as the name and job number of the target person.

[0109] In one example, if the detection result shows that the electrostatic clothing is worn in compliance, the target person can be marked as a compliant person. If the detection result shows that the electrostatic clothing is not worn in compliance, the target person can be marked as a non-compliant person. For a person whose body is blocked and marked as to be investigated, when the detection result of the electrostatic clothing wearing shows that the electrostatic clothing is worn in compliance after detection, the person to be investigated is marked as a person wearing the electrostatic clothing in compliance.

[0110] S212. Generate an alarm message including identity information.

[0111] Specifically, a video clip or video image including the target person not wearing the electrostatic clothing in compliance can be intercepted from the video data, and the video clip or video image and the identity information of the target person are sent to the server or edge terminal to record the behavior information of the target person not wearing the electrostatic clothing in compliance on the server or edge terminal, so as to conduct assessment and training on the target person, and generate a voice alarm message for broadcasting the identity information of the target person and a preset prompt message. For example, the voice alarm message is broadcast through the broadcasting device in the production area to remind and supervise the target person to wear the electrostatic clothing in a standardized manner.

[0112] After extracting the human body image from the video data in this embodiment, the human body image is respectively input into the human body key point detection sub-model, the electrostatic clothing detection sub-model, and the electrostatic cap detection sub-model to obtain the human body key points, the electrostatic clothing area, and the electrostatic cap area. Further, in the wearing detection sub-model, wearing detection is performed based on the human body key points, the electrostatic clothing area, and the electrostatic cap area to obtain the detection result of electrostatic clothing wearing. When the detection result of electrostatic clothing wearing shows that the electrostatic clothing is not worn in compliance, face recognition is performed on the human body image to obtain the identity information of the target person, and an alarm message including the identity information is generated. There is no need to arrange personnel to supervise the wearing situation of the electrostatic clothing, which saves manpower, and there is no need to install sensors on the electrostatic clothing for improvement, which reduces the cost of the electrostatic clothing. The detection of electrostatic clothing wearing through image recognition has high efficiency and can accurately detect the behavior of not wearing the electrostatic clothing in compliance.

[0113] Embodiment III

[0114] Figure 5 It is a schematic structural diagram of an electrostatic clothing wearing detection device provided in Embodiment III of the present invention. As Figure 5 shown, the electrostatic clothing wearing detection device includes:

[0115] A video data acquisition module 501, configured to acquire video data of the production area through a camera;

[0116] A human body detection and tracking module 502, configured to input the video data into a pre-trained human body detection and tracking model to obtain a human body detection frame;

[0117] The human body image extraction module 503 is used to extract the area of the human body detection frame from the video data to obtain a human body image;

[0118] The static electricity suit wearing detection module 504 is used to input the human body image into the static electricity suit wearing detection model to obtain the static electricity suit wearing detection result;

[0119] The identity information determination module 505 is used to perform face recognition on the human body image to obtain the identity information of the target person when the static electricity suit wearing detection result is that the static electricity suit is not worn in compliance;

[0120] The alarm module 506 is used to generate an alarm message including the identity information.

[0121] Optionally, the human body detection and tracking module 502 includes:

[0122] The first video image extraction unit is used to extract multiple frames of first video images from the video data;

[0123] The second video image extraction unit is used to intercept a preset close-up area from the first video image to obtain a second video image;

[0124] The human body detection unit is used to input the second video image into the human body detection and tracking model to obtain a human body detection frame.

[0125] Optionally, the static electricity suit wearing detection model includes a human body key point detection sub-model, a static electricity clothing detection sub-model, a static electricity hat detection sub-model, and a wearing detection sub-model. The static electricity suit wearing detection module 504 includes:

[0126] The image input unit is used to input the human body image into the human body key point detection sub-model, the static electricity clothing detection sub-model, and the static electricity hat detection sub-model respectively;

[0127] The human body key point extraction unit is used to extract human body key points from the human body image in the human body key point detection sub-model;

[0128] The static electricity clothing area extraction unit is used to extract the static electricity clothing area from the human body image in the static electricity clothing detection sub-model;

[0129] The static electricity hat area extraction unit is used to extract the static electricity hat area from the human body image in the static electricity hat detection sub-model;

[0130] The static electricity suit wearing detection unit is used to perform wearing detection based on the human body key points, the static electricity clothing area, and the static electricity hat area in the wearing detection sub-model to obtain the static electricity suit wearing detection result.

[0131] Optionally, the static electricity suit wearing detection unit includes:

[0132] A head key point determination subunit, configured to determine head key points in the wearable detection sub-model;

[0133] An electrostatic cap wearing judgment subunit, configured to judge whether the electrostatic cap area covers the head key points; if so, execute the electrostatic cap compliance wearing determination subunit, if not, execute the electrostatic cap non-compliance wearing determination subunit;

[0134] An electrostatic cap compliance wearing determination subunit, configured to generate a detection result of wearing an electrostatic cap in compliance;

[0135] An electrostatic cap non-compliance wearing determination subunit, configured to generate a detection result of not wearing an electrostatic cap in compliance;

[0136] An electrostatic clothing wearing judgment subunit, configured to judge whether the electrostatic clothing area covers preset target key points, where the target key points are preset key points among human key points excluding the head key points; if so, execute the electrostatic clothing compliance wearing determination subunit, if not, execute the electrostatic clothing non-compliance wearing determination subunit;

[0137] An electrostatic clothing compliance wearing determination subunit, configured to generate a detection result of wearing electrostatic clothing in compliance;

[0138] An electrostatic clothing non-compliance wearing determination subunit, configured to generate a detection result of not wearing electrostatic clothing in compliance.

[0139] Optionally, before the human body image capture module 503, it further includes:

[0140] A detection frame overlap degree calculation module, configured to calculate the overlap degree of adjacent human body detection frames when there are two or more human body detection frames;

[0141] An overlap degree judgment module, configured to judge whether the overlap degree is less than a preset overlap degree threshold; if so, execute the human body image capture module 503, if not, execute the occlusion determination module;

[0142] If so, execute the step of intercepting the area of the human body detection frame from the video data to obtain a human body image;

[0143] An occlusion determination module, configured to determine that the person corresponding to the human body detection is occluded, mark the person corresponding to the human body detection frame as a person to be investigated, and return to the human body detection and tracking module 502.

[0144] Optionally, after the electrostatic clothing wearing detection module 504, it further includes:

[0145] A to-be-investigated mark update module, configured to mark the person to be investigated as a person wearing electrostatic clothing in compliance if the electrostatic clothing wearing detection result of the person marked as to be investigated is wearing electrostatic clothing in compliance.

[0146] Optionally, the alarm module 506 includes:

[0147] A video clip or video image capturing unit, configured to capture a video clip or video image including the target person not wearing the static electricity protection suit in compliance from the video data;

[0148] A sending unit, configured to send the video clip or video image and the identity information of the target person to the server;

[0149] A voice alarm unit, configured to generate a voice alarm message for broadcasting the identity information of the target person and a preset prompt message.

[0150] The static electricity protection suit wearing detection device provided by the embodiments of the present invention can execute the static electricity protection suit wearing detection method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0151] Embodiment 4

[0152] Figure 6 FIG. shows a schematic structural diagram of an electronic device 60 that can be used to implement the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device (such as a helmet, glasses, a watch, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are only examples, and are not intended to limit the implementation of the present invention described herein and / or claimed.

[0153] As Figure 6 shown, the electronic device 60 includes at least one processor 61, and a memory communicatively connected to at least one processor 61, such as a read-only memory (ROM) 62, a random access memory (RAM) 63, etc. Among them, the memory stores a computer program executable by at least one processor. The processor 61 can execute various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 62 or the computer program loaded from the storage unit 68 into the random access memory (RAM) 63. In the RAM 63, various programs and data required for the operation of the electronic device 60 can also be stored. The processor 61, the ROM 62, and the RAM 63 are connected to each other through a bus 64. The input / output (I / O) interface 65 is also connected to the bus 64.

[0154] Multiple components in the electronic device 60 are connected to the I / O interface 65, including: an input unit 66, such as a keyboard, a mouse, etc.; an output unit 67, such as various types of displays, speakers, etc.; a storage unit 68, such as a disk, an optical disc, etc.; and a communication unit 69, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 69 allows the electronic device 60 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0155] The processor 61 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 61 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 61 executes the various methods and processes described above, such as the electrostatic suit wearing detection method.

[0156] In some embodiments, the electrostatic suit wearing detection method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 68. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 60 via the ROM 62 and / or the communication unit 69. When the computer program is loaded into the RAM 63 and executed by the processor 61, one or more steps of the electrostatic suit wearing detection method described above can be executed. Alternatively, in other embodiments, the processor 61 can be configured to execute the electrostatic suit wearing detection method by any other suitable means (e.g., by means of firmware).

[0157] Various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0158] A computer program for implementing the method of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, a special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer programs may be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0159] In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0160] In order to provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0161] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected with each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0162] A computing system can include a client and a server. The client and the server are generally far from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0163] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0164] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for detecting the wearing of electrostatic clothing, characterized in that: include: Collect video data of the production area through cameras; Inputting the video data into a pre-trained human body detection and tracking model to obtain a human body detection frame; Cutting out the area of ​​the human body detection frame from the video data to obtain a human body image; Inputting the human body image into an anti-static clothing wearing detection model to obtain an anti-static clothing wearing detection result; When the anti-static clothing wearing detection result is that the anti-static clothing is not worn in compliance with regulations, face recognition is performed on the human body image to obtain identity information of the target person; Generate warning information including the identity information.

2. The method for detecting the wearing of electrostatic clothing according to claim 1, characterized in that: Inputting the video data into a pre-trained human detection and tracking model to obtain a human detection frame includes: Extracting a plurality of frames of first video images from the video data; intercepting a preset close-up area from the first video image to obtain a second video image; The second video image is input into a human body detection and tracking model to obtain a human body detection frame.

3. The method for detecting the wearing of electrostatic clothing according to claim 1, characterized in that: The anti-static clothing wearing detection model includes a human key point detection sub-model, an anti-static clothing detection sub-model, an anti-static cap detection sub-model and a wearing detection sub-model. The human image is input into the anti-static clothing wearing detection model to obtain an anti-static clothing wearing detection result, including: Inputting the human body image into the human body key point detection sub-model, the electrostatic clothing detection sub-model, and the electrostatic hat detection sub-model respectively; Extracting human body key points from the human body image in the human body key point detection submodel; Extracting an electrostatic clothing area from the human body image in the electrostatic clothing detection sub-model; extracting an electrostatic cap area from the human body image in the electrostatic cap detection sub-model; In the wearing detection sub-model, wearing detection is performed based on the key points of the human body, the anti-static clothing area and the anti-static hat area to obtain an anti-static clothing wearing detection result.

4. The method for detecting the wearing of electrostatic clothing according to claim 3, characterized in that: In the wearing detection sub-model, wearing detection is performed based on the key points of the human body, the anti-static clothing area, and the anti-static cap area to obtain an anti-static clothing wearing detection result, including: Determining key points of the head in the wear detection sub-model; Determine whether the electrostatic cap area covers the key points of the head; If so, generate a test result of wearing an electrostatic cap in compliance with regulations; If not, a detection result of not wearing the electrostatic cap in compliance with regulations is generated; Determine whether the anti-static clothing area covers a preset target key point, where the target key point is a preset key point of the human body except the head key point; If so, generate a test result of wearing anti-static clothing in compliance with regulations; If not, a detection result is generated indicating that the anti-static clothing is not worn in compliance with regulations.

5. The method for detecting the wearing of electrostatic clothing according to any one of claims 1 to 4, characterized in that: Before extracting the area of ​​the human body detection frame from the video data to obtain the human body image, the method further includes: When there are more than two human body detection frames, the overlap degree of two adjacent human body detection frames is calculated; Determining whether the overlap is less than a preset overlap threshold; If yes, executing the step of capturing the area of ​​the human body detection frame from the video data to obtain a human body image; If not, it is determined that the person corresponding to the human body detection is blocked, the person corresponding to the human body detection frame is marked as a person to be checked, and the process returns to the step of inputting the video data into a pre-trained human body detection and tracking model to obtain a human body detection frame.

6. The method for detecting the wearing of electrostatic clothing according to claim 5, characterized in that: After inputting the human body image into the anti-static clothing wearing detection model to obtain the anti-static clothing wearing detection result, the method further includes: If the anti-static clothing wearing detection result of the person marked as the person to be checked is that the anti-static clothing is worn in compliance with regulations, the person to be checked is marked as a person who wears the anti-static clothing in compliance with regulations.

7. The method for detecting the wearing of electrostatic clothing according to any one of claims 1 to 4, characterized in that: Generating warning information including the identity information includes: Intercepting from the video data a video clip or video image including the target person not wearing the anti-static clothing in compliance with regulations; Sending the video clip or video image and the identity information of the target person to a server; Generate a voice warning message that broadcasts the identity information of the target person and a preset prompt.

8. An electrostatic clothing wearing detection device, characterized in that: include: Video data acquisition module, used to collect video data of the production area through a camera; A human body detection and tracking module, used for inputting the video data into a pre-trained human body detection and tracking model to obtain a human body detection frame; A human body image capture module, used to capture an area of ​​a human body detection frame from the video data to obtain a human body image; An anti-static clothing wearing detection module, used for inputting the human body image into an anti-static clothing wearing detection model to obtain an anti-static clothing wearing detection result; An identity information determination module, configured to perform face recognition on the human image to obtain the identity information of the target person when the anti-static clothing wearing detection result is that the anti-static clothing is not worn in compliance with regulations; An alarm module is used to generate alarm information including the identity information.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the static electricity clothing wearing detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the electrostatic clothing wearing detection method according to any one of claims 1 to 7 when executed.

Images