Self-rescuer intelligent device management system

Abstract

The invention discloses a self-rescuer intelligent device management system, and relates to the technical field of intelligent management of safety protection equipment, and the system comprises the steps: collecting a self-rescuer sealing joint surface area image at an imaging station, generating a gray image, generating an illumination normalized image based on the gray image, and carrying out the illumination normalized image; extracting a seal combination seam point set in the illumination normalized image and calculating a mass center, determining an outer side sampling direction according to the mass center and obtaining a radial gray profile, calculating a deposition intensity sequence based on the radial gray profile and obtaining a wedge-shaped deposition black band index according to the deposition intensity sequence; and further reading the available capacity of the isolation bin, sequencing the self-rescuers to be distributed according to the index, performing isolation and distributable distribution, and updating the standing book state. According to the invention, intelligent quantification and automatic shunting of the self-rescuer sealing deposition state can be realized, and safety management is improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent management technology for safety protection equipment, and in particular to an intelligent device management system for self-rescue devices. Background Technology

[0002] Self-rescue devices are critical emergency respiratory protective equipment for workers in enclosed or semi-enclosed environments such as mines and tunnels. Their safety and reliability are directly related to the lives of workers, therefore traceability management is required throughout the entire process of requisition, wearing, return, storage, and maintenance. In the special working conditions of damp and dusty environments underground, the joint between the upper and lower covers of the self-rescue device is prone to the formation of deposits due to the adhesion and compaction of dust and moisture. Long-term accumulation of these deposits or their impact or compression during wear can lead to hidden deterioration of the sealing performance at the joint, thereby affecting the normal use of the self-rescue device in emergency scenarios and posing potential risks to the safety of workers.

[0003] Existing intelligent management systems for self-rescue devices mostly focus on basic management functions such as identity binding, inventory ledger statistics, expiration date monitoring, and entry / exit record retention. During the distribution or inspection process, the status check of self-rescue devices still mainly relies on quick manual visual observation. There is a lack of precise analysis methods for the state of sediment at the joint. It is impossible to objectively quantify the degree of sediment accumulation, nor is an automatic linkage mechanism between sediment status and self-rescue device distribution and isolation actions established. As a result, self-rescue devices with significant sediment at the joint that may affect sealing performance may still enter the distribution link through the existing management process. It is difficult to achieve a closed loop of safety management for the entire life cycle of self-rescue devices and to effectively intercept potential risks and hidden dangers from the source. Summary of the Invention

[0004] The purpose of this invention is to solve the problems existing in the prior art by proposing a self-rescue device intelligent management system.

[0005] To address the problems existing in the prior art, the present invention adopts the following technical solution: A self-rescue device intelligent management system includes: The image acquisition module is used to acquire images of the sealing and bonding surface area of ​​the self-rescue device at the imaging station and generate grayscale images; An illumination normalization module is used to generate an illumination normalized image based on the grayscale image; The suture and centroid module is used to extract the set of sealing seam points in the illumination normalized image and calculate the centroid of the set of sealing seam points. The profile sampling and intensity module is used to determine the outer sampling direction based on the centroid and the points of the sealing joint, and to perform profile sampling on the illumination normalized image along the outer sampling direction to generate a deposition intensity sequence. The wedge index module is used to divide the set of sealing joint points into an upper half-ring set and a lower half-ring set according to the centroid, and to calculate the wedge-shaped depositional black band index based on the depositional intensity sequence. The capacity distribution module is used to read the available capacity of the isolation chamber, sort the self-rescue devices to be issued according to the wedge-shaped deposition black band index, select the number of self-rescue devices corresponding to the available capacity of the isolation chamber to enter the isolation queue, and put the remaining self-rescue devices into the issuance queue, and update the ledger status.

[0006] Preferably, generating an illumination-normalized image based on the grayscale image includes: Calculate the mean image of the local window of the grayscale image; The grayscale image and the local window mean image are normalized to obtain an illumination-normalized image.

[0007] Preferably, extracting the set of sealing seam points from the illumination-normalized image and calculating the centroid of the set of sealing seam points includes: In the illumination-normalized image, boundary points are extracted based on the brightness abrupt change at the sealing joint to obtain the sealing joint point set; The centroid of the sealing joint point set is obtained by averaging the coordinates of each point in the set.

[0008] Preferably, the outer sampling direction is obtained by normalizing the reverse vector from the current point in the set of sealing seam points to the centroid.

[0009] Preferably, the illumination-normalized image is profiled along the outer sampling direction to generate a deposition intensity sequence, including: For each sealing point in the set of sealing joint points, a radial grayscale profile is obtained along the outer sampling direction; In the radial grayscale profile, the darkest point grayscale value and the background grayscale value are determined, and the average value of the darkest point grayscale value and the background grayscale value is used as the half-depth level. The width of the sedimentary zone is obtained by determining the distance between the left and right intersection points of the sedimentary zone based on the half-depth level. The blackness of the deposition zone is calculated based on the grayscale value of the darkest point and the grayscale value of the background. The deposition intensity at the corresponding sealing point is obtained by multiplying the width of the deposition zone by the blackness of the deposition zone, thus forming a deposition intensity sequence.

[0010] Preferably, the upper half-ring set is a set of points whose ordinates of the sealing joint points are less than the ordinate of the centroid, and the lower half-ring set is a set of points whose ordinates of the sealing joint points are greater than the ordinate of the centroid.

[0011] Preferably, calculating the wedge-shaped sedimentary black band index based on the sedimentation intensity sequence includes: Calculate the mean sedimentation intensity sequence corresponding to the lower half-ring set and the mean sedimentation intensity sequence corresponding to the upper half-ring set, respectively; The wedge-shaped sedimentary black band index is calculated based on the mean of the sedimentary intensity sequences corresponding to the lower and upper half-ring sets. The formula for calculating the wedge-shaped sedimentary black band index is as follows: In the formula, This represents the mean of the sedimentation intensity sequence corresponding to the lower half of the ring set. This represents the mean of the sedimentation intensity sequence corresponding to the upper half-ring set.

[0012] Preferably, the number of self-rescue devices entering the isolation queue is the smaller of the available capacity of the isolation chamber and the number of self-rescue devices currently to be issued.

[0013] Preferably, updating the ledger status includes: The self-rescue devices that enter the isolation queue are written into the isolation status log record, and the image of the sealing joint surface area and the wedge-shaped deposition black band index corresponding to the self-rescue device are stored. Write the self-rescue device that enters the dispensing queue into the dispensing status log record.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention acquires stable images of the sealing joint area using a fixed fixture and annular supplementary lighting component at the imaging station. After grayscale and illumination normalization processing, the interference caused by shell curvature and uneven supplementary lighting is reduced. Then, based on the brightness change feature of the sealing joint, a point set is extracted and the centroid is calculated. The width and blackness of the sedimentary zone are adaptively determined using the half-depth method. The directional distribution of sediments is transformed into an objectively comparable quantitative indicator through sediment intensity sequence and wedge-shaped sedimentary black band index. This eliminates the reliance on manual visual observation and realizes the digital and precise characterization of the degree of sediment accumulation, avoiding errors caused by subjective judgment.

[0015] 2. After reading the available capacity of the isolation chamber, this invention sorts the self-rescue devices to be issued according to the wedge-shaped deposition black band index, automatically selects the device with the highest risk to enter the isolation queue, and includes the remaining devices in the issuance link. At the same time, the ledger status is updated synchronously and traceability data such as images and indices are retained, realizing automated risk interception under resource constraints. The traceability of self-rescue device management is ensured through full-process data retention. Self-rescue devices with hidden deterioration risks in sealing performance are blocked from entering the use stage from the source, and the safety management closed loop of the entire life cycle of self-rescue devices is improved. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a functional block diagram of a self-rescue device intelligent management system according to the present invention; Figure 2 This is a schematic diagram of the intelligent management cabinet structure for the self-rescue device of the present invention; Figure 3 This is a schematic diagram of the directional distribution of the deposition zone on the sealing interface of the present invention; Figure 4 This is a schematic diagram of the sampling direction and half-depth width measurement of the outer side of the sealing joint of the present invention. Detailed Implementation

[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0018] Example: This example provides a self-rescue device intelligent management system, see [link to example]. Figure 1 Specifically, including: The image acquisition module is used to acquire images of the sealing and bonding surface area of ​​the self-rescue device at the imaging station and generate grayscale images; An illumination normalization module is used to generate an illumination normalized image based on the grayscale image; The suture and centroid module is used to extract the set of sealing seam points in the illumination normalized image and calculate the centroid of the set of sealing seam points. The profile sampling and intensity module is used to determine the outer sampling direction based on the centroid and the points of the sealing joint, and to perform profile sampling on the illumination normalized image along the outer sampling direction to generate a deposition intensity sequence. The wedge index module is used to divide the set of sealing joint points into an upper half-ring set and a lower half-ring set according to the centroid, and to calculate the wedge-shaped depositional black band index based on the depositional intensity sequence. The capacity distribution module is used to read the available capacity of the isolation chamber, sort the self-rescue devices to be issued according to the wedge-shaped deposition black band index, select the number of self-rescue devices corresponding to the available capacity of the isolation chamber to enter the isolation queue, and put the remaining self-rescue devices into the issuance queue, and update the ledger status.

[0019] In an embodiment of the present invention, an image of the sealing joint area of ​​the self-rescue device is acquired at the imaging station and a grayscale image is generated. It should be noted that the self-rescue device refers to an isolation breathing device used for emergency respiratory protection of workers in mines, tunnels and other enclosed or semi-enclosed environments. Its shell is usually made of metal or high-strength plastic and has a sealing structure with an upper cover and a lower cover. The image acquisition position or working unit of the imaging station usually includes a fixing fixture, lighting components and a camera device, which is used to image the self-rescue device in a fixed posture. This station can be set inside a dispensing cabinet, inspection cabinet or detection module. The sealing joint area of ​​the self-rescue device refers to the sealing joint area formed between the upper cover and the lower cover of the self-rescue device. Its sealing performance directly affects the airtightness and safety performance of the self-rescue device. This joint area is generally distributed in a ring or strip shape.

[0020] Specifically, the imaging station is a fixed imaging position set within the self-rescue device intelligent device management system. It consists of a fixing clamp for restricting the self-rescue device's posture, a ring light source for providing stable illumination, and a camera and lens that work in conjunction with the ring light source. During imaging, the self-rescue device to be tested is placed and held by the fixing clamp, ensuring the sealing surface of the self-rescue device faces the camera's imaging direction and maintains a stable relative position with the camera. Simultaneously, the ring light source is activated to create uniform illumination at the sealing surface, reducing imaging fluctuations caused by reflection differences and obtaining clear information on the texture and brightness variations of the sealing seam's neighborhood. Under these posture and illumination conditions, the controller triggers the camera to acquire images of the sealing surface area. These images refer to local digital image data including the sealing area between the upper and lower covers and its surrounding neighborhood. After collection, the pixel values ​​of the red, green, and blue channels in the image of the sealing interface area are synthesized into a grayscale image using a brightness-weighted method. This reduces the dependence of subsequent processing on color changes and improves the stability of characterizing differences in brightness, such as the deposited black band. Grayscale conversion can be achieved using existing weighted conversion methods based on brightness components. For example, the three channels can be synthesized into a single-channel brightness image using the brightness weights commonly used in television encoding. This allows the resulting grayscale image to more directly reflect the differences in reflection and absorption of the sealing interface under illumination. The clamping of the imaging station and the ring supplementary lighting ensure the repeatable acquisition conditions of the sealing interface area image. Grayscale conversion transforms the image data into a brightness-based expression, reducing the interference of illumination and color differences on the subsequent quantitative characterization of the wedge-shaped deposited black band at the lower edge of the sealing interface.

[0021] In an embodiment of the present invention, extracting the set of sealing seam points from the illumination-normalized image and calculating the centroid of the set of sealing seam points includes: Calculate the mean image of the local window of the grayscale image; The grayscale image and the local window mean image are normalized to obtain an illumination-normalized image; Specifically, the grayscale image is a single-channel brightness image obtained by grayscale conversion of the sealing surface region image. Its pixel values ​​are used to characterize the differences in reflection and absorption of the sealing surface under imaging illumination conditions. The local window mean image is a background brightness image formed by taking the grayscale image as input and calculating the brightness mean of each pixel position within its neighborhood. It is used to depict the slowly changing illumination distribution of the sealing surface caused by the ring light source illumination angle, shell curvature, and surface reflection. The local window mean image can be generated by performing window sliding average or equivalent low-pass smoothing on the grayscale image. That is, at each pixel, a square or circular neighborhood centered on that pixel is taken as a local window, and the brightness of all pixels within the local window is calculated arithmetically. The average value is written to the corresponding position to obtain a local window mean image with the same size as the original grayscale image and containing only low-frequency brightness variations. After obtaining the local window mean image, the grayscale image and the local window mean image are normalized to obtain an illumination-normalized image. Normalization means eliminating the background illumination component by dividing the pixel value of the grayscale image by the corresponding pixel value of the local window mean image at the pixel level. To avoid division by zero, a very small positive number can be added to the denominator. The result can be linearly scaled after normalization to keep the output pixels within the preset image dynamic range. This makes the illumination-normalized image highlight the local dark bands and depositional variations in the neighborhood of the sealing joint, while weakening large-scale gradual shadows and highlight drift. The slow brightness fluctuations of the sealing joint surface caused by shell geometry and illumination can be separated from the image, so that subsequent extraction of the sealing joint point set, sampling of the outer profile, and calculation of the deposition intensity sequence mainly respond to the local brightness decrease and bandwidth changes caused by the wedge-shaped depositional black band, thereby improving the stability of the wedge-shaped depositional black band index calculation.

[0022] In an embodiment of the present invention, extracting the set of sealing seam points from the illumination-normalized image and calculating the centroid of the set of sealing seam points includes: In the illumination-normalized image, boundary points are extracted based on the brightness abrupt change at the sealing joint to obtain the sealing joint point set; Specifically, the illumination-normalized image is a brightness-corrected image obtained by normalizing the grayscale image according to the local window mean. Its brightness changes can more effectively reflect the real structural boundary of the sealing joint neighborhood and the local darkening caused by the deposition black band. When extracting the sealing joint point set in the illumination-normalized image, the sealing joint is regarded as an annular or band-shaped boundary formed by the joint of the upper and lower covers. This boundary is usually represented in the image as a brightness abrupt change band along the direction of the joint.

[0023] It should be noted that the sedimentary black band is a specific appearance phenomenon used to characterize the state of the sealing joint area of ​​the self-rescue device. It refers to the dark sediment accumulation band formed in the neighborhood of the sealing joint between the upper and lower covers of the self-rescue device. Under conditions such as dampness and dust in the well, being worn by the user, and being subjected to bending and compression, the sedimentary black band exhibits a directional distribution characteristic relative to the direction of gravity downward. That is, the sedimentary black band in the lower edge area is wider and darker, while the sedimentary black band in the upper edge area is relatively narrower and lighter in color. The sedimentary black band is formed by dust and moisture adhering to the micro-slit neighborhood of the sealing joint and being compacted under the action of compression and gravity. Its morphological changes are used to characterize the changes in the degree of deposition and darkening in the neighborhood of the sealing joint. The width of the sedimentary black band is used to characterize the effective extension range of the dark band in the radial direction, and the blackness of the sedimentary black band is used to characterize the degree of darkening of the darkest position relative to the surrounding background. Quantitative analysis of the width and blackness can provide data support for the digital assessment of the safety status of the self-rescue device seal and subsequent diversion management.

[0024] Specifically, when extracting the set of sealing seam points from the normalized illumination image, any of the following alternative methods can be used: Preferably, the brightness difference of pixels in the horizontal and vertical directions is calculated on the illumination-normalized image and combined to form a gradient magnitude map. The gradient magnitude map is used to characterize the intensity of brightness change at each pixel. Then, pixels with larger gradient magnitudes are selected from the gradient magnitude map as candidate boundary points, and the candidate boundary points are connected into continuous boundary segments through connectivity constraints. The boundary segments with the longest length that are closed or nearly closed around the sealing joint area are further screened. The pixel coordinates in the boundary segment are used as the sealing joint boundary points and are collected to form the sealing joint point set. The sealing joint point set refers to a set of multiple two-dimensional coordinate points used to describe the spatial position and orientation of the sealing joint in the image.

[0025] Optionally, using the edge tracking method, first perform Canny edge detection on the illumination-normalized image to obtain an edge map containing all edges in the image. Based on the preset position range of the self-rescue device's sealing seam (matching the sealing area corresponding to the self-rescue device's shell structure), locate the initial edge point. Then, starting from this initial point, track and record the coordinates pixel by pixel along the continuous direction of the edge until a closed or continuous edge curve is formed. The set of pixel coordinates of this curve is used as the sealing seam point set.

[0026] Optionally, by using morphological thinning, the region where the sealing seam is located in the illumination-normalized image is first binarized (the binarization threshold is determined based on the brightness difference between this region and other regions of the shell) to obtain a binary region containing the edge of the sealing seam. Then, morphological erosion and thinning operations are performed on this binary region in sequence, and the set of pixel coordinates corresponding to the single-pixel width continuous curve obtained after thinning is used as the sealing seam point set.

[0027] The centroid of the sealing joint point set is obtained by averaging the coordinates of each point in the sealing joint point set. Specifically, after obtaining the set of sealing seam points, the arithmetic mean of the x-coordinate and y-coordinate of each boundary point in the set is calculated to obtain the centroid of the sealing seam point set. The centroid refers to the two-dimensional coordinates used to characterize the overall geometric center position of the point set, which serves as the benchmark for determining the subsequent outer sampling direction and dividing the upper and lower ring sets.

[0028] By leveraging the inherent brightness abrupt change characteristics at the sealing joint, a point set representing the spatial morphology of the sealing joint is stably extracted from the illumination-normalized image. Furthermore, the centroid of the point set is obtained, enabling subsequent sedimentation intensity sequence calculations to perform consistent radial profile sampling in its outer neighborhood with the sealing joint as a reference. Simultaneously, the wedge-shaped sedimentary black band index can achieve directional comparison of the upper and lower halves of the ring based on the centroid.

[0029] In an embodiment of the present invention, the outer sampling direction is determined by the centroid and the points of the sealing seam set, and the illumination normalized image is profiled along the outer sampling direction to generate a deposition intensity sequence, including: For each sealing point in the set of sealing joint points, a radial grayscale profile is obtained along the outer sampling direction; Specifically, the sealing joint point set is used to characterize the spatial position of the sealing joint in the illumination-normalized image. Each sealing point is a pixel coordinate point in the sealing joint point set. To evaluate the local deposition state of the wedge-shaped depositional black band at the lower edge of the sealing joint surface at each sealing point, the outer sampling direction is determined by the centroid of the sealing joint point set and the current sealing point. The outer sampling direction refers to the radial direction from the centroid to the current sealing point and continuing to the outer neighborhood of the sealing. This direction is normalized to obtain a unit vector of the outer sampling direction, which is used for equidistant sampling on the illumination-normalized image. When obtaining the radial grayscale profile along the outer sampling direction, sampling is performed in the outer neighborhood at pixel or sub-pixel steps, starting from the current sealing point, to form a one-dimensional sequence with radial distance as the independent variable and illumination-normalized grayscale value as the dependent variable. The radial grayscale profile is used to characterize the brightness change formed by the sediment, gap shadow and shell background during the transition from the sealing joint to the outside.

[0030] In the radial grayscale profile, the darkest point grayscale value and the background grayscale value are determined, and the average value of the darkest point grayscale value and the background grayscale value is used as the half-depth level. Specifically, when determining the darkest point gray value in the radial gray profile, the sampling point with the smallest gray value is searched within the outer sampling length range and its gray value is recorded as the darkest point gray value. The darkest point corresponds to the strongest darkening position of the depositional black band or the darkening zone of the seam. To avoid interference from distant textures or local noise on the background estimation, a radial interval in the radial gray profile that is far from the sealing seam and far from the neighborhood of the darkest point is taken as the background estimation interval. The background estimation interval is a region in the neighborhood of the sealing surface where the brightness change is relatively gentle. The median of the gray values ​​in this background estimation interval is used to obtain the background gray value. The background gray value is used to characterize the reference brightness of the neighborhood of the sealing point without the dominant influence of the depositional black band. When determining the half-depth level based on the darkest point gray value and the background gray value, the arithmetic mean of the two is taken to obtain the half-depth level. The half-depth level is used to define the boundary of the depositional zone in an adaptive manner to avoid boundary drift under different reflection or different lighting conditions caused by using a fixed threshold.

[0031] The width of the sedimentary zone is obtained by determining the distance between the left and right intersection points of the sedimentary zone based on the half-depth level. Specifically, when determining the distance between the left and right intersection points of the sedimentation zone based on the half-depth level, the radial grayscale profile is searched in the direction closer to the sealing joint and in the direction farther away from the sealing joint, with the radial position of the darkest point as the center. The first intersection point where the radial grayscale changes from below the half-depth level to above the half-depth level is taken as the left intersection point and the right intersection point, respectively. The difference between the radial distance of the right intersection point and the radial distance of the left intersection point is calculated to obtain the width of the sedimentation zone. The width of the sedimentation zone is used to characterize the radial expansion range of the dark zone at the sealing point.

[0032] The blackness of the deposition zone is calculated based on the grayscale value of the darkest point and the grayscale value of the background. Specifically, when calculating the blackness of the sedimentary zone based on the darkest point gray value and the background gray value, the degree of darkening of the darkest point gray value relative to the background gray value is normalized so that the blackness of the sedimentary zone reflects the relative darkening intensity caused by sediment absorption or shading.

[0033] The deposition intensity at the corresponding sealing point is obtained by multiplying the width of the deposition zone by the blackness of the deposition zone, thus forming a deposition intensity sequence; Specifically, after obtaining the width and blackness of the sedimentary zone, the two are multiplied to obtain the sedimentation intensity of the corresponding sealing point. The sedimentation intensity is used to characterize the combined effect of a wider and darker sedimentary zone. The sedimentation intensity of each sealing point is arranged in the circumferential order of the sealing joint point set to form a sedimentation intensity sequence, which is used for subsequent statistical analysis and calculation of the wedge-shaped sedimentary black band index according to the upper and lower half ring sets.

[0034] By achieving adaptive width measurement of the boundary of the sedimentary black band at half depth level and characterizing the blackness with the degree of darkening relative to the background, the directional appearance phenomenon of the wedge-shaped sedimentary black band at the lower edge of the sealing interface is transformed into a calculable sedimentary intensity sequence, providing reliable and stable data for subsequent quantification of differences between the upper and lower half rings and intelligent cabinet diversion management.

[0035] In an embodiment of the present invention, the set of sealing joint points is divided into an upper half-ring set and a lower half-ring set according to the centroid, and the wedge-shaped depositional black band index is calculated based on the depositional intensity sequence, including: Calculate the mean sedimentation intensity sequence corresponding to the lower half-ring set and the mean sedimentation intensity sequence corresponding to the upper half-ring set, respectively; Specifically, to partition and statistically analyze the sealing seams according to their directionality, the ordinate of the centroid is used as the dividing criterion. Points with ordinates less than the centroid's ordinate form an upper ring set, while points with ordinates greater than the centroid's ordinate form a lower ring set. The upper and lower ring sets correspond to the circumferential regions of the sealing seams located above and below the centroid in the image coordinate system, respectively. This partitioning allows subsequent statistical analysis to directly characterize the distribution differences of the deposited black bands in the vertical direction. After obtaining the upper and lower ring sets... After the ring set, the deposition intensity sequence is a sequence formed by arranging the deposition intensities obtained by sampling radial grayscale profiles along the outer sampling direction of each sealing point in the sealing joint point set in a ring direction. Each sealing point corresponds to a deposition intensity, which is used to simultaneously characterize the effective width and darkening degree of the depositional zone at that location. When calculating the wedge-shaped depositional black band index based on the deposition intensity sequence, the average deposition intensity corresponding to each sealing point in the lower half ring set is used to obtain the average deposition intensity of the lower half ring, and the average deposition intensity corresponding to each sealing point in the upper half ring set is used to obtain the average deposition intensity of the upper half ring.

[0036] The wedge-shaped sedimentary black band index is calculated based on the mean of the sedimentary intensity sequences corresponding to the lower and upper half-ring sets. The formula for calculating the wedge-shaped sedimentary black band index is as follows: In the formula, This represents the mean of the sedimentation intensity sequence corresponding to the lower half of the ring set. This represents the mean of the sedimentation intensity sequence corresponding to the upper half of the ring set; Specifically, the molecules adopt It can directly characterize the direction and magnitude of the difference between the upper and lower halves. When the lower half-ring is heavier, the numerator is positive and increases with the increase of the difference. At the same time, the absolute value of the deposition intensity is affected by factors such as imaging brightness, material reflectivity, and overall contamination level. If only the difference is used, it will lead to incomparability under different equipment or lighting conditions. Therefore, the denominator adopts... The differences are normalized so that the indicators reflect the proportion of the differences to the total depositional level, thus maintaining comparability even when the overall deposition is relatively dirty or relatively clean; furthermore, because and All are non-negative mean depositional intensity values. For positive, after normalization It has a stable range of values, and when the upper and lower values ​​are close... make This indicates no obvious wedge shape, when the lower half of the ring is significantly larger. A value close to 1 indicates a significant lower-edge wedge-shaped sedimentary black band, while the upper half of the band is actually larger. A negative value can indicate that the deposition distribution direction is opposite to the expected direction. This calculation method maintains a direct expression of directional wedge differences while reducing interference from changes in absolute brightness and overall stain levels through summation normalization. The wedge-shaped depositional black band index is used to characterize the directional significance of the wedge-shaped depositional black band at the lower edge of the sealing interface. When the average depositional intensity of the lower half-ring is greater than that of the upper half-ring, the wedge-shaped depositional black band index increases, indicating that the depositional black band in the lower edge region is wider and darker and the wedge-shaped distribution is more significant.

[0037] In an embodiment of the present invention, the available capacity of the isolation chamber is read, the self-rescue devices to be issued are sorted according to the wedge-shaped deposition black band index, and the number of self-rescue devices corresponding to the available capacity of the isolation chamber are selected to enter the isolation queue. The remaining self-rescue devices are entered into the issuance queue, and the ledger status is updated, including: The self-rescue devices to be issued are sorted according to the wedge-shaped deposition black band index, and the number of self-rescue devices corresponding to the available capacity of the isolation chamber are selected to enter the isolation queue. The self-rescue devices that enter the isolation queue are written into the isolation status log record, and the image of the sealing joint surface area and the wedge-shaped deposition black band index corresponding to the self-rescue device are stored. Write the self-rescue devices that have entered the dispensing queue into the dispensing status log record; Specifically, the isolation compartment is a set of compartments in the self-rescue device intelligent device management system used to temporarily store self-rescue devices that require further processing. The available capacity of the isolation compartment refers to the number of compartments in the isolation compartment that are in an idle and placeable state, which is obtained in real time by the controller based on the occupancy sensor signal of the isolation compartment or the storage information of the compartment status. The self-rescue devices to be distributed refer to the set of self-rescue devices that are currently in the intelligent cabinet or waiting to enter the intelligent cabinet distribution link and have completed the wedge deposition black band index calculation. Each self-rescue device to be distributed has a unique identifier, which is used to establish a correspondence between the self-rescue device and the ledger. The system records the corresponding image data, wedge-shaped depositional black band index, and diversion results. After reading the available capacity of the isolation chamber, the controller reads the wedge-shaped depositional black band index corresponding to the set of self-rescue devices to be issued from the memory, and sorts the self-rescue devices in descending order using the wedge-shaped depositional black band index as the sorting key, placing self-rescue devices with larger wedge-shaped depositional black band indices at the front. This sorting means that self-rescue devices with more significant depositional intensity in the lower half-ring compared to the upper half-ring and more prominent wedge-shaped depositional black bands at the lower edge are given priority for isolation. When determining the number of self-rescue devices to enter the isolation queue based on the available capacity of the isolation chamber... The system compares the available capacity of the isolation chamber with the number of self-rescue devices to be distributed, taking the smaller of the two as the target number for the isolation queue. This ensures that the diversion action can still be performed even when isolation chamber resources are limited, without causing conflicts due to exceeding the chamber's capacity. After obtaining the target number for the isolation queue, the controller sequentially selects the corresponding number of self-rescue devices from the front of the sorting results to form the isolation queue, and the remaining self-rescue devices form the distribution queue. The controller then outputs diversion control commands for both the isolation queue and the distribution queue, driving the diversion mechanism to transfer the self-rescue devices from the isolation queue to the isolation chamber and transfer the self-rescue devices from the distribution queue to the distribution queue. The device is transported to the dispensing warehouse or dispensing channel. Simultaneously, the controller performs a synchronous update on the ledger status. The ledger is an electronic record set used to record the status of the self-rescue device throughout its entire life cycle. The ledger status includes at least the isolation status and the dispensing status. During the update, the unique identifier of each self-rescue device in the isolation queue is written into the isolation status record and associated with its wedge-shaped deposition black band index and the corresponding sealing interface area image. The unique identifier of each self-rescue device in the dispensing queue is written into the dispensing status record and associated with its wedge-shaped deposition black band index and the corresponding sealing interface area image, so that the diversion result of any self-rescue device can be traced.

[0038] By establishing a direct link between the quantitative result of the wedge-shaped sedimentary black band index and the actual resource constraint of the available capacity of the isolation chamber, priority interception of more significant self-rescue devices for the wedge-shaped sedimentary black band can be achieved through sorting and capacity matching. Isolation and disbursable diversion are performed through the diversion mechanism, and the diversion results and basis data are updated and solidified in the ledger status, thereby forming an automated closed-loop management of the risk of the wedge-shaped sedimentary black band at the lower edge of the sealing interface.

[0039] To facilitate understanding of the above embodiments, a specific application scenario of the above embodiments will be used as an example for illustration below: The application scenario is the intelligent device management system for self-rescue devices in coal mine surface lighting rooms. Before entering the mine, the self-rescue devices need to undergo automated detection and diversion management by the distribution cabinet. The on-site environment is a typical high-humidity and dusty working condition. The relative humidity in the lamp room is consistently between 85% and 95% RH (90% RH in the example). Coal dust and water vapor easily adhere to the area adjacent to the outer edge of the self-rescue device's sealing joint and, after being worn and compressed, form a deposited black band. The equipment is deployed at the imaging station inside the dispensing cabinet. The imaging station includes a fixing fixture, a ring light source, and a camera. The fixing fixture faces the self-rescue device's sealing joint surface toward the camera and constrains its posture. The ring light source provides stable illumination, and the camera captures images of the sealing joint surface area and uses them as input data. The data acquisition method involves acquiring one frame of the sealing joint surface area image for each self-rescue device to be dispensed (resolution example 1920×1080). The data is then processed within the cabinet controller, including grayscale conversion, illumination normalization, sealing joint point set extraction, deposition intensity sequence calculation, wedge-shaped deposition black band index calculation, and sorting and distribution based on the available capacity of the isolation chamber, as detailed below: After the imaging station acquires a color image of the sealing surface of a self-rescue device, the controller uses a brightness-weighted algorithm to generate a grayscale image. Taking a specific pixel in this image as an example, its red, green, and blue channel values ​​are respectively: The corresponding grayscale value is calculated using the formula: Calculated Only pixel brightness information is retained to stably characterize the darkening differences of the deposited black bands, providing a unified single-channel data foundation for subsequent image processing; an illumination-normalized image is generated based on this grayscale image. At that time, the controller uses the average value of the local window of the pixel's neighborhood as the background brightness and performs a pixel-by-pixel normalization operation. Taking the above pixel as an example, its local window average value is 140 (obtained by averaging the brightness of neighboring pixels), then the normalized brightness is... ( (Using a very small positive number to avoid a denominator of zero), this operation can suppress the overall brightness fluctuations caused by uneven supplementary lighting and the curvature of the self-rescue device shell, making the local darkening features of the deposited black band more prominent; in the illumination-normalized image, the controller extracts boundary points based on the brightness abrupt change at the sealing joint, obtaining the sealing joint point set. The centroid is obtained by taking the arithmetic mean of the coordinates of the point set. In this embodiment, the coordinates of the eight sealing joint boundary points extracted are: (100,200), (140,190), (180,200), (200,240), (180,280), (140,290), (100,280), (80,240). Therefore, the centroid coordinates are calculated as follows: , The centroid serves as a unified geometric reference, used to determine the subsequent outer sampling direction and to divide the upper and lower half-ring sets. During the deposition intensity sequence calculation phase, the controller targets each sealing point. The outer sampling direction is determined by the line connecting the centroid and the sealing point and normalized to a unit vector. After obtaining the radial grayscale profile along this direction, the deposition zone width and blackness are adaptively obtained using the half-depth method. Taking the sealing point (180, 280) as an example, the unit vector of its outer sampling direction is: The controller travels radially along this direction. Radial grayscale profile obtained by sampling within the range And determine the gray value of the darkest point in this profile. Background grayscale value (The background grayscale value is obtained by taking the median of the neighborhood intervals that are far from the seam and have a flat brightness), and then the half-depth level is calculated as follows. When the cross-section intersects the horizontal half-depth at the left and right points respectively, , At that time, the width of the sedimentary zone was The blackness of the sedimentary zone is The deposition intensity at the corresponding sealing point is By using a composite quantification method of width × blackness, the significance of the sedimentary black band at that point is characterized, and a sedimentary intensity sequence is formed. ; For the same self-rescue device, in the sedimentation intensity sequence corresponding to the 8 sealing points, the upper ring set (satisfying) The deposition intensity corresponding to the point is The lower half ring set (satisfying) The deposition intensity corresponding to the point is The average deposition intensity of the upper half ring Mean deposition intensity in the lower half ring The wedge-shaped sedimentary black band index is calculated using the formula... Calculated The function of this index is to quantify the directional wedge distribution with a wider and darker lower edge and a narrower and lighter upper edge into a single comparable indicator. After calculating the value of a single self-rescue device, the batch distribution process begins. This batch contains 4 self-rescue devices to be distributed, with corresponding wedge-shaped sedimentary black band indices as follows: Unit A B station C-stage D platform The controller reads the available capacity of the isolation chamber. ,according to Sort by size from largest to smallest and select the top few. The device enters the isolation queue (in this embodiment, the isolation queue is {C,A}), and the remaining devices ({B,D}) enter the dispensing queue. Subsequently, the controller outputs a diversion control command to drive the diversion mechanism to transfer the self-rescue devices in the isolation queue to the isolation chamber, and to transfer the self-rescue devices in the dispensing queue to the dispensing chamber or dispensing channel, and simultaneously updates the ledger status: the unique identifier of the self-rescue device in the isolation queue is associated with its sealing joint surface area image and wedge-shaped deposition black band index and written into the isolation status record, and the unique identifier of the self-rescue device in the dispensing queue is associated with the corresponding data and written into the dispensing status record, thereby realizing the automated interception of the risk of self-rescue device sealing deterioration under high humidity and dust conditions.

[0040] like Figure 2 As shown, the self-rescue device intelligent management cabinet in this embodiment includes an intelligent management cabinet body 1. An imaging station 2 is set inside the cabinet body 1. An industrial camera 3 is installed in the imaging station 2, and a ring light 4 is set below it to provide stable illumination and image acquisition for the self-rescue device body 6 located in the imaging station 2. A self-rescue device fixing clamp 5 is set below the imaging station 2 to limit and fix the self-rescue device body 6 and ensure that its posture is consistent with the imaging position. A mechanical diversion gate 7 and a sorting slide 8 are set at the lower part of the cabinet body 1. The mechanical diversion gate 7 switches the flow direction of the self-rescue device body 6 under the action of control commands, so that the self-rescue device body 6 enters the corresponding storage or processing channel along the sorting slide 8, thereby realizing the integrated arrangement of imaging detection and diversion management of the self-rescue device in the cabinet.

[0041] like Figure 3 As shown, the upper and lower covers of the self-rescue device form a circumferential sealing seam. A band-shaped deposition area appears in the vicinity of the outer side of the sealing seam. The deposition area extends along the sealing seam and shows a clear directional distribution. Under the influence of gravity, the deposition is more significant on the side closer to the lower cover, which is characterized by a larger coverage area and a higher degree of darkening of the deposition band in the lower edge area, while it is relatively narrower and has a lower degree of darkening in the upper edge area. The direction of deposition aggravation is indicated by the gravity direction arrow in the figure, which illustrates that this band-shaped deposition phenomenon is related to the adhesion and compaction process of the self-rescue device in a humid and dusty environment, and can serve as a visual representation of the appearance of the vicinity of the sealing seam.

[0042] like Figure 4As shown, the sealing joint is extracted from the illumination-normalized image and a sealing joint point set is formed. The centroid of the sealing joint point set is used as the geometric reference point. For each sealing point on the sealing joint, the outer sampling direction is determined by pointing from the centroid to the sealing point and then to the outer neighborhood of the sealing. Radial grayscale profiles are obtained in the outer neighborhood of the sealing joint along the outer sampling direction. The dashed box indicates the local sampling window and its radial sampling range at the corresponding sealing point. In the enlarged schematic diagram on the right, the radial grayscale profile is plotted with distance as the horizontal axis and grayscale value as the vertical axis. The profile curve forms grayscale valleys at the sedimentation zone. The half-depth level is obtained by determining the background grayscale value and the darkest point grayscale value and averaging them. The distance between the left and right intersections of the half-depth level and the profile curve is defined as the half-depth width, thereby realizing adaptive width measurement of the sedimentation zone in the radial direction.

[0043] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A self-rescue device intelligent management system, characterized in that, include: The image acquisition module is used to acquire images of the sealing and bonding surface area of ​​the self-rescue device at the imaging station and generate grayscale images; An illumination normalization module is used to generate an illumination normalized image based on the grayscale image; The suture and centroid module is used to extract the set of sealing seam points in the illumination normalized image and calculate the centroid of the set of sealing seam points. The profile sampling and intensity module is used to determine the outer sampling direction based on the centroid and the points of the sealing joint, and to perform profile sampling on the illumination normalized image along the outer sampling direction to generate a deposition intensity sequence. The wedge index module is used to divide the set of sealing joint points into an upper half-ring set and a lower half-ring set according to the centroid, and to calculate the wedge-shaped depositional black band index based on the depositional intensity sequence. The capacity distribution module is used to read the available capacity of the isolation chamber, sort the self-rescue devices to be issued according to the wedge-shaped deposition black band index, select the number of self-rescue devices corresponding to the available capacity of the isolation chamber to enter the isolation queue, and put the remaining self-rescue devices into the issuance queue, and update the ledger status.

2. The intelligent self-rescue device management system according to claim 1, characterized in that, Generating an illumination-normalized image based on the grayscale image includes: Calculate the mean image of the local window of the grayscale image; The grayscale image and the local window mean image are normalized to obtain an illumination-normalized image.

3. The intelligent self-rescue device management system according to claim 1, characterized in that, Extracting the set of sealing seam points from the normalized illumination image and calculating the centroid of the set of sealing seam points includes: In the illumination-normalized image, boundary points are extracted based on the brightness abrupt change at the sealing joint to obtain the sealing joint point set; The centroid of the sealing joint point set is obtained by averaging the coordinates of each point in the set.

4. The intelligent self-rescue device management system according to claim 1, characterized in that, The outer sampling direction is obtained by normalizing the reverse vector from the current point in the set of sealing seam points to the centroid.

5. The intelligent self-rescue device management system according to claim 4, characterized in that, Profiling the illumination-normalized image along the outer sampling direction to generate a deposition intensity sequence includes: For each sealing point in the set of sealing joint points, a radial grayscale profile is obtained along the outer sampling direction; In the radial grayscale profile, the darkest point grayscale value and the background grayscale value are determined, and the average value of the darkest point grayscale value and the background grayscale value is used as the half-depth level. The width of the sedimentary zone is obtained by determining the distance between the left and right intersection points of the sedimentary zone based on the half-depth level. The blackness of the deposition zone is calculated based on the grayscale value of the darkest point and the grayscale value of the background. The deposition intensity at the corresponding sealing point is obtained by multiplying the width of the deposition zone by the blackness of the deposition zone, thus forming a deposition intensity sequence.

6. The intelligent self-rescue device management system according to claim 1, characterized in that, The upper half-ring set is the set of points whose ordinates at the sealing joints are less than the ordinates at the centroid, and the lower half-ring set is the set of points whose ordinates at the sealing joints are greater than the ordinates at the centroid.

7. The intelligent self-rescue device management system according to claim 1, characterized in that, The wedge-shaped sedimentary black band index is calculated based on the aforementioned sedimentary intensity sequence, including: Calculate the mean sedimentation intensity sequence corresponding to the lower half-ring set and the mean sedimentation intensity sequence corresponding to the upper half-ring set, respectively; The sedimentation intensity difference is obtained by subtracting the mean of the sedimentation intensity sequence corresponding to the upper half ring set from the mean of the sedimentation intensity sequence corresponding to the lower half ring set. The average sedimentation intensity sequence of the lower half-ring set is added to the average sedimentation intensity sequence of the upper half-ring set to obtain the sedimentation intensity sum value; The wedge-shaped sedimentary black band index is obtained by comparing the difference in sedimentary intensity with the sum of sedimentary intensity.

8. The intelligent self-rescue device management system according to claim 1, characterized in that, The number of self-rescue devices entering the isolation queue is the smaller of the available capacity of the isolation chamber and the number of self-rescue devices currently to be issued.

9. The intelligent self-rescue device management system according to claim 1, characterized in that, Update the ledger status, including: The self-rescue devices that enter the isolation queue are written into the isolation status log record, and the image of the sealing joint surface area and the wedge-shaped deposition black band index corresponding to the self-rescue device are stored. Write the self-rescue device that enters the dispensing queue into the dispensing status log record.

Images