256results about How to "Reduce computing power requirements" patented technology

The invention provides an underground long-distance pipeline three-dimensional mapping method and system based on three-dimensional Gaussian sputtering. The method comprises the steps that a color map, a depth map and IMU data are collected; pixel coordinate conversion is carried out to obtain point cloud coordinates; inputting the depth map, the color map and the IMU data into a pose estimation algorithm, and processing to obtain a predicted pose of the IMU data, visual pose estimation and visual pose estimation information; initializing three-dimensional Gaussian, performing back propagation through a three-dimensional Gaussian loss function, and optimizing to obtain an applicable three-dimensional model; on the basis of the initialized three-dimensional Gaussian, using a GS-SLAM algorithm to update the applicable three-dimensional model and visual pose estimation information, and using a three-dimensional Gaussian loss function to process to obtain an optimized camera pose; the initial camera pose is updated through Kalman filtering, and pipeline mapping operation is completed. The technical problems that the equipment memory is large, the labor cost and the time cost are high, errors are easy to accumulate, pose deviation is easy to occur, and the model is difficult to maintain relatively high accuracy are solved.

The invention provides a power distribution network model lightweight deployment method and device based on knowledge distillation, and the method comprises the steps: carrying out the data division of a first data set, and obtaining a training set; constructing a teacher model, and training the teacher model through the training set; constructing a student model, softening output labels of the teacher model and the student model, and training the student model through the training set and the teacher model; and descending the student model to the power distribution network edge node, training the student model based on the second data set, deploying the student model on the power distribution network edge node, and enabling the student model to operate on the power distribution network edge node. According to the method, the communication pressure of cloud edge interaction can be relieved, the computing power demand of an edge side computing model is reduced, and the availability of a power distribution network artificial intelligence computing model is improved.

The invention provides a fluid simulation data prediction and noise reduction method, system and device, and relates to the technical field of fluid simulation, and the method mainly comprises the steps: 1, constructing a simulation model; the simulation model comprises an auto-encoder, a denoising model and a decoder which are arranged in sequence; the auto-encoder is used for encoding condition input data and mapping the condition input data to a potential space to obtain characterization data; the denoising model is used for introducing noise into the representation data and converting the representation data into target data through denoising processing; the decoder is used for decoding the target data and mapping the target data to a functional space to obtain simulation result data; and step 2, inputting initial condition data of fluid dynamics into the simulation model, and outputting corresponding prediction result data. According to the scheme, the space noise reduction and output generation operation is completed in the potential space, the operation efficiency can be improved, the operation cost can be reduced, the output can be smoother, and therefore the quality and stability of the output are improved.

The invention relates to the technical field of agricultural engineering, in particular to a flexible fruit adaptive initial grabbing force estimation system and method.Image data are obtained through a visual camera, and based on the image data, contour features and fruit hardness of a to-be-grabbed fruit are obtained through an image processing and neural network method; and then the estimation of the initial grabbing force of the fruit is completed. The method has the beneficial effects that the initial grabbing force of the flexible fruit is estimated only through machine vision without excessive sensors, the structure is simple, the cost is saved, meanwhile, based on industry application priori knowledge, a fruit category and maturity recognition model is designed, fruit hardness information is indirectly obtained, the model reasoning difficulty is reduced, and the working efficiency is improved. On the premise that the identification precision requirement is met, the requirement for the computing power of a computer is lowered.

The invention relates to a real-time optimization control method for a central air-conditioning system based on an agent model. The machine learning algorithm is adopted to construct the agent model used for predicting the optimal control strategy result, the real-time operation condition is input into the agent model in the actual application process, the optimal control strategy result under the operation condition is predicted, and compared with a traditional control method based on model optimization, the method has the advantages that the operation efficiency is improved, and the cost is reduced. According to the method, the calculation complexity is reduced, the calculation efficiency of the optimal control strategy is remarkably improved, and the real-time optimization control requirement of the central air conditioning system is met. The mathematical optimization model of the central air-conditioning system is established based on the equipment composition and the topological structure of the target central air-conditioning system as well as the optimization target and the constraint condition during model training, so that a large number of optimal control strategies corresponding to the operation condition sample are generated through the randomly generated operation condition; and forming a training data set for agent model training, so that the trained agent model can accurately predict an optimal control strategy.

The invention discloses an image filtering method based on eye movement tracking, a near-eye display method and system, a head-mounted display, a medium and a product, and is applied to the technical field of display of the head-mounted display. The method comprises the steps of obtaining a gaze point position of human eyes in a display area; determining a gaze point area of the gaze point in a display area based on the gaze point position; generating a target point spread function of the staring point area by combining an interpolation algorithm based on the pre-calibrated point spread function of each position in the display area; and filtering the gaze point region in the rendered image based on the target point spread function to perform defocused spot correction. According to the method, defocused spot correction can be carried out on the gazing point area with relatively high influence on the display effect, the filtering processing amount is relatively small, the required computing power is relatively low, and the display effect is improved under the condition that the frame rate or the resolution of the head-mounted display is not influenced.

The embodiment of the invention provides a communication and memory efficient large model distributed training method and a text classification method, and belongs to the technical field of artificial intelligence. The method comprises the steps of performing model disturbance on an original local classification model according to a model disturbance vector and a disturbance hyper-parameter to obtain a first reference classification model and a second reference classification model, performing text classification on language texts through the first reference classification model to obtain a first text category, and performing text classification on the language texts through the second reference classification model; performing text classification on the language text through a second reference classification model to obtain a second text category, calculating a loss difference value according to the text category label, the first text category and the second text category, obtaining an original symbol of the loss difference value, and sending the original symbol to a server, and receiving the target symbol obtained by aggregating the original symbols by the server, and updating the original local classification model according to the target symbol, the model disturbance vector and the preset learning rate to obtain the target classification model, so that the model training efficiency can be improved.

The invention discloses a semantic map construction method based on an RGBD visual segmentation algorithm, and the method comprises the steps: firstly initializing each parameter, constructing a visual segmentation neural network model, carrying out the visual semantic segmentation of a collected image, carrying out the calculation of the collected image, converting a depth image into point cloud data, carrying out the sparsification processing, and carrying out the segmentation of a visual segmentation neural network model; and then generating point cloud data with semantic information, superposing the point cloud data with external parameters of the camera, projecting the superposed point cloud data to an actual three-dimensional space, determining a spatial grid with semantic information corresponding to each moment, superposing and aging the acquired spatial grids at different moments, and constructing a complete semantic map based on data acquired in a time sequence. According to the scheme, the requirement for equipment is low, the detection range is wide, the requirement for computing power is low, robustness is high, dense sensing of the surrounding space can be achieved even if the constructed semantic map is sparse and matched with a proper frame rate, and the very high fine degree can be achieved.

The invention discloses a non-intrusive permanent magnet synchronous servo motor bearing fault on-line diagnosis method and system, and the method comprises the steps: injecting a chirp current signal in a tuning stage, calculating a frequency response function through combining with a rotating speed feedback signal, and determining a resonant frequency according to the frequency response function; calculating a theoretical characteristic order according to the motor bearing size parameter; in a motor operation stage, collecting a motor rotating speed signal and determining a theoretical characteristic frequency according to the theoretical characteristic order; determining an optimal resonance demodulation frequency band according to the resonance frequency and the theoretical characteristic frequency, and performing band-pass filtering on the motor rotating speed signal according to the optimal resonance demodulation frequency band; an envelope signal of the filtered signal is calculated online through an orthogonal demodulation algorithm, the frequency spectrum of the envelope signal is obtained through FFT analysis, threshold comparison is conducted on the frequency spectrum of the envelope signal and the amplitude corresponding to the theoretical characteristic frequency, and fault diagnosis result judgment is completed. According to the invention, on-line diagnosis of the bearing fault of the permanent magnet synchronous servo motor under the condition of low-performance chip resources is realized.

The invention discloses a permanent magnet synchronous motor sensorless control method based on a dynamic position error. A transition stage is provided, per-unit estimation position errors between a first position error signal obtained based on a high-frequency signal injection method and a second position error signal obtained based on a sliding mode observer are compared, an optimal switching speed point is dynamically determined, smooth fusion is carried out on the two position errors by using a nonlinear weighting function, and the optimal switching speed point is obtained. And finally, the estimated position and rotating speed are obtained through a phase-locked loop. And finally, position tracking is carried out by adopting different error signals according to the estimated rotating speed. According to the method, the problem of estimation value pulsation and jump caused by the fact that a traditional method depends on fixed experience switching points is solved, smooth and self-adaptive switching of the two estimation methods in the full-speed domain range is achieved, the control precision and operation stability of the system are remarkably improved, the structure is simple, and engineering implementation is easy.

The invention discloses an energy substitution-based efficient generation method for an optimal trajectory of aerodynamic-assisted orbit drop burn-up, which belongs to the field of aerospace and comprises the following steps of: establishing a spacecraft aerodynamic-assisted orbit drop flight dynamics model, extracting a motion equation in a longitudinal plane, converting an independent variable into a variable in an energy form, and generating the optimal trajectory of the aerodynamic-assisted orbit drop burn-up. Obtaining a height and energy decoupling differential equation, and parameterizing the change of the aerodynamic acceleration; designing a segmented parameterization expression of the track angle with respect to energy, and realizing high-precision fitting of the optimal track angle change; based on the fitting function, deriving an expression of terminal energy in different situations, and further predicting the trajectory speed; and corresponding speeds are calculated at different height points, a burnup optimal reference trajectory is efficiently generated, and it is guaranteed that the spacecraft is transferred to a target orbit with low burnup. According to the energy substitution-based efficient generation method for the optimal burnup trajectory of pneumatic auxiliary rail descending, provided by the invention, efficient generation of the subsequent optimal burnup trajectory can be realized no matter in a height descending section or a height ascending section.

The invention relates to an agent brain-like SLAM (Simultaneous Localization and Mapping) and end-side environment exploration method and system, belongs to three-dimensional scene construction and artificial intelligence detection and identification, and solves the problems of low precision, high computing resource consumption, inaccurate target sound source positioning and insufficient visualization effect during end-side environment exploration in the existing map construction process. Comprising the following steps: acquiring IMU data, a camera image, DVS event sequence data and a sound source signal around an intelligent agent; characterizing the DVS event sequence data into a normalized voxel grid through frequency coding, then extracting features, fusing the features with the features of the camera image to obtain fused features, and obtaining the pose of the intelligent agent based on the fused features and IMU (Inertial Measurement Unit) data; constructing a cognitive map by using a spatial position characterization mechanism based on grid cell coding in combination with a camera image and DVS data after voxel grid normalization on the basis of the pose of the intelligent agent; and inputting a sound source signal into the trained sound listening and position distinguishing model to obtain azimuth information of a sound source relative to the intelligent agent in the cognitive map.

The embodiment of the invention provides a water quality monitoring method based on edge calculation and a related device, and belongs to the technical field of data prediction. The method comprises the following steps: acquiring initial parameter data of a target water environment under a target sensor and target space information under a positioning sensor; determining the measurement reliability degree of the target sensor according to the initial parameter data, and adjusting the initial parameter data according to the measurement reliability degree to obtain target parameter data; performing data fusion on the target parameter data to obtain target fusion data of the target sensor; sending the target fusion data, the target parameter data and the target space information to a cloud server in communication connection with the edge device, so that the cloud server performs water quality prediction on the target fusion data, the target parameter data and the target space information according to a water quality prediction model to obtain a target prediction result of the target water environment; and receiving a target prediction result sent by the cloud server, and determining a target monitoring result of the target water environment according to the target prediction result.

The invention belongs to the technical field of neural network acceleration design, and provides a low-bit neural network training method and an accelerator. The bit neural network training method and the accelerator are mainly characterized in the following three aspects. Firstly, a periodic function is used in the training process to approximate the gradient of an evidence obtaining function; secondly, the low-bit computing array is adjusted when the computing parallelism of the dimensions of the input channel and the output channel can be operated; and thirdly, the low-bit neural network accelerator uses full addition convolution. According to the bit neural network training method and the accelerator, four-bit wide operation is realized, relatively good network precision is maintained, relatively high calculation parallelism and calculation speed are achieved, calculation complexity and resource requirements are reduced, and the method and the accelerator are suitable for deployment of a resource-limited platform.

The invention discloses a method, a device and equipment for monitoring a vehicle-mounted air filter and a vehicle. The method comprises the following steps of: obtaining alarm pressure of an air inlet resistance sensor according to a theoretical life curve of the vehicle-mounted air filter; the state of the engine and the air inlet resistance measured by the air inlet resistance sensor are obtained in real time; and when it is detected that the engine is in the preset operation state, the air inlet resistance is larger than or equal to the alarm pressure and lasts for preset time, the vehicle is controlled to send out a first alarm signal. The vehicle-mounted air filter is monitored through the existing air inlet resistance sensor of the vehicle, other sensors and communication systems do not need to be added, the structure is simple, and cost is lower. The air inlet resistance measured by the air inlet resistance sensor is directly compared with the alarm pressure, only the air inlet resistance condition needs to be considered, factors such as the air quality and the air inlet amount do not need to be considered, the computational logic is relatively simpler, and the requirement for the operational capability of the ECU is low. The user can be accurately guided to maintain the air filter, and meanwhile, the user can reasonably plan the maintenance time, and normal work is prevented from being influenced.

The invention relates to the technical field of plant feature recognition, in particular to a plant leaf appearance character detection method and system, and the method comprises the steps: carrying out the image correction through the edge shape of a reference plane in the same visual field range; acquiring image information including a reference plane and a to-be-detected leaf, performing color space conversion, distinguishing the reference plane and the leaf by using a set color space range, and extracting a leaf contour in the image; determining the relationship between the size of a reference plane in the image and the number of pixels corresponding to the reference plane, and obtaining the width and length of the blade according to the number of length and width pixels of the minimum bounding rectangle frame of the blade contour; performing binarization processing on the image according to a set color space range, determining a ratio of a leaf contour pixel number to a reference plane contour pixel number, and obtaining a leaf area according to the area of the reference plane; and according to the Euclidean distance between the color of the to-be-detected leaf and each color in the colorimetric card list, taking the color corresponding to the minimum Euclidean distance as the relative chromaticity information of the to-be-detected leaf.

The invention discloses a carrying control method and a material carrying system.The material carrying system comprises an upper control system, TCSs, a track and carrying equipment running along the track, the track is divided into a plurality of independent and continuous track subareas, and each track subarea and each carrying equipment are configured to be managed only by one of the multiple TCSs; each of the TCSs is configured to manage at least one track partition. According to the method and the system, the track of one factory is divided into a plurality of independent and continuous track subareas, the independent and continuous track subareas are respectively managed through a plurality of TCSs, and each TCS only needs to manage a small number of track subareas, so that the computing power requirement on each TCS can be effectively reduced, the load of each TCS during operation can be reduced, and the operation efficiency of each TCS is improved. The timeliness of carrying instruction execution is improved, the carrying efficiency is improved, the computing power resources of each TCS can be utilized in real time, the utilization rate of the computing power resources can be effectively improved, and waste of the computing power resources is avoided.

The invention relates to an object positioning method and device based on the RFID technology and a storage medium, and the method comprises the steps: obtaining the measurement data of all positioning tags, and enabling the measurement positions to comprise the angle, the distance and the signal intensity; obtaining a first position of each positioning label based on the measurement data of all the positioning labels, and forming a target point set by the first positions of all the positioning labels; all the auxiliary labels are partitioned according to positions, and an auxiliary point set corresponding to each partition is determined based on measurement data of all the auxiliary labels in each partition; respectively calculating the Hausdorff distance between the target point set and each auxiliary point set, and selecting the auxiliary point set with the minimum Hausdorff distance with the target point set as a target auxiliary point set; and determining the position of the object based on the measurement data of all the auxiliary labels in the target auxiliary point set and the measurement data of the positioning label. Compared with the prior art, the method has the advantages of high anti-interference capability, low requirement on computing power of equipment and the like.

The invention relates to the technical field of road flatness detection, and particularly discloses a road flatness detection method and system based on laser scanning, and the method comprises the steps: S1, controlling a mobile carrier to move on a road at a preset constant speed, enabling a laser scanner to collect the coordinate information of a road surface at a preset frequency, the acceleration sensor collects acceleration information of the laser scanner in the vertical direction in real time; when the road flatness of the corresponding road section is judged to be possibly substandard according to preliminary analysis, whether the road section needs to be repaired or not is further judged, so that the computing power demand of an analysis control module in the road flatness detection process can be effectively reduced; the cost of road flatness detection is reduced, the risk of crash of the analysis control module is reduced, a proper vehicle speed is ensured in the flatness detection process of the driving carrier on the expressway, and potential traffic safety hazards caused by low-speed driving are avoided.

The invention relates to a concrete structure damage detection system and method based on percussion sound and acceleration information, the system comprises a percussion hammer, a microphone, an acceleration sensor, a data acquisition instrument and a processor, and the method comprises the following steps: collecting the percussion sound and acceleration information of a concrete structure by using the percussion hammer, the microphone and the acceleration sensor; carrying out data processing on the collected percussion sound and acceleration information to construct a data set; training, verifying and testing the constructed neural network architecture by using the data set, and establishing a damage state detection model; and obtaining percussion sound and acceleration information of the current concrete structure, inputting the percussion sound and acceleration information into the damage state detection model after data processing, and outputting to obtain a damage state detection result. Compared with the prior art, the damage state of the concrete structure can be accurately, conveniently and nondestructively detected in real time, and the damage state can be classified and identified.

The invention relates to novel expiration detection equipment based on the quantitative analysis method, and the novel expiration detection equipment is applied to the fields of human health assessment, disease screening, clinical diagnosis and the like. The expiration diagnosis is an emerging in-vitro diagnosis method, the clinical application range of the expiration diagnosis is gradually expanded, real-time and on-line detection of human expiration is achieved, a quantitative method for markers in expiration is established, reliable endogenous marker experimental data is obtained, a human expiration marker data model is constructed, and the real-time detection of the human expiration is achieved. Powerful support can be provided for exhaled air physiology research, clinical application of exhaled air diagnosis and environmental exposure research. And quantitative analysis is carried out on healthy people and sick people in different areas, and hit rules of different people are set according to weight values, so that the accuracy of quantitative analysis is improved, and the requirement on the computing power of data analysis is reduced.

The invention provides a battery temperature control method and device, a medium and electronic equipment. An electrothermal coupling model is constructed based on an off-line resistance and capacitance parameter identification method, a thermoelectric similarity principle, a three-dimensional unsteady state heat transfer theory and a Bernardi heat generation rate formula, the electrothermal coupling model is adopted to process initial data of a battery to obtain a battery temperature field, and the battery temperature field is calculated according to the battery temperature field. The battery overall temperature prediction value at the target time point is determined, so that compared with an existing scheme, a large amount of computing power requirements are omitted for temperature prediction, and finally the first processing mode or the second processing mode is executed according to the size of the battery overall temperature prediction value. Therefore, the problems that in an existing scheme, a large amount of computing power is needed in the process of controlling the TMS to stop the battery temperature to reduce power consumption, and operation on embedded software is difficult are solved.

The invention discloses an image matting model training method, an image matting method and device, equipment and a medium. The image matting model training method comprises the following steps: acquiring a data set for training; constructing a background probability model formula; marking each pixel point of all the image frames by using the sum obtained by the background probability model formula to obtain a probability thermodynamic diagram of whether the pixel is the background, and obtaining a binary portrait foreground label according to the thermodynamic diagram; using the chrominance component of the data set for training and the binarized portrait foreground label to train a color threshold model of the multivariate nonlinear regression model; brightness and chrominance components of image blocks are randomly acquired in a bbox range, and a nonlinear regression model is trained to represent a calculation model for fitting an edge part of a foreground portrait into an expected fuzzy weight under blue or green backgrounds under different conditions. According to the method, the foreground imaging effect can be optimized during superposition operation, and the computing power of an image matting algorithm is reduced.

The invention discloses a parameter adaptive optical frequency comb repetition frequency phase locking system and method, and belongs to the field of optical frequency combs. The system comprises an optical fiber optical frequency comb, a repetition frequency detection module, a phase discriminator module, a parameter adaptive phase lock control module, a PZT frequency actuator and a frequency reference source module, the repetition frequency detection module is used for detecting a repetition frequency signal of the optical fiber optical frequency comb, and the frequency reference source module generates a target repetition frequency signal; the two paths of signals generate repetition frequency phase errors through the phase discriminator module, the repetition frequency phase errors enter the parameter self-adaptive phase lock control module, the parameter self-adaptive phase lock control module generates phase lock signals through a parameter self-adaptive phase lock controller, and control over repetition frequency of the optical fiber optical frequency comb is achieved through a PZT frequency actuator. The phase-locked controller is based on a single-neuron network PID controller, and can optimize PID control parameters in real time according to the current phase-locked condition so as to improve the phase-locked precision and the autonomous locking capability of the repetition frequency of the optical frequency comb in a complex dynamic environment.

The invention discloses an instruction-driven end-side agent control method and system. A screen content understanding model is trained and tested; a screen content understanding data set required for making screen content understanding is collected; performing increment fine tuning training of a screen content understanding model by using the screen content understanding data set; training and testing an instruction translation model; collecting an instruction translation data set required for completing instruction translation; performing increment fine tuning training of an instruction translation model by using the instruction translation data set; and combining the intelligent agent and executing a user request: combining the screen content understanding model and the instruction translation model into the intelligent agent, and calling an Android ADB toolbox to execute the user request. According to the method, the phenomenon of low efficiency of an exploring type intelligent agent is avoided, and the requirement for high computing power of the multi-mode LLM is reduced.

The invention provides an underlying surface engineering and surface feature intelligent identification method based on small sample learning, which comprises the steps of collecting surface feature sample data, and further comprises the following steps: constructing a learning sample set; performing feature extraction; and carrying out sample identification. According to the underlying surface engineering and surface feature intelligent identification method based on small sample learning, a machine learning technology and a manifold learning algorithm are introduced into underlying surface engineering and surface feature intelligent identification, and a small sample feature learning technology is utilized to analyze and mine historical image samples, so that the underlying surface engineering and surface feature intelligent identification method is realized. And a newly collected image sample can be quickly identified and intelligently judged.

The invention discloses a wheel-legged robot based on a stability control technology and a control method of the wheel-legged robot. The wheel-legged robot comprises a fixed frame; the control box is mounted on the fixed frame; the obstacle avoidance detection module, the signal receiving module, the IMU module and the controller are all mounted in the control box; the power supply is mounted on the fixed frame; the leg connecting rod mechanisms are symmetrically arranged below the fixed frame; the pair of driving wheels are respectively connected to the bottom ends of the first connecting rod mechanism and the second connecting rod mechanism; a hub motor is embedded in the axis of the driving wheel and used for driving the driving wheel to roll. The robot is simple in structure, and efficient and stable control over the robot is achieved by controlling the hub motor, the front servo motor and the rear servo motor; and the control method is simple and efficient, and has low requirements on computing power.

The invention belongs to the crossing field of artificial intelligence and barrier-free assistance technology, and particularly discloses a sign language translation method and device based on a large language model.The method is applied to a terminal and comprises the steps that key points are extracted from sign language video frames; based on the coordinates, types and global indexes of the key points in the sign language video frames, obtaining fusion features through feature fusion; based on the fusion features corresponding to the sign language video frames in the frame window, time sequence modeling and classification are carried out, and a word prediction result of the frame window is obtained; a sliding window voting and deduplication mode is adopted, the candidate word sequence is preprocessed, a preprocessed word sequence is obtained, and the candidate word sequence is formed by connecting word prediction results of all frame windows in series according to a time sequence; based on the preprocessed word sequence, word error correction and statement generation are carried out by calling a large language model, and a sign language translation result is generated. According to the invention, sign language video translation can be realized for the terminal, and the accuracy and fluency of the translation result can be guaranteed.

The invention provides a three-dimensional fluorescence spectrum analysis method based on unsupervised contribution modeling. The method comprises the following steps: acquiring an original three-dimensional fluorescence spectrum of a sample; carrying out independent component analysis on the three-dimensional fluorescence spectrum, and reconstructing the spectrum by utilizing an analysis result; and carrying out modeling on the reconstructed spectrum by using a multivariate curve resolution-alternating least square algorithm to obtain an analysis report. According to the method, interference signals such as scattering in an original three-dimensional fluorescence spectrum are automatically removed through combination of two two-dimensional unsupervised modeling algorithms, then modeling is carried out on a reconstructed spectrum, dependence of three-dimensional fluorescence spectrum analysis on data, tools, platforms and personnel is reduced, good fault tolerance and accuracy are guaranteed while flexibility is improved, and the method is suitable for popularization and application. And reliable technical support is provided for water quality monitoring.

The invention relates to a machine learning classification privacy protection method and device based on homomorphic encryption and a storage medium, and the method comprises the steps: supposing that a data owner does not have the homomorphic encryption and decryption capability, introducing a conversion mechanism of an AES ciphertext and a homomorphic ciphertext into a participant, and guaranteeing that the data owner does not have the homomorphic encryption and decryption capability except the data owner; other participants cannot pry the plaintext information of the data. According to the method, on the premise that the security is not reduced, the requirement on the computing power of a data owner is reduced, and the overall efficiency of the model is not excessively influenced.