123 results about "Invariant feature" patented technology

The invention discloses a metal surface quality detection method and system, and relates to the technical field of metal surface quality detection. The method is used for solving the problems of low microdefect detection precision, weak technological parameter relevance and closed-loop control deficiency of the high-reflection surface. The metal surface is irradiated through multi-angle coherent light field serialization, the phase offset of interference fringes is analyzed to generate three-dimensional shape data, and reflection noise interference is restrained. Defect depth gradient is extracted based on dynamic segmentation of process parameter constraint, deposition temperature and pressure deviation are quantified through deconvolution calculation, and process deviation feature distribution is constructed. Finite element simulation is utilized to generate a process-morphology mapping atlas library, cross-domain invariance features are extracted through depth constraint manifold alignment and comparative learning, and a causal correlation model of defect types and process parameters is established. And dynamically adjusting process parameters according to the weight gradient, and reflowing data to update the manifold rule. And high-precision three-dimensional defect detection, process deviation traceability and adaptive parameter optimization are realized.

The invention provides a multi-modal sentiment analysis method and system based on main modal two-stage guidance, and relates to the technical field of sentiment analysis. Inputting the multi-modal data into a multi-modal sentiment analysis model, and extracting language, visual and acoustic features from the multi-modal data through a feature extraction module; semantically decoupling the multi-modal features into modal invariant features and modal unique features through a feature space distribution alignment module, and realizing feature distribution alignment dominated by language modals through alignment reconstruction constraints; performing self-attention modeling on the modal invariant feature through an attention enhancement module to obtain a first enhanced feature, and adaptively enhancing the visual and acoustic unique features through a cross-modal attention mechanism by taking the language unique feature as a dominant feature to obtain a second enhanced feature; the first enhanced feature and the second enhanced feature are fused through the emotion prediction module, an emotion intensity prediction result is obtained through regression prediction, and the accuracy and robustness of emotion analysis in a complex scene are improved.

The invention provides a large model tuning method and system based on multi-modal information and AI, and relates to the technical field of artificial intelligence, and the method comprises the steps: extracting multi-modal parameter features and interaction features, and dynamically adjusting a contribution coefficient to achieve feature migration, and obtaining a cross-modal fusion vector; aligning the modal features by adopting an adversarial mechanism to obtain domain invariant features; constructing a prompt chain based on the modal confidence, optimizing the prompt chain by using AI reinforcement learning, and iterating to converge to obtain an optimization result; and finally updating the large model parameters. According to the method, effective fusion of multi-modal information is realized, and the tuning effect and generalization ability of the man-machine interaction large model are improved.

The invention provides a user consumption behavior multi-dimensional portrait analysis method and system based on a neural network, and relates to the technical field of data analysis, and the method comprises the steps: obtaining user historical consumption behavior data, and constructing a basic feature vector; key time sequence features are determined through a sub-sequence dynamic pruning algorithm and entropy value weighted mapping; extracting sequence features by adopting a bidirectional long-short-term memory network; constructing a multi-task adversarial feature extraction network to obtain scene invariant features; performing feature fusion to obtain multi-dimensional combined features; constructing a feature index tree to calculate user similarity; and hierarchical clustering is carried out to obtain a consumption behavior portrait. According to the invention, high-precision user portraits are realized, and scene adaptability and calculation efficiency are improved.

The invention discloses a cross-subject electroencephalogram emotion recognition method and system based on dynamic domain invariant representation decoupling and recombination, and belongs to the technical field of artificial intelligence. The invention provides a non-personalized decoupling and recombination framework for cross-subject electroencephalogram emotion recognition, and aims to separate emotion-related individual invariant features from cross-subject individual invariant features through complex electroencephalogram signal characterization obtained through dynamic decoupling, so that individual differences are eliminated while emotion classification performance is guaranteed. Specifically, the method comprises the following steps: carrying out original EEG data analysis and preprocessing by using MATLAB and Python MNE libraries; frequency spectrum and space features of EEG signals are extracted through a multi-channel frequency spectrum space self-attention mechanism module, and capture of emotional features is enhanced in combination with a self-attention mechanism and a cross-attention mechanism; a joint distribution alignment method based on a category prototype is adopted, and decoupled feature distribution is optimized, so that invariant features in subjects and invariant features among subjects have higher distinction degree in emotion classification; the optimized decoupling features are recombined through a linear network, the two features are coordinated to perform more sufficient emotion representation extraction, and emotion classification is performed through a multi-layer perceptron. According to the method, excellent cross-subject emotion recognition performance is obtained on multiple data sets, and the emotion recognition rate of the electroencephalogram signals in a cross-subject scene can be effectively improved.

The invention relates to the technical field of artificial intelligence, and discloses an unmanned aerial vehicle flight control-oriented intelligent obstacle recognition method, which comprises the following steps of constructing an adaptive wind disturbance fuzzy kernel based on unmanned aerial vehicle attitude data, and synthesizing a degraded image for training; a specific obstacle recognition model integrating a frequency domain and space domain joint feature decomposition module, a turbulence invariant feature enhancement module guided by physical prior information and a double-branch anti-fuzzy feature extraction network is adopted, and a fuzzy robustness contrast loss function is combined to carry out progressive training so as to learn feature representation insensitive to fuzziness; after the trained model is deployed, the wind disturbance fuzzy intensity of a real-time input image is estimated on line, the internal workflow of the model is dynamically adjusted according to the wind disturbance fuzzy intensity, the attention weight is adaptively adjusted, and the output of a trunk or an anti-fuzzy branch is selected. According to the invention, through coupling of the physical model and deep learning, the recognition robustness, accuracy and stability of the model in a dynamic wind disturbance environment are improved.

The invention provides a cross-domain fault diagnosis method based on a self-learning convolution domain adversarial network, and the method comprises the following steps: constructing a self-adaptive feature extractor based on self-learning convolution, and achieving the dynamic perception of signal features through a convolution parameter self-learning mechanism; constructing a self-learning convolution domain adversarial network comprising a self-learning convolution-based self-adaptive feature extractor, a domain classifier with a gradient inversion layer and a fault classifier; and inputting the source domain data set and the target domain data set into the self-learning convolutional domain adversarial network by adopting a cross-domain joint training strategy, synchronously optimizing cross-domain feature distribution confusion loss and source domain classification loss, and realizing feature alignment of the source domain and the target domain. According to the method, the advantage that self-learning convolution can dynamically adjust convolution kernel parameters according to different domain data is fully utilized, so that the model better adapts to cross-domain data distribution with difference in the feature extraction process, invariant features of a source domain and a target domain can be effectively extracted and aligned, and the cross-domain fault diagnosis performance of the model is remarkably improved.

The invention relates to the technical field of multi-modal data processing, and discloses a multi-modal sentiment analysis method and system based on feature decoupling and variational optimization, and the system comprises a feature decoupling module, a modal variational alignment module, an adversarial representation module and a multi-path interactive fusion network. According to the system, modal features are separated into modal specific features and modal invariant features through a feature decoupling technology, uncertainty modeling and re-parameterization processing are carried out on exclusive features by adopting a single-modal uncertainty perception fusion technology, and robustness of feature learning is enhanced by adopting an adversarial training mechanism for shared features. Model optimization is carried out by combining a cross-modal attention mechanism modeling inter-modal interaction relationship and a multi-level loss function, so that efficient feature extraction, robust feature fusion and accurate emotional intensity prediction of multi-modal data are realized.

The invention provides a multi-source domain invariant acoustic feature extraction method and system for an equipment operation state, and belongs to the technical field of equipment maintenance, and the method comprises the steps: constructing a multi-source domain invariant acoustic feature extraction network based on a DANN model, and the network comprises a feature extractor, a classifier, a domain discriminator and a multi-domain acoustic feature class boundary constraint module; the feature extractor extracts high-dimensional features of sound signals, the classifier carries out fault mode recognition, and the domain discriminator realizes domain prediction. The multi-domain constraint module generates an embedding space, and calculates the maximum mean value difference, the local maximum mean value difference and the Euclidean distance among different source domain features. The network constructs a loss function by taking minimization of inter-domain difference and classification loss and maximization of inter-domain distance and domain discrimination loss as targets, and carries out adversarial training through multi-source tagged acoustic data. According to the method, domain invariant features are extracted by using adversarial learning and hidden space constraint alignment of multi-source domain acoustic features, so that the influence of feature offset under a cross-working-condition condition is effectively reduced, and the fault mode recognition accuracy is improved.

The invention discloses a series arc fault detection method based on DSCAN and feature fusion. The method comprises the following steps: collecting source domain data and target domain data; preprocessing data of a source domain and a target domain to obtain multi-dimensional features such as a grayscale image after GASF conversion, an IMF component after VMD adaptive decomposition, a frequency domain component obtained by DT-CWT and the like; constructing a four-channel feature fusion module, inputting the original current signals and the converted multi-dimensional features into corresponding channel networks, extracting signal features, and splicing the signal features; and constructing a DSCAN network to learn domain invariant features of a source domain and a target domain, and detecting a series arc fault through a model. Through multi-dimensional feature fusion and cross-domain transfer learning, arc fault features in a complex load scene are effectively extracted, the robustness of the model to different load environments is enhanced, the problem that a traditional method is insufficient in detection precision under mixed loads is solved, efficient and accurate detection of series arc faults is achieved, and the fault detection efficiency is improved. And a reliable technical scheme is provided for electrical safety monitoring.

This invention belongs to the field of remote sensing data processing technology and discloses a method, system, terminal, and storage medium for multimodal remote sensing data representation and fusion. The method includes: acquiring and preprocessing a multimodal image dataset to obtain an initial image dataset; inputting the initial image dataset into corresponding autoencoders to obtain corresponding time-invariant features, and pre-training the corresponding autoencoders using feature reconstruction; classifying the obtained features using the geographic location information of the initial image data to generate positive and negative sample pairs, and adjusting the trained autoencoders using a contrastive loss function constrained by cosine similarity; based on the adjusted autoencoders, extracting features with spatiotemporal invariant information from the corresponding modal images, and performing feature fusion using a spatiotemporal invariant information integration method that considers modal importance; and outputting the fused image features. This invention improves the fusion accuracy of multimodal remote sensing image features.

The invention discloses a training and online diagnosis method and system for a ship fault diagnosis model. The method comprises the following steps: generating a dynamic weight based on multi-source domain and target domain data; using the weight to guide a feature extractor and a domain discriminator to carry out adversarial training so as to learn domain invariant features; fusing the weighted supervision loss and domain adversarial loss, and combining a target domain regularization item optimization model; screening a target domain high-confidence sample through a dynamic threshold value to generate a pseudo label and performing iterative training; and finally, performing real-time fault diagnosis by utilizing the trained model, and adaptively updating model parameters according to online data distribution change. According to the method, the problem of negative migration in multi-source domain fusion is effectively solved, and the generalization ability, the diagnosis precision and the long-term operation stability of the model in the target domain are improved.

The invention discloses a domain offset mitigation method and system based on a power equipment image, and relates to the technical field of cross-domain detection, and the method comprises the steps: constructing a generalization model based on insulator defect detection; based on an IRM risk extrapolation method, regularization items are added to extract invariant features, so that the aligned sample loss of each domain is the same; by measuring the Zensen-Shannon divergence of each domain feature of the selected layer in the generalization model, the distribution difference in the feature space is minimized, so that the similarity of the feature space is the same, and the domain offset between the source domain and the target domain is reduced. According to the method, the generalization of real data is enhanced by reducing the difference between the synthetic domain and the real domain to the greatest extent, so that the problem of domain migration is solved.

This invention provides a method, apparatus, device, and storage medium for fault diagnosis of rotating components of a crane, relating to the field of mechanical fault diagnosis technology. The method includes: processing source and target domain data of the crane's rotating components in a unified format to output a normalized vibration signal; aligning feature distributions through a shared encoder and adversarial training to output domain-invariant features; constructing conditional embedding vectors based on the domain-invariant features and generating target domain fault data using a diffusion model; combining the generated fault data with real data to form a diagnostic dataset, extracting multimodal features, fusing them, and outputting the fault diagnosis result through a classifier. This invention effectively solves the problem of scarce crane fault data through a diffusion model and, combined with multimodal feature fusion technology, improves the accuracy, robustness, and cross-domain generalization ability of fault diagnosis.

The application discloses a hyperspectral wetland image classification method based on a graph capsule neural network, which comprises the following steps: S1, learning feature transformation is performed on an adversarial domain self-adaptive framework, so that source domain samples and target domain samples of a hyperspectral wetland image are matched in features; S2, a graph capsule neural domain self-adaptive network structure is constructed, domain-invariant features and domain-related features are extracted, and transferable features are discovered and shared across domains; and S3, a coupling structure two-classifier is designed, the two-classifier is trained by using the source domain samples, classification differences of the target domain samples are maximized, and precise classification of the hyperspectral wetland image is realized by identifying a classification boundary. Meanwhile, the application discovers transferable knowledge and realizes cross-domain sharing, enhances effective discrimination of a class boundary, and finally realizes precise classification of the hyperspectral wetland image under conditions of unknown regions, complex scenes, and lack, deficiency and imbalance of data types.

The application discloses a feature extraction method and system based on difficult sample mining and multi-granularity division, which is used for a cross-view geographical image retrieval task. The method first preprocesses cross-view street view images and satellite images, and uses a generative model to generate cross-view images, reducing the visual difference between different view images. Then a two-stage difficult sample mining model is constructed, including a sampling strategy based on geographical location and visual similarity, mining difficult negative samples in different ranges, and enhancing the inter-class discrimination ability. Then a multi-granularity feature division module is introduced, the image features are extracted through a ResNet50 backbone network, and the features are divided and fused according to different granularities to obtain rich and robust view-invariant feature representation. Finally, the satellite image to be retrieved is input into the trained feature extraction model, the features are extracted, and similarity matching is performed with the street view image library to obtain the cross-view retrieval result.

The embodiment of the invention discloses a weld defect intelligent detection method and device based on multi-modal feature fusion. The method comprises the steps that an excitation unit is controlled to apply standardized mechanical knocking excitation to a target weld joint, and response signals generated by the weld joint are collected; transforming the time domain signals corresponding to the acoustic signal and the vibration signal based on Lie group transformation to obtain an invariant feature vector so as to construct a multi-modal feature vector; the weight of each feature in the multi-modal feature vector is determined based on the physical parameters of the target welding seam, and the weighted distance between the multi-modal feature vector and each clustering center in a standard database is calculated based on the weight, so that a target clustering center is determined; and in response to the fact that the target weighted distance corresponding to the target clustering center is larger than a distance threshold value, determining the defect type of the target weld joint based on the similarity between the multi-modal feature vector and each defect feature vector. In the embodiment of the invention, higher-precision and more reliable weld defect identification can be realized, and the whole process can be automatically carried out without manual operation.

A method and system for explainable classification of a target point cloud may include receiving a target point cloud comprising N points in a multidimensional space, applying a feature extraction model to extract, for each point, a permutation-invariant feature vector comprising local feature entries, applying a bottleneck function on the feature vectors to produce a global feature vector having F global feature entries, applying a classification model on the global feature vector to classify the target point cloud according to classification criteria, for one or more points, applying an aggregation function over the local feature entries of the respective feature vector, prior to the bottleneck function, to obtain an aggregation value, and indicating importance of the points in the classification based on their aggregation values.

The invention discloses a recognition method of a reversely-installed fastener, a storage medium and equipment, and the method comprises the steps: firstly obtaining an original image of a track, carrying out the data enhancement and domain extension of the original image, inputting the enhanced track image and a domain extension image into a fastener recognition model, and carrying out the iterative training, thereby obtaining a trained fastener recognition model; inputting a track image to be recognized to obtain a fastener recognition result of the track image; the fastener identification model is based on Faster R-CNN and comprises a backbone network, a region candidate network, a domain comparison module and a detection head, the domain comparison module introduces a class-level domain comparison learning mechanism to encourage a detector to understand correlation and difference between fasteners in different fields on the class level, the model is guided to mine fastener class-level domain invariant features, and the detection head is used for identifying the fasteners in the class-level domain. According to the method, the problem that the reversely-installed fastener is difficult to recognize in multiple scenes is solved, the generalization and accuracy of track fastener recognition are greatly improved, the cost is saved, and the method has important practical application significance.

The invention provides a civil engineering structure apparent damage small sample intelligent diagnosis method and system, and relates to the field of artificial intelligence. Comprising the following steps: S1, inputting a labeled source domain data set and a labeled small sample target domain data set, and carrying out data preprocessing; s2, training an encoder through an adversarial training mode containing domain adversarial loss, and extracting domain invariant features; s3, extracting domain invariant features from the target domain data set by using the trained encoder to obtain a target domain invariant feature set; using the target domain invariant feature set training condition to generate an adversarial network, and obtaining a target domain sample set; combining the target domain data set and the target domain sample set to form a mixed training set; s4, using the mixed training set to finely adjust a diagnosis network; and S5, extracting features of a target domain image to be diagnosed by the trained encoder, and inputting the features into the trained diagnosis network to obtain a damage identification result. According to the invention, high-precision, high-robustness and high-adaptability intelligent structural damage diagnosis is realized.

The invention discloses a color three-dimensional image zero-watermark processing method based on eight-element matrix factorization, which comprises the following steps of: firstly, cutting square invariant feature region images at the centers of left and right viewpoints of a three-dimensional image, and carrying out non-subsampled dual-tree complex wavelet transform on the invariant feature region images in different channels; averaging the four low-frequency sub-bands obtained by each channel image to obtain a low-frequency mean sub-band, then performing block scrambling on the low-frequency mean sub-band from each channel, selecting a part of sub-blocks to construct an octuple matrix, and performing real number field octuple singular value decomposition on the octuple matrix; obtaining a maximum singular value matrix and then binarizing the maximum singular value matrix to obtain a robust binary feature matrix; and finally, carrying out XOR on the binary watermark which is scrambled and encrypted by the four-wing chaotic system and the cyclic code and the robust binary feature matrix to obtain a zero-watermark image. Before watermark authentication, a KAZE matching algorithm is used to correct left and right viewpoints of a to-be-authenticated stereo image. The method has relatively high robustness on common non-geometric attacks and geometric attacks.

The application belongs to the field of electroencephalogram signal processing and classification, and particularly discloses an electroencephalogram transfer learning classification method and system based on an aligned reference source domain, which comprises the following steps: obtaining electroencephalogram signals of multiple subjects, calculating the Euclidean distance between each electroencephalogram data for each subject, taking the subject with high intra-class similarity and low inter-class similarity as a reference source domain, and taking the remaining subjects as a target domain to construct a training set; training a classification model based on the reference source domain through the training set, wherein the classification model comprises a feature extractor, a feature resolver and a classifier, the feature extractor extracts reference standard features and target domain features according to the electroencephalogram data of the reference source domain and the target domain; the feature resolver performs feature unwinding on the target domain features to obtain class-related features and domain-invariant features, so that the class-related features are aligned to the reference standard features; and the classifier determines the class of the electroencephalogram data of the target domain according to the class-related features. The application can effectively improve the classification accuracy of electroencephalogram signals.

The embodiment of the application discloses a domain generalization target detection method based on a large model guide and related equipment, a training image set with the same content is obtained, and a foreground enhanced image set is generated in combination with foreground mask guidance, which not only helps to enrich the diversity of data enhancement from the root by means of multiple training data, but also focuses on the target foreground for feature enhancement through the foreground mask, thereby effectively retaining the positioning characteristics of the target while improving the richness of data distribution and avoiding damage to the positioning accuracy of the target. The style of the foreground enhanced image set and the to-be-predicted image is fused to realize feature adaptation, and then a hyper-domain invariant feature encoding is used to obtain a feature map after feature encoding. On the basis of retaining key feature information, the feature map after feature encoding efficiently extracts domain invariant features with strong generalization, thereby avoiding feature information loss and improving feature representation capability, so that accurate target detection results are obtained.

The application discloses a multi-source open set image classification method based on deep adversarial domain self-adaptation, an electronic device and a computer storage medium, relates to the field of transfer learning, and solves the problem that existing multi-source open set transfer learning methods regard unknown classes in a target domain as a large class, ignore the diversity of the classes, and are prone to causing confusion of the unknown classes. The application provides the following technical scheme: the multi-source open set image classification method comprises the following steps: taking a public data set as a test data set, dividing target data set into known classes and unknown classes according to alphabetical order, and pre-processing the target domain data set; introducing a contrast regular term into the pre-processed target data set to obtain domain-invariant features of a source domain data set; constructing a decision boundary for unknown class samples in the target domain data set; predicting the probability of the unknown class samples belonging to the target domain data set; and iteratively clustering the obtained unknown class samples by using an improved unsupervised clustering algorithm. The application is also suitable for the field of multi-source open set image classification methods.

The invention provides a knowledge distillation method for cargo identification model lightweight, and relates to the technical field of knowledge distillation, and the method comprises the steps: constructing a prototype representation learning module and a feature enhancement module, and enabling local semantic prototypes extracted from a large-scale teacher model, such as invariant features of'peel texture ', 'leaf edge' and the like, to be subjected to feature extraction; and the knowledge is migrated to a lightweight student model, so that efficient transmission of the knowledge is realized. The feature enhancement module performs selective enhancement on intermediate features of the student model based on a prototype related attention mechanism, and improves the ability of the student model to discriminate similar goods. Meanwhile, a joint optimization objective including embedded layer distillation loss, prototype consistency loss and classification loss is designed, and the problem of performance reduction caused by cross-task label space difference and data distribution imbalance is relieved.

The invention discloses an unsupervised domain adaptive method based on a cascade learning strategy, and aims to decouple the training process of a source domain and a target domain, prevent semantic information confusion and realize fine-grained adaptation to the target domain. Specifically, the proposed two-stage cascade learning framework mainly comprises two stages of learning processes, in the first stage, a low-rank adaptation fine tuning module is used for performing fine tuning of a source field on a visual language pre-training model so as to learn feature information which is related to categories and invariable in field; and in the second stage, learning knowledge of a target domain by freezing the fine-tuned visual language pre-training model and a low-rank adaptive fine-tuning module and introducing text prompt, and refining a pseudo tag by means of knowledge of a source domain. According to the method, semantic learning and a domain-specific adaptation process are effectively decoupled, so that the performance of the target domain is improved.

The invention belongs to the technical field of fault diagnosis, and discloses a double-path incremental fault diagnosis method fusing an attention mechanism and a prototype, and the method comprises the steps: designing a double-branch network, and carrying out the division of labor through a heterogeneous network: extracting a time domain feature through a Hu moment-based convolutional neural network, extracting a frequency domain feature through a scale invariant feature conversion-based convolutional neural network, and carrying out the classification of the time domain feature and the frequency domain feature; the time domain branch network captures the global features of the operation trend of the equipment, and the frequency domain branch network extracts the local features of abnormal signals of a specific frequency band to form a time-frequency complementary feature space, thereby achieving the comprehensiveness of feature coverage. Considering that local frequency domain features of a high frequency band are generally more sensitive to early faults and time domain global features can visually reflect the overall operation state, an attention mechanism is introduced to realize adaptive distribution of weights, and the accuracy of feature extraction is further enhanced. Random projection is adopted to increase feature dimensions, the linear separability of features is improved, the problem of category overlapping in a low-dimensional feature space is solved, and category boundaries are expanded.

The application discloses a water surface target and underwater target classification method based on unsupervised domain adaptation, and belongs to the technical field of underwater acoustic target depth classification. The application solves the problems of poor generalization and target classification performance of traditional deep learning methods under small sample conditions. The classification model constructed by the application introduces sound field elevation angle structure simulation data to assist training, and extracts domain invariant features of simulation data and measured data through adversarial training, so that the classification performance degradation problem caused by the difference in feature distribution between domains is relieved. The application can realize effective classification of water surface targets and underwater targets under small sample conditions, and does not need to know the class labels of measured data, effectively improving the generalization and target classification performance of the model. The method can be applied to the classification of water surface targets and underwater targets.

The application relates to the technical field of EEG emotion recognition, and provides a cross-subject EEG emotion recognition method and system based on decoupled hybrid meta-learning, which comprises the following steps: constructing a backbone network which is decoupled and connected in series by a cross-subject invariant feature encoding module and a sparse dynamic graph topology adaptation module; firstly, pre-training the cross-subject invariant feature encoding module based on a supervised contrast loss and an adversarial loss, fixing the parameters of the cross-subject invariant feature encoding module after training convergence, and outputting general node features; then, constructing a double-path hybrid meta-learning architecture: in the inner loop stage, a classification head is accessed for parameter adaptation, and then a scene semantic memory reasoning module is accessed for memory construction; in the outer loop stage, the prediction results of the parameter optimization path and the memory cognitive path are weightedly fused, and global meta-parameters are updated. Through the method, the accuracy and robustness of cross-subject EEG emotion recognition are improved.

A method for extracting text information from images includes obtaining an extraction request associated with live data comprising an image; generating, using a prediction model, rotational variant features and rotational invariant features associated with the live data; generating, using the prediction model, text embeddings associated with the rotational variant features using overlapping kernel-based embedding on the live data; generating, using the prediction model, attention values for each pixel in the live data using context attention; applying a trained language model to the text embeddings, attention values, and the live data to generate predictions; and performing extraction actions based on the predictions.