6807 results about "Imaging Feature" patented technology

The invention discloses a multi-modal medical image data intelligent processing system, relates to the field of medical image analysis, and is applied to multi-modal medical image whole-process analysis of CT, MRI, PET, ultrasound and the like. According to the system, different modal image features are extracted and fused through a cross-modal manifold fusion network; a semantic guidance dynamic registration engine optimizes registration parameters to ensure that the registration error is less than or equal to 1.5 mm; the multi-task collaborative diagnosis network realizes multiple tasks such as disease classification; the clinical knowledge embedding and interpretable module generates a structured report and is in butt joint with an HIS system. Meanwhile, the model is optimized through a federated learning architecture, the adaptability of newly added data is improved by more than or equal to 20%, and intelligent processing and analysis of multi-modal medical images are realized.

The invention provides a magnetic core intelligent cutting parameter self-adaptive optimization system based on multi-mode perception, and relates to the technical field of data processing.The method comprises the steps that a multi-mode sensor module is integrated on magnetic core cutting equipment, and the module comprises a force sensor, a visual sensor and a temperature sensor; the acquisition units are respectively used for acquiring cutting force dynamic signals, cutting track image sequences and cutter temperature time sequence data in real time; magnetic core surface texture features and three-dimensional contour data are captured through a visual sensor, and an initial cutting parameter set is generated in combination with a magnetic core material type recognition result, associated parameters in a historical process database and preset process constraint conditions; and first workpiece trial cutting is executed based on the initial cutting parameter set, multi-modal data fusion collection is synchronously started, cutting force frequency domain feature vectors, a tool temperature change rate curve and cutting surface defect image features are obtained, and multi-modal data are obtained. According to the invention, multi-objective collaborative optimization of processing efficiency and energy consumption is realized.

The invention discloses a weldment welding seam automatic detection method and device based on machine vision, and relates to the technical field of machine vision intelligent detection. The weldment welding seam automatic detection method and device based on machine vision comprises the steps that S1, surface images and forming feature data of a weldment are collected and preprocessed to construct a standardized image feature data set; s2, the boundary clearness of the weld joint is evaluated by combining the edge strength and the contour coherence, and the main contour extraction range is dynamically adjusted; s3, analyzing abnormal focusing characteristics of the candidate area, and adjusting a defect labeling range and a detection priority; and S4, integrating the boundary definition and the abnormal focusing features, analyzing the structure abnormality, and dynamically controlling and verifying a resource allocation strategy. The problems that in the weldment detection process, obvious light reflection and texture blurring phenomena exist in a heat affected area at a weld joint, a traditional image enhancement and edge extraction algorithm is difficult to stably recognize microdefects, and the credibility of a detection result is reduced are solved.

The invention relates to the technical field of computer vision detection, in particular to an unmanned aerial vehicle target detection method based on frequency-space joint attention and dynamic fusion, and the method comprises the steps: obtaining an unmanned aerial vehicle image data set, carrying out the preprocessing, and dividing a training set and a test set; constructing a target detection model, inputting the training set into the target detection model to extract image features, sequentially performing frequency domain detail enhancement, spatial domain salient region extraction and multi-scale feature adaptive fusion based on the image features, and establishing a feature sequence; screening the feature sequence to obtain an initial target query, and finishing target classification and positioning on the initial target query through a decoder; training a target detection model by using the training set, and inputting the test set into the trained target detection model to generate a detection result; on the premise that the real-time reasoning advantage of RT-DETR is kept as much as possible, the problems that in an unmanned aerial vehicle scene, a target is prone to missing detection, the scale change is large, the background is complex, and the target is fuzzy are effectively solved, and the detection precision is improved.

The invention discloses a cut workpiece defect detection method and system based on image feature feedback, and relates to the technical field of image processing.The method comprises the steps that cut workpiece technological characteristics are obtained, a preset defect type library is constructed, a hardware system is built, and parameters are initialized; synchronously acquiring a multi-view original image, and storing and associating annotation information; de-noising the original image, enhancing the contrast, and extracting a region of interest ROI; extracting texture, shape, edge and gray features from the ROI, and screening through a Relief-F algorithm to obtain an optimal feature subset; inputting into an SVM (Support Vector Machine) model for reasoning, and screening to obtain an effective defect detection result; and calculating an evaluation index and generating a feedback signal, and performing iterative optimization after adjusting parameters. The system comprises an acquisition module, a master control module, a data processing module and a display module. Through the precise design and closed-loop feedback of the whole process, the precision, efficiency and long-term adaptability of defect detection of the complex cutting workpiece are improved, and the industrial quality management and control requirements are met.

The present invention relates to the technical field of deep learning and computer vision. Disclosed is an unmanned aerial vehicle image-based small object detection method for target areas. The present invention crops images of obvious small objects in certain target areas, and annotates the small objects of different categories to form a raw training and testing dataset, so as to ensure the accuracy of data required in the early stage of the algorithm and further ensure the scientificity of the algorithm; uses the computing capability of an improved YOLOv7 detection model to collect image features of different degrees in the dataset, the improved YOLOv7 detection model using YOLOv7 as a basic model and adding to a neck network an MS-CET module, which is constituted by an improved self-attention mechanism and convolution module SPPCSP, and a BHC-FB module, which is constituted by bidirectional mixed convolution modules NConv and RPConv connected in parallel; and finally fuses different feature layers as a final judgment basis of an unmanned aerial vehicle for small object detection in the target areas, to further check the accuracy of the algorithm and criteria for dataset selection, thereby improving recognition accuracy.

The invention discloses a wind turbine generator data analysis and fault diagnosis method and system based on big data and artificial intelligence. According to the method, a blade image, a vibration signal, audio data and operation parameters are synchronously acquired through an unmanned aerial vehicle multi-mode sensor and a ground monitoring system, and a multi-source heterogeneous data set is constructed; after the data is classified and preprocessed, image features, vibration time-frequency domain features and operation parameter key value pairs are extracted respectively; dimensionality reduction is carried out by using an auto-encoder, feature-level space-time alignment is realized through an improved DTW algorithm, and a multi-dimensional fault feature matrix is generated; a hierarchical diagnosis model including a GRU auto-encoder, an MLP network and an attention mechanism CNN is constructed, and training is carried out by taking minimization of sub-model deviation as an optimization target; and finally, fusing multi-source features to realize fault classification, and generating a visual diagnosis report. According to the method, efficient fusion and accurate diagnosis of multi-source heterogeneous data are realized, and the accuracy and the real-time performance of fault detection of the wind turbine generator are remarkably improved.

The invention discloses a medical image segmentation method and system based on guidance information and a multi-dimensional attention mechanism. The method comprises the following steps: collecting an original dermatoscope image for preprocessing; constructing a segmentation model, wherein the segmentation model comprises a double-path image encoder, a guide information encoder and a mask decoder; the two-way image encoder is used for extracting local detail features and global context semantic information in the image; the guide information encoder is used for converting a coarse segmentation mask predicted by the last round of network into guide feature information; the mask decoder fuses the image feature information and the guide feature information, gradually restores and refines the coarse-grained feature map, and finally outputs an accurate lesion segmentation mask; constructing a loss function, and training the segmentation model by using the preprocessed data; and inputting a to-be-segmented original dermatoscope image into the trained segmentation model, and outputting a lesion region segmentation mask of the image. According to the method, the segmentation precision and the model generalization ability can be improved, and the multi-scale lesion processing ability is enhanced.

The invention discloses a storage checking method based on a multi-modal sensing technology, a robot and a storage system, and belongs to the technical field of storage management and intelligent sensing fusion. Multi-modal data such as a visual image, space depth, radio frequency sensing and infrared temperature are collected, and an image feature vector, a three-dimensional point cloud model, a radio frequency response matrix and a temperature map are constructed; generating a fusion recognition vector through a multi-channel fusion network based on an attention mechanism, and dynamically adjusting a modal weight; constructing an article space distribution map, and marking a perception missing region; automatically complementing low-confidence region data based on a priority scheduling algorithm; performing joint verification on the original fusion result and the completion result to form a final inventory list; if the confidence coefficient of a certain article is lower than an early warning threshold continuously for multiple times, triggering an abnormal alarm and generating a traceable sensing sequence; the method is suitable for a high-precision inventory task in a complex storage scene, and has the advantages of high recognition robustness, intelligent completion mechanism, traceable abnormity and the like.

The invention discloses a cross-modal interactive image restoration method fusing text semantic guidance and visual structure priori, which comprises the following steps of: firstly, acquiring natural language description input by a user and an image to be restored, and generating a semantic segmentation map of the image through a semantic segmentation model; encoding the text and image semantics by using a pre-trained cross-modal encoding model to obtain text and semantic features; guiding a semantic alignment attention module through Prompt to realize deep fusion of multi-modal semantic features and image space features; structural enhancement and regulation of image features are realized by constructing a text guide weight graph, performing element-level modulation on the text guide weight graph and the optimized semantic segmentation graph, constructing a cross-modal structure semantic feature graph and generating a structural modulation factor; a four-stage image restoration network is adopted, and a high-quality restoration image conforming to semantic guidance and structure prior is generated step by step. According to the method, the semantic consistency, the structural integrity and the visual reality sense of an image restoration result are improved.

The invention discloses a deep forgery detection method based on a visual language model, and relates to the field of image forensics. The deep forgery detection method based on the visual language model aims to combine multi-source information to improve the discrimination capability of the model on a real image and a generated image. The method comprises the following steps: firstly, extracting image features through an image encoder of a pre-trained CLIP model; meanwhile, a frequency domain enhanced counterfeit perception adapter is embedded in the image encoder to mine potential anomalies of counterfeit images in the image domain and the frequency domain. Secondly, a manual feature extraction module is provided, discriminative low-dimensional features are extracted from the four aspects of the edge, the texture, the frequency and the symmetry of the image, and the discriminative low-dimensional features are used as auxiliary information input in the forgery detection process, so that the robustness and the interpretability of the model are improved; meanwhile, the text cue words are converted into feature vectors through a text encoder of a pre-training CLIP model; and finally, the model predicts a forgery score by calculating the cosine similarity between the image features and the text features so as to realize the discrimination of the authenticity of the image. According to the method, the problem that the detection capability of the model on the cross-dataset is insufficient is effectively improved.

The invention discloses a brain tumor multi-mode large model construction method, device and equipment and a storage medium, and is applied to the technical field of brain tumor imagines.The method comprises the steps that pixel-concept level alignment is conducted on a multi-mode MRI image and a pathological text; constructing a multi-modal feature fusion network for fusing image features and text features by adopting an attention mechanism of pathology perception and combining medical semantic information; training the multi-modal feature fusion network to generate an analysis report and a segmentation result; according to the technical scheme of multi-task cooperation, cross-modal pathological semantic accurate alignment, pathological knowledge graph injection and lightweight and continuous optimization parallelization, full-process coverage of brain tumor accurate segmentation, analysis report generation and prognosis prediction is achieved, the problems that a traditional model lacks pathological semantic support and is insufficient in clinical adaptability are solved, and the clinical adaptability of the traditional model is improved. And the deployment feasibility and the dynamic optimization capability are also considered.

The embodiment of the invention discloses an image annotation method and system applied to brain MRI image segmentation, and the method comprises the steps: obtaining a brain MRI image data set of a target object, and the brain MRI image data set comprises original image sequences of a plurality of scanning levels; performing multi-modal feature fusion processing on the original image sequence to generate an enhanced image feature set; calling a multi-layer cascade segmentation network to perform hierarchical feature extraction on the enhanced image feature set to obtain a multi-scale anatomical structure feature map; and performing region boundary optimization processing based on the multi-scale anatomical structure feature map, and generating a marked brain structure segmentation image. Therefore, the boundary of each structure of the brain can be accurately defined, the segmented image is more accurate and clearer, and the image segmentation and marking of the brain MRI image can be accurately and clearer realized.

The invention discloses a concrete structure apparent defect recognition method and device based on an inspection robot, and the method comprises the steps: collecting an apparent image of a target concrete structure in real time through the inspection robot in response to an apparent defect recognition instruction of the target concrete structure; obtaining a defect identification model based on YOLOv5; inputting the processed apparent image into a defect identification model, carrying out global image feature extraction and context image feature extraction on the apparent image through each Transform module in the backbone network, and carrying out merging processing on the global image features and the context image features through an SPPF module; performing key feature enhancement on the merged image features through a channel attention layer and a space attention layer of each CBAM module in the neck network, performing feature fusion on the enhanced merged image features to obtain fused image features corresponding to each CBAM module, and performing defect identification on each fused image feature through the head network to obtain a fused image feature corresponding to each CBAM module; and obtaining a defect identification result output by each prediction head module.

The invention discloses a small target recognition method and system for a multi-modal fusion image in a complex environment, and belongs to the technical field of computer vision and image recognition, and the method comprises the steps: obtaining a visible light image, an infrared image and environment sensor data; image registration is carried out on visible light and infrared images, and a multi-scale image feature pyramid is constructed. And respectively extracting visible light and infrared image features to obtain visible light and infrared imaging feature data. And performing multi-modal data fusion on the visible light and infrared imaging feature data based on a cross-modal attention mechanism, and adaptively adjusting a fusion weight based on environmental sensor data to generate fusion features. And performing space-time enhancement processing on the fusion feature to obtain an enhanced fusion feature. And performing target tracking detection on the small target, and outputting position and category information of the small target. According to the method, the small target recognition capability in a severe environment is remarkably improved, and high precision and robustness can still be kept in a foggy, low-visibility and dark scene.

The invention discloses an extreme sea condition parameter identification system based on deep learning, and relates to the technical field of ship navigation auxiliary equipment, in particular to a self-adaptive sea condition identification device which is used for acquiring image data and inertial measurement data of a current sea condition; the wave field visual depth estimation module is used for extracting visible light image features and infrared image features of a wave area from image data of the current sea condition, fusing the extracted visible light image features and infrared image features, using an encoder-decoder architecture and fusing an energy function to obtain a pixel-level wave height field, and outputting the pixel-level wave height field. The three-dimensional reconstruction of the wave surface is realized; the multi-modal data fusion module uses a filter dynamic model and a cost function to eliminate space-time asynchronous errors between inertial measurement data and visual perception data, performs multi-modal data fusion, and outputs wave field real-time parameterization information. According to the invention, the sea condition parameter real-time high-precision identification capability of the autonomous unmanned ship or the offshore carrying platform can be improved.

A method, apparatus, non-transitory computer readable medium, and system for image generation includes obtaining an input image and an input prompt, where the input image depicts an object and the input prompt describes a lighting condition for the object, generating relighted image features based on the input image and the input prompt, where the relighted image features represent the object with the lighting condition, and generating a synthetic image based on the relighted image features, where the synthetic image depicts the object with the lighting condition.

The invention provides an L1-L4 lumbar vertebra segment-oriented multi-feature fusion diagnosis system and method, and the system comprises an image preprocessing module which is used for receiving a lumbar vertebra CT image sequence of a patient; a centrum anatomy partition module; the multi-dimensional image feature extraction module is used for extracting four types of quantitative features from each sub-region; the clinical multi-modal data coding module is used for independently acquiring and processing three types of clinical data: a multi-modal graph attention fusion network; and the segment-level diagnosis output module outputs diagnosis results of three levels. Through a parallel processing architecture and an optimized feature extraction algorithm, the whole diagnosis process only needs 45 seconds from data input to report generation, time is saved compared with manual film reading, and the consistency of diagnosis results is remarkably improved.

The invention provides a semi-supervised target detection method for a visible light-infrared multi-mode fusion scene. The method comprises the following steps: constructing a semi-supervised visible light-infrared multi-modal image data set based on an LLVIP data set; on the basis of a YOLOv11 model architecture, constructing a target detection model oriented to multi-modal image feature fusion, and training the target detection model by using a semi-supervised visible light-infrared multi-modal image data set in a deep learning end-to-end mode to obtain a trained target detection model; and inputting a to-be-detected multi-modal image into the trained target detection model, and outputting a target detection result of the to-be-detected multi-modal image by the trained target detection model. The method is based on a semi-supervised learning normal form, so that the precision and robustness of target detection in a multi-modal fusion scene are improved, and the requirements for high efficiency and reliability of target recognition in practical application scenes such as intelligent traffic and intelligent security and protection are met.

The invention discloses a target detection method based on a YOLO model, electronic equipment and a storage medium, and relates to the technical field of target detection. Comprising the following steps: inputting an aerial image of an unmanned aerial vehicle into a trained target YOLO model; the target YOLO model comprises a backbone network, a neck network and a head network, and a feature extraction module in the backbone network performs multi-scale feature extraction by adopting a double-branch architecture attention mechanism; performing multi-scale feature extraction on the aerial image by adopting a dual-branch architecture attention mechanism through a feature extraction module in the backbone network, and constructing to obtain a plurality of layers of first comprehensive image features of the aerial image; performing feature fusion on the first comprehensive image features of different levels through a neck network to obtain second comprehensive image features of multiple levels; and inputting the multiple levels of second comprehensive image features into a head network to obtain a target detection result of the aerial image. According to the invention, the accuracy of small target detection can be improved.

The invention discloses a video multi-mode sentiment analysis method and device based on multi-layer perceptron fusion, and relates to the technical field of sentiment analysis, and the method comprises the steps: S1, extracting text features, image features and audio features in a video; extracting time sequence information in the image features and the audio features to obtain time sequence image features and time sequence audio features; s2, constructing a video multi-modal sentiment analysis model comprising a multi-modal feature capture module, a multi-layer perceptron fusion module and a sentiment classifier, and constructing a loss function according to modal similarity and modal heterogeneity between modals; s3, training the model; and S4, inputting the text features, the time sequence image features and the time sequence audio features into the trained model to obtain emotion polarity probability distribution. According to the method, similarity loss and heterogeneity loss are constructed, and sequence, channel and modal dimension fusion is carried out by using a multi-layer perceptron, so that the calculation complexity and memory consumption are reduced, and the integrity and discrimination capability of the multi-modal emotion features are improved.

The invention discloses a three-dimensional shielding target tracking method based on multi-modal space-time interaction, and relates to the technical field of target tracking. The method comprises the following steps: acquiring a point cloud and an image and preprocessing to obtain global fusion features; obtaining an initial detection frame and region-of-interest features through region proposal network processing; projecting the non-empty voxel point cloud to the image features, and reconstructing shielded target features; convolution and neural network processing are utilized to obtain a refined detection frame; screening legal detection frames through distance calculation and legality judgment; the bipartite graph and the self-adaptive channel graph are adopted for convolution, and appearance correlation scores are calculated; and matching the detection frame and the trajectory based on a Hungary algorithm to realize whole-course tracking. The target identification accuracy and robustness are improved, the shielding problem is solved, the accuracy of the detection frame is ensured, the correlation accuracy is improved by using the bipartite graph and the adaptive convolution, the nodes are matched in combination with the geometric cost matrix, and whole-course tracking and error calibration are realized.

The invention relates to the technical field of earthquakes, in particular to an optical remote sensing image feature landslide information extraction method. A double-time-phase NDVI time sequence analysis strategy is adopted, and landslide area recognition is achieved by constructing a vegetation index difference chart before and after an earthquake. A landslide mass area before an earthquake presents a high NDVI value due to complete vegetation coverage, and a vegetation index of the area in an image after a disaster is significantly attenuated due to surface disturbance, so that an obvious change response is formed. If the normalized vegetation index does not change, a non-landslide area can be determined, and the greater the change of the normalized vegetation index is, the greater the possibility of landslide occurrence is, and a landslide preselection area is determined; then Otsu threshold segmentation is carried out by combining cloud layer features and water body features, cloud layer and water body change parts are effectively eliminated, object-oriented geometric shape rule fine recognition is carried out on a preselected area, a DEM is adopted to calculate a slope value to constrain low-lying or flat earth surface changes, and efficient and accurate landslide information extraction is achieved.

Examples described herein relate to personalized image generation using combined image features. A plurality of input images is provided by a user of an interaction application. Each of the plurality of input images depicts at least part of a subject. Each input image is encoded to obtain an identity representation. The identity representations obtained from the plurality of input images are combined to obtain a combined identity representation associated with the subject. A personalized output image is generated via a generative machine learning model. The generative machine learning model processes the combined identity representation and at least one additional image generation control to generate the personalized output image. At a user device, the personalized output image is presented in a user interface of the interaction application.

The invention discloses a peripheral vein puncture blood vessel pressure monitoring system for image enhancement examination based on multi-modal fusion, and relates to the technical field of medical image assistion.The peripheral vein puncture blood vessel pressure monitoring system comprises the steps that image data, sensor data and personalized data of a patient are collected; generating an image feature vector, a sensor feature vector and a personalized feature vector based on the preprocessed multi-modal data, carrying out feature interaction on a multi-modal feature line, carrying out representation fusion by combining Transform with contrast learning, and generating a fusion representation vector; the maximum tolerance pressure, the extravasation risk level and the recommended injection flow rate range of the target puncture point under the current injection flow rate are predicted and evaluated in the blood vessel pressure resistance prediction model, injection suggestions and injection alarms of the target puncture point are generated, and injection parameters are automatically adjusted. According to the method, the personalized safety pressure threshold values of patients with different blood vessel conditions are accurately evaluated, and the safety and accuracy of peripheral vein puncture are remarkably improved.

The invention belongs to the related technical field of brain image processing, and provides an Alzheimer's disease classification method and system based on topology perception and a group hypergraph in order to solve the problem of inaccurate classification of the Alzheimer's disease in the prior art. Constructing a dynamic function connection network sequence through a sliding window strategy; a local topology perception encoder and a global topology perception encoder are respectively used for extracting local topology features and global topology features of each time window, deep interaction and fusion are carried out, and comprehensive feature representation of a tested level is generated; according to the method, each subject is used as a hypergraph node, hyperedges are constructed on the basis of comprehensive feature representation of a subject level and by combining feature similarity calculated by diffusion tensor imaging features and clinical embedded features of the subject, then a group hypergraph is constructed, a classification result is obtained by using a hypergraph neural network, and the early classification diagnosis accuracy of the Alzheimer's disease is effectively improved.

The invention discloses a detection method for printing effect verification, and the method comprises the steps: collecting scanning image information and environment sensing information, employing an improved OTSU algorithm and an illumination compensation module to process the scanning image information, and extracting a standardized image feature matrix containing the characteristics of character integrity, edge sharpness and the like; identifying a defect mode by using an EAST text detection algorithm and morphological operation, and generating a quantitative evaluation vector; the vector and environment sensing information are subjected to space-time alignment, a printing quality degradation model is constructed through a density peak value clustering algorithm, and a multi-dimensional quality index set is established; and finally, based on the graph attention network and the time sequence convolutional network, analyzing the incidence relation between the printing quality and the equipment state, and outputting a verification report containing a quality score, a defect positioning graph and a life prediction curve. According to the method, accurate quantitative evaluation of the printing quality is realized, a dynamic association model of the quality, the environmental parameters and the equipment aging is established, and an intelligent decision basis is provided for printing quality maintenance.

The invention provides an artificial influence weather-based image recognition method and system, and the method comprises the steps: firstly obtaining a meteorological image data set of a target region containing a plurality of original meteorological images in sequential arrangement, carrying out the feature extraction of the original meteorological images, and obtaining a meteorological element distribution feature and a dynamic evolution feature; and then, based on a preset weather influence analysis model, carrying out weather influence analysis on the meteorological element distribution characteristics and the dynamic evolution characteristics, generating weather intervention decision characteristics, finally, according to the weather intervention decision characteristics, generating an operation strategy of artificial influence weather, and sending the operation strategy to operation execution equipment to trigger weather intervention operation. Therefore, meteorological image features can be comprehensively and deeply analyzed, the operation strategy is scientifically and accurately generated, and the effect and efficiency of artificial influence weather operation are improved.

The invention discloses an ice condition prediction method based on multi-feature fusion and physical constraint, and belongs to the technical field of hydrological forecasting. Firstly, various types of data of a target area are collected and preprocessed to serve as a data set, features of the various types of data are extracted, feature fusion is conducted on obtained image features, time sequence features and environment features through a multi-head attention mechanism, and a fusion feature vector is generated. Secondly, adopting a physical information neural network model, taking the fusion feature vector as input, taking ice thickness and ice stress as output layers, carrying out constraint by using a composite loss function, carrying out model optimization by using a verification set, carrying out processing through a full connection layer in the network, and carrying out end-to-end training and regularization of a prediction model; and finally, evaluating the final prediction model obtained by training through the test set, and verifying the effectiveness and generalization ability of the final prediction model. The method combines multi-source and multi-mode observation data, can effectively capture complex relations between ice surface changes and various factors, can improve prediction precision, stability and reliability, and can improve model training efficiency and generalization ability.

The invention provides a multi-modal information fused steel pipe inner surface defect area segmentation method and system, and the method comprises the steps: obtaining a steel pipe inner surface defect RGB image and a depth map at the same time, and carrying out the preprocessing and marking; constructing a support data set and a query data set; the support and query image feature extractor is used for respectively extracting support image multi-scale aggregation features FS, support image multi-mode semantic features IS, query image multi-scale aggregation features FQ and query image multi-mode semantic features IQ by sharing the multi-mode feature extraction backbone network; the FS, the IS, the FQ and the IQ are input into a multi-feature fusion device, a graph semantic guide module is supported to generate class guide features FA by using the FS and the MS, a similar prior feature generation module is supported to generate similar prior features FM by using the IS, the IQ and the MS, the FA, the FM and the FQ are spliced on a channel dimension, and a fusion feature graph is output and decoded by a multi-mode decoder to obtain defect area segmentation output. The method can be used for segmenting the defect area on the inner surface of the steel pipe.