505results about How to "Avoid slow recognition" patented technology

The invention discloses an industrial equipment real-time monitoring system based on edge computing, and the system comprises the following steps: collecting, processing and analyzing the operation data of industrial equipment through distributed edge computing nodes, constructing a node cooperation mechanism, and completing the processing of the data; the dynamic data scheduling module is used for dynamically optimizing a data transmission path and task allocation according to node performance; the intelligent model dynamic migration module is used for adjusting lightweight model deployment in combination with node resource conditions; the multi-modal fault sensing module is used for extracting and analyzing multi-modal features by using a cross-node collaborative algorithm; the edge node self-adaptive management module is used for adjusting node operation parameters in real time and repairing an abnormal state; and the equipment relevance reasoning module is used for constructing an equipment relevance model through a graph calculation method and generating a real-time relevance graph. According to the method, efficient monitoring and anomaly detection of industrial equipment are realized through resource optimization and intelligent analysis of collaborative edge computing nodes.

The invention provides an intelligent access method and system for small accessories of an unmanned aerial vehicle, and relates to the technical field of intelligent access. Generating an action feature vector through an action semantic understanding model; identifying key action nodes and generating a basic action sequence; constructing an action semantic mapping model; performing mapping analysis based on the execution equipment capability; generating an equipment control instruction; collecting action execution state data; establishing an execution evaluation model; generating a compensation instruction and updating the control instruction; and intelligent access operation is completed. According to the method, intelligent storage and taking of the unmanned aerial vehicle accessories are achieved, and the operation efficiency and precision are improved.

The invention relates to the technical field of fault diagnosis, in particular to an intelligent monitoring system for a sewage treatment pipeline, which comprises a topological structure analysis module for collecting pipeline network nodes and communication information, calculating the connection number and geometric attributes of multiple nodes, identifying the pipeline length and the connection direction between adjacent nodes, and analyzing the pipeline network nodes; and calculating the length of a loop formed among multiple nodes and a topological structure, analyzing structural stability features, and obtaining topological geometric feature data. According to the invention, by analyzing the geometric and topological data of the sewage treatment pipeline system, the understanding of the stability of the pipeline structure is optimized, so that the accuracy of fault prediction is improved, the flow and pressure changes are monitored in real time, and the response to dynamic adjustment and abnormal phenomena in the system is more sensitive and timely; the flow gradient and the pressure difference are analyzed, so that the recognition capability of an abnormal mode is enhanced, a problem area can be quickly recognized and processed, the operation efficiency is improved, and the environmental risk and the maintenance cost are reduced.

The invention discloses a text keyword desensitization method, system and equipment based on a large model and a medium, belongs to the technical field of large model and natural language processing, and aims to solve the technical problem of how to automatically identify and desensitize sensitive information, improve the accuracy of sensitive information identification and further improve the text processing efficiency. According to the technical scheme, the method comprises the steps of data collection; preprocessing the data; marking sensitive information; segmenting the data; selecting a basic model; finely adjusting the model; training the fine-tuned large language model, and obtaining and storing the trained fine-tuned large language model; preprocessing the to-be-processed text data; detecting sensitive information, formulating a desensitization strategy and implementing desensitization processing, performing desensitization verification on the text after desensitization processing, and optimizing and improving the fine-tuned large language model according to a verification result; and packaging the optimized and improved fine-tuned large language model into an API (Application Program Interface) service for an application program to call.

The invention belongs to the technical field of safety monitoring of subway tunnels, and particularly relates to a subway tunnel water leakage monitoring method and system based on multi-modal data fusion, and the method comprises the steps: obtaining multi-source sensor time sequence data in a subway tunnel, carrying out the feature extraction, and extracting feature parameters related to water leakage; carrying out fusion analysis on the characteristic parameters through a data fusion algorithm, and evaluating the position and severity of water leakage, including extracting time-space correlation characteristics of data by using a dynamic time warping method; after feature extraction is performed on the data set and the space-time correlation features by using a convolutional neural network and a recurrent neural network, feature fusion is performed; and performing anomaly detection and classification on the fused high-level features by using a neural network method to obtain the position and severity of the leakage water. The advanced data fusion technology and the feature extraction algorithm are adopted, various collected data can be analyzed in real time, abnormal conditions can be found in time, and early warning is carried out.

The invention discloses a bad traffic monitoring method and system based on an LSTM and a clustering algorithm, belongs to the technical field of network security, and efficiently converges network traffic by using a BPF technology. The method comprises the following steps: accurately identifying fragmented traffic through an LSTM network, and capturing a complex dependency relationship by using a special memory unit structure of the LSTM network; performing session recombination by adopting an FCM clustering algorithm, restoring a complete network session, and quickly identifying malicious traffic through feature matching; and finally, timely treatment measures are taken for the confirmed malicious traffic. According to the method, rapid and accurate detection and identification of bad website traffic fragmentation behaviors are realized, and the technical problem that a traditional method is difficult to deal with is solved; through efficient caching and recombination mechanisms, the efficiency and accuracy of malicious traffic processing are improved; intelligent disposal measures are adopted, and it is ensured that malicious traffic is effectively restrained in time; the overall protection capability of network security is improved, and enterprise network resources and user data security are protected.

The invention relates to a river water ecological space global monitoring method and system, and the method comprises the steps: recognizing and segmenting a water body image from an RGB image photographed by an unmanned plane based on a U-Net deep learning model, and forming a water body recognition image; converting and reconstructing the RGB image aerially photographed by the unmanned aerial vehicle into a water body hyperspectral image based on a hyperspectral information continuation reconstruction model; the U-Net deep learning model extracts a multispectral segment sequence of pixel points of a water body in the water body hyperspectral image and sends the multispectral segment sequence to the classification regression model; and calculating the multispectral segment sequence based on a classification regression model so as to label a concentration classification result of chlorophyll a in a water body hyperspectral image, thereby realizing visualization of chlorophyll concentration. The method gets rid of limitation of technical means and time acquisition of hyperspectral data acquisition, and realizes the effect of acquiring river global images at any time and at any moment.

The invention relates to the technical field of PCBs (printed circuit boards), in particular to an intelligent PCB detection and repair method, which comprises the following steps: S1, initializing module configuration: automatically identifying the type and structure information of a PCB; s2, a dynamic detection stage: collecting electrical parameters of the PCB, and generating fault position information; s3, circuit dynamic characteristic compensation: analyzing the nonlinear distortion characteristic of the circuit, and calculating and executing dynamic compensation; s4, an automatic repair stage: generating a repair scheme, including solder joint repair, line reconstruction or signal compensation; and S5, an intelligent recheck stage: performing comprehensive retest on the repaired PCB, verifying the performance of the repaired area, and generating a repair evaluation report. According to the invention, the repair precision and efficiency are improved, the whole repair process is intelligent and automatic, the risk of manual intervention is reduced, and the overall repair efficiency and the long-term reliability of the PCB circuit are improved.

The invention discloses a security protection method and system for an information management system and a storage medium, and relates to the technical field of security protection.During operation of the system, when a user logs in or carries out sensitive operation, multiple identity verifications are carried out, an identity verification coefficient AC is obtained through calculation, various security parameters in the system are collected and updated in real time, and the security protection efficiency is improved. A high-strength encryption algorithm is applied to data to be transmitted, an encrypted data packet is generated, a transmission encryption coefficient EC is obtained through calculation, corresponding access permissions are set for different data based on user roles, identities and access content, an abnormal monitoring coefficient ADC is obtained, and a comprehensive safety coefficient SEC is obtained through calculation according to an identity verification coefficient AC, the transmission encryption coefficient EC and the abnormal monitoring coefficient ADC. The overall safety of the system is dynamically evaluated, abnormal behaviors are recognized and logs are recorded by comparing the comprehensive safety coefficient SEC with a preset threshold value, and when the system detects an abnormal access mode, a safety alarm is triggered and countermeasures are taken.

The invention discloses an AI-based marketing commodity short video generation method and device, a medium and a product, and relates to the field of short video generation. The method comprises the following steps: constructing a short video industry knowledge base; obtaining question information input by a user; searching answers from a short video industry knowledge base according to the question information; obtaining a video selected from the answers by the user as a reference short video, and obtaining commodity information uploaded by the user; the commodity information comprises commodity selling points and commodity materials; retrieving a reference copywriting from the short video industry knowledge base according to the commodity selling point, taking the reference copywriting as knowledge supplementary information, generating a script copywriting according to the reference short video and the knowledge supplementary information, and generating picture information according to the script copywriting and the commodity material; and editing the picture information to obtain a short video of the commodity information. The output efficiency of the running amount materials can be improved.

The invention discloses an unmanned warehouse shelf-oriented automatic cargo checking method and system, and the method is characterized in that the method comprises the following steps: employing a target detection algorithm to extract a cargo boundary frame from captured continuous cargo images; for each cargo bounding box, calculating shape features, color features, texture features, size features and local features of the cargo in the cargo bounding box, and splicing feature vectors of the cargo according to the features; pre-defining cargo templates of different types of cargos, matching the cargo templates with the feature vectors of the cargos so as to obtain a category label of each cargo bounding box, and iteratively optimizing the target detection algorithm through a loss function; and tracking the goods extracted by the target detection algorithm by using a Kalman filtering target tracking algorithm, and obtaining the position and quantity information of the goods from the continuous goods images.

The invention discloses an intelligent operation and maintenance fault diagnosis system based on a communication network, which belongs to the technical field of communication networks and comprises a data acquisition preprocessing module, a network dynamic evolution module, a quantum auxiliary diagnosis module, a fault prediction evaluation module, a fault decision module, a fault operation and maintenance module and a fault feedback module. By integrating the data acquisition and preprocessing module, the network dynamic evolution module and the quantum auxiliary diagnosis module, the system can acquire and process the state information of a communication network in real time, and quickly position and accurately identify a fault scene in combination with the high efficiency of quantum calculation, so that the fault diagnosis time is remarkably shortened, the diagnosis accuracy is improved, and the fault diagnosis efficiency is improved. Through the fault prediction and evaluation module in combination with the results of network dynamic evolution and quantum-assisted diagnosis, prediction of a plurality of future network states can be generated, different fault scenes are simulated for evaluation, potential risks are effectively predicted, and meanwhile, through risk evaluation and sorting, operation and maintenance personnel are helped to preferentially process high-risk scenes, and the operation and maintenance efficiency is improved.

The invention, which relates to the technical field of the cloud security system, discloses a monitoring method for a cloud security system comprising a data acquisition and preprocessing system, a threat detection and identification system, a behavior analysis and association analysis system, a risk assessment and early warning system, a response strategy making and execution system, and a monitoring report and log management system. The data acquisition and preprocessing system is used for acquiring network traffic, system logs and user behavior data in real time, and the threat detection and recognition system is used for performing threat detection and recognition on the preprocessed data. The data analysis and machine learning technology extracts significant features related to threats from a large amount of monitoring data, and the features can help a machine learning model better understand and classify potential threats, so that the accuracy and efficiency of threat detection are improved, the system automatically learns and updates threat features, and the threat detection efficiency is improved. And rapid identification and response to novel threats are realized.

The invention discloses a multi-agent cooperative control system based on multi-mode brain-computer interface-visual tracking, and the system comprises the following modules: a signal collection and processing module which employs an 8-lead electroencephalogram cap to collect electroencephalogram original data, selects occipital region parts (PO7, PO8, P3, P4, Pz, O1, O2 and Oz) as an electroencephalogram data collection region, and carries out the EEG data collection; the feature extraction and classification module is used for extracting SSVEP signal features by using a space-time convolutional network, optimizing classification precision in combination with an integrated learning strategy, extracting EOG signals based on a Transform model, integrating an LMS adaptive filter and denoising the EEG signals in combination with wavelet transform, expanding the frequency tracking range of the adaptive filter, and reducing signal loss to the greatest extent; the cooperative control strategy design module is used for mapping the electroencephalogram signals into control instructions of different intelligent agents, and cooperative control over the unmanned aerial vehicle and the trolley is achieved; and the target detection and tracking module is used for optimizing an unmanned aerial vehicle visual tracking system based on a PP-YOLOE model and realizing accurate tracking and automatic adjustment of the trolley.

The invention discloses a gas leakage point leakage detection system and method based on an e-dog, and belongs to the technical field of gas leakage detection. Through the autonomous navigation capability of the mobile platform module, the system can automatically avoid obstacles in a complex environment and inspect along a predetermined path, optimizes the inspection path in real time after finding a gas leakage point, improves the inspection efficiency, guarantees the high-density inspection of a key area, and the rapid movement and precise positioning capability of an e-dog, and improves the inspection efficiency. The gas leakage detection is quicker and more efficient, the risk that personnel enter a dangerous environment is reduced, the system has the functions of autonomous navigation, obstacle avoidance, remote control and the like, the intelligent level of detection is improved, the system is suitable for detection requirements of various complex environments and different types of gases, and through cooperative work of the gas detection module and the data acquisition and processing module, the detection efficiency is improved. The gas concentration data can be accurately acquired, the position of the leakage point can be judged, and the leakage point can be quickly and accurately positioned based on the combination of the GPS and the gas concentration data.

The invention belongs to the technical field of coal processing, and particularly discloses an intelligent coal gangue and coal quality sorting method and system based on an image hyperspectral technology, and the method comprises the steps: obtaining a hyperspectral image on a to-be-detected sample transmission line, converting the hyperspectral image into a grayscale image, and extracting the edge information of a to-be-detected sample; segmenting a spectral image of each to-be-detected sample, and extracting position information of each to-be-detected sample at the same time; determining the dynamic position of each sample to be detected; transmitting sound waves to each to-be-tested sample, and receiving sound wave signals passing through each to-be-tested sample; preprocessing the spectral image and the sound wave signal of each to-be-detected sample, and respectively extracting spectral features, texture features, spatial features and sound wave features to form a comprehensive feature vector; and inputting the comprehensive feature vector into a trained coal gangue recognition model to obtain a coal gangue and coal recognition result. According to the method, the sorting precision of the coal gangue and the coal quality is remarkably improved, so that the model can still keep relatively high recognition precision in a complex environment.

The invention discloses an underground diaphragm wall steel bar lap joint bending resistance calculation method based on big data, and belongs to the technical field of underground engineering design. In order to solve the problems that the condition change of a construction site is difficult to respond in real time, the construction parameters are difficult to dynamically adjust and the complex construction environment is difficult to adapt, the data set quality and the bending resistance prediction precision are effectively improved through multi-source data acquisition and key feature screening, and a physical prediction model of a steel bar lap joint is established. According to the method, factors such as stress transmission and binding power are comprehensively considered, so that the model has good adaptability under different working conditions, model parameters are dynamically adjusted, the model responds to changes of environments and construction errors in real time, the weight is automatically adjusted when monitoring data fluctuates, accurate prediction of the bending resistance is achieved, potential failure risks are effectively prevented, and the construction quality is improved. And a failure mode pre-judgment and multi-level risk management mechanism helps a construction team to quickly identify a high-risk area and flexibly adjust a construction scheme.

The invention relates to a terminal equipment intelligent fault diagnosis method based on electric power internet of things edge computing. The method comprises the following steps: constructing a cloud edge interconnection architecture; in a data acquisition stage, real-time denoising and standardization processing are carried out by utilizing an edge computing layer, and data availability and access speed are guaranteed by adopting an edge caching technology; data from different sensors are integrated by using a data fusion technology, the accuracy of fault detection and diagnosis is improved, multi-source data association analysis is applied, and multi-angle early warning is performed for abnormal behaviors; event mode identification and processing are realized on an edge calculation layer by using CEP, and potential fault signals are quickly responded; all transmission data of the communication layer uses an end-to-end encryption protocol to ensure data security, and edge device intrusion detection and firewall protection are implemented to prevent hostile attacks. According to the invention, the accuracy and efficiency of power equipment fault diagnosis are improved, and the safety and operation stability of the system are remarkably enhanced.

The invention discloses an outdoor fire hydrant water pressure real-time monitoring system, and relates to the technical field of fire fighting. When the system operates, a data acquisition and preprocessing module performs efficient fusion of multi-source data, and provides accurate water pressure data Px (t); the correlation analysis module evaluates the relationship between adjacent fire hydrants by using the geographic position data and the correlation coefficient, and can discover potential risks in time and generate an early warning identifier; the anomaly detection and evaluation module dynamically analyzes the water pressure data and the early warning identifier to provide an accurate fire hydrant anomaly risk evaluation strategy scheme; the path analysis module accurately judges an abnormal propagation path by analyzing the abnormal index gradient Gxy (t) of the adjacent fire hydrants, and reveals a potential regional problem; according to the method, the effective regional exception handling strategy is generated and executed immediately, the response speed and the decision-making ability are enhanced, and particularly, when regional or systematic problems are handled, early warning can be performed in advance, and optimal countermeasures can be taken.

The invention relates to the technical field of hotel management, in particular to a smart hotel guest room control system and method and a storage medium, and the method comprises the steps: obtaining the booking information of a guest room through a continuous check-in judgment module, and judging whether the guest room is continuously checked in; the travel and demand acquisition module acquires travel information and self-defined cleaning demand data of the guests after confirming continuous check-in, and determines a cleaning time period. And the cleaning time management module detects the state of the guest according to the cleaning time period and the cleaning interval time and judges whether the guest is cleaned or not. And the reminding and feedback module is used for sending information and receiving feedback when the guests do not go out. And when the guest goes out or is allowed to clean, the cleaning and quality inspection module collects a cleaned guest room image through the hotel robot, inputs the guest room image into the trained guest room image semantic extraction model, generates first semantic information, analyzes the guest room image semantic extraction model, compares the analyzed guest room image semantic extraction model with a guest room arrangement rule, and notifies a cleaner if the guest room image semantic extraction model does not pass the guest room arrangement rule. According to the invention, the satisfaction degree of the guests is improved through the control of the hotel guest room cleaning mode.

The invention provides a multi-modal data management system and method, equipment and a medium, and the system comprises a multi-modal data quality index creation module which is used for creating at least one multi-modal data quality index for multi-modal data; the multi-modal data quality detection task creation module is used for creating a multi-modal data quality detection task for the multi-modal data according to the at least one multi-modal data quality index and distributing the multi-modal data quality detection task to the multi-modal data detection module; the multi-modal data detection module is used for executing a multi-modal data quality detection task so as to carry out quality detection on the multi-modal data; and the multi-modal data detection result generation module is used for generating and displaying a multi-modal data quality detection report according to the execution result of the multi-modal data quality detection task, and the quality and management efficiency of the multi-modal data can be remarkably improved.

The invention discloses an Internet of Things service management system based on regional digital economy, and relates to the technical field of data processing. The Internet of Things service management system based on regional digital economy comprises a data minimization and anonymization module, an end-to-end encryption module, an access control and identity authentication module, a data security and auditing module and a security management and response module, and the data minimization and anonymization module is used for collecting personal sensitive information data; the end-to-end encryption module is used for encrypting a personal sensitive information data transmission process; the access control and identity authentication module is used for multi-factor authentication; the data security and audit module is used for protecting personal sensitive information data; and the security management and corresponding module is used for dynamically adjusting security policies and measures according to the privacy security state evaluation value. The problem that in the prior art, as the number of devices increases, more and more personal data are collected, and once sensitive data are integrated and analyzed, personal privacy may be exposed is solved.

The invention provides an excavator control method and system oriented to non-structural working conditions. According to the excavator control method and system oriented to the non-structural working condition, on one hand, according to the dynamic remodeling characteristic of earthwork in the excavation process, the excavation action track is planned through the earthwork state obtaining model and the excavation action planning model, the excavator control method and system can adapt to the non-structural working condition, and it is ensured that the operation effect is stable; on the second aspect, through real-time sensing and data feedback, the excavation action track can be adjusted in real time in the operation process, optimization is conducted according to the earthwork state, and the operation flexibility and effect can be improved; thirdly, by autonomously planning an excavation action track, the dependence on a manual operator can be reduced, the requirement of manual intervention is reduced, and the autonomous operation capability is improved; and fourthly, the mining action track is optimized through the mining action planning model, so that the energy consumption in the mining process can be reduced, the operation efficiency can be improved, and the operation cost can be reduced.

The invention discloses a network security and situation awareness equipment data monitoring and early warning system for a centralized control system, and relates to the technical field of equipment situation monitoring, and the monitoring system monitors the risks in the aspects of time synchronization, port activity, electronic element corrosion, part supply chain and memory resource management of equipment in real time. And an alarm is given out in time by rapidly flickering the warning lamps with different colors. And the strategy module automatically generates corresponding coping strategies according to different early warning alarms, such as triggering of a time synchronization mechanism, blocking of abnormal ports, arrangement of equipment maintenance, replacement of parts, adjustment of memory allocation and the like. The comprehensive management scheme not only can effectively prevent system faults, but also can ensure that actions can be taken quickly when abnormity occurs, so that potential safety hazards and operation interruption are reduced. Through automated risk management and response, the system facilitates the improvement of the reliability of device operation and the security of the network.

The invention discloses a benthonic animal community structure recognition method and system based on image recognition, and relates to the technical field of benthonic animal community structure recognition, and the method comprises the steps: carrying out the image collection of a target water area through an underwater camera, and transmitting the collected image to a computer. According to the method, the image recognition technology is adopted, rapid and accurate recognition and classification of the benthonic animals are achieved, the working efficiency is greatly improved, recognition is carried out by utilizing the image recognition technology and combining features such as shapes, textures and colors of the benthonic animals, the confusion rate and the error rate in the classification and counting process are effectively reduced, and the classification accuracy is improved. The accuracy and reliability of the result are improved, the abnormal change of the zoobenthos community structure can be found in time by utilizing a real-time data analysis technology, timely early warning and response are provided for protecting the ecological environment of the water body, targeted suggestions and measures are automatically provided, a user is helped to better improve and protect the ecological environment of the water body, and the user experience is improved. And the scientificity and effectiveness of decision making are improved.

The invention discloses an intelligent thermal work order system based on AI algorithm data processing, particularly relates to the field of place name mapping and conflict processing, is used for solving the problems of space conflict and precision in matching of dialect place names and standard place names, and aims at solving the problems of space conflict and precision in matching of dialect place names and standard place names by collecting and analyzing implicit association between place name description in historical work orders and actual work order coordinates. And by combining public GIS (Geographic Information System) data, community text information and a local database, automatically processing and updating geographical name mapping. When facing place names with dialect characteristics or outdated place names, effective matching can be carried out, and the place name mapping dictionary can be continuously updated and optimized online through machine learning and natural language processing technologies; in this way, the mapping mechanism can adapt to urban development, language transition and user habit changes in real time, it is ensured that the work order system can accurately and rapidly recognize and locate the place name, and then the scheduling efficiency and the accuracy of resource allocation are effectively improved.

The invention discloses a coal face working state detection method based on YOLOv8. The method comprises the following steps of 1, obtaining a data set of all working states of a coal face; step 2, a YOLOv8-EST network is constructed; 3, inputting the working state data set of the coal face into the improved YOLOv8 model for training to obtain a working state detection model of the coal face; and 4, inputting a to-be-identified coal face state picture into the coal face working state detection model to obtain a detection result of the working state of the to-be-identified coal face. According to the invention, the improved YOLOv8 target detection algorithm is constructed by adopting the YOLOv8 target detection algorithm, fusing the improved Swin Transform module and fusing the EMA attention mechanism, so that the real-time, efficient and accurate detection of the working state of the coal face is realized, and the production safety and efficiency are improved.

The invention provides a bit number silk screen adding method and device, equipment and a medium, relates to the technical field of PCB design, and is used for solving the problem that in the prior art, bit number silk screen adding efficiency and accuracy are low. The method comprises the steps that Allegro software is started, a design file of the printed circuit board is imported, and all devices of the printed circuit board are displayed; in response to a skill program calling instruction input by a user, displaying an option box; in response to a setting operation of a user in the option box, obtaining a placement parameter; in response to a first click operation of a user on a target device in all the devices through a mouse, attaching the bit number of the target device to the mouse in a silk-screen manner; and in response to a second click operation of the user at a target position through the mouse, placing the bit number silk screen of the target device at the target position according to the placement parameters. The user can add the bit number silk screen of the device only by clicking the mouse, so that the efficiency and the accuracy are improved.

The invention discloses a laser obstacle clearance remote control system for remote obstacle clearance of an overhead line, relates to the technical field of laser obstacle clearance, and aims to solve the problem that laser obstacle clearance equipment cannot accurately acquire obstacles. Through integration and optimization of the steps of image processing, dynamic adjustment, batch normalization and the like, the processing efficiency of the image data can be remarkably improved, the laser obstacle clearing equipment can respond to and recognize obstacles more quickly, obstacle features in the optimized image data and an obstacle model are trained and calculated by using a computer learning algorithm, and the obstacle recognition efficiency is improved. And the adaptive capability and robustness of the system can be greatly improved. By continuously learning new obstacle features, the system can gradually improve the recognition performance, calculate the obstacle features and the operation mode through a particle filtering method, and adjust the direction of the laser beam in real time according to the calculation result, thereby ensuring the accurate tracking of the obstacle position.

The invention relates to a woolen sweater defect detection method. The method is characterized by comprising the following four main steps: a data preprocessing step, a model construction step, a model training step and a defect detection step. Wherein in the model construction step, strategies such as an attention mechanism and a network structure optimization are introduced, so that the identification capability of the model on woolen sweater defects is improved; in the detection process, automatic positioning and classification of defect areas are realized through a series of key functions. The method is high in detection efficiency, low in comprehensive detection cost and high in detection accuracy.