Laboratory management platform based on large model and visualization technology

By adopting large-scale models and visualization technologies in the laboratory management platform, combining image recognition and 3D visualization technologies, the problem of lack of natural language processing and intelligent interaction functions in the existing technology is solved, and efficient management and resources of the laboratory environment are achieved. Reasonable allocation to ensure safety and efficiency in the laboratory.

CN119048971APending Publication Date: 2024-11-29ANHUI HEXIN TECH DEV

Patent Information

Application Number
CN202410945977.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-15
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The existing laboratory management platform lacks natural language processing capabilities and intelligent interaction functions, and resource scheduling relies on static rules. The 2D visual interface is limited in terms of intuitiveness and interactivity, making it difficult to effectively manage laboratory environment and resources.

Method used

Using a laboratory management platform based on large models and visualization technology, including environmental acquisition module, image acquisition module, data analysis module, 3D visualization module and early warning module, the monitoring images and environmental information in the laboratory are judged through image recognition and analysis, and the monitoring images and environmental information in the laboratory are used. 3D visualization technology displays the analysis results and issues early warnings when abnormalities are abnormal.

Benefits of technology

It realizes effective management of surveillance images and environment in the laboratory, quickly and accurately determines whether the laboratory staff privately store experimental materials, and promptly warns through 3D visualization technology to ensure the reasonable allocation and use of laboratory resources, and improves the laboratory Management efficiency and security.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119048971A_ABST
    Figure CN119048971A_ABST
Patent Text Reader

Abstract

The invention discloses a laboratory management platform based on a large model and a visualization technology, which belongs to the field of laboratory management platforms and comprises an environment acquisition module, an image acquisition module, a data analysis module, a 3D visualization module and an early warning module. The image acquisition module is used for acquiring monitoring image information in a laboratory; the environment acquisition module is used for acquiring temperature and humidity information and noise information in a laboratory; the data analysis module is used for analyzing according to the collected information and outputting an analysis result; the 3D visualization module is used for displaying an analysis result for a platform manager to check; and the early warning module is used for deciding whether to give out early warning according to the analysis result. According to the invention, the monitoring image and environment in the laboratory can be effectively managed, the analysis result can be displayed through the 3D visualization technology, and early warning can be carried out in time when abnormity occurs.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of laboratory management platforms, in particular to laboratory management platforms based on large models and visualization technologies. Background Art

[0002] The laboratory management platform is a management system based on information technology that integrates laboratory resource management, security monitoring, data analysis and other functions. It is an important tool for laboratory management, aiming to improve the laboratory's operational efficiency, achieve optimal resource allocation, and promote the laboratory's innovative development.

[0003] In order to achieve more comprehensive and efficient management of the laboratory, the laboratory management platform can be combined with large models and visualization technology. The large model is an important model in the field of AI, usually referring to a deep learning model with a very large number of parameters and data. Visualization technology is a theory, method and technology that uses computer graphics and image processing technology to convert data into graphics or images and display them on the screen for interactive processing.

[0004] Existing laboratory management platforms usually consist of multiple subsystems, including but not limited to laboratory information management system, laboratory resource management system, laboratory visualization management system, etc.

[0005] Laboratory information management systems usually use database management and web-based user interfaces to help laboratories manage various assets such as laboratory equipment, security equipment, and environmental monitoring equipment. Queries and operations are mainly performed through menus and forms, lacking natural language processing capabilities and intelligent interactive functions; laboratory resource management systems rely on manual input and scheduling, using database and spreadsheet tools to record and manage equipment, reagents, and consumables. Resource scheduling is based on static rules and pre-set schedules, and lacks intelligent prediction and optimization algorithms; laboratory visualization management systems generally use a 2D interface, using charts, dashboards, and reports to display data. 2D visualization can provide a certain degree of data display and analysis, but has significant limitations in intuitiveness and interactivity.

[0006] Therefore, those skilled in the art provide a laboratory management platform based on large models and visualization technology to solve the problems raised in the above background technology. Summary of the invention

[0007] The purpose of the present invention is to provide a laboratory management platform based on large models and visualization technology, which can effectively manage the monitoring images and environment in the laboratory, display the analysis results through 3D visualization technology, and issue timely warnings when abnormalities occur, so as to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] The laboratory management platform based on large models and visualization technology includes environment acquisition module, image acquisition module, data analysis module, 3D visualization module and early warning module;

[0010] The image acquisition module is used to collect monitoring image information in the laboratory; the environment acquisition module is used to collect temperature, humidity and noise information in the laboratory; the data analysis module is used to analyze the collected information and output the analysis results; the 3D visualization module is used to display the analysis results for platform managers to view; the early warning module is used to decide whether to issue an early warning based on the analysis results.

[0011] As a further solution of the present invention: the specific process of the data analysis module analyzing the collected laboratory monitoring image information is as follows:

[0012] Step 1: Perform image recognition on the collected images to determine whether the staff in the laboratory are facing away from the surveillance camera and blocking the experimental materials. If so, proceed to the next step. If not, output the analysis result that the surveillance image is normal;

[0013] Step 2: Construct an experimental material analysis and judgment model to output the judgment result on whether the material should be used;

[0014] Step 3: Use image recognition analysis to determine whether the staff member has experimental materials in his hand after turning around. If there are experimental materials, count the amount and type of the experimental materials;

[0015] Step 4: Based on the analysis and judgment results of steps 2 and 3, assess whether the staff has privately hidden the experimental materials;

[0016] Step 5: If the staff member does not hide the experimental materials, the analysis result is output as normal monitoring images. If the staff member hides the experimental materials, the analysis result is output as abnormal monitoring images and the risk level of hiding experimental materials is calculated.

[0017] As a further solution of the present invention: in step 2, the specific analysis and judgment process of the experimental material analysis and judgment model is:

[0018] S201: marking the monitoring image before the experimental material is blocked by the staff as P1, and marking the monitoring image after the experimental material is blocked by the staff as P2;

[0019] S202: Compare the similarity of the positions of the containers storing the experimental materials in P1 and P2. If the similarity value between the two is less than a preset value, it means that the placement of the experimental materials has changed during the process of being blocked by the staff; if the similarity value between the two is greater than or equal to the preset value, it means that the placement of the experimental materials has not changed during the process of being blocked by the staff;

[0020] S203: Perform image recognition on P1 and P2 to determine whether the experimental materials in the transparent container of P2 are reduced compared with those in P1. If there is a reduction, it means that the staff has taken the experimental materials in the transparent container. The amount of experimental materials taken from the transparent container is counted and marked as A1, and the type of experimental materials taken from the transparent container is marked as A2.

[0021] S204: extract features from P1 and P2, select opaque containers storing experimental materials in P1 and P2, and mark them as Bi and bi, respectively, where i=1···n, and n is a positive integer;

[0022] S205: Compare Bi with bi one by one, and use image processing software to perform pixel-level difference analysis on the color and shape of the container during the comparison process, and calculate the difference degree C;

[0023] S206: If the difference C is greater than the preset value c in the comparison between any item Bi and bi, it means that the staff has taken the experimental material in the opaque container;

[0024] S207: If the analysis result shows that the staff member has taken the experimental materials in the transparent container or the staff member has taken the experimental materials in the opaque container, it is determined that the staff member has taken the experimental materials.

[0025] As a further solution of the present invention: in step 4, the specific process of evaluating whether the staff member has privately kept the experimental materials is as follows:

[0026] If it is determined in step 3 that the staff member does not have any experimental materials in his hand after turning around, but it is determined in step 2 that the staff member has taken the experimental materials, it means that the staff member has kept the experimental materials privately;

[0027] If it is determined in step 3 that the staff member has experimental materials in his hand after turning around, and it is determined in step 2 that the staff member has taken the experimental materials, then proceed to the next step;

[0028] The amount of experimental materials counted in step 3 is marked as a1, and the type of experimental materials counted in step 3 is marked as a2. If the staff member takes the experimental materials in the transparent container in step 2, a2 ​​is matched with A2. If the two do not match, it means that the staff member has privately kept the experimental materials. If the two match, proceed to the next step;

[0029] Compare a1 with A1. If the difference between the two is less than the preset value, it means that the staff did not hide the experimental materials privately. If the difference between the two is greater than or equal to the preset value, proceed to the next step.

[0030] The historical time taken by the staff to take the experimental materials with the amount a1 and the type a2 is marked as Ti, i=1···n, and the total time the staff turned their back to the surveillance camera to block the experimental materials is marked as t;

[0031] Calculate the difference between t and Ti. If any difference between Ti and t is within the preset time, it means that the staff member has not hidden the experimental materials. Conversely, if any difference between Ti and t is not within the preset time, it means that the staff member has hidden the experimental materials.

[0032] As a further solution of the present invention: in the step five, the specific process of calculating the risk level of secret possession of experimental materials is: if the staff member secretes the experimental materials, and the placement of the experimental materials has not changed during the process of being blocked by the staff member, the output risk level is level two; if the staff member secretes the experimental materials, and the placement of the experimental materials has changed during the process of being blocked by the staff member, the output risk level is level one.

[0033] As a further solution of the present invention: the specific process of the data analysis module analyzing the collected temperature information in the laboratory is:

[0034] The collected temperatures of various areas in the laboratory are marked as Di, i = 1···n;

[0035] Then the temperature difference between two adjacent areas is marked as di, i = 1···n-1;

[0036] The temperature of each area of ​​the laboratory at the same time point on the previous day is marked as Ei, i = 1···n;

[0037] Then, the temperature difference between two adjacent areas of the laboratory at the same time point the previous day is marked as ei, i = 1···n-1;

[0038] Compare Di and Ei one by one. If the difference between the two is greater than the preset value, the corresponding Di is planned to the first sequence;

[0039] Compare di and ei one by one. If the difference between the two is greater than the preset value, the corresponding di will be planned to the second sequence;

[0040] Count the number of all Di in the first sequence and mark it as X1, then count the number of all di in the second sequence and mark it as Y1;

[0041] Calculate the temperature anomaly weight value Z1 = 60% * X1 + 40% * Y1;

[0042] Compare Z1 with the preset value z1. If Z1>z1, it means that the laboratory temperature is abnormal. If Z1≤z1, it means that the laboratory temperature is normal.

[0043] As a further solution of the present invention: the specific process of the data analysis module analyzing the collected humidity information in the laboratory is:

[0044] The collected humidity of each area in the laboratory is marked as Fi, i = 1···n;

[0045] Then the humidity difference between two adjacent areas is marked as fi, i = 1···n-1;

[0046] The humidity in each area of ​​the laboratory at the same time point on the previous day is marked as Gi, i = 1···n;

[0047] Then, the humidity difference between two adjacent areas of the laboratory at the same time point the previous day is marked as gi, i = 1···n-1;

[0048] Compare Fi and Gi one by one. If the difference between the two is greater than the preset value, the corresponding Fi will be planned to the third sequence;

[0049] Compare fi and gi one by one. If the difference between them is greater than the preset value, the corresponding fi will be planned to the fourth sequence;

[0050] Count the number of all Fi in the third sequence and mark it as X2, then count the number of all fi in the fourth sequence and mark it as Y2;

[0051] Calculate the humidity anomaly weight value Z2 = 60% * X2 + 40% * Y2;

[0052] Compare Z2 with the preset value z2. If Z2>z2, it means that the laboratory humidity is abnormal. If Z2≤z2, it means that the laboratory humidity is normal.

[0053] As a further solution of the present invention: the specific process of the data analysis module analyzing the noise information collected in the laboratory is:

[0054] The noise volume collected in real time is marked as H, and the location of the noise source is marked as Q;

[0055] Compare H with the preset noise volume h. If H>h, it means that the noise in the laboratory is abnormal. If H≤h, it means that the noise in the laboratory is normal.

[0056] According to the location Q of the noise source, it is judged whether the noise is generated in the laboratory. If so, it means that the noise in the laboratory is abnormal. If not, it means that the noise outside the laboratory is abnormal.

[0057] As a further solution of the present invention: the 3D visualization module includes a 3D holographic projection device, and the 3D image projected by the 3D holographic projection device is divided into four areas, which are respectively used to display the analysis results of the monitoring image information in the laboratory, the analysis results of the temperature information in the laboratory, the analysis results of the humidity information in the laboratory, and the analysis results of the noise information in the laboratory.

[0058] As a further solution of the present invention: the specific warning process of the early warning module is: if there are abnormalities in the analysis results of the monitoring image information in the laboratory, the analysis results of the temperature information in the laboratory, the analysis results of the humidity information in the laboratory, and the analysis results of the noise information in the laboratory, the 3D image area corresponding to the abnormal analysis result flashes red.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] 1. This application can effectively manage the monitoring images and environment in the laboratory, and display the analysis results through 3D visualization technology, and issue timely warnings when abnormalities occur. Among them, for the private hiding of experimental materials in the laboratory, this application uses the collected monitoring images to analyze and judge from multiple angles, and then quickly and accurately determine whether the staff in the laboratory have privately hidden experimental materials. In addition, for the temperature and humidity information in the laboratory, from the real-time collected temperature and humidity and the temperature and humidity difference of adjacent areas, combined with previous records, it is possible to quickly and effectively determine whether there are abnormalities in the temperature and humidity in the laboratory. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is the structural block diagram of the laboratory management platform based on big model and visualization technology. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] As mentioned in the background technology of this application, research has found that existing laboratory visualization management systems generally use a 2D interface and use charts, dashboards and reports to display data. 2D visualization can provide a certain degree of data display and analysis, but it has great limitations in intuitiveness and interactivity and has certain defects.

[0064] In order to solve the above-mentioned defects, the present application discloses a laboratory management platform based on large models and visualization technology, which can effectively manage the monitoring images and environment in the laboratory, display the analysis results through 3D visualization technology, and provide timely warnings when abnormalities occur.

[0065] The following will describe in detail how the solution of the present application solves the above technical problems in conjunction with the accompanying drawings.

[0066] See also Figure 1 In the embodiment of the present invention, the laboratory management platform based on the large model and visualization technology includes an environment acquisition module, an image acquisition module, a data analysis module, a 3D visualization module and an early warning module; the image acquisition module is used to collect monitoring image information in the laboratory; the environment acquisition module is used to collect temperature, humidity and noise information in the laboratory; the data analysis module is used to analyze the collected information and output the analysis results; the 3D visualization module is used to display the analysis results for the platform management personnel to view; the early warning module is used to decide whether to issue an early warning based on the analysis results. This application can effectively manage the monitoring images and environment in the laboratory, and display the analysis results through 3D visualization technology, and issue early warnings in time when abnormalities occur.

[0067] In this embodiment, the specific process of the data analysis module analyzing the collected laboratory monitoring image information is as follows: Step 1: Perform image recognition on the collected image to determine whether the staff in the laboratory is facing away from the monitoring camera to block the experimental materials. If so, proceed to the next step. If not, output the analysis result that the monitoring image is normal; Step 2: Construct an experimental material analysis and judgment model to output the judgment result of whether the material is taken; Step 3: Use image recognition analysis to determine whether there are experimental materials in the hands of the staff after turning around. If there are experimental materials, count the amount and type of the experimental materials; Step 4: According to the analysis and judgment results of Steps 2 and 3, evaluate whether the staff has privately hidden experimental materials; Step 5: If the staff does not privately hide the experimental materials, output the analysis result that the monitoring image is normal. If the staff privately hides the experimental materials, output the analysis result that the monitoring image is abnormal and calculate the risk level of privately hidden experimental materials. Privately hiding experimental materials will lead to unreasonable allocation and use of laboratory resources, thereby affecting the normal progress of other scientific research projects. The public materials of the laboratory should be reasonably allocated and used to ensure the smooth progress of scientific research. In addition, the behavior of privately hiding experimental materials will disrupt the management order of the laboratory and lead to chaotic laboratory management. Laboratories should establish sound management systems and supervision mechanisms to prevent similar behaviors from happening. In public experiments, keeping experimental materials privately will destroy the atmosphere of teamwork and the basis of trust, affecting the overall effectiveness of the scientific research team and the output of scientific research results.

[0068] In this embodiment, in step 2, the specific analysis and judgment process of the experimental material analysis and judgment model is as follows: S201: mark the monitoring image before the experimental material is blocked by the staff as P1, and then mark the monitoring image after the experimental material is blocked by the staff as P2; S202: compare the similarity of the positions of the containers storing the experimental materials in P1 and P2. If the similarity value between the two is less than the preset value, it means that the placement of the experimental materials has changed during the process of being blocked by the staff; if the similarity value between the two is greater than or equal to the preset value, it means that the placement of the experimental materials has not changed during the process of being blocked by the staff; S203: perform image recognition on P1 and P2 to determine whether there is a decrease in the experimental materials in the transparent container in P2 compared to P1. If there is a decrease, it means that the staff has taken the experimental materials in the transparent container, and the number of experimental materials in the transparent container that have been blocked is counted. The amount taken is marked as A1, and the type of experimental material taken from the transparent container is marked as A2; S204: extract features from P1 and P2, screen out the opaque containers storing experimental materials in P1 and P2, and mark them as Bi and bi respectively, i=1···n, where n is a positive integer; S205: compare Bi and bi one by one, and during the comparison process, use image processing software to perform pixel-level difference analysis on the color and shape of the container, and calculate the difference C; S206: if the difference C is greater than the preset value c in any comparison between Bi and bi, it means that the staff has taken the experimental materials in the opaque container; S207: if the analysis result shows that the staff has taken the experimental materials in the transparent container or the staff has taken the experimental materials in the opaque container, it is judged that the staff has taken the experimental materials. This setting can quickly and accurately determine whether the staff has taken the experimental materials. It should be noted that common image processing software includes Beyond Compare, DiffImg, imgdiff, and Photoshop (Adobe Photoshop). Among them, Beyond Compare is not only a powerful file comparison tool, but also has image comparison function. It can achieve pixel-level detail processing, display differences through grayscale images, fast speed and small memory usage. In addition, it also provides simplified control, fast processing capabilities, and can easily generate comparison reports to meet the needs of designers and ordinary users. DiffImg is a lightweight software whose main function is to compare two RGB images of the same size and display the difference results. It can quickly obtain the differences between two images through automatic analysis of software algorithms. Although the function is relatively simple, it is suitable for quick comparison. imgdiff is an open source, lightweight but efficient command line tool for comparing the pixel differences between two images. Its core goal is to quickly and accurately identify and highlight small changes between images.Compared with existing tools, imgdiff excels in performance, especially when processing large images. In addition, it provides a wealth of parameter options, including color difference threshold adjustment, output image mode setting, etc. Although Photoshop is primarily an image processing software, it also has some tools and techniques for comparing image differences. For example, you can use the "History" panel to view the difference before and after an image is edited, or use the "Difference" blending mode to compare the difference between two layers. However, these features are not specifically designed for pixel-level difference analysis.

[0069] In this embodiment, in step 4, the specific process of evaluating whether the staff member has privately kept the experimental materials is as follows: if it is determined in step 3 that there are no experimental materials in the hands of the staff member after turning around, but it is determined in step 2 that the staff member has taken the experimental materials, it means that the staff member has privately kept the experimental materials; if it is determined in step 3 that there are experimental materials in the hands of the staff member after turning around, and it is determined in step 2 that the staff member has taken the experimental materials, then proceed to the next step; mark the amount of experimental materials counted in step 3 as a1, and then mark the type of experimental materials counted in step 3 as a2; if the staff member has taken the experimental materials in the transparent container in step 2, then match a2 with A2; if the two do not match, it means that the staff member has privately kept the experimental materials. Materials, if the two match, proceed to the next step; compare a1 and A1, if the difference between the two is less than the preset value, it means that the staff has not hidden the experimental materials, if the difference between the two is greater than or equal to the preset value, proceed to the next step; mark the historical time the staff has used to take the experimental materials with a quantity of a1 and a type of a2 as Ti, i=1···n, and then mark the total time the staff has turned their backs to the surveillance camera to block the experimental materials as t; calculate the difference between t and Ti, if any difference between Ti and t is within the preset time, it means that the staff has not hidden the experimental materials, conversely, if any difference between Ti and t is not within the preset time, it means that the staff has hidden the experimental materials. This application can quickly and accurately determine whether the staff has hidden the experimental materials by comparing the experimental materials and their containers before and after.

[0070] In this embodiment, in step 5, the specific process of calculating the risk level of hiding experimental materials is as follows: if the staff hides the experimental materials, and the placement of the experimental materials does not change during the process of being blocked by the staff, the output risk level is level 2; if the staff hides the experimental materials, and the placement of the experimental materials changes during the process of being blocked by the staff, the output risk level is level 1. If the staff wants to keep the placement of the experimental materials in the previous position after hiding the experimental materials, it means that the staff has been intentional for a long time, rather than acting on impulse, and the harm caused is greater.

[0071] In this embodiment, the specific process of the data analysis module analyzing the collected temperature information in the laboratory is as follows: the collected temperatures of each area in the laboratory are marked as Di, i=1···n; the temperature difference between two adjacent areas is marked as di, i=1···n-1; the temperatures of each area of ​​the laboratory at the same time point on the previous day are marked as Ei, i=1···n; the temperature difference between two adjacent areas of the laboratory at the same time point on the previous day is marked as ei, i=1···n-1; Di is compared with Ei one by one. Yes, if the difference between the two is greater than the preset value, the corresponding Di will be planned to the first sequence; di and ei are compared one by one, if the difference between the two is greater than the preset value, the corresponding di will be planned to the second sequence; the number of all Di in the first sequence is counted and marked as X1, and then the number of all di in the second sequence is counted and marked as Y1; the temperature anomaly weight value Z1=60%*X1+40%*Y1 is calculated; Z1 is compared with the preset value z1, if Z1>z1, it means that the laboratory temperature is abnormal, if Z1≤z1, it means that the laboratory temperature is normal. For the temperature information in the laboratory, from the real-time collected temperature and the temperature difference of the adjacent areas, combined with previous records, it is quickly and effectively determined whether the temperature in the laboratory is abnormal.

[0072] In this embodiment, the specific process of the data analysis module analyzing the collected humidity information in the laboratory is as follows: the collected humidity of each area in the laboratory is marked as Fi, i=1···n; the humidity difference between two adjacent areas is marked as fi, i=1···n-1; the humidity of each area of ​​the laboratory at the same time point on the previous day is marked as Gi, i=1···n; the humidity difference between two adjacent areas of the laboratory at the same time point on the previous day is marked as gi, i=1···n-1; Fi is compared with Gi one by one Yes, if the difference between the two is greater than the preset value, the corresponding Fi will be planned to the third sequence; compare fi with gi one by one, if the difference between the two is greater than the preset value, the corresponding fi will be planned to the fourth sequence; count the number of all Fi in the third sequence and mark it as X2, and then count the number of all fi in the fourth sequence and mark it as Y2; calculate the humidity anomaly weight value Z2=60%*X2+40%*Y2; compare Z2 with the preset value z2, if Z2>z2, it means that the laboratory humidity is abnormal, if Z2≤z2, it means that the laboratory humidity is normal. For the humidity information in the laboratory, from the real-time collected humidity and the humidity difference of adjacent areas, combined with previous records, it is possible to quickly and effectively determine whether the humidity in the laboratory is abnormal.

[0073] In this embodiment, the specific process of the data analysis module analyzing the noise information collected in the laboratory is as follows: mark the real-time collected noise volume as H, and then mark the location of the noise source as Q; compare H with the preset noise volume h, if H>h, it means that the noise in the laboratory is abnormal, if H≤h, it means that the noise in the laboratory is normal; determine whether the noise is generated in the laboratory according to the location Q of the noise source, if so, it means that the noise in the laboratory is abnormal, if not, it means that the noise outside the laboratory is abnormal. This setting can quickly determine whether the noise in the laboratory is abnormal.

[0074] In this embodiment, the 3D visualization module includes a 3D holographic projection device. The 3D image projected by the 3D holographic projection device is divided into four areas, which are used to display the analysis results of the monitoring image information in the laboratory, the analysis results of the temperature information in the laboratory, the analysis results of the humidity information in the laboratory, and the analysis results of the noise information in the laboratory. The 3D holographic projection device uses optical principles to suspend the 3D image in the air through air or special media to achieve a true three-dimensional display effect.

[0075] In this embodiment, the specific warning process of the early warning module is: if there is an abnormality in the analysis results of the monitoring image information in the laboratory, the analysis results of the temperature information in the laboratory, the analysis results of the humidity information in the laboratory, and the analysis results of the noise information in the laboratory, the 3D image area corresponding to the abnormal analysis result flashes red. This application uses a visual 3D holographic projection device to display the above analysis results in different regions, which is convenient for staff to view and can also issue early warnings in real time.

[0076] The present invention can effectively manage the monitoring images and environment in the laboratory, and display the analysis results through 3D visualization technology, and timely warn when an abnormality occurs. Among them, for the phenomenon of privately hiding experimental materials in the laboratory, this application uses the collected monitoring images to analyze and judge from multiple angles, and then quickly and accurately determine whether the staff in the laboratory have privately hidden experimental materials. In addition, for the temperature and humidity information in the laboratory, from the real-time collected temperature and humidity and the temperature and humidity difference of adjacent areas, combined with previous records, it is possible to quickly and effectively determine whether there is an abnormality in the temperature and humidity in the laboratory.

[0077] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

[0078] What is described above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A laboratory management platform based on large models and visualization technology, characterized by: It includes environment acquisition module, image acquisition module, data analysis module, 3D visualization module and early warning module; The image acquisition module is used to collect monitoring image information in the laboratory; the environment acquisition module is used to collect temperature, humidity and noise information in the laboratory; the data analysis module is used to analyze the collected information and output the analysis results; the 3D visualization module is used to display the analysis results for platform managers to view; the early warning module is used to decide whether to issue an early warning based on the analysis results.

2. The laboratory management platform based on large models and visualization technology according to claim 1, characterized in that: The specific process of the data analysis module analyzing the collected laboratory monitoring image information is as follows: Step 1: Perform image recognition on the collected images to determine whether the staff in the laboratory are facing away from the surveillance camera and blocking the experimental materials. If so, proceed to the next step. If not, output the analysis result that the surveillance image is normal; Step 2: Construct an experimental material analysis and judgment model to output the judgment result on whether the material should be used; Step 3: Use image recognition analysis to determine whether the staff member has experimental materials in his hand after turning around. If there are experimental materials, count the amount and type of the experimental materials; Step 4: Based on the analysis and judgment results of steps 2 and 3, assess whether the staff has privately hidden the experimental materials; Step 5: If the staff member does not hide the experimental materials, the analysis result is output as normal monitoring images. If the staff member hides the experimental materials, the analysis result is output as abnormal monitoring images and the risk level of hiding experimental materials is calculated.

3. The laboratory management platform based on large models and visualization technology according to claim 2 is characterized in that: In step 2, the specific analysis and judgment process of the experimental material analysis and judgment model is as follows: S201: marking the monitoring image before the experimental material is blocked by the staff as P1, and marking the monitoring image after the experimental material is blocked by the staff as P2; S202: Compare the similarity of the positions of the containers storing the experimental materials in P1 and P2. If the similarity value between the two is less than a preset value, it means that the placement of the experimental materials has changed during the process of being blocked by the staff; If the similarity value between the two is greater than or equal to the preset value, it means that the placement of the experimental materials has not changed during the process of being blocked by the staff; S203: Perform image recognition on P1 and P2 to determine whether the experimental materials in the transparent container of P2 are reduced compared with those in P1. If there is a reduction, it means that the staff has taken the experimental materials in the transparent container. The amount of experimental materials taken from the transparent container is counted and marked as A1, and the type of experimental materials taken from the transparent container is marked as A2. S204: extract features from P1 and P2, select opaque containers storing experimental materials in P1 and P2, and mark them as Bi and bi, respectively, where i=1···n, and n is a positive integer; S205: Compare Bi with bi one by one, and use image processing software to perform pixel-level difference analysis on the color and shape of the container during the comparison process, and calculate the difference degree C; S206: If the difference C is greater than the preset value c in the comparison between any item Bi and bi, it means that the staff has taken the experimental material in the opaque container; S207: If the analysis result shows that the staff member has taken the experimental materials in the transparent container or the staff member has taken the experimental materials in the opaque container, it is determined that the staff member has taken the experimental materials.

4. The laboratory management platform based on large model and visualization technology according to claim 3 is characterized in that: In step 4, the specific process of evaluating whether the staff member has privately kept the experimental materials is as follows: If it is determined in step 3 that the staff member does not have any experimental materials in his hand after turning around, but it is determined in step 2 that the staff member has taken the experimental materials, it means that the staff member has kept the experimental materials privately; If it is determined in step 3 that the staff member has experimental materials in his hand after turning around, and it is determined in step 2 that the staff member has taken the experimental materials, then proceed to the next step; The amount of experimental materials counted in step 3 is marked as a1, and the type of experimental materials counted in step 3 is marked as a2. If the staff member takes the experimental materials in the transparent container in step 2, a2 ​​is matched with A2. If the two do not match, it means that the staff member has privately kept the experimental materials. If the two match, proceed to the next step; Compare a1 with A1. If the difference between the two is less than the preset value, it means that the staff did not hide the experimental materials privately. If the difference between the two is greater than or equal to the preset value, proceed to the next step. The historical time taken by the staff to take the experimental materials with the amount a1 and the type a2 is marked as Ti, i=1···n, and the total time the staff turned their back to the surveillance camera to block the experimental materials is marked as t; Calculate the difference between t and Ti. If any difference between Ti and t is within the preset time, it means that the staff member has not hidden the experimental materials. Conversely, if any difference between Ti and t is not within the preset time, it means that the staff member has hidden the experimental materials.

5. The laboratory management platform based on large models and visualization technology according to claim 4 is characterized in that: In step five, the specific process of calculating the risk level of secret possession of experimental materials is as follows: if the staff member secretes the experimental materials, and the placement of the experimental materials has not changed during the process of being blocked by the staff member, the output risk level is level two; if the staff member secretes the experimental materials, and the placement of the experimental materials has changed during the process of being blocked by the staff member, the output risk level is level one.

6. The laboratory management platform based on large model and visualization technology according to claim 5 is characterized in that: The specific process of the data analysis module analyzing the collected temperature information in the laboratory is as follows: The collected temperatures of various areas in the laboratory are marked as Di, i = 1···n; Then the temperature difference between two adjacent areas is marked as di, i = 1···n-1; The temperature of each area of ​​the laboratory at the same time point on the previous day is marked as Ei, i = 1···n; Then, the temperature difference between two adjacent areas of the laboratory at the same time point the previous day is marked as ei, i = 1···n-1; Compare Di and Ei one by one. If the difference between the two is greater than the preset value, the corresponding Di is planned to the first sequence; Compare di and ei one by one. If the difference between the two is greater than the preset value, the corresponding di will be planned to the second sequence; Count the number of all Di in the first sequence and mark it as X1, then count the number of all di in the second sequence and mark it as Y1; Calculate the temperature anomaly weight value Z1 = 60% * X1 + 40% * Y1; Compare Z1 with the preset value z1. If Z1>z1, it means that the laboratory temperature is abnormal. If Z1≤z1, it means that the laboratory temperature is normal.

7. The laboratory management platform based on large models and visualization technology according to claim 6, characterized in that: The specific process of the data analysis module analyzing the collected humidity information in the laboratory is as follows: The collected humidity of each area in the laboratory is marked as Fi, i = 1···n; Then the humidity difference between two adjacent areas is marked as fi, i = 1···n-1; The humidity in each area of ​​the laboratory at the same time point on the previous day is marked as Gi, i = 1···n; Then, the humidity difference between two adjacent areas of the laboratory at the same time point the previous day is marked as gi, i = 1···n-1; Compare Fi and Gi one by one. If the difference between the two is greater than the preset value, the corresponding Fi will be planned to the third sequence; Compare fi and gi one by one. If the difference between them is greater than the preset value, the corresponding fi will be planned to the fourth sequence; Count the number of all Fi in the third sequence and mark it as X2, then count the number of all fi in the fourth sequence and mark it as Y2; Calculate the humidity anomaly weight value Z2 = 60% * X2 + 40% * Y2; Compare Z2 with the preset value z2. If Z2>z2, it means that the laboratory humidity is abnormal. If Z2≤z2, it means that the laboratory humidity is normal.

8. The laboratory management platform based on large models and visualization technology according to claim 7, characterized in that: The specific process of the data analysis module analyzing the collected noise information in the laboratory is as follows: The noise volume collected in real time is marked as H, and the location of the noise source is marked as Q; Compare H with the preset noise volume h. If H>h, it means that the noise in the laboratory is abnormal. If H≤h, it means that the noise in the laboratory is normal. According to the location Q of the noise source, it is judged whether the noise is generated in the laboratory. If so, it means that the noise in the laboratory is abnormal. If not, it means that the noise outside the laboratory is abnormal.

9. The laboratory management platform based on large models and visualization technology according to claim 8, characterized in that: The 3D visualization module includes a 3D holographic projection device. The 3D image projected by the 3D holographic projection device is divided into four areas, which are used to display the analysis results of the monitoring image information in the laboratory, the analysis results of the temperature information in the laboratory, the analysis results of the humidity information in the laboratory, and the analysis results of the noise information in the laboratory.

10. The laboratory management platform based on large model and visualization technology according to claim 9, characterized in that: The specific warning process of the early warning module is: if there are abnormalities in the analysis results of the monitoring image information in the laboratory, the analysis results of the temperature information in the laboratory, the analysis results of the humidity information in the laboratory, and the analysis results of the noise information in the laboratory, the 3D image area corresponding to the abnormal analysis result flashes red.

Citation Information

Patent Citations

  • Wireless terminal acquisition equipment comprehensive management and control system based on Internet of Things

    CN116755493A

  • Multi-information fusion laboratory monitoring system and method

    CN118094461A

  • KR20230147774A

Cited By

  • Experimental operation specification degree evaluation system based on composite decomposition

    CN121236700A