Individual protection safety helmet for limited space operation

Abstract

The invention discloses an individual protection safety helmet for limited space operation. The safety helmet is provided with a camera device used for acquiring real-time image data of an operation environment; the gas sensor is used for acquiring real-time gas data of a working environment; the controller is used for performing risk assessment on the real-time image data through a risk assessment model to obtain a risk assessment result, determining a gas detection result according to the real-time gas data, and sending a risk prompt signal to the alarm under the condition that the risk assessment result and / or the gas detection result meet a risk prompt condition; and the alarm is used for carrying out risk prompt according to the risk prompt signal. According to the safety helmet, real-time multi-dimensional (gas and image) risk detection can be carried out on the working environment of a worker, and when any risk exists, a risk prompt corresponding to the existing risk is carried out, so that the worker can avoid the risk in a targeted manner or evacuate in time; the safety helmet can ensure the operation safety of a worker in a limited space.

Description

Technical Field

[0001] The invention relates to the technical field of computer application, and in particular to an individual protective helmet for working in a limited space. Background Art

[0002] Confined space refers to a closed or partially closed space. When workers work in this environment, they are exposed to risks such as internal and external mechanical injuries, internal space collapse, and poisoning and suffocation. Therefore, before entering a confined space for work, you need to wear a personal protective helmet and conduct a comprehensive risk assessment of the outside and inside of the confined space so that you can take appropriate preventive measures in a timely manner to ensure the personal safety of workers.

[0003] In related technologies, the safety helmets worn by workers can perform certain risk detection on the working environment and issue risk warnings when operational risks occur. However, the judgment criteria are mostly relatively simple, making it difficult to deal with potential risks from multiple aspects at the work site, and it is impossible to ensure the safety of workers in limited working spaces from multiple dimensions. Summary of the invention

[0004] The present invention provides an individual protective helmet for working in a confined space, which can perform a full-process real-time risk assessment of the external and internal environments of the work site, so as to solve the technical problem in the related art that the safety helmet is difficult to ensure the safety of workers working in a confined work space.

[0005] According to the present invention, a personal protective helmet for confined space operations is provided, on which a gas sensor, a camera device, a controller and an alarm are deployed; wherein:

[0006] The camera device is used to obtain real-time image data of the working environment;

[0007] The gas sensor is used to obtain real-time gas data of the working environment;

[0008] The controller is used to perform risk assessment on the real-time image data through a risk assessment model to obtain a risk assessment result, determine a gas detection result according to the real-time gas data, generate a risk warning signal when the risk assessment result and / or the gas detection result meet a risk warning condition, and send the risk warning signal to the alarm, wherein the risk assessment model is obtained by training a convolutional neural network based on training samples;

[0009] The alarm is used to receive the risk warning signal and issue a risk warning according to the risk warning signal.

[0010] According to another aspect of the present invention, there is provided a personal protective helmet for confined space operations, the helmet also being provided with an infrared detector; wherein:

[0011] The infrared detector is used to obtain real-time wearing data of the target helmet, wherein the real-time wearing data represents the contact between the helmet and the head of the target wearer;

[0012] The controller is further configured to determine a hat-off detection result according to the real-time wearing data, generate a hat-off alarm signal when the hat-off detection result meets a hat-off alarm condition, and send the hat-off warning signal to the alarm;

[0013] The alarm is also used for receiving the hat-off warning signal and performing a hat-off alarm according to the hat-off warning signal.

[0014] According to another aspect of the present invention, there is provided a personal protective helmet for confined space operations, on which a navigator is also deployed; wherein:

[0015] The controller is further used to modify a preset original operation map according to the real-time image data to obtain a target operation map, determine real-time navigation data according to the real-time positioning data and the target operation map, generate a real-time navigation signal according to the real-time navigation data, and send the real-time navigation signal to the navigator, wherein the operation map at least includes an operation end point;

[0016] The navigator is used for navigating according to the received real-time navigation signal.

[0017] According to another aspect of the present invention, there is provided a personal protective helmet for confined space operations, the helmet also being provided with a positioning device; wherein:

[0018] The positioning device is used to obtain real-time positioning data of the helmet;

[0019] The controller is also used to send the real-time positioning data to a target cloud server through a communication device, wherein the target cloud server is a cloud server of a hard hat management platform.

[0020] According to another aspect of the present invention, there is provided a personal protective helmet for confined space operations, the helmet also having a call button; wherein,

[0021] The distress button is used to generate a distress signal in response to a button triggering operation, and send the distress signal to the controller;

[0022] The controller is further used to send the distress signal to the target cloud server through the communication device.

[0023] According to another aspect of the present invention, there is provided a personal protective helmet for confined space operations, the helmet is also provided with a memory, wherein:

[0024] The memory is used to store the real-time image data to obtain stored image data;

[0025] The controller is further used to send the stored image data to the target cloud server through the communication device.

[0026] The technical solution of the embodiment of the present invention is to deploy a gas sensor, a camera device, a controller and an alarm on the safety helmet; wherein the camera device is used to obtain real-time image data of the working environment; the gas sensor is used to obtain real-time gas data of the working environment; the controller is used to perform risk assessment on the real-time image data through a risk assessment model to obtain a risk assessment result, determine the gas detection result according to the real-time gas data, and generate a risk warning signal when the risk assessment result and / or the gas detection result meet the risk warning condition, and send the risk warning signal to the alarm, wherein the risk assessment model is obtained by training a convolutional neural network based on training samples; the alarm is used to receive the risk warning signal and give a risk warning according to the risk warning signal. The present invention integrates a gas sensor, a camera device, a controller and an alarm on a safety helmet at the same time, so that the safety helmet can perform real-time multi-dimensional (gas and image) risk detection on the working environment of the staff, and give risk warnings corresponding to the existing risks when any risk exists, so that the staff can timely understand the risks existing in the working environment, and avoid the risks or evacuate in time when there are operational risks, thereby effectively ensuring the working safety of the staff in a limited space on the basis of ensuring the work progress.

[0027] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended 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

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0029] Figure 1 is a connection relationship diagram of a deployment device on a safety helmet provided according to the first embodiment of the present invention;

[0030] Figure 2 is a flow chart of risk assessment and prompting provided according to an embodiment of the present invention;

[0031] Figure 3 is a connection relationship diagram of a deployment device on a safety helmet provided according to the second embodiment of the present invention;

[0032] Figure 4 The figure is an overall connection diagram of a deployment device on a safety helmet provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0034] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0035] Embodiment 1

[0036] Figure 1 1 is a connection diagram of a deployment device on a safety helmet according to the first embodiment of the present invention. This embodiment is applicable to situations where the safety helmet performs risk detection and early warning on the surrounding working environment. Figure 1 As shown, a gas sensor, a camera device, a controller and an alarm may be deployed on the safety helmet.

[0037] Wherein, the camera device is used to obtain real-time image data of the working environment;

[0038] The gas sensor is used to obtain real-time gas data of the working environment;

[0039] The controller is used to perform risk assessment on the real-time image data through a risk assessment model to obtain a risk assessment result, determine a gas detection result according to the real-time gas data, generate a risk warning signal when the risk assessment result and / or the gas detection result meet a risk warning condition, and send the risk warning signal to the alarm, wherein the risk assessment model is obtained by training a convolutional neural network based on training samples;

[0040] The alarm is used to receive the risk warning signal and issue a risk warning according to the risk warning signal.

[0041] The camera device can be used to obtain real-time image data of the working environment around the helmet. In the embodiment of the present invention, the camera device can be preset according to the scene requirements, and is not specifically limited here. Exemplarily, the camera device can be a "depth camera".

[0042] The real-time image data can be understood as image data acquired in real time. Optionally, the real-time image data can be a real-time video. The real-time video can include an image sequence. The image sequence can include multiple frames of images.

[0043] The risk assessment model can perform risk assessment on the operating environment based on the input real-time image data and output a risk assessment result, wherein the risk assessment result can indicate whether the operating environment has an operating risk.

[0044] In an embodiment of the present invention, the risk assessment model may include a collapse detection model for detecting inside a confined space, and a moving object detection model for detecting inside or outside a confined space (the moving object refers to a large mobile device).

[0045] Optionally, the controller is used to input the real-time image data into a collapse detection model and a moving object detection model, respectively, to obtain a collapse detection result and a moving object detection result, respectively, wherein the collapse detection result represents the collapse risk level of the working environment, and the moving object detection result represents whether there are moving objects of the target category in the working environment, and the collapse detection model is obtained by training the convolutional neural network based on collapse samples, and the moving object detection model is obtained by training the convolutional neural network based on mobile object samples.

[0046] Wherein, the collapse detection model can be used to perform collapse detection on the input real-time image data and output the collapse detection result. In an embodiment of the present invention, the collapse detection model can be obtained by training the convolutional neural network based on collapse samples. Wherein, the collapse samples can include collapse positive samples and / or collapse negative samples. The collapse positive sample can include a collapsed image of the working environment of the limited working space and a sample label representing the collapse. The collapse negative sample can include a non-collapsed image of the working environment of the limited working space and a sample label representing the non-collapse. The convolutional neural network (CNN) can be understood as a deep learning model. Wherein, the collapse detection result can represent the collapse risk level of the working environment. Wherein, the collapse risk level can represent the probability that the working environment has a collapse risk. Exemplarily, the collapse risk level can include no collapse, primary collapse, and secondary collapse. The probability that the working environment characterized by the secondary collapse has a collapse risk is higher than the probability that the working environment characterized by the primary collapse has a collapse risk.

[0047] The mobile object detection model can be used to detect mobile objects (i.e., large mobile equipment) on the input real-time image data, and output the mobile object detection result (i.e., large mobile equipment detection result). In an embodiment of the present invention, the mobile object detection model can be obtained by training the convolutional neural network based on mobile object samples. Wherein, the mobile object samples can include object positive samples and / or object negative samples. The object positive sample can include an image of a movable object of a target category. Exemplarily, the object positive sample is an image of a large mobile equipment. The object negative sample can include an image of a movable object of other categories. Exemplarily, the object negative sample can be an image of a non-large mobile equipment. Exemplarily, the mobile object of the target category can be a large mobile equipment, such as a crane or an excavator. Wherein, the mobile object detection result characterizes whether there is a large mobile machinery in the working environment. Whether there is a large machinery can be determined based on the recognition of each frame of the real-time image data (image sequence), and whether the large machinery is moving can be determined based on the before-after comparison of the images in the real-time image data (image sequence). The large mobile machinery detection result (i.e., the mobile object detection result) can characterize the mechanical injury risk level of the working environment. The mechanical injury risk level can characterize the probability of the existence of mechanical injury risk in the working environment. Exemplarily, the mechanical injury risk level can include no mechanical injury, first-level mechanical injury, and second-level mechanical injury. The probability of the existence of mechanical injury risk in the working environment characterized by the second-level mechanical injury is higher than the probability of the existence of collapse risk in the working environment characterized by the first-level mechanical injury.

[0048] Based on the above technical solution, it is possible to simultaneously conduct risk assessments of internal collapse in confined spaces and risk assessments of large mobile machinery inside and outside confined spaces for the working environment, thereby improving the comprehensiveness of risk assessments and further ensuring the working safety of staff.

[0049] The gas sensor can be used to monitor the gas in the working environment around the helmet in real time, that is, it can be used to obtain gas data in real time.

[0050] The real-time gas data can be understood as gas data acquired in real time. In the embodiment of the present invention, the real-time gas data can be set in advance according to user needs, and is not specifically limited here.

[0051] Optionally, the controller is used to determine a preset target concentration threshold, and determine the gas detection result based on the target concentration threshold and the real-time gas data, wherein the target concentration threshold is related to the target wearer of the helmet, and the real-time gas data includes at least the gas concentration of the target gas, and the target gas includes at least oxygen, toxic gas and harmful gas.

[0052] Wherein, the real-time gas data may include at least the gas concentration of the target gas. Wherein, the target gas may be understood as the gas to be monitored. The gas concentration may be understood as the concentration of the target gas. In an embodiment of the present invention, the target gas may include at least oxygen, toxic gas and harmful gas. The gas concentration may include at least the concentration of oxygen, the concentration of toxic gas and the concentration of harmful gas. Exemplarily, the toxic gas may include carbon monoxide, etc. The harmful gas may include methane and sulfur hexafluoride, etc.

[0053] The target concentration threshold value may be determined based on the physical fitness data of the target wearer of the helmet. The physical fitness data may characterize the target wearer's tolerance to hypoxia, toxic gas, and harmful gas, etc. It is understandable that the physical fitness data of different target wearers may be different or the same. Therefore, when the target wearers of the helmet are different, the target concentration threshold values ​​may be different or the same. Among them, the target wearer may refer to the staff member wearing the helmet.

[0054] The gas detection result can indicate whether the current working environment is suitable for the target wearer to continue working. The gas detection result can include whether the wearer can continue working and / or whether the wearer can continue working. If the gas detection result is that the wearer can continue working, the gas detection result can also include the duration of the wearer's continued working. For example, the gas detection result is that the wearer can continue working for 1 hour, 30 minutes, or 5 minutes.

[0055] The risk warning condition can be used to determine whether a risk warning is required. In an embodiment of the present invention, the risk warning condition can be preset according to the scenario requirements and is not specifically limited here. Optionally, the risk warning condition may include an assessment warning condition and / or a gas warning condition. The assessment warning condition is used to determine whether an early warning is issued for the risk assessment result. The gas warning condition is used to determine whether an early warning is issued for the gas detection result.

[0056] The risk warning signal may be related to the specific circumstances in which the risk assessment result and / or the gas detection result meet the risk warning condition.

[0057] The alarm device can be used to receive a signal sent by the controller and issue an alarm corresponding to the received signal.

[0058] Exemplarily, the alarm may include a lighting lamp and / or a voice announcer, etc.

[0059] Exemplarily, the controller is used to generate an assessment risk signal when the risk assessment result meets the assessment prompt condition and the gas detection result does not meet the gas prompt condition, and send the assessment risk signal to the lighting lamp and / or voice announcer. Furthermore, the lighting lamp is used to receive the assessment risk signal and flash a green or red light at a first frequency according to the assessment risk signal; the voice announcer is used to receive the assessment risk signal and make a voice announcement such as "Please note that there is a risk of collapse" or "There is a risk of injury from large mobile machinery" according to the assessment risk signal.

[0060] Alternatively, the controller is used to generate a gas risk signal when the risk assessment result does not meet the assessment prompt condition and the gas detection result meets the gas prompt condition, and send the gas risk signal to the lighting lamp and / or voice announcer. Further, the lighting lamp is used to receive the gas risk signal and flash a blue light of a first frequency according to the gas risk signal; the voice announcer is used to receive the gas risk signal and make a voice announcement such as "Please note that there are toxic and harmful gases" or "The oxygen concentration is low" according to the gas risk signal.

[0061] Alternatively, the controller is used to generate multiple risk signals when the risk assessment result meets the assessment prompt condition and the gas detection result meets the gas prompt condition, and send the multiple risk signals to the lighting lamp and / or voice announcer. Further, the lighting lamp is used to receive the multiple risk signals and flash the red and blue lights of the second frequency alternately according to the multiple risk signals; the voice announcer is used to receive the multiple risk signals and make voice announcements such as "There is a high operational risk, please evacuate immediately" according to the multiple risk signals.

[0062] Based on the above technical solution, it is possible to simultaneously detect the oxygen concentration, toxic gas and harmful gas concentration in the working environment, and issue warnings for the gas detection results based on thresholds set based on the personalized physical characteristics of the helmet wearer (worker), thereby realizing personalized warnings for the workers and avoiding the situation where some workers have special physical conditions and have been poisoned or fainted due to lack of oxygen without being warned when warnings are issued based on uniformly set thresholds, thereby ensuring the work safety of different individuals.

[0063] As an alternative embodiment of the above embodiment, Figure 2 FIG. 1 is a flow chart of risk assessment and prompting provided according to an embodiment of the present invention. Figure 2 As shown, the risk assessment and warning process can be:

[0064] Start the real-time risk assessment model and gas risk detection program for confined space operations with smart helmets; use the risk assessment model to detect large mobile machinery and collapse risks on the impact data captured by the camera, and determine whether there are mechanical injury risks and collapse risks outside and inside the current working space. Specifically, the processed image data is input into the shared convolutional layer to obtain the feature map, and then the feature map is input into the region proposal network to obtain the candidate region and feature information. Finally, the classifier is used to determine whether there are large mobile machinery outside the working environment corresponding to the image data, and whether the inside is a space prone to collapse.

[0065] If there is any large moving machinery, it is considered that there is a risk of mechanical injury. The controller changes the color of the helmet light to red, and voice reminds the staff to stop entering the confined space for work. If there is no large moving machinery, it is considered that there is no risk of mechanical injury, and the color of the helmet light will not be changed.

[0066] If there is a risk of collapse, the controller will control the color of the lighting to change to green to remind the staff to stop working and evacuate the current space; if there is no risk of collapse, it will continue to detect whether there are large moving machines in the current space. Specifically, the large mobile machinery detection model is used to perform motion recognition of large machinery detection on the image data captured by the camera.

[0067] The gas sensor performs real-time monitoring simultaneously, records the type and concentration of the gas detected, and also stores the data. Furthermore, gas risk detection is performed.

[0068] Determine whether there are toxic and harmful gases in the types of collected gases (toxic and harmful gases include toxic gases and harmful gases). If there are no toxic and harmful gases, determine whether the oxygen concentration is greater than the first threshold. If it is greater than the first threshold, broadcast "high oxygen concentration" and the remaining working time to the staff. If it is not greater than the first threshold, further determine whether the oxygen concentration is less than the second threshold. If it is less than the second threshold, broadcast "low oxygen concentration" and the remaining working time to the staff to prompt the staff to exit as needed; if it is not less than the second threshold, the gas detection is considered safe. Among them, the first threshold is set according to the symptoms of oxygen intoxication in the human body, and the second threshold is set according to the minimum oxygen concentration that the human body can withstand. The remaining working time is set according to the duration of the symptoms of oxygen intoxication or hypoxia in the human body.

[0069] If there are toxic and harmful gases in the collected gas types, it is determined whether the gas concentration of the toxic and harmful gas exceeds the third threshold, wherein the third threshold is determined according to the concentration of the gas that poisons the staff. If the gas concentration of the toxic and harmful gas exceeds the third threshold, the color of the lighting is controlled to be blue to prompt the staff to stop working and evacuate the current space. If the gas concentration of the toxic and harmful gas does not exceed the third threshold, it is further determined whether the oxygen concentration is greater than the first threshold. If it is greater than the first threshold, the "oxygen concentration is high" and the remaining working time are broadcast to the staff. If it is not greater than the first threshold, it is further determined whether the oxygen concentration is less than the second threshold. If it is less than the second threshold, the "oxygen concentration is low" and the remaining working time are broadcast to the staff; if it is not less than the second threshold, the gas detection is considered safe. Among them, the remaining working time is determined according to the pre-set mapping relationship between toxic and harmful gases and working time. It should be noted that when risk detection is performed on oxygen concentration and toxic and harmful gas concentration alone, if the detection result does not have a risk, there are still situations such as staff fainting or poisoning. That is, it is believed that the effects of oxygen concentration and toxic and harmful gas concentration on the human body are interactive. Therefore, in this embodiment, the concentration of the two gases is comprehensively detected at the same time, which can avoid the occurrence of such a situation and further ensure the safety of the operators.

[0070] The technical solution of the embodiment of the present invention is to deploy a gas sensor, a camera device, a controller and an alarm on the safety helmet; wherein the camera device is used to obtain real-time image data of the working environment; the gas sensor is used to obtain real-time gas data of the working environment; the controller is used to perform risk assessment on the real-time image data through a risk assessment model to obtain a risk assessment result, determine the gas detection result according to the real-time gas data, and generate a risk warning signal when the risk assessment result and / or the gas detection result meet the risk warning condition, and send the risk warning signal to the alarm, wherein the risk assessment model is obtained by training a convolutional neural network based on training samples; the alarm is used to receive the risk warning signal and give a risk warning according to the risk warning signal. The present invention integrates a gas sensor, a camera device, a controller and an alarm on a safety helmet at the same time, so that the safety helmet can perform real-time multi-dimensional (gas and image) risk detection on the working environment of the staff, and give risk warnings corresponding to the existing risks when any risk exists, so that the staff can timely understand the risks existing in the working environment, and avoid the risks or evacuate in time when there are operational risks, thereby effectively ensuring the working safety of the staff in a limited space on the basis of ensuring the work progress.

[0071] Embodiment 2

[0072] Figure 3 1 is a connection diagram of a deployment device on a safety helmet according to the second embodiment of the present invention. This embodiment is a further addition to the deployment device for the safety helmet in the above embodiment. Figure 3 As shown, an infrared detector can also be deployed on the helmet.

[0073] Wherein, the infrared detector is used to obtain real-time wearing data of the helmet, wherein the wearing data represents the contact between the helmet and the head of the target wearer;

[0074] The controller is further configured to determine a hat-off detection result according to the real-time wearing data, generate a hat-off alarm signal when the hat-off detection result meets a hat-off alarm condition, and send the hat-off warning signal to the alarm;

[0075] The alarm is also used for receiving the hat-off warning signal and performing a hat-off alarm according to the hat-off warning signal.

[0076] The infrared detector can be understood as a detector that works by detecting infrared rays emitted by the human body, that is, the infrared detector can be used to obtain wearing data representing the contact between the helmet and the head of the target wearer in real time.

[0077] The hat removal detection result can indicate whether the safety helmet is detached from the target wearer. Optionally, the hat removal detection result can include whether the helmet has been removed and / or whether the helmet has not been removed.

[0078] Optionally, the hat-off alarm condition may be used to determine whether to perform a hat-off warning for the hat-off detection result. The hat-off alarm condition may include that the hat-off detection result is that the hat has been taken off.

[0079] Exemplarily, the alarm (including the voice announcer) is also used to receive the hat-off warning signal, and make a voice announcement such as "take off the safety helmet" according to the hat-off warning signal.

[0080] Based on the above embodiment, an infrared detector is deployed on the safety helmet to realize the detection of hat removal, thereby ensuring that the workers wear the safety helmet in real time during the operation in a confined space.

[0081] Figure 4 1 is a diagram showing the overall connection relationship of a deployment device on a helmet according to an embodiment of the present invention. Figure 4 As shown, the helmet may also be equipped with a gas sensor, a camera device, a controller and an alarm, an infrared detector, a navigator, a positioning device, a communication device, a call button, a memory and a lighting lamp.

[0082] Optionally, a navigator may also be deployed on the safety helmet; wherein,

[0083] The controller is further used to modify a preset original operation map according to the real-time image data to obtain a target operation map, determine real-time navigation data according to the real-time positioning data and the target operation map, generate a real-time navigation signal according to the real-time navigation data, and send the real-time navigation signal to the navigator, wherein the operation map at least includes an operation end point;

[0084] The navigator is used for navigating according to the received real-time navigation signal.

[0085] The original operation map may be understood as an operation map pre-imported into the helmet for the controller to calculate navigation data. In an embodiment of the present invention, the original operation map may be a map drawn before the operation based on historical operation experience.

[0086] The target operation map can be understood as a modified operation map. In an embodiment of the present invention, the target operation map and the original operation map can be different or the same. The operation map includes at least an operation end point and may also include an operation route. Under normal circumstances, the operation end point in the target operation map is the same as the operation end point in the original operation map, and the operation route corresponding to the target operation map and the operation route in the original operation map can be different or the same.

[0087] The real-time navigation data can be understood as real-time navigation data. The real-time navigation data can at least include real-time navigation direction.

[0088] Specifically, the controller may transmit the real-time navigation data to the navigator based on the generated real-time navigation signal.

[0089] The navigator may be used to output navigation data corresponding to the real-time navigation signal. In an embodiment of the present invention, the navigation data may at least include a navigation direction. Optionally, the navigator may include a voice announcer or a micro screen.

[0090] Based on the above-mentioned embodiment scheme, the controller can make real-time corrections to the pre-imported original work map based on the real-time acquired image data to obtain a target work map that is highly consistent with the actual work environment, thereby improving the accuracy of work navigation through the target work map and improving work efficiency.

[0091] Optionally, a positioning device may be deployed on the safety helmet; wherein,

[0092] The positioning device is used to obtain real-time positioning data of the helmet;

[0093] The controller is also used to send the real-time positioning data to a target cloud server through a communication device, wherein the target cloud server is a cloud server of a hard hat management platform.

[0094] Wherein, the positioning device can be used to obtain the positioning data of the helmet in real time. In an embodiment of the present invention, the positioning device can be a micro-positioner.

[0095] The real-time positioning data can be understood as positioning data acquired in real time. Optionally, the positioning data can include latitude and longitude data and / or location data of the helmet in the operation map.

[0096] The target cloud server is a cloud server of a hard hat management platform, wherein the hard hat management platform can monitor and display information of one or more hard hats in real time.

[0097] The communication device can be used to transmit the information of the hard hat terminal to the hard hat management platform.

[0098] Based on the above embodiment scheme, the effect of transmitting the positioning data of the hard hat terminal to the hard hat management platform in real time is achieved, so that the hard hat management platform can monitor the hard hat terminal in real time and send relevant instructions to the hard hat terminal based on the monitoring information.

[0099] Optionally, a call button may be arranged on the helmet; wherein,

[0100] The distress button is used to generate a distress signal in response to a button triggering operation, and send the distress signal to the controller;

[0101] The controller is further used to send the distress signal to the target cloud server through the communication device.

[0102] The call button can be used to call for help. The button triggering operation can be understood as an operation to trigger the call button. Exemplarily, the button triggering operation can be pressing or continuously pressing the call button.

[0103] Based on the above embodiment, the emergency call button is deployed on the safety helmet to realize a one-button emergency call function.

[0104] Optionally, a memory may be disposed on the safety helmet, wherein:

[0105] The memory is used to store the real-time image data to obtain stored image data;

[0106] The controller is further used to send the stored image data to the target cloud server through the communication device.

[0107] Wherein, the memory can be used to store image data in real time.

[0108] Based on the above-mentioned embodiment scheme, the helmet management platform can replay and analyze the operation process based on the stored image data.

[0109] Optionally, the controller is also used to determine a target data volume of the data to be processed, and when the target data volume exceeds a data volume threshold, send the data to be processed to a target cloud server, and receive a data processing result returned by the target cloud server, wherein the data to be processed includes at least one of the real-time image data, the real-time gas data, the real-time wearing data, and the real-time positioning data.

[0110] The data to be processed can be understood as data to be processed. In an embodiment of the present invention, the data to be processed may include image data to be risk assessed acquired in real time by the camera device, gas data to be detected acquired in real time by the gas sensor, wearing data to be detected by hat removal acquired in real time by the infrared detector, and positioning data to be stored acquired in real time by the positioning device.

[0111] The target data volume can be understood as the data volume of the data to be processed. Generally, the data volume of the data to be processed at different time points can be different or the same.

[0112] The data volume threshold can be understood as a threshold used to determine whether the data volume reaches the upper limit of the controller's computing capacity. In the embodiment of the present invention, the data volume threshold can be preset according to the scene requirements and is not specifically limited here.

[0113] The data processing result can be understood as the result obtained by the target cloud server performing operations on the data to be processed. Corresponding to the above-mentioned data to be processed, the data processing result can include risk assessment results, gas detection results, cap removal detection results, and stored image data.

[0114] Based on the above embodiment scheme, when the computing intensity corresponding to the data volume of the data to be processed exceeds the computing capacity upper limit of the controller, the data to be processed is uploaded to the target cloud server, so that the target cloud server performs the calculation of the data to be processed (usually, the computing capacity of the target cloud server is stronger than the controller on the safety helmet), and returns the calculation result, thereby avoiding the occurrence of excessive data volume, excessive controller load, long calculation time and delayed feedback of processing results, thereby ensuring the efficient operation of various functions of the safety helmet and the smooth progress of the operation.

[0115] Optionally, the alarm is also used to receive a target signal and perform a light alarm and / or voice alarm corresponding to the target signal, wherein the target signal includes the risk warning signal or the cap removal alarm signal.

[0116] Based on the above embodiment scheme, the alarm is implemented to perform targeted warnings corresponding to the target signal based on the received target signal, such as hat removal warnings or risk warnings, so that the staff can understand the specific risks that trigger the warnings.

[0117] The safety helmet of the present invention is equipped with a gas sensor, a camera device, a controller and an alarm, an infrared detector, a navigator, a positioning device, a communication device, a call button, a memory and a lighting lamp. The safety helmet of the present invention can simultaneously realize functions such as image acquisition, risk assessment (collapse risk assessment and large mobile equipment detection), gas data acquisition, gas risk detection (oxygen concentration, toxic gas concentration and harmful gas concentration), risk alarm, cap removal detection, cap removal warning, navigation, positioning, communication between the safety helmet and the management platform, call for help, data storage and lighting, so that the workers can carry out safe and efficient limited space operations by wearing the helmet safely.

Claims

1. A personal protective helmet for working in confined spaces, characterized in that: The safety helmet is equipped with a gas sensor, a camera device, a controller and an alarm; wherein, The camera device is used to obtain real-time image data of the working environment; The gas sensor is used to obtain real-time gas data of the working environment; The controller is used to perform risk assessment on the real-time image data through a risk assessment model to obtain a risk assessment result, determine a gas detection result according to the real-time gas data, generate a risk warning signal when the risk assessment result and / or the gas detection result meet a risk warning condition, and send the risk warning signal to the alarm, wherein the risk assessment model is obtained by training a convolutional neural network based on training samples; The alarm is used to receive the risk warning signal and issue a risk warning according to the risk warning signal.

2. The safety helmet according to claim 1, characterized in that: The controller is used to input the real-time image data into a collapse detection model and a moving object detection model respectively to obtain a collapse detection result and a moving object detection result respectively, wherein the collapse detection result represents the collapse risk level of the working environment, and the moving object detection result represents whether there are moving objects of the target category in the working environment, the collapse detection model is obtained by training the convolutional neural network based on collapse samples, and the moving object detection model is obtained by training the convolutional neural network based on moving object samples.

3. The safety helmet according to claim 1, characterized in that: The controller is used to determine a preset target concentration threshold, and determine the gas detection result based on the target concentration threshold and the real-time gas data, wherein the target concentration threshold is related to the target wearer of the helmet, and the real-time gas data at least includes the gas concentration of the target gas, and the target gas includes at least oxygen, toxic gas and harmful gas.

4. The safety helmet according to claim 1, characterized in that: Also includes an infrared detector; wherein, The infrared detector is used to obtain real-time wearing data of the helmet, wherein the wearing data represents the contact between the helmet and the head of the target wearer; The controller is further configured to determine a hat-off detection result according to the real-time wearing data, generate a hat-off alarm signal when the hat-off detection result meets a hat-off alarm condition, and send the hat-off warning signal to the alarm; The alarm is also used for receiving the hat-off warning signal and performing a hat-off alarm according to the hat-off warning signal.

5. The safety helmet according to claim 1, characterized in that: Also included is a navigator; wherein, The controller is further used to modify a preset original operation map according to the real-time image data to obtain a target operation map, determine real-time navigation data according to the real-time positioning data and the target operation map, generate a real-time navigation signal according to the real-time navigation data, and send the real-time navigation signal to the navigator, wherein the operation map at least includes an operation end point; The navigator is used for navigating according to the received real-time navigation signal.

6. The safety helmet according to claim 1, characterized in that: Also includes a positioning device; wherein, The positioning device is used to obtain real-time positioning data of the helmet; The controller is also used to send the real-time positioning data to a target cloud server through a communication device, wherein the target cloud server is a cloud server of a hard hat management platform.

7. The safety helmet according to claim 6, characterized in that: It also includes a call button; The distress button is used to generate a distress signal in response to a button triggering operation, and send the distress signal to the controller; The controller is further used to send the distress signal to the target cloud server through the communication device.

8. The safety helmet according to claim 6, characterized in that: Also includes: Memory, where The memory is used to store the real-time image data to obtain stored image data; The controller is further used to send the stored image data to the target cloud server through the communication device.

9. The safety helmet according to claims 1, 4 and 6, characterized in that: The controller is further used to determine a target data volume of the data to be processed, and when the target data volume exceeds a data volume threshold, send the data to be processed to a target cloud server, and receive a data processing result returned by the target cloud server, wherein the data to be processed includes at least one of the real-time image data, the real-time gas data, the real-time wearing data, and the real-time positioning data.

10. The safety helmet according to claims 1 and 4, characterized in that: The alarm is also used for receiving a target signal and performing a light alarm and / or a voice alarm corresponding to the target signal, wherein the target signal includes the risk warning signal or the cap removal alarm signal.

Images