A matching detection method for personnel qualification and individual protective equipment for hazardous chemical operation

Abstract

The present application relates to the technical field of image recognition, and discloses a matching detection method for personnel qualification and individual protective equipment for dangerous chemical operation; a target protection level and an air-tightness benchmark are determined according to a region category, dynamic images and real-time face images are collected; an identity mark is matched with a qualification database to determine a permitted operation range, identity is verified according to a motion characteristic signal, and a wearing protection level is identified by recognizing an equipment mark; a fitting degree is determined according to dynamic characteristics of clothing wrinkles and a level is matched, and air-tightness comparison is performed according to the continuity distribution of a face boundary and an equipment edge; operation permission instructions are generated according to a matching result and an air-tightness comparison result, and an identity mark and a region category are associated to save archives, so that linkage detection of personnel qualification and wearing state of protective equipment is realized.

Description

Technical Field

[0001] This invention relates to the field of image recognition technology, and in particular to a method for matching and detecting the qualifications of personnel and their personal protective equipment in hazardous chemical operations. Background Technology

[0002] At the entry point for hazardous chemical spill response, personnel qualification verification and personal protective equipment (PPE) inspection are used to confirm the tasks that operators can undertake and whether the current equipment meets the requirements for entering the corresponding area. Personnel qualification verification is usually completed through document scanning, card reading, and qualification database queries, comparing certificate numbers with file records to verify the validity of the certificate and the scope of permitted operations. PPE inspection can be carried out through on-site images, reading equipment markings, and manual observation to identify the equipment category, protection level, and wearing status. Equipment registration information can explain the nominal attributes of the equipment, and personnel qualification files can explain the authorized scope of operation. The two types of information form the verification results.

[0003] However, when an operator has passed the qualification verification but their current wearing status changes, it is difficult to link the same operator's qualifications, permitted work scope, and current equipment protection level and fit status into a single test result by simply checking the personnel file and equipment registration information separately. Therefore, it is difficult to form a work permit judgment corresponding to the requirements of the target work area based on this. Summary of the Invention

[0004] This invention provides a matching detection method for personnel qualifications and personal protective equipment (PPE) in hazardous chemical operations. Its main purpose is to solve the problem that personnel qualifications and PPE wearing status cannot be detected in conjunction in the existing technology.

[0005] To achieve the above objectives, the present invention provides a method for matching and testing the qualifications of personnel and personal protective equipment in hazardous chemical operations, comprising:

[0006] S1: Obtain the area category of the hazardous chemical leak area and the identification of the operators, determine the target protection level and airtightness benchmark based on the area category, and collect dynamic images and real-time facial images of the operators;

[0007] S2: Match the qualification level with the identity identifier in the qualification database, compare the qualification level with the target protection level, and obtain the permitted operation scope;

[0008] S3: Extract motion feature signals and clothing fold dynamic features from dynamic images, match the motion feature signals with motion feature samples in the qualification database to obtain the identity verification result, and when the identity verification result is valid, identify the protective equipment logo in the dynamic image to obtain the wearing protection level.

[0009] S4: Identify the fit of protective equipment based on the dynamic characteristics of clothing folds, generate a measured value of the protection level by combining the wearing protection level, match the measured value of the protection level with the permitted work range, and obtain the matching result of the protective equipment.

[0010] S5: Extract the distribution of the continuity between the facial boundary and the edge of the protective equipment in the real-time face image to obtain the airtight fit state. Compare the airtight fit state with the airtight reference to obtain the airtight comparison result.

[0011] S6: Generate a work permit instruction based on the matching results of protective equipment and the airtightness comparison results, associate the work permit instruction with the identity identifier and area category, and save it as a work permit file.

[0012] In one possible implementation, the qualification level is matched against the identity identifier in the qualification database, and the qualification level is compared with the target protection level to obtain the permitted scope of operation, including:

[0013] The identity identifier is broken down into the current certificate code and the historical certificate code, and candidate current files and candidate historical files are retrieved respectively.

[0014] The historical certificate code is compared with the historical certificate code recorded in the candidate current file. If they do not match, the certificate renewal record of the candidate historical file is traced to the current valid file. If they match, the candidate current file is selected to obtain a unique matching file.

[0015] Read the review status identifier of the unique matching file. If the assessment conclusion is passed and the assessment date is within the specified validity period before the current date, extract the qualification level. Compare the hazardous chemical media category, highest protection level and operation type in the qualification level with the corresponding requirements of the target protection level to obtain the permitted operation scope.

[0016] In one possible implementation, the qualification level is matched against the identity identifier in the qualification database, and the qualification level is compared with the target protection level to obtain the permitted scope of operation, including:

[0017] The inclusion comparison between the hazardous chemical media category in the qualification level and the leakage media type in the target protection level is performed to obtain the media matching result;

[0018] When the media matching result is valid, the highest protection level in the qualification level is compared with the lowest protection level in the target protection level in order of protection capability from strong to weak to obtain the strength matching result.

[0019] Based on the intensity matching results and the leakage concentration level of the area category, the concentration is divided into zones to obtain candidate operation ranges. The task project number of each spatial sub-area within the candidate operation range is matched with the corresponding number of the operation type in the qualification level. The boundary coordinates of the successfully matched spatial sub-area and the operation tasks are associated to obtain the permitted operation range.

[0020] In one possible implementation, motion feature signals and clothing fold dynamic features are extracted from the dynamic image. The motion feature signals are then matched with motion feature samples in a qualification database to obtain an authentication result. If the authentication result is valid, the protective equipment identifier in the dynamic image is identified to determine the wearing protection level, including:

[0021] A stage sequence is formed based on the changing time sequence of motion feature signals and dynamic features of clothing folds. Continuous motion segments that are consistent with the sequence of motion feature samples are retained to obtain sequentially consistent segments.

[0022] From the sequentially consistent segments, select the spatially matching segments whose joint bending direction matches the crease change, and compare the crease extension direction of the spatially matching segments with the corresponding moment of the motion feature sample to obtain the equipment linkage segment;

[0023] The coverage of the equipment linkage section is checked in stages. The ratio of the total duration of the equipment linkage section to the total duration of all action stages is used as the coverage level. The identity verification result is generated based on the comparison result of the coverage level and the preset qualified threshold. When the identity verification result is valid, the protective equipment logo is identified to obtain the wearing protection level.

[0024] In one possible implementation, motion feature signals and clothing fold dynamic features are extracted from the dynamic image. The motion feature signals are then matched with motion feature samples in a qualification database to obtain an authentication result. If the authentication result is valid, the protective equipment identifier in the dynamic image is identified to determine the wearing protection level, including:

[0025] When the identity verification result is valid, the micro-deformation region and the deformation-stable region are distinguished according to the dynamic characteristics of clothing folds. The candidate regions for identification that have abrupt changes in reflected light intensity and form closed edges are extracted from the deformation-stable regions.

[0026] Compare the edge orientation of the candidate identification area with the crease extension direction of the micro-deformation area, and retain intersecting or independent valid identification edges;

[0027] The pattern enclosed by the effective marking edge and the arrangement of the internal color blocks are compared with the pre-stored marking sample. The protection level marking information of the matching sample is read, and the wearing protection level is determined according to the protection level code in the protection level marking information.

[0028] In one possible implementation, the fit of protective equipment is identified based on the dynamic characteristics of clothing folds. This is combined with the wear protection level to generate a measured protection level value. The measured protection level value is then matched with the permitted work area to obtain the matching result of the protective equipment, including:

[0029] By tracking the creases in the dynamic features of clothing folds frame by frame, the distribution of crease bands is obtained, and the sliding crease bands that have undergone positional migration and the stationary crease bands that have fixed positions are distinguished.

[0030] The convergence node is identified based on the adjacency relationship between the slip crease zone and the dwell crease zone, and the directional concentration area is identified according to the direction of the creases around the convergence node;

[0031] The closed-bound areas within the directional concentration region are designated as the fit-support zone, and the open-bound areas are designated as the fit-detachment zone. By combining the spatial envelopment relationship, the dynamic contact state between the protective equipment and the operator's body is analyzed to obtain the fit degree. Based on the fit degree, the wearing protection level is corrected to form the measured protection level value. This value is then compared with the lower limit of the protection level for the permitted work area to obtain the matching result of the protective equipment.

[0032] In one possible implementation, the fit of protective equipment is identified based on the dynamic characteristics of clothing folds. This is combined with the wear protection level to generate a measured protection level value. The measured protection level value is then matched with the permitted work area to obtain the matching result of the protective equipment, including:

[0033] Continuous crease regions with fixed positions are selected from the dynamic features of clothing folds, and the cross-sections where the continuous crease regions intersect with the edges of protective equipment are extracted.

[0034] Based on the situation where the edge of the protective equipment is divided into two parts by the cross section and the connection between the two parts, the gap connection state is obtained. The wearing protection level in the gap connection state is downgraded according to the degree of fit, and the actual protection level value is obtained.

[0035] The measured value of the protection level is compared with the lower limit of the protection level in the permitted operation range in order of protection capability. If the lower limit of the protection level is reached, the matching result is considered passed; otherwise, it is considered failed, thus obtaining the matching result of the protective equipment.

[0036] In one possible implementation, the continuous distribution of facial boundaries and the edges of protective equipment in a real-time face image is extracted to obtain the airtight fit state. This airtight fit state is then compared with an airtight reference to obtain the airtight comparison result, including:

[0037] Samples are taken from the facial boundary toward the edge of the protective equipment to form a grayscale transition zone, and the grayscale gradient interruption section is marked as a micro-gap region;

[0038] By arranging the micro-gap regions and gray-scale gradient continuous segments sequentially along the facial boundary, an edge continuity distribution is obtained, connecting spatially adjacent micro-gap regions into potential leakage paths.

[0039] The positions at both ends of the potential leakage path are compared with the two sides of the contact area of ​​the face boundary and the edge of the protective equipment to obtain the airtight fit status indicating whether the potential leakage path penetrates the contact area. Then, the airtight fit status is compared with the allowable gap range of the corresponding face area in the airtight reference to obtain the airtight comparison result.

[0040] In one possible implementation, the continuous distribution of facial boundaries and the edges of protective equipment in a real-time face image is extracted to obtain the airtight fit state. This airtight fit state is then compared with an airtight reference to obtain the airtight comparison result, including:

[0041] The edge continuity distribution corresponding to the airtight fit state of the facial area is matched with the allowable gap range of the airtight reference to obtain the area continuity distribution;

[0042] The regions where grayscale gradients are interrupted in the continuous distribution of the region are taken as micro-gap regions. The discontinuities at both ends of the micro-gap regions are extracted and connected to form discontinuous segments. The position and extension length of the discontinuous segments are obtained. The facial region, start position and end position of each discontinuous segment are recorded and arranged in the order of extension along the facial boundary.

[0043] Discontinuous segments whose extension length exceeds the allowable discontinuity range of the corresponding facial area are marked as airtightness non-compliant segments. The abnormal segment distribution is obtained by arranging them by position. When there are no airtightness non-compliant segments in the abnormal segment distribution, a qualified airtightness comparison result is generated. When there are airtightness non-compliant segments, an unqualified airtightness comparison result is generated.

[0044] In one possible implementation, a work permit instruction is generated based on the matching results of the protective equipment and the airtightness comparison results. This work permit instruction is then associated with an identity identifier and area category and saved as a work permit file, including:

[0045] Read the measured value of the protection level corresponding to the matching result of the protective equipment, compare it with the protection level range allowed by the permitted operation range in order of protection capability, and obtain the protection level matching conclusion.

[0046] Read the airtightness comparison result and compare it with the allowable gap boundary range of the airtightness reference to obtain the airtightness boundary comparison conclusion.

[0047] By correlating the location of the local detachment section in the airtight boundary comparison conclusion with the sealing structure characteristics of the protective equipment, the airtightness verification result is obtained. Based on the protection level adaptation conclusion and the airtightness verification result, an operation permit order is generated to allow or prohibit entry for operation. The operation permit order is then associated with the identity identifier and area category and saved as an operation permit file.

[0048] Beneficial effects

[0049] This solution uses the qualification level obtained by matching the identity identifier with the qualification database to determine the scope of permitted operations, and matches the motion feature signals in the dynamic image with the motion feature samples. When the identity verification result is valid, the protective equipment identifier in the dynamic image is then identified. Since the qualification verification and equipment identifier identification are associated with the same operator, the personnel qualification, permitted scope of operation and the level of protective equipment worn can be included in the same detection, reducing the problem of inconsistent detection objects when querying separately.

[0050] This solution identifies the fit degree based on the dynamic characteristics of clothing folds and matches the measured protection level obtained by combining the protection level with the permitted work area. At the same time, it compares the airtightness of the continuous distribution of the facial boundary and the edge of the protective equipment. Since the work permit instruction uses both the matching result of the protective equipment and the airtightness comparison result, and is associated with and saved with the identity identifier and area category, it can form a permit file containing personnel, area and the test results of this equipment, providing corresponding records for subsequent verification and permit judgment. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating a method for matching and testing personnel qualifications and personal protective equipment in hazardous chemical operations, as provided in an embodiment of the present invention.

[0052] Figure 2 A flowchart of the airtight fit detection sub-process of a matching detection method for personnel qualifications and personal protective equipment in hazardous chemical operations, provided as an embodiment of the present invention;

[0053] Figure 3 The flowchart illustrates the regional category determination and protection level assessment of a matching detection method for personnel qualifications and personal protective equipment in hazardous chemical operations, provided as an embodiment of the present invention.

[0054] Figure 4 A flowchart illustrating the comprehensive matching determination of personnel qualifications and personal protective equipment in a method for matching personnel qualifications and personal protective equipment in hazardous chemical operations, provided as an embodiment of the present invention.

[0055] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0056] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0057] This application provides a method for matching and testing the qualifications of personnel and personal protective equipment (PPE) for handling hazardous chemicals. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for matching and testing the qualifications of personnel and PPE for handling hazardous chemicals can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0058] Reference Figures 1 to 4 As shown in this embodiment, a method for matching and testing the qualifications of personnel and personal protective equipment for handling hazardous chemicals includes:

[0059] S1: Obtain the area category of the hazardous chemical leak area and the identification of the operators, determine the target protection level and airtightness benchmark based on the area category, and collect dynamic images and real-time facial images of the operators;

[0060] Specifically, the hazardous chemical leak area refers to the spatial range in which a hazardous chemical leak accident occurs. This range is determined based on the location of the leak source, the direction of diffusion of the leaked medium, and the diffusion distance.

[0061] The area category is a classification of hazardous chemical spill areas based on the scope and severity of the hazard and the location of the spill source. Specifically, the area categories include high-hazard areas, medium-hazard areas, low-hazard areas, and safe areas. The high-hazard area is the area closest to the spill source, where the concentration of hazardous chemicals is the highest. The concentrations of hazardous chemicals in the medium-hazard and low-hazard areas decrease sequentially. The safe area is the area where the concentration is below the hazard threshold.

[0062] The target protection level is determined based on the area category of the hazardous chemical spill zone, and represents the level of protective equipment that operators must wear when entering the area. Specifically, protection levels are divided into Level 1, Level 2, and Level 3 protection according to the hazard of the leaked medium. Level 1 protection is the highest level of protection and is applicable to situations requiring the highest level of protection for skin, respiratory organs, and eyes. Level 2 protection is applicable to situations requiring the highest level of respiratory protection but with slightly lower requirements for skin protection. Level 3 protection is applicable when the type and concentration of airborne substances are known. For example, when the area category is a high-hazard zone, the corresponding target protection level is Level 1 protection; when the area category is a moderate-hazard zone, the corresponding target protection level is Level 2 protection.

[0063] The airtightness standard is a pre-established standard for judging the airtightness of protective equipment corresponding to the target protection level. The airtightness standard includes the allowable gap range divided according to the facial area. This allowable gap range is determined according to the standard airtightness parameters corresponding to the model of protective equipment.

[0064] Dynamic images are video image sequences obtained by continuously capturing images of an operator performing a specified sequence of actions through an image acquisition device; the specified sequence of actions includes raising both arms, squatting, and turning left and right.

[0065] Real-time face images are single-frame images obtained by capturing the operator's facial area through an image acquisition device at the same time as or adjacent to the acquisition of dynamic images.

[0066] Furthermore, after a hazardous chemical leak occurs, the spatial extent of the leak area is first determined based on the location of the leak source, the direction of diffusion of the leaking medium, and the diffusion distance. Then, based on the concentration levels of hazardous chemicals within this spatial area, the area is divided into different zone categories such as heavily hazardous zone, moderately hazardous zone, lightly hazardous zone, and safe zone.

[0067] Based on the identified area category, determine the target protection level that operators should achieve when entering the area for work. For example, when the area category is a severely hazardous area, the target protection level is determined to be Level 1 protection, and when the area category is a moderately hazardous area, the target protection level is determined to be Level 2 protection.

[0068] At the same time, based on the model of the protective equipment corresponding to the target protection level, the airtightness qualification standard of the protective equipment of that model is retrieved from the pre-stored standard airtightness parameters and used as the airtightness benchmark.

[0069] After the operator puts on protective equipment, instruct the operator to stand within the field of view of the image acquisition device and perform the standard sequence of movements: raising both arms, squatting, and turning left and right.

[0070] During the process of the operator performing the above standard action sequence, the image acquisition device continuously captures the operator's full-body image to obtain dynamic images; at the same time, at a designated moment during the dynamic image acquisition, the image acquisition device focuses on the operator's facial area to capture real-time facial images.

[0071] S2: Match the qualification level with the identity identifier in the qualification database, compare the qualification level with the target protection level, and obtain the permitted operation scope;

[0072] In this embodiment, the qualification level is matched with the identity identifier in the qualification database, and the qualification level is compared with the target protection level to obtain the permitted scope of operation, including:

[0073] The identity identifier is broken down into the current certificate code and the historical certificate code, and candidate current files and candidate historical files are retrieved respectively.

[0074] The historical certificate code is compared with the historical certificate code recorded in the candidate current file. If they do not match, the certificate renewal record of the candidate historical file is traced to the current valid file. If they match, the candidate current file is selected to obtain a unique matching file.

[0075] Read the review status identifier of the unique matching file. If the assessment conclusion is passed and the assessment date is within the specified validity period before the current date, extract the qualification level. Compare the hazardous chemical media category, highest protection level and operation type in the qualification level with the corresponding requirements of the target protection level to obtain the permitted operation scope.

[0076] The identity identifier is the identification information used in this method to associate operators and their qualification records. The disassembly steps described later are applicable to situations where this identification information records both the current certificate code and the historical certificate code according to the known field rules.

[0077] The current certificate code is the serial number of the operator's latest valid qualification certificate, and its function is to directly point to the latest qualification record.

[0078] Historical certificate codes are serial numbers of expired or renewed certificates held by the operator before the current certificate.

[0079] Disassembly is an operation that extracts the current certificate code and the historical certificate code from the specified bits of the identity identifier according to the certificate code arrangement rules.

[0080] The qualification database is a pre-built electronic dataset used to store all records of changes in the qualifications of personnel handling hazardous chemicals.

[0081] The retrieval process uses the current certificate code as an index to search the qualification database, retrieve the certificate file that is completely consistent with the code, and use it as a candidate current file.

[0082] The current candidate file is an electronic record set containing the operator's latest certificate information and review status.

[0083] Candidate historical archives are electronic records that record the operator's past certificate information, retrieved from the qualification database using historical certificate codes as an index.

[0084] Reverse consistency comparison is the process of verifying whether the historical certificate codes recorded in the current candidate file match the historical certificate codes as a benchmark.

[0085] The reverse comparison result is a logical value indicating whether the historical certificate code is consistent with the historical certificate code recorded in the candidate current file. It is used to verify whether the personnel represented by the candidate current file and the historical certificate code constitute a valid upstream and downstream connection.

[0086] The certificate renewal record is the associated index path in the old certificate file that points to the new certificate file when a certificate expires or is lost and needs to be replaced.

[0087] Step-by-step tracing is an iterative addressing operation that follows the replacement pointer recorded in the candidate historical archives, from the old archive to the new archive, then from the new archive to the updated archive, and so on, until the final archive that is currently in a valid state is found.

[0088] The currently valid files are the latest qualification files that are in a valid state, obtained through step-by-step tracking.

[0089] The unique matching file is the final, definitive electronic record that accurately reflects the operator's current legal qualifications.

[0090] The review status indicator is a flag stored in a unique matching file, which indicates whether the operator's professional qualifications have undergone the legally mandated review and assessment within the prescribed time period.

[0091] A valid status means that the review assessment conclusion is passed and the most recent review assessment date is within the prescribed validity period before the current date; the review conclusion and the assessment date are used together for judgment, and the validity period cannot be ignored based solely on the "review passed" mark.

[0092] The qualification level is a qualification record read from a unique matching file, including the type of hazardous chemical medium used for subsequent comparison, the highest level of protection approved, and the type of operation; the level code is used to indicate the order of protection capabilities, and a single level code is not interpreted as containing all categories and tasks at the same time.

[0093] Furthermore, firstly, the identification information provided by the operator is obtained, which is a string recorded on the face of the professional qualification certificate card.

[0094] For an identity identifier that records both the current certificate code and the historical certificate code according to the field rules, the processing device locates the start and end positions of the two fields according to the rules and reads the two certificate codes respectively. The field rules here must be consistent with the record format of the actual input identifier, and the unrecorded historical certificate codes cannot be inferred solely from the year of issuance or the version of issuance.

[0095] After the code decomposition is completed, the current certificate code is used as the primary key index to initiate the first round of search query to the qualification database. The record entry that is completely consistent with the code is located in the certificate main table of the database. The entire electronic file corresponding to the entry is extracted and temporarily stored in the cache area as a candidate current file.

[0096] At the same time, using the historical certificate code as a secondary key index, a second round of parallel retrieval is initiated in the same database to locate and extract record entries that match the historical code from the certificate history table as candidate historical archives.

[0097] Read the field in the current candidate file that specifically stores information about the previous certificate. This field records the old certificate code that was used when the current certificate was renewed.

[0098] The old certificate code recorded in this field is compared character by character with the historical certificate code originally extracted from the identity identifier. This comparison operation produces a logical judgment result. When the two code strings are completely consistent, the reverse comparison result is marked as consistent, indicating that the input old certificate information matches the previous certificate recognized by the current file, that is, the current file correctly inherits the qualification history represented by the old certificate.

[0099] When there are any differences between the two encoded strings, the reverse comparison result is marked as inconsistent, indicating that the old certificate is not the direct predecessor of the current file, and there may be an intermediate replacement process between the two that is not recorded in the current file.

[0100] If the reverse comparison results are inconsistent, there is a pointer field in the structure of the candidate historical archive that specifically stores the information on the destination of the replacement. This field records the new certificate code generated after the old certificate is replaced.

[0101] Read the certificate replacement destination field, retrieve the next certificate file using the new certificate code, and continue the query along the certificate replacement record; if the certificate replacement destination field is empty, it only means that the record does not contain a subsequent file, and it cannot be used to determine that the certificate is still valid. The obtained end file still needs to be verified according to the review status mark and validity period conditions described later. Only after the verification is passed can it be regarded as the current valid file.

[0102] If the reverse comparison results indicate a match, there is no need to perform the above-mentioned step-by-step tracking operation. Instead, the previously extracted candidate current file is directly identified as the unique matching file.

[0103] After identifying a unique matching file, a final verification of the file's eligibility is performed; the data segment in the unique matching file used to store review assessment information is located, and the review status identifier field is read from that segment.

[0104] This field records the date and conclusion of the most recent review assessment. The assessment conclusion recorded in the review status identifier is compared with the preset valid judgment conditions. When the assessment conclusion is passed and the assessment date is within the specified validity period before the current date, the review status identifier is determined to be in an effective state.

[0105] Only after confirming that the review status indicator is valid, the hazardous chemical media category, the highest approved protection level, and the operation type corresponding to the qualification level are read from the unique matching file as input for subsequent media matching, coverage comparison, and operation task association.

[0106] In this embodiment, the qualification level is matched with the identity identifier in the qualification database, and the qualification level is compared with the target protection level to obtain the permitted scope of operation, including:

[0107] The inclusion comparison between the hazardous chemical media category in the qualification level and the leakage media type in the target protection level is performed to obtain the media matching result;

[0108] When the media matching result is valid, the highest protection level in the qualification level is compared with the lowest protection level in the target protection level in order of protection capability from strong to weak to obtain the strength matching result.

[0109] Based on the intensity matching results and the leakage concentration level of the area category, the concentration is divided into zones to obtain candidate operation ranges. The task project number of each spatial sub-area within the candidate operation range is matched with the corresponding number of the operation type in the qualification level. The boundary coordinates of the successfully matched spatial sub-area and the operation tasks are associated to obtain the permitted operation range.

[0110] Specifically, the category of hazardous chemical media is the classification of hazardous chemicals that the operator is authorized to handle, as recorded in the operator's qualification certificate. Examples include categories of toxic gases, corrosive liquids, flammable liquids, or oxidizing substances.

[0111] The type of leaked medium refers to the physical state and hazard category of the substance actually leaked in the hazardous chemical leak area. The physical state category includes gaseous leaks, liquid leaks, and solid leaks, while the hazard category includes toxic substances, corrosive substances, flammable substances, and explosive substances.

[0112] The inclusion comparison involves comparing the range of hazardous chemical types covered by the hazardous chemical media category with the actual types of substances leaked corresponding to the leaked media type, one by one, to determine whether the hazardous chemical media category fully covers all the types of hazards involved in the leaked media type.

[0113] The media matching result includes the judgment conclusion drawn after the comparison is completed. This conclusion is used to indicate whether the types of hazardous chemicals covered by the operator's qualifications are sufficient to cover the types of leaked media actually present in the current leak area.

[0114] A valid media matching result means that the comparison conclusion is that the hazardous chemical media category fully covers all hazard types involved in the leaked media type, that is, the operator's qualification scope includes all hazard categories of the current leaked media.

[0115] The highest protection level is the highest level of protective equipment that the operator is authorized to wear, as recorded in the operator's qualification level. This level reflects the strongest level of protection that the operator can safely use after training and assessment.

[0116] The minimum protection level is the minimum level of protective equipment required for a hazardous chemical spill area within the target protection level. This level is the protection threshold that must be met to enter the area for operations.

[0117] The coverage comparison is performed in descending order of protection capability to determine whether the highest protection level allowed by the qualification level can cover the lowest protection level required by the target protection level. This embodiment stipulates that Level 1 protection is higher than Level 2 protection, and Level 2 protection is higher than Level 3 protection. The serial numbers in the level names cannot be compared directly in the way that the larger the value, the stronger the protection capability.

[0118] The strength matching result is a judgment conclusion drawn after the coverage comparison is completed. This conclusion is used to indicate whether the highest level of protection authorized to the operator is sufficient to meet the minimum protection requirements for entering the current leak area.

[0119] The strength matching result is a judgment conclusion drawn after the coverage comparison is completed. When the highest protection level is not lower than the lowest protection level, the strength matching result is valid.

[0120] Area category is the classification of hazardous chemical spill areas based on the hazardous characteristics and diffusion range of the spilled substances. This category includes high-hazard areas, medium-hazard areas, and low-hazard areas.

[0121] Leakage concentration level is a classification of the concentration level of the leaked substance in the air at different spatial locations within a hazardous chemical leak area. This level is used to characterize the differences in the degree of danger at different locations within the area.

[0122] Concentration zoning is the operation of distinguishing spatial sub-regions based on the concentration distribution provided by on-site monitoring equipment; intensity matching results are used to determine whether the corresponding sub-region can be included in the candidate operation range of the operator, and are not used to change the acquired on-site concentration, nor do they indicate whether the operator can tolerate a certain concentration of hazardous chemicals.

[0123] The candidate work area is a set of one or more spatial sub-regions that the operator can enter under the current protection capabilities, obtained after the concentration zoning is completed.

[0124] The job type is the specific operational task that the operator is authorized to perform under the qualification level, as recorded in the operator's qualification level. This type includes leak sealing operations, tank transfer operations, recovery operations, neutralization and covering operations, and dilution operations, etc.

[0125] Consistent association involves matching the actual operational tasks to be performed in each spatial sub-region of the candidate operation scope with the operation type one by one to determine whether the operator's operation type matches the operational task requirements in that spatial sub-region.

[0126] The permitted scope of work is the sub-area of ​​space that the operator is allowed to enter and perform operations under the current qualifications and protection capabilities, which is finally determined after the consistency association is completed, and the sum of the types of operational tasks that are allowed to be performed within that area.

[0127] Furthermore, firstly, the operator located in the safe zone outside the hazardous chemical spill area reads the hazardous chemical media category record in their qualification file through a mobile display terminal they carry. This record lists all the hazardous chemical types and their corresponding hazard attribute groups that the operator is authorized to handle in the form of a text list.

[0128] Meanwhile, the mobile display terminal obtains real-time identification results of the leaked medium type from the on-site monitoring equipment in the hazardous chemical leak area. The real-time identification results include the physical properties of the leaked substance at normal temperature and pressure and the hazard category label to which the substance belongs.

[0129] The operator manually triggers the comparison button on the mobile display terminal. The mobile display terminal compares each authorized hazardous chemical type in the hazardous chemical medium category record with the actual substance type corresponding to the leaked medium type. During the comparison process, the physical properties of the actual leaked substance are matched with the physical description in the authorization list. At the same time, the hazard category label of the actual leaked substance is matched with the hazard attribute group in the authorization list. The mobile display terminal outputs a valid medium matching signal only when the authorization list simultaneously covers the physical properties of the actual leaked substance and all hazard category labels.

[0130] After the mobile display terminal outputs a valid medium matching signal, the operator continues to read the highest protection level value recorded in their qualification level through the mobile display terminal. This value comes from the upper limit protection capability level code indicated on the operator's qualification certificate for the highest level of protective equipment.

[0131] At the same time, the mobile display terminal retrieves the lowest protection level value from the safety management database of the hazardous chemical leak area. This value is the lowest level code of protective equipment that must be worn to enter the area, calculated based on the toxicity threshold of the leaked medium type and the danger boundary of the leak concentration level.

[0132] The mobile display terminal performs coverage comparison in descending order of protection capabilities corresponding to Level 1, Level 2, and Level 3 protection. When the highest protection level authorized for use by the operator reaches the lowest protection level required by the target area, a valid coverage comparison signal is output; otherwise, no valid coverage comparison result is formed.

[0133] After the mobile display terminal outputs a valid coverage comparison signal, the operator can view the concentration distribution map of the hazardous chemical leak area through the mobile display terminal. This concentration distribution map is drawn by multiple gas concentration detectors deployed in the leak area, which collect the concentration values ​​of the leaked substances at each detection point in real time. The map is marked with different shades of color to indicate the concentration gradient zones that gradually decrease from the center of the leak source to the periphery.

[0134] The operator manually draws a dividing line on the concentration distribution map, starting from the center of the leak source and extending along the direction of decreasing concentration. The mobile display terminal divides the leak area into three spatial sub-regions based on this dividing line: an inner high-concentration zone, a middle medium-concentration zone, and an outer low-concentration zone.

[0135] The mobile display terminal compares the protection requirements of each spatial sub-area with the protection level that the operator is authorized to use, retains the sub-areas that meet the coverage conditions, and generates a candidate work range list. The list records the boundary position and area of ​​the retained sub-areas in coordinate ranges. It is not possible to simply remove the inner circle and retain the remaining sub-areas without comparing the actual protection requirements.

[0136] Operators can access the list of job types recorded in their qualification level via a mobile display terminal. This list records the specific job tasks that the operator is authorized to perform, in the form of operation item number and corresponding job name. These tasks include leak sealing, tank transfer, recovery, neutralization and covering, and dilution.

[0137] At the same time, the mobile display terminal reads the name of the actual operation task to be completed in each spatial sub-region from the task execution list corresponding to each spatial sub-region in the candidate job range list.

[0138] Operators select each spatial sub-region from the candidate job range list one by one, and match the task execution list of that sub-region displayed on the mobile display terminal with the job type list in their own qualification level item by item. During the matching, the task project number is used as the unique associated key value for one-to-one correspondence.

[0139] After completing the matching and comparison of all sub-regions, the mobile display terminal combines the successfully matched sub-regions and their corresponding executable task names to generate a permitted work scope. This permitted work scope is output in tabular form, with each row of the table recording the boundary coordinates of a spatial sub-region and a list of job task names that the operator is allowed to perform within that sub-region.

[0140] S3: Extract motion feature signals and clothing fold dynamic features from dynamic images, match the motion feature signals with motion feature samples in the qualification database to obtain the identity verification result, and when the identity verification result is valid, identify the protective equipment logo in the dynamic image to obtain the wearing protection level.

[0141] In this embodiment, motion feature signals and clothing fold dynamic features are extracted from the dynamic image. The motion feature signals are matched with motion feature samples in the qualification database to obtain the identity verification result. When the identity verification result is valid, the protective equipment identifier in the dynamic image is identified to obtain the wearing protection level, including:

[0142] A stage sequence is formed based on the changing time sequence of motion feature signals and dynamic features of clothing folds. Continuous motion segments that are consistent with the sequence of motion feature samples are retained to obtain sequentially consistent segments.

[0143] From the sequentially consistent segments, select the spatially matching segments whose joint bending direction matches the crease change, and compare the crease extension direction of the spatially matching segments with the corresponding moment of the motion feature sample to obtain the equipment linkage segment;

[0144] The coverage of the equipment linkage section is checked in stages. The ratio of the total duration of the equipment linkage section to the total duration of all action stages is used as the coverage level. The identity verification result is generated based on the comparison result of the coverage level and the preset qualified threshold. When the identity verification result is valid, the protective equipment logo is identified to obtain the wearing protection level.

[0145] Specifically, motion feature signals are digital time-series data that describe the changes in the position and posture of an operator's limb joints in space over time when the operator completes a preset standard motion sequence.

[0146] The temporal relationship refers to the sequential order and logical interval of different action stages in the action feature signal on the time axis.

[0147] The dynamic characteristics of clothing folds are the dynamic trajectory data of the position movement, extension direction change and shape change of the fold area on the surface of protective equipment caused by limb bending, as presented in continuous image frames.

[0148] The change sequence is a record of the occurrence time and duration of each dynamic change event in the dynamic characteristics of clothing folds on the time axis.

[0149] The action phase is the smallest action unit with a single movement trend obtained by dividing a complete standard action sequence according to the change of limb movement direction or the extreme point of speed.

[0150] A phase sequence is an ordered set formed by arranging multiple action phases after segmentation in chronological order.

[0151] The qualification database is a pre-stored digital collection containing the identity information of certified operators and their corresponding standard action feature samples.

[0152] The motion feature sample is a standardized temporal template data recorded in the qualification database, showing the changes in the angles of the limb joints of a certified operator over time when performing a standard motion sequence. The motion feature sample in the qualification database is established in the following way: when an operator participates in a qualification certification assessment, he / she is required to wear standard protective equipment and complete a set of preset standard motion sequences in front of an image acquisition device; the system automatically extracts the motion feature signals and clothing fold dynamic features during the process, associates them with the operator's identity identifier, and stores them in the qualification database as motion feature samples.

[0153] A continuous action segment is a combination of one or more adjacent action stages that are extracted from a stage sequence and remain continuous on the timeline without interruption.

[0154] A sequence-consistent segment is a part of a continuous action sequence in which the order of the action stages is exactly the same as the order of the action stages recorded in the action feature sample.

[0155] The moment of change of direction is the critical point in time when an operator's limb joint changes from extension to flexion or from flexion to extension during the execution of an action.

[0156] The direction of joint flexion is the direction of the limb joint's movement relative to the previous moment at the moment of change of direction, specifically characterized by the flexion or extension of the joint.

[0157] The crease extension direction is the main direction of the wrinkles formed on the surface of the protective equipment on the image plane, and this direction changes as the limb joints bend.

[0158] A spatially congruent segment is a sequence segment in which the trend of change in the direction of joint bending during that time period is synchronized with the trend of change in the direction of crease extension at the corresponding moment.

[0159] The corresponding time is a point in time within the spatial alignment segment, which has the same time offset as the time point in the action feature sample.

[0160] The equipment linkage segment is the part of the spatial matching segment where the direction of crease extension during that time period is the same as the direction of crease extension at the corresponding moment recorded in the motion feature sample.

[0161] Determine whether the equipment linkage segment fully covers each action stage of the action characteristic signal on the time axis to obtain the coverage degree;

[0162] The time axis is a coordinate reference used to represent the continuous time process of an action feature signal from start to finish.

[0163] Coverage degree is the proportion of the total duration of the equipment linkage segment on the time axis to the total duration of all action stages of the action characteristic signal; the specific calculation process of the coverage degree includes:

[0164] First, arrange all equipment linkage segments according to their start times. For overlapping time intervals, count them only once. Subtract the start time from the end time of each non-overlapping interval to obtain the duration of each interval. Add up the durations of each interval to obtain the total equipment linkage duration.

[0165] Then, the duration of all action stages in the action feature signal is statistically calculated using the same time unit to obtain the total action duration; when the action stage continuously covers the acquisition interval from the start frame to the end frame, this duration is equal to the difference between the end time and the start time.

[0166] Divide the total duration of equipment linkage by the total duration of action to obtain the dimensionless coverage. This division is only performed when the total duration of action is greater than zero and the equipment linkage segment belongs to the action phase being counted. If there is a lack of effective actions, the result of valid identity cannot be formed by this ratio.

[0167] The identity verification result is generated based on whether the coverage meets the preset standard, and is used to indicate whether the operator currently wearing protective equipment is the person corresponding to the identity identifier in the qualification database, and is either affirmative or negative.

[0168] Furthermore, firstly, a continuous data stream of motion feature signals is acquired, which records the spatial angle values ​​of the shoulder, elbow, wrist, hip, knee, and ankle joints in each frame of an image during a set of standard movements performed by the operator after wearing protective equipment, including raising both arms, squatting, and turning left and right.

[0169] Simultaneously, dynamic features of clothing folds are acquired. These features record the position movement trajectory and extension direction change curve of the folds on the surface of the protective equipment located at curved parts such as the elbow, armpit, knee, and groin during the above-mentioned actions on continuous image frames.

[0170] Subsequently, based on the inflection point when the joint angle changes from increasing to decreasing or from decreasing to increasing in the motion characteristic signal, and combined with the corresponding moments when the fold pattern in the clothing fold dynamic characteristics shows stagnation, reversal, or acceleration, the entire motion signal is divided into multiple continuous and non-overlapping segments from beginning to end. Each segment corresponds to a complete limb extension or bending movement. All segments are then arranged in chronological order to form a stage sequence.

[0171] Starting from the beginning of the phase sequence, one segment is taken out in sequence, and the action type corresponding to the segment is compared with the first action segment in the historical action sample of the operator stored in the qualification database. If the action types are the same, the comparison continues to the next segment until a segment does not match the segment type at the corresponding position in the sample or the entire phase sequence has been compared. At this point, multiple consecutive segments that have been successfully matched in the phase sequence without interruption are extracted to form a consistent sequence segment.

[0172] For each action segment in the sequentially consistent segment, locate the instantaneous change direction of the joint angle at the beginning of the segment, such as whether the elbow joint is extending or flexing. At the same time, observe whether the creases on the surface of the protective equipment located on the outside of the joint converge toward the center of the joint or spread to both sides of the joint. If the direction of extension or flexion of the joint and the direction of convergence or spread of the creases show spatial consistency, mark the action segment as a spatially consistent segment. In this way, traverse all segments in the sequentially consistent segment and filter out all segments that meet the spatial consistency condition.

[0173] For each spatial matching segment, the extension angle of the crease in each frame of the image within the segment is numerically compared with the extension angle of the crease recorded at the same time position in the motion feature sample. If the angle difference between the two is within the allowable range, the frame is marked as a consistent frame, and consecutive consistent frames are connected into equipment linkage segments. Each spatial matching segment may contain one or more equipment linkage segments.

[0174] According to the aforementioned statistical method of removing duplicate time intervals, the total duration of equipment linkage and the total duration of all action phases are calculated. When the total duration of action is greater than zero, the ratio between the two is calculated to obtain the coverage. Since the numerator and denominator use the same time unit and the numerator interval belongs to the denominator interval, this ratio represents the proportion of time covered by the equipment linkage segment.

[0175] The coverage value is compared with a pre-defined qualification threshold. When the coverage reaches or exceeds the qualification threshold, it is determined that the currently collected motion feature signal and the motion feature sample recorded in the qualification database originate from the same operator, and a valid identity verification result is generated.

[0176] When the coverage level is below the acceptable threshold, it is determined that the two are not from the same person, and an invalid identity verification result is generated. At this time, the operator is refused entry into the work area and prompted to re-verify their identity.

[0177] In this embodiment, motion feature signals and clothing fold dynamic features are extracted from the dynamic image. The motion feature signals are matched with motion feature samples in the qualification database to obtain the identity verification result. When the identity verification result is valid, the protective equipment identifier in the dynamic image is identified to obtain the wearing protection level, including:

[0178] When the identity verification result is valid, the micro-deformation region and the deformation-stable region are distinguished according to the dynamic characteristics of clothing folds. The candidate regions for identification that have abrupt changes in reflected light intensity and form closed edges are extracted from the deformation-stable regions.

[0179] Compare the edge orientation of the candidate identification area with the crease extension direction of the micro-deformation area, and retain intersecting or independent valid identification edges;

[0180] The pattern enclosed by the effective marking edge and the arrangement of the internal color blocks are compared with the pre-stored marking sample. The protection level marking information of the matching sample is read, and the wearing protection level is determined according to the protection level code in the protection level marking information.

[0181] Specifically, the surface of protective equipment refers to the outer surface of the protective equipment that is directly exposed to the field of view of the image acquisition device.

[0182] Microscopic deformation areas are localized regions on the surface of protective equipment where the position or direction of creases changes between adjacent image frames when an operator performs a specified sequence of actions. These changes are limited to surface changes that are actually recognizable in the image, and the physical depth of the crease is not obtained directly from ordinary images.

[0183] The deformation-stable region is a local surface area on the protective equipment where the position and direction of the crease remain basically unchanged between adjacent frames. The creases in this region exhibit a relatively fixed spatial distribution pattern in a continuous image sequence.

[0184] The determination is made by comparing the spatial position offset and the change in the extension direction of the creases on the surface of the protective equipment between adjacent frames in the dynamic image. Local areas where the spatial position offset of the crease exceeds the offset threshold are classified as micro-deformation areas, and local areas where the spatial position offset of the crease does not exceed the offset threshold are classified as deformation-stable areas.

[0185] Reflected light intensity is the illumination value recorded at each pixel position on the surface of protective equipment in a dynamic image. This value depends on the reflection characteristics of the protective equipment surface material to incident light from the environment.

[0186] Spatial difference is the numerical difference distribution of reflected light intensity between different pixel positions on the surface of protective equipment in a dynamic image.

[0187] The edge of a sudden change in reflected light intensity is the line connecting pixel positions where the reflected light intensity value changes abruptly when traveling in a certain direction within a deformation-stable region, and the difference in reflected light intensity on both sides of this line exceeds the intensity change threshold.

[0188] Extraction involves scanning the reflected light intensity values ​​pixel by pixel along the horizontal and vertical directions within the image range of the deformation-stable region. Pixels whose reflected light intensity values ​​change rate exceeds a threshold are marked, and adjacent marked pixels are connected into continuous lines to obtain the edges of abrupt changes in reflected light intensity.

[0189] A closed region is a closed image region formed by multiple edges with abrupt changes in reflected light intensity connected end to end. There is an overall intensity difference between the internal and external reflected light intensity distributions of this closed region.

[0190] The candidate region for identification is a candidate image area on the surface of protective equipment that may carry information about the protection level identification, and is defined by a region with closed edges.

[0191] Edge direction refers to the angle of extension of each edge within the candidate region in the image plane.

[0192] The crease extension direction is the direction in which the crease pattern extends within the image plane in the micro-deformation region. This direction changes in different image frames as the operator performs a specified action.

[0193] The comparison involves checking the directional consistency of the extension angle of each edge in the candidate region with the extension angle of the crease in the micro-deformation region, and determining whether the included angle is less than the directional consistency threshold.

[0194] The edge that intersects with the crease extension direction is the edge where the angle between the extension direction angle and the crease extension direction angle in the micro-deformation region is greater than the direction consistency threshold and less than the vertical angle threshold.

[0195] An independent edge is an edge whose angle between the extension direction angle and the crease extension direction angle in the micro-deformation region is greater than or equal to the vertical angle threshold. The direction of this edge is not parallel or approximately parallel to any of the crease extension directions.

[0196] Effective identifying edges are those edges selected from all edges in the candidate identifying region that do not align with the crease extension direction in the micro-deformation region. Specifically, they include edges that intersect with the crease extension direction and independent edges.

[0197] The pattern enclosed by the effective marking edges is a graphic outline formed by connecting the effective marking edges within the marking candidate area. This outline defines the shape boundary of the protection level marking in the image.

[0198] The color blocks inside the pattern are several connected regions divided by different reflected light intensity values ​​within the pattern boundary formed by the effective identification edge. The reflected light intensity value within each connected region remains consistent or changes gradually, while there are abrupt changes in reflected light intensity between adjacent connected regions.

[0199] The color block arrangement order is the spatial distribution order of each color block within the pattern boundary, including the relative positional relationship of each color block in the horizontal and vertical directions.

[0200] Identification is based on the shape characteristics of the pattern enclosed by the edge of the valid mark and the arrangement order of the color blocks inside the pattern. It compares the shape and color block arrangement with each mark sample in the pre-stored protection level mark pattern library. When the shape and color block arrangement of the pattern are consistent with the shape and color block arrangement of a certain mark sample, the protection level mark information corresponding to the mark sample is determined as the protection level mark information of the protective equipment.

[0201] Protection rating information is a combination of text, numbers, symbols, or graphics printed on the surface of protective equipment to indicate the level of protection that the equipment can provide.

[0202] The protection level code is a level code recorded in the protection level identification information in the form of numbers, letters or a combination of numbers and letters. This code corresponds to a unique protection level according to the grading rules in the national standard for protective equipment.

[0203] The process involves extracting the protection level code from the protection level identification information and mapping the code to the protection equipment level classification standard to obtain the corresponding protection level value.

[0204] The protection level is the numerical value of the level of protection that the protective equipment currently worn by the operator can provide when it is in good working order and meets the standards. It is obtained by mapping the protection level code.

[0205] The operator wears protective gear and stands facing the image acquisition device, which continuously captures images of the operator in chronological order to obtain dynamic images including the surface of the protective gear.

[0206] During the filming process, the operator performed the aforementioned sequence of actions, including raising both arms, squatting, and turning left and right, to record the changes in the surface wrinkles of the protective equipment in consecutive image frames.

[0207] Read the dynamic image and track the spatial position of the creases on the surface of the protective equipment in adjacent frames. Measure the positional offset of each crease between adjacent frames and the angle of change of the crease extension direction. Mark the local area where the positional offset exceeds the set offset threshold as the micro-deformation area, and mark the local area where the positional offset does not exceed the offset threshold as the deformation stable area.

[0208] After the region is divided, the reflected light intensity values ​​of each pixel position are scanned pixel by pixel along the horizontal and vertical directions within the image range of the deformation-stable region. The rate of change of reflected light intensity between adjacent pixels is calculated in real time. Pixel positions where the rate of change exceeds the rate of change threshold are marked, and spatially adjacent marked pixels are connected sequentially to form continuous lines. These lines are the edges of abrupt changes in reflected light intensity.

[0209] Then, find the combination of edges that form a closed region from these edges, and record each such closed region as a candidate region for identification.

[0210] For each candidate region, the extension direction angle of each edge in the region is measured one by one, and the extension direction angle of each crease in the micro-deformation region is also measured. The extension direction angle of each edge is compared with the extension direction angle of each crease in the micro-deformation region one by one, and the included angle between the two is calculated.

[0211] When the included angle is greater than the direction consistency threshold and less than the vertical angle threshold, the edge is determined to be an edge intersecting the crease extension direction. When the included angle is greater than or equal to the vertical angle threshold, the edge is determined to be an independent edge. The intersecting edges and independent edges are selected together as valid identification edges, while edges with included angles less than or equal to the direction consistency threshold are removed.

[0212] The selected valid identifier edges are reconnected within the identifier candidate area, forming a new graphic contour. Multiple connected color blocks divided by abrupt changes in reflected light intensity within the graphic contour are extracted, and the spatial arrangement order of each color block is recorded sequentially in the horizontal and vertical directions.

[0213] The shape features of the graphic outline and the arrangement of the internal color blocks are matched one by one with the shape outline and color block sequence of each logo sample in the pre-stored protection level logo pattern library. When a logo sample with the same shape and color block arrangement is found, the protection level logo information corresponding to the logo sample is read. This information contains the protection level code recorded in the form of numbers, letters or a combination of numbers and letters.

[0214] The protection level code is extracted from the protection level identification information. Based on the level classification mapping relationship specified in the national standard for protective equipment, the code is converted into the corresponding protection level value, which is the wearing protection level of the protective equipment currently worn by the operator.

[0215] S4: Identify the fit of protective equipment based on the dynamic characteristics of clothing folds, generate a measured value of the protection level by combining the wearing protection level, match the measured value of the protection level with the permitted work range, and obtain the matching result of the protective equipment.

[0216] In this embodiment, the fit of protective equipment is identified based on the dynamic characteristics of clothing folds. Combined with the wear protection level, a measured protection level value is generated. This measured protection level value is then matched with the permitted work area to obtain the matching result of the protective equipment, including:

[0217] By tracking the creases in the dynamic features of clothing folds frame by frame, the distribution of crease bands is obtained, and the sliding crease bands that have undergone positional migration and the stationary crease bands that have fixed positions are distinguished.

[0218] The convergence node is identified based on the adjacency relationship between the slip crease zone and the dwell crease zone, and the directional concentration area is identified according to the direction of the creases around the convergence node;

[0219] The closed-bound areas within the directional concentration region are designated as the fit-support zone, and the open-bound areas are designated as the fit-detachment zone. By combining the spatial envelopment relationship, the dynamic contact state between the protective equipment and the operator's body is analyzed to obtain the fit degree. Based on the fit degree, the wearing protection level is corrected to form the measured protection level value. This value is then compared with the lower limit of the protection level for the permitted work area to obtain the matching result of the protective equipment.

[0220] Specifically, translation trajectory discrimination is the process of tracking the spatial position of each crease area on the surface of the protective equipment frame by frame in the image frames acquired at different times, and determining whether each crease has moved as a whole on the surface of the protective equipment.

[0221] Crease distribution refers to the spatial layout of creases formed by marking the location and direction of each crease on the surface of protective equipment.

[0222] A continuous translation trajectory crease is a crease band in a continuous frame sequence of a dynamic image, in which the spatial position of the crease undergoes continuous translation over time and the trajectory of the motion is a continuous curve.

[0223] A slip crease is a crease on the surface of protective equipment that experiences relative sliding displacement due to the bending or stretching of limb joints within a continuous translational trajectory crease.

[0224] A stationary crease is a crease that remains fixed relative to the surface of protective equipment in a continuous frame sequence of dynamic images and does not undergo significant positional shift with limb movements.

[0225] A convergence node is a spatial point where the end of a slip crease intersects with one or more dwell creases.

[0226] The change in orientation refers to the change in the angle of deflection of the spatial extension direction of each crease near the convergence node relative to the spatial extension direction of the crease away from the convergence node.

[0227] A directional concentration area is a localized area on the surface of protective equipment where multiple creases converge in the same spatial direction.

[0228] The fit and support area is a continuous region on the surface of protective equipment where the creases are distributed in a closed loop and have closed boundaries. This region represents the part of the protective equipment that maintains contact with the operator's body surface and bears the body's support force.

[0229] The contact separation zone is an area on the surface of protective equipment where the creases are open and the boundaries are not closed. This zone represents the distribution range of the part where the protective equipment separates from the operator's body surface and forms a suspended space.

[0230] The fit is the overall spatial distribution of the contact relationship between various parts of the protective equipment surface and the operator's body surface, including the distribution of the fit support area and the distribution of the fit separation area.

[0231] The follow-up contact state refers to the dynamic relationship in which the inner surface of the protective equipment moves synchronously with the movement of the operator's body surface and maintains contact during the operation of the protective equipment.

[0232] In this embodiment, the degree of fit is calculated based on the dimensionless ratio of the pixel area of ​​the visible area of ​​the protective equipment and the area of ​​fit-detachment within the same image. This ratio is used to describe the area proportion of the non-detachment area in the image. This two-dimensional pixel area ratio is not equivalent to the ratio of the actual contact area of ​​the inner surface of the equipment to the total inner surface area. The specific calculation is as follows.

[0233] First, count the number of different pixels in the same image that are classified as the non-fitting area. Pixels in the overlapping area are counted only once to obtain the total area of ​​the non-fitting area.

[0234] Then, the number of different pixels in the visible area of ​​the protective equipment in the same image is counted to obtain the total area of ​​the equipment. Both the total area of ​​the detachment zone and the total area of ​​the equipment are represented by pixel count.

[0235] When the total area of ​​the equipment is greater than zero and the decoupling area is located within the statistical equipment area, the total area of ​​the decoupling area is divided by the total area of ​​the equipment to obtain the decoupling area ratio; if no effective equipment area is detected, this division is not performed, and no conclusion of good fit is output based on this.

[0236] Finally, subtract the percentage of the area that is detached from the result by one to obtain the degree of fit.

[0237] Furthermore, after acquiring dynamic images of operators performing standard action sequences, the location and extension direction of the crease area on the surface of the protective equipment in each image frame are extracted from the dynamic images, and the spatial position of each crease area between adjacent frames is tracked and compared frame by frame according to the acquisition time sequence.

[0238] When the position coordinates of a certain crease area are continuously translated in multiple consecutive frames of images and the translation trajectory shows a continuous curve shape, the crease area is marked as a continuous translation trajectory crease. From the continuous translation trajectory creases, creases that have relative sliding displacement on the surface of the protective equipment due to the bending or extension of limb joints are selected as slip creases. At the same time, creases that maintain a fixed position relative to the surface of the protective equipment in the continuous frame sequence of dynamic images are marked as stationary creases.

[0239] Tracing the sliding crease band along its extension direction to its end, when a point is identified at the end of the sliding crease band that intersects with the spatial position of one or more stationary crease bands, this point is marked as a convergence node; after identifying each convergence node, with each convergence node as the center, the deflection angle of the extension direction of each crease band near the convergence node position relative to the extension direction of the crease band away from the convergence node position is measured, and the change value of each deflection angle is recorded.

[0240] Then, the extension direction of each crease band is scanned on the surface of the protective equipment according to its spatial position. When multiple crease bands in a certain local area are found to converge in the same spatial direction, that local area is marked as the direction concentration area.

[0241] After marking all directional concentrated areas, the boundary shape of each directional concentrated area is identified. Continuous areas with closed loops and closed boundaries are divided into fit-support areas, and areas with open folds and unclosed boundaries are divided into fit-separation areas. The spatial distribution of all fit-support areas and fit-separation areas is sorted out to generate the fit state between the protective equipment and the operator's body.

[0242] Finally, the spatial positional relationship between each bonding separation zone and the adjacent bonding support zone is tracked to determine whether each bonding separation zone partially or completely surrounds the corresponding bonding support zone from the outside. Combined with whether the surrounded bonding support zone moves synchronously with the movement of the operator's body surface during the operation of the operator's limbs, the degree of fit between the protective equipment and the operator's body is comprehensively determined.

[0243] In this embodiment, the fit of protective equipment is identified based on the dynamic characteristics of clothing folds. Combined with the wear protection level, a measured protection level value is generated. This measured protection level value is then matched with the permitted work area to obtain the matching result of the protective equipment, including:

[0244] Continuous crease regions with fixed positions are selected from the dynamic features of clothing folds, and the cross-sections where the continuous crease regions intersect with the edges of protective equipment are extracted.

[0245] Based on the situation where the edge of the protective equipment is divided into two parts by the cross section and the connection between the two parts, the gap connection state is obtained. The wearing protection level in the gap connection state is downgraded according to the degree of fit, and the actual protection level value is obtained.

[0246] The measured value of the protection level is compared with the lower limit of the protection level in the permitted operation range in order of protection capability. If the lower limit of the protection level is reached, the matching result is considered passed; otherwise, it is considered failed, thus obtaining the matching result of the protective equipment.

[0247] Specifically, time-series filtering involves determining the spatial location of each crease region at different times from multiple consecutive frames of images, based on the temporal order of dynamic image acquisition, and filtering out crease regions that continuously appear on the timeline.

[0248] A persistent and fixed crease region refers to a crease region that can be identified from the first frame to the last frame of a dynamic image in a time series, without disappearing at any intermediate moment, and whose spatial coordinates do not change significantly over time. This region corresponds to the part where the protective equipment is in close contact with the operator's body, and there is no relative sliding between the surface of the protective equipment and the surface of the body below.

[0249] A persistent crease region is a set of crease regions selected from dynamic images that simultaneously meet the conditions of persistence and fixed position. Each crease region in this set indicates that there is no relative slippage between the protective equipment and the body in that local area.

[0250] The edge of protective equipment is the outline boundary line of the protective equipment in the image. This boundary line corresponds to the opening part of the protective equipment or the seam between different parts. It is a key reference line for judging whether the protective equipment completely covers the operator's body.

[0251] A crossing section is the section where the continuous crease area intersects with the edge of the protective equipment in the direction of spatial extension; that is, the continuous line segment where the boundary line of the continuous crease area crosses the edge line of the protective equipment.

[0252] The segmented state refers to the impact on the integrity and continuity of the edge of the protective equipment after the crossing section intersects with it. It manifests as the spatial distribution pattern in which the edge of the protective equipment is divided into several independent line segments by the crossing section.

[0253] The gap continuity state refers to the distance between the two parts when the edge of the protective equipment is divided into two parts by a cross section, and whether the gap forms a continuous and continuous gap channel in space, which connects the inner space and the outer space of the protective equipment.

[0254] The protection rating is the rated protection level of the protective equipment as indicated by the markings on its surface. It represents the nominal protection capability of the equipment in its intact condition.

[0255] Rating correction is an operation that adjusts the nominal value of the protective equipment's protection level based on the reduction in protective capability caused by poor fit or gaps in the current wearing state.

[0256] The specific process of the downgrade process is as follows:

[0257] First, arrange Level 1 protection, Level 2 protection, and Level 3 protection in descending order of protective capability to determine the position of the wearable protection level within this sequence.

[0258] Then, the position is moved one level in the direction of weaker protection to obtain the corrected level position.

[0259] If the protection level is already at level 3, the corrected level position will still be the position corresponding to level 3 protection.

[0260] Finally, the protection level corresponding to the corrected protection level position is taken as the measured value of the protection level.

[0261] The measured protection level is the effective protection level value of the protective equipment under the current actual wearing condition, obtained after level correction. This value comprehensively considers the influence of the equipment's nominal protection capability and the wearing quality on the protection effect.

[0262] The lower limit of the protection level is the minimum effective protection level required for the permitted scope of work. The effective protection level of the protective equipment actually worn by the operator must not be lower than this lower limit before entering the work area.

[0263] Level comparison is an operation that compares the measured value of the protection level with the position of the lower limit of the protection level in the protection level sequence to determine whether the measured value meets or exceeds the lower limit requirement.

[0264] The matching result is the output conclusion of the level comparison, indicating whether the effective protection level of the protective equipment in the current wearing state meets the protection level requirements of the permitted work area.

[0265] Furthermore, based on the dynamic characteristics of the garment folds, the spatial position of each fold area is identified frame by frame along the time axis of the dynamic image. Fold areas that can be stably identified from the start frame to the end frame of the dynamic image and whose spatial coordinates have not changed significantly are selected and grouped into continuous fold areas. These areas reflect the parts where the protective equipment fits tightly against the body without relative slippage.

[0266] After identifying the continuous crease regions, the boundary direction of each continuous crease region and the edge contour line of the protective equipment are tracked in the image space. The spatial intersection detection of the boundary line of each continuous crease region and the edge contour line of the protective equipment is performed, and the continuous line segment that intersects with the boundary line is extracted as the crossing segment.

[0267] After obtaining the crossing section, observe the impact of the crossing section on the integrity of the protective equipment edge, and determine whether the protective equipment edge is divided into one segment or multiple segments by the crossing section. If it is divided into two independent line segments, further measure the distance between these two line segments and determine whether the interval forms a continuous and through gap channel in space to obtain the gap through state.

[0268] The protection level is adjusted based on the gap penetration status and fit: when the gap penetration status indicates that the edge of the protective equipment is divided into two parts and there is a continuous gap channel between the two parts, the protection level is adjusted down one level in the protection level sequence, that is, the nominal protection level is replaced with the next lower level of protection level, and the actual protection level value is obtained.

[0269] The measured value of the protection level is compared with the lower limit of the protection level specified in the permitted work area within the protection level sequence: if the measured value of the protection level is not lower than the lower limit of the protection level in the protection level sequence, the matching result is deemed successful, indicating that the effective protection capability of the protective equipment currently worn by the operator, after considering the fit quality, meets the minimum requirements for entering the work area; conversely, if the measured value of the protection level is lower than the lower limit of the protection level, the matching result is deemed unsuccessful, indicating that the effective protection capability of the protective equipment is insufficient, and the operator is not allowed to enter the work area in the current wearing state.

[0270] S5: Extract the distribution of the continuity between the facial boundary and the edge of the protective equipment in the real-time face image to obtain the airtight fit state. Compare the airtight fit state with the airtight reference to obtain the airtight comparison result.

[0271] In this embodiment, the continuous distribution of the facial boundary and the edge of the protective equipment in the real-time face image is extracted to obtain the airtight fit state. The airtight fit state is compared with the airtight reference to obtain the airtight comparison result, including:

[0272] Samples are taken from the facial boundary toward the edge of the protective equipment to form a grayscale transition zone, and the grayscale gradient interruption section is marked as a micro-gap region;

[0273] By arranging the micro-gap regions and gray-scale gradient continuous segments sequentially along the facial boundary, an edge continuity distribution is obtained, connecting spatially adjacent micro-gap regions into potential leakage paths.

[0274] The positions at both ends of the potential leakage path are compared with the two sides of the contact area of ​​the face boundary and the edge of the protective equipment to obtain the airtight fit status indicating whether the potential leakage path penetrates the contact area. Then, the airtight fit status is compared with the allowable gap range of the corresponding face area in the airtight reference to obtain the airtight comparison result.

[0275] Specifically, the facial boundary is the visible outer contour of the operator's face in a real-time face image. This contour is defined by the pixel grayscale transition positions between the facial skin area and the external environment or other object areas, including the outer edges of facial areas such as the forehead, cheeks, and jaw.

[0276] The grayscale transition zone is a strip-shaped area located between the facial boundary and the edge of the protective equipment. The grayscale value of the pixels in this strip-shaped area gradually transitions from the grayscale value of the facial skin area to the grayscale value of the protective equipment material. This grayscale transition zone represents the spatial adjacency relationship between the facial skin surface and the protective equipment surface, as well as the degree of tightness between the two.

[0277] The grayscale gradient interruption section is the section within the grayscale transition zone where the pixel grayscale value changes abruptly along the direction perpendicular to the facial boundary. In this section, the grayscale change sequence that should have been a continuous gradient becomes a step jump. This step jump indicates that there is a non-contact gap between the facial surface and the protective equipment surface at this location.

[0278] The micro-gap region is a localized, tiny spatial separation area between the facial boundary and the edge of the protective equipment due to poor fit. This region manifests as an interrupted section in the grayscale transition zone where grayscale changes are discontinuous. The width and extension length of this micro-gap region reflect the geometric scale of airtightness failure at this local location.

[0279] The grayscale gradient continuous section is a segment within the grayscale transition zone where the pixel grayscale value changes continuously and smoothly along the direction perpendicular to the facial boundary. Within this segment, the grayscale value gradually transitions from the grayscale value of the facial skin area to the grayscale value of the protective equipment material. This continuous change indicates that the facial surface and the protective equipment surface maintain close contact at this location.

[0280] Edge continuity distribution is an ordered sequence formed by sequentially arranging each micro-gap region and each gray-level gradient continuous segment within the gray-level transition zone along the extension direction of the facial boundary. The spatial position, segment length, and category of each segment in the sequence, whether it belongs to a micro-gap region or a gray-level gradient continuous segment, together constitute the content of this continuity distribution.

[0281] Spatial adjacency refers to two micro-gap regions in a continuous edge distribution being adjacent to each other along the extension direction of the facial boundary and having no other micro-gap regions between them, or two micro-gap regions forming a continuous connected relationship in space through one or more micro-gap regions.

[0282] A potential leakage path is one or more continuous paths formed by connecting spatially adjacent micro-gap regions in the edge continuity distribution. The starting point of the continuous path is located at one side boundary of the contact area between the face boundary and the edge of the protective equipment, and the ending point is located at the other side boundary of the contact area, or both ends of the continuous path are located at the same side boundary of the contact area.

[0283] The contact area is the continuous surface area formed by the part of the respiratory protective equipment that actually comes into contact with the operator's face. This contact area extends along the boundary of the face and is the functional area where the respiratory protective equipment achieves an airtight seal. Harmful media need to break through the sealing barrier of this contact area to enter the interior of the respiratory protective equipment.

[0284] A through-path is a state in which a potential leak path extends continuously from one boundary of the contact area to the other boundary of the contact area. When a potential leak path is in a through-path state, the potential leak path passes through the entire contact area, connecting the external environment of the contact area with the internal space of the respiratory protective equipment.

[0285] Airtight fit refers to the effectiveness of the seal between the respiratory protective equipment and the operator's face within the contact area. This status includes the determination of whether there are potential leakage paths within the contact area and whether any existing potential leakage paths are open.

[0286] Furthermore, after acquiring a real-time facial image of the operator's front face, all pixels in the image are first scanned row by row and column by column. Based on the location of local abrupt changes in pixel grayscale values, the visible outer contour of the face is marked as the facial boundary, and the outline of the respiratory protective equipment is marked as the edge of the protective equipment.

[0287] The first pixel on the facial boundary is selected as the starting point. A straight line with the normal direction of the facial boundary at that point is drawn through the starting point. The line moves pixel by pixel from the starting point toward the edge of the protective equipment along the normal direction. The gray value of each pixel is recorded at each pixel position until the corresponding pixel on the edge of the protective equipment is reached. The entire pixel line segment covered by this movement process is the normal sampling line corresponding to the starting point.

[0288] After completing the normal sampling of the first starting point, move along the extension direction of the facial boundary to the next adjacent pixel as the new starting point. Repeat the above operation of drawing perpendicular lines, moving pixel by pixel and recording gray values ​​until all pixels on the facial boundary are traversed. The set of pixels covered by all normal sampling lines together constitutes the gray-scale transition zone.

[0289] For each normal sampling line, gray values ​​are read sequentially from the end closest to the face boundary. The absolute value of the difference between the gray values ​​of adjacent pixels is taken to obtain the magnitude of the adjacent gray value difference. These absolute values ​​are accumulated and divided by the number of adjacent pixel pairs to obtain the average absolute gray value difference of the sampling line. When the absolute gray value difference in a continuous segment is not greater than the average value, the segment is marked as a continuous gray value gradient segment.

[0290] When the absolute value of the difference between adjacent gray levels suddenly increases and significantly exceeds the average absolute gray level difference in a certain continuous segment, the continuous segment is determined to be a gray level gradient interruption segment, the position of the continuous pixels covered by the segment is marked, and the corresponding image area is recorded as a micro gap region.

[0291] After completing the above determination and marking for all normal sampling lines, the segment marking results on the normal sampling lines corresponding to each starting point are read sequentially from one end of the facial boundary to the other end along the extension direction of the facial boundary. Segments with the same marking category between adjacent starting points are merged into a whole segment. Then, the merged whole segments are arranged sequentially according to their positions to form an edge continuity distribution sequence from one end of the facial boundary to the other end.

[0292] In this sequence, all segments classified as micro-gap regions are identified, and it is checked whether adjacent micro-gap regions are separated only by a small number of continuous gray-scale gradient segments. If the total length of the separated continuous gray-scale gradient segments is less than the length of any of the adjacent micro-gap regions, the adjacent micro-gap regions are determined to be spatially adjacent and connected together. This process is repeated until all spatially adjacent micro-gap regions are connected. Each continuous connected body formed after connection is a potential leakage path.

[0293] The actual extension direction of each potential leakage path in the image space is compared with the spatial range of the contact area. The starting end of the potential leakage path is tracked to see if it is located at the outer boundary of the contact area and the ending end is located at the inner boundary of the contact area. If a potential leakage path satisfies both the starting end being at the outer boundary and the ending end being at the inner boundary, it is determined that the path runs through the entire contact area. At this time, the airtight fit is marked as unqualified, and the position and length of the through path are recorded as the basis for unqualification.

[0294] If, after traversing all potential leakage paths, no path is found that simultaneously crosses both boundaries of the contact area, then it is determined that none of the potential leakage paths penetrate the contact area, and the airtight fit is marked as qualified.

[0295] In this embodiment, the continuous distribution of the facial boundary and the edge of the protective equipment in the real-time face image is extracted to obtain the airtight fit state. The airtight fit state is compared with the airtight reference to obtain the airtight comparison result, including:

[0296] The edge continuity distribution corresponding to the airtight fit state of the facial area is matched with the allowable gap range of the airtight reference to obtain the area continuity distribution;

[0297] The regions where grayscale gradients are interrupted in the continuous distribution of the region are taken as micro-gap regions. The discontinuities at both ends of the micro-gap regions are extracted and connected to form discontinuous segments. The position and extension length of the discontinuous segments are obtained. The facial region, start position and end position of each discontinuous segment are recorded and arranged in the order of extension along the facial boundary.

[0298] Discontinuous segments whose extension length exceeds the allowable discontinuity range of the corresponding facial area are marked as airtightness non-compliant segments. The abnormal segment distribution is obtained by arranging them by position. When there are no airtightness non-compliant segments in the abnormal segment distribution, a qualified airtightness comparison result is generated. When there are airtightness non-compliant segments, an unqualified airtightness comparison result is generated.

[0299] Specifically, the facial region is a number of independent local areas of the operator's face divided according to the anatomical structure, including the forehead region, the bridge of the nose region, the left cheek region, the right cheek region, the jaw region, and the chin region. Each facial region has its own independent geometric contour and curvature characteristics.

[0300] The permissible gap range is a value set individually for each facial region in the airtightness standard. It represents the maximum gap width and maximum gap extension length that are allowed to exist between the edge of the respiratory protective equipment and the facial boundary in that region. The permissible gap ranges are different for different facial regions.

[0301] Spatial correspondence is the process of assigning the continuous distribution of edges detected along the entire facial boundary in an airtight fit state to the corresponding facial regions pre-divided in the airtight reference according to the spatial location of each detection point.

[0302] Regional continuity distribution is a subset of edge continuity distributions obtained for each facial region after the spatial correspondence operation is completed. Each facial region corresponds to an independent regional continuity distribution.

[0303] A discontinuity point is the boundary between a microscopic gap region and a continuous grayscale transition section in a continuous regional distribution. At this location, the grayscale transition zone between the facial boundary and the edge of the protective equipment jumps from a continuous gradual state to an interrupted state.

[0304] A discontinuous segment is a continuous line segment formed by connecting multiple discontinuities that appear sequentially along the facial boundary in a continuous distribution of a region according to their spatial adjacency. Each discontinuous segment corresponds to a continuously extending microscopic gap region.

[0305] The extension length is the spatial distance between the starting discontinuity and the ending discontinuity along the direction of the facial boundary of the discontinuous segment.

[0306] The permissible discontinuity range is a threshold for the maximum extension length of discontinuous segments that are allowed to exist in each facial region, set individually in the airtightness benchmark. When the extension length of the discontinuous segment does not exceed this threshold, the airtightness of the region is considered to be within an acceptable range.

[0307] A section that fails to meet airtightness standards is a discontinuous section in a continuous regional distribution whose length exceeds the allowable discontinuity range corresponding to that facial region.

[0308] The abnormal segment distribution is a spatial distribution set formed by arranging all the airtightness-deficient segments according to their respective facial region locations and extension order along the facial boundary. This set records the location, start position, end position, and extension length of each airtightness-deficient segment.

[0309] The airtightness comparison result is the final judgment based on whether there are any substandard airtightness sections in the abnormal section distribution. When there are no substandard airtightness sections in the abnormal section distribution, the airtightness comparison result is qualified; when there is at least one substandard airtightness section, the airtightness comparison result is unqualified.

[0310] Furthermore, the previously obtained data on the continuous distribution of the entire edge in the airtight bonding state is used to determine the location of each detection point on the facial boundary in the actual image space, and then compared with the spatial range of each facial region pre-divided in the airtight reference.

[0311] The determination rule is that if the detection point falls within the forehead area, it is classified as a continuous distribution of the forehead area; if it falls within the bridge of the nose area, it is classified as a continuous distribution of the bridge of the nose area; if it falls within the left cheek area, it is classified as a continuous distribution of the left cheek area, and so on, until all detection points on the facial boundary have completed the location determination, thus obtaining the independent continuous distribution of each facial region.

[0312] After obtaining the regional continuity distribution of each facial region, discontinuity extraction is performed separately for each regional continuity distribution, including: along the facial boundary corresponding to the facial region, scanning the category label of each segment in the regional continuity distribution point by point from one end to the other end; whenever the segment category changes from a gray-scale gradient continuous segment to a micro-gap region, the transition position is recorded as a discontinuity; whenever the segment category changes from a micro-gap region to a gray-scale gradient continuous segment, the transition position is also recorded as a discontinuity.

[0313] After extracting all discontinuities, the two end discontinuities belonging to the same micro-gap region are connected in series to form a discontinuous segment. One micro-gap region corresponds to one discontinuous segment. The starting point of the discontinuous segment is the starting discontinuity position of the micro-gap region, and the ending point is the ending discontinuity position of the micro-gap region.

[0314] After obtaining each discontinuous segment, measure the actual spatial distance of each discontinuous segment from the starting point to the ending point along the facial boundary direction, and take this distance value as the extension length of the discontinuous segment.

[0315] The extension length of each discontinuous segment is compared with the allowable discontinuity range of the facial region where the discontinuity segment is located in the airtightness standard. If the extension length is less than or equal to the allowable discontinuity range, the discontinuous segment is not marked; if the extension length is greater than the allowable discontinuity range, the discontinuous segment is marked as an airtightness non-compliance segment.

[0316] After traversing all discontinuous segments and completing the above comparison and marking, all discontinuous segments marked as airtightness failure segments are arranged according to their respective facial region locations and the order of extension along the facial boundary to generate an abnormal segment distribution.

[0317] Finally, check whether there are any substandard airtight sections recorded in the abnormal section distribution. If the abnormal section distribution is empty and there are no substandard airtight sections, the comparison result between the airtight bonding state and the airtight benchmark is judged as qualified. If one or more substandard airtight sections are recorded in the abnormal section distribution, the comparison result between the airtight bonding state and the airtight benchmark is judged as unqualified. Thus, the final airtight comparison result is obtained.

[0318] It should be noted that, in order to improve the reliability of the test results and eliminate the limitations of single visual detection which is affected by lighting and skin color, this embodiment adopts a multimodal detection strategy that combines visual images with pressure sensing when performing airtightness testing.

[0319] Specifically, the protective equipment (such as a full-face mask) has a flexible pressure sensor array pre-installed at the edge seal to collect real-time data on the contact pressure distribution between the edge of the equipment and the facial skin.

[0320] While performing grayscale transition band analysis on real-time face images, the system simultaneously acquires contact pressure distribution data. The spatial correspondence between pressure data and images requires the correspondence between the sensor unit position and the image position. Normalization only changes the numerical scale and cannot generate positional correspondence independently. Contact pressure should be compared with a contact threshold using the same pressure unit. If the comparison is performed after normalization, the threshold should be converted to the same scale according to the same rules. The normalized pressure values ​​are organized according to the corresponding image positions to obtain a normalized pressure matrix, which records the normalized pressure values.

[0321] Subsequently, the normalized pressure matrix is ​​used to spatially correct the preliminary airtight bonding state obtained based on the grayscale transition band analysis:

[0322] If the grayscale transition zone analysis results show that a certain area is a continuous grayscale gradient section (visually appears to fit), but the pressure value corresponding to this area in the normalized pressure matrix is ​​lower than the preset contact threshold, then the judgment result of this area will be forcibly corrected to a micro gap area.

[0323] Conversely, if the grayscale transition zone analysis results show that there is a suspected shadow (suspected gap) in a certain area, but the pressure value of that area in the normalized pressure matrix is ​​significantly higher than the threshold required by the airtightness benchmark, then it is determined that the area is a misjudgment caused by the shadow of the light, and it is corrected to a grayscale gradual continuous section.

[0324] Visual features and contact pressure features provide local fit information from different sources, and the spatial correspondence between the two is used to verify suspected gaps in the image. Contact pressure data can supplement image information, but normal contact pressure itself cannot prove that there is no gas leak, nor can it guarantee the accuracy of airtightness detection.

[0325] S6: Generate a work permit instruction based on the matching results of protective equipment and the airtightness comparison results, associate the work permit instruction with the identity identifier and area category, and save it as a work permit file.

[0326] In this embodiment, a work permit instruction is generated based on the matching results of the protective equipment and the airtightness comparison results. The work permit instruction is then associated with the identity identifier and area category and saved as a work permit file, including:

[0327] Read the measured value of the protection level corresponding to the matching result of the protective equipment, compare it with the protection level range allowed by the permitted operation range in order of protection capability, and obtain the protection level matching conclusion.

[0328] Read the airtightness comparison result and compare it with the allowable gap boundary range of the airtightness reference to obtain the airtightness boundary comparison conclusion.

[0329] By correlating the location of the local detachment section in the airtight boundary comparison conclusion with the sealing structure characteristics of the protective equipment, the airtightness verification result is obtained. Based on the protection level adaptation conclusion and the airtightness verification result, an operation permit order is generated to allow or prohibit entry for operation. The operation permit order is then associated with the identity identifier and area category and saved as an operation permit file.

[0330] Specifically, the protection level range is the numerical range between the lowest and highest effective protection levels specified in the permitted work area. The lower limit of this range is the protection threshold that must be met to enter the work area.

[0331] The protection level matching conclusion is a judgment result obtained by comparing the measured value of the protection level with the protection level range. It is used to indicate whether the actual protection capability of the currently worn protective equipment, after considering the wearing quality, falls within the protection level range required for the permitted operation.

[0332] The gap boundary range is the upper limit of the allowable gap width and the upper limit of the allowable gap extension length set in the airtight standard for each contact section between the respiratory protective equipment and the face. Different facial areas have their own independent gap boundary ranges.

[0333] The airtight boundary comparison conclusion is a judgment result obtained by comparing the actual gap parameters of each contact section in the airtight bonding state with the gap boundary range of the corresponding section in the airtight reference one by one. It is used to indicate whether the airtight state of each contact section is within the range allowed by the airtight reference.

[0334] The distribution of local discontinuities is a spatial distribution set formed by arranging the discontinuous segments marked as airtight in the airtight boundary comparison results according to their respective facial region locations and extension order along the facial boundary. This set records the location, start position, end position, and extension length of each airtight segment.

[0335] Sealing structural features are the physical structural attributes of respiratory protective equipment at the sealing parts that come into contact with the face, including the width of the sealing lip, the thickness of the sealing lip, the direction of the sealing lip's extension, the coverage area of ​​the sealing lip in different areas of the face, and the contour shape of the sealing lip's edge.

[0336] Spatial correspondence is an operation that compares the actual spatial coordinates of each substandard segment in the local disengagement segment location distribution with the structural parameters of the sealing structure features of the respiratory protective equipment at the same spatial coordinates.

[0337] The airtightness verification result is a re-evaluation of the airtightness effectiveness based on whether each local separation section determined after spatial correspondence operation overlaps with the key sealing parts in the sealing structure features.

[0338] A work permit is a digital instruction signal generated based on the protection level compatibility conclusion and the airtightness verification results, which allows operators to enter the hazardous chemical spill area to perform specified work operations.

[0339] Furthermore, the operator stands in a safe position outside the hazardous chemical spill area, facing the image acquisition device with their face directly facing the camera, so that the image acquisition device can capture a real-time facial image of their face area.

[0340] The value is read from the obtained measured value of the protection level, and the allowed protection level range is read from the permitted operation range. The measured value of the protection level is compared with the lower limit and upper limit of the protection level range respectively. When the measured value of the protection level is greater than or equal to the lower limit of the protection level range and less than or equal to the upper limit of the protection level range, the protection level is deemed to be adapted successfully; otherwise, it is deemed to be adapted unsuccessfully.

[0341] The actual gap width and actual gap extension length values ​​of each contact segment between the face boundary and the edge of the protective equipment are read from the obtained airtight fit status. At the same time, the allowable gap boundary range of the corresponding face area is read from the airtight reference. The actual gap width value of each contact segment is compared with the upper limit of the allowable width in the gap boundary range, and the actual gap extension length value is compared with the upper limit of the allowable extension length in the gap boundary range. When the actual gap width value of a contact segment exceeds the upper limit of the allowable width or the actual gap extension length value exceeds the upper limit of the allowable extension length, the contact segment is marked as an airtight non-compliant segment. After traversing all contact segments and completing the above comparison and marking, an airtight boundary comparison conclusion containing the location information of all airtight non-compliant segments is generated.

[0342] The actual spatial coordinates of each airtight section on the face boundary are extracted from the airtight boundary comparison results. At the same time, the sealing structure feature data of the respiratory protective equipment is read. This data records the spatial distribution coordinates and extension direction of the sealing lip in each face area. The coordinates of each airtight section are spatially overlapped and compared with the coordinates of the sealing lip in the sealing structure feature. If the location of the local detachment section coincides with the low pressure area detected in the pressure sensor array in spatial coordinates, it is confirmed that there is a real risk of airtight failure at that location, and an airtight verification result with the risk of sealing structure failure is generated.

[0343] If the local detachment segment only exists in the visual image, but the corresponding physical pressure distribution data shows normal contact pressure, then the detachment segment is determined to be a false defect caused by light and shadow interference, and an airtightness verification result with normal airtightness is generated.

[0344] Read the protection level adaptation conclusion and the airtightness verification result. When the protection level adaptation conclusion is that the adaptation is passed and the airtightness verification result is that there is no risk of sealing structure failure, generate a work permit order that allows entry into the work. When the protection level adaptation conclusion is that the adaptation is not passed or the airtightness verification result is that there is a risk of sealing structure failure, generate a work permit order that prohibits entry into the work.

[0345] This method can be executed by the aforementioned terminal or server. The processing device sequentially completes qualification file matching, dynamic image analysis, equipment matching, airtightness comparison, and license file storage. The aforementioned input-output relationship is retained between each processing process, and these specific processing processes cannot be replaced by module division.

[0346] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0347] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0348] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0349] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for matching and testing the qualifications of personnel and personal protective equipment in hazardous chemical handling, characterized in that, The method includes: S1: Obtain the area category of the hazardous chemical leak area and the identification of the operators, determine the target protection level and airtightness benchmark based on the area category, and collect dynamic images and real-time facial images of the operators; S2: Match the qualification level with the identity identifier in the qualification database, compare the qualification level with the target protection level, and obtain the permitted operation scope; S3: Extract motion feature signals and clothing fold dynamic features from dynamic images, match the motion feature signals with motion feature samples in the qualification database to obtain the identity verification result, and when the identity verification result is valid, identify the protective equipment logo in the dynamic image to obtain the wearing protection level. S4: Identify the fit of protective equipment based on the dynamic characteristics of clothing folds, generate a measured value of the protection level by combining the wearing protection level, match the measured value of the protection level with the permitted work range, and obtain the matching result of the protective equipment. S5: Extract the distribution of the continuity between the facial boundary and the edge of the protective equipment in the real-time face image to obtain the airtight fit state. Compare the airtight fit state with the airtight reference to obtain the airtight comparison result. S6: Generate a work permit instruction based on the matching results of protective equipment and the airtightness comparison results, associate the work permit instruction with the identity identifier and area category, and save it as a work permit file.

2. The matching and testing method for personnel qualifications and personal protective equipment in hazardous chemical operations as described in claim 1, characterized in that, Based on the identity identifier, the qualification level is matched with the qualification database. The qualification level is then compared with the target protection level to obtain the permitted scope of operation, including: The identity identifier is broken down into the current certificate code and the historical certificate code, and candidate current files and candidate historical files are retrieved respectively. The historical certificate code is compared with the historical certificate code recorded in the candidate current file. If they do not match, the certificate renewal record of the candidate historical file is traced to the current valid file. If they match, the candidate current file is selected to obtain a unique matching file. Read the review status identifier of the unique matching file. If the assessment conclusion is passed and the assessment date is within the specified validity period before the current date, extract the qualification level. Compare the hazardous chemical media category, highest protection level and operation type in the qualification level with the corresponding requirements of the target protection level to obtain the permitted operation scope.

3. The matching and testing method for personnel qualifications and personal protective equipment in hazardous chemical operations as described in claim 1, characterized in that, Based on the identity identifier, the qualification level is matched with the qualification database. The qualification level is then compared with the target protection level to obtain the permitted scope of operation, including: The inclusion comparison between the hazardous chemical media category in the qualification level and the leakage media type in the target protection level is performed to obtain the media matching result; When the media matching result is valid, the highest protection level in the qualification level is compared with the lowest protection level in the target protection level in order of protection capability from strong to weak to obtain the strength matching result. Based on the intensity matching results and the leakage concentration level of the area category, the concentration is divided into zones to obtain candidate operation ranges. The task project number of each spatial sub-area within the candidate operation range is matched with the corresponding number of the operation type in the qualification level. The boundary coordinates of the successfully matched spatial sub-area and the operation tasks are associated to obtain the permitted operation range.

4. The matching and testing method for personnel qualifications and personal protective equipment in hazardous chemical operations as described in claim 1, characterized in that, Motion feature signals and clothing fold dynamic features are extracted from dynamic images. These motion feature signals are then matched with motion feature samples in the qualification database to obtain an authentication result. If the authentication result is valid, the protective equipment identifier in the dynamic image is recognized to determine the wearing protection level, including: A stage sequence is formed based on the changing time sequence of motion feature signals and dynamic features of clothing folds. Continuous motion segments that are consistent with the sequence of motion feature samples are retained to obtain sequentially consistent segments. From the sequentially consistent segments, select the spatially matching segments whose joint bending direction matches the crease change, and compare the crease extension direction of the spatially matching segments with the corresponding moment of the motion feature sample to obtain the equipment linkage segment; The coverage of the equipment linkage section is checked in stages. The ratio of the total duration of the equipment linkage section to the total duration of all action stages is used as the coverage level. The identity verification result is generated based on the comparison result of the coverage level and the preset qualified threshold. When the identity verification result is valid, the protective equipment logo is identified to obtain the wearing protection level.

5. The matching and testing method for personnel qualifications and personal protective equipment in hazardous chemical operations as described in claim 1, characterized in that, Motion feature signals and clothing fold dynamic features are extracted from dynamic images. These motion feature signals are then matched with motion feature samples in the qualification database to obtain an authentication result. If the authentication result is valid, the protective equipment identifier in the dynamic image is recognized to determine the wearing protection level, including: When the identity verification result is valid, the micro-deformation region and the deformation-stable region are distinguished according to the dynamic characteristics of clothing folds. The candidate regions for identification that have abrupt changes in reflected light intensity and form closed edges are extracted from the deformation-stable regions. Compare the edge orientation of the candidate identification area with the crease extension direction of the micro-deformation area, and retain intersecting or independent valid identification edges; The pattern enclosed by the effective marking edge and the arrangement of the internal color blocks are compared with the pre-stored marking sample. The protection level marking information of the matching sample is read, and the wearing protection level is determined according to the protection level code in the protection level marking information.

6. The matching and testing method for personnel qualifications and personal protective equipment in hazardous chemical operations as described in claim 1, characterized in that, The fit of protective equipment is determined by identifying the dynamic characteristics of clothing folds. Combined with the wear protection level, a measured protection level value is generated. This measured protection level value is then matched with the permitted work area to obtain the matching result of the protective equipment, including: By tracking the creases in the dynamic features of clothing folds frame by frame, the distribution of crease bands is obtained, and the sliding crease bands that have undergone positional migration and the stationary crease bands that have fixed positions are distinguished. The convergence node is identified based on the adjacency relationship between the slip crease zone and the dwell crease zone, and the directional concentration area is identified according to the direction of the creases around the convergence node; The closed-bound areas within the directional concentration region are designated as the fit-support zone, and the open-bound areas are designated as the fit-detachment zone. By combining the spatial envelopment relationship, the dynamic contact state between the protective equipment and the operator's body is analyzed to obtain the fit degree. Based on the fit degree, the wearing protection level is corrected to form the measured protection level value. This value is then compared with the lower limit of the protection level for the permitted work area to obtain the matching result of the protective equipment.

7. The matching and testing method for personnel qualifications and personal protective equipment in hazardous chemical operations as described in claim 1, characterized in that, The fit of protective equipment is determined by identifying the dynamic characteristics of clothing folds. Combined with the wear protection level, a measured protection level value is generated. This measured protection level value is then matched with the permitted work area to obtain the matching result of the protective equipment, including: Continuous crease regions with fixed positions are selected from the dynamic features of clothing folds, and the cross-sections where the continuous crease regions intersect with the edges of protective equipment are extracted. Based on the situation where the edge of the protective equipment is divided into two parts by the cross section and the connection between the two parts, the gap connection state is obtained. The wearing protection level in the gap connection state is downgraded according to the degree of fit, and the actual protection level value is obtained. The measured value of the protection level is compared with the lower limit of the protection level in the permitted operation range in order of protection capability. If the lower limit of the protection level is reached, the matching result is considered passed; otherwise, it is considered failed, thus obtaining the matching result of the protective equipment.

8. The matching and testing method for personnel qualifications and personal protective equipment in hazardous chemical operations as described in claim 1, characterized in that, Extract the continuity distribution of facial boundaries and protective equipment edges from real-time face images to obtain the airtight fit status. Compare the airtight fit status with an airtight benchmark to obtain the airtightness comparison results, including: Samples are taken from the facial boundary toward the edge of the protective equipment to form a grayscale transition zone, and the grayscale gradient interruption section is marked as a micro-gap region; By arranging the micro-gap regions and gray-scale gradient continuous segments sequentially along the facial boundary, an edge continuity distribution is obtained, connecting spatially adjacent micro-gap regions into potential leakage paths. The positions at both ends of the potential leakage path are compared with the two sides of the contact area of ​​the face boundary and the edge of the protective equipment to obtain the airtight fit status indicating whether the potential leakage path penetrates the contact area. Then, the airtight fit status is compared with the allowable gap range of the corresponding face area in the airtight reference to obtain the airtight comparison result.

9. The matching and testing method for personnel qualifications and personal protective equipment in hazardous chemical operations as described in claim 1, characterized in that, Extract the continuity distribution of facial boundaries and protective equipment edges from real-time face images to obtain the airtight fit status. Compare the airtight fit status with an airtight benchmark to obtain the airtightness comparison results, including: The edge continuity distribution corresponding to the airtight fit state of the facial area is matched with the allowable gap range of the airtight reference to obtain the area continuity distribution; The regions where grayscale gradients are interrupted in the continuous distribution of the region are taken as micro-gap regions. The discontinuities at both ends of the micro-gap regions are extracted and connected to form discontinuous segments. The position and extension length of the discontinuous segments are obtained. The facial region, start position and end position of each discontinuous segment are recorded and arranged in the order of extension along the facial boundary. Discontinuous segments whose extension length exceeds the allowable discontinuity range of the corresponding facial area are marked as airtightness non-compliant segments. The abnormal segment distribution is obtained by arranging them by position. When there are no airtightness non-compliant segments in the abnormal segment distribution, a qualified airtightness comparison result is generated. When there are airtightness non-compliant segments, an unqualified airtightness comparison result is generated.

10. The matching and testing method for personnel qualifications and personal protective equipment in hazardous chemical operations as described in claim 7, characterized in that, Based on the matching results and airtightness comparison results of the protective equipment, a work permit order is generated. This work permit order is then associated with the identification and area category and saved as a work permit file, including: Read the measured value of the protection level corresponding to the matching result of the protective equipment, compare it with the protection level range allowed by the permitted operation range in order of protection capability, and obtain the protection level matching conclusion. Read the airtightness comparison result and compare it with the allowable gap boundary range of the airtightness reference to obtain the airtightness boundary comparison conclusion. By correlating the location of the local detachment section in the airtight boundary comparison conclusion with the sealing structure characteristics of the protective equipment, the airtightness verification result is obtained. Based on the protection level adaptation conclusion and the airtightness verification result, an operation permit order is generated to allow or prohibit entry for operation. The operation permit order is then associated with the identity identifier and area category and saved as an operation permit file.

Images