8753 results about "Semantic information" patented technology

The invention belongs to the technical field of multi-modal data processing, and particularly relates to a multi-modal document retrieval enhancement generation method based on a large model, which comprises the following steps: receiving query content input by a user for a multi-modal document; processing the query content by adopting an embedded model, generating a query vector representing user query semantic information, and storing the query vector in a vector database; analyzing the multi-modal document to obtain long text information, segmenting the long text information into data blocks by adopting a recursive partitioning strategy, and numbering and marking the data blocks; carrying out vectorization processing on the data blocks by adopting an embedded model to generate document vectors, storing the document vectors into a vector database, and constructing a hierarchical index structure; retrieving in a vector database based on the query vector, and returning a retrieval result; and processing a retrieval result by utilizing a large language model to generate response content conforming to the query intention of the user. According to the method, the multi-modal document can be effectively analyzed and processed, and the accuracy and comprehensiveness of analysis are improved.

The invention discloses an intelligent operation and maintenance management and alarm system based on a large model agent, and the system comprises a data collection and fusion module which is used for processing original data collected from an IT system to obtain an operation and maintenance data graph, and the original data at least comprise logs, indexes and link tracking data; the anomaly detection module is used for performing analysis based on the preprocessed data and the time sequence characteristics to determine abnormal data; the alarm optimization module is used for analyzing the semantic information corresponding to the abnormal data so as to determine the priority of the corresponding alarm; the root cause analysis module is used for performing fault reasoning on the abnormal data based on a pre-constructed knowledge graph and generating fault interpretation according to data corresponding to the abnormal data; and the automatic operation and maintenance decision module is used for generating an automatic repair strategy according to the abnormal data and the fault interpretation and executing the automatic repair strategy to repair the abnormal data.

The invention provides an unstructured text data calculation method and system, and relates to the technical field of natural language processing, and the method comprises the steps: carrying out the multi-level processing of input unstructured text data, dynamically adjusting the word segmentation granularity according to the word frequency distribution, constructing a part-of-speech co-occurrence matrix in combination with context semantic information, and extracting entity information. And fusing the part-of-speech co-occurrence matrix and the entity information to generate a hierarchical semantic tag sequence. Feature extraction units with different convolution kernel sizes are adopted to extract feature representation, and cosine similarity between different semantic hierarchies is calculated to establish an association weight matrix. A semantic enhancement vector is constructed based on entity information, and adversarial training is carried out to obtain a multi-modal semantic feature matrix. And calculating the semantic similarity between the fusion feature vectors for clustering, setting difficulty weight sorting according to the complexity, consistency and ambiguity of class clusters, then inputting the class clusters into a classifier, and iteratively optimizing a classification result to obtain a calculation result.

The invention provides a construction scene prediction method and device fusing an image, a text and a BIM modal, and relates to the technical field of intelligent construction prediction management. According to the method, the BIM semantic graph is constructed by extracting the BIM semantic information of the BIM model, and the target detection recognition of the construction site video is carried out in combination with the YOLO model to obtain the target detection result; performing cross-modal alignment with the CLIP to realize deep fusion of the multi-modal data of the image, the text and the BIM to obtain a multi-modal heterogeneous graph; and inputting the multi-modal heterogeneous graph into a space-time sequence model for prediction, outputting prediction results of construction scenes at a plurality of moments in the future, and dynamically mapping the prediction results to a digital twinborn platform to realize risk early warning and visual display. The construction dynamic change can be captured in real time, the construction progress and risk can be accurately predicted, and the intelligent level of construction management is improved.

The invention discloses an intelligent real-time emotion evaluation method for social media and online text data based on a multi-modal knowledge graph, and the method comprises the steps: firstly obtaining and preprocessing multi-modal original data; secondly, designing single-mode feature extraction of dialogue texts, voices and facial expressions based on BERT-BiLSTM, AST and ResNet-50 pre-training models; furthermore, sharing among multiple modes and private emotion semantic information are considered, and complementation of cross-mode emotion information is promoted by enhancing repeated emotion information and utilizing uniqueness of private information. Meanwhile, a multi-mode knowledge graph method is provided from bottom to top to carry out emotion real-time evaluation. Analyzing user behaviors by using DeepFM and generating user features, and capturing external knowledge layer features in combination with K-BERT; finally, on the basis of a TransformerEncoder structure, cross-modal emotion information interaction and feature fusion are achieved, real-time recognition and classification of multi-modal emotion perception are completed through a Softmax layer, and therefore the purpose of intelligent real-time emotion evaluation is achieved.

The invention provides a large language model-driven code review and optimization method, which belongs to the field of computer science, and comprises the following steps of: firstly, performing intensive training on a model through rich code samples and accurate manual annotation to ensure that the model has accurate code understanding and analysis capability; and then, performing deep analysis on an input source code by utilizing the trained large language model, and constructing an abstract syntax tree (AST) to capture deep semantic information of the code. And finally, the model automatically analyzes the codes, identifies potential defects, non-standard coding practices and optimization possibilities, and generates specific optimization suggestions, including code repair schemes and guidance following optimal practices, according to identification results, so that efficient support is provided for developers, and the quality of software products is ensured.

The invention discloses a hierarchical semantic-driven retrieval enhancement generation method and system, a natural hierarchical relationship and a semantic boundary of a document are effectively reserved by constructing a tree hierarchical structure based on a document chapter title, and a recursive semantic boundary splitting strategy is adopted to refine overlong text nodes, so that the semantic integrity is ensured, and the retrieval enhancement generation efficiency is improved. And the model input length limitation is met, and the semantic information is prevented from being lost. Meanwhile, node knowledge point extraction and abstract generation are achieved through a large language model, top-down multi-level title path transmission and bottom-up content aggregation are combined, the structural perception and semantic expression ability of nodes is enhanced, and in the retrieval stage, based on similarity distribution of query and node semantic expression, an adaptive retrieval threshold value is dynamically calculated, and the retrieval efficiency is improved. A fixed top-k retrieval strategy is replaced, intelligent screening of different query and hierarchical nodes is achieved, information coverage and redundancy suppression are balanced, and retrieval efficiency and accuracy are remarkably improved.

The invention provides a vehicle-mounted video monitoring system and a vehicle-mounted video monitoring method integrating 360-degree look-around perception and multi-mode intelligent analysis. According to the system, video images around a vehicle are collected through a vehicle-mounted 360-degree all-round camera, high-precision vehicle position information is obtained through a vehicle-mounted multi-sensor fusion positioning technology, videos and position time are stored in an associated mode, and a space-time video database is constructed. The system utilizes a multi-mode perception fusion technology to intelligently analyze a video, extracts semantic information, and supports a user to quickly retrieve a target video clip through natural language interaction. Meanwhile, the system can monitor abnormal driving behaviors and vehicle surrounding environment risks in real time and trigger early warning when necessary. According to the invention, advanced technologies such as 360-degree looking-around perception, multi-sensor fusion positioning, multi-modal intelligent analysis, natural language interaction and the like are innovatively and deeply fused with vehicle-mounted video monitoring, so that the limitation of a traditional vehicle-mounted monitoring system is broken through, the intelligent and humanized level is greatly improved, the development direction of a vehicle-mounted video monitoring technology is guided, and the vehicle-mounted video monitoring system has a wide application prospect. The method is of great significance in guaranteeing driving safety of intelligent automobiles and promoting development of intelligent traffic.

The invention discloses a lightweight malicious network traffic detection method based on heterogeneous modal feature fusion, and mainly solves the problems of low feature extraction efficiency and insufficient single modal feature representation of the existing method. Comprising the following steps: acquiring and optimizing a network flow data set, preprocessing the network flow data set, and extracting and generating spatial feature, time sequence feature and behavior pattern feature vectors; an improved self-attention mechanism network is constructed, a lightweight heterogeneous modal feature fusion model LMF is designed, multi-modal feature deep fusion is performed through dynamic weight distribution, a lightweight classification model is trained, and the model is utilized to detect network malicious traffic. Through the lightweight heterogeneous modal feature fusion model LMF and in combination with a dynamic weight distribution mechanism, spatial distribution, time sequence dependence and behavior semantic information of network traffic are deeply mined, data processing efficiency, malicious traffic detection accuracy and system robustness are improved, and the method is suitable for efficient malicious traffic identification and defense in the field of network security.

The invention relates to the technical field of image processing, can be applied to business scenes of medical health, financial science and technology and the like, and discloses an image sensitive character desensitization method which comprises the following steps: extracting a plurality of feature maps from a to-be-processed image, and inputting a feature pyramid network to generate pyramid features, fusing a spatial attention mechanism to enhance the response of a sensitive region in each level of feature map, generating candidate region coordinates through a region candidate network, mapping the candidate region coordinates to a to-be-processed image to obtain a local image, extracting text semantic information in the local image by using a codec based on a self-attention mechanism, and obtaining a local image; and performing sensitive word detection in combination with the semantic classification model, and performing local texture repair operation on the sensitive area according to a detection result to generate a desensitized image. According to the method, a space attention mechanism and a semantic understanding module are introduced on the basis of multi-scale feature fusion, so that the capability of detecting small-size sensitive characters in an image is improved.

The present invention relates to the technical field of knowledge graph completion tasks, and particularly relates to a knowledge graph link prediction method. The method comprises: using a precoding model to obtain an embedding layer vector, and constructing a corresponding masked triple; adding a corresponding position code to each element in the masked triple, so as to obtain a corresponding input sequence, inputting the input sequence into a trained main masking model, and outputting an entity classification probability; and on the basis of the entity classification probability, predicting potential candidate entities. The method further comprises: concatenating semantic information corresponding to the embedding layer vector and structural information obtained by an embedding model, so as to obtain fused head entity and relation representations, and constructing a corresponding fused masked triple; and adding a corresponding position code to each element in the fused masked triple, so as to obtain a corresponding fused input sequence. The present invention uses a precoding method, thereby effectively reducing the training burden on a model, and improving the inference speed of a model; and a fusion module is used before inputs are fed into a main masked model, thereby ensuring the integrity of textual description information and improving prediction accuracy.

The invention discloses a low-altitude airspace management method and system based on data analysis, and relates to the technical field of air traffic management and low-altitude airspace planning, and the method comprises the steps: 1, multi-scale dynamic grid division: adjusting the grid granularity in real time according to the aircraft density and environment complexity, generating a multi-scale grid by adopting an optimized quadtree data structure, and carrying out the multi-scale dynamic grid division; boundary smoothing processing is carried out on boundary areas of the grids with different scales; 2, intelligent airspace layering: dynamically adjusting the thickness of a height layer according to a risk score, and setting a buffer layer to enhance the flight safety; and 3, intelligent semantic annotation: giving dynamic semantic information such as risk levels and traffic priorities to the grid units, embedding airspace control regulations and flight rules, and realizing multi-source data fusion and real-time updating of semantic information. According to the method, the problems of static division limitation, boundary problem, insufficient real-time performance and the like in the prior art are solved, and the airspace resource utilization rate and the flight safety are improved.

The invention discloses a community intelligent monitoring and emergency linkage method and system fusing BIM spatial semantics, and the method comprises the steps: constructing a BIM scene map, and obtaining the attributes and mutual relationships of components and spatial regions in a BIM model; mapping a dynamic target detected in video monitoring into the BIM model, and obtaining spatial semantic information of the dynamic target; based on BIM spatial semantic information of a dynamic target, a target-environment interaction graph is constructed, a graph neural network model is used for training and reasoning, and specific complex events related to spatial contexts are recognized; taking the BIM model as a space-time reference, fusing multi-source heterogeneous data, and reconstructing by adopting a graph-based event association algorithm to form a complete event chain containing an atomic event sequence and an association relationship; and when an emergency event or an event chain is detected to indicate an emergency state, combining BIM preset information and real-time sensor data, dynamically generating an optimal emergency plan, and performing visual commanding and dispatching through a BIM three-dimensional scene and augmented reality.

The invention discloses an end-to-end monocular visual odometer method fusing space-time semantic information. According to the method, continuous image sequence frames are collected through a color monocular camera, and a multi-information fusion end-to-end deep learning framework is constructed; a heterogeneous training domain is adopted to set various data set course sharing parameter fusion training, continuous image sequences are input, and the end-to-end deep learning framework is dynamically coupled with hidden state feature vectors output historically, so that a feature mapping relation of time sequence perception is formed; and interpretable feature decoupling of the static background elements and the dynamic entity objects in the scene is realized. After iterative feature fusion, the system outputs sparse depth and camera motion poses which conform to scene geometric constraints, so that a camera trajectory estimation model with high robustness and strong generalization ability in a complex environment is constructed. According to the method, the positioning precision and stability of the monocular vision odometer are remarkably improved.

The invention discloses a multi-target scene visual SLAM (Simultaneous Localization and Mapping) method fusing target semantics and Gaussian splashing. The method comprises the steps that input data are preprocessed, two-dimensional space masks and semantic information corresponding to the two-dimensional space masks are extracted, global semantic identifiers are distributed to the two-dimensional space masks corresponding to a background and each target, and a target semantic segmentation map is obtained; reading first frame RGBD data as a current frame, acquiring a target semantic segmentation map corresponding to the current frame, and initializing camera pose parameters; respectively establishing an initial target Gaussian splashing model for the background and each target; adopting a Gaussian splash algorithm to render all the target Gaussian splash models to a next frame, and generating an RGB rendering image, a depth rendering image and a target semantic rendering image of the next frame; and designing a loss function between the next frame of rendered image and the corresponding real image, optimizing the pose of the camera, and updating the parameters of the target Gaussian splash model. According to the invention, simultaneous positioning and map construction of multiple target scenes are realized.

The invention discloses an unmanned aerial vehicle small target aerial photography detection method and system based on RT-DETR, a medium and equipment, and belongs to the technical field of unmanned aerial vehicle visual detection, aerial photography images are collected based on an unmanned aerial vehicle, the images are input into a trained target detection model, and target position and category information is obtained. The method comprises the following steps: firstly, extracting low-layer features of an image, and simultaneously capturing information of a channel dimension and a space dimension based on an efficient multi-scale attention mechanism; and outputting features through the convolution residual block. Key features are processed based on a single-scale feature interaction module; for the coded feature map, fusing information of a shallow layer and information of a deep layer step by step in an up-sampling and transverse connection mode; in a down-sampling stage, context semantic information of a target is reserved, and global and local information interaction is enhanced. According to the method, the accuracy and robustness of small target detection are remarkably improved, and the practicability and deployment value in actual application scenes such as unmanned aerial vehicle aerial photography and remote sensing monitoring are expanded.

The invention relates to the technical field of data processing, and discloses a cross-border e-commerce commodity recommendation system and method based on multi-source data fusion. The system comprises an acquisition module for performing data acquisition to obtain a cross-border unified data warehouse and federal learning cooperation data; the classification module performs text classification processing to obtain a user preference analysis result and an interpretable attention mark; a quantization module carries out quantization processing to obtain a cross-border selection feature matrix; the fusion module carries out weighted fusion processing to obtain a basic comprehensive score and a weight convergence detection result; the evaluation module performs risk evaluation processing to obtain risk probability distribution and cross-border compliance evaluation results; and the sorting module carries out real-time processing through lightweight preprocessing and a flow-type calculation pipeline to obtain a comprehensive score sorting list of cross-border selected products. The problem that a traditional cross-border e-commerce product selection method cannot effectively integrate multi-source heterogeneous data and cannot analyze user feedback text semantic information is solved.

The invention relates to the technical field of building model design, in particular to a building three-dimensional model lightweight design method and system based on artificial intelligence, and the method comprises the steps: extracting geometric-semantic features of a building model through a multi-scale curvature filtering and semantic segmentation network, constructing a fusion feature vector matrix, and obtaining a fusion feature vector matrix; by utilizing an integrated graph convolutional network and a self-adaptive neural simplification network of a double-branch attention mechanism, differential resampling is executed based on vertex importance weight, a simplified intermediate model is generated, a surface microstructure is recovered by means of a generative adversarial network, grid holes are corrected by combining Delou inner triangulation, and a surface microstructure is obtained. A non-uniform rational B-spline curved surface is adopted to reconstruct a key decoration component and output a lightweight model, so that the problems of insufficient geometric feature retention and semantic information splitting in the traditional technology are solved, intelligent, efficient and lightweight of a building three-dimensional model is realized, visual fidelity is ensured while data compression is performed, and the construction quality is improved. And the digital management requirement of the whole life cycle of the building is met.

The invention discloses a Transform-CNN medical image segmentation method and system based on multi-scale fusion semantic enhancement, and relates to the technical field of image segmentation, and the method comprises the steps: collecting medical image data, and constructing a medical image data set; constructing a medical image segmentation network model; training the medical image segmentation network model through the medical image data set; and inputting real-time acquired data into the trained medical image segmentation network model to obtain a medical image segmentation result. According to the invention, double encoders, LGDA modules and the like are adopted to capture multi-scale global features of kidney tumors and enhance representation, and the LGDA module is used for adapting to size and form changes of the kidney tumors; the MLCF module is used for supplementing information of the main encoder; the PSE module captures multi-scale global semantic information and integrates local context information, local features are fused through the SAM module, and kidney tumor positioning in an endoscope image is achieved.

The invention belongs to the field of safety research of advanced persistent threat detection, and particularly relates to an APT detection method based on semantic enhancement and an attention mechanism. The method comprises the following steps: collecting and analyzing a system log, and constructing a traceability graph according to the analyzed system log; obtaining a summary sentence of each node according to attributes of the nodes of the traceability graph and types of edges, inputting the summary sentences into a BERT model, and extracting attribute semantic feature vector representation of the nodes in the traceability graph; generating a sub-graph of each node according to the nodes in the traceability graph, extracting a feature matrix of each sub-graph by using the self-attention graph convolutional network, and taking the feature matrix as a node feature vector representation of each sub-graph; and inputting the node feature vector representation of the to-be-detected node into the lightweight classifier, and carrying out anomaly detection. According to the method, the semantic information of the log is deeply mined and understood, the coding structure and causal information of the traceability graph are deeply mined, and the accuracy and efficiency of APT attack detection are improved.

The invention relates to the technical field of medical data management, in particular to a medical data document classification and marking system which comprises a document structure analysis module, a semantic information identification module, a classification and marking module and a document intelligent label module. According to the method, document visualization and retrieval efficiency is improved by analyzing document content and extracting titles, texts and appendixes, analytic performance and operability of data are enhanced by identifying key information and semantic relations and accurately extracting and labeling medical terms, medicine research data and medical record records, storage is optimized according to priorities in combination with classification and marking, and the method is high in practicability and easy to popularize and use. Clinical values and application scenes of documents are clarified, document management processes are refined, management compatible with clinical data formats is formed through structured storage of marked data and classification label generation, storage normalization and retrieval efficiency are improved, the requirement for accurate information calling in clinical tests is met, document management capacity is improved, and the document management efficiency is improved. And support is provided for improvement of clinical decision-making and research efficiency.

The invention discloses a cockpit active recommendation system and method based on a knowledge graph and semantic reasoning, and relates to the technical field of intelligent cockpits. The system receives natural language voice input of a user, executes voice recognition and semantic analysis, extracts user intention, keywords and slot entities, generates structured semantic information, constructs or calls a knowledge graph structure with semantic relation edges in combination with environment context information, and obtains the knowledge graph structure with the semantic relation edges. Semantic path reasoning is carried out based on the path dependence weight and the semantic similarity, a semantic edge label guided graph attention mechanism is introduced to calculate a path consistency score, a candidate recommendation set is generated, the semantic fitting degree and the path score are fused to sort and output recommendation content, and the graph edge weight and the user portrait are updated based on user feedback. According to the method, semantic understanding precision, recommendation path interpretability and system adaptive capacity are improved, and the method is suitable for personalized voice recommendation, man-machine interaction and scene linkage control tasks in an intelligent cockpit.

The invention discloses a weld defect 3D point cloud detection method, and relates to the technical field of defect detection, and the method comprises the following steps: S1, point cloud data acquisition and preprocessing; s2, point cloud data enhancement and feature extraction; s3, weld defect detection and identification; and S4, explaining a detection result. According to the 3D point cloud detection method for the weld defect, the point cloud data is enhanced by means of point cloud completion and density increase, so that incomplete structural information and semantic information of a target due to few obtained target point clouds are avoided, the influence of point cloud sparsity on a detection result is effectively relieved, and the detection accuracy is improved. Therefore, false detection or missing detection of weld defects is effectively reduced, meanwhile, detection results are labeled and explained in various modes of colors, shapes and labels, various presentation modes of generating a three-dimensional model, labeling defect positions and providing detection reports are provided, and the detection results can be presented to users in an easy-to-understand mode.

Embodiments of the present invention provide a server, a display device, and a digital human processing method. The method comprises: receiving image data; inputting the image data into a shared encoder to obtain a first depth encoding feature; inputting the first depth encoding feature into a segmentation decoder to obtain a segmentation mask; inputting the segmentation mask and the depth encoding feature into a guided flow architecture module to obtain a refined feature; inputting the depth encoding feature and the refined feature into an image matting decoder to obtain a face image having undergone image matting; performing image quality enhancement processing on said face image to obtain a face enhanced image; determining face keypoint data; and generating standard data. According to the embodiments of the present invention, a segmentation mask is obtained by means of a segmentation decoder, and the segmentation mask provides, for an image matting decoder, semantic information obtained by means of segmentation, such that the accuracy of an image matting result is higher, the portrait edge segmentation is cleaner, the quality and precision of standard data are improved, and human assistance information is not needed.

The invention relates to a medical image processing technology, and discloses a medical image segmentation and labeling method and system based on multi-modal information fusion, and the method comprises the steps: constructing a model structure of a medical image segmentation and labeling model based on multi-modal information fusion; preprocessing the data set, constructing the data set by using a random sampling method, and dividing the data set into a training set, a verification set and a test set according to a proportion; training and convergence of the model: training the model through a loss function and an optimizer until the model is converged; and S3, cutting an organ to be segmented from the original image, inputting the cut organ to the model trained in the step 3, inputting a text prompt word for semantic segmentation, and taking a mask predicted by the model as a model segmentation labeling result. According to the method, the model structure is constructed, then the medical image data set is subjected to necessary preprocessing, and then the medical image data is segmented in combination with the guiding text semantic information, so that the efficiency of segmenting and marking the medical image from 0 to 1 is effectively improved.

The invention discloses a large-format image segmentation method and system fusing frequency domain perception and state space modeling. According to the method, three core parts including an encoder, a state space modeling module and a frequency domain perception feature fusion module are included, a feature extraction network based on a ResNet structure is adopted by the encoder, multi-scale feature extraction of an input image is achieved, and effective capture of deep semantic information is ensured. The state space module realizes adaptive optimization of image data through visual state space modeling, and constructs a hybrid architecture in combination with a wavelet multi-scale convolutional network, thereby effectively solving the limitation of the traditional CNN in the aspect of remote dependence capture. And the frequency domain perception feature fusion module effectively optimizes the feature reconstruction process through interactive fusion of multi-frequency domain information, so that efficient and fine pixel-level semantic segmentation is realized. The decoder can integrate a low-frequency global structure and high-frequency edge detail information, and the recovery effect of edges and textures in a high-resolution image is remarkably enhanced.

The invention discloses a multi-agent unmanned aerial vehicle cluster collaborative search method based on a semantic communication technology. The method comprises the following steps: defining a partially observable Markov decision process of an unmanned aerial vehicle cluster multi-agent system according to a search task demand; initializing a semantic communication protocol of interaction between the unmanned aerial vehicle cluster and the environment; summarizing and abstracting semantic information of the unmanned aerial vehicle cluster; unmanned aerial vehicle cluster collaborative search task allocation and path planning are carried out by using semantic information; dynamic adjustment and real-time task optimization of the multi-agent system are realized in combination with semantic information; and evaluating a task completion effect. According to the method, the problem of incomplete search coverage of the unmanned aerial vehicle cluster in the complex environment is effectively solved, the search efficiency and accuracy of the unmanned aerial vehicle cluster in the complex environment can be effectively improved, the search efficiency and task completion quality of the unmanned aerial vehicle cluster are remarkably improved, and the intelligent level and robustness of the system are enhanced.

The invention discloses a building engineering automatic compliance examination method based on AI and BIM, and the method comprises the steps: obtaining a to-be-examined standard text, carrying out the natural language processing, and generating a semantic information group; by adopting a semantic mapping rule, mapping the semantic information group subjected to the information conversion algorithm and entity attributes in the IFC ontology structure, and by utilizing a logic rule generator, converting a mapped target NLP deep learning rule knowledge model into a logic rule; obtaining a to-be-examined BIM model, constructing a plurality of specific IDM views, extracting entity attributes corresponding to a specific field from the to-be-examined BIM model, mapping the entity attributes in each specific IDM view and the entity attributes in the IFC ontology structure, and converting the mapped target IFC engineering feature semantic model into a logic fact by using a preset conversion tool; and performing automatic compliance review on the logic rules and the logic facts by utilizing a rule engine through a rule reasoning algorithm and an ontology semantic reasoning algorithm to obtain a rule check report.

The invention relates to the technical field of crack identification, and discloses a crack identification analysis method and system for a building shear wall, and the method comprises the steps: carrying out the multi-angle shooting and image standardization processing of the surface of the building shear wall, and obtaining a standardized image data matrix; performing multi-model degradation processing and multi-angle image combination to obtain target enhanced image data; inputting the image into a convolution denoising auto-encoder for noise separation to obtain a denoising feature image matrix; inputting a dynamic graph convolutional neural network to carry out local feature and global feature extraction to obtain a local crack feature matrix and a global crack feature matrix, and carrying out feature fusion to obtain a fused feature matrix; and inputting the twin metric learning network to carry out feature measurement and crack classification, and outputting a target crack classification result, thereby realizing effective capture of deep semantic information between crack areas of the building shear wall, and realizing efficient integration of crack features of different scales.

The invention belongs to the technical field of target detection and recognition, and particularly relates to a thinking chain-based large and small model collaborative target detection and recognition method. According to the method, the small model is responsible for most of easy-to-detect targets, the calculation pressure of the large model is reduced, the large model is responsible for suspected samples, vision and language multi-mode reasoning is combined, the overall false detection rate and the omission ratio are both reduced, confidence evaluation is conducted through the joint probability, automatic screening and manual rechecking of uncertain results are achieved, the reliability of key results is guaranteed, and the method is suitable for large-scale popularization and application. According to the'pseudo thinking chain + pseudo label 'method, by means of reasoning and labels generated by the model, data dependence on manual labeling is reduced, only low-confidence samples are manually confirmed, the human intervention range is narrowed, the human cost is remarkably saved, and semantic information with finer granularity is provided for the model by introducing phrase-level feature descriptors. And the identification capability of complex target attributes and states is improved.