On-duty state detection method and device for operating personnel in limited space

By collecting images in a limited space and evaluating their quality and the characteristics of the operators, combining facial and wearable information, the problem of incomplete detection is solved, the accuracy and safety of on-duty personnel are improved, and the timely response to potential risks is ensured. Identify and respond.

CN119027880APending Publication Date: 2024-11-26GUANGDONG POWER GRID CO LTD +1
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411177818.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-26
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

In a limited space operation environment, the existing technology does not have comprehensive considerations for on-the-job personnel, resulting in low detection accuracy and safety, which may lead to inability to respond in a timely manner and identify potential risks when an accident occurs.

Method used

By collecting images from a limited space, determining whether their quality is qualified, combining the facial features and clothing characteristics of the worker, determining whether they are the target personnel, and evaluating their on-the-job status based on the working status characteristics to ensure multiple considerations in the detection.

Benefits of technology

It improves the accuracy and safety of on-duty personnel testing, ensures that potential risks can be identified and responded to in a timely manner in a limited space, and reduces the possibility of accidents.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027880A_ABST
    Figure CN119027880A_ABST
Patent Text Reader

Abstract

The invention discloses an on-duty state detection method and device for an operator in a limited space, and relates to the technical field of image recognition. The method comprises the following steps: determining a qualified image from a plurality of images collected in a limited space; judging whether the operator in the qualified image is a target operator or not according to the facial features and the clothes wearing features of the operator in the qualified image; under the condition that the operating personnel in the qualified image are target personnel, acquiring working state characteristics of the target personnel in the qualified image; and determining the on-duty state of the target personnel according to the working state characteristics. According to the method and the device, the technical problems of low on-duty personnel detection accuracy and low on-duty personnel safety caused by incomplete consideration of factors for on-duty personnel detection in a limited space in the prior art are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to a method and device for detecting the on-the-job status of workers in a confined space. Background Art

[0002] In the power industry, especially in confined space working environments (such as underground cable wells), the safety risks are high and the consequences of accidents are serious. Therefore, it is necessary to strengthen the monitoring and management of on-site workers to ensure that they comply with safe operating procedures. These confined space operations are usually characterized by closedness, poor ventilation and small space. In such a complex working environment, it is easy to cause workers to suffer from dangerous conditions such as hypoxia and suffocation.

[0003] Therefore, in a confined space working environment, detecting the identity and compliance of on-the-job personnel is crucial to improving the safety level of the operation. However, the current power grid confined space operation supervision method based on face recognition still has some shortcomings. Specifically, the traditional face recognition system has a slow processing speed, and the factors considered in the detection process of on-the-job personnel are not comprehensive enough. In the confined space operation of the power grid, any safety hazard may cause serious consequences. If the face recognition processing speed is slow, the information of the on-the-job personnel cannot be identified and confirmed in time, which may lead to the inability to respond quickly when an accident occurs. Incomplete detection factors may cause key safety information to be ignored, resulting in potential risks not being identified and controlled in time.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The present application provides a method and device for detecting the on-the-job status of workers in a confined space, so as to at least solve the technical problem in the prior art that the factors considered in the detection of on-the-job workers in a confined space are not comprehensive enough, resulting in low accuracy of on-the-job worker detection and low safety of on-the-job workers.

[0006] According to one aspect of the present application, a method for detecting the on-the-job status of workers in a confined space is provided, including: determining a qualified image from multiple images collected in the confined space, wherein an image parameter value of the qualified image is greater than a preset parameter value; judging whether the worker in the qualified image is a target person based on facial features and clothing features of the worker in the qualified image, wherein the target person is a trusted person who meets the preset clothing requirements in the confined space, and the trusted person is a worker allowed to enter the confined space; in the case where the worker in the qualified image is the target person, obtaining the working status features of the target person in the qualified image; determining the on-the-job status of the target person based on the working status features, wherein the on-the-job status is used to characterize the difference between the actual working status of the target person and the preset working status.

[0007] Optionally, determining a qualified image from multiple images captured within a limited space includes: determining an image evaluation value based on the image captured within the limited space, wherein the image evaluation value is used to quantify the clarity of the image; obtaining a target evaluation value, wherein the target evaluation value is a minimum image quality value corresponding to a preset image quality; determining that the image is a qualified image when the evaluation value is greater than or equal to the target evaluation value; and determining that the image is an unqualified image when the evaluation value is less than the target evaluation value.

[0008] Optionally, determining an evaluation value of an image based on an image captured in a limited space includes: determining an image resolution, an image contrast, and a difference between image brightness and a preset brightness based on the image captured in the limited space; determining an evaluation value of the image based on the image resolution, image contrast, and a difference between image brightness and a preset brightness.

[0009] Optionally, judging whether the operator in the qualified image is the target person based on the facial features and clothing features of the operator in the qualified image, includes: determining facial feature values ​​based on the facial features of the operator in the qualified image, wherein the facial feature values ​​are used to quantify the overall facial appearance of the operator; determining clothing feature values ​​based on the clothing features of the operator in the qualified image, wherein the clothing feature values ​​are used to quantify the similarity between the actual wearing condition of the trusted person and the preset wearing requirements; judging whether the operator in the qualified image is the target person based on the facial feature values ​​and clothing feature values.

[0010] Optionally, facial feature values ​​are determined based on the facial features of the operator in the qualified image, including: determining the operator's eye distance, atrium length, and lateral mouth distance based on the facial features; determining the operator's facial feature values ​​based on the eye distance, atrium length, and lateral mouth distance.

[0011] Optionally, determining a clothing feature value based on clothing features of the operator in the qualified image includes: determining a first difference value for the operator based on the clothing features, wherein the first difference value is the difference between the number of safety equipment actually worn on the operator's whole body and a first target number, the first target number being the number of safety equipment that the operator needs to wear on the whole body in a limited space; determining a second difference value for the operator based on the clothing features, wherein the second difference value is the difference between the number of safety equipment actually worn on the operator's key parts and a second target number, the second target number being the number of safety equipment that needs to be worn on the operator's key parts, namely, parts of the operator's body; determining the clothing feature value based on the first difference value and the second difference value.

[0012] Optionally, judging whether the operator in the qualified image is a target person based on facial feature values ​​and wearing clothing feature values ​​includes: obtaining a first feature value set and a second feature value, wherein the first feature value set is a set of preset facial feature values ​​of all trusted persons in a limited space, and the second feature value is used to characterize the minimum wearing clothing feature value corresponding to when the wearing state of the trusted person meets the preset wearing requirements; when the facial feature value is included in the first feature value set, determining that the operator in the qualified image is a trusted person; when the wearing clothing feature value of the operator in the qualified image is greater than or equal to the second feature value, determining that the operator in the qualified image is a target person; when the wearing clothing feature value of the operator in the qualified image is less than the second feature value, determining that the operator in the qualified image is a non-target person.

[0013] Optionally, in the case that the operator in the qualified image is the target person, after obtaining the working status characteristics of the target person in the qualified image, the on-the-job status detection method for the operator in the confined space includes: determining the target person's movement frequency in the confined space, the target person's actual heart rate and the target person's stationary time in the confined space according to the target person's working status characteristics; determining the target person's working status characteristic value according to the movement frequency, the actual heart rate and the stationary time, wherein the working status characteristic value is used to quantify the similarity between the target person's actual working status and the preset working status.

[0014] Optionally, determining the on-the-job status of the target personnel according to the work status characteristics includes: obtaining a third characteristic value, wherein the third characteristic value is used to characterize the minimum work status characteristic value corresponding to the actual work status of the target personnel that meets the preset work status; when the work status characteristic value is greater than or equal to the third characteristic value, determining that the on-the-job status of the target personnel is a normal state; when the work status characteristic value is less than the third characteristic value, determining that the on-the-job status of the target personnel is an abnormal state.

[0015] According to another aspect of the present application, there is also provided an on-the-job status detection device for workers in a confined space, comprising: a first determination unit, which determines a qualified image from images collected in the confined space, wherein the qualified image is an image that meets the preset image requirements in the confined space, and the image at least includes image resolution, image contrast, and the absolute difference between image brightness and preset brightness; a first judgment unit, which judges whether the worker in the qualified image is a target person based on the facial features and clothing features of the worker in the qualified image, wherein the target person is a trusted person who meets the preset clothing requirements in the confined space, and the trusted person is a worker allowed to enter the confined space; a first acquisition unit, which acquires the working status characteristics of the target person in the qualified image when the worker in the qualified image is the target person; a second determination unit, which determines the on-the-job status of the target person based on the working status characteristics, wherein the on-the-job status is used to characterize the difference between the actual working status of the target person and the preset working status requirements.

[0016] From the above content, it can be seen that in this application, first determine whether the quality of the image collected in the limited space is qualified to ensure that the acquisition equipment can accurately capture and record important information at the work site to avoid the loss of key details due to image blur or distortion. Then, after determining that the acquired image is a qualified image, determine whether the operator is a trusted person based on the facial features of the operator in the qualified image, and then determine whether the trusted person is a target person based on the clothing features of the operator in the qualified image. Finally, if the operator in the qualified image is the target person, determine the on-the-job status of the target person based on the work status characteristics of the target person.

[0017] Compared with the prior art that detects the on-the-job status of operators based on a single factor, the present application analyzes the on-the-job status of operators through multiple aspects such as image quality, operator identity, operator wearing status and operator working status, thereby improving the accuracy of on-the-job personnel detection and the safety of on-the-job personnel, and further solving the technical problem that the prior art does not consider comprehensive factors for on-the-job personnel detection in a limited space, resulting in low accuracy of on-the-job personnel detection and low safety of on-the-job personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 It is a flowchart of an optional on-the-job status detection method for workers in a limited space according to an embodiment of the present application;

[0020] Figure 2 It is a schematic diagram of an optional on-the-job status detection device for workers in a confined space according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application 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 interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] It should also be noted that the information and data collected by this application are authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation portals for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or institutions. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or institution through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or institution.

[0024] According to an embodiment of the present application, an example of a method for detecting the on-the-job status of workers in a confined space is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0025] It should be noted that an intelligent detection system can be used as the execution subject of the on-the-job status detection method for workers in a confined space provided in the embodiment of the present application. It is understandable that the on-the-job status detection method for workers in a confined space provided in the embodiment of the present application can also be executed by other systems or devices, and the embodiment of the present application does not specifically limit this.

[0026] Figure 1 is a flow chart of an optional method for detecting the on-the-job status of workers in a limited space according to an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:

[0027] Step S101, determining a qualified image from a plurality of images collected in a limited space.

[0028] In step S101, the image parameter value of the qualified image is greater than the preset parameter value.

[0029] Optionally, the intelligent detection system acquires images within a limited space through monitoring equipment (such as a depth camera, a 3D scanning device).

[0030] Optionally, the information in the image includes but is not limited to image resolution, image contrast, and image brightness.

[0031] Step S102, judging whether the operator in the qualified image is the target person based on the facial features and clothing features of the operator in the qualified image.

[0032] In step S102, the target person is a trusted person who meets the preset wearing requirements in the limited space, and the trusted person is a working person allowed to enter the limited space.

[0033] Optionally, the facial features include but are not limited to the distance between the operator's eyes, the length of the atrium, and the lateral distance between the mouth.

[0034] Optionally, the wearing clothing features include, but are not limited to, the number of safety equipment worn on the operator's entire body and the number of safety equipment worn on key parts of the operator's body.

[0035] Step S103: when the operator in the qualified image is the target person, obtain the working status features of the target person in the qualified image.

[0036] Optionally, the working status characteristics include but are not limited to the working movement frequency, heart rate, and stationary time of the operator.

[0037] Step S104, determining the on-the-job status of the target personnel according to the working status characteristics.

[0038] In step S104, the on-the-job status is used to characterize the difference between the actual working status of the target personnel and the preset working status.

[0039] Optionally, the intelligent detection system can determine the on-the-job status of the target person based on the working status characteristics of the target person.

[0040] It can be known from the contents of step S101 to step S104 that in the present application, a qualified image is first determined from a plurality of images collected in a confined space, wherein the image parameter value of the qualified image is greater than the preset parameter value, and then, based on the facial features and clothing features of the operator in the qualified image, it is determined whether the operator in the qualified image is a target person, wherein the target person is a trusted person who meets the preset clothing requirements in the confined space, and the trusted person is an operator allowed to enter the confined space, and then, if the operator in the qualified image is the target person, the working status features of the target person in the qualified image are obtained, and finally, the on-the-job status of the target person is determined based on the working status features, wherein the on-the-job status is used to characterize the difference between the actual working status of the target person and the preset working status.

[0041] From the above content, it can be seen that in this application, first determine whether the quality of the image collected in the limited space is qualified to ensure that the acquisition equipment can accurately capture and record important information at the work site to avoid the loss of key details due to image blur or distortion. Then, after determining that the acquired image is a qualified image, determine whether the operator is a trusted person based on the facial features of the operator in the qualified image, and then determine whether the trusted person is a target person based on the clothing features of the operator in the qualified image. Finally, if the operator in the qualified image is the target person, determine the on-the-job status of the target person based on the work status characteristics of the target person.

[0042] Compared with the prior art that detects the on-the-job status of operators based on a single factor, the present application analyzes the on-the-job status of operators through multiple aspects such as image quality, operator identity, operator wearing status and operator working status, thereby improving the accuracy of on-the-job personnel detection and the safety of on-the-job personnel, and further solving the technical problem that the prior art does not consider comprehensive factors for on-the-job personnel detection in a limited space, resulting in low accuracy of on-the-job personnel detection and low safety of on-the-job personnel.

[0043] In an optional embodiment, the intelligent detection system determines an evaluation value of an image based on an image captured in a limited space, wherein the evaluation value of the image is used to quantify the clarity of the image, and then obtains a target evaluation value, wherein the target evaluation value is a minimum image quality value corresponding to a preset image quality. When the evaluation value is greater than or equal to the target evaluation value, the image is determined to be a qualified image, and when the evaluation value is less than the target evaluation value, the image is determined to be an unqualified image.

[0044] Optionally, the intelligent detection system obtains an image quality assessment value (the above-mentioned assessment value) based on a comprehensive analysis of the image data collected by the monitoring equipment, and compares the image quality assessment value with the image quality reference assessment value (the above-mentioned target assessment value) stored in a database. If the image quality assessment value is higher than or equal to the image quality reference assessment value, it indicates that the image quality is qualified; if the image quality assessment value is lower than the image quality reference assessment value, it indicates that the image quality is unqualified.

[0045] Optionally, the intelligent detection system can use a comprehensive analysis method to determine the image quality assessment value based on the actual collected data and then make decisions. Data-based decision-making is usually more accurate and reliable than relying solely on subjective judgment, thereby reducing the risk of wrong decisions. In addition, by establishing clear image quality assessment standards and databases, the intelligent detection system can track and record the performance of monitoring equipment over a long period of time, providing data support for subsequent analysis, while also enhancing the credibility of the monitoring equipment and helping operators better identify and handle potential safety incidents.

[0046] From the above content, it can be seen that the intelligent detection system can ensure that the monitoring equipment can accurately capture and record important information on the scene by judging whether the quality of the images collected by the monitoring equipment is qualified, avoid the loss of key details due to blurred or distorted images, ensure the clarity of the image, and provide reliable data for subsequent analysis of the on-the-job status of the operators.

[0047] In an optional embodiment, the intelligent detection system determines the image resolution, image contrast, and the difference between the image brightness and the preset brightness based on the image captured in a limited space, and then determines the evaluation value of the image based on the image resolution, image contrast, and the difference between the image brightness and the preset brightness.

[0048] Optionally, image resolution is used to measure image clarity, image contrast is used to measure the difference in brightness of an image, and the absolute value of the difference between image brightness and reference brightness (the above-mentioned preset brightness) is used to measure the overall brightness of the image.

[0049] Optionally, the intelligent detection system obtains image resolution through basic information of the surveillance camera, obtains image contrast through the surveillance camera and an image processing algorithm, and obtains image brightness through the sensor and exposure control of the surveillance camera.

[0050] It should be noted that image brightness directly affects the perception of contrast. Appropriate brightness can make the image contrast more obvious, and resolution itself does not directly affect brightness. However, during image acquisition and processing, higher-resolution images can usually better present details under different brightness conditions, because high resolution means that the image contains more pixels and can capture more details and information. This is also particularly important in terms of contrast performance, because high-resolution images can make subtle differences in light and dark more obvious. Therefore, appropriate resolution, contrast, and brightness can improve image clarity and detail performance.

[0051] Optionally, the intelligent detection system normalizes the resolution, contrast, brightness and absolute value of the difference between the captured image and the reference brightness, and then evaluates the image resolution to ensure that the details of the captured image are clear and discernible, which is convenient for subsequent analysis and processing, such as face recognition or object monitoring, and then judges the visualization quality of the image based on the absolute value of the difference between the brightness and the reference brightness to ensure that the monitoring screen is clearly visible under different lighting conditions.

[0052] Optionally, the image quality evaluation value α is calculated as shown in formula (1):

[0053]

[0054] Among them, α is the image quality assessment value, fb is the image resolution, db is the image contrast, ld is the absolute value of the difference between the image brightness and the reference brightness, ε1 is the compensation factor of the set image resolution, ε2 is the compensation factor of the set image contrast, and ε3 is the compensation factor of the absolute value of the difference between the set image brightness and the reference brightness.

[0055] Optionally, the intelligent detection system establishes a mapping relationship between historical image resolution and compensation factor of image resolution, a mapping relationship between historical image contrast and compensation factor of image contrast, and a mapping relationship between the absolute value of the difference between historical image brightness and reference brightness and the compensation factor of the absolute value of the difference based on the image data historically collected by the monitoring device. A mapping set is constructed based on these mapping relationships, and this mapping set is stored in a database. Then, based on the mapping set in the database, the compensation factors corresponding to the image resolution collected by the current monitoring device, the image contrast collected by the monitoring device, and the absolute value of the difference between the image brightness collected by the monitoring device and the reference brightness can be obtained.

[0056] From the above content, it can be seen that the intelligent detection system determines the image resolution, image contrast and the difference between the image brightness and the preset brightness based on the images collected in the limited space to obtain the image quality assessment value, and quantifies the clarity of the image. Through this method of quantitative evaluation of image quality, the deficiencies of the monitoring equipment can be discovered and corrected in a timely manner to ensure that the image resolution, brightness and contrast are qualified. It can not only evaluate the quality of the monitoring equipment, but also help operators identify which equipment or locations may need to be upgraded or maintained, so as to reasonably allocate resources and extend the service life of the equipment.

[0057] In an optional embodiment, the intelligent detection system determines facial feature values ​​based on the facial features of the operator in the qualified image, wherein the facial feature values ​​are used to quantify the overall facial appearance of the operator, and then determines wearing clothing feature values ​​based on the wearing clothing features of the operator in the qualified image, wherein the wearing clothing feature values ​​are used to quantify the similarity between the actual wearing conditions of the trusted person and the preset wearing requirements, and finally determines whether the operator in the qualified image is the target person based on the facial feature values ​​and the wearing clothing feature values.

[0058] Optionally, the intelligent detection system obtains facial feature values ​​based on a comprehensive analysis of the facial features of the workers in the qualified image, obtains clothing feature values ​​based on a comprehensive analysis of the clothing features of the workers in the qualified image, and then determines whether the worker in the qualified image is the target person based on the facial feature values ​​and clothing feature values ​​obtained by analysis.

[0059] From the above content, it can be seen that the intelligent detection system can comprehensively judge the facial information and wearing conditions of the operators to ensure that the operators meet the preset standards. This method not only improves the accuracy and reliability of identity authentication, but also effectively strengthens the management of the operators' wearing specifications. Ultimately, it realizes the intelligent monitoring of the operators' identity and wearing requirements, and enhances overall safety and compliance.

[0060] In an optional embodiment, the intelligent detection system determines the operator's eye distance, atrium length and lateral mouth distance based on facial features, and then determines the operator's facial feature values ​​based on the eye distance, atrium length and lateral mouth distance.

[0061] Optionally, the interocular distance refers to the straight-line distance between the two eyes, which is an important component of facial features, the facial atrium length refers to the distance from the center of the forehead to the bottom of the nose, and the facial mouth lateral distance refers to the straight-line distance between the left and right end points of the lips.

[0062] Optionally, the intelligent detection system will normalize the worker's eye distance, atrium length, and lateral distance between mouths to facilitate comparison and analysis. By analyzing features such as eye distance, atrium length, and lateral distance between mouths, the intelligent detection system can more accurately identify and match the corresponding worker among a large number of faces, reducing misidentification and missed identification.

[0063] Optionally, the facial feature value ω is calculated as shown in formula (2):

[0064]

[0065] Among them, ω is the facial feature value of the operator in the qualified image, yj is the interocular distance, zt is the atrium length, zc is the lateral distance of the mouth, σ1 is the compensation factor of the set interocular distance, σ2 is the compensation factor of the set atrium length, σ3 is the compensation factor of the set interocular distance, and e is a natural constant.

[0066] Optionally, the intelligent detection system establishes a mapping set of the interocular distance, atrium length, and lateral mouth distance of the workers in the historical images and the compensation factors of the interocular distance, atrium length, and lateral mouth distance based on the historically collected image data, and stores this mapping set in a database. Then, based on the database, the compensation factors corresponding to the interocular distance, atrium length, and lateral mouth distance of the workers collected by the current monitoring device can be obtained.

[0067] From the above content, it can be seen that the intelligent detection system first extracts the operator's eye distance, atrium length and lateral mouth distance based on the image, and then calculates the operator's facial feature values ​​based on these data, quantifies the operator's facial features, and provides a basis for the subsequent determination of trusted personnel.

[0068] In an optional embodiment, the intelligent detection system determines a first difference value of the operator based on the characteristics of the worn clothing, wherein the first difference value is the difference between the number of safety equipment actually worn on the operator's whole body and the first target number, and the first target number is the number of safety equipment that the operator needs to wear on the whole body in a limited space. Then, a second difference value of the operator is determined based on the characteristics of the worn clothing, wherein the second difference value is the difference between the number of safety equipment actually worn on the operator's key parts and the second target number, and the second target number is the number of safety equipment that needs to be worn on the operator's key parts, which are part of the operator's body parts. Finally, the characteristic value of the worn clothing is determined based on the first difference value and the second difference value.

[0069] Optionally, the actual number of safety equipment worn on the whole body refers to the number of devices (such as wearable devices, electronic tags, identification documents, etc.) worn on the whole body of the operator identified in the image, and the number of reference devices worn (the first goal mentioned above) refers to the number of equipment worn on the whole body set under specific standards. By calculating the difference between these two numbers, the credibility of the trusted person in compliance or identity recognition can be evaluated.

[0070] Optionally, the number of safety equipment actually worn on key parts refers to the number of devices worn by the trusted personnel on specific parts (such as wrists, necks, feet, etc.), and the reference number of wearing parts (the second target number mentioned above) is the number of equipment worn on specific parts based on existing standards or requirements. If the difference between the two is large, it may indicate that the operator may not comply with the regulations or there may be safety hazards.

[0071] Optionally, the intelligent detection system normalizes the absolute value of the difference between the number of safety equipment actually worn on the operator's whole body and the number of reference equipment worn, and the absolute value of the difference between the number of safety equipment actually worn on the operator's key parts and the number of reference wearing parts, to facilitate comparison and analysis. By analyzing the wearing status of the equipment, it can effectively monitor whether the operator wears safety equipment (such as helmets, safety belts, etc.) in accordance with regulations, thereby improving the safety of the working environment and reducing the risk of accidents. When the intelligent detection system detects that there is a large difference in the number of equipment worn on the whole body or the number of equipment worn on certain parts, it can quickly remind the operator to make timely adjustments to ensure the effectiveness of safety measures. It can also evaluate the behavior of the operators, such as analyzing whether there is improper wearing, arbitrary removal, etc., so as to provide a basis for improving management measures. Collecting and analyzing wearing data can help management understand the overall compliance rate and compliance.

[0072] Optionally, the calculation method of the clothing characteristic value δ is as shown in formula (3):

[0073]

[0074] Among them, δ is the characteristic value of the clothing worn by the trusted person, cn is the first difference of the trusted person, bn is the second difference of the trusted person, τ1 is the compensation factor of the set first difference of the trusted person, τ2 is the compensation factor of the set second difference of the trusted person, and e is a natural constant.

[0075] Optionally, the intelligent detection system establishes a mapping set of the first difference, the second difference and the compensation factor of the first difference, and the compensation factor of the second difference of the operating personnel in the historical images based on the historically collected image data, and stores this mapping set in a database. Then, the compensation factor of the first difference and the compensation factor of the second difference of the operating personnel collected by the current monitoring device can be obtained based on the database.

[0076] From the above content, it can be seen that the intelligent detection system obtains the characteristic value of wearing clothing according to the number of safety equipment actually worn by the operators and the number of safety equipment actually worn on key parts, quantifies the wearing status of the operators, and provides a basis for the subsequent determination of target personnel. By establishing evaluation standards and databases for equipment wearing status, monitoring and management work can be more standardized. Enterprises or organizations can clarify the equipment wearing standards of various types of operators in different places, improve management efficiency, and timely compare and analyze the equipment wearing status of operators, which helps to promptly discover personnel who fail to wear equipment as required, and take corresponding measures to correct them and reduce potential risks. Data-based analysis can not only judge the eligibility of equipment wearing status, but also accumulate a large amount of data for subsequent analysis.

[0077] In an optional embodiment, the intelligent detection system obtains a first feature value set and a second feature value, wherein the first feature value set is a set of preset facial feature values ​​of all trusted persons in a limited space, and the second feature value is used to characterize the minimum wearing clothing feature value corresponding to when the wearing state of the trusted person meets the preset wearing requirements. When the facial feature value is included in the first feature value set, the operator in the qualified image is determined to be a trusted person. When the wearing clothing feature value of the operator in the qualified image is greater than or equal to the second feature value, the operator in the qualified image is determined to be a target person. When the wearing clothing feature value of the operator in the qualified image is less than the second feature value, the operator in the qualified image is determined to be a non-target person.

[0078] Optionally, the intelligent detection system compares the facial feature value with the facial reference feature values ​​(the above-mentioned facial preset feature values) collected by all monitoring devices stored in the database. If there is a facial reference feature value equal to the facial feature value, the operator corresponding to the facial feature value is a trusted person. If there is no facial reference feature value equal to the facial feature value, the operator corresponding to the facial feature value is an untrusted person. After determining that the operator in the image is an untrusted person, an alarm is immediately issued to warn the operator. In this way, potential security threats can be quickly discovered and illegal activities can be prevented. Once an untrusted person is identified, the relevant operator can Measures can be taken quickly. At the same time, automatic identification and alarm of untrusted personnel reduce the pressure of manual monitoring and improve efficiency. It allows managers to focus more on areas that really need attention, rather than checking each screen one by one. The system helps reduce human misjudgment and false alarms through automated processing of feature value comparisons. The intelligent detection system can also store alarm records in the database to provide data support for subsequent incident investigation and analysis, which helps to understand the patterns and trends of security incidents. By analyzing the frequency and location of untrusted personnel, security managers can adjust and optimize security strategies and resource allocation to ensure enhanced monitoring and preventive measures in high-risk areas.

[0079] Optionally, after determining the trusted person, the intelligent detection system detects the wearing status of the trusted person, and compares the trusted person's wearing clothing characteristic value with the trusted person's equipment wearing clothing reference characteristic value (the above-mentioned second characteristic value) stored in the database. If the trusted person's wearing clothing characteristic value is higher than or equal to the equipment wearing clothing reference characteristic value, then the trusted person's wearing status is qualified and he is the target person. If the trusted person's wearing clothing characteristic value is lower than the equipment wearing clothing reference characteristic value, then the trusted person's wearing status is unqualified and he is a non-target person. In this way, it is ensured that all operating personnel wear the required equipment as required, which helps to respond quickly in emergency situations, reduce safety hazards, and effectively improve overall safety.

[0080] From the above content, it can be seen that the intelligent detection system can determine whether the operator is a trusted person who meets the preset clothing requirements in a confined space by analyzing the operator's facial feature values ​​and clothing feature values. In this way, the reliability of the operator's identity and the safety of the operator can be ensured.

[0081] In an optional embodiment, the intelligent detection system determines the target person's movement frequency in a limited space, the target person's actual heart rate, and the target person's stationary time in a limited space based on the target person's work status characteristics, and then determines the target person's work status characteristic value based on the movement frequency, actual heart rate, and stationary time, wherein the work status characteristic value is used to quantify the similarity between the target person's actual work status and the preset work status.

[0082] Optionally, the intelligent detection system obtains the target person's movement frequency through an acceleration sensor, obtains the target person's active heart rate through a heart rate monitor, and obtains the target person's stationary time through a motion tracker.

[0083] Optionally, movement frequency refers to the number of times or distance the target person moves within a certain period of time, reflecting the target person's activity level and work intensity. For example, in a specific work environment, frequent movement may mean high-intensity work, while a lower movement frequency may mean that the work is easier or inactive; heart rate refers to the amplitude of heart rate changes, which can usually reflect the work intensity or physical load of the target person; stationary time refers to the time that the target person stays in place or does not move in the same position, which is usually used to evaluate the work mode of the staff.

[0084] Optionally, the working state characteristic value is calculated as shown in formula (4):

[0085]

[0086] Among them, γ is the working status characteristic value of the target person, bp is the movement frequency, xl is the absolute value of the difference between the heart rate and the reference activity heart rate, st is the static time, μ1 is the compensation factor of the set movement frequency, μ2 is the compensation factor of the absolute value of the difference between the set activity heart rate and the reference activity heart rate, and μ3 is the compensation factor of the set static time.

[0087] Optionally, the intelligent detection system establishes, based on historical data, a mapping relationship between the historically measured movement frequency of the operating personnel and the compensation factor of the movement frequency, a mapping relationship between the absolute value of the difference between the heart rate and the reference activity heart rate and the compensation factor of the absolute value of the difference between the heart rate and the reference activity heart rate, and a mapping relationship between the stillness time and the compensation factor of the stillness time. A mapping set is constructed based on these mapping relationships, and this mapping set is stored in a database. Then, based on the mapping set in the database, the compensation factors corresponding to the movement frequency, the absolute value of the difference between the heart rate and the reference activity heart rate, and the stillness time collected by the current monitoring device can be obtained.

[0088] Optionally, the intelligent detection system will normalize the target personnel's movement frequency, the absolute value of the difference between the active heart rate and the reference active heart rate, and the static time for easy comparison and analysis. The movement frequency and static time can be used to evaluate the target personnel's work efficiency, understand the target personnel's activity level in different working situations, and help managers identify efficient and inefficient work modes. By monitoring the heart rate, an automated early warning mechanism can be set up. When abnormal data is found (such as a heart rate that is too high or a movement frequency that is too low), the management or other operating personnel will be notified in a timely manner to prevent potential safety hazards. The work status characteristic values ​​under different working states can be used to analyze the target personnel's work habits and behavior patterns, thereby providing data support for personalized management.

[0089] From the above content, it can be seen that the intelligent detection system determines the target person's working status characteristic value based on the target person's movement frequency, actual heart rate and stillness time, which provides a basis for the subsequent determination of the target person's on-the-job status, and can prevent potential safety hazards by monitoring the target person's movement frequency, actual heart rate and stillness time.

[0090] In an optional embodiment, the intelligent detection system obtains a third characteristic value, wherein the third characteristic value is used to characterize the minimum working state characteristic value corresponding to the preset working state when the actual working state of the target personnel meets the preset working state. When the working state characteristic value is greater than or equal to the third characteristic value, the on-the-job state of the target personnel is determined to be a normal state. When the working state characteristic value is less than the third characteristic value, the on-the-job state of the target personnel is determined to be an abnormal state.

[0091] Optionally, the intelligent detection system compares the target person's work status characteristic value with the work status reference characteristic value (the third characteristic value mentioned above) stored in the database. If the target person's work status characteristic value is lower than the work status reference characteristic value, the target person's work status is unqualified, that is, the on-the-job status is abnormal; if the target person's work status characteristic value is higher than or equal to the work status reference characteristic value, the target person's work status is qualified, that is, the on-the-job status is normal. The target person with a qualified work status is marked as an actual on-the-job person, and all the on-the-job personnel stored in the database except the actual on-the-job personnel are marked as off-the-job personnel.

[0092] From the above content, it can be seen that the intelligent detection system can identify the actual on-duty personnel and the off-duty personnel by judging the working status of the target personnel. In this way, work tasks can be effectively allocated to ensure that there are suitable candidates for key positions and to achieve more effective resource allocation. For target personnel with unqualified working status, they can be marked as off-duty in a timely manner, which can effectively reduce safety hazards, ensure the safety of the workplace, and allow qualified on-duty personnel to give full play to their advantages and improve overall work efficiency.

[0093] According to an embodiment of the present application, an embodiment of a device for detecting the on-the-job status of workers in a confined space is also provided. Figure 2 is a schematic diagram of an optional on-the-job status detection device for workers in a limited space according to an embodiment of the present application, such as Figure 2 As shown, the device for detecting the on-the-job status of workers in a limited space includes: a first determining unit 201 , a first judging unit 202 , a first acquiring unit 203 , and a second determining unit 204 .

[0094] Optionally, the first determination unit 201 is used to determine a qualified image from the images collected in the limited space, wherein the qualified image is an image that meets the preset image requirements in the limited space, and the image at least includes image resolution, image contrast, and the absolute difference between image brightness and preset brightness; the first judgment unit 202 is used to judge whether the operator in the qualified image is a target person based on the facial features and clothing features of the operator in the qualified image, wherein the target person is a trusted person who meets the preset clothing requirements in the limited space, and the trusted person is an operator allowed to enter the limited space; the first acquisition unit 203 is used to acquire the working status characteristics of the target person in the qualified image when the operator in the qualified image is the target person; the second determination unit 204 is used to determine the on-the-job status of the target person based on the working status characteristics, wherein the on-the-job status is used to characterize the difference between the actual working status of the target person and the preset working status requirements.

[0095] Optionally, the first determination unit 201 includes: a first determination subunit, a first acquisition subunit, a second determination subunit, and a third determination subunit. The first determination subunit is used to determine an evaluation value of an image based on an image collected in a limited space, wherein the evaluation value of the image is used to quantify the clarity of the image; the first acquisition subunit is used to obtain a target evaluation value, wherein the target evaluation value is a minimum image quality value corresponding to a preset image quality; the second determination subunit is used to determine that the image is a qualified image when the evaluation value is greater than or equal to the target evaluation value; and the third determination subunit is used to determine that the image is an unqualified image when the evaluation value is less than the target evaluation value.

[0096] Optionally, the first determination subunit includes: a first determination module and a second determination module. The first determination module is used to determine the image resolution, image contrast, and the difference between the image brightness and the preset brightness according to the image collected in the limited space; the second determination module is used to determine the evaluation value of the image according to the image resolution, image contrast, and the difference between the image brightness and the preset brightness.

[0097] Optionally, the first judgment unit 202 includes: a fourth determination subunit, a fifth determination subunit and a first judgment subunit. The fourth determination subunit is used to determine a facial feature value based on the facial features of the operator in the qualified image, wherein the facial feature value is used to quantify the overall face of the operator; the fifth determination subunit is used to determine a clothing feature value based on the clothing features of the operator in the qualified image, wherein the clothing feature value is used to quantify the similarity between the actual wearing condition of the trusted person and the preset wearing requirement; the first judgment subunit is used to determine whether the operator in the qualified image is a target person based on the facial feature value and the clothing feature value.

[0098] Optionally, the fourth determination subunit includes: a third determination module and a fourth determination module. The third determination module is used to determine the interocular distance, atrium length and lateral distance between the mouths of the operator according to the facial features; the fourth determination module is used to determine the facial feature value of the operator according to the interocular distance, atrium length and lateral distance between the mouths.

[0099] Optionally, the fifth determination subunit includes: a fifth determination module, a sixth determination module and a seventh determination module. The fifth determination module is used to determine the first difference of the operator according to the characteristics of the clothing worn, wherein the first difference is the difference between the number of safety equipment actually worn by the operator and the first target number, and the first target number is the number of safety equipment that the operator needs to wear on the whole body in a limited space; the sixth determination module is used to determine the second difference of the operator according to the characteristics of the clothing worn, wherein the second difference is the difference between the number of safety equipment actually worn on the key parts of the operator and the second target number, and the second target number is the number of safety equipment that needs to be worn on the key parts of the operator, and the key parts are part of the body of the operator; the seventh determination module is used to determine the characteristic value of the clothing worn according to the first difference and the second difference.

[0100] Optionally, the first judgment subunit includes: a first acquisition module, an eighth determination module, a ninth determination module and a tenth determination module. The first acquisition module is used to acquire a first feature value set and a second feature value, wherein the first feature value set is a set of preset facial feature values ​​of all trusted persons in a limited space, and the second feature value is used to characterize the minimum clothing feature value corresponding to the wearing state of the trusted person when it meets the preset wearing requirements; the eighth determination module is used to determine that the operator in the qualified image is a trusted person when the facial feature value is included in the first feature value set; the ninth determination module is used to determine that the operator in the qualified image is a target person when the clothing feature value of the operator in the qualified image is greater than or equal to the second feature value; the tenth determination module is used to determine that the operator in the qualified image is a non-target person when the clothing feature value of the operator in the qualified image is less than the second feature value.

[0101] Optionally, the first acquisition unit 203 includes: a sixth determination subunit and a seventh determination subunit. The sixth determination subunit is used to determine the target person's movement frequency in the limited space, the target person's actual heart rate, and the target person's stationary time in the limited space according to the target person's work status characteristics; the seventh determination subunit is used to determine the target person's work status characteristic value according to the movement frequency, the actual heart rate, and the stationary time, wherein the work status characteristic value is used to quantify the similarity between the target person's actual work status and the preset work status.

[0102] Optionally, the second determination unit 204 includes: a second acquisition subunit, an eighth determination subunit, and a ninth determination subunit. The second acquisition subunit is used to acquire a third characteristic value, wherein the third characteristic value is used to characterize the minimum working state characteristic value corresponding to the actual working state of the target person in accordance with the preset working state; the eighth determination subunit is used to determine that the on-the-job state of the target person is a normal state when the working state characteristic value is greater than or equal to the third characteristic value; the ninth determination subunit is used to determine that the on-the-job state of the target person is an abnormal state when the working state characteristic value is less than the third characteristic value.

[0103] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0104] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0105] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0106] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0107] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0108] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0109] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for detecting the on-the-job status of workers in a limited space, characterized in that: include: Determining a qualified image from a plurality of images collected within a limited space, wherein an image parameter value of the qualified image is greater than a preset parameter value; According to the facial features and clothing features of the operator in the qualified image, determine whether the operator in the qualified image is a target person, wherein the target person is a trusted person who meets the preset clothing requirements in the limited space, and the trusted person is an operator allowed to enter the limited space; When the operator in the qualified image is the target person, obtaining working status features of the target person in the qualified image; The on-the-job status of the target person is determined according to the working status characteristics, wherein the on-the-job status is used to characterize the difference between the actual working status of the target person and the preset working status.

2. The method for detecting the on-the-job status of workers in a limited space according to claim 1, characterized in that: Determine qualified images from multiple images collected in a limited space, including: Determining an evaluation value of the image according to the image collected in the limited space, wherein the evaluation value of the image is used to quantify the clarity of the image; Obtaining a target evaluation value, wherein the target evaluation value is a minimum image quality value corresponding to a preset image quality; When the evaluation value is greater than or equal to the target evaluation value, determining that the image is a qualified image; In the case where the evaluation value is less than the target evaluation value, the image is determined to be a failed image.

3. The method for detecting the on-the-job status of workers in a limited space according to claim 2, characterized in that: Determining an evaluation value of the image according to the image collected in the limited space includes: Determining, according to the image collected in the limited space, the image resolution, the image contrast, and the difference between the image brightness and a preset brightness of the image; An evaluation value of the image is determined according to the image resolution, the image contrast, and a difference between the image brightness and a preset brightness.

4. The method for detecting the on-the-job status of workers in a limited space according to claim 1, characterized in that: Judging whether the operator in the qualified image is a target person according to the facial features and clothing features of the operator in the qualified image includes: Determining facial feature values ​​according to facial features of the operator in the qualified image, wherein the facial feature values ​​are used to quantify the overall facial appearance of the operator; Determining a clothing feature value according to the clothing feature of the operator in the qualified image, wherein the clothing feature value is used to quantify the similarity between the actual wearing condition of the trusted operator and the preset wearing requirement; It is determined whether the operator in the qualified image is the target person according to the facial feature value and the clothing feature value.

5. The method for detecting the on-the-job status of workers in a limited space according to claim 4, characterized in that: Determining facial feature values ​​according to facial features of the operator in the qualified image includes: Determine the interocular distance, atrium length, and lateral distance of the mouth of the operator according to the facial features; The facial feature value of the operator is determined according to the interocular distance, the atrium length, and the lateral distance of the mouth.

6. The method for detecting the on-the-job status of workers in a limited space according to claim 4, characterized in that: Determining a clothing feature value according to the clothing feature of the operator in the qualified image includes: Determining a first difference value of the operator according to the wearing clothing feature, wherein the first difference value is the difference between the number of safety equipment actually worn by the operator and a first target number, and the first target number is the number of safety equipment that the operator needs to wear in the limited space; Determine a second difference of the operator according to the wearing clothing feature, wherein the second difference is the difference between the number of safety equipment actually worn on the key part of the operator and a second target number, and the second target number is the number of safety equipment required to be worn on the key part of the operator, and the key part is a part of the body of the operator; The wearing clothing feature value is determined according to the first difference and the second difference.

7. The method for detecting the on-the-job status of workers in a limited space according to claim 4, characterized in that: Judging whether the operator in the qualified image is a target person according to the facial feature value and the clothing feature value includes: Obtaining a first feature value set and a second feature value, wherein the first feature value set is a set of preset facial feature values ​​of all trusted persons within the limited space, and the second feature value is used to represent the minimum wearing clothing feature value corresponding to when the wearing state of the trusted person meets the preset wearing requirement; In a case where the facial feature value is included in the first feature value set, determining that the operator in the qualified image is the trusted person; When the clothing feature value of the operator in the qualified image is greater than or equal to the second feature value, determining the operator in the qualified image as the target person; When the clothing feature value of the operator in the qualified image is smaller than the second feature value, the operator in the qualified image is determined to be a non-target person.

8. The method for detecting the on-the-job status of workers in a limited space according to claim 1, characterized in that: In the case where the operator in the qualified image is the target person, after obtaining the working status features of the target person in the qualified image, the on-the-job status detection method for the operator in the limited space includes: Determine the moving frequency of the target person in the limited space, the actual heart rate of the target person, and the stationary time of the target person in the limited space according to the working status characteristics of the target person; The working state characteristic value of the target person is determined according to the movement frequency, the actual heart rate and the stationary time, wherein the working state characteristic value is used to quantify the similarity between the actual working state of the target person and the preset working state.

9. The method for detecting the on-the-job status of workers in a limited space according to claim 8, characterized in that: Determining the on-the-job status of the target personnel according to the working status characteristics includes: Obtaining a third characteristic value, wherein the third characteristic value is used to represent a minimum working state characteristic value corresponding to when the actual working state of the target person meets the preset working state; When the working status characteristic value is greater than or equal to the third characteristic value, determining that the on-the-job status of the target person is a normal status; When the working status characteristic value is less than the third characteristic value, it is determined that the on-the-job status of the target person is an abnormal state.

10. A device for detecting the on-the-job status of workers in a limited space, characterized in that: include: A first determining unit is configured to determine a qualified image from images collected in a limited space, wherein the qualified image is an image that meets preset image requirements in the limited space, and the image at least includes image resolution, image contrast, and an absolute difference between image brightness and preset brightness; A first judgment unit is configured to judge whether the operator in the qualified image is a target person according to the facial features and clothing features of the operator in the qualified image, wherein the target person is a trusted person who meets the preset clothing requirements in the limited space, and the trusted person is an operator allowed to enter the limited space; a first acquisition unit, which, when the operator in the qualified image is the target person, acquires a working state feature of the target person in the qualified image; The second determining unit determines the on-the-job status of the target person according to the working status characteristics, wherein the on-the-job status is used to characterize the difference between the actual working status of the target person and the preset working status requirement.

Citation Information

Patent Citations

  • Personnel off-post detection method based on video information, control device and storage medium

    CN113936293A

  • Intelligent supervision system for safe operation

    CN115169777A

  • Limited space operator physical condition monitoring system based on wearable device

    CN116687363A

  • Image processing method and system for improving face recognition rate

    CN117011921A

  • Operator safety protection early warning device in limited space

    CN209625395U