Detailed Description
The embodiment of the application provides a GMP workshop intelligent monitoring management method and system, and solves the technical problem that in the prior art, the workshop is managed mainly in a manual or machine-assisted mode, so that the intelligent degree is low. Acquiring all-around image information of a GMP workshop through an intelligent camera device, clustering the acquired images according to the type of the acquired images, screening out redundant data, and obtaining clustering results representing personnel images and equipment images; furthermore, an intelligent recognition model is trained based on encrypted data acquired by big data of a plurality of different enterprises, image abnormal data in a clustering result is analyzed by the trained intelligent model, and the GMP workshop is managed based on abnormal information. Training samples are increased by collecting data of multiple enterprises, model identification capacity is improved, data privacy among different enterprises is guaranteed by means of encryption training, management is carried out according to abnormal data, automatic management is achieved, and the technical effect of improving GMP workshop management intellectualization is achieved.
Summary of the application
In the production process of a medicine enterprise, in order to ensure the safety of the production process and the quality of produced medicines, the workshop which strictly meets the requirements of a GMP quality safety management system is generally required to work, and the strict management of the GMP workshop is the premise of ensuring good operation. Traditional management mode mainly relies on personnel training and manual monitoring to formulate management standard to realize the management of GMP workshop, but in recent years with the rise of intelligent manufacturing, automatic or semi-automatic production has been realized gradually to medicine enterprise, and traditional management mode has been unable to adapt to present development current situation, but at present, does not appear effective other management modes. However, in the prior art, the workshop is managed mainly in a manual or machine-assisted mode, so that the technical problem of low intelligent degree exists.
In view of the above technical problems, the technical solution provided by the present application has the following general idea:
the embodiment of the application provides a GMP workshop intelligent monitoring management method, wherein the method is applied to a GMP workshop intelligent monitoring management system, the system is in communication connection with an intelligent camera device, and the method comprises the following steps: acquiring a first image information set of a first GMP workshop through the intelligent camera device; performing cluster analysis on the first image information set to obtain a first cluster result, wherein the first cluster result comprises first person image information and first equipment image information; acquiring a first encryption parameter, and constructing a first anomaly identification model based on the first image information set and the first encryption parameter, wherein the first encryption parameter is acquired through multi-party GMP workshop image information aggregation analysis; inputting the first person image information and the first equipment image information into the first anomaly identification model to obtain a first identification result; and managing the first GMP workshop according to the first identification result.
Having thus described the general principles of the present application, various non-limiting embodiments thereof will now be described in detail with reference to the accompanying drawings.
Example one
As shown in fig. 1, an embodiment of the present application provides a GMP workshop intelligent monitoring and management method, where the method is applied to a GMP workshop intelligent monitoring and management system, and the system is in communication connection with an intelligent camera apparatus, and the method includes:
s100: acquiring a first image information set of a first GMP workshop through the intelligent camera device;
specifically, the intelligent camera device is a device for performing all-around video monitoring on each angle of the first GMP workshop, and a plurality of intelligent camera devices in communication connection with the GMP workshop intelligent monitoring management system are preferably deployed, and the deployment position is set based on the fact that the first GMP workshop can be comprehensively monitored, especially key areas such as a vent opening and a production line; the first image information set of the first GMP workshop is an image set which is obtained by extracting and storing real-time monitoring video information acquired by the intelligent camera device based on a time sequence and can represent the all-angle information of the first GMP workshop. The video information is converted into real-time image data stored along with time sequence, so that the data is efficiently processed, a data base is provided for monitoring management based on the omnibearing image information, and the accuracy of the analysis of a subsequent model is further ensured.
S200: performing cluster analysis on the first image information set to obtain a first cluster result, wherein the first cluster result comprises first person image information and first equipment image information;
specifically, the first clustering result is obtained by performing clustering dimensionality reduction on the first image information set based on different attention elements, the original first image information set is redundant data derived from a plurality of different intelligent camera devices, and image information with a high similarity degree or a small information amount can be screened out through clustering analysis, so that redundancy is reduced. Further, clustering results of different attention elements are obtained, which are as follows: the first person image information whose attention element is a crowd, the first device image information whose attention element is a device, and the important region data information whose attention element is an important region: vents, safety channels, production areas, etc. After the first image information set is clustered according to different attention elements, the images corresponding to the clusters can be called quickly only by reading the attention elements during subsequent model analysis, and all image data do not need to be traversed, so that the data processing efficiency is improved.
S300: acquiring a first encryption parameter, and constructing a first anomaly identification model based on the first image information set and the first encryption parameter, wherein the first encryption parameter is acquired through multi-party GMP workshop image information aggregation analysis;
specifically, the first encryption parameter refers to a parameter of an image recognition model which is provided by a third-party collaboration system and integrates monitoring management training of a plurality of same-type medicine enterprises based on a GMP (good manufacturing practice) workshop; through the interaction of model parameters among multiple enterprises, the data islands among the traditional enterprises are broken through, the data islands are guaranteed to be broken through in an encryption mode, the private data inside the enterprises cannot be leaked, and the safety of data interaction and the feasibility of a system are improved; the first anomaly identification model is an intelligent model based on neural network training, the neural network model is a multi-level intelligent model constructed by simulating human neurons, the learning ability is very strong, and each set of training data comprises: and determining various types of abnormal identification information based on the GMP workshop quality management standard by the first image information set. And after the output value of the model is converged, stopping training, extracting the model parameters, encrypting and uploading the model parameters to a third-party cooperative system, integrating the similar model parameters uploaded by a plurality of enterprises through the third-party cooperative system to obtain the first encryption parameters, returning the first encryption parameters to each enterprise, and updating the model by each enterprise according to the first encryption parameters to achieve the technical effect of improving the accuracy of the processing result.
S400: inputting the first person image information and the first equipment image information into the first anomaly identification model to obtain a first identification result;
specifically, since the elements in which the abnormal situation mainly occurs in the first GMP plant are people and equipment, the first person image information and the first equipment image information are input to the first abnormality recognition model to be recognized, abnormal behaviors in the first person image information and the first equipment image information are discriminated, and image data of the abnormal behaviors is used as the first recognition result; the abnormal behavior includes two examples without limitation: abnormal behavior of the personnel: stoop pickup, no passing of the hands, door opening interference, no wearing of an eye mask, sitting on the ground and a table, presence above a sterile object, and the like; equipment abnormality information: such as bottle breakage in a bottle washing machine, mismatch in the transport speed on the conveyor belt and instrument speed parameters, etc. The abnormal behavior information of different image data is represented based on the first recognition result, and the obtained recognition result is accurate due to the fact that the recognition model trained based on the multi-party big data has strong recognition capability.
S500: and managing the first GMP workshop according to the first identification result.
Specifically, after the first recognition result is obtained, the abnormal information represented by the first recognition result is processed one by one, for example, without limitation: when the bottle in the bottle washing machine is broken, alarming processing is carried out in time, and abnormal information is sent to relevant workers for adjustment; performing alarm processing in real time when the personnel behavior is abnormal; when the articles are found to be accumulated in a key area, such as an air outlet, an alarm is given and information is sent to relevant personnel for processing. Through the real-time monitoring of the first GMP workshop, abnormal behaviors are managed in time, and the quality standard of the first GMP workshop is ensured to meet the standard requirement.
Further, constructing a first anomaly identification model based on the first set of image information and the first encryption parameter, step S300 includes:
s310: obtaining a first cooperation system, and downloading a first initial model from the first cooperation system;
s320: inputting the first person image information and the first equipment image information into the first initial model for training to obtain a first training parameter;
s330: encrypting the first training parameter to obtain a first encryption result;
s340: uploading the first encryption result to the first cooperation system, and acquiring the first encryption parameter by the first cooperation system by aggregating the first encryption result and the second encryption result until a kth encryption result, wherein the first encryption result and the second encryption result until the kth encryption result are information provided by GMP workshops of K different factories, and K is a natural number greater than or equal to 2;
s350: and updating the first initial model by using the first encryption parameter to obtain the first anomaly identification model.
Specifically, steps S310 to S350 are a process of constructing the first anomaly identification model, where the first collaboration system, i.e., a third party assisting a plurality of participants in training an updated model, is involved; multiple parties, i.e., parties that provide data and use the model. The following describes the construction process of the model from the first collaboration system side and from the pharmaceutical factory side, respectively:
the first implementation mode comprises the following steps:
the first cooperation system distributes an original model, namely the first initial model, to each pharmaceutical factory, each pharmaceutical factory performs model training, receives model parameter encryption results after training from the first pharmaceutical factory, the second pharmaceutical factory, the third pharmaceutical factory to the Kth pharmaceutical factory is completed, and the first encryption result is obtained.
The second embodiment:
the method comprises the steps that a first pharmaceutical factory, a second pharmaceutical factory, a third pharmaceutical factory and a Kth pharmaceutical factory download an original model from a first cooperation system, namely a first initial model, use multiple groups of first person image information and first equipment image information to train the first initial model until the first initial model converges, extract model parameters and encrypt the model parameters, send the model parameters to the first cooperation system as a first encryption result, after receiving information that the first cooperation system aggregates a plurality of K training parameter models which are similar to those sent by the first pharmaceutical factory, namely the first abnormity identification model is updated, and when the model needs to be called, send request information to the first cooperation system, and cooperatively call the request information to process data. The intelligent model can be used for integrating multi-party data, increasing the data volume and achieving the technical effect of improving the accuracy of processing results
Further, based on the inputting of the first person image information and the first device image information into the first abnormality recognition model, a first recognition result is obtained, and step S400 includes:
s410: performing feature extraction on the first person image information to obtain first feature information;
s420: performing feature extraction on the first equipment image information to obtain second feature information;
s430: inputting the first characteristic information and the second characteristic information into the first anomaly identification model to obtain a first identification result, wherein the first identification result comprises first person anomaly information and first equipment anomaly information;
s440: and managing the first GMP workshop according to the first person abnormal information and the first equipment abnormal information.
Specifically, the first person image information and the first device image information need to be preprocessed before being input into the first abnormality recognition model: the first feature information is a result obtained by extracting the first person image information and the behavior features related to the behavior norms based on the behavior actions related to the behavior norms established by the persons in the first GMP workshop, and a preferred feature extraction model based on convolutional neural network training is used for feature extraction, and convolution can be used as a feature extractor in machine learning, so that the extracted feature information has concentration and representativeness; the second characteristic information is extracted from the equipment image on the production line based on the production flow and the running state, and the extraction mode is preferably the same as the first characteristic information; further, the first recognition result can be obtained by inputting the first feature information and the second feature information into the first anomaly recognition model, and the first person anomaly information and the first equipment anomaly information are obtained correspondingly because the feature information of the first person image information and the first equipment image information is input; further, the management of the personnel and the equipment is performed based on the first person abnormality information and the first equipment abnormality information, respectively, in the following manner: and early warning, management and record storage. Through carrying out feature extraction on the first person image information and the first equipment image information based on the established behavior specifications in the first GMP workshop, when the first identification model is input, the redundancy of data can be reduced, the identification efficiency is improved, and the technical purpose of real-time monitoring is ensured.
Further, based on the feature extraction of the first person image information, obtaining first feature information, step S410 further includes:
s411: acquiring a first key area according to the first image information set;
s412: matching the first person image information with the first key area to obtain first key area person image information;
s413: performing personnel behavior feature extraction on the first key area personnel image information to obtain first personnel behavior feature information;
s414: extracting the characteristics of the personnel position of the personnel image information in the first key area to obtain the characteristic information of the first personnel position;
s415: performing characteristic extraction of personnel number on the personnel image information of the first key area to obtain characteristic information of the first personnel number;
s416: and taking the first person behavior characteristic information, the first person position characteristic information and the first person quantity characteristic information as the first characteristic information.
Specifically, the first key area is an area with limits on behaviors, identities and quantity of people, two areas without limits such as a canning room are lifted, and the number of people entering and exiting the box area is strictly limited; some experimental areas are forbidden to come in and go out of irrelevant people, and have larger requirements on identity rights. The first key area personnel image information is used for screening out the first personnel image information based on the first key area, screening out personnel images which do not belong to the first key area, and avoiding interference of irrelevant data; the first person behavior feature information is a result obtained by performing feature extraction on person action information in the person image information of the first key area based on a convolutional neural network; the first person position characteristic information is position information of each person in the first important area, which is preferably determined by constructing a spatial grid coordinate diagram in the first GMP workshop; the first person number characteristic information is a result obtained by performing feature extraction on the person number information in the person image information of the first key area based on a convolutional neural network. Further, the first person behavior feature information, the first person position feature information and the first person quantity feature information are used as the first feature information, when the first feature information is input into the first abnormality identification model, one or more of conditions that a person behavior violation occurs in the first key area, a person crosses a border, and the number of persons exceeds a preset value are obtained, early warning is performed, and a product control department is notified preferentially to process the conditions, so that the technical effect of intelligently managing the first GMP workshop is achieved.
Further, based on the feature extraction performed on the first device image information to obtain second feature information, step S420 further includes:
s421: acquiring a first key equipment image according to the first image information set;
s422: acquiring the operating parameters of the first key equipment to obtain the real-time working parameter characteristic information of the first key equipment;
s423: performing feature extraction on the operating state of the first key equipment based on the first key equipment image to obtain operating state feature information of the first key equipment;
s424: and taking the real-time working parameter characteristic information and the running state characteristic information of the first key equipment as the second characteristic information.
Specifically, the first key equipment image is an equipment instrument with a large potential safety hazard or a large effect in a production process or a parameter state which is difficult to control; preferably, the equipment including the first key equipment is in communication connection with the GMP workshop intelligent monitoring management system, and the operation parameters of the first key equipment can be read in real time and used as the real-time working parameter characteristic information of the first key equipment; further, the real-time operating parameter characteristic information is matched with the historical normal state information of the first device, and the matching result is taken as a standard operating state, including but not limited to: conveying information such as product conveying speed and yield; the operating state characteristic information of the first key equipment is data representing the abnormal degree of the operating state of the first key equipment, which is obtained by comparing the operating state of the first key equipment acquired in real time with a standard operating state, and the larger the difference degree of the comparison result is, the larger the abnormal degree of the operating state of the first key equipment is. Furthermore, the real-time working parameter characteristic information and the running state characteristic information of the first key equipment are used as the second characteristic information, the abnormal information of the first key equipment is obtained after the first abnormal recognition model is input, and the reason of the abnormality can be quickly traced through accurate positioning and the specific situation of the abnormal information, so that the adjustment is convenient. The processing mode, the abnormal condition and the abnormal factor are preferably stored correspondingly, so that the root tracing can be quickly carried out when similar abnormal information is encountered in the later step, and the reason can be found and processed.
Further, as shown in fig. 2, the method further includes step S600:
s610: acquiring a second key area according to the first image information set;
s620: extracting material characteristics of the image information of the second key area to obtain first material characteristic information;
s630: adjusting the first identification result based on the first material characteristic information to obtain a second identification result, wherein the second identification result comprises first material abnormal information;
s640: and managing the first GMP workshop according to the first material abnormity information.
Specifically, the second important area is an area sensitive to material accumulation, and may or may not overlap with the first important area; the first material characteristic information is characteristic information extracted from the image information of the second key area based on material characteristics, and the first material characteristic information includes, but is not limited to: characteristic information such as type, quantity and position of the materials; when the first material characteristic information is input into the first anomaly identification model, and the first material anomaly information is determined, for example, waste is accumulated in a ventilation opening, so that ventilation is influenced. And adjusting the first identification result through the first material characteristic information to obtain the second identification result. Further, based on the second recognition result, elements such as abnormal personnel, equipment, and materials may be managed, so as to ensure the quality standard of the first GMP workshop, and other elements may also be monitored and managed in the same manner, which is not limited herein.
Further, as shown in fig. 3, based on the system being further communicatively connected to an MES system, the method further includes step S700:
s710: obtaining first abnormal space-time information according to the first identification result;
s720: matching is carried out on the MES system according to the first abnormal spatiotemporal information to obtain first production data of the first GMP workshop, wherein the first production data and the spatiotemporal information of the first recognition result are the same;
s730: evaluating the relevance between the first identification result and the first production data to obtain a first relevance degree;
s740: obtaining a first preset association threshold, and judging whether the first association meets the first preset association threshold;
s750: and if the first correlation degree meets the first preset correlation degree threshold value, performing correlation identification on the first recognition result and the first production information, and uploading an identification result to the MES.
Specifically, the first abnormal spatio-temporal information is occurrence time node data of the abnormal information extracted from the first recognition result and position data in the first GMP workshop; the MES system is a functional module for quality management of the first GMP workshop, and is used together with the first anomaly identification to manage the quality information of the first GMP workshop; the first production data of the first GMP workshop are various types of production data of corresponding space and time obtained by matching and extracting in the MES system based on the time node and the position information which are the same as the first abnormal space and time information, and include but are not limited to: information such as order sheet, production quality, raw materials, production plan, deviation event and the like; the first association degree is data for evaluating the association degree of the first production data and the abnormal information extracted from the first identification result, the specific association degree is determined according to the actual production condition, and is not limited herein, when the association degree is higher, the corresponding first association degree is higher, and the influence of the abnormal information extracted from the first identification result on the first production data is represented to be larger; the first relevance threshold is the lowest first relevance value at which abnormal information extracted from the preset first identification result has great influence on first production data; further, the first relevance threshold is compared with the first relevance, if the first relevance is greater than or equal to the first preset relevance threshold, the first relevance meets the first preset relevance threshold, and the first recognition result and the first production information are subjected to simultaneous identification and fed back to the MES system. The method and the system ensure that which type of deviation data of a batch of medicines occurs in a certain process can be inquired on the MES system, further the inquiry of event information including video information can be carried out, and the tracing of the production condition of each batch of products is also ensured.
To sum up, the form page design method and system for user-defined metadata provided by the embodiment of the present application have the following technical effects:
1. the embodiment of the application provides a GMP workshop intelligent monitoring management method and system, wherein an intelligent camera device is used for acquiring all-around image information of a GMP workshop, clustering the acquired images according to the type of the acquired images, and screening out redundant data to obtain a clustering result representing personnel images and equipment images; furthermore, an intelligent recognition model is trained based on encrypted data acquired by big data of a plurality of different enterprises, image abnormal data in a clustering result is analyzed by the trained intelligent model, and the GMP workshop is managed based on abnormal information. Training samples are increased by collecting data of multiple enterprises, model identification capacity is improved, data privacy among different enterprises is guaranteed by means of encryption training, management is carried out according to abnormal data, automatic management is achieved, and the technical effect of improving GMP workshop management intellectualization is achieved.
2. And simultaneously identifying the first identification result and the first production information, and feeding back the identification result and the first production information to the MES system, so that the MES system can inquire which type of deviation data of a certain batch of medicines in a certain process, further inquire event information including video information, and ensure the tracing of the production condition of each batch of products.
Example two
Based on the same inventive concept as the GMP workshop intelligent monitoring and management method in the foregoing embodiment, as shown in fig. 4, an embodiment of the present application provides a GMP workshop intelligent monitoring and management system, where the system of the party includes:
a first obtaining unit 11, where the first obtaining unit 11 is configured to obtain, by using the smart camera device, a first image information set of a first GMP plant;
a second obtaining unit 12, where the second obtaining unit 12 is configured to perform cluster analysis on the first image information set to obtain a first clustering result, where the first clustering result includes first person image information and first device image information;
a third obtaining unit 13, where the third obtaining unit 13 is configured to obtain a first encryption parameter, and construct a first anomaly identification model based on the first image information set and the first encryption parameter, where the first encryption parameter is obtained through multi-party GMP plant image information aggregation analysis;
a fourth obtaining unit 14, where the fourth obtaining unit 14 is configured to input the first person image information and the first device image information into the first abnormality recognition model, and obtain a first recognition result;
a first management unit 15, where the first management unit 15 is configured to manage the first GMP workshop according to the first identification result.
Further, the system further comprises:
the first downloading unit is used for acquiring a first collaboration system and downloading a first initial model from the first collaboration system;
the first training unit is used for inputting the first person image information and the first equipment image information into the first initial model for training to obtain a first training parameter;
the first encryption unit is used for encrypting the first training parameter to obtain a first encryption result;
a fifth obtaining unit, configured to upload the first encryption result to the first collaboration system, where the first collaboration system obtains the first encryption parameter by aggregating the first encryption result and the second encryption result until a kth encryption result, where the first encryption result, the second encryption result until the kth encryption result are information provided by GMP plants of K different plants, and K is a natural number greater than or equal to 2;
a sixth obtaining unit, configured to update the first initial model using the first encryption parameter, and obtain the first anomaly identification model.
Further, the system further comprises:
a seventh obtaining unit, configured to perform feature extraction on the first person image information to obtain first feature information;
an eighth obtaining unit, configured to perform feature extraction on the first device image information to obtain second feature information;
a ninth obtaining unit, configured to input the first feature information and the second feature information into the first abnormality recognition model, and obtain a first recognition result, where the first recognition result includes first person abnormality information and first device abnormality information;
and the second management unit is used for managing the first GMP workshop according to the first person abnormal information and the first equipment abnormal information.
Further, the system further comprises:
a tenth obtaining unit, configured to obtain a first key area according to the first image information set;
an eleventh obtaining unit, configured to match the first person image information with the first key region to obtain first key region person image information;
a twelfth obtaining unit, configured to perform feature extraction of a person behavior on the first key region person image information, and obtain first person behavior feature information;
a thirteenth obtaining unit, configured to perform feature extraction on the position of the person in the first key area to obtain first person position feature information;
a fourteenth obtaining unit, configured to perform feature extraction on the number of people from the first key area person image information to obtain first person number feature information;
a first setting unit configured to take the first person behavior feature information, the first person position feature information, and the first person quantity feature information as the first feature information.
Further, the system further comprises:
a fifteenth obtaining unit, configured to obtain a first key device image according to the first image information set;
a sixteenth obtaining unit, configured to collect the operation parameters of the first key device, and obtain real-time working parameter feature information of the first key device;
a seventeenth obtaining unit, configured to perform feature extraction on the operation state of the first key apparatus based on the first key apparatus image, and obtain operation state feature information of the first key apparatus;
and the second setting unit is used for taking the real-time working parameter characteristic information and the running state characteristic information of the first key equipment as the second characteristic information.
Further, the system further comprises:
an eighteenth obtaining unit, configured to obtain a second important region according to the first image information set;
a nineteenth obtaining unit, configured to perform material feature extraction on the image information of the second important region to obtain first material feature information;
a twentieth obtaining unit, configured to adjust the first identification result based on the first material characteristic information to obtain a second identification result, where the second identification result includes first material abnormality information;
and the third management unit is used for managing the first GMP workshop according to the first material abnormal information.
Further, the system further comprises:
a twenty-first obtaining unit, configured to obtain first abnormal spatiotemporal information according to the first recognition result;
a twenty-second obtaining unit, configured to perform matching in the MES system according to the first abnormal spatio-temporal information, and obtain first production data of the first GMP workshop, where the first production data is the same as the spatio-temporal information of the first recognition result;
a twenty-third obtaining unit configured to evaluate a correlation between the first recognition result and the first production data, and obtain a first degree of correlation;
the first judging unit is used for obtaining a first preset association threshold value and judging whether the first association meets the first preset association threshold value or not;
and the first execution unit is used for performing association identification on the first recognition result and the first production information if the first association degree meets the first preset association degree threshold value, and uploading the identification result to the MES system.
Exemplary electronic device
The electronic device of the embodiment of the present application is described below with reference to figure 5,
based on the same inventive concept as the GMP workshop intelligent monitoring management method in the foregoing embodiment, the embodiment of the present application further provides a GMP workshop intelligent monitoring management system, including: a processor coupled to a memory for storing a program that, when executed by the processor, causes a system to perform the method of any of the first aspects
The electronic device 300 includes: processor 302, communication interface 303, memory 301. Optionally, the electronic device 300 may also include a bus architecture 304. Wherein the communication interface 303, the processor 302 and the memory 301 may be interconnected by a bus architecture 304; the bus architecture 304 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus architecture 304 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 5, but this is not intended to represent only one bus or type of bus.
Processor 302 may be a CPU, microprocessor, ASIC, or one or more integrated circuits for controlling the execution of programs in accordance with the teachings of the present application.
The communication interface 303 is a system using any transceiver or the like, and is used for communicating with other devices or communication networks, such as ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), wired access network, and the like.
The memory 301 may be, but is not limited to, a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, an electrically erasable Programmable read-only memory (EEPROM), a compact disk read-only memory (CD-ROM) or other optical disk storage, optical disk storage (including compact disk, laser disk, optical disk, digital versatile disk, blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. The memory may be self-contained and coupled to the processor through a bus architecture 304. The memory may also be integral to the processor.
The memory 301 is used for storing computer-executable instructions for executing the present application, and is controlled by the processor 302 to execute. The processor 302 is configured to execute the computer executable instructions stored in the memory 301, so as to implement the GMP plant intelligent monitoring management method provided by the above-mentioned embodiment of the present application.
Optionally, the computer-executable instructions in the embodiments of the present application may also be referred to as application program codes, which are not specifically limited in the embodiments of the present application.
The embodiment of the application provides a GMP workshop intelligent monitoring management method and system, wherein an intelligent camera device is used for acquiring all-around image information of a GMP workshop, clustering the acquired images according to the type of the acquired images, and screening out redundant data to obtain a clustering result representing personnel images and equipment images; furthermore, an intelligent recognition model is trained based on encrypted data acquired by big data of a plurality of different enterprises, image abnormal data in a clustering result is analyzed by the trained intelligent model, and the GMP workshop is managed based on abnormal information. Training samples are increased by collecting data of multiple enterprises, model identification capacity is improved, data privacy among different enterprises is guaranteed by means of encryption training, management is carried out according to abnormal data, automatic management is achieved, and the technical effect of improving GMP workshop management intellectualization is achieved.
Those of ordinary skill in the art will understand that: the various numbers of the first, second, etc. mentioned in this application are only used for the convenience of description and are not used to limit the scope of the embodiments of this application, nor to indicate the order of precedence. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., A and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one" means one or more. At least two means two or more. "at least one," "any one," or similar expressions refer to any combination of these items, including any combination of item(s) or items. For example, at least one (one ) of a, b, or c, may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or multiple.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the application to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable system. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device including one or more available media integrated servers, data centers, and the like. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
The various illustrative logical units and circuits described in this application may be implemented or operated upon by general purpose processors, digital signal processors, Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other programmable logic systems, discrete gate or transistor logic, discrete hardware components, or any combination thereof. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing systems, e.g., a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital signal processor core, or any other similar configuration.
The steps of a method or algorithm described in the embodiments herein may be embodied directly in hardware, in a software element executed by a processor, or in a combination of the two. The software cells may be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. For example, a storage medium may be coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC, which may be disposed in a terminal. In the alternative, the processor and the storage medium may reside in different components within the terminal. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Although the present application has been described in conjunction with specific features and embodiments thereof, it will be evident that various modifications and combinations may be made thereto without departing from the spirit and scope of the application. Accordingly, the specification and figures are merely exemplary of the present application as defined in the appended claims and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations.