Method and system for identifying wearing condition of safety helmet during maintenance of sample preparation system

Abstract

The invention belongs to the technical field of safety monitoring, and discloses a method and a system for identifying the wearing condition of a safety helmet when a sample preparation system is overhauled. Comprising the following steps: obtaining a maintainer image: processing the maintainer image to obtain a to-be-detected image; obtaining a reference image corresponding to the to-be-detected image, wherein the reference image is an image that the safety helmet is correctly worn; performing gray processing on the to-be-detected image to obtain a first gray image; performing gray processing on the reference image to obtain a second gray image; based on the first gray level image and the second gray level image, attribute features of the to-be-detected image and the reference image are extracted; calculating the similarity A of the to-be-detected image and the reference image through the attribute features; when the similarity A is within a safety threshold range, it is judged that the safety helmet is correctly worn; and when the similarity A is out of the safety threshold range, judging that the safety helmet is not worn correctly. According to the invention, the safety helmet wearing condition of the maintainer can be identified, the manual detection cost is reduced, and the personal safety of the maintainer is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of safety monitoring technology, and in particular to a method and system for identifying whether safety helmets are worn during sample preparation system maintenance. Background Technology

[0002] In coal-based industries such as power, steel, and cement, automated sample preparation technology has become a trend. It has fundamentally changed the traditional manual sample preparation method, achieving full automation from sample collection to preparation, greatly reducing the need for human intervention. This technology not only improves the accuracy and repeatability of sample preparation but also reduces the possibility of human error, thereby enhancing the quality control level of the entire industrial production process. Especially with the integration of robotics, the development of automated sample preparation systems has entered a new stage. Robotic sample preparation systems, with their high transparency, intelligence, and efficiency, have become a major highlight in the field of industrial sample preparation. This system can prepare individual samples independently or continuously, greatly improving sample preparation efficiency and reducing labor costs.

[0003] However, all equipment has its maintenance cycle, even highly automated robotic sample preparation systems. During regular maintenance and overhauls of robotic sample preparation systems, operation must be paused, introducing a new safety issue—how to ensure the safety of maintenance personnel during maintenance. In industrial environments, due to the complexity of equipment and limited space, maintenance personnel face various potential safety risks, especially mechanical collision injuries. To effectively protect the safety of maintenance personnel during operations, especially in situations where falling objects or mechanical collisions may occur, wearing a safety helmet has become a basic safety measure. Safety helmets can significantly reduce the risk of head injuries and are an indispensable part of the personal protective equipment (PPE) for maintenance personnel.

[0004] Currently, in the monitoring of the sample preparation system's work area, it is both time-consuming and labor-intensive to rely on long-term viewing of the monitoring screens by human eyes or to arrange for personnel to regularly inspect the work area and observe whether maintenance personnel are wearing safety helmets during maintenance work. In large-scale centralized monitoring, a sufficient number of inspection or monitoring personnel are needed to frequently check in order to timely monitor all monitoring screens or the safety of maintenance personnel in the patrol area. This not only wastes manpower, but also causes inspection and monitoring personnel to miss abnormal images due to fatigue. At the same time, additional personnel entering and leaving the maintenance work area also increases the risk of accidents. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for identifying the wearing status of safety helmets during the maintenance of a sample preparation system, which is used to identify whether maintenance personnel are wearing safety helmets, reduce manual inspection costs, and ensure the personal safety of maintenance personnel.

[0006] To achieve this objective, the present invention adopts the following technical solution:

[0007] Methods for identifying whether safety helmets are worn during sample preparation system maintenance include:

[0008] Acquire, acquire images of maintenance personnel;

[0009] Preprocessing involves processing the image of the maintenance personnel to obtain the image to be inspected; and obtaining a reference image corresponding to the image to be inspected, wherein the reference image is an image of a safety helmet being worn correctly.

[0010] Grayscale processing is performed on the image to be detected to obtain a first grayscale image; grayscale processing is performed on the reference image to obtain a second grayscale image.

[0011] Extraction: Based on the first grayscale image and the second grayscale image, extract the attribute features of the image to be detected and the reference image, respectively;

[0012] The similarity A between the image to be detected and the reference image is calculated based on the attribute features.

[0013] If the similarity A is within the safety threshold range, it is determined that the maintenance personnel have correctly worn the safety helmet; if the similarity A is outside the safety threshold range, it is determined that the maintenance personnel have not correctly worn the safety helmet.

[0014] As an optional method for identifying the wearing status of safety helmets during sample preparation system maintenance, the preprocessing step includes one or more processing methods such as sharpening processing, noise reduction processing, contrast enhancement processing, size adjustment processing, and head position positioning processing.

[0015] As an alternative method for identifying the wearing status of safety helmets during sample preparation system maintenance, the image to be detected and the reference image are adjusted to the same size before grayscale processing.

[0016] As an optional method for identifying the wearing status of safety helmets during the maintenance of a sample preparation system, the similarity A is calculated using a structural similarity index.

[0017] As an optional method for identifying the wearing status of safety helmets during the maintenance of a sample preparation system, the similarity A is obtained by comprehensively using a first comparison term, a second comparison term, and a third comparison term, wherein the first comparison term, the second comparison term, and the third comparison term are the average brightness l(x,y), contrast c(x,y), and structural similarity s(x,y) of the image to be detected and the reference image, respectively.

[0018] As an alternative method for identifying the wearing status of safety helmets during sample preparation system maintenance, the attribute features include color attributes, shape attributes, and texture attributes.

[0019] As an optional method for identifying the wearing status of safety helmets during sample preparation system maintenance, when it is determined that the maintenance personnel are not wearing safety helmets correctly, an alarm is triggered or a real-time alarm message is sent to the maintenance personnel.

[0020] As an optional method for identifying the wearing status of safety helmets during sample preparation system maintenance, when it is determined that the maintenance personnel are not wearing safety helmets correctly, the personnel's identity information and the corresponding information on the incorrect wearing of safety helmets are stored.

[0021] The identification system, based on the method for identifying the wearing status of safety helmets during the maintenance of the sample preparation system as described in any of the above schemes, includes:

[0022] The image acquisition module is used to capture and collect images of the maintenance personnel;

[0023] The image processing module is used to perform grayscale conversion on the image to be detected and the reference image;

[0024] The feature extraction module extracts attribute features from the image to be detected and the reference image;

[0025] The similarity calculation module is used to calculate the similarity A between the image to be detected and the reference image.

[0026] As an optional solution for the identification system, the identification system also includes an alarm module.

[0027] Beneficial effects:

[0028] In the first aspect of this invention, images of maintenance personnel are collected before they enter the construction site. The initially acquired images are preprocessed to ensure the efficiency and accuracy of the comparison. Grayscale processing simplifies calculations, reducing the required storage space and computing resources, thus accelerating image processing and analysis. Furthermore, grayscale processing highlights features, making feature extraction more effective. Further, attribute features of the first and second grayscale images are extracted from the image to be detected and the reference image, respectively, to ensure the accuracy of subsequent comparisons. A similarity A is calculated to determine the degree of similarity between the two images, ensuring the reliability of the judgment criteria and the accuracy of the judgment results. A safe threshold range exists for similarity A. When the calculated similarity A is within the safe threshold range, the system or a human determines that the maintenance personnel are correctly wearing safety helmets and meet the entry requirements. If the calculated similarity A is outside the safe threshold range (e.g., exceeding the safe threshold range), it is determined that the maintenance personnel are not correctly wearing safety helmets, and the system or a human indicates that the maintenance personnel do not meet the entry requirements. This identification method can identify whether maintenance personnel are wearing safety helmets, eliminating the need for manual identification of each individual, thus reducing the cost of manual inspection. In addition, the accuracy of the identification results can be guaranteed during the image preprocessing and comparison process, thereby ensuring the personal safety of maintenance personnel.

[0029] In a second aspect of the invention, the identification system can effectively reduce the detection cost of manually identifying whether maintenance personnel are wearing safety helmets, thereby ensuring the personal safety of maintenance personnel. Attached Figure Description

[0030] Figure 1 This is a flowchart of a method for identifying the wearing status of safety helmets during the maintenance of a sample preparation system, provided in an embodiment of the present invention. Detailed Implementation

[0031] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0032] In the description of this invention, unless otherwise explicitly specified and limited, the terms "connected," "linked," and "fixed" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0033] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0034] In the description of this embodiment, the terms "upper," "lower," "right," etc., refer to the orientation or positional relationship shown in the accompanying drawings. They are used only for ease of description and simplification of operation, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention. In addition, the terms "first" and "second" are used only for distinction in description and have no special meaning.

[0035] This embodiment focuses on a smart fuel sampling system. Based on advanced automation, information, and artificial intelligence technologies, this system intelligently upgrades the traditional manual sampling process, achieving unmanned and automated control. The system can accurately and efficiently process various coal samples, providing accurate and reliable samples for subsequent testing and analysis. During the regular maintenance and overhaul of the smart fuel sampling system, ensuring the safety of maintenance personnel is paramount.

[0036] Please see the appendix Figure 1 The first aspect of this embodiment relates to a method for identifying the wearing status of a safety helmet during sample preparation system maintenance (hereinafter referred to as the "identification method"), which includes the following steps:

[0037] Step S01: Acquire images of maintenance personnel.

[0038] Before maintenance personnel enter the construction site, surveillance cameras or other image acquisition devices can be installed at predetermined locations, such as in front of a security door. When maintenance personnel are in the predetermined location, full-body or upper-body images of the maintenance personnel can be directly captured through the shooting function of the surveillance camera or image acquisition device. It should be noted that the images of maintenance personnel can be captured once or multiple times. Capturing multiple times is beneficial to obtain complete images of the maintenance personnel from multiple angles to obtain more complete identification information, which can be provided and selected for subsequent preprocessing steps.

[0039] Step S02: Preprocessing. Process the image of the maintenance personnel to obtain the image to be inspected; obtain the reference image corresponding to the image to be inspected, which is an image of the safety helmet being worn correctly.

[0040] The images of maintenance personnel initially acquired need to be preprocessed to filter out interference and retain identifiable information, ensuring the efficiency and accuracy of the comparison. In this step, the collected images of maintenance personnel are processed to obtain the image to be tested. Simultaneously, a corresponding reference image is matched onto the image to be tested. The reference image should be a pre-taken image of the same maintenance personnel correctly wearing a safety helmet, so that it can be used as a standard image for subsequent comparison.

[0041] Step S03: Grayscale processing. The image to be detected is processed in grayscale to obtain a first grayscale image; the reference image is processed in grayscale to obtain a second grayscale image.

[0042] Grayscale processing allows for the use of internal algorithms to convert both the image to be detected and the reference image into grayscale, resulting in a first grayscale image and a second grayscale image, respectively. Grayscale processing simplifies computation, requiring less storage space and computational resources, thus accelerating image processing and analysis. Furthermore, key features in an image (such as edges and textures) are independent of color but primarily depend on brightness variations; grayscale processing can highlight these features, making feature extraction more effective.

[0043] Step S04: Extraction. Based on the first grayscale image and the second grayscale image, extract the attribute features of the image to be detected and the reference image, respectively.

[0044] Furthermore, for the first grayscale image and the second grayscale image, the attribute features of the image to be detected and the reference image are extracted respectively, especially the head features of the maintenance personnel, i.e. the safety helmet attributes, to ensure the accuracy of subsequent comparisons.

[0045] Step S05: Compare and calculate the similarity A between the image to be detected and the reference image based on attribute features;

[0046] The similarity A is calculated to determine the degree of similarity between the two, ensuring the reliability of the judgment criteria and the accuracy of the judgment results.

[0047] Step S06: Determine if the similarity A is within the safety threshold range, then the maintenance personnel are determined to have worn safety helmets correctly; if the similarity A is outside the safety threshold range, then the maintenance personnel are determined to have not worn safety helmets correctly.

[0048] In this embodiment, there is a safety threshold range for similarity A. When the calculated similarity A is within the safety threshold range, the system or a human determines that the maintenance personnel have correctly worn their safety helmets and that the maintenance personnel meet the access conditions. Once the calculated similarity A is outside the safety threshold range (e.g., exceeding the safety threshold range), it is determined that the maintenance personnel have not correctly worn their safety helmets, and the system or a human prompts the maintenance personnel that they do not meet the access conditions.

[0049] This identification method can identify whether maintenance personnel are wearing safety helmets, eliminating the need for manual identification of each individual, thus reducing the cost of manual inspection. In addition, the accuracy of the identification results can be guaranteed during the image preprocessing and comparison process, thereby ensuring the personal safety of maintenance personnel.

[0050] Optionally, step S02 includes one or more processing methods such as sharpening, noise reduction, contrast enhancement, size adjustment, and head position positioning.

[0051] The preprocessing steps include, but are not limited to, sharpening, noise reduction, contrast enhancement, size adjustment, and head positioning. Multiple processing methods can even be used together to simplify the process and improve efficiency.

[0052] Image sharpening can utilize existing algorithms and sharpening tools to enhance image clarity; image denoising can use filters to remove noise, further filtering out interference and improving image clarity while simplifying calculations; contrast enhancement improves visual appeal, making the image more suitable for machine analysis and suppressing unwanted information; resizing ensures a clear image size; and head positioning ensures the accurate placement of the helmeted head within the image.

[0053] Optionally, the image to be detected and the reference image are resized to the same size before grayscale processing.

[0054] Unifying the image to be detected and the reference image to the same size ensures data consistency and avoids processing complexity caused by different sizes. This consistency is crucial for subsequent grayscale processing and analysis, as it ensures that the algorithm compares and calculates on the same basis. Furthermore, adjusting to the same size facilitates the standardization of input size construction, thereby simplifying the data processing flow. In image processing, images of different sizes may require different processing methods and parameters. By unifying all images to the same size, the same processing methods and parameters can be used to process all images, thus simplifying the processing flow. Further, in some safety helmet recognition scenarios, it may be necessary to focus on a specific key area in the image, such as the head of a maintenance worker. By unifying the image size, this key area can be better focused on, and irrelevant information can be removed, thereby improving the accuracy and robustness of the algorithm.

[0055] Optionally, the similarity A is calculated using the Structural Similarity Index (SSIM). Specifically, the similarity A is obtained by combining the first comparison term, the second comparison term, and the third comparison term, where the first comparison term, the second comparison term, and the third comparison term are the average brightness l(x,y), the contrast c(x,y), and the structural similarity s(x,y) of the image to be detected and the reference image, respectively.

[0056] The Structural Similarity Index (SSIM) is a metric that measures the similarity (A) between two images. Based on the human visual system's perception of image structure, it assesses image similarity (A) from three aspects: brightness, contrast, and structure. By modeling distortion as a combination of these three different factors, it more accurately evaluates image similarity (A). Specifically, the SSIM calculates the similarity between two images in terms of average brightness l(x,y), contrast c(x,y), and structural similarity s(x,y), and then multiplies these similarity values ​​to obtain the final SSIM(x,y) value.

[0057] Specifically, calculate the average brightness l(x, y) (mean) of the two images:

[0058]

[0059] Where μ x and μ y C1 is the local mean of the image position coordinates x and y, and C1 is a constant set to avoid the denominator being zero.

[0060] Calculate the contrast c(x, y) (standard deviation) of the two images:

[0061]

[0062] Where σ x and σy C1 is the local standard deviation of the image position coordinates x and y, and C2 is a constant set to avoid the denominator being zero.

[0063] The structural similarity s(x, y) between two images is calculated by comparing their local covariances.

[0064]

[0065] Where σ xy C1 is the local covariance of the image position coordinates x and y, and C2 is a constant set to avoid the denominator being zero.

[0066] Combining the three comparison terms above, the final SSIM value is calculated:

[0067] SSIM(x,y)=l(x,y)·c(x,y)·s(x,y)

[0068] To calculate the SSIM(x, y) of the entire image, a small sliding window (exemplarily 11x11 pixels) is typically used to slide across the image, and the SSIM(x, y) value within each window is calculated. The SSIM(x, y) values ​​from all windows are then averaged to obtain the SSIM of the entire image. Constants C1, C2, and C3 are used to avoid zero denominators or to make the comparison terms more numerically stable. They are determined by the dynamic range L of the image; for example, for an 8-bit image, L = 2. 8 -1 = 255, where C1 = (K1·L)2; C2 = (K2·L)2; C3 = C2 / 2. Here, K1 and K2 are very small constants, and K1 ≤ 0.01, K2 ≤ 0.03.

[0069] Through the above steps, the SSIM values ​​of the two images are obtained. The SSIM value ranges from [-1, 1], where 1 indicates that the two images are completely identical and 0 indicates that the two images are completely dissimilar.

[0070] A positive value close to 1 indicates that the two images are very similar.

[0071] A positive value close to 0 indicates that the two images have a low degree of similarity.

[0072] Negative values ​​indicate that the two images are not only dissimilar, but also have a certain negative correlation in structure.

[0073] An SSIM value of -1 indicates an extremely rare occurrence, as it requires very extreme variations in image contrast, brightness, and structure. If an SSIM value of -1 is found, the system needs to check for potential problems with the input image data, such as an image where all pixel values ​​have been incorrectly inverted.

[0074] A SSIM value of -0.1 indicates a decrease in image quality or a significant change in image content, but it does not necessarily mean the image is completely unusable. This value needs to be evaluated in the context of the specific application scenario. For example, in video compression or image transmission applications, an SSIM value of -0.1 might mean that the reconstructed image quality is poor, requiring the acquisition of a new image for recalculation of the similarity.

[0075] Optionally, the attribute features include color attributes, shape attributes, and texture attributes.

[0076] In this embodiment, the similarity A involves three dimensions of attribute features: color attribute, shape attribute, and texture attribute, especially the color attribute, shape attribute, and texture attribute of the safety helmet in the image.

[0077] Optionally, when it is determined that the maintenance personnel are not wearing safety helmets correctly, an alarm is triggered or a real-time alarm message is sent to the maintenance personnel.

[0078] In this embodiment, alarm information can be displayed directly on the monitoring screen or transmitted in real time to the mobile device worn by maintenance personnel; additionally, alarm signals can be emitted by audible and visual alarms. Triggering the alarm can promptly remind maintenance personnel that they are not wearing their safety helmets correctly.

[0079] Furthermore, when it is determined that the maintenance personnel are not wearing safety helmets correctly, the system stores the maintenance personnel's identity information and the corresponding information regarding the incorrect wearing of safety helmets.

[0080] If it is determined that the maintenance personnel failed to wear safety helmets correctly, the personnel's identification information can be further linked to the information regarding the failure to wear safety helmets correctly, and the relevant information can be reported to the safety management personnel so that appropriate safety measures can be taken.

[0081] The second aspect of this embodiment also relates to an identification system based on the above-described method for identifying the wearing status of safety helmets during sample preparation system maintenance. This identification system includes an image acquisition module, an image processing module, a feature extraction module, and a similarity calculation module. The image acquisition module is used to capture and collect images of maintenance personnel; the image processing module is used to perform grayscale conversion on the image to be detected and a reference image; the feature extraction module extracts attribute features from the image to be detected and the reference image; and the similarity calculation module is used to calculate the similarity between the image to be detected and the reference image. This system can integrate the hardware and software of existing conventional machine vision processing systems to realize the related tasks of image acquisition, image processing, feature extraction, and similarity calculation.

[0082] This identification system effectively reduces the detection cost of manually identifying whether maintenance personnel are wearing safety helmets, and ensures the personal safety of maintenance personnel.

[0083] Furthermore, the identification system also includes an alarm module.

[0084] The alarm module can be a conventional audible and visual alarm. When maintenance personnel do not wear safety helmets correctly, the audible and visual alarm will be triggered to notify the person or safety management personnel in a timely manner.

[0085] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art will be able to make various obvious changes, readjustments, and substitutions without departing from the scope of protection of the present invention. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for identifying the wearing status of safety helmets during sample preparation system maintenance, characterized in that, include: Acquire, acquire images of maintenance personnel; Preprocessing involves processing the images of the maintenance personnel to obtain the images to be inspected; Obtain a reference image corresponding to the image to be detected, wherein the reference image is an image of a helmet being worn correctly; Grayscale processing: The image to be detected is processed to obtain a first grayscale image; The reference image is processed to obtain a second grayscale image; Extraction: Based on the first grayscale image and the second grayscale image, extract the attribute features of the image to be detected and the reference image, respectively; The similarity A between the image to be detected and the reference image is calculated based on the attribute features. If the similarity A is within the safety threshold range, it is determined that the maintenance personnel have correctly worn the safety helmet; if the similarity A is outside the safety threshold range, it is determined that the maintenance personnel have not correctly worn the safety helmet.

2. The method for identifying the wearing status of safety helmets during sample preparation system maintenance according to claim 1, characterized in that, The preprocessing steps include one or more processing methods such as sharpening, noise reduction, contrast enhancement, size adjustment, and head position positioning.

3. The method for identifying the wearing status of safety helmets during sample preparation system maintenance according to claim 1, characterized in that, Before the grayscale processing, the image to be detected and the reference image are adjusted to the same size.

4. The method for identifying the wearing status of safety helmets during sample preparation system maintenance according to claim 1, characterized in that, The similarity A is calculated using the structural similarity index.

5. The method for identifying the wearing status of safety helmets during sample preparation system maintenance according to claim 4, characterized in that, The similarity A is obtained by combining the first comparison term, the second comparison term, and the third comparison term, where the first comparison term, the second comparison term, and the third comparison term are the average brightness l(x,y), contrast c(x,y), and structural similarity s(x,y) of the image to be detected and the reference image, respectively.

6. The method for identifying the wearing status of safety helmets during sample preparation system maintenance according to claim 1, characterized in that, The attribute features include color attributes, shape attributes, and texture attributes.

7. The method for identifying the wearing status of safety helmets during sample preparation system maintenance according to claim 1, characterized in that, When it is determined that the maintenance personnel are not wearing safety helmets correctly, an alarm is triggered or a real-time alarm message is sent to the maintenance personnel.

8. The method for identifying the wearing status of safety helmets during sample preparation system maintenance according to claim 1, characterized in that, When it is determined that the maintenance personnel are not wearing safety helmets correctly, the personnel's identity information and the corresponding information regarding the incorrect wearing of safety helmets are stored.

9. An identification system, characterized in that, Based on the method for identifying the wearing status of safety helmets during sample preparation system maintenance as described in any one of claims 1-8, the identification system includes: The image acquisition module is used to capture and collect images of the maintenance personnel; The image processing module is used to perform grayscale conversion on the image to be detected and the reference image; The feature extraction module extracts attribute features from the image to be detected and the reference image; The similarity calculation module is used to calculate the similarity between the image to be detected and the reference image.

10. The identification system according to claim 9, characterized in that, The identification system also includes an alarm module.

Images