21results about How to "Fully understand" patented technology

The invention discloses a remote sensing image segmentation method based on multi-scale wavelet transform and Mama. The method comprises the following four steps: firstly, carrying out preprocessing and division on an ISPRS Potsdam data set and a Vaihingen data set; then constructing a segmentation network, wherein the network comprises a local detail extraction branch, a spatial semantic extraction branch, a cross-branch feature fusion part and a decoder; training and optimizing the network by using the training set; and finally, performing segmentation reasoning on a test image by using the trained model. According to the method, the high-frequency detail extraction capability of the image is enhanced through multi-scale wavelet transform, the long-distance dependency relationship is modeled by using the visual state space block, and effective fusion of the features is realized through the double-branch fusion module, so that the feature extraction capability and the semantic segmentation precision are improved, and meanwhile, the training efficiency and the stability are optimized.

The application discloses a movie classification system based on multi-modal deep representation collaborative federated learning, and the method comprises the following steps: firstly, initializing local model parameters of each client with public model parameters; then, the client updates the local model parameters according to the total loss on the local movie multi-modal data, obtains the local multi-modal deep representation, and uploads the local multi-modal deep representation and the local model parameters to the server; the server calculates the global multi-modal deep representation, obtains the initialization local model parameters of each client in the next global round through personalized parameter aggregation, and distributes the local model parameters to each client; finally, each client replaces the local model parameters. Except for the initialization, the above training process is repeatedly executed until the local model parameters of each client converge. For the to-be-recognized movie data, only the category thereof is recognized in the local client. While protecting the privacy, the application effectively improves the classification accuracy of each client on the movie multi-modal data.

The application is suitable for the technical field of network drama auditing, and provides an artificial intelligence full-process auditing method and system for network drama, which comprises the following steps: analyzing a script based on a natural language model, and constructing a script knowledge graph comprising a macro-story layer, a meso-plot layer and a micro-element layer; the macro-story layer is used for representing the core theme and value orientation of the whole drama; the meso-plot layer automatically divides the script into multiple plot units; the micro-element layer is used for representing the entity elements and element development track within each plot unit; based on the script knowledge graph, dynamic risk identification is performed on the script to determine potential risk plot units; the rough cut sample and the script knowledge graph are aligned, and the risk index of the sample segment corresponding to the potential risk plot unit is calculated. The multi-level script knowledge graph constructed based on the natural language model can comprehensively and systematically understand the script content, and improves the accuracy and consistency of the auditing result.

The application discloses a student programming answer prediction method combining semantic understanding and structured modeling, collects student programming homework data and pre-processes the data to obtain data samples, translates the data samples to obtain an English feature set; uses low-rank adaptation to fine-tune a pre-trained semantic understanding model, constructs input features, and predicts an answer correctness probability; performs graph neural network structured modeling based on the association relationship between problems and concepts, performs message propagation and feature updating to obtain an embedding set, inputs the embedding set and student embedding into a prediction layer to obtain a probability of answering a question correctly; and inputs the probability of answering a question correctly and the probability of answering a question correctly after fusion into a multilayer perceptron for nonlinear mapping to predict the probability of answering a question correctly. The application solves the problems of difficulty in effectively processing non-standardized codes submitted by students, insufficient semantic understanding of question texts and low prediction accuracy in programming knowledge tracking, and provides a scientific basis for personalized teaching and learning resource scheduling.

The application relates to the technical field of intelligent decision-making, and discloses an abnormal text recognition method, device and equipment and a computer readable storage medium. The method comprises the following steps: matching a combined word group with a sensitive word group library to obtain a target sensitive word group and generate a sensitive word feature vector; obtaining a word-level feature vector and a character-level feature vector from a to-be-recognized text; splicing the word-level feature vector and the character-level feature vector to obtain a text feature vector; obtaining context information between each word in a word segmentation sequence to generate a semantic feature vector; fusing a rule feature vector, the text feature vector and the semantic feature vector to generate a fusion feature vector; and inputting the fusion feature vector into an abnormal text classifier to calculate a probability value. The application can be applied to a financial technology, medical health and other business system platform, and can recognize text deformation and countermeasures.

The invention provides an auxiliary bid evaluation method, a cue word generation method, a cue word generation device, equipment and a medium, the auxiliary bid evaluation method comprises the following steps: obtaining review standard information and text information and image information of bid inviting and tendering files, the bid inviting and tendering files comprising a purchase file for bid inviting and a to-be-reviewed bid inviting file; semantic analysis is conducted on the review standard information, a review point used for reviewing the bidding file and a review object of the review point are obtained, and the review object comprises text information and / or image information; according to the review point and the review object, analyzing and extracting the image information and / or the text information to obtain support information of the review point; and performing comparative analysis on the review point and the support information to obtain a review result of the bidding document. The auxiliary bid evaluation method provided by the embodiment of the invention is more intelligent, the bid document is more comprehensively evaluated, the evaluation result is more accurate, and the universality is higher.

The invention provides an image processing method and device, equipment and a storage medium, and relates to the field of image processing. The method comprises the steps of obtaining a to-be-processed image, wherein the to-be-processed image comprises a text; performing multi-dimensional recognition processing on a text in the to-be-processed image to obtain a text noise feature of the to-be-processed image; obtaining a text editing prompt word; and performing image de-noising and redrawing according to the text editing prompt words and the text noise features, and generating a target image after the text is edited. According to the method, feature extraction can be comprehensively carried out on the text in the to-be-processed image from multiple dimensions to obtain the text noise features, the text noise features serve as the basis of noise pixel elimination, the obtained text editing prompt words serve as the basis of generation of the redrawing content, and the target image after the text is edited is generated.

The invention relates to the technical field of computers, and discloses a schedule management method, device and equipment, a storage medium and a computer program product.The method comprises the steps that when a user voice instruction is detected, schedule items are extracted from the user voice instruction through a schedule extraction analysis model, and the emotional state expressed by the user voice instruction is analyzed; according to the schedule items and the emotional state, the certainty degree of the user for the schedule is predicted through a certainty degree prediction network, and the schedule of the user is managed according to the schedule items and the certainty degree; the schedule items are directly extracted from the user voice instruction, so that the operation process of schedule management can be simplified, the schedule management efficiency is improved, and the reliability of the user on the schedule is predicted according to the schedule items and the emotional state of the user, so that the natural language instruction of the user can be fully understood, and the user experience is improved. And the multi-task processing accuracy is improved.

This invention discloses a method and apparatus for ranking responses in multi-turn dialogues using common sense enhancement. The method includes: extracting knowledge from a common sense knowledge graph based on first dialogue data to construct an entity subgraph using the extracted entities as context nodes; inputting the preprocessed first dialogue data and entity subgraph into a response network model to output two representation vectors for the context nodes; training and optimizing the response network model using the similarity between the two representation vectors and the loss between the predicted response output and the actual response output as objective functions to obtain a trained response network model; inputting second dialogue data into the trained response network model for multi-turn dialogue response output, and ranking the multi-turn dialogue response outputs to obtain a ranked response result. This invention can significantly improve the efficiency of ranking response candidates.

The application discloses a kind of deep learning methods for identifying phosphorylation sites of SARS-CoV-2 infection, comprising: first, collect the phosphorylation site dataset of known SARS-CoV-2 infected human A549 cells;Then five feature encoding methods are used to extract the feature representation of peptide sequence respectively;The features extracted by the five feature encoding methods are vector splicing;Then a deep learning model is constructed based on convolutional neural network, gating mechanism and bidirectional gated recurrent unit;The encoding vector of positive and negative samples and its label are used to train the deep learning neural network model;The trained model is used to predict unknown peptide sequence, the application makes full use of the advantages of convolutional neural network, gating mechanism and bidirectional gated recurrent unit network, not only realizes the synergistic effect of feature extraction, information integration and time sequence processing, but also provides an innovative and accurate method for efficient identification of SARS-CoV-2 infection phosphorylation sites.

The application provides a surgery navigation method and device based on a multimodal large language model, wherein the method comprises: acquiring user voice input; resampling the user voice input to obtain resampled audio; performing mel-frequency spectrum conversion and normalization on the resampled audio to obtain voice features; determining a voice intent corresponding to the user voice input based on the voice features; performing similarity retrieval in a preset material document based on the voice intent by a retriever to obtain target application program interface text; splicing the target application program interface text and the voice intent to obtain text input; and inputting the text input and visual input into a pre-trained large language model to obtain a text answer output by the pre-trained large language model, wherein the text answer is used to control a preset machine to perform surgery navigation. The application can provide more accurate real-time navigation in neurosurgery.

The application provides an intelligent video analysis method based on large model scheduling, which comprises the following steps: video data acquisition and preprocessing, video content feature extraction and analysis, large model dynamic scheduling and task allocation, video analysis based on the large model and result generation, result fusion and post-processing, online incremental learning and model updating, and the like. The method can efficiently and automatically analyze, has the advantages of quantifying behavior reliability, flexible adaptation to scenes, and support for complex decisions. Meanwhile, multi-dimensional information fusion improves the accuracy of behavior description and the practical application value of the analysis method.

The invention discloses a household electrical appliance control method and household electrical appliance equipment, relates to the field of intelligent household electrical appliances, and aims to perform time sequence prediction by combining different prompt word templates under the synergistic effect of a pre-trained time sequence model and a pre-trained large language model so as to obtain a more reasonable and accurate prediction time sequence. Therefore, the operation of the target household appliance is controlled according to the control instruction generated by the predicted time sequence, so that the operation result of the target household appliance better meets the requirements of the user, and the intelligent experience of the user on the household appliance is improved.

The present application relates to the technical field of computer simulation, in particular to a kind of anti-collision wall stress test digital twin system and method.The system includes: physical anti-collision wall module is laid with multiple sensors to collect actual data;Data acquisition module is responsible for the transmission after pre-processing sensor data;Data processing module is simulated, analyzed and optimized by constructing and calibrating parameterized finite element model in combination with artificial intelligence algorithm on the stress condition of anti-collision wall;User interaction module realizes result display and parameter input interaction;The method is based on the above-mentioned system, covers the complete process from model construction to result output and feedback.By the present application, high-precision, real-time anti-collision wall mechanical property testing and evaluation can be realized, which can effectively reduce the testing cost and improve the efficiency, provide strong support for the structural safety protection and optimization design of anti-collision wall, and is suitable for anti-collision wall structure safety detection and performance prediction in highway, bridge, tunnel and other scenes.

This invention discloses a method and system for medical image segmentation and 3D reconstruction based on region growing. The method includes the following steps: S1, acquiring the original medical image data of the target object and preprocessing it to generate a preprocessed 2D image sequence; S2, selecting seed points on the preprocessed 2D image sequence; S3, segmenting based on the seed points to generate segmented regions; S4, optimizing the boundaries of the segmented regions to generate an optimized binary mask; S5, generating the final binary mask; S6, completing the 3D model reconstruction; S7, spatially registering and fusing the reconstructed 3D model. The semi-automated segmentation process of this invention reduces the subjective differences of manual delineation, achieving over 90% consistency in reconstruction results for the same image across different operators, providing a reproducible technical basis for multi-center studies and efficacy evaluation.

The invention relates to the technical field of network security, in particular to a code vulnerability analysis method and device, computer equipment, a storage medium and a product. The method comprises the following steps: constructing a time sequence code attribute graph corresponding to a historical version code according to the historical version code corresponding to a to-be-detected code; according to the time sequence code attribute graph, determining an evolution perception soft prompt corresponding to the to-be-detected code; and performing vulnerability analysis on the to-be-detected code according to the evolution perception soft prompt to obtain a vulnerability detection result of the to-be-detected code. According to the method, the historical evolution process of the to-be-detected code is fully understood when vulnerability analysis is carried out on the to-be-detected code, and the vulnerability analysis accuracy of the to-be-detected code is ensured.

This invention provides an automated handling system that enables an autonomous mobile robot with a width greater than the spacing between the legs of a shelf to enter the shelf. The automated handling system transports the shelf by the autonomous mobile robot entering under the shelf, which has multiple legs on the side of its bottom surface. At least one of the multiple legs has a U-shaped member, wherein the two ends of the U-shaped member on the open side are arranged in the vertical direction.

A potentially event monitoring includes a plurality of cameras monitoring a monitored area. The cameras capture video streams showing a portion of the monitored area and perform video analysis on the captured video streams. When a camera identifies an event of interest within the captured video stream, the camera generates event information associated with the identified event of interest. The camera uses relative position information to determine which of the plurality of cameras is positioned to capture the event of interest now and / or in the future and instructs that camera to track the event.

The application provides a welding seam extraction method and device based on multi-modal information fusion, equipment and storage medium, relating to the technical field of computer vision. The method comprises: acquiring an RGB image and a 3D point cloud of a welding seam; performing coarse positioning processing on the RGB image to obtain mask pixel coordinates of the welding seam region; based on the alignment relationship between the 3D point cloud and the mask pixel coordinates of the welding seam region, the initial welding seam region is cut from the welding seam region to obtain the 3D point cloud of the initial welding seam region; using plane segmentation and nearest neighbor search based on the KD tree data structure, the 3D point cloud of the initial welding seam region is processed to extract welding seam feature points; and the welding seam feature points are fitted to obtain welding path points. The application solves the problem that the traditional method is difficult to simultaneously consider anti-redundant point cloud interference and accurate welding seam three-dimensional information extraction in a complex industrial scene, and improves the processing efficiency and extraction accuracy of the welding seam in the welding region.

The application relates to the technical field of natural language processing, and provides a mask enhancement named entity recognition method based on a large model, which comprises the following steps: collecting to-be-recognized text data; preprocessing to obtain an input sequence, inputting the trained recognition model to obtain a recognition result; the recognition model training process comprises the following steps: performing mask processing on the training input sequence based on a set mask strategy to obtain a mask input sequence, inputting the mask input sequence into a BERT model to obtain entity and mask context representation features; performing a named entity recognition task and a mask prediction task and sharing parameters to obtain entity prediction values and mask prediction values; calculating a first loss function based on the entity context representation features and the entity prediction values, and calculating a second loss function based on the mask prediction values; updating model parameters; evaluating model performance, and repeating training until the performance reaches a set requirement. The application can fully understand semantics, has strong generalization capability, strong context dependency capturing capability, and less misrecognition and missed recognition.

The present application relates to the field of small sample target detection, especially to a multi-detection head two-stage small sample target detection method based on improved SPP structure, mainly comprising the following steps: designing five-branch Spatial Pyramid Pooling (SPP) structure; adding the improved SPP structure into a new detection head branch, and inserting the branch into an existing network structure to construct a detection network; taking a new class small sample data set with a small amount of labeled information as input to fine-tune the parameters of the detection head part of the detection network; inputting a to-be-detected data set into the detection network to obtain a detection result. Compared with the prior art, the present application has the following advantages: the detection performance of the model on various size targets is improved; the SPP structure is integrated in the new detection head, which can more flexibly and efficiently process targets of various scales, reduce the calculation cost, reduce the number of parameters, and improve the stability of feature expression, which helps to further improve the accuracy of target detection.