1149results about How to "Reduce resource requirements" patented technology

The invention discloses an electric power facility visual monitoring system and method, and relates to the technical field of electric power facility operation and maintenance, the system comprises the following components: a multi-source heterogeneous data acquisition module, a multi-source heterogeneous data processing module, a real-time risk assessment and maintenance strategy adjustment module and a visual display module; according to the method, the electric parameters, the environment data, the image video data, the audio data and other multi-source heterogeneous data are integrated, the intelligent fusion algorithm and the association analysis model are utilized, the potential association relationship among different types of data is deeply mined, and therefore the operation condition of the electric power facility can be reflected more comprehensively and accurately, and the operation efficiency of the electric power facility is improved. According to the method, the accuracy and integrity of monitoring data are improved, the system can timely find and pre-judge potential risks, powerful decision support is provided for operation and maintenance personnel, and safety risks and maintenance cost increase caused by incomplete or inaccurate monitoring are effectively avoided.

The invention discloses a multi-task collaborative robot scheduling method and system oriented to an intelligent factory, and the method comprises the steps: receiving the input data of a plurality of tasks to be executed by the intelligent factory, the input data comprising the task type, the dependency relationship between the tasks, the task priority, the resource demand of each task, and the expected completion time of each task; determining an initial task execution sequence of the tasks according to the task types, the dependency relationships and the task priorities; obtaining current resource state data of all available robots in the intelligent factory; according to the resource demand of each task, the expected completion time of each task, the initial task execution sequence and the current resource state data of all available robots, a scheduling algorithm is adopted to generate a multi-task cooperative scheduling rule; and controlling the plurality of target robots to execute the plurality of tasks according to the multi-task cooperative scheduling rule. According to the method, the resource utilization rate of the robot in the intelligent factory can be effectively improved, and the task execution efficiency is optimized.

The invention discloses a video monitoring safety control system and method based on the Internet of Things, particularly relates to the technical field of video monitoring safety control, and is used for solving the problem of low video monitoring control efficiency. Multi-source data are collected through an Internet of Things sensor network, the time sequence consistency is ensured by adopting a clock synchronization protocol, the data are preprocessed, the accuracy is ensured, then correlation analysis is carried out by utilizing a Bayesian network and principal component analysis, historical information and scene modes are extracted, risk levels of all areas are determined, and the risk levels of all the areas are determined. And determining a resource scheduling priority according to the risk level. And by simulating scene change, the risk early warning capability of the system is improved. According to the system, camera resources are dynamically scheduled through a collaborative optimization algorithm, monitoring resources of a high-risk area are preferentially allocated, redundant monitoring is reduced, power consumption and bandwidth use are optimized, and therefore the response speed and precision of video monitoring in a complex environment are improved, and the safety of video monitoring is improved.

The invention discloses a large model service platform for high-efficiency concurrent computing, particularly relates to the technical field of data analysis, and realizes optimal management of the large model service platform for high-efficiency concurrent computing through a task allocation module, a node health assessment module, a node division module, a dynamic adjustment module and a fault identification module. The task distribution module reasonably distributes computing resources according to task priorities and real-time monitoring data; the node health assessment module accurately assesses node health conditions through preprocessing and feature extraction; the node division module classifies the nodes according to the health assessment result; the dynamic adjustment module dynamically adjusts a resource allocation strategy by using fuzzy logic; the fault recognition module recognizes and repairs fault nodes in time by further analyzing the health conditions of the high-load nodes; the overall performance and the resource utilization rate of the system are improved, efficient and reliable execution of intelligent financial services is ensured, and the stability and the reliability of the system are enhanced.

The invention discloses a cloud computing storage, calculation and training resource allocation method based on reinforcement learning and dynamic resource scheduling. The method comprises the following steps: S1, establishing a unified resource scheduling framework; s2, constructing a state space and an action space; s3, generating a state representation for predicting future resource demands; s4, initializing a reward function of key performance index weighted comprehensive evaluation; s5, generating an initial storage resource, computing resource and training resource allocation strategy; s6, continuously adjusting a reward function of the reinforcement learning model based on the self-supervised learning feedback signal; s7, dynamically adjusting a distribution scheme of the storage resources, the computing resources and the training resources through quick response of the resource scheduling framework; s8, optimizing resource allocation among a plurality of cloud computing nodes or platforms; s9, automatically identifying different stages of the current task through a self-supervised learning model; and S10, finally generating an optimal allocation scheme. According to the staged resource scheduling method, the condition of overload or insufficiency of resources is effectively avoided, and the task execution time is greatly shortened.

The invention discloses a method for analyzing a matching degree between a demand and an output result based on text semantics. The method comprises the following steps: step 1, labeling a data set; step 2, technical document preprocessing; 3, training and predicting a single-parameter model; 4, integrating prediction results of the multi-parameter model; the method has the beneficial effects thatthe method is simple; deep learning and the NLP technology are applied to the field of project association degree calculation of enterprise project management for the first time. Calculating an association matching degree between the two projects according to project requirements and result description; the associated project positioning difficulty is effectively reduced; meanwhile, the demand side can be helped to quickly and efficiently locate high-quality projects adapting to the demand of the demand side; time and resource investment for achievement screening and matching are greatly reduced, the association matching degree between projects is calculated by means of text data of existing project achievement technical documents and project declaration guidelines, and then large enterprises are assisted in screening high-quality projects with the high matching degree in the project bidding and tendering link.

The invention relates to the technical field of building engineering, in particular to a steel structure construction quality monitoring system based on an unmanned aerial vehicle and three-dimensional scanning, which comprises a data acquisition module, a data preprocessing module, a construction defect identification and risk assessment module and a quality assessment module, the data acquisition module acquires quality monitoring data of a construction site in real time; the data preprocessing module is used for preprocessing the acquired quality monitoring data; the construction defect identification and risk assessment module identifies steel structure defects in the construction process and assesses the severity of the steel structure defects; and the quality evaluation module performs comprehensive quality evaluation on the preprocessed quality monitoring data and provides corresponding repair suggestions. According to the method, key problem areas can be quickly identified, a construction resource allocation strategy is optimized, and comprehensive and accurate data support is provided for steel structure quality evaluation in a complex construction environment.

The invention discloses a novel dry bulk cargo port multi-scene simulation method based on a large language model, belongs to the technical field of port production and operation, and realizes the expansion of training samples by using the general knowledge of the large language model; based on an Adam optimizer, training the initial model by adopting the training set, and verifying the initial model by adopting the verification set; evaluating a multi-scene simulation result under the guidance of the model based on a test result of the test set, and performing iterative expansion training by utilizing a prompt learning mechanism in combination with multi-scene data to obtain a trained large language model; extracting key semantics of scene features through the trained large language model according to a current operation scene of a port; performing simulation resource configuration reasoning on the extracted key semantics; a novel dry bulk port simulation model in a corresponding scene is constructed by adopting AnyLogic simulation, and a big language model training corpus and a training set are updated according to a simulation preview result and port actual operation data. According to the invention, the production operation decision can be optimized.

The invention discloses a real-time monitoring and early warning method for the leakage state of a long-line type public facility, and relates to the technical field of monitoring and early warning of the leakage state of the long-line type public facility, and the method specifically comprises the following steps: analyzing environment dynamic response information stored in a central monitoring system; respectively generating a soil humidity change coefficient, an environment pressure response index and a temperature influence compensation coefficient; constructing a leakage risk assessment model for the generated soil humidity change coefficient, the environment pressure response index and the temperature influence compensation coefficient, and generating a leakage risk coefficient; and determining a preset leakage risk coefficient threshold value, comparing the determined leakage risk coefficient threshold value with the generated leakage risk coefficient, evaluating whether a leakage risk exists in the pipe gallery area close to the earth surface, and respectively generating a normal signal and an abnormal signal according to an evaluation result. According to the invention, the problem of underground pipe gallery leakage monitoring is solved, abnormal signals are accurately distinguished, and the monitoring accuracy and safety are improved.

The invention relates to the technical field of biology, and discloses a chemically modified small nucleic acid drug zero sample prediction method and system based on a large language model and multi-view deep learning. The method comprises the following steps: collecting chemically modified RNA data; performing Token conversion by using an SMILES token regular expression, and performing embedding by using a pre-trained large language model; chemical modification position information is obtained and combined with SMILES in an embedded mode; the RNA sequence is embedded; processing the comprehensive embedded vector by using a self-attention mechanism; fusing RNA sequence embedding and comprehensive embedding by adopting a cross attention mechanism; performing feature extraction and dimension reduction by using a three-dimensional convolutional neural network; and inputting the features into a full-connection layer to predict the drug efficiency. According to the method, high-precision zero-sample prediction of novel chemical modification is realized, the generalization ability and prediction precision of the model are remarkably improved, and a flexible, efficient and accurate method is provided for design and development of RNAi drugs.

The invention provides an energy consumption simulation method and system for a refrigerating machine room, and is applied to the field of energy consumption data processing. By establishing the real-time monitoring and analysis system, the machine room can effectively integrate dispersed energy consumption data, the peak and valley load time periods are defined, and the energy consumption prediction and management strategy is dynamically adjusted. According to the systematic method, environmental factors and equipment efficiency are combined, energy consumption influences of various load types are deeply analyzed, it is ensured that a machine room can flexibly cope with dynamic changes of equipment, resource waste and operation cost are reduced, meanwhile, key influence factors are recognized, combined load testing is carried out, and the testing efficiency is improved. Management personnel can formulate targeted cooling strategies and optimization schemes, the energy efficiency is improved, decision support of data driving is achieved, and therefore the overall operation efficiency and the sustainable development capacity of the machine room are improved.

The invention discloses a remote sensing image ground object segmentation method based on hierarchical feature extraction and decoding, which relates to the technical field of remote sensing image processing, and comprises the following steps: step 1, collecting and preprocessing a remote sensing image, and constructing a training sample set; 2, constructing an image segmentation network, wherein the image segmentation network comprises a hierarchical encoder, a receptive field module unit and a network decoder which are connected in sequence; 3, training and optimizing the image segmentation network by adopting the training sample set to obtain a remote sensing image ground object segmentation model; and 4, collecting a remote sensing image to be segmented, and inputting the remote sensing image to the remote sensing image ground object segmentation model to obtain a ground object segmentation result. According to the method, the advantages of the advanced visual basic model and the U-shaped network architecture are combined, so that rapid and accurate segmentation of the ground target is realized.

The invention discloses a receipt identification method and device, equipment and a medium, and relates to the technical field of bill content identification. The method comprises the following steps: acquiring a receipt image through a preset acquisition mode, and identifying the receipt image by using a preset optical character identification model to determine text content and corresponding position information in the receipt image; performing information matching on the receipt image in a preset multi-mode large model prompt word bank to obtain a first prompt word, and inputting the first prompt word, the text content and the position information into a preset multi-mode large model to obtain a key information key value pair; and performing information matching on the receipt image in a preset language large model prompt word library to obtain a second prompt word, and inputting the second prompt word, the text content, the position information and the key information key value pair into a preset language large model to obtain a target key information key value pair. Through the technical scheme of the invention, the accuracy of identifying the key field content of the non-standard receipt can be improved.

The invention discloses an electric power data anomaly detection analysis method and system based on big data, and relates to the technical field of big data, and the method comprises the steps: obtaining an interaction influence relation matrix Iinter between devices through the real-time collection of a feature vector matrix Xpre of the devices; and using maximum likelihood estimation to carry out parameter optimization on the equipment association interaction influence analysis model to iterate an interaction influence relation matrix Iinter, predicting an abnormal mode under the interaction influence between the equipment, obtaining an abnormal model matrix Aetect, and predicting the future interaction operation state of the equipment according to the generated comprehensive abnormal score Sabn of the equipment. According to the method, the problem of cross-equipment abnormity is solved, the fault positioning and emergency response capabilities of the power system are improved, equipment shutdown and resource waste caused by unidentified abnormity are also remarkably reduced, and the stability of the power system is guaranteed.

The invention relates to the technical field of surface defect detection, and provides an injection molding surface defect detection method for an engineering plastic mold. The method comprises the following steps: transferring an engineering plastic mold to a defect detection area, and acquiring injection molding surface images of the mold by an image acquisition device to obtain a plurality of injection molding surface images; a splicing coordinate system is established, the multiple images are spliced on the splicing coordinate system, and an injection molding surface stereogram is obtained; constructing a defect library, and detecting the stereogram based on the defect library to obtain a defect detection result; mold injection molding constraints are extracted, and when the detection result does not meet the constraints, a maintenance instruction is generated according to the detection result and the injection molding constraints; and based on the maintenance instruction, the mold is transferred to the maintenance area for maintenance. The technical problems that the detection efficiency of the injection molding surface defects of the engineering plastic mold is low, the accuracy rate is insufficient, and subsequent maintenance is not timely are solved, efficient and accurate detection and automatic maintenance processes of the injection molding surface defects are achieved, and the technical effects of improving the mold quality and the production efficiency are achieved.

The invention provides a training strategy determination method for a computing power cluster using chips of various models in a mixed mode and a related device, and relates to the technical field of graphic processing chips, computing power clusters, large models, model training and the like. The method comprises the following steps: determining each target model chip forming a computing power cluster and a corresponding quantity proportion; for each target model chip, determining performance evaluation of the target model chip in a preset performance type and weight distribution of each performance type under each alternative training strategy; for each alternative training strategy, performing weighting processing on the performance evaluation under each corresponding preset performance type in combination with each corresponding number proportion by using the weight distribution, matched with each target model chip, of the alternative training strategy, so as to obtain corresponding comprehensive training evaluation; and determining a target training strategy based on the evaluation parameter of each comprehensive training evaluation. By applying the scheme, the training strategy which is suitable for being used by the computing power cluster which mixedly uses multiple types of chips during training of a large model can be quickly determined at low cost.

The invention discloses a blood vessel segmentation method and device based on large model fine tuning, equipment and a medium, and the method comprises the steps: obtaining a blood vessel image, carrying out the preprocessing of the obtained blood vessel image, and obtaining a training set; performing fine tuning on a pre-trained image segmentation model by adopting the training set, so that the image segmentation model learns related features of blood vessel segmentation; wherein the low-rank adaptive fine tuning technology is adopted to perform fine tuning on the image segmentation model: two parallel low-rank matrixes are inserted into a Transform layer in the image encoder, the input is processed by the two low-rank matrixes in sequence, and the processing result and the original output are added to obtain the final output. According to the method, the defects of a large model in a blood vessel segmentation task are overcome through efficient parameter fine tuning and a multi-scale feature fusion module. Accurate blood vessel segmentation can be realized under limited annotation data, the difficulty and cost of medical image segmentation tasks are greatly reduced, and the method can be widely applied to the technical field of medical image processing.

The invention relates to the technical field of medical drug dose analysis, in particular to a deep learning-based advanced nasopharyngeal carcinoma multi-target dose prediction system, which comprises a data acquisition module, an automatic sketching module, a dose prediction model construction module, a diversified dose generation module, a model training and optimization module, a result output and visualization module and the like. The automatic sketching module is used for realizing accurate and automatic sketching by using a multi-modal fusion and knowledge distillation framework; the dose prediction module accurately extracts features based on a UNetr network model and generates a multi-channel dose prediction result; the diversified dose generation module generates diversified complementary dose distribution through a specific method and a constraint mechanism; the model training and optimizing module effectively improves the model performance; the result output and visualization module presents results in various visual modes for doctors to make decisions; the nasopharynx cancer treatment dose prediction accuracy and individuation degree can be improved, the automatic sketching precision and efficiency can be improved, the clinical decision process can be optimized, and development of related technologies can be promoted.

The invention discloses a dam effect quantity prediction method and system, a medium and equipment, and relates to the technical field of safety engineering, dam monitoring data is acquired, influence factors of a dam effect quantity are determined, and effect quantity influence factor samples are constructed; performing dimension reduction processing on the influence factor sample through a principal component analysis method to obtain an influence factor matrix after decoupling and dimension reduction; coupling the long and short-term memory neural network with an artificial fish swarm algorithm, constructing a dam effect quantity prediction model, performing parameter setting on the dam effect quantity prediction model, and optimizing the initial weight and the number of neurons of the long and short-term memory neural network by using the artificial fish swarm algorithm; training the optimized long and short term memory neural network until convergence, and obtaining a trained dam effect quantity prediction model; and performing prediction and anti-normalization processing to obtain an effect quantity prediction result. And the accuracy, the stability and the efficiency of dam effect quantity prediction are improved.

The embodiment of the invention provides a subway station large passenger flow risk management and control method and system, and the method comprises the steps: dividing a key region according to a station structure, obtaining the risk level of the key region through employing a passenger flow data-based risk threshold dividing method, and calculating the risk level of a station; taking the passenger flow management and control comprehensive cost and the station coupling passenger flow risk as optimization targets, and enabling the risk level of the station to meet a preset risk level; and carrying out optimization solution on the optimization target, selecting an optimal solution according to different scenes, and taking the optimal solution as a station large passenger flow risk management and control strategy. According to the method, the large passenger flow risk in the station is prevented by generating the dynamic management and control strategy of the passenger flow risk.

The invention discloses a digital twinborn-based campus comprehensive service management method and system, and relates to the technical field of campus canteen service management. According to the method, a digital twinborn model is established, and a plurality of catering areas in a canteen are monitored; the utilization rate Rct is calculated by extracting the diner number characteristic, the food supply state and the queuing duration characteristic in the regional real-time state set. Therefore, the system can judge whether the resource allocation of each catering area is reasonable or not in advance. According to the method, a food consumption rate Fcx, a dining table turnover coefficient Pds, a customer average action path length Zyr, a tableware equipment load rate Ssy and a saturation evaluation index Bpzs are calculated by performing preprocessing and standardization processing on a regional real-time state set, and a system automatically calculates an unbalance rate Shs when the saturation evaluation index Bpzs exceeds a saturation threshold value K; whether current personnel scheduling, food supply and equipment use are in a balanced state or not is analyzed, and an optimization instruction is triggered in time, so that resource waste and reduction of service quality are avoided.

The invention relates to the technical field of multi-agent systems, and discloses a multi-agent and multi-mode alignment interaction method and system for a review scene, and the method comprises the steps: mapping real-time multi-mode data into a local belief vector through a heterogeneous agent cluster, and calculating the spatial divergence to determine a global cognitive entropy; interaction time sequence gating logic based on entropy flow differential is established, and active alignment is triggered when the global cognitive entropy is larger than a convergence threshold value and a first-order derivative is smaller than a damping threshold value; according to the method, through entropy flow differential control and an active semantic disturbance closed loop, decision oscillation of the multi-agent system in the non-cooperative environment is eliminated, and the reliability of the multi-agent system in the non-cooperative environment is improved. Mathematical convergence of the belief state in an adversarial disguise scene is realized, and the system identification precision in a complex interactive game is improved.

The invention discloses a power system time synchronization tester detection management method, and relates to the technical field of power system testing and monitoring, and the method comprises the steps: S1, connecting each time synchronization tester to a central management platform, and starting initial time synchronization; s2, generating a synchronous correction instruction based on the time deviation data reported by each tester; s3, sending a synchronous correction instruction to each tester, and performing time deviation correction; s4, monitoring the time deviation of each tester in real time, and ensuring the synchronization precision after correction; s5, when it is detected that the time deviation exceeds a preset threshold value, triggering a re-correction process; according to the power system time synchronization tester detection management method, the problem of synchronization coordination caused by inconsistent time deviation correction in a multi-device synchronization process in the prior art is solved.

The invention discloses a wind power blade unmanned aerial vehicle inspection method and device, and belongs to the technical field of wind driven generators, and the wind power blade unmanned aerial vehicle inspection method comprises the steps: carrying out the unmanned aerial vehicle inspection of the front surface of a wind power blade to form first inspection information, and carrying out the unmanned aerial vehicle inspection of the back surface of the wind power blade to form second inspection information, the first inspection information comprises a first track point data set and a blade front image set, and the second inspection information comprises a second track point data set and a blade back image set; according to the first inspection information and the second inspection information, in combination with a wind power blade defect judgment condition, obtaining a wind power blade unmanned aerial vehicle inspection result; according to the wind power blade unmanned aerial vehicle inspection result, whether an unmanned aerial vehicle re-inspection mode is started is judged, if yes, the unmanned aerial vehicle re-inspection mode is started, and the wind power blade unmanned aerial vehicle inspection result is updated; and if not, the re-inspection mode is not started. The method can reduce the influence on the wind power plant operation and power generation efficiency, and improves the accuracy of the wind power blade unmanned aerial vehicle inspection result.

The embodiment of the invention discloses a message parsing system and a message parsing method.The message parsing system comprises a fixed parser and a programmable parser, the fixed parser is configured to receive and parse an original message, and in response to the fact that the message is not parsed completely, the programmable parser is configured to parse the original message; message analysis information including the analyzed result and the message header is sent to a programmable analyzer, and the programmable analyzer is configured to continue to perform message analysis according to the received analyzed result and the received message header to obtain a message analysis result. Therefore, according to the embodiment of the invention, the fixed parser and the programmable parser can be combined, the message protocol which can be parsed by the fixed parser is parsed by the fixed parser, and the message protocol which cannot be parsed by the fixed parser is parsed by the programmable parser, so that the message parsing system can meet the low-delay requirement and the flexibility requirement; and meanwhile, the system resource demand is also reduced.

The invention belongs to the technical field of information, and discloses a time-phased physical examination reservation and management system based on intelligent optimization, and the system comprises a physical examination reservation client which is used for a user to complete the selection of a physical examination package, the reservation of a time period and the like online through a mobile phone client according to the actual condition of the user, the self-service machine only needs to print a physical examination guide sheet and sign in and take a number on the site on the same day, and then the whole physical examination process is completed according to system prompts; and the system management end is used for project maintenance, personnel management, system management and the like of physical examination center management personnel. According to the invention, by introducing a time-phased physical examination appointment mode, an intelligent hospital guide service is provided for a user, and a set of time-phased physical examination appointment and management system based on intelligent optimization is designed. The result is that the user only needs to make physical examination before the reservation time period through the online physical examination reservation, and examination is carried out according to the intelligent hospital guide prompt. Through time-phased appointment management, physical examination persons can be averagely arranged in all time periods, the number of physical examination persons on site is effectively controlled, the physical examination queuing time of the users is shortened, and the satisfaction degree of the physical examination users is further improved.

The invention discloses a feeding method and device capable of automatically adding pig raising feed, and aims to optimize feed management in the pig raising process through an automatic and intelligent technology. Comprising the following steps: monitoring environmental parameters such as temperature, humidity and air quality in a pig house in real time through an environment monitoring device, and transmitting data to a central control system; the central control system dynamically generates and adjusts a feed formula according to environmental parameters and growth stages and health conditions of pigs, and accurately mixes feed components through an automatic proportioning device. The feed trough sensor monitors the residual amount of feed in the feed trough, and when the feed is insufficient, the system automatically controls the feed conveying device to supplement the feed. In the feeding process, the central control system adjusts the feed adding frequency and the feeding amount according to real-time data, and the cleaning device is combined with the ultrasonic cleaning and nano-coating technology, so that the cleanliness and long-term stable operation of the equipment are ensured.

The invention discloses a respiratory rehabilitation training data management method and system based on artificial intelligence, and relates to the technical field of data processing, the respiratory rehabilitation training data management method based on artificial intelligence comprises the following steps: obtaining training data of a respiratory rehabilitation person, and preprocessing the training data to obtain a multi-source data set; constructing a teacher network based on the high-performance model, training the teacher network through the multi-source data set, constructing a student network, and training the student network based on the trained teacher network; optimizing the trained student network by using a differential evolution algorithm to obtain a student network optimized model; and training a reinforcement learning agent by using data generated by the model after student network optimization, and automatically adjusting rehabilitation training parameters of the respiratory rehabilitation personnel through the reinforcement learning agent. According to the invention, computing resource and storage requirements are reduced, data-driven decision is realized, and a personalized rehabilitation training scheme is provided.

The invention discloses a lithium ion battery state-of-charge estimation method and device based on transfer learning, and belongs to the technical field of battery management systems, and the method comprises the steps: obtaining current, voltage and temperature original data, and carrying out the normalization processing of the current, voltage and temperature original data, and constructing time sequence data; sequentially performing convolution processing and pooling processing on the time sequence data, and substituting the time sequence data into an LSTM network to extract time sequence features; carrying out transfer learning processing on the time sequence characteristics to obtain an SOC value of primary version prediction; and the SOC value after Kalman filtering correction is obtained from the SOC value predicted by the primary edition through a Kalman filter.

The invention relates to the technical field of metal button surface defect real-time detection methods, in particular to a metal button surface defect real-time detection method based on dry deep learning, which comprises the following steps: acquiring a to-be-detected metal button image and preprocessing the to-be-detected metal button image to obtain a to-be-detected metal button image; and inputting the preprocessed image into a deep learning network model to extract a defect region, optimizing and verifying the extracted defect region, and finally generating a statistical result of defect detection. According to the method, dynamically regulated progressive boundary construction, adaptive gradient mapping, Gaussian fitting detail enhancement, multi-dimensional difference correction factors, sectional defect region generation and other algorithms are innovatively introduced in sequence, so that boundary recognition and detail optimization are progressive layer by layer. The innovative multi-stage optimization design enables the data output of each step to be connected with the subsequent steps, guarantees the continuity and consistency of defect detection, and avoids the problems of boundary error and information loss caused by algorithm independence in a traditional method.