Medical waste tracking management system and method combined with face recognition

Abstract

The application discloses a medical waste tracking management system and method combined with face recognition, and relates to the technical field of medical waste management.The system comprises the following steps: obtaining target medical waste; calling predetermined characteristic indexes to collect multi-source characteristics to obtain target characteristic index information and generate an initial point code label; performing similar clustering of the target medical waste; matching and determining a predetermined waste disposal decision according to the waste clustering result; calling an intelligent image collector to perform dynamic monitoring to obtain a target processing image; matching target personnel characteristic information in a personnel database; and obtaining a target point code label.The application solves the technical problems of the prior art, such as the fact that medical waste information tracking is not accurate and complete, and the fact that there is a lack of dynamic monitoring of the processing process, which leads to insufficient traceability, accuracy and safety of the medical waste processing process, and achieves the technical effect of improving the traceability, accuracy and safety of the waste processing process.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical waste management, and particularly relates to a medical waste tracking management system and method combined with face recognition. BACKGROUND

[0002] With the increasing strictness of medical waste treatment requirements and the complexity of management, the existing medical waste management method faces many challenges, especially in the tracking and monitoring links. Medical waste usually includes needle, medicine bottle, surgical instrument and the like, and its management involves multiple aspects such as environmental protection and public health safety. In recent years, the medical industry has gradually introduced Internet of Things technology, bar code scanning technology, intelligent sensors and the like, and through automatic monitoring of equipment and data collection, real-time tracking management of medical waste is carried out, but there are still some limitations, for example, the diversity of medical waste is not handled finely and completely, and real-time monitoring and tracking of personnel behavior cannot be effectively realized, thereby affecting the transparency of medical waste treatment, the responsibility tracing ability and the intelligent management of the waste treatment process, and improper treatment may lead to serious health and safety risks.

[0003] Therefore, in the current medical waste tracking management related technology, there is a technical problem that the medical waste information tracking is not accurate and complete, and the dynamic monitoring of the treatment process is lacking, thereby leading to insufficient traceability, accuracy and safety of the medical waste treatment process. SUMMARY

[0004] The medical waste tracking management system and method combined with face recognition provided by the present application solve the technical problem that the medical waste information tracking is not accurate and complete, and the dynamic monitoring of the treatment process is lacking in the prior art, thereby leading to insufficient traceability, accuracy and safety of the medical waste treatment process, and achieve the technical effect of improving the traceability, accuracy and safety of the waste treatment process.

[0005] The application provides a medical waste tracking management system combined with face recognition. The system comprises: a target medical waste acquisition module for acquiring target medical waste, wherein the target medical waste refers to any waste meeting predetermined waste constraints; a target characteristic index information collection module for collecting target characteristic index information by collecting multi-source characteristics of the target medical waste according to predetermined characteristic indexes, and generating an initial point code label of the target characteristic index information; a waste clustering result obtaining module for performing similar clustering of the target medical waste by taking the target characteristic index information as a constraint to obtain a waste clustering result; a predetermined waste disposal decision determination module for determining a predetermined waste disposal decision of the target medical waste according to the waste clustering result; a target processing image obtaining module for calling an intelligent image collector to dynamically monitor a processing process of the target medical waste based on the predetermined waste disposal decision to obtain a target processing image; a target personnel characteristic information matching module for extracting a target face slice in the target processing image based on face recognition technology, and matching target personnel characteristic information of a target personnel corresponding to the target face slice in a personnel database; and a target point code label obtaining module for adding the target personnel characteristic information to the initial point code label to obtain a target point code label, wherein the target point code label is used for tracking management of the target medical waste.

[0006] In a possible implementation, the medical waste tracking management method combined with face recognition further performs the following processing: the predetermined characteristic indexes at least include a source, an infection route, a pathogen type, and a risk level.

[0007] In a possible implementation, the medical waste tracking management method combined with face recognition further performs the following processing: a target image of the target medical waste is collected; target label text information is obtained by recognizing text on the target image; a target source of the target medical waste is determined by analyzing the target label text information; and the target source is added to the target characteristic index information.

[0008] In a possible implementation, the medical waste tracking management method combined with face recognition further performs the following processing: first characteristic index information of a first medical waste is acquired; when the target characteristic index information and the first characteristic index information meet a predetermined clustering constraint, a first waste clustering cluster is formed based on the target medical waste and the first medical waste; and the waste clustering result is formed according to the first waste clustering cluster.

[0009] In a possible implementation, a predetermined weight distribution of the predetermined characteristic indicators is acquired; the analysis and calculation of the target characteristic indicator information and the first characteristic indicator information are sequentially performed in combination with the predetermined weight distribution, to respectively obtain a target characteristic coefficient of the target medical waste and a first characteristic coefficient of the first medical waste; a first characteristic deviation rate is obtained by comparing the target characteristic coefficient and the first characteristic coefficient; the first characteristic deviation rate is analyzed by introducing a clustering support evaluation function, to obtain a first clustering support; it is determined whether the first clustering support meets a predetermined support threshold; and if not, the first waste clustering cluster is adjusted and intervened.

[0010] In a possible implementation, the medical waste tracking management method combined with face recognition further performs the following processing: an expression of the clustering support evaluation function is as follows:

[0011] ;

[0012] wherein, is the first characteristic coefficient of the target characteristic coefficient , and k is the amplitude of the clustering support evaluation function. , is a clustering attenuation factor, and .

[0013] In a possible implementation, the medical waste tracking management method combined with face recognition further performs the following processing: a first image in the plurality of images is extracted, the first image corresponding to a first processing stage; a first face slice in the first image is acquired by using a semantic recognition principle; first personnel characteristic information of a first personnel corresponding to the first face slice is matched in the personnel database; the target personnel is formed based on the first personnel, and the target personnel characteristic information is formed based on the first personnel characteristic information.

[0014] The application also provides a medical waste tracking management method combined with face recognition, which comprises the following steps: obtaining a target medical waste, wherein the target medical waste refers to any waste meeting a predetermined waste constraint; collecting target feature index information of the target medical waste by collecting multi-source features of the target medical waste according to the predetermined feature index, and generating an initial point code label of the target feature index information; performing similar clustering on the target medical waste by taking the target feature index information as a constraint to obtain a waste clustering result; determining a predetermined waste disposal decision of the target medical waste according to the waste clustering result; dynamically monitoring a disposal process of the target medical waste based on the predetermined waste disposal decision by calling an intelligent image collector to obtain a target disposal image; extracting a target face slice in the target disposal image based on face recognition technology, and matching target personnel feature information of a target personnel corresponding to the target face slice in a personnel database; adding the target personnel feature information to the initial point code label to obtain a target point code label, wherein the target point code label is used for tracking management of the target medical waste.

[0015] The medical waste tracking management system and method combined with face recognition provided in the application can obtain a target medical waste, collect target feature index information by collecting multi-source features according to a predetermined feature index, and generate an initial point code label; perform similar clustering on the target medical waste; determine a predetermined waste disposal decision according to a waste clustering result; dynamically monitor a disposal process by calling an intelligent image collector to obtain a target disposal image; match target personnel feature information in a personnel database; and obtain a target point code label. The technical problems that the medical waste information tracking is not accurate and complete, the disposal process is not dynamically monitored, and the traceability, accuracy and safety of the medical waste disposal process are insufficient in the prior art are solved, and the technical effects of improving the traceability, accuracy and safety of the waste disposal process are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. The flowcharts are used to illustrate the operations performed by the system according to the embodiments of the present application in the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously according to needs. Meanwhile, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 The medical waste tracking management system combined with face recognition provided in the embodiments of the present application is shown in the structural schematic diagram.

[0018] Figure 2A flowchart of a medical waste tracking management method combined with face recognition provided by an embodiment of the present application is shown.

[0019] Reference signs: target medical waste acquisition module 10, target feature index information collection module 20, waste clustering result acquisition module 30, predetermined waste disposal decision determination module 40, target processing image acquisition module 50, target personnel feature information matching module 60, target point code label acquisition module 70. DETAILED DESCRIPTION

[0020] The above description is only a summary of the technical solutions of the present application. In order to make the technical means of the present application more clearly understood, and to be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0021] In order to make the purposes, technical solutions and advantages of the present application more clear, the following will further describe the present application in combination with the drawings, and the described embodiments should not be regarded as limiting the present application. All other embodiments obtained by those skilled in the art without making creative labor, belong to the scope of protection of the present application.

[0022] In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subset of all possible embodiments, and can be combined with each other without conflict. The term "first\second" is only to distinguish similar objects, and does not represent the specific order of the objects. The terms "include" and "have" and any variations, are intended to cover non-exclusive inclusion, for example, a process, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0023] The embodiments of the present application provide a medical waste tracking management system combined with face recognition, as shown in Figure 1 The system comprises:

[0024] The target medical waste acquisition module 10 is used to acquire target medical waste, and the target medical waste refers to any waste that meets the predetermined waste constraints.

[0025] Preferably, the target medical waste is obtained, which refers to any waste in the medical waste that meets the predetermined waste constraints, for example, medical waste can be divided into different types according to its nature, such as infectious waste (such as needle, medical equipment, etc. that has been in contact with patients with infectious diseases), harmless waste, drug waste, chemical waste, plastic bags and ordinary hospital supplies, etc. The target medical waste meets the standard of the predetermined waste constraints, and is more accurately managed through identification and tracking.

[0026] The target feature index information collection module 20 is configured to collect the target feature index information of the target medical waste by calling the predetermined feature index, and generate an initial point code label of the target feature index information.

[0027] The specific configuration of the target feature index information collection module 20 further comprises that the predetermined feature index at least includes source, transmission path, pathogen type and risk level.

[0028] Preferably, by calling the multiple key feature indexes of the target medical waste (such as source, transmission path, pathogen type, risk level, etc.), comprehensive feature information is collected, and an initial point code label (such as a two-dimensional code or a bar code) is generated based on these information, which is used to uniquely identify the target medical waste. Specifically, the predetermined feature index refers to the key information required to be collected in the medical waste management process in order to accurately describe and track the target waste. These indexes can help to distinguish and identify the important features of the waste, such as the type, source, infectivity, risk level, etc., and thus provide a basis for subsequent management and processing. The predetermined feature index at least includes source, transmission path, pathogen type and risk level, wherein the source refers to the source type of the medical waste, which may be from different medical institutions such as hospitals, clinics, laboratories, etc. For example, whether it is waste generated by departments such as infectious disease department, emergency department, etc. The transmission path refers to the transmission mode of the waste, which usually includes air transmission, contact transmission, droplet transmission, etc., which is helpful to evaluate the potential harm of the waste and the processing measures to be taken. The pathogen type refers to the type of pathogen that may be contained in the waste, such as bacteria, viruses, fungi, etc. Different types of pathogens may cause different health risks. The risk level of medical waste is classified according to its harmfulness to the environment and health, which usually includes different levels such as high risk, medium risk and low risk. High-risk waste such as infectious waste needs to be strictly treated, while low-risk waste can be treated by conventional methods.

[0029] Preferably, the multi-source feature collection of the target medical waste means that different types of information related to the target waste are obtained from multiple different sources or channels according to predetermined feature indicators, such as collecting the generation time, treatment requirements, storage location, historical treatment records and other data of the target medical waste, and comprehensively understanding the characteristics of the target waste by integrating information from different sources. The multi-source data obtained as target feature indicator information includes all collected feature data (such as source, transmission route, pathogen type and risk level, etc.) and related data, and then an initial point code label is generated according to these feature data. Specifically, the initial point code label is a unique identifier (such as a two-dimensional code, a bar code, etc.) of the target waste, which contains the feature information (i.e. target feature indicator information) of the target waste. The detailed data related to the waste can be easily obtained by scanning these labels. Each medical waste can be identified by a unique point code label (such as a two-dimensional code, a bar code, etc.), thereby avoiding confusion, and the detailed data of the waste such as source, pathogen type, risk level, etc. can be quickly obtained by using a scanning device, which is convenient for tracking, management and recording, thereby realizing efficient management of medical waste from generation to treatment, and ensuring the traceability and safety of the waste.

[0030] Further, the specific configuration of the target feature indicator information collection module 20 further includes collecting a target image of the target medical waste; recognizing the text on the target image to obtain target label text information; analyzing the target label text information to determine the target source of the target medical waste; and adding the target source to the target feature indicator information.

[0031] Preferably, the image information of the target medical waste is obtained by some image acquisition device (such as a smart camera, mobile device or scanner, etc.), and the target image, that is, the image of the waste itself or the image information of the label, barcode, etc. on the waste is obtained. Then, the text in the image is recognized by using text recognition technology (OCR, optical character recognition), and the text information is extracted and converted into a data format that can be stored and processed. It may be the content written on the label, packaging, bag, etc. attached to the waste, which may include the type of medical waste, the source of the waste (such as a specific department or department), the production date or processing requirements, and the specific identification of the waste, etc. The text information output is analyzed to determine the source of the target medical waste. For example, the waste may come from different departments such as infectious diseases, emergency departments, operating rooms, etc., the category of waste (such as infectious waste, sharps waste, pharmaceutical waste, etc.) and the person responsible for waste treatment. After identifying and analyzing the source of the target waste, it is added to the characteristic indicator information of the target medical waste. The target characteristic indicator information is a collection of all key characteristic data of the waste, such as transmission route, pathogen type, risk level, etc. The relevant information of the target source is stored together with other characteristics (such as risk level, transmission route, etc.) for accurate tracking and management in the waste management system.

[0032] The waste clustering result obtaining module 30 is used to perform similarity clustering of the target medical waste using the target characteristic index information as a constraint to obtain a waste clustering result.

[0033] Preferably, cluster analysis is used to classify multiple medical wastes based on the key characteristic information of medical waste (i.e., target characteristic index information) to identify similar types of wastes. In medical waste management, clustering algorithms (K-means clustering, hierarchical clustering, DBSCAN, etc.) are used to classify similar wastes (e.g., wastes with similar sources, similar transmission routes, and similar pathogen types) into the same category by analyzing the characteristic indicators of different wastes (e.g., sources, risk levels, etc.). Specifically, the characteristic information of the target medical waste (e.g., sources, transmission routes, etc.) is used to classify the wastes into the same category. The input data is information such as infection pathways, risk levels, etc., and the similarity between each waste is calculated. Distance metrics (such as Euclidean distance, Manhattan distance, etc.) are usually used to measure the similarity of wastes. Based on the similarity information, a clustering algorithm is used to group similar wastes. For example, wastes with the same risk level and infection pathway may be clustered together. Finally, the waste clustering result is obtained, that is, multiple groups, each group represents a type of waste with similar characteristics, which helps to distinguish different types of waste, optimize the treatment process, improve resource allocation efficiency, and ensure that resources and equipment are reasonably allocated.

[0034] Further, the specific configuration of the waste clustering result obtaining module 30 further comprises: obtaining first characteristic index information of a first medical waste; when the target characteristic index information and the first characteristic index information meet a predetermined clustering constraint, forming a first waste clustering cluster based on the target medical waste and the first medical waste; and forming the waste clustering result according to the first waste clustering cluster.

[0035] Preferably, the first medical waste is a kind of medical waste existing in the system, and the first characteristic index information of the first medical waste is obtained from the existing waste data, that is, the detailed characteristic description of the waste, including the type, source, infectivity, risk level and the like of the waste. The characteristic information of the target medical waste is compared with the characteristic information of the first medical waste to determine whether a certain predetermined clustering constraint is met, wherein the predetermined clustering constraint is a preset condition, including that some characteristics must be consistent, such as the target waste and the first waste coming from the same department, the risk level of the target waste and the first waste being similar, and the like. Then, the target waste and the first waste meeting the clustering constraint are aggregated into a waste clustering cluster, representing a kind of medical waste with similar characteristics. For example, all wastes coming from the same department, having similar transmission routes and the same pathogen type are aggregated into the same clustering cluster. Finally, the waste clustering result is integrated and output, representing the grouping situation of all wastes meeting the clustering condition. Each clustering cluster contains some wastes similar in characteristics, which can be processed, managed or tracked according to their similarity, helping to accurately classify and effectively manage medical waste, and ensuring the safety and efficiency of the waste treatment process.

[0036] Further, the specific configuration of the waste clustering result obtaining module 30 further comprises: obtaining a predetermined weight distribution of the predetermined characteristic index; sequentially analyzing and calculating the target characteristic index information and the first characteristic index information in combination with the predetermined weight distribution, respectively obtaining a target characteristic coefficient of the target medical waste and a first characteristic coefficient of the first medical waste; comparing the target characteristic coefficient and the first characteristic coefficient to obtain a first characteristic deviation rate; introducing a clustering support evaluation function to analyze the first characteristic deviation rate to obtain a first clustering support; determining whether the first clustering support meets a predetermined support threshold; and if not, adjusting and intervening the first waste clustering cluster.

[0037] The specific configuration of the waste clustering result obtaining module 30 further comprises that the expression of the clustering support evaluation function is as follows:

[0038] ;

[0039] wherein, is the first characteristic coefficient a target feature coefficient of the target medical waste and a first feature coefficient of the first medical waste, wherein the target feature coefficient represents a comprehensive coefficient of all features of the target medical waste after weighted calculation, and the first feature coefficient represents a comprehensive coefficient of all features of the first medical waste after weighted calculation. k is a magnitude of the clustering support evaluation function, and , is a clustering attenuation factor, and .

[0040] Preferably, based on the weighted feature indicators, the clustering optimization is performed by analyzing the similarity of the medical wastes to ensure that the clustering result is more reasonable and accurate, and the predetermined weight distribution of the predetermined feature indicators is obtained, wherein each medical waste has a predetermined weight for each feature indicator (such as source, transmission route, pathogen type, risk level, etc.), which reflects the importance of each feature in the clustering process, and is usually set based on the management needs of the medical wastes or the importance of the medical wastes, for example, the transmission route may account for 30%, the pathogen type may account for 40%, and the risk level may account for 30%. Then, according to the set weight, the feature information of the target medical waste and the first medical waste is analyzed and calculated, that is, each pair of features (such as the transmission route of the target medical waste and the transmission route of the first medical waste) is compared, and each feature is weighted and calculated according to the set weight to obtain the feature coefficients of the two medical wastes. The target feature coefficient represents the comprehensive coefficient of all features of the target medical waste after weighted calculation, and the first feature coefficient represents the comprehensive coefficient of all features of the first medical waste after weighted calculation.

[0041] Preferably, after the target feature coefficient and the first feature coefficient are calculated, the two coefficients are compared, and the first feature deviation rate is calculated to reflect the difference between the target medical waste and the first medical waste in the weighted feature indicators. If the deviation rate is high, it means that there is a large difference between the two medical wastes in the features. Then, the first feature deviation rate is analyzed according to the clustering support evaluation function. Specifically, the clustering support evaluation function is used to evaluate the credibility and stability of the clustering result, and the clustering is quantified based on the deviation rate to ensure that the clustering result can reflect the real features of the medical wastes. If the first feature deviation rate is very high, the clustering support may be low, which means that the target medical waste and the first medical waste have large differences in features and may not be suitable for clustering together. It is judged whether the first clustering support exceeds the preset support threshold. The support threshold is a threshold set according to the application requirements, which is used to determine whether the two medical wastes can be reasonably gathered in the same cluster. For example, the support threshold can be set to 80%, which means that if the clustering support is less than 80%, it is considered that the clustering is not stable enough and the two medical wastes cannot be merged. If the first clustering support fails to reach the predetermined threshold, the first medical waste cluster is adjusted and intervened, for example, the members of the clustering cluster are adjusted, the target medical waste is compared with other medical wastes, or new features are added to optimize the clustering result, until the satisfactory support threshold is reached.

[0042] The predetermined waste disposal decision determination module 40 is configured to determine a predetermined waste disposal decision of the target medical waste according to the waste clustering result matching.

[0043] Preferably, after clustering analysis of the medical waste, the treatment scheme of the waste in each cluster is determined according to the clustering result. Specifically, according to the clustering result of the waste, each cluster is analyzed, and a corresponding treatment scheme is formulated for each cluster. In combination with the characteristics of the cluster (such as the type of waste, the risk level, etc.), the waste in the cluster is associated with the predetermined treatment decision. The predetermined waste disposal decision is a preset treatment scheme based on the characteristics of the medical waste. It is a treatment method set in advance according to the category, characteristics, risk level, etc. of the waste. For example, for a waste cluster with strong infectivity and high risk, special treatment methods such as high-temperature disinfection and chemical disinfection can be selected. For a low-risk waste cluster, only conventional harmless treatment (such as incineration or landfill) can be required. For some medical waste (such as some packaging materials), recycling treatment methods can be adopted. Through data-driven decision-making, the accuracy, efficiency and safety of waste management are greatly improved.

[0044] The target processing image obtaining module 50 is configured to call the intelligent image collector to dynamically monitor the processing process of the target medical waste based on the predetermined waste disposal decision, and obtain a target processing image.

[0045] Preferably, image acquisition technology (such as a camera, an intelligent image sensor, etc.) is used to monitor and record the state of the target medical waste in the processing process in real time, and corresponding image data, i.e. target processing image, is generated. Ensure that the processing process of the waste conforms to the predetermined treatment decision, and can timely discover possible problems or abnormalities. Specifically, the intelligent image collector is used to efficiently monitor the medical waste processing process, i.e. to track and monitor the state of the target medical waste in the processing process in real time, to ensure that the processing process is in accordance with the predetermined waste disposal decision. Images or videos are obtained during the waste processing process as target processing images. The target processing image can include images of the waste at each stage of the processing process, whether the operator processes according to the correct steps, etc., thereby effectively improving the transparency, safety and efficiency of waste management.

[0046] The target personnel feature information matching module 60 is configured to extract a target face slice in the target processing image based on face recognition technology, and match target personnel feature information of a target personnel corresponding to the target face slice in a personnel database.

[0047] Preferably, in the medical waste treatment process, the face recognition technology is used to identify and track the person in the treatment image and obtain detailed information about the personnel, that is, based on the face recognition algorithm, the face part in the image is located and extracted through image processing technology to generate a face feature area (i.e. target face slice), the target face slice refers to the face area related to the operator extracted from the target image, and the slice refers to the area in the image for representing the face (i.e. face feature area). Specifically, the face recognition technology identifies and locates the face in the image by analyzing and processing image data, such as using convolutional neural network (CNN) to extract facial feature points (such as eyes, nose, mouth, etc.) from the image and compare, and finally extract the accurate face area; then match the target personnel feature information of the target personnel corresponding to the target face slice in the personnel database, wherein the personnel database refers to a database storing information of all personnel participating in medical waste treatment, usually including personal identity, job position, work record, processing capacity, and other characteristic information related thereto of each operator. Specifically, after extracting the target face slice, the face recognition algorithm is used to compare the target face slice extracted from the treatment image with the face features in the personnel database, such as calculating the distance and angle between the face feature points to judge the accuracy of the match, and when the matching face feature is found, it is associated with the corresponding operator, and the detailed information (such as name, position, work record, etc.) of the operator is extracted as the target personnel feature information, that is, the personal information and identity characteristic data related to the operator, which can improve the transparency of the medical waste treatment process, ensure the compliance of the operation process, reduce human errors and improper behavior, and improve the safety of waste treatment.

[0048] Further, the specific configuration of the target personnel feature information matching module 60 further includes that the target treatment image includes a plurality of images of a plurality of treatment stages, wherein the plurality of treatment stages at least include a packaging stage, a transportation stage and a storage stage, a first image in the plurality of images is extracted, the first image corresponds to a first treatment stage; a first face slice in the first image is obtained by using semantic recognition principle; first personnel feature information of a first personnel corresponding to the first face slice is matched in the personnel database; the first personnel is used to form the target personnel, and the first personnel feature information is used to form the target personnel feature information.

[0049] Preferably, the waste disposal involves multiple stages, each stage generates a corresponding set of images, which collectively constitute the target disposal image, so the target disposal image includes multiple images of multiple disposal stages, and each stage can include different operations and management activities, and the multiple disposal stages at least include a packaging stage, a transportation stage and a storage stage. Specifically, the packaging stage refers to the initial classification, packaging and labeling of waste at the medical institution or waste generation point, image collection is performed to record the waste packaging and whether the label is clear; the transportation stage refers to the transportation of waste from the generation point to the transfer station or the processing station, image collection is performed to monitor the loading of the transportation vehicle and the state of the waste during transportation; the storage stage refers to temporarily storing the waste in a specific area before final disposal, image collection is performed to monitor the storage location, storage time, environmental conditions (such as temperature, humidity), etc.

[0050] Preferably, in the process of medical waste disposal, personnel information is extracted and analyzed from multiple images through image recognition technology to ensure that the identity of the operator is associated with the disposal process and generate a tracking record. Specifically, any one of the multiple images is selected as a first image, which records the first stage of waste disposal, and then a first face slice in the first image is obtained using semantic recognition principles. Semantic recognition refers to analyzing semantic information in an image through machine learning and computer vision technology, such as recognizing objects, people, scenes, etc. in the image. Here, it refers to recognizing the face area in the image, i.e. extracting the area showing the operator's face (i.e. face slice) from the first image, such as using face detection algorithms (e.g. Haar cascade, deep learning, etc.) to locate and extract the face area in the image; then through face recognition technology, the first face slice extracted from the first image is compared with all facial features in the personnel database, and by calculating the similarity of facial features, the operator matching the face slice is found, and the detailed information (such as name, position, etc.) of the operator is returned as the first personnel feature information. Finally, the target personnel is formed according to the first personnel and the target personnel feature information is formed according to the first personnel feature information, i.e. the complete identity information of the target personnel is constructed according to the matched first personnel feature information, such as the target personnel feature information including other related data (such as disposal task record, operation qualification, identity authentication, etc.) of the personnel in addition to facial features, which is used to form a complete file of the personnel in the waste disposal process, which helps to ensure the identity of all personnel in the waste disposal process, thereby improving the safety, transparency and compliance of waste management.

[0051] The target point code label obtaining module 70 is configured to add the target personnel feature information to the initial point code label to obtain a target point code label, wherein the target point code label is used for tracking management of the target medical waste.

[0052] Preferably, in the medical waste treatment process, the target personnel characteristic information is added to the initial point code label, that is, the identity information of the operator is associated with the tracking label (such as a two-dimensional code, a bar code, etc.) of the waste itself, specifically, the characteristic information of the target personnel is added on the basis of the original initial point code label to form an updated target point code label, which can record the basic information of the waste and the information of the specific operator participating in the treatment, and provide a more comprehensive tracking tool, which can not only track the source and treatment process of the waste, but also track the detailed identity of the operator participating in each link. The target point code label is used for the whole process management of the waste, the current state, the historical treatment record of the waste and the information of the operator can be obtained in real time, once any problem or accident (such as improper treatment, leakage, misplacement, etc.) occurs, the specific waste and its treatment personnel can be quickly located, and the responsibility of each operator can be determined, so that the responsibility person of each link is ensured, thereby improving the operation transparency and enhancing the safety and compliance of the treatment process.

[0053] In the foregoing, with reference to Figure 1 The medical waste tracking management system combined with face recognition according to the embodiments of the application is described in detail. Next, with reference to Figure 2 The medical waste tracking management method combined with face recognition according to the embodiments of the application will be described.

[0054] The medical waste tracking management method combined with face recognition, as shown in Figure 2 The method includes: acquiring a target medical waste, the target medical waste refers to any waste meeting a predetermined waste constraint; calling a predetermined characteristic index to collect multi-source characteristics of the target medical waste to obtain target characteristic index information, and generating an initial point code label of the target characteristic index information; performing similar clustering of the target medical waste with the target characteristic index information as a constraint to obtain a waste clustering result; matching and determining a predetermined waste treatment decision of the target medical waste according to the waste clustering result; calling an intelligent image collector to dynamically monitor a treatment process of the target medical waste based on the predetermined waste treatment decision to obtain a target treatment image; extracting a target face slice in the target treatment image based on a face recognition technology, and matching target personnel characteristic information of a target personnel corresponding to the target face slice in a personnel database; adding the target personnel characteristic information to the initial point code label to obtain a target point code label, wherein the target point code label is used for tracking management of the target medical waste.

[0055] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or the server, the various units and modules included are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific name of each functional unit is only for the convenience of mutual differentiation, and is not used to limit the protection scope of the present application.

[0056] The above detailed description does not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. The medical waste tracking and management system combined with face recognition is characterized by: The system comprises: A target medical waste acquisition module is used to acquire target medical waste, where the target medical waste refers to any waste that meets the predetermined waste constraints; A target characteristic indicator information collection module is used to retrieve predetermined characteristic indicators to perform multi-source characteristic collection on the target medical waste to obtain target characteristic indicator information, and generate an initial point code label for the target characteristic indicator information; A waste clustering result obtaining module, configured to perform similarity clustering of the target medical waste using the target characteristic index information as a constraint to obtain a waste clustering result; A predetermined waste disposal decision determination module, configured to determine a predetermined waste disposal decision for the target medical waste according to the waste clustering result matching; a target processing image acquisition module, configured to call an intelligent image collector to dynamically monitor the treatment process of the target medical waste based on the predetermined waste treatment decision, and obtain a target processing image; A target person feature information matching module is used to extract a target face slice from the target processed image based on face recognition technology, and match the target person feature information of the target person corresponding to the target face slice in a person database; A target point code label acquisition module is used to add the target person's characteristic information to the initial point code label to obtain a target point code label, wherein the target point code label is used to track and manage the target medical waste; The waste clustering result obtaining module performs the following steps: Obtaining first characteristic indicator information of first medical waste; When the target characteristic index information and the first characteristic index information meet a predetermined clustering constraint, forming a first waste cluster based on the target medical waste and the first medical waste; forming the waste clustering result according to the first waste clustering cluster; Obtaining a predetermined weight distribution of the predetermined characteristic indicators; performing analysis and calculation on the target characteristic index information and the first characteristic index information in sequence in combination with the predetermined weight distribution, and obtaining a target characteristic coefficient of the target medical waste and a first characteristic coefficient of the first medical waste respectively; Comparing the target characteristic coefficient with the first characteristic coefficient to obtain a first characteristic deviation rate; Introducing a cluster support evaluation function to analyze the first feature deviation rate to obtain a first cluster support; Determining whether the first cluster support meets a predetermined support threshold; If not, adjusting and intervening in the first waste cluster; The expression of the cluster support evaluation function is as follows: ； in, Refers to the first characteristic coefficient The target characteristic coefficient The first cluster support, k refers to the magnitude of the cluster support evaluation function, and , is the clustering decay factor, and .

2. The medical waste tracking and management system combined with face recognition according to claim 1 is characterized in that: The predetermined characteristic indicators include at least the source, route of infection, pathogen type and risk level.

3. The medical waste tracking and management system combined with face recognition according to claim 2 is characterized in that: The target characteristic indicator information collection module performs the following steps: collecting a target image of the target medical waste; Recognizing the text on the target image to obtain target label text information; Analyzing the target label text information to determine the target source of the target medical waste; The target source is added to the target feature indicator information.

4. The medical waste tracking and management system combined with face recognition according to claim 1 is characterized in that: The target person feature information matching module performs the following steps: The target processing image includes a plurality of images of a plurality of processing stages, wherein the plurality of processing stages include at least a packaging stage, a transportation stage, and a storage stage, and a first image is extracted from the plurality of images, where the first image corresponds to a first processing stage; Acquire a first face slice in the first image using a semantic recognition principle; matching first person feature information of a first person corresponding to the first face slice in the person database; The target person is formed based on the first person, and the target person characteristic information is formed based on the first person characteristic information.

5. A medical waste tracking and management method incorporating facial recognition, the method being applied to the medical waste tracking and management system incorporating facial recognition as claimed in any one of claims 1 to 4, the method comprising: Obtaining target medical waste, where the target medical waste refers to any waste that meets predetermined waste constraints; Retrieving predetermined characteristic indicators to perform multi-source characteristic collection on the target medical waste to obtain target characteristic indicator information, and generating an initial point code label for the target characteristic indicator information; Performing similarity clustering of the target medical waste using the target characteristic index information as a constraint to obtain a waste clustering result; Determining a predetermined waste disposal decision for the target medical waste based on the waste clustering result matching; Invoking an intelligent image collector to dynamically monitor the treatment process of the target medical waste based on the predetermined waste treatment decision to obtain a target treatment image; Extracting a target face slice from the target processed image based on face recognition technology, and matching target person feature information of a target person corresponding to the target face slice in a person database; The target person characteristic information is added to the initial point code tag to obtain a target point code tag, wherein the target point code tag is used to track and manage the target medical waste.

Images