199results about How to "Effective study" patented technology

The invention discloses a singular value-based pre-trained visual model parameter fine tuning method, which comprises the following steps of: for a pre-trained visual model, selecting levels which need to be subjected to parameter fine tuning from the visual model according to actual requirements, and respectively performing singular value decomposition on a weight matrix of each layer to be subjected to fine tuning; obtaining an optimal rank-r approximation matrix and a residual matrix used for reconstructing a weight matrix, forming a singular value vector by the first r singular values as a trainable parameter, freezing other model parameters, inputting an image sample in a training sample set into a visual model to update the trainable parameter, and obtaining an optimal rank-r approximation matrix and a residual matrix used for reconstructing the weight matrix; and reconstructing by adopting the updated trainable parameters to obtain a weight matrix. According to the method, singular value decomposition is adopted to initialize the low-rank matrix, learning of downstream tasks on new domain knowledge is enhanced while pre-training information is reserved, the number of trainable parameters is remarkably reduced, the knowledge and adaptability of the model are effectively reserved, and the fine adjustment effect of the model is improved.

The invention discloses a remote sensing image ship detection method based on SPD-YOLOv8, which belongs to the technical field of satellite remote sensing, is used for ship detection, and comprises the following steps: acquiring a satellite remote sensing image ship data set in an experimental sea area, cutting a remote sensing image, and performing radiation correction and geometric correction; dividing the data set into a training set, a test set and a verification set; performing image enhancement on the data set; an SPD-YOLOv8 ship target detection model is constructed; performing ship target detection by using the ship target detection model; parameter tuning and result evaluation are carried out; deploying the ship target detection model to a server; and abnormal behaviors and illegal activities are detected and early warned. Compared with the prior art, the method has the advantages that the detection effect of low-resolution images and small targets is improved, the accuracy, real-time performance and response speed of the detection model are improved, ship targets of different sizes and types can be learned and generalized more effectively, and the robustness and practicability of a detection algorithm are improved.

The invention discloses an electric power system state estimation method based on graph contrast learning. The method comprises the following steps: constructing graph sequence data of electric power system state estimation; constructing an electric power system state estimation model based on graph contrast learning, and putting the graph sequence data into the electric power system state estimation model based on graph contrast learning; inputting the node feature sequence and the masked edge features into a hybrid encoder of a dynamic edge convolutional neural network combining a self-attention mechanism and a residual block and a graph convolutional neural network with a dynamic propagation strategy of a gating mechanism to comprehensively process the features of nodes and edges; comparing an updating mechanism and a sampling symmetry training strategy of model parameters of the learning framework through a graph; and calculating a node voltage phase angle by using the branch phase angle difference, and outputting a state estimation result. According to the method, the internal structure and mode of the power system data can be effectively learned by combining graph contrast learning and a deep neural network technology, and the capability of identifying and filtering abnormal data is improved.

The invention relates to the technical field of novel power system optimization operation, in particular to a new energy high-uncertainty scene generation method and device based on a CWGAN-GP model, and the method comprises the steps: correcting an initial generative adversarial network through employing a Wasserstein distance and a gradient penalty term, so as to obtain a final generative adversarial network, and the target convolutional neural network is introduced into the final generative adversarial network to obtain a conditional deep convolutional generative adversarial network of improved deep convolution, so that WGAN-GP is introduced into the conditional deep convolutional generative adversarial network of improved deep convolution to obtain a CWGAN-GP model of improved deep convolution. And a target L-ISODATA clustering algorithm is applied to cluster the generated wind and light scenes so as to generate a high-uncertainty new energy power generation output scene. Therefore, the problems that in the related technology, when the GAN processes high-dimensional data, complex space-time characteristics are difficult to effectively capture, and due to lack of prior information, the training direction is prone to deviation, and the scene generation effect and stability are affected are solved.

The invention discloses a multi-scale perception feature anomaly positioning method based on knowledge distillation and VAE, relates to the technical field of industrial image anomaly detection, and is used for unsupervised image surface defect anomaly detection. Comprising the steps of image generation, pixel-level abnormal score calculation, teacher clone network, student encoder, multi-scale feature fusion, abnormal region segmentation and the like. According to the method, useful feature representations are extracted from input data by using a clone network C, activation values of features of each layer are extracted to a VAE encoder, and a multi-level double attention module is introduced for an image reconstruction module. The innovation not only relieves the problem of shallow feature interference existing in the model, but also improves the accuracy of the model for obtaining the image perception features and positioning the abnormality at the pixel level. Finally, the method achieves an excellent anomaly detection effect on a public data set, is obviously improved compared with other VAE methods, and proves the practicability and effectiveness of the method in a real scene.

The invention provides a greenhouse climate prediction control method and device, electronic equipment and a storage medium, and relates to the technical field of agriculture, and the method comprises the steps: obtaining climate data and crop data in a greenhouse, and carrying out the multi-source data fusion, and obtaining multi-source fusion data; according to the multi-source fusion data, taking a crop growth state as output, training and verifying a deep neural network, and obtaining a nonlinear dynamic model of the greenhouse climate system; through a Bayesian inference technology, taking parameters of the nonlinear dynamic model as random variables, taking climate data inside and outside a greenhouse as observation data, and establishing posterior distribution of parameter uncertainty, so as to obtain a posterior distribution model of output uncertainty; establishing an optimization problem of a multi-target random opportunity constraint MPC according to a nonlinear dynamic model of a greenhouse climate system, a posterior distribution model of parameter uncertainty and targets of a greenhouse production system; and solving to obtain control input and a control strategy of the greenhouse.

The invention provides an influence maximization method and system based on deep reinforcement learning and a graph neural network, and belongs to the field of influence maximization in social network analysis. The method aims to optimize selection of seed nodes through a reinforcement learning technology so as to realize maximum propagation of information in a network, and the graph neural network can accurately capture an interaction relationship between nodes in a graph structure by virtue of a message transmission mechanism of the graph neural network, which is highly matched with an information propagation process in a cascade propagation model. Therefore, in the influence maximization model based on the deep reinforcement learning and the graph neural network, the graph neural network is combined with the deep reinforcement learning, the embedded representations of the nodes are learned by using the graph neural network, and the embedded representations are used as the input of the deep Q network so as to fully capture the complex structure and node relationship in the network. In addition, a multi-head attention mechanism is introduced to aggregate information of each round, expression of key features is enhanced, and information redundancy is effectively reduced.

The invention discloses an lncRNA-miRNA association prediction method of a multi-channel attention mechanism, and relates to an lncRNA-miRNA association prediction method. The method aims to solve the problem that more effective and robust feature representation may not be learned due to the fact that feature representation of multiple data sources cannot be properly fused and similarity and difference between samples cannot be learned through a contrast learning method in an existing method. According to the method, multi-source data of lncRNA and miRNA is obtained, and multiple similarity matrixes and coding matrixes are obtained through calculation; in combination with a graph convolutional network and a Transform, capturing information from a local angle and a global angle; fusing feature representations of different data sources by using a multi-channel attention mechanism; a contrast learning method is introduced to further optimize feature representation, so that the consistency between different modal features of the same lncRNA or miRNA is ensured, and the discrimination capability of the model is enhanced; in the final prediction stage, KAN is utilized to obtain a prediction score associated with lncRNA-miRNA, and the parameter and calculation efficiency are further improved.

The invention provides a student attention assessment method and system based on big data analysis, and the method comprises the steps: capturing a class image of a student through a camera, collecting the facial expression of the student from the class image of the student, and recognizing the expression type of the student according to the facial expression of the student; acquiring student posture images from the student class images, and judging whether the students raise hands to answer questions or not according to the student posture images; attention scores are calculated from changes in the facial expression types and the body features of the students, and whether the attention of the students is concentrated is determined based on the attention scores. Through the method and the corresponding system, the attention of the student can be evaluated by using a plurality of sequences recorded by the video, and a more real result can be obtained.

The invention discloses a tornado identification and path prediction method based on X-band dual-polarization radar data, which mainly utilizes six observed quantities of two lowest elevation angles of a radar, combines a deep learning model Unet, introduces an attention mechanism and position coding information, optimizes a data transmission process, and constructs a weak supervision type lightweight identification model TD-Unet. The model carries out feature extraction on multi-elevation-angle radar data, and generates a tornado identification probability and a geographic azimuth map. In the recognition result, a geographic coordinate sequence where the tornado is located is further extracted, a particle filtering algorithm is optimized based on tornado motion features, a path prediction algorithm is constructed, and path prediction is achieved through the extracted real-time tornado coordinate sequence. According to the method, the accuracy and efficiency of tornado identification can be improved, a scientific and feasible method is provided for path prediction, and powerful technical support is provided for meteorological early warning and disaster prevention and control.

The invention discloses a real-time three-dimensional multi-machine conflict resolution method based on graph reinforcement learning, and the method comprises the steps: constructing a three-dimensional air traffic interaction scene, carrying out the modeling of the scene into a graph, and obtaining the convolution features of the graph as model input; and establishing a conflict buffer model, and designing a reward function to guide the aircraft to minimize an additional flight distance while ensuring a safety interval. A D3QN reinforcement learning algorithm is adopted for training, the optimal avoidance action of the aircraft is obtained, and real-time multi-aircraft conflict resolution in the high-density airspace is achieved.

The invention belongs to the field of medical ICD automatic coding, and particularly relates to a medical ICD coding classification method based on a graph neural network, and the method comprises the steps: obtaining a data set, and carrying out the preprocessing of the data set, and obtaining a text vector matrix; constructing a graph structure according to the text vector matrix, wherein the graph structure comprises a node feature matrix and an adjacent matrix; inputting the graph structure into a graph neural network model, outputting a prediction ICD coding result, and training the model by adopting a multi-label loss function based on cross entropy; inputting to-be-coded medical record data into the trained graph neural network model to obtain a predicted ICD code; according to the method, the semantic similarity between the medical terms, the level distance between the ICD codes and the co-occurrence frequency relation are included into a graph structure, the complex relation between words and labels can be comprehensively captured, the potential relevance between the ICD codes is better utilized, and the context understanding ability and accuracy of a classification model are enhanced.

The invention discloses a small-scale greenhouse gas emission prediction method based on a pre-training model, and belongs to the field of greenhouse gas emission prediction. According to the method, a greenhouse gas monitoring micro station is arranged in a park, data such as greenhouse gas concentration and meteorological factors are collected in real time, and high-precision traceability of a greenhouse gas emission source and accurate accounting of greenhouse gas emission can be achieved in combination with a high-precision meteorological numerical value mode, a backward trajectory tracking algorithm, a receptor model and a diffusion model; the greenhouse gas emission amount is predicted in real time by calling the pre-training model, and the prediction efficiency and accuracy are remarkably improved. Different from a mode that a traditional model depends on a large amount of historical data training, the pre-training model can be finely adjusted on the basis of a small amount of existing data, so that effective learning and prediction are realized. According to the method, a large amount of time and computing resources are saved, the environmental change can be quickly responded, and real-time and accurate emission prediction is provided.

The invention provides an eight-segment brocade action evaluation method and system based on dynamic planning and deep learning. The method comprises the steps of collecting eight-segment brocade action video data; a matching path, a DTW distance and a single frame score are obtained through an eight-segment brocade action recognition model based on MediaPipe and DTW, and a feature sequence of hub actions output by the eight-segment brocade action recognition model is subjected to uniform frame extraction processing and then is input into an eight-segment brocade action evaluation model based on LSTM to obtain classification labels of user hub actions. And reflecting errors of the user on key actions according to the classification labels. According to the method, skeleton points are extracted through a Mediape framework, the total score is obtained according to the DTW distance, the similarity between the overall recruitment of the user and the standard action is evaluated according to the rhythm score, and the rhythm stability of the action can be reflected in detail according to the single-frame score; the problem that the length of a video sequence is unstable is solved by adopting a uniform frame extraction method, and the model is ensured to effectively learn and recognize action features.

The invention discloses a multi-view middle school experiment step detection method and system based on differential feature enhancement, and the method comprises the steps: carrying out the preprocessing of a video, extracting original and differential video frame segments, obtaining a visual time sequence feature and a differential motion feature through a feature extractor, inputting the features into an encoder, and learning a space-time dependence relation; and carrying out feature fusion by using a cross attention mechanism, and finally outputting a classification result from the fused multi-view visual time sequence features through a decoder. The method can more accurately capture the change and action characteristics of the experiment steps, improves the precision and robustness of the classification model, can effectively solve the problems that the experiment steps in manual detection consume a large amount of manpower resources, the conventional method does not fully explore the time sequence relation, and the efficiency is low, improves the detection precision, and reduces the detection cost. Therefore, the method has relatively high application value and application prospect.

The invention provides an All-In-One image restoration method based on deep learning in the technical field of image restoration, and the method comprises the steps: S1, obtaining a large number of original images, carrying out the randomization processing of each original image, obtaining a corresponding noise image, and constructing a data set based on each original image and the noise image; s2, creating an image restoration model based on the prompt network, the encoder and the decoder, and setting a loss function of the image restoration model; the prompt network, the encoder and the decoder are connected in sequence; the encoder is provided with a plurality of data stream architecture blocks; the decoder is provided with a plurality of prompt blocks; s3, training the image restoration model through the data set, and continuously optimizing hyper-parameters of the image restoration model in the training process; and S4, deploying the trained image restoration model, and performing image restoration by using the deployed image restoration model. The method has the advantages that the generalization ability and the restoration quality of image restoration are greatly improved.

The invention discloses a facial action unit identification method based on CLIP and multi-mode mask prompt learning. The method comprises the following steps: 1, carrying out facial action unit identification on the basis of the CLIP and the multi-mode mask prompt learning; the method comprises the steps that 1, a multi-modal shared AU prompt (AU-prompt) and an attention mask are designed based on an AU prior position, the AU-prompt comprises a plurality of learnable tokens, the attention mask is designed according to prior positions of different AU facial muscle changes, and a specific AU attention mask is obtained; 2, in a multi-mode prompt learning process, introducing an attention mask of a specific AU into different layers of a CLIP model text and an image encoder for prompt learning, and learning local detail features of the AU and global association features among different AUs to obtain local features and global features; and step 3, fusing the local features and the global features, calculating loss with the text features, and finally realizing AU identification. According to the method, the facial action unit can be effectively recognized under the condition that the labeled data is limited.

The invention discloses a cognitive communication interference method and system based on Transform and deep reinforcement learning, and the method comprises the steps: taking Transform Encoder as a deep reinforcement learning network, processing the state information of enemy communication, and selecting an interference action according to the processed information through DDQN; the action with low power consumption is selected by designing a reward function and establishing an invalid interference list, and the convergence speed of the algorithm is improved. The system comprises a communication interference system model construction module, an interference decision module, a state and action space definition module, a reward function design module, a network construction module, an invalid interference list establishment module, a training module and a communication interference module. According to the method, multiple communication signals can be successfully interfered, the interaction frequency can be reduced, the convergence speed is improved, and the performance is relatively stable.

According to the deep learning-based unstructured road travelable area identification method and program, a road image is input into a travelable area identification model for image identification, and a road identification result is obtained. A backbone network of the driving area identification model is provided with a C2f-iRMB module, a neck network is provided with a feature fusion module, and the segmentation detection head is fused and integrated with a SeaFormer module. Compared with an original YOLOv8n-seg model, the driving area identification model integrates position information in a feature image into channel attention through a C2f-iRMB module, semantic segmentation is performed on a road image through a neck network, and target detection and target image segmentation are performed through head segmentation detection. The accuracy, the recall rate, the mAP, the FPS and the FLOPS of the travelable area identification model are greatly improved, and the method is suitable for travelable area identification and segmentation of unstructured roads.

The invention belongs to the technical field of attitude estimation, and particularly discloses an object six-degree-of-freedom pose estimation method based on deep learning. Comprising the following steps: S1, obtaining an original RGBD image, and processing the original RGBD image through a mask technology; s2, constructing a pyramid network, inputting the processed RGBD image into the pyramid network, endowing different features with different weights, and generating rotation and translation prediction of an object; and S3, constructing a loss function, carrying out training and verification through the 6D attitude estimation data set model, and evaluating the precision and robustness of the model when the model processes a 6D attitude estimation task. The invention provides an object six-degree-of-freedom pose estimation method based on deep learning. The object six-degree-of-freedom pose estimation method comprises a pyramid network. The network structure extracts rich color and geometric information in a complex scene, so that accurate attitude estimation is realized.

The invention discloses an aluminum electrolysis cell internal magnetic field prediction method based on a physical information network (PINN), and the method comprises the following steps: inputting related data of a to-be-predicted magnetic field into a trained PINN model, and outputting prediction data; the specific content of constructing the trained PINN model comprises the following steps: S1, constructing a physical model of an internal magnetic field of a target aluminum electrolysis cell; s2, a PINN model is constructed, a loss function is constructed according to the physical model of the internal magnetic field of the target aluminum electrolysis cell, physical loss in the back propagation process is calculated through the loss function through iterative training and back propagation, if a preset condition is met, training is completed, and the PINN model used for predicting the internal magnetic field of the target is obtained; and S3, verification of the PINN model is completed, and a trained PINN model is obtained.

The invention discloses a multi-task reinforcement learning method based on an emergence hybrid expert model, and the method comprises the following steps: constructing a multi-task reinforcement learning network which comprises a plurality of feedforward neural networks, and connecting a routing network in front of each feedforward neural network; performing clustering analysis on each feedforward neural network, and equally dividing the feedforward neural network into a plurality of expert modules; training each routing network based on a strategy gradient descent method, and updating parameters of corresponding expert modules; selecting experiment tasks in the multi-task learning environment; and for each experiment task, activating a corresponding expert module based on the trained routing network to perform decoupling processing, and completing multi-task reinforcement learning. According to the method, an efficient and stable optimization strategy is provided for multi-task learning, and the problem of mutual interference between parameter sharing and tasks in multi-task learning is effectively relieved.

The invention belongs to the technical field of deep learning, and particularly relates to a multi-dimensional data fusion processing method for medical text data, which comprises the following steps: acquiring multi-dimensional data, and inputting the multi-dimensional data into a trained fusion model to obtain a multi-dimensional data fusion result; the step of training the fusion model comprises the sub-steps of obtaining a multi-dimensional data set, and dividing the multi-dimensional data set into a training set and a test set by adopting a data splitting technology; performing feature extraction on the data in the training set, performing tensor conversion on the extracted features, and reconstructing the features after tensor conversion; inputting the reconstructed feature data into the fusion model for training, adjusting model parameters, and finishing model training when a model loss function is minimum; in a model definition stage, a DNN deep learning network architecture is used to ensure that a model can effectively learn and predict a target variable; through the constructed multi-dimensional data fusion processing method, the key information is directly extracted from the original medical text obtained in multiple inspection modes, and the working efficiency of the user is improved.

The invention relates to a robot action generation model training method and an action generation method, and the model training comprises the steps: constructing an automobile wire harness operation data set, and extracting a human body posture; constructing an instruction set based on the data set and the human body posture; pre-training a motion word segmentation device; obtaining a text-action vocabulary; pre-training a language model based on the data set, the instruction set and the text-action vocabulary; and constructing an operation data set, and performing fine tuning on the pre-trained language model based on the operation data set to complete training. According to the generation method, a humanoid action sequence is output through a model trained through a model training method, and the humanoid action sequence is redirected to a robot to complete robot action generation. Compared with the prior art, the generality of the generation model is improved, the robot action more conforming to the automobile wire harness operation can be generated, and the flexibility of the robot for operating the automobile wire harness is improved.

The invention discloses a rail passenger flow short-term prediction method based on dynamic multi-graph fusion space-time deep learning, and the method comprises the steps: carrying out the convolution of historical passenger flow feature data, and forming input data; then, a fusion graph convolution module is constructed, the fusion graph convolution module comprises multiple predefined graph convolution and dynamic graph convolution, and causal time is introduced into the fusion graph convolution module to form a space-time learning device; processing the input data by using a space-time learning device; and finally, the output of the space-time learner is connected through a jump connection layer, the jump connection layer introduces the output of the space-time learner into an output layer, and a prediction result is obtained through the output layer. According to the method, the combination of the predefined graph convolution and the dynamic graph convolution is used as the adjacent matrix, the complex spatial features of the passenger flow can be effectively learned, a causal time multi-head attention mechanism is introduced, and multi-layer spatial-temporal feature extraction can be performed, so that the spatial-temporal feature extraction capability of the model is improved; and more accurate feature representation is provided for short-term passenger flow prediction.

The invention discloses an image classification method for dynamically adjusting temperature knowledge distillation based on information entropy, and the method comprises the steps: obtaining an image with a label, and constructing a target temperature fitting function; the residual error convolution temperature prediction network outputs a dynamic distillation temperature, and accepts supervised training of a target temperature fitting function at the same time; and softening a classification result of the teacher network and a classification result of the student network after logits fusion by using the distillation temperature to realize knowledge distillation training and optimization. According to the method, the information entropy is introduced as a basis for dynamically adjusting the temperature, and the distillation temperature is adaptively adjusted according to the sample complexity. In the training process, the classification results of the teacher network and the student network and the information entropy are used for generating the target temperature. The target temperature is used as a supervision signal to be input into the residual convolution temperature prediction network, the image classification task is optimized, the image sample complexity difference is better dealt with, and the classification accuracy and robustness of the model in the image classification task are improved.

The invention discloses a point cloud semantic segmentation method and system based on extended attention and multi-prototype fusion. The method comprises the following steps: training a to-be-processed point cloud sample to obtain a pre-training model; screening the to-be-processed point cloud sample to obtain a target point cloud sample; performing feature extraction on the target point cloud sample by using the pre-training model to obtain sample feature data; building an expanded attention block model, and training the expanded attention block model based on multi-view prototype loss and the sample feature data to obtain a trained neural network; and performing semantic segmentation on the point cloud data based on the trained neural network. Therefore, a small amount of existing information can be fully extracted, and a segmentation network model with stronger generalization ability is constructed.

The invention provides a visual feature generation method for improving visual language model combinatorial reasoning, and belongs to the technical field of cross-modal semantic enhancement and image feature learning, and the method comprises the steps: constructing a data set, and obtaining a semantic offset between a positive description and a negative description to generate a difference vector; generating visual disturbance through the difference vector, guiding image feature modification, and obtaining visual features which keep semantic consistency with negative description of the text; and the generated visual negative features are applied to contrast learning of a visual-language model, and the performance of the model in a combined reasoning task is optimized. The method has the advantages that disturbance vectors for visual features can be generated through semantic offset guidance of positive and negative samples of a text, complementary information between the text and an image is fully mined, and the quality of generation of the visual negative features is improved; complex semantic information is processed more effectively in the evaluation process, and finally the ability of the model in processing the combinatorial reasoning task is improved.

The invention belongs to the field of target detection, and particularly relates to a construction method of an astronomical image point source target detection model with depth feature matching. The method comprises the following steps: obtaining a training set of astronomical images containing a point source target; the training set of the astronomical image containing the point source target is input into a PSDetNet network, and the PSDetNet network is composed of a feature extraction module and a matching positioning module; utilizing a feature extraction module to obtain a feature map with point source template features; searching and positioning point source features in the feature map by using a matching and positioning module to obtain a target detection model; the model is trained using an improved cross entropy loss function. According to the invention, high integrity and purity can still be maintained at a low signal-to-noise ratio; an astronomical image with a point source target can be converted into a feature map with a point source similar to a point source template; the network efficiency can be effectively improved, and the point source in the astronomical image can be accurately detected with high purity and high integrity.

The invention discloses an artificial intelligence interactive English learning system, and relates to the technical field of English learning, one end of a wireless network module is in electrical input connection with a firewall module, the other end of the wireless network module is in electrical output connection with a selection module, and one end of the selection module is in electrical output connection with an administrator module, a teacher module and a student module; meanwhile, automatic answering of learning questions is achieved, answering results are fed back to the dialog box module to be displayed, the dialog box module is in two modes of a machine answering module and a manual answering module, the interactive teaching module is in electrical output connection with the training module, and the training module is used for training the learning questions through the multiple modes. The interactive learning function module is used for training test questions, such as games, questions and answers, corresponding tasks are completed only by inputting correct test question answering content during training, the communication module is used for communication and interaction between students and communication and interaction between the students and teachers, and the interactive learning function module can provide a functional chat room for the students to communicate and learn.