288results about How to "Easy to learn" patented technology

The invention relates to a traffic flow prediction optimization method based on an intelligent optimization algorithm, which is applied to the technical field of traffic flow prediction. The system comprises a multi-modal data acquisition and fusion module, a hybrid deep learning prediction module, an intelligent parameter optimization module, an integrated learning module, a multi-scale error correction module, a prediction result analysis and visualization module, a distributed learning and privacy protection module and the like. Wherein the hybrid deep learning prediction module adopts an attention mechanism to fuse models such as CNN, Transform and LSTM, and introduces an adaptive error correction mechanism, so that the prediction precision is remarkably improved; the intelligent parameter optimization module accelerates the model search efficiency by using technologies such as quantum computing and swarm intelligence optimization; the integrated learning module and the adversarial training strategy enhance the robustness and generalization of the model; and the multi-scale error correction module adopts a time sequence decomposition and self-adaptive online learning method to improve the accuracy of long-term and short-term prediction.

The invention discloses a building construction safety risk assessment method and system, and relates to the technical field of artificial intelligence, and the method comprises the steps: constructing a neural network model, and recognizing an unsafe behavior mode of a worker; receiving a data stream from a construction site in real time, dynamically adjusting risk assessment parameters according to a current construction state and environment change, and predicting potential safety risks; according to the prediction result, generating corresponding prevention measure suggestions; key indexes and abnormal conditions are displayed through a monitoring interface, and a detailed analysis report is generated; according to the invention, through dynamic assessment and timely response to the safety risk of the construction site, the risk assessment parameters can be rapidly adjusted according to the environmental change, the potential risk can be predicted, and the prevention measures and suggestions can be provided, so that the occurrence probability of safety accidents is effectively reduced, and the construction efficiency is improved.

The invention provides an automatic task allocation method and system for an elevator emergency rescue platform, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining elevator operation state data and environment information data, constructing an emergency task model, and obtaining classified emergency tasks; inputting the classified emergency tasks into a deep reinforcement learning model to generate a task allocation scheme; the deep reinforcement learning model is formed by combining a deep Q network structure with a priority scanning algorithm, and a priority experience playback mechanism is introduced; and constructing a multi-objective optimization module comprising a time constraint, a space constraint and a resource constraint, and optimizing the task allocation scheme by the multi-objective optimization module based on an improved particle swarm algorithm to obtain an optimal task allocation scheme.

The invention discloses a multi-scale feature fusion fault prediction method for a high-speed rail bearing. The method comprises the following steps: acquiring a fault vibration signal of the high-speed rail bearing through a vibration signal comparison technology; noise and interference signals of the high-speed rail bearing fault original vibration signals are removed through an adaptive filter; performing multi-level decomposition on the vibration signal by using wavelet packet decomposition, extracting sub-band signals of different frequency bands, calculating instantaneous frequency and energy distribution of the sub-band signals by using Hilbert-Huang transform, and forming a time-frequency-energy three-dimensional feature map; carrying out weight distribution on each feature dimension in the three-dimensional feature map by adopting a self-adaptive weighting algorithm; constructing a training data set through the time-frequency-energy three-dimensional feature map and the fault type label; and training the long short-term memory network LSTM by using the training data set, constructing a fault prediction model, and performing fault prediction on the high-speed rail bearing. According to the invention, accurate prediction of the high-speed rail bearing fault is realized, and the accuracy of fault detection is improved.

The invention discloses a carbon emission prediction method and system based on multi-source heterogeneous data contrast learning, and relates to the technical field of carbon emission prediction. The method comprises the following steps: acquiring to-be-predicted carbon emission data, and preprocessing the carbon emission data; extracting features of the carbon emission data by using a feature selection method; processing the enhanced representation vector by using the carbon emission prediction model, performing time series data learning and fusion on the enhanced representation vector to obtain a comprehensive representation vector and a weighted fusion vector, performing carbon emission prediction by using the comprehensive representation vector and the weighted fusion vector, and calculating the comprehensive representation vector by using comparative learning to obtain a carbon emission prediction result. And optimizing a carbon emission prediction result according to a calculation result. According to the method, the time sequence and multi-source isomerism of carbon emission are fully considered, time sequence feature information from different time scales is fused, and a comparative learning method is introduced to learn a more effective feature representation vector, so that the accuracy of carbon emission prediction is improved.

The invention discloses a cross-modal foreground distillation method for multi-view three-dimensional target detection, and belongs to the technical field of automatic driving perception. According to the method, a laser radar or a multi-modal three-dimensional target detection model is used as a teacher model, a multi-view aerial view BEV perception three-dimensional target detection model is used as a student model, and the model performance is improved through multi-scale feature distillation; a space attention mechanism and a foreground mask module are introduced, and the student model can simulate feature responses of the teacher model in different space positions under multiple scales, focus on a foreground target and suppress background noise; the adaptive module ensures the consistency of the multi-scale features in feature dimensions and spatial resolution, and finally, the difference between the features and regression output is measured through loss functions such as mean square error and KL divergence, so that efficient cross-modal and multi-scale foreground feature distillation is realized. Finally, the multi-view three-dimensional target detection precision and efficiency are improved, and meanwhile, the distillation requirements of multi-scale features are fully considered.

The invention provides a heterogeneous model training method and system based on federated learning. The method comprises the following steps: each client initializes a local heterogeneous model, and divides the local heterogeneous model into a feature extractor and a classifier; the server initializes a global adapter; the server issues the global adapter to each client; each client uses the received global adapter as a to-be-updated local adapter, and uses local data to train the to-be-updated local adapter, the feature extractor and the classifier to generate the current optimal local adapter, the feature extractor and the classifier; wherein in the training process, the local adapter to be updated and the feature extractor share a classifier; each client uploads the current optimal local adapter to the server; and the server aggregates all the received local adapters to obtain an updated global adapter, and iteratively updates the global adapter until the local heterogeneous model on each client meets a preset requirement.

The invention provides a method and a device for generating a multiphase flow-containing solid rock three-dimensional model based on a diffusion model, and a medium, and can solve the problems that a multiphase pore and a rock skeleton structure cannot be generated at the same time and only a single-phase structure or a pure skeleton image can be reconstructed in the prior art. According to the method, the watershed algorithm is introduced to accurately segment the multiphase features, Gaussian noise is used as diffusion model input, and the model gradually learns' de-noising 'from the noise to obtain a real and complex multiphase rock image in training, so that a high-fidelity multiphase distributed rock internal image can be obtained only by inputting random noise in an inference stage.

The invention discloses a time series data anomaly detection model construction method and an anomaly detection method and system, and the method comprises the steps: reconstructing time series data through employing a model constructed based on a TCN-LSTM-CMA auto-encoder, and generating reconstructed data; calculating a reconstruction error by comparing the time sequence data and the reconstruction data before and after reconstruction; and according to the calculated reconstruction error, carrying out anomaly detection by adopting an isolated forest algorithm, and identifying abnormal data in the time series data. The invention provides a brand-new TCN-LSTM-CMA auto-encoder model construction method, and the method is combined with an isolated forest method to carry out time series data anomaly detection. According to the method, the defects of an existing anomaly detection method in the aspects of precision, adaptability and robustness are overcome. The method has a wide application prospect in the aspect of time series data anomaly detection in the fields of industrial monitoring, network security and the like.

The invention belongs to the technical field of medical image processing, and relates to a medical image registration method based on multi-scale deformation decomposition. Firstly, a multi-scale feature extraction network is constructed, the method focuses on extracting features of a mobile medical image, and the features are transmitted to a hierarchical feature fusion module to be processed; secondly, the hierarchical feature fusion network is used for establishing a mapping relation between the mobile medical image and the fixed medical image; in a feature decoding stage, a mobile decoding feature network and a fusion decoding feature network are respectively constructed, multi-scale decoding features are utilized to output a multi-scale deformation field, and a coarse-to-fine registration strategy is gradually realized by fusing the deformation fields of adjacent stages layer by layer. By designing the double-branch network structure and connecting the gating updating modules in the adjacent stages, the complex deformation processing capability of the registration network is enhanced, so that the final registration effect is improved.

The invention discloses a long time sequence prediction method based on a generative adversarial space-time attention network, and relates to the technical field of deep learning and time sequence prediction, the generative adversarial space-time attention network comprises a generative adversarial network and a space-time attention network, and the training process comprises the following steps: obtaining a time sequence data set and carrying out deletion processing; constructing a generative adversarial network, training the generative adversarial network by using the time sequence data after missing processing, the generative adversarial network being used for recovering the time sequence data after missing processing; preprocessing the recovered time series data set, and obtaining a training set and a test set; a space-time attention network is constructed, the training set and the test set are used to train the space-time attention network, and a space-time attention network model is used to predict the recovered time series data; and performing error analysis on a predicted result and an original time sequence data set. The method aims at solving the problem of missing values of the time sequence data, and meanwhile the limitation of a time sequence prediction model in the process of processing complex, high-dimensional and nonlinear time sequence data is solved.

The invention provides a semi-supervised image classification method and system based on pseudo labels and embedded clustering matching, and belongs to the technical field of image processing. The method comprises the following steps: acquiring a marked image and an unmarked image, and performing enhancement processing on the unmarked image; for the marked image, calculating supervision loss based on a semi-supervised image classification model; inputting the enhanced unmarked image into a semi-supervised image classification model, screening false labels based on a dynamic threshold value and a learning state of a category, and calculating unsupervised loss; respectively mapping the weakly enhanced image and the strongly enhanced image to an embedding space to generate K-Means clustering results which are respectively used for generating a target image and a strongly enhanced embedded clustering image, and calculating image matching loss; total loss is established based on supervised loss, unsupervised loss and graph matching loss, and the performance of the semi-supervised image classification model is improved by minimizing the total loss. Consistency regularization is applied to a single sample and a sample affinity level, so that confirmation deviation is effectively relieved.

The invention relates to the field of audio signal processing, in particular to a sperm whale cry enhancement method based on a VQ-MAE network, which comprises the following steps: step 1, performing noise reduction processing on a sperm whale cry audio signal, and adjusting the amplitude range of the audio signal; 2, processing a frequency spectrum image, and taking an extracted feature vector as a part of MAE network input; 3, processing the audio signal through an LSTM network, and extracting the time sequence characteristics of the original audio signal; 4, designing and training a multi-mode mask encoder (MAE) network, and training fusion weights of different modes; and step 5, performing masking, and selecting a strategy training network with the best masking effect to perform data enhancement. According to the method, the time domain, frequency domain and spatial domain features of the multi-modal network fusion sound signals are introduced innovatively, semantic details in the wry of the sperm whale are captured and enhanced more accurately and deeply, and powerful support is provided for classification and recognition of the wry.

The invention relates to the technical field of intelligent robots, and provides an intelligent question answering method and device, electronic equipment and a storage medium, the intelligent question answering method comprises the following steps: obtaining voice information corresponding to a question asked by a user, and carrying out text conversion on the voice information to obtain a question text of the user; obtaining a user portrait model of the user, wherein the user portrait model is generated according to at least one of a mood feature, an emotion feature, a dialogue habit feature and an age feature of the user; and according to the question text and the user portrait model, querying a plurality of corresponding candidate answers, further determining a target answer from the plurality of candidate answers, performing secondary processing on the target answer to enable the processed target answer to accord with the cognitive level of the age group of the user, and outputting the target answer. The accuracy and correlation of question answering are effectively improved, and the user experience and loyalty are improved.

The invention discloses a blind hyperspectral unmixing method and system based on end member query design and a Transform attention mechanism, and relates to the field of hyperspectral image processing. The problems that a current simplified mixing mechanism is mainly a linear mixing model, ground feature abundance and corresponding end member accuracy are low, a deep learning blind hyperspectral unmixing method depends on good end member initialization, and unmixing performance is seriously reduced during end member initialization are solved. The method comprises the following steps: constructing an abundance mapping module, extracting features in a hyperspectral image, and performing abundance decoding on the features to form an estimated abundance image; mapping the abundance graph into abundance embedding; constructing a Transform module fitted by a hybrid mechanism, and combining end member query, reconstruction query and abundance embedding through the Transform module fitted by the hybrid mechanism to obtain reconstruction query of end member information and abundance information; the original hyperspectral image is remodeled according to the reconstructed query. The method is applied to the field of blind hyperspectral image unmixing.

The invention relates to the technical field of machine learning and software development, in particular to a software development system based on a machine learning algorithm, which comprises an acquisition processing unit, a model training unit, an interaction prediction unit and an optimization feedback unit, and is characterized in that the acquisition processing unit constructs a database, extracts key elements such as variable names and converts the key elements into vectors; the model training unit adopts recurrent neural network hierarchical training and gradually introduces intermediate and complex structure codes and learns logic relation and semantic relation from code snippets related to a basic grammar structure, the interaction prediction unit screens and extracts the code snippets according to code function module labels and historical use frequency, the code snippets are input into a model after vector conversion, and the model is used for prediction. A probability weighting algorithm is used for predicting subsequent fragments, an optimization feedback unit calculates the similarity between predicted fragments and actual fragments, an optimization mechanism is automatically triggered if a threshold value is not reached, parameters are adjusted by utilizing reinforcement learning, and prediction accuracy is used as a key part of a reward function so as to improve the system performance.

The invention provides an electric power operation image detection and distillation method based on yov10. The method comprises the following steps: acquiring an electric power operation image data set; training a teacher model by using the power work image data set; and freezing parameters of the teacher model, and distilling knowledge of the teacher model to the student model. The teacher model YOLOv10-X similar to the student model YOLOv10-N in structure is used for knowledge distillation, it is ensured that probability distribution of the teacher model YOLOv10-X and the student model YOLOv10-N in the feature space has high similarity, and the student model can better learn and adapt to feature representation of the teacher model; according to the method, the similarity distillation loss between pixels is adopted to restrain the student model, so that the student model can learn the strong feature extraction capability of the teacher model; the student model not only can improve the capturing capability of local details of the image, but also can improve the robustness and precision in a complex scene, and is suitable for an application scene having requirements for the detection speed and accuracy in power production work.

The invention belongs to the technical field of natural language processing, and discloses a natural language processing platform and method based on deep learning, and the platform comprises a data collection module which is used for collecting font data, text data, image data and environment data; the data processing module is used for preprocessing the collected font data, text data, image data and environment data and obtaining a comprehensive feature data set; the character recognition module is used for predicting character recognition results through a character recognition model and judging whether the character recognition results are consistent or not; the semantic recognition module is used for generating corresponding character paraphrases through a semantic recognition model; collecting character recognition results and corresponding character paraphrases through the software terminal; the voice output module is used for converting the character recognition result and the paraphrases of the corresponding characters into natural voices and outputting the natural voices by adopting a TTS technology; through the implementation of the technology, students can be helped to better learn and understand.

The invention belongs to the technical field of AI image recognition, and particularly relates to a visualization method and system based on AI image discrimination, a terminal and a storage medium, and the method comprises the steps: S1, obtaining the actual type of a target object, and carrying out the regional processing of the target object according to the actual type; s2, acquiring actual view information of each sub-region through AI imaging equipment; s3, performing optimization preprocessing on the actual view information to obtain preprocessed view information; s4, extracting and processing the preprocessed view information to obtain actual image features; and S5, comparing the preprocessed view information with the actual image features through an AI image recognition technology to obtain a comparison result. According to the method, a large amount of complex image data can be automatically identified and analyzed, the need of manual intervention is reduced, the data processing efficiency is remarkably improved, and the learning and adaptive capacity of the system to new conditions is also enhanced.

The present invention relates to the field of solid-state drive fault prediction, and in particular to a solid-state drive fault prediction method and system based on artificial intelligence. The method comprises the following steps: fine-grained division of the solid-state drive into storage units, and serialization of the storage units to construct a storage unit sequence for each partition; obtaining a solid-state drive operation log; performing multi-time point state evolution on the storage unit sequence for each partition based on the solid-state drive operation log, and performing deep dynamic evolution learning to generate a state evolution law for each storage unit; performing periodic access load peak calculation on the solid-state drive operation log, thereby generating a simulated access data packet for each storage unit; performing dynamic pressure access simulation based on the state evolution law of each storage unit and the simulated access data packet of each storage unit to obtain response state evaluation data for each storage unit. The present invention realizes efficient and accurate solid-state drive fault prediction.

The invention discloses a pointer instrument detection and identification reading method based on improved YOLOv8n, and the method comprises the steps: obtaining the image data of a pointer instrument, and sequentially carrying out the data expansion and dial marking, and obtaining a first marking data set; performing optimization adjustment based on partial convolution modules in a backbone network and a neck network of the YOLOv8n network structure and a convolution module in a head network to obtain an improved YOLOv8n network model; training the improved YOLOv8n network model based on the first annotation data set to obtain a dial region detection model; inputting the first annotation data set into a dial region detection model to obtain a dial data set; performing secondary annotation based on the dial data set to obtain a second annotation data set, and training the dial area detection model to obtain a target frame detection model; inputting the to-be-detected instrument image into the target frame detection model to obtain a target frame; and obtaining the pointer reading of the to-be-measured instrument based on the target frame. And the reading identification efficiency and accuracy of the pointer type instrument are improved.

The embodiment of the invention provides a training method for a long-tail image recognition model, and belongs to the technical field of data processing, and the method specifically comprises the steps: sampling a picture through employing a balance sampling strategy and a random sampling strategy, and carrying out the data enhancement of a random sampling picture, and obtaining a first sample; executing the CamMix operation to obtain a second sample; inputting the first sample and the second sample into a backbone network to obtain sample features; obtaining cross entropy loss, and taking parameters in the linear classifier as prototype features which are not subjected to L2 normalization; enabling the prototype features and the sample features to pass through a multi-layer perceptron network to obtain L2 normalized features, and inputting the L2 normalized features into a prototype decoupling contrast learning loss function to obtain contrast loss; and obtaining a total loss function by combining the cross entropy loss and the comparison loss, and guiding backbone network training according to the total loss function to obtain a long-tail image recognition model. Through the scheme of the invention, the model identification accuracy and robustness are improved.

The invention belongs to the technical field of image recognition, and particularly relates to a skin disease classification recognition system based on deep learning, which comprises a skin image acquisition module, a skin image preprocessing module, a skin state classification module and a state classification optimization module, according to the system, a skin state recognition model is constructed, and a multi-head self-attention mechanism is introduced, so that skin state differences can be distinguished; local features are extracted through a space exploration CNN block, an AssemFormer block learns a global dependency relationship, the local features and the global features are fused, and a lesion mode can be understood; scale variant attention blocks are used for highlighting skin features of suspected lesion areas, Monte Carlo attention blocks are used for carrying out random sampling pooling calculation on feature representation, attention map enhancement blocks are used for generating attention maps, skin state categories are output, and skin state prediction labels are generated for each skin image, so that skin lesion details can be highlighted; quantitative reference is provided for medical staff, and skin state judgment is effectively assisted.

A portable electronic nose used for beer identification and based on a convolutional neural network is characterized in that a vertical annular gas sensor array is arranged, the gas sensor array is connected with a preprocessing module through an ADC analog-to-digital converter, and the preprocessing module is connected with a beer identification module; the gas sensor array is used for collecting beer gas signals and transmitting the beer gas signals to the preprocessing module; the preprocessing module is used for preprocessing the beer gas signal and then transmitting preprocessed beer gas data to the beer identification module; and the beer identification module is a two-dimensional convolutional neural network model CC-2DCNN based on a channel attention mechanism and a cosine annealing hot restart strategy, and is used for carrying out beer identification according to the beer gas data and outputting a beer identification result. The method has the advantages that accuracy and reliability of beer identification are improved, and a reliable solution is provided for detecting counterfeit beer and preventing beer deterioration.

The invention provides an online video training method based on a large model, and the method comprises the steps: constructing teaching data, and displaying the teaching data to a user in a visual manner; carrying out vectorization processing on the teaching data, dividing the teaching data through the similarity between vectors, and adding knowledge point positioning for the teaching data according to a division result; generating a multi-question-type question bank according to the characteristics of different question types in combination with teaching data with knowledge point positioning, and selecting part of questions from the multi-question-type question bank to interact with a user; semantic migration is carried out on the questions so that the questions interacted with the user each time are not completely the same, part of teaching data is played back according to the interaction result of the user and the questions, and personalized review plans and learning suggestions are generated. The system has the beneficial effects that long videos are divided into short videos, and a quick retrieval module and an autonomous question setting module are added, so that the interactivity is increased, the learning interest of students is improved, the learning efficiency is improved, and the learning effect is improved.

The invention relates to the technical field of cross-modal data processing and graph convolution, and discloses a cross-modal data decoupling method, characterization method and characterization system.The decoupling method comprises the following steps that S1, feature extraction is conducted, specifically, features of each modal data are extracted, and the features of different modal data are synchronized in the time dimension; and S2, feature decoupling: constructing a shared mapping module based on an attention mechanism, performing feature decoupling of different modal data, and outputting the decoupled modal features. According to the method, the problem that cross-modal information is difficult to accurately understand and process in the prior art is solved.

The invention discloses a visual target tracking technology based on a local fine-grained target. The visual target tracking technology comprises a visual target tracking system, an establishment method of the visual target tracking system and a visual target tracking method. The visual target tracking system excavates deep features between a template image containing a target to be tracked and a search image, and performs target tracking through classification and regression. The template image is input into a hierarchical self-adaptive mask mechanism module for self-adaptive local fine-grained mask processing, and a template mask image is obtained; the template image feature extraction network and the search image feature extraction network extract deep features of a template image and a search image respectively, deep fusion and relation modeling are performed on the features through the Transform encoder and the Transform decoder, and then prediction of a target bounding box in the search image is performed through classification and regression. The system can track a fine-grained target under the condition that the structure priori is not clear.

The invention belongs to the technical field of ultrasonic sound velocity image reconstruction, and particularly relates to an implicit neural representation network-based ultrasonic fault sound velocity imaging method, which comprises the following steps of: dividing grids in a fault plane imaging region, performing block division on all grids, determining two-dimensional coordinates of any grid in each grid block, and determining a two-dimensional coordinate of any grid in each grid block; converting the two-dimensional coordinate into a high-dimensional feature vector; all the high-dimensional feature vectors are input into a pre-trained implicit neural representation network, output values serve as sound velocity distribution values of corresponding grid blocks, space arrangement is conducted on the sound velocity distribution values of all the grid blocks, the sound velocity distribution values of all the grid blocks in the arrangement result are mapped to the actual sound velocity value range, and a sound velocity image of a fault plane imaging area is obtained; and performing forward modeling on the sound velocity image to obtain sound field data generated by simulation, comparing the sound field data with actually observed sound field data, optimizing network parameters, and taking the final sound velocity image as a reconstruction result. According to the method, the ultrasonic fault sound velocity imaging precision can be ensured without an accurate initial sound velocity model.

The invention discloses a software defect positioning method based on feature crossing and a KAN network. Firstly, different types of features are extracted from a defect report and a source code file, then cross correlation of the features is carried out through a cross layer, and a hidden relation between the features is mined; the KAN is further used for learning a parameterized nonlinear activation function, the process of fitting a polynomial function by the feature cross network is converted into the process of fitting a plurality of one-dimensional functions, dimension disasters are overcome, changes of complex functions are accurately captured and adapted, complex high-order feature interaction information in defect positioning is accurately captured, and the defect positioning accuracy is improved. Therefore, the defect positioning performance is improved. And finally, inputting a result output by the KAN network into a full connection layer, calculating a final similar score, and carrying out descending sorting according to the final similar score to obtain a defect positioning result. The method improves the defect positioning precision, has advantages in the aspect of calculation efficiency, and is especially suitable for key business application scenes with strict time requirements.

The invention discloses a facial acne grading method based on collaborative mapping multi-scale self-distillation, is applied to the field of computers, and aims to solve the problem of low accuracy of acne severity grading in the prior art. According to the method, the convolutional neural network model is trained by adopting a self-distillation architecture, and a novel distillation module optimization training process is provided in combination with the characteristics of acnes, so that the model can notice the uniqueness of acne images, and the severity grading precision of the model for facial acnes is improved. The method can automatically identify the acne features and grade the severity of the acne features, assists doctors in diagnosis and treatment, and accelerates the healing period of the acne.