Intelligent nematode detection method and system based on VR simulation

Abstract

The invention discloses a nematode intelligent detection method and system based on VR simulation, and belongs to the technical field of quarantine training management. The system comprises a data acquisition module, a training analysis module, a resource management module and a visual storage module. The data acquisition module is used for acquiring historical records of all objects using the simulation training model in the training room; the training analysis module is used for analyzing a historical record and calculating a risk coefficient of each object when training is started; the resource management module is used for dynamically allocating resources for each object according to the risk coefficient in the training process; and the visual storage module is used for displaying the training result of each object through a visual interface after the training is completed. According to the technical scheme, through a dynamic risk response type resource allocation algorithm and high-precision quarantine biological characteristic modeling, dual jump of operation authenticity and resource efficiency of the VR training system under limited computing power is realized.

Description

Technical Field

[0001] The present invention relates to the field of quarantine training management technology, and specifically to a nematode intelligent detection method and system based on VR simulation. Background Art

[0002] Plant quarantine, a core tool for preventing the cross-border spread of pests and ensuring the ecological security of agriculture and forestry, requires a high degree of practical experience in its technical training. However, traditional training models, limited by the high costs, high risks, and scarce resources of physical laboratories, struggle to meet the large-scale, standardized training needs of grassroots quarantine personnel.

[0003] Currently, virtual simulation technology is commonly used during nematode detection process training for quarantine personnel to simulate daily training. However, certain drawbacks still exist during actual practice, specifically an imbalance between rigid computing power scheduling and the flexible requirements of training scenarios. During training, different students' chosen scenario modes and operating habits lead to significant variations in computing power requirements. When some students perform high-magnification microscopic observations, particularly when rendering 3D nematode motion using a 40x objective, the computing power requirements for a single scene can surge to 3 to 5 times that of conventional separation steps. Fixed resource allocation can lead to command delays, resulting in screen freezes and inaccurate tactile feedback, severely impacting the smoothness and realism of training. For other students, redundant computing power is present when performing routine operations, resulting in overall resource waste. Under the rigid constraints of device computing power, traditional static resource allocation strategies severely limit the system's performance in multi-student collaborative training scenarios. Therefore, an efficient and intelligent quarantine training management technology solution is needed to address these issues. Summary of the Invention

[0004] The purpose of the present invention is to provide a nematode intelligent detection method and system based on VR simulation to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides a nematode intelligent detection method based on VR simulation, comprising:

[0006] S100: Collect historical records of all objects in the training room using the simulation training model.

[0007] S200: When training starts, analyze historical records and calculate the risk coefficient of each subject.

[0008] S300 , dynamically allocating resources to each object according to the risk factor during the training process.

[0009] S400: After the training is completed, the training results of each object are displayed through a visual interface.

[0010] In S100, the simulation training model refers to a virtual scene established based on plant quarantine requirements in a real environment using VR technology. The subject refers to a natural person who undergoes quarantine training using the simulation training model.

[0011] The historical records include the entire operation record of each training session. Each operation record includes the selected mode and the instruction set for that mode. The instruction set includes the generation time, execution time, and the computational cost of each instruction. Instructions refer to the operation instructions performed by the subject during quarantine training.

[0012] Comprehensive historical training data for all subjects in the training room is collected. The VR simulation platform records the complete operational process of each training session, including the user's selected training mode and the set of operational instructions. The data storage structure ensures the traceability of each operation's spatiotemporal attributes and resource requirements, providing the raw data foundation for subsequent analysis.

[0013] S200 includes:

[0014] S201: When a subject in a training room begins quarantine training using a simulation training model, analyze the mode m1 selected by subject Q. Filter operation records with mode m1 from the historical records of all subjects, and obtain the instruction set of each operation record.

[0015] S202: Calculate the sum of the computational effort of all instructions in each instruction set as the total computational effort, and the generation time difference of all instructions as the consumption time. Divide the total computational effort by the consumption time to obtain the reference computing power of the corresponding instruction set. The average value of the reference computing power of all instruction sets is used to obtain the computing power threshold CP of mode m1. t .

[0016] S203. Obtain the history of object Q and filter out all instruction sets with mode m1. Obtain the computational load hc and execution time ed of each instruction in each instruction set, and calculate the computing power of each instruction according to the formula hc÷ed. Substitute the formula into the risk factor FONE of object Q. Q :

[0017]

[0018] Where h is the number of instruction sets, is the average computing power of all instructions in the kth instruction set, j is the number of all instructions in the kth instruction set, is the computing power of the ith instruction in the kth instruction set. Similarly, the risk coefficient of each object is calculated separately.

[0019] During the training startup phase, the system performs the following analyses based on the specific training mode selected by the user:

[0020] Mode threshold calculation: Extract all instruction sets of the same mode from historical data, calculate the total computational amount and time consumption range of each instruction set respectively, and then deduce the computing power threshold of the mode.

[0021] Individual assessment: Based on the historical operation records of the current object, the fluctuation of its instruction computing power in this mode is counted, and the personal risk coefficient is finally calculated through weighted mean square error.

[0022] This method quantifies the positive correlation between user operation stability and potential error risk.

[0023] S300 includes:

[0024] S301. Analyze the computing power threshold CP of each object selection mode and substitute it into the formula: e×CP÷CP sum Calculate the initial allocated computing power YCP of each object. Among them, e is the total computing power of the system, CP sum The total amount of hashrate threshold for selecting mode for all objects.

[0025] S302, obtain the risk factor of each object and calculate the average value FONE of the risk factors of all objects ave The objects with YCP greater than CP are taken as outbound objects, and the objects with YCP less than CP are taken as inbound objects. The estimated outbound computing power CP of each outbound object is calculated according to the formula. c , and the estimated computing power CP of each incoming object r :

[0026]

[0027] In the formula, FONE c To retrieve the risk factor of the object, FONE r The risk factor of the transferred object.

[0028] S303. Calculate the sum of the estimated call-out computing power of all call-out objects as the total call-out computing power CCP sum The sum of the expected incoming computing power of all incoming objects is taken as the total incoming computing power RCP sum .

[0029] S304, When CCP sum Not less than RCP sum When substituting into the formula: RCP sum ×CP c ÷CCP sum Calculate the actual outgoing computing power of each outgoing object separately. Subtract the actual outgoing computing power from the outgoing object's initial allocated computing power to get the final allocated computing power. Add the expected incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power.

[0030] S305, When CCP sum Less than RCP sum When substituting into the formula: CCP sum ×CP r ÷RCP sum Calculate the actual incoming computing power for each incoming object. Subtract the expected outgoing computing power from the initial allocated computing power of the outgoing object to get the final allocated computing power. Add the actual incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power.

[0031] S306: Configure system resources according to the final allocated computing power of each object, update the risk coefficient in real time according to the switching of the modes of each object in the training room, recalculate the final allocated computing power and synchronously modify the configuration method of system resources.

[0032] The final allocated computing power for each object also includes optimized computing power scheduling for the specificity of nematode detection:

[0033] High load priority scenario:

[0034] Motion trajectory prediction: Simulating the migration of nematode groups in soil pores requires real-time collision detection.

[0035] Ultramicroscopic rendering: High-definition reconstruction of the oral structures of a coleopteran insect carrying a nematode.

[0036] Elastic degradation mechanism:

[0037] In normal mode, turn off unnecessary surface scattering ray tracing.

[0038] In emergency mode, AI-assisted recognition is automatically activated when it occupies 30% of the computing power bandwidth.

[0039] In S400, after the same mode training of each object is completed, the operation instructions and training results are displayed through a visual interface, and the instruction set is automatically recorded to generate an operation record and store it in the history record.

[0040] After training is complete, the system displays two core dimensions through an interactive interface:

[0041] Operational Process Reconstruction: Presents the sequence of user commands in chronological order, supporting retrospective review of key operational nodes, such as nematode feature recognition. Resource Consumption Statistics: Graphically displays the distribution of computational effort at each stage. All newly generated instruction sets are automatically archived to the corresponding historical records, forming a closed-loop data flow.

[0042] The nematode intelligent detection system based on VR simulation includes a data acquisition module, a training and analysis module, a resource management module, and a visualization storage module.

[0043] The data acquisition module is used to collect the historical records of all objects in the training room using the simulation training model.

[0044] The training analysis module is used to analyze historical records and calculate the risk factor of each object when training begins.

[0045] The resource management module is used to dynamically allocate resources to each object according to the risk factor during the training process.

[0046] The visualization storage module is used to display the training results of each object through a visualization interface after the training is completed.

[0047] The simulation training model is based on the practical pain points of plant quarantine training, and is implemented by combining virtual reality technology with scenarios and functional requirements. It includes:

[0048] Scenario restoration requirements: simulate the entire laboratory testing process.

[0049] Skill standardization requirements: Build repeatable standardized operation modules.

[0050] Safety training needs: virtualization of high-risk links.

[0051] Focusing on the core processes of plant quarantine laboratories, the following virtual simulation functions have been developed:

[0052] Full-process scenario modeling: Step-by-step reconstruction of the process from receiving wood samples to splitting wood to nematode separation to microscopic examination. Design of interactive logic for virtual microscope operation. Integration of key technical specifications: Standardized operation prompts. Nematode separation time control threshold.

[0053] Build a 3D biological model library to support quarantine and identification. Key technical points include:

[0054] Accurate morphological modeling: 1:1 restoration of key taxonomic features of the pine wood nematode, including stylet length and spur morphology. Dynamic model switching between developmental stages. Multimodal display: supports 360° rotation and local magnification. Pathological feature comparison.

[0055] The data collection module includes a history record collection unit and an operation instruction collection unit.

[0056] The history collection unit is used to collect operation records for the entire training process. Each operation record includes the selected mode and the instruction set used in that mode. The instruction set includes the generation time, execution time, and the computing power consumed by executing each instruction.

[0057] The operation instruction collection unit is used to collect the operation instructions of the subject during quarantine training.

[0058] The training analysis module includes a threshold analysis unit and a coefficient calculation unit.

[0059] The threshold analysis unit is used to calculate the computing power threshold of each mode.

[0060] First, when the subjects in the training room are undergoing quarantine training, the mode m1 selected by subject Q is analyzed. The operation records with mode m1 are screened out from the historical records of all subjects, and the instruction set of each operation record is obtained.

[0061] Secondly, the sum of the computational effort of all instructions in each instruction set is calculated as the total computational effort, and the extreme difference in the generation time of all instructions is calculated as the consumed time. The total computational effort is divided by the consumed time to obtain the reference computing power of the corresponding instruction set.

[0062] Finally, the average value of the reference computing power of all instruction sets is used to obtain the computing power threshold CP of mode m1. t Similarly, the computing power threshold of each object selection mode is calculated separately.

[0063] The coefficient calculation unit is used to calculate the risk coefficient of the object.

[0064] First, get the history of object Q and filter out all instruction sets with mode m1.

[0065] Secondly, obtain the computational load hc and execution time ed of each instruction in each instruction set, and calculate the computing power of each instruction according to the formula hc÷ed.

[0066] Finally, according to the formula: Calculate the risk factor FONE of object Q Q .

[0067] Where h is the number of instruction sets, is the average computing power of all instructions in the kth instruction set, j is the number of all instructions in the kth instruction set, is the computing power of the i-th instruction in the k-th instruction set.

[0068] And so on, the risk coefficient of each object is calculated separately.

[0069] The resource management module includes a computing power allocation unit and an execution supervision unit.

[0070] The computing power allocation unit is used to calculate the final allocated computing power for each object.

[0071] First, analyze the computing power threshold CP of each object selection mode and substitute it into the formula: e×CP÷CP sum Calculate the initial allocated computing power YCP of each object.

[0072] e is the total computing power of the system, CP sum The total amount of hashrate threshold for selecting mode for all objects.

[0073] Secondly, obtain the risk factor of each object and calculate the average value of the risk factor of all objects FONEave The object whose YCP is greater than CP is taken as the object to be called out, and the object whose YCP is less than CP is taken as the object to be called in.

[0074] According to the formula: Calculate the estimated call-out computing power CP for each call-out object separately c , and the estimated computing power CP of each incoming object r .

[0075] FONE c To retrieve the risk factor of the object, FONE r The risk factor of the transferred object.

[0076] Then, the sum of the estimated call-out computing power of all call-out objects is calculated as the total call-out computing power CCP sum The sum of the expected incoming computing power of all incoming objects is taken as the total incoming computing power RCP sum .

[0077] Finally, when the CCP sum Not less than RCP sum When, according to the formula: RCP sum ×CP c ÷CCP sum Calculate the actual outgoing computing power of each outgoing object separately. Subtract the actual outgoing computing power from the outgoing object's initial allocated computing power to get the final allocated computing power. Add the expected incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power.

[0078] When the CCP sum Less than RCP sum When substituting into the formula: CCP sum ×CP r ÷RCP sum Calculate the actual incoming computing power for each incoming object. Subtract the expected outgoing computing power from the initial allocated computing power of the outgoing object to get the final allocated computing power. Add the actual incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power.

[0079] The execution supervision unit configures system resources according to the final allocated computing power of each object.

[0080] The risk factor is updated in real time according to the switching of each object mode in the training room, the final allocated computing power is recalculated, and the configuration of system resources is modified synchronously.

[0081] The visualization storage module is used to display the operation instructions and training results through a visual interface after the training of each object is completed, generate operation records and store them in the history record.

[0082] The overall model training forms an integrated platform of "teaching, learning, training and combat":

[0083] Teaching mode: 3D dynamic demonstration of standard operating procedures.

[0084] Practical training mode: simulates emergency response to an epidemic. Automatically scores operational compliance.

[0085] Evaluation and feedback mechanism: Generate statistical reports on operation time and error types. Recommend repeated training of key skill points.

[0086] Compared with the prior art, the present invention has the following beneficial effects:

[0087] Intelligent dynamic resource allocation mechanism: This mechanism uses historical data analysis to calculate precise model computing power thresholds and object risk factors, establishing a computing power allocation algorithm based on real-time parameter adjustment. Compared to the traditional fixed quota allocation model, this mechanism dynamically allocates computing resources based on the trainee's operational stability, ensuring that high-complexity scenarios receive the necessary computing power while preventing low-risk operations from occupying redundant resources.

[0088] Application of Virtual Reality Technology: VR technology is used to create a realistic plant quarantine training environment, simulating the entire laboratory testing process, from sample receipt to microscopic examination, to achieve safety training and skill standardization for on-site operations. After training is completed, a visual interface displays the training results for each subject, presenting the user command sequence in chronological order and providing resource consumption statistics, enabling the restoration of the operation process and the tracing of key nodes.

[0089] Integrated teaching, learning, training and combat platform: This solution forms an integrated platform of "teaching, learning, training and combat" to adapt to different training needs. It includes dynamic demonstration of standard operating procedures and actual combat training mode, which effectively improves the systematicness and practicality of training.

[0090] In summary, the present invention provides a safe, controllable, and quantitatively controllable digital infrastructure for quarantine talent training, which has significant industry innovation significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0092] Figure 1 This is a flow chart of the nematode intelligent detection method based on VR simulation of the present invention;

[0093] Figure 2 It is a structural diagram of the nematode intelligent detection system based on VR simulation of the present invention. DETAILED DESCRIPTION

[0094] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0095] See also Figure 1 The present invention provides a nematode intelligent detection method based on VR simulation, comprising:

[0096] S100: Collect historical records of all objects in the training room using the simulation training model.

[0097] S200: When training starts, analyze historical records and calculate the risk factor of each subject.

[0098] S300 , dynamically allocating resources to each object according to the risk factor during training.

[0099] S400: After the training is completed, the training results of each object are displayed through a visual interface.

[0100] In S100, the simulation training model refers to a virtual scene established based on plant quarantine requirements in a real environment using VR technology. The subject refers to a natural person who undergoes quarantine training using the simulation training model.

[0101] The historical records include the entire operation record of each training session. Each operation record includes the selected mode and the instruction set for that mode. The instruction set includes the generation time, execution time, and the computational cost of each instruction. Instructions refer to the operation instructions performed by the subject during quarantine training.

[0102] Comprehensive historical training data for all subjects in the training room is collected. The VR simulation platform records the complete operational process of each training session, including the user's selected training mode (e.g., basic mode / advanced mode) and the set of operational instructions (including each instruction's generation timestamp, execution duration, and corresponding computing resource consumption). The data storage structure ensures the traceability of each operation's spatiotemporal attributes and resource requirements, providing the raw data foundation for subsequent analysis.

[0103] S200 includes:

[0104] S201: When a subject in a training room begins quarantine training using a simulation training model, analyze the mode m1 selected by subject Q. Filter operation records with mode m1 from the historical records of all subjects, and obtain the instruction set of each operation record.

[0105] S202: Calculate the sum of the computational effort of all instructions in each instruction set as the total computational effort, and the generation time difference of all instructions as the consumption time. Divide the total computational effort by the consumption time to obtain the reference computing power of the corresponding instruction set. The average value of the reference computing power of all instruction sets is used to obtain the computing power threshold CP of mode m1. t .

[0106] S203. Obtain the history of object Q and filter out all instruction sets with mode m1. Obtain the computational load hc and execution time ed of each instruction in each instruction set, and calculate the computing power of each instruction according to the formula hc÷ed. Substitute the formula into the risk factor FONE of object Q. Q :

[0107]

[0108] Where h is the number of instruction sets, is the average computing power of all instructions in the kth instruction set, j is the number of all instructions in the kth instruction set, is the computing power of the ith instruction in the kth instruction set. Similarly, the risk coefficient of each object is calculated separately.

[0109] During the training startup phase, the system performs the following analyses based on the specific training mode selected by the user:

[0110] Mode threshold calculation: Extract all instruction sets of the same mode from historical data, calculate the total computational amount and time consumption range of each instruction set respectively, and then deduce the computing power threshold of the mode.

[0111] Individual assessment: Based on the historical operation records of the current object, the fluctuation of its instruction computing power in this mode is counted, and the personal risk coefficient is finally calculated through weighted mean square error.

[0112] This method quantifies the positive correlation between user operation stability and potential error risk.

[0113] S300 includes:

[0114] S301. Analyze the computing power threshold CP of each object selection mode and substitute it into the formula: e×CP÷CP sum Calculate the initial allocated computing power YCP of each object. Among them, e is the total computing power of the system, CP sum The total amount of hashrate threshold for selecting mode for all objects.

[0115] S302, obtain the risk factor of each object and calculate the average value FONE of the risk factors of all objects aveThe objects with YCP greater than CP are taken as outbound objects, and the objects with YCP less than CP are taken as inbound objects. The estimated outbound computing power CP of each outbound object is calculated according to the formula. c , and the estimated computing power CP of each incoming object r :

[0116]

[0117] In the formula, FONE c To retrieve the risk factor of the object, FONE r The risk factor of the transferred object.

[0118] S303. Calculate the sum of the estimated call-out computing power of all call-out objects as the total call-out computing power CCP sum The sum of the expected incoming computing power of all incoming objects is taken as the total incoming computing power RCP sum .

[0119] S304, When CCP sum Not less than RCP sum When substituting into the formula: RCP sum ×CP c ÷CCP sum Calculate the actual outgoing computing power of each outgoing object separately. Subtract the actual outgoing computing power from the outgoing object's initial allocated computing power to get the final allocated computing power. Add the expected incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power.

[0120] S305, When CCP sum Less than RCP sum When substituting into the formula: CCP sum ×CP r ÷RCP sum Calculate the actual incoming computing power for each incoming object. Subtract the expected outgoing computing power from the initial allocated computing power of the outgoing object to get the final allocated computing power. Add the actual incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power.

[0121] S306: Configure system resources according to the final allocated computing power of each object, update the risk coefficient in real time according to the switching of the modes of each object in the training room, recalculate the final allocated computing power and synchronously modify the configuration method of system resources.

[0122] The final allocated computing power for each object also includes optimized computing power scheduling for the specificity of nematode detection:

[0123] High load priority scenario:

[0124] Motion trajectory prediction: Simulating the migration of nematode groups in soil pores requires real-time collision detection (CPU-intensive).

[0125] Supermicroscopic rendering: High-definition reconstruction of the oral structure of a coleopteran insect carrying a nematode (multi-GPU parallel processing is triggered when GPU memory usage exceeds 40%).

[0126] Elastic degradation mechanism:

[0127] In normal mode (basic training), turn off unnecessary surface scattering ray tracing.

[0128] In emergency mode (such as simulating an epidemic outbreak), AI-assisted recognition is automatically activated when it occupies 30% of the computing power bandwidth.

[0129] In S400, after the same mode training of each object is completed, the operation instructions and training results are displayed through a visual interface, and the instruction set is automatically recorded to generate an operation record and store it in the history record.

[0130] After training is complete, the system displays two core dimensions through an interactive interface:

[0131] Operational Process Reconstruction: Presents the sequence of user commands in chronological order, supporting retrospective review of key operational nodes, such as nematode feature recognition. Resource Consumption Statistics: Graphically displays the distribution of computational effort at each stage. All newly generated instruction sets are automatically archived to the corresponding historical records, forming a closed-loop data flow.

[0132] The nematode intelligent detection system based on VR simulation includes a data acquisition module, a training and analysis module, a resource management module, and a visualization storage module.

[0133] The data acquisition module is used to collect the historical records of all objects in the training room using the simulation training model.

[0134] The training analysis module is used to analyze historical records and calculate the risk factor of each object when training begins.

[0135] The resource management module is used to dynamically allocate resources to each object according to the risk factor during the training process.

[0136] The visualization storage module is used to display the training results of each object through a visualization interface after the training is completed.

[0137] The simulation training model is based on the practical pain points of plant quarantine training (such as the risks of handling high-risk pests and the limitations of physical laboratory space), and is implemented by combining virtual reality technology (VR) with scenarios and functional requirements. It includes:

[0138] Scene restoration requirements: simulate the entire laboratory testing process (such as wood sample pretreatment, nematode separation, and microscopic observation).

[0139] Skill standardization requirements: Build repeatable standardized operation modules (such as key points for morphological identification of pine wood nematodes).

[0140] Safety training needs: Virtualization of high-risk links (such as harmless treatment operations of live pests).

[0141] Focusing on the core processes of plant quarantine laboratories, the following virtual simulation functions have been developed:

[0142] Full-process scenario modeling: A step-by-step reconstruction of the process from receiving wood samples to splitting wood to nematode isolation to microscopic examination. Interactive logic design for virtual microscope operations (focusing, switching objectives, etc.). Key technical specifications implemented: Standardized operational prompts (e.g., sample number verification, mandatory disinfection step verification). Nematode isolation time control thresholds (e.g., international standard duration for the Baermann funnel method).

[0143] Build a 3D biological model library to support quarantine and identification. Key technical points include:

[0144] Accurate morphological modeling: 1:1 reconstruction of key taxonomic features of the pine wood nematode, including stylet length and spur morphology. Dynamic model switching between developmental stages (larvae and adults). Multimodal display: 360° rotation and zoom (e.g., 3D slice viewing of female nematodes). Comparison of pathological features (e.g., differences in texture between healthy and diseased xylem).

[0145] The data collection module includes a history record collection unit and an operation instruction collection unit.

[0146] The history collection unit is used to collect operation records for the entire training process. Each operation record includes the selected mode and the instruction set used in that mode. The instruction set includes the generation time, execution time, and the computing power consumed by executing each instruction.

[0147] The operation instruction collection unit is used to collect the operation instructions of the subject during quarantine training.

[0148] The training analysis module includes a threshold analysis unit and a coefficient calculation unit.

[0149] The threshold analysis unit is used to calculate the computing power threshold of each mode.

[0150] First, when the subjects in the training room are undergoing quarantine training, the mode m1 selected by subject Q is analyzed. The operation records with mode m1 are screened out from the historical records of all subjects, and the instruction set of each operation record is obtained.

[0151] Secondly, the sum of the computational effort of all instructions in each instruction set is calculated as the total computational effort, and the extreme difference in the generation time of all instructions is calculated as the consumed time. The total computational effort is divided by the consumed time to obtain the reference computing power of the corresponding instruction set.

[0152] Finally, the average value of the reference computing power of all instruction sets is used to obtain the computing power threshold CP of mode m1. t Similarly, the computing power threshold of each object selection mode is calculated separately.

[0153] The coefficient calculation unit is used to calculate the risk coefficient of the object.

[0154] First, get the history of object Q and filter out all instruction sets with mode m1.

[0155] Secondly, obtain the computational load hc and execution time ed of each instruction in each instruction set, and calculate the computing power of each instruction according to the formula hc÷ed.

[0156] Finally, according to the formula: Calculate the risk factor FONE of object Q Q .

[0157] Where h is the number of instruction sets, is the average computing power of all instructions in the kth instruction set, j is the number of all instructions in the kth instruction set, is the computing power of the i-th instruction in the k-th instruction set.

[0158] And so on, the risk coefficient of each object is calculated separately.

[0159] The resource management module includes a computing power allocation unit and an execution supervision unit.

[0160] The computing power allocation unit is used to calculate the final allocated computing power for each object.

[0161] First, analyze the computing power threshold CP of each object selection mode and substitute it into the formula: e×CP÷CP sum Calculate the initial allocated computing power YCP of each object.

[0162] e is the total computing power of the system, CP sum The total amount of hashrate threshold for selecting mode for all objects.

[0163] Secondly, obtain the risk factor of each object and calculate the average value of the risk factor of all objects FONE ave The object whose YCP is greater than CP is taken as the object to be called out, and the object whose YCP is less than CP is taken as the object to be called in.

[0164] According to the formula: Calculate the estimated call-out computing power CP for each call-out object separately c , and the estimated computing power CP of each incoming object r .

[0165] FONE cTo retrieve the risk factor of the object, FONE r The risk factor of the transferred object.

[0166] Then, the sum of the estimated call-out computing power of all call-out objects is calculated as the total call-out computing power CCP sum The sum of the expected incoming computing power of all incoming objects is taken as the total incoming computing power RCP sum .

[0167] Finally, when the CCP sum Not less than RCP sum When, according to the formula: RCP sum ×CP c ÷CCP sum Calculate the actual outgoing computing power of each outgoing object separately. Subtract the actual outgoing computing power from the outgoing object's initial allocated computing power to get the final allocated computing power. Add the expected incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power.

[0168] When the CCP sum Less than RCP sum When substituting into the formula: CCP sum ×CP r ÷RCP sum Calculate the actual incoming computing power for each incoming object. Subtract the expected outgoing computing power from the initial allocated computing power of the outgoing object to get the final allocated computing power. Add the actual incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power.

[0169] The execution supervision unit configures system resources according to the final allocated computing power of each object.

[0170] The risk factor is updated in real time according to the switching of each object mode in the training room, the final allocated computing power is recalculated, and the configuration of system resources is modified synchronously.

[0171] The visualization storage module is used to display the operation instructions and training results through a visual interface after the training of each object is completed, generate operation records and store them in the history record.

[0172] The overall model training forms an integrated platform of "teaching, learning, training and combat":

[0173] Teaching mode: 3D dynamic demonstration of standard operating procedures (SOPs) (e.g., parameter settings for a nematode separation centrifuge).

[0174] Practical training mode: simulates emergency response to epidemics (e.g., emergency handling of live nematodes intercepted in imported timber). Automatic scoring of operational compliance (e.g., alarm for deviation in splitting angle).

[0175] Evaluation and feedback mechanism: Generate statistical reports on operation time and error types. Recommend repeated training of key skill points (e.g., triggering special exercises when the failure rate of depth of field adjustment on a stereo microscope exceeds 30%).

[0176] Example 1:

[0177] Assume that the computing power threshold of the selected mode for object Q1 is 15 TFLOPS. Its instruction set contains three instructions: A1, A2, and A3, with computing powers of 8 TFLOPS, 7 TFLOPS, and 9 TFLOPS respectively. Substitute the formula to calculate the risk coefficient of object Q1:

[0178]

[0179] The risk factor of object Q1 is 4.42.

[0180] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0181] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A VR-based intelligent nematode detection method, characterized by: The method includes: S100, collecting historical records of all subjects in the training room using the simulation training model; S200, when training begins, analyzing historical records and calculating the risk factor of each subject; S300, dynamically allocating resources to each object according to the risk factor during training; S400: After the training is completed, the training results of each object are displayed through a visual interface.

2. The VR simulation-based nematode intelligent detection method according to claim 1, characterized in that: In S100, the simulation training model refers to a virtual scene established based on plant quarantine requirements in a real environment using VR technology; the subject refers to a natural person who undergoes quarantine training using the simulation training model; The historical records include the operation records of the entire training process. Each operation record includes the selected mode and the instruction set under the mode. The instruction set includes the generation time, execution time and computing power consumed by executing the instructions. Instructions refer to the operation instructions of the subject during quarantine training.

3. The VR simulation-based nematode intelligent detection method according to claim 2, characterized in that: S200 includes: S201. When a subject in the training room begins quarantine training using the simulation training model, the mode m1 selected by subject Q is analyzed; operation records with mode m1 are filtered out from the historical records of all subjects, and the instruction set of each operation record is obtained; S202. Calculate the sum of the computational effort of all instructions in each instruction set as the total computational effort, and the range of the generation time of all instructions as the consumed time. Divide the total computational effort by the consumed time to obtain the reference computing power of the corresponding instruction set. Calculate the average value of the reference computing power of all instruction sets to obtain the computing power threshold CP of mode m1. t ; S203. Obtain the historical records of object Q and filter out all instruction sets with mode m1; obtain the computational load hc and execution time ed of each instruction in each instruction set, and calculate the computing power of each instruction according to the formula hc÷ed; substitute the formula into the formula to calculate the risk factor FONE of object Q. Q : Where h is the number of instruction sets, is the average computing power of all instructions in the kth instruction set, j is the number of all instructions in the kth instruction set, is the computing power of the i-th instruction in the k-th instruction set; and so on, the risk coefficient of each object is calculated separately.

4. The VR simulation-based nematode intelligent detection method according to claim 3, characterized in that: S300 includes: S301. Analyze the computing power threshold CP of each object selection mode and substitute it into the formula: e×CP÷CP sum Calculate the initial allocated computing power YCP of each object; where e is the total computing power of the system, CP sum The total amount of computing power threshold for selecting mode for all objects; S302, obtain the risk factor of each object and calculate the average value FONE of the risk factors of all objects ave ; The object with YCP greater than CP is regarded as the outbound object, and the object with YCP less than CP is regarded as the inbound object; the expected outbound computing power CP of each outbound object is calculated according to the formula c , and the estimated computing power CP of each incoming object r : In the formula, FONE c To retrieve the risk factor of the object, FONE r The risk factor of the transferred object; S303. Calculate the sum of the estimated call-out computing power of all call-out objects as the total call-out computing power CCP sum The sum of the expected incoming computing power of all incoming objects is taken as the total incoming computing power RCP sum ; S304, When CCP sum Not less than RCP sum When substituting into the formula: RCP sum ×CP c ÷CCP sum Calculate the actual outgoing computing power of each outgoing object separately; subtract the actual outgoing computing power from the outgoing object's initial allocated computing power to get the final allocated computing power, and add the expected incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power; S305, When CCP sum Less than RCP sum When substituting into the formula: CCP sum ×CP r ÷RCP sum Calculate the actual incoming computing power for each incoming object separately; subtract the expected outgoing computing power from the initial allocated computing power of the outgoing object to get the final allocated computing power, and add the actual incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power; S306: Configure system resources according to the final allocated computing power of each object, update the risk coefficient in real time according to the switching of the modes of each object in the training room, recalculate the final allocated computing power and synchronously modify the configuration method of system resources.

5. The VR simulation-based nematode intelligent detection method according to claim 4, characterized in that: In S400, after the same mode training of each object is completed, the operation instructions and training results are displayed through a visual interface, and the instruction set is automatically recorded to generate an operation record and store it in the history record.

6. The VR simulation-based nematode intelligent detection system is characterized by: The system includes data acquisition module, training analysis module, resource management module and visual storage module; The data acquisition module is used to collect the historical records of all objects in the training room using the simulation training model; The training analysis module is used to analyze historical records and calculate the risk factor of each subject at the beginning of training; The resource management module is used to dynamically allocate resources to each object according to the risk factor during the training process; The visualization storage module is used to display the training results of each object through a visualization interface after the training is completed.

7. The VR simulation-based nematode intelligent detection system according to claim 6, characterized in that: The data acquisition module includes a history record acquisition unit and an operation instruction acquisition unit; The history record collection unit is used to collect the operation records of the entire training process. Each operation record includes the selected mode and the instruction set under this mode; the instruction set includes the generation time, execution time and computing power consumed by executing each instruction. The operation instruction collection unit is used to collect the operation instructions of the subject during quarantine training.

8. The VR simulation-based nematode intelligent detection system according to claim 7, characterized in that: The training analysis module includes a threshold analysis unit and a coefficient calculation unit; The threshold analysis unit is used to calculate the computing power threshold of each mode; First, when the subjects in the training room are undergoing quarantine training, the mode m1 selected by subject Q is analyzed; the operation records with mode m1 are filtered out from the historical records of all subjects, and the instruction set of each operation record is obtained; Secondly, the sum of the computational effort of all instructions in each instruction set is calculated as the total computational effort, and the range of the generation time of all instructions is calculated as the consumption time. The total computational effort is divided by the consumption time to obtain the reference computing power of the corresponding instruction set. Finally, the average value of the reference computing power of all instruction sets is used to obtain the computing power threshold CP of mode m1. t ; Similarly, calculate the computing power threshold of each object selection mode separately; The coefficient calculation unit is used to calculate the risk coefficient of the object; First, obtain the history of object Q and filter out all instruction sets with mode m1; Secondly, obtain the computational load hc and execution time ed of each instruction in each instruction set, and calculate the computing power of each instruction according to the formula hc÷ed; Finally, according to the formula: Calculate the risk factor FONE of object Q Q ; Where h is the number of instruction sets, is the average computing power of all instructions in the kth instruction set, j is the number of all instructions in the kth instruction set, is the computing power of the i-th instruction in the k-th instruction set; And so on, the risk coefficient of each object is calculated separately.

9. The VR simulation-based nematode intelligent detection system according to claim 8, characterized in that: The resource management module includes a computing power allocation unit and an execution supervision unit; The computing power allocation unit is used to calculate the final allocated computing power for each object; First, analyze the computing power threshold CP of each object selection mode and substitute it into the formula: e×CP÷CP sum Calculate the initial allocated computing power YCP of each object; e is the total computing power of the system, CP sum The total amount of computing power threshold for selecting mode for all objects; Secondly, obtain the risk factor of each object and calculate the average value of the risk factor of all objects FONE ave ; The object whose YCP is greater than CP is taken as the outgoing object, and the object whose YCP is less than CP is taken as the incoming object; According to the formula: Calculate the estimated call-out computing power CP for each call-out object separately c , and the estimated computing power CP of each incoming object r ; FONE c To retrieve the risk factor of the object, FONE r The risk factor of the transferred object; Then, the sum of the estimated call-out computing power of all call-out objects is calculated as the total call-out computing power CCP sum The sum of the expected incoming computing power of all incoming objects is taken as the total incoming computing power RCP sum ; Finally, when the CCP sum Not less than RCP sum When, according to the formula: RCP sum ×CP c ÷CCP sum Calculate the actual outgoing computing power of each outgoing object separately; subtract the actual outgoing computing power from the outgoing object's initial allocated computing power to get the final allocated computing power, and add the expected incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power; When the CCP sum Less than RCP sum When substituting into the formula: CCP sum ×CP r ÷RCP sum Calculate the actual incoming computing power for each incoming object separately; subtract the expected outgoing computing power from the initial allocated computing power of the outgoing object to get the final allocated computing power, and add the actual incoming computing power to the initial allocated computing power of the incoming object to get the final allocated computing power; The execution supervision unit configures system resources according to the final allocated computing power of each object; The risk factor is updated in real time according to the switching of each object mode in the training room, the final allocated computing power is recalculated, and the configuration of system resources is modified synchronously.

10. The VR simulation-based nematode intelligent detection system according to claim 9, characterized in that: The visualization storage module is used to display the operation instructions and training results through a visual interface after the training of each object is completed, generate operation records and store them in the history record.

Images