381 results about "Boundary detection" patented technology

The invention belongs to the technical field of artificial intelligence, and relates to an artificial intelligence-based electric energy quality disturbance identification and positioning method, which comprises the steps of constructing an electric energy quality disturbance signal data set, performing segmented preprocessing on electric energy quality disturbance voltage data, enhancing time-frequency joint features and encoding disturbance sensitive areas. And constructing a deep learning model for power quality disturbance identification and positioning, and identifying and positioning the power quality disturbance. According to the invention, through adaptive denoising processing, boundary detection and multi-resolution time-frequency feature extraction, the identification precision and positioning precision of power quality disturbance are significantly improved; self-adaptive wavelet denoising and dynamic segmentation are combined, noise interference is effectively suppressed, and the edge characteristics of voltage sudden change points are kept; according to the dual-task sharing network, disturbance identification and positioning tasks are cooperatively optimized, so that the network can consider disturbance classification and time positioning at the same time; and through Bayesian reasoning, the system can output confidence estimation, provides credibility quantification of identification and positioning results, and effectively improves the reliability of the system.

The invention belongs to the technical field of document content extraction, and particularly discloses a document content extraction method and system based on multimodal model collaboration, a terminal and a medium. Comprising the following steps: identifying the type of an input to-be-processed document, and judging the document type; on the basis of the type identification result, calling a multi-modal model to analyze the document content, and outputting space coordinates, visual features and semantic features of document elements; generating a content sequence according with a reading habit through a semantic sequence reconstruction algorithm; paragraph boundary detection, paragraph recombination and semantic association modeling of charts and texts are completed based on the multilayer attention network and the graph neural network; grammar error correction, format optimization and title hierarchy generation are carried out by using a large language model and a hierarchical classification network; and converting the identification result into a structured output file. According to the method, the processing requirements of different types of documents can be considered, and high-precision analysis and efficient output are realized under the scenes of complex layouts, multiple languages and formula tables.

The invention relates to the technical field of industrial production intelligent process control, in particular to an AI-based electrolytic manganese process production quality intelligent prediction method and system, and the system comprises a drift calibration module, a structure matching module, a process parameter inversion module, a path correction module and a deviation verification module. According to the method, dynamic compensation of deviation between input parameter drift and output parameters is achieved through the response relation between input and output of all links, the actual influence degree of fluctuation of all the parameters on the target control parameters and product quality is effectively recognized according to correlation matching of the periodic change of the production process and the fluctuation frequency of the target control parameters, and the accuracy of the product quality is improved. Potential interference of key process parameters on stability of each process is locked in time, diffusion starting time and space of impurities are clearly traced through path judgment of residual liquid concentration trend and raw material adding position, and migration path parameters are further adjusted based on ion retention behaviors of boundary detection points.

The invention relates to a distributed trajectory planning method and system for an unmanned aerial vehicle cluster in an obstacle environment. The method comprises the steps of initializing a system, constructing a local Euclidean symbol distance site map, and realizing real-time interaction of state information of the unmanned aerial vehicle. An initial collision-free path is generated using jump point search. And solving an optimal control point through an L-BFGS algorithm through multi-constraint trajectory optimization in combination with a load swing dynamics model, four-rotor dynamics limitation, cluster collision avoidance and environment obstacle avoidance requirements. And a dynamic time redistribution strategy is adopted, and the track time interval is adjusted according to speed and acceleration overrun conditions. The method further comprises an adaptive re-planning mechanism, and local target points are updated in real time and adjacent aircraft collaborative optimization is triggered based on local map boundary detection and quadrotor track safety detection. According to the method, efficient, safe and stable trajectory planning of the hanging load quad-rotor unmanned aerial vehicle cluster in a complex environment is realized, the requirement of autonomously and efficiently completing tasks is met, and the task execution efficiency and safety are improved.

The invention belongs to the technical field of remote sensing image intelligent processing and agricultural information extraction, and particularly relates to a remote sensing large model pre-training and multi-granularity boundary supervision natural field type extraction method, which comprises the following steps: firstly, pre-training a model on a large-scale space-time spectrum remote sensing data set, and combining anchor point sensing mask and geographic information coding; secondly, inputting the multi-scale features into a multi-branch structure sensing network, and outputting a semantic segmentation prediction map through high and low resolution double input and dynamic attention fusion; generating a multi-scale field boundary label through morphological operation, and outputting a boundary prediction map through multi-task supervision after domain enhancement of features by a frequency space double-domain enhancement module; and finally, aligning the two images and performing pixel-level operation to obtain a high-precision extraction result. Through the method, the global classification error during cross-region migration is greatly reduced, the method is adaptive to a low-pixel wide ridge, the boundary detection value and the closure rate of a small-scale field are improved, and the extraction precision of a large field and a small field is considered.

The invention discloses an image edge cutting method and device based on boundary detection, equipment and a storage medium, and relates to the technical field of image boundary detection, and the method comprises the steps: carrying out the preprocessing of an original image, and obtaining a target image; determining a scene type of the target image; based on an edge detection strategy corresponding to the scene type, performing edge detection on the target image to obtain an edge image; performing straight line detection on the edge image to obtain a plurality of candidate straight line segments; and based on a boundary positioning rule, determining a target boundary line segment from the plurality of candidate straight line segments, so as to carry out edge cutting on the original image according to the target boundary line segment. According to the invention, a more reliable image edge cutting effect can be generated.

The invention discloses a sign language translation model, a sign language translation system and a sign language translation method based on full-modal alignment. The sign language translation method comprises the steps of extracting multi-modal features of hand, face and body postures from an input video and performing preliminary fusion; deep secondary fusion and alignment are carried out through multi-scale time sequence coding and a cross-modal collaborative attention mechanism, and a spatial-temporal feature sequence of full-modal alignment is generated; performing boundary detection and dynamic segmentation on the feature sequence by using a CTC-based sequence prediction model, and outputting a discrete sign language word sequence with a timestamp; and finally, capturing a sign language grammar structure of the sequence through a graph structure enhanced Transform encoder, inputting the sign language grammar structure into a Transform decoder integrating grammar consistency loss, and generating a target text conforming to target natural language grammar and semantic rules. According to the method, the problems of continuous sign language action adhesion and grammar structure difference are effectively solved, and the sign language translation accuracy and the natural language generation fluency are greatly improved.

The invention discloses a pulmonary nodule boundary detection method based on semantic segmentation. The method comprises the following steps: generating an edge prior image set; obtaining a multi-scale boundary alignment feature map set and a deformation offset parameter set; a post-gating deformation offset parameter set is obtained, the multi-scale boundary alignment feature image set is updated, cross-scale feature fusion is executed, and a fusion feature image set is generated; establishing a node set and an edge set by utilizing node features in the fusion feature graph set and a spatial adjacency relationship, and initializing a node feature vector set; forming a fusion feature map set enhanced by multi-hop reasoning; and outputting a pulmonary nodule region mask prediction image set and a pulmonary nodule boundary probability prediction image set. According to the method, fracture between pseudo boundaries and slices is effectively inhibited, so that a three-dimensional structure prediction result is more coherent, and high-readability and high-consistency boundary segmentation output is provided for a clinical CAD (Computer Aided Design) system.

The invention relates to the technical field of voice signal processing, and discloses a voice segmentation intelligent editing system based on deep learning. The system comprises a voice feature extraction module, a segmentation boundary detection module, a semantic content analysis module, an editing strategy generation module and a real-time quality evaluation module. The voice feature extraction module collects multi-dimensional voice features and timestamp information, and verifies feature integrity and timeliness; the segmentation boundary detection module identifies voice pause intervals and semantic turning nodes and divides segmentation units and boundary types; a semantic content analysis module extracts text content and emotion features of each segment, and analyzes semantic topic relevance and information density; an editing strategy generation module formulates a segmentation retention rule and a sequence adjustment scheme, and matches user preferences and scene demands; the real-time quality evaluation module monitors voice fluency and information integrity in the editing process and analyzes splicing errors and user feedback. According to the system, intelligent processing of the whole voice editing process is realized.

The invention provides an anti-collision method, device and equipment based on obstacle classification in a medical environment, a medium and a program product, and relates to the field of intelligent anti-collision. The anti-collision method comprises the following steps: acquiring point cloud data of a target body in a medical environment; carrying out clustering reconstruction processing on the point cloud data of the first dynamic obstacle in the target body to obtain a reconstructed cluster, and carrying out space division on the cluster to generate a high-dynamic convex boundary; performing recursive space division on the point cloud data of the second dynamic obstacle and the static obstacle in the target body to generate a low dynamic convex boundary, and performing recursive space division on the point cloud data of the to-be-controlled anti-collision body to generate a target convex boundary; and expanding the boundary of the target convex boundary based on the safety distance, detecting the position relationship between the expanded boundary and the low-dynamic convex boundary and the position relationship between the expanded boundary and the high-dynamic convex boundary, generating a control instruction according to the position relationship, and controlling the movement of the to-be-controlled anti-collision body.

The invention discloses a remote sensing image semantic segmentation method and equipment based on an enhanced MambaVision architecture. The method comprises the following steps: performing initial feature extraction on a to-be-segmented high-resolution remote sensing image through a patch embedding module; the initial features are input into a MambaVision backbone network to extract multi-scale features, the backbone network comprises four layering stages, the stage 1 and the stage 2 adopt adaptive expansion convolution blocks to balance high and low frequency features, and the stage 3 and the stage 4 adopt two-dimensional selective scanning and a multi-head attention mechanism to capture global dependence; performing fusion and up-sampling on the multi-scale features through an MFMS decoder, and optimizing texture and boundary detection; the decoder comprises an MFMS attention module and a channel mixer; and performing multi-scale convolution fusion on the output of the MFMS attention module based on the channel mixer to generate a segmentation result. According to the method, through two-dimensional selective scanning, adaptive expansion convolution and a multi-frequency multi-scale MFMS decoder, the segmentation precision and efficiency are improved, and the method is suitable for high-resolution and large-size remote sensing images.

The invention belongs to the technical field of three-dimensional point cloud processing, and particularly discloses an automatic tooth segmentation method and system for oral cavity scanning point clouds, and the method comprises the following steps: collecting and preprocessing dentition three-dimensional point clouds, extracting point cloud features containing coordinates and normal directions, and constructing a neighborhood structure; setting a first-stage multi-branch network, and performing tooth and gingiva coarse segmentation and instance initial clustering on the point cloud; setting a second-stage semantic refinement network, and carrying out local cutting on the original point cloud; setting a boundary detection and boundary offset network to enhance a tooth contact area, and refining a boundary instance by clustering to obtain a boundary enhanced instance tag; and performing label fusion, and propagating the fused label to the original high-resolution point cloud through a nearest neighbor algorithm to obtain a final semantic label and a final instance label. By adopting the technical scheme, high-precision instance-level segmentation of the complete dentition point cloud is realized through semantic segmentation, boundary detection and clustering cooperative work based on a biased field.

The invention discloses an ultrasonic image denoising method based on variational mode decomposition and local space sparse fusion, which comprises the following steps of: firstly, adaptively optimizing key parameters of variational mode decomposition by using a grey wolf optimization algorithm to realize stable and efficient decomposition of an ultrasonic image; then, classifying the modal components according to the structural features of the modal components, and implementing differentiated denoising strategies for different types of modals to separate noise and reserve useful information; after modal reconstruction, a sparse expression method based on local space information is further introduced, according to the method, accurate boundary detection is carried out through gradient vector flow and gray scale proportion analysis, self-adaptive partitioning is carried out on an image according to boundary information, and finally sparse reconstruction is carried out through double dictionaries trained for different areas. According to the method, speckle noise in the ultrasonic image can be effectively suppressed, and meanwhile, the capability of keeping the edge and detail information of a tissue structure is remarkably improved, so that the ultrasonic image with higher quality is obtained.

The invention relates to the technical field of pupil boundary detection, in particular to a multi-algorithm fusion pupil boundary detection method and system, and the method comprises the following steps: obtaining an eye image through an infrared camera, converting the eye image into a grey-scale map, optimizing the grey-scale map, carrying out the median filtering noise reduction, maintaining the edge information, and determining the pupil boundary through a Sobel operator and Hough transform. And the pupil center displacement is analyzed, the motion consistency is evaluated, the pupil position is corrected, and the pupil positioning accuracy is optimized. According to the method, the image preprocessing process is optimized by applying infrared camera shooting and image processing algorithms, accurate detection, median filtering and local contrast enhancement of pupil boundaries are achieved through multi-algorithm fusion, and particularly noise and edge information in the images are optimized; according to the method, the edge detection precision and the image definition are effectively improved, the pupil boundary is accurately mapped through combined use of the Sobel operator and Hough transform, and the misjudgment rate is greatly reduced.

The invention discloses a safety fence protection method and device based on the Internet of Things technology, and belongs to the technical field of safety alarm and Internet of Things application, and the method comprises the following steps: respectively collecting the distance, heat source distribution and movement track information of a target protection area through an ultrasonic sensor, an infrared sensor and a millimeter wave radar; generating a virtual fence model with an initial alarm boundary by adopting a confidence weighted fusion algorithm based on environmental condition dynamic adjustment; environmental change parameters such as obstacle distribution and meteorological conditions are monitored in real time, when the change exceeds a preset threshold value, a dynamic boundary adjustment strategy is triggered, and the fence alarm boundary shape is optimized through a boundary node position redistribution mechanism; a distributed computing framework is adopted to decompose boundary detection tasks, and the tasks are dynamically allocated according to the real-time resource utilization rate of edge computing nodes. According to the invention, the problems of insufficient real-time performance, limited dynamic adjustment capability and low system integration degree in a complex environment in the prior art can be effectively solved.

The invention provides a road boundary detection method and a false target identification method and system, and the method comprises the steps: detecting the operation information of a vehicle in real time, and updating the historically accumulated point cloud information based on the motion information of the vehicle; performing screening processing on the point cloud information based on a preset screening rule to obtain final point cloud information; performing rasterized indexing on the final point cloud information based on a preset rasterized indexing rule to obtain a preliminary road boundary point; and based on the self-vehicle motion information and the final point cloud information, carrying out smoothing processing on the initial road boundary point to obtain a final road boundary. According to the method provided by the invention, the technical problem that the road edge identification technology cannot accurately identify the road boundary in special-shaped road boundaries or irregular road scenes, such as forks, roundabout, small roads in villages and the like, is effectively solved.

The invention discloses a hybrid PDF (Portable Document Format) document analysis and knowledge fragment construction method and device, and belongs to the technical field of digital document intelligent processing. The method comprises the following steps: firstly, carrying out layout element identification and semantic classification on a PDF page through a deep learning model, and calculating the confidence of the PDF page; when the confidence coefficient is lower than or equal to a threshold value, a correction module based on vertical and horizontal projection analysis is automatically switched to carry out boundary detection and result error correction; the fused layout result is converted into structured JSON data, and the structured JSON data is further mapped into a Markdown format with a reserved title level; and finally, knowledge fragment segmentation is carried out based on a Markdown hierarchical relationship, so that each fragment carries a complete title context path. According to the method, the problems of semantic deficiency, insufficient stability and context splitting in the prior art are effectively solved, and the robustness and accuracy of complex format PDF and low-quality scanning copy analysis and the interpretability of downstream knowledge retrieval are remarkably improved.

The invention discloses an industrial product surface defect detection system and method based on a segmentation cutting model, and belongs to the technical field of detection. According to the method, all segmentation models are successfully applied to an industrial defect detection task through a double-adapter strategy, a scale perception feature attention sensor is designed, and the sensitivity of defect detection is enhanced by fusing a Transform-based multi-scale feature and a mixed attention mechanism; meanwhile, an efficient decoder comprising an adaptive feature enhancement module is constructed, the boundary detection capability is optimized through local convolution and cross-layer compensation, the defect detection method is remarkably superior to a traditional defect detection method, and the robustness of the model to small defects and complex backgrounds is improved. The invention provides an efficient and accurate solution for industrial defect detection, and has important engineering application value.

The invention provides a multi-task learning optimized corpus annotation automation system, and aims to solve the problems of low efficiency and poor accuracy of a traditional corpus annotation method. The system comprises a shared encoder, a multi-task decoder, a dynamic adaptive module and an online incremental learning module. The shared encoder adopts a Transform architecture to carry out word segmentation, sliding window segmentation and time embedding processing on an input corpus to generate global semantic representation; the multi-task decoder completes tasks such as event boundary detection, time sequence reconstruction and causal relationship extraction in parallel; the dynamic adaptive module adjusts the loss weight in real time according to the consistency score output by each task, and enhances the adaptive ability of the model; and the online incremental learning module realizes continuous optimization of the model through expert auditing and model parameter updating. According to the system, through multi-task learning and self-adaptive optimization, the efficiency and accuracy of corpus tagging are remarkably improved, and the system is suitable for multiple fields such as news reports, social media analysis and legal document processing.

The invention discloses an RGB-D salient target detection method based on decoupling contrast learning, and designs a saliency detection framework integrating expression enhancement, modal collaborative perception and structural discrimination learning by combining a structural heterogeneity problem in multi-modal modeling and utilizing the frequency domain structural advantage of a deep mode and the long-distance modeling capability of Transform. By introducing wavelet convolution and Transform joint modeling, a cross-modal interaction parallel fusion mechanism and a pixel-level structure perception contrast learning strategy, high-precision, multi-scale and boundary clear detection of a salient target area in a complex scene is realized. The method can effectively solve the problems of large information difference between modes of the RGB and the depth map, difficulty in structure alignment, fuzzy boundary prediction, weak feature expression ability and the like, significantly improves semantic consistency and structural integrity of the salient region, and has good cross-modal generalization ability and robustness.

The invention discloses an electric ship remote intelligent control method and system based on cloud-side cooperation, and relates to the technical field of cloud-side cooperation computing, and the method comprises the steps: carrying out the feature extraction of images around a ship and time-space aligned multi-modal data, obtaining visual features, motion features and environment features, carrying out the feature fusion, and generating an environment situation vector; inputting the environment situation vector into a channel boundary detection model, analyzing the emergency degree of the current situation, generating a collision avoidance instruction, and generating a ship control decision through a three-level collaborative decision; combining a ship control decision with ship control characteristics, simulating a traffic development situation in a navigation area, comparing with historical ship navigation conflict data, and identifying potential conflict data; through collaborative fusion of the ship-end decision-making unit, the shore-end decision-making unit and the cloud-end decision-making unit, an optimal control strategy is generated in combination with respective advantages, intelligent decision-making of global optimization in a complex environment is realized, and adaptability and safety of ship control are enhanced.

The invention discloses a driving environment sensing method, device and equipment for an underground roadway and a medium. The method comprises the following steps: acquiring environmental point cloud data in the driving process of a vehicle in an underground roadway; based on the environment point cloud data, multi-dimensional environment perception detection is carried out on the underground roadway to obtain environment perception detection data, and the multi-dimensional environment perception detection comprises at least one of ground slope detection, obstacle detection, roadway boundary detection and roadway intersection detection; correspondingly, the environment perception detection data comprises at least one of ground slope data, obstacle information, roadway boundary information and roadway intersection information. Through the technical scheme of the invention, the problem that the underground view of the underground vehicle is limited in illumination restriction and narrow roads is solved, the effect of multi-dimensional environment perception is realized, the occurrence of safety accidents such as collision and scratch of the underground mine car is effectively avoided, and the safety of underground personnel and property of a coal mine is further ensured.

The invention relates to the technical field of trajectory control, in particular to a multi-modal motion trajectory real-time tracking method and system for machine control, which performs weight redistribution through channel noise variance and amplitude difference, performs weighted stacking and continuous updating in combination with Kalman filtering, reduces state estimation variance and channel bias transfer, and improves the tracking accuracy of a multi-modal motion trajectory. Time consistency and amplitude consistency of fusion track state data are improved, an overshoot peak value is inhibited, a deviation attenuation process is shortened, abnormal samples in a deviation sequence are rejected through an isolated forest, interference of outliers on parameter updating is reduced, and speed difference and acceleration difference are compared for a correction control sequence, so that the accuracy of correction control is improved. The boundary detection rate and the determinacy of the stage number are improved, the energy change rate is screened through the speed difference and the curvature value of the adjacent stages, the transfer relation is determined, the time sequence coherence and the geometric continuity of stage switching are kept, and the time sequence stability and the stage connection coherence of the control instruction are kept in the multi-degree-of-freedom space.

Embodiments of the present disclosure provide a cleaning method, a window wiping robot, and a storage medium. According to the embodiments of the present disclosure, when a window wiping task needs to be performed, the window wiping robot firstly spins to detect a boundary of a glass window to be cleaned. If the boundary of the glass window to be cleaned is detected during spinning, the window wiping robot performs the window wiping task directly along the boundary. The cleaning mode may save the boundary detection time and help to improve the cleaning efficiency when starting to perform the window wiping task without detecting the boundary up, down, left, and right.

The invention relates to the technical field of optical fiber communication device detection, and provides an optical fiber connector end face 3D shape detection method and system, and the method comprises the steps: obtaining a multi-step phase shift interference image sequence of an optical fiber connector end face; performing interface segmentation on the multi-step phase shift interference image sequence by adopting a fiber core boundary detection network based on radial gradient perception to obtain interface position data; based on the interface position data and the multi-step phase shift interference image sequence, performing phase unpacking by adopting a dual-coordinate system fusion phase unpacking algorithm to obtain initial phase data; performing global reconstruction on the initial phase data by adopting a regional confidence coefficient weighted reconstruction algorithm to obtain global phase field data; performing spherical parameter fitting on the global phase field data by adopting an IEC (International Electrotechnical Commission) standard constraint hierarchical particle swarm optimization algorithm to obtain spherical geometric parameters; and generating a 3D shape detection result of the end face of the optical fiber connector according to the spherical geometric parameters. According to the invention, the precision and reliability of 3D shape detection of the end face of the optical fiber connector are improved.

The invention discloses a network security entity identification method based on fine-grained denoising diffusion boundary detection, and relates to the field of network security, in particular to a network security entity identification method. The objective of the invention is to solve the problems that effective generalization is difficult to realize, network security entity features are difficult to fully extract and the boundary identification error rate is high due to insufficient targeted research of the existing network security entity identification method. The method comprises the following steps of: 1, acquiring marked network security entity data; 2, constructing an entity recognition model based on fine-grained denoising diffusion boundary detection, and obtaining a trained entity recognition model based on fine-grained denoising diffusion boundary detection; and 3, obtaining to-be-detected network security entity data, inputting the to-be-detected network security entity data into the trained entity recognition model based on the fine-grained denoising diffusion boundary detection, and outputting the category of the entity by the trained entity recognition model based on the fine-grained denoising diffusion boundary detection.

A harvesting machine is configured to gather harvested material into the harvesting machine during a harvesting operation. A conveyance subsystem configured to convey the harvested material from the harvesting machine to a receiving vehicle during the harvesting operation. An image capture system generates an image of the receiving vehicle and an image processor uses a machine learning system to recognize a boundary of the receiving vehicle. A control system determines a position of the receiving vehicle boundary relative to the harvesting machine, and generates a control signal based on the identified receiving vehicle boundary.

The invention provides a smart home event segmentation and anomaly detection method, and the method comprises the steps: obtaining a plurality of pieces of time sequence data detected by a plurality of smart home sensors, and carrying out the preprocessing and coding processing of the time sequence data, and obtaining a plurality of original feature vectors; each original feature vector corresponds to one event and comprises an equipment code, a state feature and a time feature of the corresponding sensor; performing divide-and-conquer processing on features of different dimensions in the original feature vector, and then performing splicing and linear projection to obtain an initial hidden state matrix; inputting the initial hidden state matrix into a Transform encoder comprising a multi-layer encoder and a classification head; each layer of encoder adopts a multi-head self-attention mechanism to carry out processing, residual connection and layer normalization on the hidden state matrix output by the previous layer, and the final hidden state of each event is obtained; and performing boundary detection or anomaly detection based on the final hidden state of each event. According to the invention, the reliability and accuracy of smart home event segmentation and anomaly detection can be improved.

A computer-implemented method can determine a linguistic boundary condition for synthetic data generation in a multi-class classification problem. The method includes analyzing empirical labelled data using linguistic and vector representation techniques. The method further includes deriving a synthetic boundary conditional (SBC) model based on the analysis of the empirical labelled data and identifying a boundary location for performant synthetic data generation using the SBC model.

The invention relates to a multi-cavity injection mold insert defect identification method, which comprises the following steps: acquiring a high-resolution image through an industrial camera, and carrying out graying, de-noising, histogram equalization and other preprocessing to improve the image quality. Then, an edge thermodynamic diagram is generated in combination with Canny edge detection and a deep learning model, a multi-channel segmentation network is guided to extract multi-scale features and dynamically enhance edge response, and continuity and accuracy of defect boundary detection are improved by means of an adaptive feature fusion module. Furthermore, the defect area is repaired through dynamic threshold segmentation and morphological operation, and finally visual labeling and automatic output of the high-precision defect area are achieved. According to the method, parameters can be adaptively optimized to deal with image blurring and other situations, a perfect model parameter adaptation mechanism is provided, and the defect recognition robustness and engineering applicability under complex working conditions are remarkably improved.