A cigarette production site collaborative inspection method, device, equipment and medium

Abstract

The application discloses a cigarette production site cooperative inspection method, device, equipment and medium, establishes an inspection video cooperative mapping table, acquires an inspection task and determines a target inspection point, calls a fixed video acquisition device to acquire a fixed visual angle video picture, identifies the fixed visual angle video picture to obtain a fixed visual angle judgment result, performs coverage credibility evaluation on a target point area, generates a fixed visual angle inspection record when the credibility is available, determines a target intelligent safety helmet terminal from a plurality of intelligent safety helmet terminals when the credibility is unavailable, sends an inspection cooperative instruction, receives a mobile visual angle video or picture, identifies a mobile visual angle judgment result, performs consistency determination on the double visual angle judgment results, generates a confirmation type, a review type or a supplementary blind type inspection result, generates a unique inspection event identifier and stores multi-source data in association. The application can improve the coverage credibility, remote cooperative efficiency and process traceability reliability of the cigarette production site inspection.

Description

Technical Field

[0001] This invention relates to on-site inspection technology for cigarette production, and more particularly to a collaborative inspection method, device, equipment, and medium for on-site cigarette production. Background Technology

[0002] In cigarette production sites, fixed video surveillance systems are widely used for inspections of key areas and critical locations due to their advantages such as continuous data acquisition, stable coverage, and ease of centralized management. However, since the installation position and shooting angle of fixed cameras cannot be dynamically adjusted once determined, in actual working environments, their images are easily affected by factors such as equipment obstruction, stacked cigarette packs, personnel movement, and changes in lighting. This can lead to problems such as the target inspection point being invisible, severely obstructed, blurry, having abnormal lighting, incomplete targets, or excessively tilted shooting angles in the actual video footage, even if the target inspection point is within the theoretical coverage area of ​​the camera. In such cases, if inspection personnel or automated systems rely solely on the fixed-view video footage for judgment, misjudgments or omissions are highly likely. This is a reliability defect caused by the inherent limitations of the viewing angle in fixed monitoring solutions.

[0003] To overcome the aforementioned shortcomings, existing technologies have introduced solutions that link fixed monitoring with mobile inspection equipment. When the mobile inspection equipment detects environmental parameters exceeding normal ranges, the system retrieves fixed monitoring footage from the corresponding sub-area for verification. While this approach introduces a mobile perspective as a supplement, its linkage triggering mechanism relies on passive detection of abnormal events. It cannot proactively identify non-abnormal but judgment-affecting image quality issues in regular inspection tasks, such as occlusion, blurriness, or insufficient angle in the fixed-view footage. In other words, even if the fixed-view footage is completely unusable for judgment due to occlusion or blurriness, the system will not invoke the mobile perspective to fill the gap unless an environmental anomaly is detected, leading to a risk of missing judgment criteria at that location during inspection tasks.

[0004] Furthermore, existing collaborative solutions, after acquiring both fixed and mobile viewpoints, simply display or switch between them separately, lacking a mechanism for consistent comparison of the judgment results from each viewpoint. When the judgments from the two viewpoints conflict, the system cannot automatically identify the conflict, nor can it generate inspection conclusions with different confidence levels. Instead, it relies on manual secondary judgment, reducing the decision-making efficiency and reliability of the automated inspection system. Simultaneously, due to the lack of a unified event identification and association mechanism, multi-source data such as fixed video recordings, mobile video recordings, photographed images, broadcast commands, and on-site confirmation information are stored in a scattered manner, making it difficult to form a complete and traceable chain of inspection evidence. Summary of the Invention

[0005] This invention provides a method, apparatus, equipment, and medium for collaborative inspection of cigarette production sites, in order to improve the coverage reliability, remote collaboration efficiency, and process traceability reliability of on-site inspections of cigarette production.

[0006] In a first aspect, the present invention provides a collaborative inspection method for cigarette production sites, comprising: Establish and store a collaborative mapping table for inspection videos, which associates inspection points, coverage areas of fixed video acquisition devices, smart safety helmet terminal positioning information, and wearer access information. Obtain inspection tasks and determine target inspection points; According to the inspection task, a fixed video acquisition device covering or associated with the target inspection point is invoked to obtain a fixed-view video image. Image recognition is performed on the fixed-viewpoint video footage to obtain the fixed-viewpoint determination result; A point-level coverage confidence assessment is performed on the image region corresponding to the target inspection point in the fixed-view video frame to obtain the fixed-view coverage confidence. When the credibility of the fixed viewpoint coverage does not meet the preset credibility condition, the target smart helmet terminal is determined from multiple smart helmet terminals. Send an inspection collaboration instruction to the target smart safety helmet terminal. The inspection collaboration instruction includes one or more of the following: start a video stream to upload a moving perspective video image, perform remote photography, prompt the wearer to move to a designated inspection point, or prompt the wearer to adjust the shooting angle. Receive moving-view video footage or remotely captured images from the target smart safety helmet terminal; Image recognition is performed on the received moving-view video footage or remotely captured images to obtain the moving-view determination result; The consistency between the fixed-view judgment result and the moving-view judgment result is determined, and a confirmation inspection result, a verification inspection result, or a blind spot inspection result is generated based on the determination result. A unique inspection event identifier is generated for the inspection task, and the multi-source data generated during the inspection process is associated and stored based on the unique inspection event identifier to form a traceability link for the inspection process.

[0007] Optionally, a point-level coverage confidence assessment is performed on the image region corresponding to the target inspection point in the fixed-view video frame to obtain the fixed-view coverage confidence, including: Calculate the visibility score V, occlusion ratio O, sharpness score S, illumination condition score L, target integrity score T, and shooting angle adequacy score A; Through formula Calculate the coverage confidence C at a fixed viewpoint, where to The weights are 1, and the sum of the weights is 1. Wherein, the visibility score V is the ratio of the intersection area of ​​the target rectangular region projected onto the image plane and the image boundary to the area of ​​the target rectangular region; The occlusion ratio O is the ratio of the number of pixels within the target rectangular area that are covered by non-target objects to the total number of pixels in that area; The sharpness score S is obtained based on Laplace variance normalization: ,in The variance of the target region after Laplace filtering. This is the pre-calibrated optimal sharpness reference value; The illumination status score L is calculated based on the mean μ and standard deviation σ of the grayscale image of the target area: ,in , And L=0 when μ<40 or μ>220; The target integrity score T is the ratio of the number of actual visible key points of the target inspection object in the image to the number of key points in the standard template; The sufficiency score A for the shooting angle is calculated as a piecewise function based on the angle θ between the camera's optical axis and the normal vector of the target inspection point: when θ ≤ θ good When A=1, when θ ≥ θ bad When θ = 0, A = 0. good < θ < θ bad hour , where θ good =30°, θ bad =75°.

[0008] Optionally, determining the target smart helmet terminal from multiple smart helmet terminals includes: Acquire location information, online status, battery status, wearer permissions, distance information, communication quality, and current task status of multiple smart safety helmet terminals; Calculate the collaborative inspection score for each smart safety helmet terminal. The collaborative inspection score is obtained by weighted summation of distance score, online status score, battery level score, access score, communication quality score, and idle status score. The smart helmet terminal with the highest collaborative inspection score that is not lower than the preset minimum threshold is selected as the target smart helmet terminal.

[0009] Optionally, the quantification method for each sub-score in the collaborative inspection scoring is as follows: Distance rating: Where d is the spatial distance between the smart safety helmet terminal positioning and the target inspection point. The preset maximum effective distance; Online status score: 1 when the terminal is online and the heartbeat is normal, and 0 when offline or the heartbeat timeout occurs; Battery rating: ; Access score: 1 is given when the wearer has access or operation permission to the target inspection point; otherwise, 0 is given and the terminal is removed from the candidate set. Communication quality rating: based on RSSI normalization. ; Idle state score: 1 when the terminal is not currently performing other inspection tasks and has no unprocessed SOS; 0.5 when it is performing other tasks but is allowed to preempt; and 0 when it is performing a high-priority task or handling an SOS.

[0010] Optionally, a consistency determination is made between the fixed-viewpoint judgment result and the moving-viewpoint judgment result, and a confirmation-type inspection result, a verification-type inspection result, or a blind-spot-filling inspection result is generated based on the determination result, including: When the fixed-view judgment result is the same as the moving-view judgment result, a confirmatory inspection result is generated. When the fixed-view judgment result differs from the moving-view judgment result, a verification inspection result is generated, and the fixed-view video clip, the moving-view video clip, the remotely captured image, the fixed-view coverage credibility, the moving-view recognition result, and the manual verification entry are simultaneously pushed to the central control terminal. When the confidence level of the fixed viewpoint coverage is lower than the preset confidence threshold, resulting in the absence of fixed viewpoint judgment results, the judgment results of the moving viewpoint are used to generate blind spot inspection results.

[0011] Optional collaborative inspection methods at the cigarette production site also include: When an SOS alarm message is received from the smart safety helmet terminal, the alarm personnel, alarm time, and alarm location are determined based on the alarm message. Based on the alarm location, query the inspection video collaborative mapping table to determine the fixed video acquisition device covering the alarm location or adjacent to the alarm location, and simultaneously take over the moving view video screen of the smart safety helmet terminal. Simultaneously push fixed-view video footage and moving-view video footage to the central control unit; An SOS event identifier is generated for the SOS alarm, and the SOS alarm information, alarm personnel information, alarm location, fixed-view video recording, moving-view video recording, remote broadcast recording, on-site handling recording, and alarm closure recording are associated and stored based on the SOS event identifier.

[0012] Optional collaborative inspection methods at the cigarette production site also include: When the fixed video acquisition device corresponding to the inspection task is unavailable, the video stream is interrupted, the image quality is lower than the preset quality threshold, or the fixed viewpoint coverage reliability is lower than the preset reliability threshold, the video stream is automatically switched to the target smart safety helmet terminal. When the target smart helmet terminal is unavailable, offline, has a communication quality lower than a preset communication threshold, or has a battery level lower than a preset battery level, it automatically switches to a backup smart helmet terminal or a backup fixed video acquisition device. The backup smart helmet terminal is determined based on the ranking results of the collaborative inspection scoring.

[0013] Secondly, the present invention also provides a collaborative inspection device for cigarette production sites, comprising: The mapping relationship management module establishes and stores a collaborative mapping table for inspection videos. The mapping table associates inspection points, coverage areas of fixed video acquisition devices, smart safety helmet terminal positioning information, and wearer permission information. The inspection task acquisition module is used to acquire inspection tasks and determine target inspection points. The fixed video recall module is used to recall fixed video acquisition devices that cover or are associated with the target inspection point according to the inspection task, and to acquire fixed-view video images. A fixed-viewpoint recognition module is used to perform image recognition on the fixed-viewpoint video frame to obtain a fixed-viewpoint judgment result; The coverage credibility assessment module is used to perform point-level coverage credibility assessment on the image area corresponding to the target inspection point in the fixed-view video frame to obtain the fixed-view coverage credibility. The smart helmet scheduling module is used to determine the target smart helmet terminal from multiple smart helmet terminals when the credibility of the fixed view coverage does not meet the preset credibility conditions. The collaborative instruction sending module is used to send inspection collaborative instructions to the target smart safety helmet terminal. The inspection collaborative instructions include one or more of the following: starting a video stream to upload moving perspective video images, performing remote photography, prompting the wearer to move to a designated inspection point, or prompting the wearer to adjust the shooting angle. The mobile video receiving module is used to receive mobile-view video footage or remotely captured images transmitted back by the target smart safety helmet terminal. The motion perspective recognition module is used to perform image recognition on received motion perspective video footage or remotely captured images to obtain motion perspective judgment results. The dual-view judgment module is used to determine the consistency between the fixed-view judgment result and the moving-view judgment result, and generate a confirmation inspection result, a verification inspection result, or a blind spot inspection result based on the judgment result. The record and traceability module is used to generate a unique inspection event identifier for the inspection task, and to associate and store multi-source data generated during the inspection process based on the unique inspection event identifier to form an inspection process traceability link.

[0014] Thirdly, the present invention also provides an electronic device, comprising: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the collaborative inspection method for cigarette production sites provided in the first aspect of the present invention.

[0015] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the collaborative inspection method for cigarette production sites as provided in the first aspect of the invention.

[0016] This invention assesses the reliability of coverage at the point-to-point level in fixed-view video footage, distinguishing between "theoretical coverage" and "reliable coverage," thus avoiding misjudgments or omissions due to image quality issues. When a fixed viewpoint is unreliable, a target smart helmet terminal is identified from multiple smart helmet terminals, and a collaborative command is initiated to automatically schedule smart helmets to fill in the gaps. By verifying the consistency between the fixed-view and moving-view assessment results, different types of inspection results are generated. By generating a unique inspection event identifier, multi-source data from the inspection process is associated and stored, supporting post-inspection auditing, accountability tracking, and problem analysis.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart of a collaborative inspection method for cigarette production sites provided by the present invention; Figure 2 This invention provides a structural schematic diagram of a collaborative inspection device for cigarette production sites. Figure 3 This is a schematic diagram of the structure of an electronic device provided by the present invention.

[0020] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] Figure 1 This is a flowchart of a collaborative inspection method for cigarette production sites provided by the present invention. This embodiment is applicable to collaborative inspections using fixed video capture devices and smart safety helmets, and is applied to a collaborative inspection management platform. This method can be executed by the collaborative inspection device for cigarette production sites provided by the present invention. This device can be implemented in software and / or hardware, and is typically configured in electronic devices, such as... Figure 1 As shown, the collaborative inspection method for cigarette production sites includes the following steps: S101. Establish and store a collaborative mapping table for inspection videos. The mapping table associates inspection points, coverage areas of fixed video acquisition devices, smart safety helmet terminal positioning information, and wearer permission information.

[0024] In this embodiment of the invention, the inspection collaboration management platform acquires a spatial map of the cigarette production plant or workshop area, a register of fixed video acquisition devices, a register of smart safety helmet terminals, a register of wearers, and a register of inspection points. The platform then divides the spatial map into multiple inspection areas or spatial grids. The platform determines the installation location, shooting direction, field of view, coverage area, video stream address, and availability status of each fixed video acquisition device on the spatial map. Simultaneously, it determines the terminal number, communication address, online status, location information, wearer information, wearer permissions, and terminal capability information for each smart safety helmet terminal. Thus, an inspection video collaboration mapping table is established and stored, which associates inspection points, the coverage area of ​​fixed video acquisition devices, the location information of smart safety helmet terminals, and the wearer permission information.

[0025] The inspection video collaborative mapping table can adopt the field structure shown in the table below: S102. Obtain inspection tasks and determine target inspection points.

[0026] Inspection tasks can be generated by the inspection planning system, manual configuration, central control terminal, or triggered by abnormal events. An inspection task must include at least the inspection area, inspection points, inspection objects, inspection time, inspection task type, and inspection task risk level. Inspection task types can include equipment status inspection, cigarette production workshop inspection, safety boundary inspection, anomaly verification inspection, maintenance site inspection, emergency response inspection, or personnel safety inspection.

[0027] S103. Based on the inspection task, call the fixed video acquisition device covering or associated with the target inspection point to obtain fixed-angle video footage.

[0028] The inspection collaboration management platform queries the inspection video collaboration mapping table based on the inspection task to determine one or more fixed video acquisition devices corresponding to the target inspection area or target inspection point, and acquires fixed-view video images captured by the fixed video acquisition devices.

[0029] S104. Perform image recognition on the fixed-view video footage to obtain the fixed-view judgment result.

[0030] The inspection and collaborative management platform performs image recognition on fixed-view video footage to obtain fixed-view judgment results. For example, it can identify the operating status of cigarette manufacturing equipment, cigarette-making process parameters, and workshop environmental conditions.

[0031] S105. Perform point-level coverage reliability assessment on the image area corresponding to the target inspection point in the fixed-view video frame to obtain the fixed-view coverage reliability.

[0032] The inspection collaboration management platform determines whether the target inspection point is located within the coverage area of ​​the fixed video acquisition device and maps the target inspection point to the target image area in the fixed-view video frame. Subsequently, the platform analyzes the visibility, occlusion ratio, clarity, lighting conditions, target integrity, and sufficiency of the shooting angle of the target image area to obtain the reliability of the fixed-view coverage.

[0033] In some embodiments of the present invention, the fixed-viewpoint coverage reliability assessment process is as follows: 1. Calculate the visibility score V, occlusion ratio O, sharpness score S, illumination condition score L, target integrity score T, and shooting angle adequacy score A.

[0034] 2. Through the formula Calculate the coverage confidence C at a fixed viewpoint, where to The weights are 1 and the sum of the weights is 1.

[0035] The visibility score V is the ratio of the intersection area of ​​the target rectangular region projected onto the image plane and the image boundary to the area of ​​the target rectangular region.

[0036] The occlusion ratio O is the ratio of the number of pixels within the target rectangular area that are covered by non-target objects to the total number of pixels in that area.

[0037] The sharpness score S is obtained based on Laplace variance normalization: ,in The variance of the target region after Laplace filtering. This is the pre-calibrated optimal sharpness reference value.

[0038] The illumination condition score L is calculated based on the mean μ and standard deviation σ of the grayscale image of the target area: ,in , And L=0 when μ<40 or μ>220.

[0039] The target integrity score T is the ratio of the number of actual visible key points of the target inspection object in the image to the number of key points in the standard template.

[0040] The sufficiency score for the shooting angle is A, calculated using a piecewise function based on the angle θ between the camera's optical axis and the normal vector of the target inspection point: when θ ≤ θ good When A=1, when θ ≥ θ bad When θ = 0, A = 0. good < θ < θ bad hour , where θ good =30°, θ bad =75°.

[0041] For example, in a specific embodiment of the present invention, assuming the target inspection point is a dashboard of a cigarette manufacturing equipment, and the size of the target rectangular area of ​​the dashboard in the fixed camera image is 200×200 pixels, analysis shows: Visibility score V: The dashboard is completely within the frame, V=1.0.

[0042] Occlusion ratio O: The lower right corner of the dashboard is obscured by a tool, about 800 pixels, out of a total of 40,000 pixels, O=0.02.

[0043] Sharpness rating S: Laplace variance =1200, reference value =1500, S=0.8.

[0044] Illumination condition score L: mean gray value μ=110, standard deviation σ=35. =140, =100, then =1 - |110 - 140| / 100 = 0.7 =35 / 40=0.875, L=0.6×0.7+0.4×0.875=0.42+0.35=0.77.

[0045] Target integrity score T: There are 10 key points on the dashboard, and 8 were detected, so T=0.8.

[0046] Angle sufficiency score A: Included angle θ = 45°. =30°, =75°, A=1-(45-30) / (75-30)=1-15 / 45=0.667.

[0047] The inspection task type is equipment status inspection, with weights set as w1=0.1, w2=0.2, w3=0.25, w4=0.1, w5=0.25, w6=0.1. The calculation is C=0.1×1.0 + 0.2×(1-0.02) + 0.25×0.8 + 0.1×0.77 + 0.25×0.8 + 0.1×0.667 = 0.1 + 0.196 + 0.2 + 0.077 + 0.2 + 0.0667 = 0.8397. The preset confidence condition is C≥0.8 and A≥0.5, which is met; therefore, the smart safety helmet is not scheduled.

[0048] When the credibility of the fixed-view coverage meets the preset credibility conditions, the platform generates a fixed-view inspection record based on the fixed-view video footage. For example, when the fixed view can clearly observe the equipment status, area status, or personnel operation status at the inspection point, the platform saves the fixed-view video clip, screenshot, credibility score, inspection time, and fixed-view judgment result.

[0049] If C=0.65 in another scenario, the trust condition is not met, triggering scheduling and executing step S106.

[0050] S106. When the credibility of fixed viewpoint coverage does not meet the preset credibility conditions, determine the target smart safety helmet terminal from multiple smart safety helmet terminals.

[0051] When the credibility of fixed viewpoint coverage does not meet the preset credibility conditions, the platform determines that it is necessary to move the viewpoint to fill the blind spot or re-verify, and obtains the real-time status of multiple smart helmet terminals, calculates the collaborative inspection score of each terminal, and selects the terminal with the highest score and not lower than the minimum threshold as the target smart helmet terminal.

[0052] In some embodiments of the present invention, determining the target smart helmet terminal from a plurality of smart helmet terminals includes: 1. Obtain the location information, online status, battery status, wearer permissions, distance information, communication quality, and current task status of multiple smart safety helmet terminals.

[0053] 2. Calculate the collaborative inspection score for each smart safety helmet terminal. The collaborative inspection score is obtained by weighted summation of distance score, online status score, battery level score, access score, communication quality score, and idle status score.

[0054] 3. Select the smart helmet terminal with the highest collaborative inspection score that is not lower than the preset minimum threshold as the target smart helmet terminal.

[0055] The quantification methods for each sub-score in the collaborative inspection scoring are as follows: Distance rating: Where d is the spatial distance between the smart safety helmet terminal positioning and the target inspection point. This is the preset maximum effective distance.

[0056] Online status score: 1 when the terminal is online and has a normal heartbeat, and 0 when offline or when the heartbeat timeout occurs.

[0057] Battery rating: .

[0058] Permission score: 1 if the wearer has access or operation permission to the target inspection point, otherwise 0 and the terminal is removed from the candidate set.

[0059] Communication quality rating: based on RSSI normalization. .

[0060] Idle state score: 1 when the terminal is not currently performing other inspection tasks and has no unprocessed SOS; 0.5 when it is performing other tasks but is allowed to preempt; and 0 when it is performing a high-priority task or handling an SOS.

[0061] For example, in a specific embodiment of the present invention, assume the target inspection point is P-001, with a risk level of "medium". The statuses of the three candidate smart safety helmet terminals are as follows: H1: Distance 8m, online, battery 85%, authorized, RSSI=-65dBm, idle.

[0062] H2: Distance 25m, Online, Battery 40%, Permission granted, RSSI=-75dBm, Idle.

[0063] H3: Distance 5m, offline, battery 90%, authorized, RSSI=-55dBm, idle.

[0064] First, remove offline H3 files.

[0065] Calculate the sub-scores: H1: Distance score = 1 - 8 / 30 = 0.733; Battery score = 0.85; Communication score = (-65 + 100) / 60 = 0.583; Others are 1.

[0066] H2: Distance score = 1 - 25 / 30 = 0.167; Battery score = 0.4; Communication score = (-75 + 100) / 60 = 0.417.

[0067] Weights (in risk): α1=0.25, α2=0.15, α3=0.15, α4=0.25, α5=0.10, α6=0.10.

[0068] H1 total score = 0.25×0.733+0.15×1+0.15×0.85+0.25×1+0.10×0.583+0.10×1 =0.18325+0.15+0.1275+0.25+0.0583+0.1=0.86905.

[0069] H2 total score = 0.25×0.167+0.15×1+0.15×0.4+0.25×1+0.10×0.417+0.10×1 =0.04175+0.15+0.06+0.25+0.0417+0.1=0.64345.

[0070] H1 was selected as the target smart helmet terminal.

[0071] S107. Send an inspection collaboration instruction to the target smart safety helmet terminal. The inspection collaboration instruction includes one or more of the following: start video streaming and upload moving perspective video footage, perform remote photography, prompt the wearer to move to the designated inspection point, or prompt the wearer to adjust the shooting angle.

[0072] The platform sends inspection collaboration instructions to the target smart safety helmet terminal. These instructions can include activating video streaming, remotely taking photos, playing voice broadcasts, prompting the wearer to move to a designated inspection point, instructing the wearer to adjust the camera angle, or requiring on-site confirmation.

[0073] S108. Receive the moving-view video footage or remotely captured images transmitted back by the target smart safety helmet terminal.

[0074] The platform receives data transmitted back from the target smart safety helmet terminal. This transmitted data may include moving-view video footage, remotely captured images, voice broadcast results, or on-site confirmation results. The platform can further analyze the moving-view video footage or remotely captured images to identify inspection targets, abnormal targets, or personnel status, and generate inspection anomaly events or normal inspection results.

[0075] S109. Perform image recognition on the received moving-view video footage or remotely captured image to obtain the moving-view judgment result.

[0076] The inspection and collaborative management platform performs image recognition on moving-view video footage or remotely captured images to obtain judgment results regarding the moving perspective. For example, it can identify the operating status of cigarette manufacturing equipment, cigarette-making process parameters, and workshop environmental conditions.

[0077] S110. Make a consistency determination between the fixed-view judgment result and the moving-view judgment result, and generate a confirmation inspection result, a verification inspection result or a blind spot inspection result based on the determination result.

[0078] In this embodiment of the invention, the platform performs consistency determination on the fixed-view judgment result and the moving-view judgment result: if both exist and are the same, a confirmation inspection result is generated; if both exist but are different, a verification inspection result is generated and pushed for manual verification; if one is missing, the other is used to generate a blind spot inspection result.

[0079] In one specific embodiment of the present invention, after receiving video footage or remotely captured images from a moving perspective, the platform identifies the inspection object, abnormal target, or personnel status based on the moving perspective data, and obtains the moving perspective judgment result. The platform compares the fixed perspective judgment result with the moving perspective judgment result to generate different types of collaborative inspection results.

[0080] When both the fixed-view and moving-view assessment results indicate normalcy, or both indicate the same anomaly type, the platform generates a confirmatory inspection result. The confirmatory inspection result indicates that the fixed-view and moving-view results corroborate each other, making the inspection conclusion highly reliable.

[0081] When the fixed-view judgment result and the moving-view judgment result are inconsistent, the platform generates a verification inspection result and simultaneously pushes the fixed-view video clip, the moving-view video clip, the remotely photographed image, the fixed-view coverage credibility, the moving-view recognition result, and the manual verification entry to the central control terminal.

[0082] When the fixed-view judgment result is missing or the fixed-view coverage confidence level is lower than the preset confidence threshold, while the moving-view judgment result is valid, the platform uses the moving-view judgment result to generate a blind-spot-filling inspection result. The blind-spot-filling inspection result indicates that the conclusion of this inspection mainly comes from the moving first-person view of the smart safety helmet.

[0083] S111. Generate a unique inspection event identifier for the inspection task, and store the multi-source data generated during the inspection process based on the unique inspection event identifier to form a traceability link for the inspection process.

[0084] The platform generates a unique inspection event identifier and stores it in association. The platform unifies and links inspection task records, fixed video call records, fixed-view video clip records, smart safety helmet terminal call records, moving-view video clip records, remote photo records, broadcast message records, on-site confirmation records, SOS alarm records, abnormal event records, and inspection result records to form a traceability link for the inspection process.

[0085] In some embodiments of the present invention, when the smart safety helmet terminal uploads SOS alarm information, the platform determines the alarm personnel, alarm time, and alarm location based on the SOS alarm information. The platform queries the inspection video collaborative mapping table based on the alarm location to determine the fixed video acquisition device covering or adjacent to the alarm location, and simultaneously takes over the moving-view video feed of the smart safety helmet terminal.

[0086] The platform simultaneously pushes fixed-view and moving-view video feeds to the central control unit. The fixed-view feed is used to observe the surrounding environment of the alarm location, equipment operating status, and personnel distribution; the moving-view feed is used to observe the first-person perspective of the alarm personnel, the close-up situation at the scene, and the emergency response process.

[0087] The platform generates an SOS event identifier for SOS alarms and, based on the SOS event identifier, associates and stores SOS alarm information, alarm personnel information, alarm location, fixed-view video recording, moving-view video recording, remote broadcast recording, on-site handling recording, and alarm closure recording.

[0088] In some embodiments of the present invention, the platform also supports automatic switching of video sources. When the fixed video acquisition device corresponding to the inspection task is unavailable, the video stream is interrupted, the image quality is lower than a preset quality threshold, or the fixed viewpoint coverage reliability is lower than a preset reliability threshold, the platform automatically switches to the video stream of the target smart safety helmet terminal.

[0089] When the target smart helmet terminal is unavailable, offline, has a communication quality below a preset communication threshold, or has a battery level below a preset battery threshold, the platform automatically switches to a backup smart helmet terminal or a backup fixed video acquisition device. The backup smart helmet terminal can be determined based on the ranking results of the collaborative inspection scoring.

[0090] The platform can also automatically generate inspection task status based on the progress of the inspection tasks. Inspection task statuses include: pending inspection, fixed-view assessment in progress, fixed-view confirmed, moving-view blind spot filling in progress, manual review in progress, anomaly handling in progress, completed or archived, etc. Upon completion of the inspection task, the platform automatically generates an inspection report based on a unique inspection event identifier. The inspection report can include task information, fixed video recordings, moving video recordings, photo recordings, broadcast recordings, on-site confirmation records, SOS records, anomaly events, and handling results.

[0091] The collaborative inspection method for cigarette production sites provided by this invention can distinguish between "theoretical coverage" and "reliable coverage" by conducting point-level coverage reliability assessments of fixed-view video footage, thus avoiding misjudgments or omissions due to image quality issues. When the fixed viewpoint is unreliable, a target smart safety helmet terminal is identified from multiple smart safety helmet terminals, and a collaborative command is initiated to automatically schedule smart safety helmets to fill in the blind spots. By determining the consistency between the fixed-view and moving-view judgment results, different types of inspection results are generated. By generating a unique inspection event identifier, multi-source data during the inspection process is associated and stored, supporting post-event auditing, accountability tracking, and problem analysis.

[0092] Figure 2 This is a schematic diagram of the structure of a collaborative inspection device for cigarette production sites provided by the present invention, as shown below. Figure 2 As shown, the device includes: The mapping relationship management module 201 establishes and stores a collaborative mapping table for inspection videos, which associates inspection points, coverage areas of fixed video acquisition devices, smart safety helmet terminal positioning information, and wearer permission information. The inspection task acquisition module 202 is used to acquire inspection tasks and determine target inspection points. The fixed video calling module 203 is used to call the fixed video acquisition device that covers or is associated with the target inspection point according to the inspection task, and to obtain video images from a fixed perspective. The fixed-viewpoint recognition module 204 is used to perform image recognition on the fixed-viewpoint video frame to obtain a fixed-viewpoint judgment result. The coverage credibility assessment module 205 is used to perform point-level coverage credibility assessment on the image area corresponding to the target inspection point in the fixed-view video frame to obtain the fixed-view coverage credibility. The smart helmet scheduling module 206 is used to determine the target smart helmet terminal from multiple smart helmet terminals when the credibility of the fixed view coverage does not meet the preset credibility conditions. The collaborative instruction sending module 207 is used to send inspection collaborative instructions to the target smart safety helmet terminal. The inspection collaborative instructions include one or more of the following: starting a video stream to upload a moving perspective video image, performing remote photography, prompting the wearer to move to a designated inspection point, or prompting the wearer to adjust the shooting angle. The mobile video receiving module 208 is used to receive mobile perspective video images or remotely captured images transmitted back by the target smart safety helmet terminal. The motion perspective recognition module 209 is used to perform image recognition on the received motion perspective video footage or remotely captured images to obtain the motion perspective judgment result. The dual-view judgment module 210 is used to determine the consistency between the fixed view judgment result and the moving view judgment result, and generate a confirmation inspection result, a verification inspection result or a blind spot inspection result based on the judgment result. The record traceability module 211 is used to generate a unique inspection event identifier for the inspection task, and to associate and store multi-source data generated during the inspection process based on the unique inspection event identifier to form an inspection process traceability link.

[0093] In some embodiments of the present invention, the coverage confidence assessment module 205 includes: The sub-score calculation submodule is used to calculate the visibility score V, occlusion ratio O, sharpness score S, illumination condition score L, target integrity score T, and shooting angle adequacy score A; The feasibility calculation submodule is used to calculate the feasibility using the formula. Calculate the coverage confidence C at a fixed viewpoint, where to The weights are 1, and the sum of the weights is 1. Wherein, the visibility score V is the ratio of the intersection area of ​​the target rectangular region projected onto the image plane and the image boundary to the area of ​​the target rectangular region; The occlusion ratio O is the ratio of the number of pixels within the target rectangular area that are covered by non-target objects to the total number of pixels in that area; The sharpness score S is obtained based on Laplace variance normalization: ,in The variance of the target region after Laplace filtering. This is the pre-calibrated optimal sharpness reference value; The illumination status score L is calculated based on the mean μ and standard deviation σ of the grayscale image of the target area: ,in , And L=0 when μ<40 or μ>220; The target integrity score T is the ratio of the number of actual visible key points of the target inspection object in the image to the number of key points in the standard template; The sufficiency score A for the shooting angle is calculated as a piecewise function based on the angle θ between the camera's optical axis and the normal vector of the target inspection point: when θ ≤ θ good When A=1, when θ ≥ θ bad When θ = 0, A = 0. good < θ < θ bad hour , where θ good =30°, θ bad =75°.

[0094] In some embodiments of the present invention, the smart safety helmet scheduling module 206 includes: The safety helmet acquisition submodule is used to acquire the location information, online status, battery status, wearer permissions, distance information, communication quality and current task status of multiple smart safety helmet terminals; The collaborative inspection score calculation submodule is used to calculate the collaborative inspection score of each smart safety helmet terminal. The collaborative inspection score is obtained by weighted summation of distance score, online status score, battery level score, permission score, communication quality score and idle status score. The target safety helmet determination submodule is used to select the smart safety helmet terminal with the highest collaborative inspection score that is not lower than a preset minimum threshold as the target smart safety helmet terminal.

[0095] In some embodiments of the present invention, the quantification method of each sub-score in the collaborative inspection scoring is as follows: Distance rating: Where d is the spatial distance between the smart safety helmet terminal positioning and the target inspection point. The preset maximum effective distance; Online status score: 1 when the terminal is online and the heartbeat is normal, and 0 when offline or the heartbeat timeout occurs; Battery rating: ; Access score: 1 is given when the wearer has access or operation permission to the target inspection point; otherwise, 0 is given and the terminal is removed from the candidate set. Communication quality score: based on RSSI normalization, ; Idle state score: 1 when the terminal is not currently performing other inspection tasks and has no unprocessed SOS; 0.5 when it is performing other tasks but is allowed to preempt; and 0 when it is performing a high-priority task or handling an SOS.

[0096] In some embodiments of the present invention, the dual-view determination module 210 includes: The first result generation submodule is used to generate a confirmatory inspection result when the fixed view judgment result and the moving view judgment result are the same. The second result generation submodule is used to generate a verification inspection result when the fixed viewpoint judgment result is different from the moving viewpoint judgment result, and simultaneously push the fixed viewpoint video clip, the moving viewpoint video clip, the remotely photographed image, the fixed viewpoint coverage credibility, the moving viewpoint recognition result and the manual verification entry to the central control terminal. The third result generation submodule is used to generate blind spot inspection results when the fixed viewpoint coverage confidence is lower than the preset confidence threshold, resulting in the loss of fixed viewpoint judgment results.

[0097] In some embodiments of the present invention, the collaborative inspection device for cigarette production sites further includes: The alarm information determination module is used to determine the alarm personnel, alarm time, and alarm location based on the alarm information when it receives SOS alarm information uploaded by the smart safety helmet terminal. The adjacent device determination module is used to query the inspection video collaborative mapping table according to the alarm location to determine the fixed video acquisition device that covers the alarm location or is adjacent to the alarm location, and at the same time take over the moving view video screen of the smart safety helmet terminal. The video push module is used to simultaneously push fixed-view video footage and moving-view video footage to the central control terminal; The alarm information association module is used to generate an SOS event identifier for an SOS alarm, and to associate and store the SOS alarm information, alarm personnel information, alarm location, fixed-view video recording, moving-view video recording, remote broadcast recording, on-site handling recording, and alarm closure recording based on the SOS event identifier.

[0098] In some embodiments of the present invention, the collaborative inspection device for cigarette production sites further includes: The first switching module is used to automatically switch to the video stream of the target smart safety helmet terminal when the fixed video acquisition device corresponding to the inspection task is unavailable, the video stream is interrupted, the image quality is lower than the preset quality threshold, or the fixed viewpoint coverage reliability is lower than the preset reliability threshold. The second switching module is used to automatically switch to a backup smart helmet terminal or a backup fixed video acquisition device when the target smart helmet terminal is unavailable, offline, has a communication quality lower than a preset communication threshold, or has a battery level lower than a preset battery level threshold. The backup smart helmet terminal is determined based on the ranking results of the collaborative inspection scoring.

[0099] The aforementioned collaborative inspection device for cigarette production sites can execute the collaborative inspection method for cigarette production sites provided in the foregoing embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the collaborative inspection method for cigarette production sites.

[0100] Figure 3 This is a schematic diagram of an electronic device provided by the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0101] like Figure 3 As shown, the electronic device includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer programs stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0102] Multiple components in the electronic device are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, optical disk, etc.; and a communication unit 19, such as a network card, modem, wireless transceiver, etc. The communication unit 19 allows the electronic device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0103] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as collaborative inspection methods in cigarette production sites.

[0104] In some embodiments, the collaborative inspection method for cigarette production sites can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on an electronic device via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the collaborative inspection method for cigarette production sites described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the collaborative inspection method for cigarette production sites by any other suitable means (e.g., by means of firmware).

[0105] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0106] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0107] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0108] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0109] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0110] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0111] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the collaborative inspection method for cigarette production sites as provided in any embodiment of this application.

[0112] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0113] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and no limitation is imposed herein.

[0114] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A collaborative inspection method for cigarette production sites, characterized in that, include: Establish and store a collaborative mapping table for inspection videos, which associates inspection points, coverage areas of fixed video acquisition devices, smart safety helmet terminal positioning information, and wearer permission information. Obtain inspection tasks and determine target inspection points; According to the inspection task, call the fixed video acquisition device that covers or is associated with the target inspection point to obtain video footage from a fixed perspective. Image recognition is performed on the fixed-viewpoint video footage to obtain the fixed-viewpoint determination result; A point-level coverage confidence assessment is performed on the image region corresponding to the target inspection point in the fixed-view video frame to obtain the fixed-view coverage confidence. When the credibility of the fixed viewpoint coverage does not meet the preset credibility condition, the target smart helmet terminal is determined from multiple smart helmet terminals. Send an inspection collaboration instruction to the target smart safety helmet terminal. The inspection collaboration instruction includes one or more of the following: start a video stream to upload a moving perspective video image, perform remote photography, prompt the wearer to move to a designated inspection point, or prompt the wearer to adjust the shooting angle. Receive moving-view video footage or remotely captured images from the target smart safety helmet terminal; Image recognition is performed on the received moving-view video footage or remotely captured images to obtain the moving-view determination result; The consistency between the fixed-view judgment result and the moving-view judgment result is determined, and a confirmation inspection result, a verification inspection result, or a blind spot inspection result is generated based on the determination result. A unique inspection event identifier is generated for the inspection task, and the multi-source data generated during the inspection process is associated and stored based on the unique inspection event identifier to form a traceability link for the inspection process.

2. The collaborative inspection method for cigarette production sites according to claim 1, characterized in that, A point-level coverage reliability assessment is performed on the image region corresponding to the target inspection point in the fixed-view video frame to obtain the fixed-view coverage reliability, including: Calculate the visibility score V, occlusion ratio O, sharpness score S, illumination condition score L, target integrity score T, and shooting angle adequacy score A; Through formula Calculate the coverage confidence C at a fixed viewpoint, where to The weights are 1, and the sum of the weights is 1. Wherein, the visibility score V is the ratio of the intersection area of ​​the target rectangular region projected onto the image plane and the image boundary to the area of ​​the target rectangular region; The occlusion ratio O is the ratio of the number of pixels within the target rectangular area that are covered by non-target objects to the total number of pixels in that area; The sharpness score S is obtained based on Laplace variance normalization: ,in The variance of the target region after Laplace filtering. This is a pre-calibrated optimal sharpness reference value; The illumination status score L is calculated based on the mean μ and standard deviation σ of the grayscale image of the target area: ,in , And L=0 when μ<40 or μ>220; The target integrity score T is the ratio of the number of actual visible key points of the target inspection object in the image to the number of key points in the standard template; The sufficiency score A for the shooting angle is calculated as a piecewise function based on the angle θ between the camera's optical axis and the normal vector of the target inspection point: when θ ≤ θ good When A=1, when θ ≥ θ bad When θ = 0, A = 0. good < θ < θ bad hour , where θ good =30°, θ bad =75°.

3. The collaborative inspection method for cigarette production sites according to claim 1 or 2, characterized in that, The target smart helmet terminal was identified from multiple smart helmet terminals, including: Acquire location information, online status, battery status, wearer permissions, distance information, communication quality, and current task status of multiple smart safety helmet terminals; Calculate the collaborative inspection score for each smart safety helmet terminal. The collaborative inspection score is obtained by weighted summation of distance score, online status score, battery level score, access score, communication quality score, and idle status score. The smart helmet terminal with the highest collaborative inspection score that is not lower than the preset minimum threshold is selected as the target smart helmet terminal.

4. The collaborative inspection method for cigarette production sites according to claim 3, characterized in that, The quantification method for each sub-score in the collaborative inspection scoring is as follows: Distance rating: Where d is the spatial distance between the smart safety helmet terminal positioning and the target inspection point. The preset maximum effective distance; Online status score: 1 when the terminal is online and the heartbeat is normal, and 0 when offline or the heartbeat timeout occurs; Battery rating: ; Access score: 1 is given when the wearer has access or operation permission to the target inspection point; otherwise, 0 is given and the terminal is removed from the candidate set. Communication quality rating: based on RSSI normalization. ; Idle state score: 1 when the terminal is not currently performing other inspection tasks and has no unprocessed SOS; 0.5 when it is performing other tasks but is allowed to preempt; and 0 when it is performing a high-priority task or handling an SOS.

5. The collaborative inspection method for cigarette production sites according to claim 1, characterized in that, The consistency between the fixed-viewpoint judgment result and the moving-viewpoint judgment result is determined, and a confirmation-type inspection result, a verification-type inspection result, or a blind-spot-filling inspection result is generated based on the determination result, including: When the fixed-view judgment result is the same as the moving-view judgment result, a confirmatory inspection result is generated. When the fixed-view judgment result differs from the moving-view judgment result, a verification inspection result is generated, and the fixed-view video clip, the moving-view video clip, the remotely captured image, the fixed-view coverage credibility, the moving-view recognition result, and the manual verification entry are simultaneously pushed to the central control terminal. When the confidence level of the fixed viewpoint coverage is lower than the preset confidence threshold, resulting in the absence of fixed viewpoint judgment results, the judgment results of the moving viewpoint are used to generate blind spot inspection results.

6. The collaborative inspection method for cigarette production sites according to claim 1, characterized in that, Also includes: When an SOS alarm message is received from the smart safety helmet terminal, the alarm personnel, alarm time, and alarm location are determined based on the alarm message. Based on the alarm location, query the inspection video collaborative mapping table to determine the fixed video acquisition device covering the alarm location or adjacent to the alarm location, and simultaneously take over the moving view video screen of the smart safety helmet terminal. Simultaneously push fixed-view video footage and moving-view video footage to the central control unit; An SOS event identifier is generated for the SOS alarm, and the SOS alarm information, alarm personnel information, alarm location, fixed-view video recording, moving-view video recording, remote broadcast recording, on-site handling recording, and alarm closure recording are associated and stored based on the SOS event identifier.

7. The collaborative inspection method for cigarette production sites according to claim 1, characterized in that, Also includes: When the fixed video acquisition device corresponding to the inspection task is unavailable, the video stream is interrupted, the image quality is lower than the preset quality threshold, or the fixed viewpoint coverage reliability is lower than the preset reliability threshold, the video stream is automatically switched to the target smart safety helmet terminal. When the target smart helmet terminal is unavailable, offline, has a communication quality lower than a preset communication threshold, or has a battery level lower than a preset battery level, it automatically switches to a backup smart helmet terminal or a backup fixed video acquisition device. The backup smart helmet terminal is determined based on the ranking results of the collaborative inspection scoring.

8. A collaborative inspection device for cigarette production sites, characterized in that, include: The mapping relationship management module establishes and stores a collaborative mapping table for inspection videos. The mapping table associates inspection points, coverage areas of fixed video acquisition devices, smart safety helmet terminal positioning information, and wearer permission information. The inspection task acquisition module is used to acquire inspection tasks and determine target inspection points. The fixed video recall module is used to recall fixed video acquisition devices that cover or are associated with the target inspection point according to the inspection task, and to acquire fixed-view video images. A fixed-viewpoint recognition module is used to perform image recognition on the fixed-viewpoint video frame to obtain a fixed-viewpoint judgment result; The coverage credibility assessment module is used to perform point-level coverage credibility assessment on the image area corresponding to the target inspection point in the fixed-view video frame to obtain the fixed-view coverage credibility. The smart helmet scheduling module is used to determine the target smart helmet terminal from multiple smart helmet terminals when the credibility of the fixed view coverage does not meet the preset credibility conditions. The collaborative instruction sending module is used to send inspection collaborative instructions to the target smart safety helmet terminal. The inspection collaborative instructions include one or more of the following: starting a video stream to upload moving perspective video images, performing remote photography, prompting the wearer to move to a designated inspection point, or prompting the wearer to adjust the shooting angle. The mobile video receiving module is used to receive mobile-view video footage or remotely captured images transmitted back by the target smart safety helmet terminal. The motion perspective recognition module is used to perform image recognition on received motion perspective video footage or remotely captured images to obtain motion perspective judgment results. The dual-view judgment module is used to determine the consistency between the fixed-view judgment result and the moving-view judgment result, and generate a confirmation inspection result, a verification inspection result, or a blind spot inspection result based on the judgment result. The record and traceability module is used to generate a unique inspection event identifier for the inspection task, and to associate and store multi-source data generated during the inspection process based on the unique inspection event identifier to form an inspection process traceability link.

9. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the collaborative inspection method for cigarette production sites as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the collaborative inspection method for cigarette production sites as described in any one of claims 1-7.

Images