495 results about "Dynamic feature" patented technology

The invention discloses a nondestructive testing method for internal defects of a concrete structure fusing big data feature extraction and deep learning. According to the method, through multi-source data collaboration and dynamic feature fusion, the accuracy and robustness of concrete structure defect detection are remarkably improved. In a data processing link, ultrasonic electromagnetic induction infrared thermal imaging data and the like acquired by a multi-source nondestructive testing technology are subjected to collaborative preprocessing, so that the influence of noise interference and environmental fluctuation is eliminated, and standardized input is provided for feature extraction. The dynamic weight distribution network further combines the relevance of each modal feature in a historical defect sample, adjusts fusion weights of different modals in real time, reinforces ultrasonic features with great contribution to cavity recognition or infrared features sensitive to cracks, effectively compresses redundant information, and improves the accuracy of cavity recognition. According to the method, features and data-driven deep features of the fused feature vectors are manually designed at the same time, so that the limitation of single-modal data is avoided, and a model can more accurately capture multi-dimensional features of defects.

The invention discloses a long-time pedestrian re-identification method based on dual-path cooperation and key frame guided reconstruction. The method comprises the steps of firstly collecting a pedestrian video to be recognized, and extracting a video feature sequence; space and time position coding is introduced into the video feature sequence; capturing local fine-grained dynamic features through a local dynamic feature capturing path, and modeling long-range time sequence association through a cross-frame global feature modeling path; then, dual-path feature complementation is realized through bidirectional gating interaction; further screening out key frames, and realizing feature reconstruction through a full-frame attention propagation mechanism; and finally fusing the dual-path fusion features, the key frame guide reconstruction features and the refined features to generate pedestrian identity features. And processing pedestrian identity features to obtain standardized feature vectors, performing similarity comparison on the standardized feature vectors and pedestrian features in an image library, and returning a matching list. According to the method, video time sequence information is fully utilized, and the problem of insufficient robustness caused by appearance change in long-time pedestrian re-identification is effectively solved.

The invention provides an intelligent data alignment method and system based on time sequence dynamic multi-source embedding mapping. The method belongs to the technical field of multi-modal data fusion and spatio-temporal information processing. The method comprises the following steps: performing time sequence dynamic feature extraction on a multi-source heterogeneous data source to generate a heterogeneous data sequence containing a time dependency relationship; and constructing a time sequence dynamic multi-source embedded manifold space based on a manifold learning theory, mapping a heterogeneous data sequence to a unified evolution geometric structure representation space, and generating embedded manifold data. Through the method, heterogeneous data from various different data sources can be effectively processed, unified mapping is carried out through time sequence dynamic feature extraction and a manifold learning technology, cross-source alignment of the data is achieved, and the method is particularly suitable for a data scene needing to consider a time dependency relationship.

The invention discloses a dynamic feature retrieval generation and management method based on a cross attention mechanism, which belongs to the technical field of natural language processing, and comprises the following steps: step 1, converting knowledge base document fragments into atomic knowledge units, each atomic knowledge unit comprising question and answer pairs, codes as key value pairs, and adding dynamic priority weights; self-attention query is replaced with a double-channel query structure, one channel is used for cross attention, a cross attention query vector is generated through linear transformation, and key value pairs of atomic knowledge units are used; and step 3, training a cross attention adapter, freezing the weight of the language model, optimizing parameters of the adapter, and dynamically adjusting a loss function by using pre-judgment parameters based on input sequence context complexity. By means of the method, deep correlation description of user input and knowledge fragments can be achieved, semantic ambiguity is eliminated, knowledge injection and context information are balanced, distortion is avoided, and the strict requirement for semantic precision in the professional field is met.

The invention belongs to the technical field of image processing, and discloses a nonferrous metal target detection method and system for a scraped car. According to the method, through deep coupling of a GAN bimodal fusion network and a double-backbone network, a scraped car nonferrous metal target detection model is constructed, so that nonferrous metals in scraped cars are efficiently and accurately sorted. A bimodal fusion network is introduced into a model input layer, a fusion image with infrared thermal saliency and visible light texture features is generated, and the problem of cross-modal information splitting is solved; according to the dual-backbone network, standard convolution is replaced by a three-branch structure of an MGHCM module, large target contours, middle target semantics and small target details are synchronously captured, and self-adaptive shunting and efficient fusion of multi-scale features are achieved in cooperation with dynamic feature routing of a DHFBlock module; and the detection head network is optimized into a rotating frame prediction and cross-modal cross entropy classification mechanism so as to adapt to the placement scene of the metal fragments at any angle.

The invention relates to the technical field of data analysis, and discloses a multi-modal time sequence anomaly analysis method, device, equipment and medium, and the method comprises the steps: collecting visual data, audio data and process text data, carrying out the preprocessing and standardization processing of different types of data, constructing a multi-modal data set with aligned timestamps, and storing the multi-modal data set in a database; the method comprises the following steps: extracting a time-frequency dynamic feature and a semantic vector feature, extracting a map structure feature, a time-frequency dynamic feature and a semantic vector feature, fusing the features by using a cross-modal attention mechanism to generate a fused feature vector, finally performing analysis processing based on the fused feature vector, and outputting an analysis result. According to the method, the multi-modal data set with consistent time is constructed, the structural features of various modals are extracted, and the cross-modal attention mechanism is introduced to realize deep fusion of the feature level, so that the problems of single information utilization and weak feature relevance of the existing detection means are solved, and the comprehensiveness of defect detection and the accuracy of fault diagnosis are improved.

The invention relates to a three-dimensional model compression transmission method and system based on dynamic feature perception, and relates to the technical field of three-dimensional live-action modeling, and the method comprises the steps: obtaining the original three-dimensional data of a three-dimensional model, carrying out the multi-modal feature extraction based on a model scene, dividing a key region, a transition region and a non-key region, and generating a three-dimensional model partition map, compressing each region according to the partition map label and a preset compression algorithm to generate a multi-resolution LOD sequence, determining a transmission data hierarchy in combination with a network state and equipment parameters, and finally rendering the model according to the transmission data hierarchy, user behavior information and an environment state to obtain a target three-dimensional model. The technical effects of effectively compressing, transmitting and rendering the three-dimensional model according to various factors such as different area characteristics, network states, equipment parameters and user behaviors of the model and improving the transmission efficiency and the rendering quality are achieved.

The invention relates to a robust unmanned aerial vehicle detection method based on dynamic feature fusion and context attention, and belongs to the technical field of image processing. Aiming at the problems of small target feature loss, semantic gap, background noise interference and the like caused by a fixed convolution kernel scale, one-way feature fusion and a static attention mechanism in an existing unmanned aerial vehicle aerial image target detection method, the method comprises the following steps: constructing a detection model comprising a backbone network, a neck network and a detection head network; a feature rearrangement and extraction module is designed in the backbone network to enhance feature learning, an enhanced double-flow feature fusion pyramid is designed in the neck network to optimize multi-scale feature fusion, and a dynamic multi-scale context attention mechanism is designed in the detection head network to suppress irrelevant background noise. The method effectively improves the accuracy and robustness of small target detection, and achieves a clearer and more stable detection effect in a complex environment.

The invention relates to the technical field of audio and video processing, and provides an audio and video object intelligent tracking optimization method and system combined with deep learning. The method comprises the following steps: performing cross-modal feature collaborative extraction on an audio stream and a video frame sequence by acquiring a synchronous audio and video data group, and generating a multi-modal feature set containing audio time domain dynamic features and video space structure features; inputting the multi-modal feature set into a pre-trained association enhancement network to generate a cross-modal semantic aligned association feature sequence; constructing a tracking stability evaluation model based on the associated feature sequence, and outputting a stability index; tracking parameters are dynamically adjusted according to the stability index, an initial tracking result is calibrated, and an optimized tracking trajectory is output. Therefore, the precision and stability of object tracking in a complex scene are improved by deeply fusing the dual-mode characteristics of the audio and the video, mining the internal association between the modes and combining a dynamic evaluation and calibration mechanism.

The invention belongs to the technical field of data centers, discloses an energy efficiency evaluation method of an artificial intelligence data center, and solves the problem that a static evaluation method is disjointed with the real-time operation state of the data center in the prior art by constructing a dynamic feature library and an AI adaptive weight model. The dynamic feature library collects and associates service scene labels and time-space coupling data in real time, and provides a scene basis for an AI adaptive weight model, so that weight adjustment is converted from passive response index mutation in the prior art to active adaptive service scene change. Nonlinear influences of dynamic factors such as server load fluctuation and environment adjustment on energy efficiency are effectively captured, so that an energy efficiency evaluation result better fits the actual operation state of the data center, energy efficiency factor relevance changes caused by service switching of the AI data center are adapted, scene-based self-adaptive adjustment of weights is achieved through a deep learning model, and energy efficiency evaluation accuracy is improved. The limitation that the fixed weight distribution ignores the difference of different service scenes in the prior art is solved.

The invention relates to the technical field of data analysis, and discloses a data analysis method and device based on multi-modal data fusion, equipment and a medium, and the method comprises the steps: collecting visual data, audio data and process text data, carrying out the preprocessing and standardization processing of different types of data, constructing a multi-modal data set with aligned timestamps, and storing the multi-modal data set in a database; the method comprises the following steps: extracting a time-frequency dynamic feature and a semantic vector feature, extracting a map structure feature, a time-frequency dynamic feature and a semantic vector feature, fusing the features by using a cross-modal attention mechanism to generate a fused feature vector, finally performing analysis processing based on the fused feature vector, and outputting an analysis result. According to the method, the multi-modal data set with consistent time is constructed, the structural features of various modals are extracted, and the cross-modal attention mechanism is introduced to realize deep fusion of the feature level, so that the problems of single information utilization and weak feature relevance of the existing detection means are solved, and the comprehensiveness of defect detection and the accuracy of fault diagnosis are improved.

The invention relates to the technical field of image target detection, and provides a target lightweight detection method and system based on attention feature enhancement, and the method comprises the steps: collecting visible light target image data; constructing a target lightweight detection model: designing a lightweight backbone network and a special attention mechanism; designing an adaptive feature fusion module; designing a detection head; designing a loss function; designing a model lightweight strategy; and training the target lightweight detection model through the training set, inputting the test set into the trained target lightweight detection model after training is completed, and outputting a target lightweight detection result. According to the scheme of the invention, the method focuses on the design of an efficient lightweight network architecture, and through the introduction of a special attention mechanism, dynamic feature fusion and a model compression strategy, the detection precision is ensured, meanwhile, the demand for computing resources is remarkably reduced, and the real-time detection of a visible light image target is realized.

The invention relates to the technical field of face verification, and discloses a face identity verification data processing method based on dynamic feature extraction, which comprises the following steps: acquiring a continuous face video frame sequence and calculating a full-pixel instantaneous velocity vector to generate an original dense optical flow field, selecting rigid region anchor points to calculate a rigid affine transformation matrix and construct a theoretical rigid motion field, performing differential stripping on the theoretical rigid motion field from the original dense optical flow field, and extracting a non-rigid micro-motion residual field; mapping the non-rigid micro-motion residual field to a facial muscle topological grid to generate a time sequence feature tensor, and calculating a geodesic line distance between a covariance matrix of the time sequence feature tensor and a reference dynamic feature in a Riemannian manifold space; when the geodesic distance is smaller than a threshold value, verification is passed, through a rigid-non-rigid orthogonal decomposition mechanism, the special viscoelastic micro-motion and cooperation law of biological soft tissue is captured by utilizing a residual field, and the high-simulation mask is effectively defended.

The invention relates to the technical field of mineral processing visual detection, and discloses a visual analysis system for detecting the grade of a phosphorite flotation froth layer. The system comprises an image acquisition and decomposition module, a parallel feature extraction module, a dynamic feature fusion module, a foam evolution analysis module and a grade decision output module. The system performs multi-scale decomposition on a foam image, extracts physical and semantic features in parallel, constructs a dynamic fusion network based on bidirectional mapping to perform iterative interaction, and generates a multi-modal feature descriptor. Therefore, self-organizing growth of a foam evolution graph is driven, an evolution track of a key foam primitive is positioned and tracked, a foam grade state vector is formed, and finally, a regulation and control decision is output in combination with external control parameters. According to the system, deep fusion of physical and semantic features and deep analysis of the foam dynamic evolution process are achieved, the accuracy and predictability of foam grade state sensing are improved, and an effective means is provided for accurate control over the flotation process.

The invention discloses a video anti-shake method and system based on multi-scale fusion and adaptive smoothing, relates to the technical field of video image processing, and aims to effectively solve the image quality problem caused by shake in a video shooting process. Gradient histograms and wavelet energy distribution characteristics of video frames are extracted through graying and normalization processing, and the gradient histograms and the wavelet energy distribution characteristics are input into a jitter type recognition network to recognize translation, rotation and Z-axis jitter probabilities. And further extracting motion, frequency domain and edge features, and generating multi-modal coupling features through combination of a dynamic feature interaction network and a dot product attention mechanism. And constructing a motion trajectory by using the features, optimizing the trajectory by using a texture perception double-layer smoothing strategy, introducing an adaptive penalty term into a dynamic planning cost function, and outputting a smooth motion compensation parameter. And finally, processing the boundary region through motion compensation and image extrapolation to generate an anti-shake video frame. Through multi-scale feature fusion and a self-adaptive smoothing strategy, the video anti-shake effect is effectively improved, and the method is suitable for complex scenes.

The invention provides a weak supervision video anomaly detection method and system based on potential energy field damping dynamics. The method comprises the following steps: inputting a video into an anomaly detection model to obtain an original video feature sequence; calculating fluctuation potential energy by using the original video feature sequence and further obtaining a global inertia proxy vector; calculating interaction potential energy through the global inertia proxy vector and the original video feature vector; nonlinear self-adaptive damping force is generated through interaction potential energy; applying a nonlinear adaptive damping force to the original video feature vector to obtain a purified dynamic feature vector; constructing a loss function by using the purified dynamic feature vector, and training the model to obtain an optimized model; and inputting the video into the optimized model to obtain a final frame-level anomaly detection result. According to the method, anomaly detection is reconstructed from a classification problem to a signal decoupling and energy dissipation problem in a physical system, and the physical nature of a depolarization mechanism is explained from the theoretical level.

The invention relates to the technical field of computer vision, in particular to an image target detection system and method based on deep learning, and the system comprises a dynamic feature alignment unit, a motion blur compensation unit and a feature fusion control unit. A dynamic feature alignment unit generates spatial deformation parameters through a deformable convolutional layer and an offset prediction sub-network, resamples a shallow high-resolution feature map, and realizes deep and shallow feature space alignment, and a motion blur compensation unit generates a motion vector based on brightness gradient field difference, constructs a mask and weights a suppression blur region, so as to realize deep and shallow feature space alignment. The feature fusion control unit analyzes local entropy and target size distribution, dynamically distributes feature weights and feeds back and optimizes offset parameters, a closed-loop learning loop is formed by the method, and the problems of inaccurate feature alignment, fuzzy interference and poor scene adaptation are solved.

The invention belongs to the technical field of computer vision, and discloses a target identification tracking method based on a self-supervision mechanism, and the method comprises the steps: enhancing a self-supervision pre-training module through causality, constructing a causal sample pair through unlabeled video data, learning universal features through combining with comparison loss, and achieving the high-precision tracking without large-scale manual labeling. A multi-modal feature fusion and dynamic calibration mechanism further reduces dependence on annotated data, is especially suitable for industrial inspection, field monitoring and other scenes where data acquisition is difficult, significantly reduces time and labor costs in a data preparation stage, and broadens the application range of the technology in resource limited scenes; a causal reasoning and physical constraint mechanism is introduced, a dynamic relation between targets is modeled through a space-time causal graph, unreasonable tracks are filtered in combination with a physical rule, and complex conditions such as shielding, rapid movement and extreme weather are effectively dealt with; the dynamic feature calibration module corrects feature drift in real time, and ensures stable model performance in long-term tracking.

The invention discloses a power equipment image-text fusion labeling method and system based on a single-double mixing tower, and the method comprises the steps: carrying out the visual feature extraction to obtain image features, carrying out the text feature extraction to obtain text features, mining the deep features of an image and a text, maintaining the modal specificity, and carrying out the recognition of the image and the text. Processing the acquired image features and text features by adopting a cross attention mechanism to generate a bidirectional attention matrix, calculating a dynamic weight value based on the acquired bidirectional attention matrix, and generating weighted image features and weighted text features based on the dynamic weight value; and performing dynamic feature fusion on the weighted image feature and the weighted text feature to obtain a fusion result feature, and performing end-to-end multi-modal labeling based on the fusion result feature, so that the image and the text can be accurately associated, the labeling efficiency and accuracy are improved, adaptive feature fusion can be realized, the real-time problem of heterogeneous feature fusion is solved, and the real-time performance of the heterogeneous feature fusion is improved. The time consumed by multi-modal alignment is reduced from the minute level of manual intervention to the millisecond level.

The invention relates to an industrial protocol adaptation method, system and device based on dynamic feature recognition and a medium. The method comprises the following steps: acquiring original communication data of industrial equipment; performing feature extraction on the original communication data to obtain a depth feature vector; performing feature matching on the depth feature vector in a known protocol feature library to obtain a protocol matching result; if the protocol matching result is the unknown protocol parameter, marking the depth feature vector as unknown to obtain an unknown feature vector; clustering all the unknown feature vectors based on the protocol matching result to obtain candidate protocol groups; and for each group of candidate protocol groups, generating a new protocol analysis rule set. The method can support parallel operation of multiple communication protocols, is suitable for different manufacturer devices, automatically classifies unknown protocols through a machine learning algorithm, and enhances self-learning and self-adaptive capabilities.

The invention relates to a pressing plate operation behavior analysis and error prevention method and system based on image recognition and machine learning. The method comprises the following steps: acquiring a pressing plate operation area video stream through image acquisition equipment, and segmenting a video into independent operation event segments based on motion detection and trajectory analysis; extracting key point time sequence data of hands of an operator by using a human body posture estimation model, and constructing a dynamic feature sequence fusing a spatial relationship, kinematics and posture semantics; performing multi-level similarity comparison on the dynamic feature sequence and a standard operation template, and judging operation compliance by a machine learning model in combination with a dynamic threshold value; triggering graded early warning and intervention according to a judgment result; and an incremental learning mechanism is adopted, and the template and the threshold value are adaptively optimized based on historical data. According to the invention, accurate and intelligent analysis and active error prevention of the whole operation process of the pressing plate are realized, and the safety level of electric power operation is effectively improved.

The invention discloses an SAR directed target detection method based on multi-scale context sensing, and belongs to the technical field of remote sensing image processing and computer vision. According to the method, firstly, through a scattering characteristic guided multi-scale dynamic characteristic enhancement network, a multi-branch structure and a dynamic fusion mechanism are utilized to enhance characteristic expressions of targets of different scales; and secondly, designing a task-adaptive regional context sensing module, fusing local details and semantic contexts, and improving the target discrimination capability under a complex background. And a decoupling type progressive refining detection head is adopted to predict a target category, a directed bounding box and a direction angle, and the positioning precision is optimized through progressive regression. Semantic consistency loss is introduced in the training stage, and end-to-end optimization is achieved in combination with a multi-task joint strategy. According to the method, the problems of large target scale difference, strong background interference, changeable directions and the like in the SAR image are effectively solved, and the accuracy and robustness of directed target detection are remarkably improved.

The invention discloses a dynamic scene visual SLAM method based on deep learning, and the method comprises the following steps: obtaining an original scene image, and extracting ORB feature points in an image scene; based on the improved target detection network, performing dynamic target detection on the original scene image to obtain a dynamic target frame in the image scene; associating the dynamic target frame and the ORB feature points by adopting a key frame synchronization mechanism, and removing the ORB feature points corresponding to the dynamic target frame; obtaining a feature scene image after primary elimination; performing secondary elimination on unmatched ORB feature points in adjacent feature scene images by adopting epipolar geometric constraint to obtain a static feature scene image; and performing pose estimation and / or map construction based on the ORB feature points corresponding to the static feature scene image. According to the method, the dynamic feature points of the dynamic target are extracted by improving the target detection network, and the dynamic feature points are eliminated twice by adopting the epipolar geometric constraint, so that the positioning precision is improved, and accurate map construction is facilitated.

The invention discloses a gear defect real-time detection method based on dynamic feature fusion and windmill convolution, and relates to the technical field of gear defect detection.The method comprises the following steps that S1, a defect data set of a gear is taken, two data enhancement strategies of Mosai c and Mi xp are adopted for the data set, the problem that original sample defects are few is solved, and the defect detection accuracy is improved; the generalization ability of the model is improved; s2, images of the data set are input into an improved YOLOv12 model, the model adaptively changes the weight of each convolution kernel through a dynamic convolution module, and the features of the images are efficiently extracted; s3, inputting the original features into windmill convolution, so that the model can accurately perform Gaussian distribution on infrared small target pixels, and remarkably enlarges a receptive field while enhancing bottom-layer feature extraction; and S4, processing the gear image by using the improved model, and outputting a gear defect detection result including but not limited to information of the position, the type and the confidence coefficient of the defect through feature fusion of the neck part and multi-scale detection of the head part.

The invention discloses an SLAM dynamic point semantic filtering method based on DDMA-SAM, and belongs to the technical field of synchronous positioning and mapping. A decoupling distillation mechanism is introduced, an image encoder in an original SAM model is subjected to lightweight optimization, a DDMA-SAM semantic network integrating a multi-scale aggregation detection module and an efficient mask decoding module is constructed, and the SLAM dynamic point semantic filtering method based on DDMA-SAM is obtained. And the segmentation performance is improved while the model parameters are greatly compressed. Based on the semantic network, providing a semantic prior and geometric consistency combined-driven double filtering strategy; based on a semantic mask and a confidence threshold, carrying out preliminary dynamic point identification; in combination with the epipolar geometric constraint and the triangulation reprojection error of random sampling consistency estimation, fine elimination of dynamic feature points is realized, and only static points are reserved to participate in camera pose estimation. According to the method, the real-time performance of the system is kept, and meanwhile, the mapping quality and the track stability in a dynamic scene are effectively improved.

The invention relates to the technical field of computer vision, in particular to a text-guided creativity and time consistency-oriented video coloring method and system. By introducing a time deformable attention block, according to the position and shape change dynamic state of an object in a video frame, dynamic features of an instance are captured in the time dimension, so that feature representation consistency is kept, and color flickering and shifting in the coloring process are effectively prevented; semantic representation of a text noun concept is adjusted through a cross-modal pre-fusion module, mask cross attention in the module is used for enhancing understanding of the model on the noun concept and instance perception, and color distribution accuracy is improved; the global structure of the colored video is maintained through gray level video information introduced in the video decoding process; a cross-fragment fusion mechanism is introduced during reasoning, so that the long-term consistency of video coloring is maintained; through text guidance, the user can color the video according to own demands and creativity.

The invention discloses a dynamic scene real-time SLAM system fusing target detection and optical flow. The system comprises an image input module; a target detection module; an adaptive depth estimation module; when the current frame does not receive the mask from the target detection module, the self-adaptive mask compensation tracking module predicts a dynamic area mask of the current frame from the mask of the previous frame based on an optical flow method and a motion model, and when the system judges that a key frame needs to be inserted, the current frame enters the dynamic point filtering module to be processed; if not, directly carrying out local mapping and closed-loop detection; the dynamic point filtering module is used for eliminating dynamic feature points in a dynamic region and retaining static feature points based on antipode constraint and motion consistency analysis of a reverse optical flow; and a pose estimation and mapping module. According to the method, dynamic object interference is effectively identified and eliminated, high positioning precision is ensured to be obtained in a high dynamic scene, the operation speed of the system is greatly improved, and the real-time requirement is met.

The invention discloses a point cloud identification method based on dynamic feature fusion and full-process dynamic parameter adjustment. The method comprises the following steps: step 1, collecting an original point cloud of an industrial part and removing invalid points, constructing a training set in combination with a CAD model point cloud, calculating an average point spacing based on a k-d tree, dynamically adjusting the leaf size of a voxel grid, and generating a standardized point cloud; 2, calculating a point cloud normal vector, extracting a CVFH feature descriptor and an SHOT feature descriptor, dynamically fusing the two types of features based on the average point spacing and the spatial range of the point cloud, and carrying out smoothing processing; step 3, using the fusion features to train a KNN classification model, establishing a mapping relation between the features and target categories, and storing model parameters; 4, after the test point cloud is processed in the step 1 and the step 2, model parameters are input, and a prediction result with the highest confidence coefficient is output and output in a log. According to the method, the problem of low recognition precision caused by insufficient global and local feature capture of a complex industrial part by a single feature descriptor in the prior art is solved.

The invention relates to the technical field of deep learning, in particular to a small sample fine-grained image classification method based on multilayer feature dynamic interactive fusion, which comprises the following steps: inputting a support set and a query set into an image classification model, and outputting a prediction category of a query sample; the image classification model processing steps are as follows: S201, extracting to obtain multi-scale features; s202, obtaining interaction features after interaction optimization of each support sample through a multi-layer feature interaction module; s203, obtaining fusion features of each support sample through a dynamic feature fusion module; s204, projecting the fusion features to a corresponding feature space; s205, calculating category prototype representation of each category in each feature space; s206, calculating the cosine similarity between the query sample and the category prototype representation of each category in each feature space; and S207, determining a prediction category of the query sample based on the cosine similarity. According to the method, efficient cross-layer feature aggregation can be realized, and the capability of capturing fine-grained differences is enhanced through an attention mechanism of spatial perception.

The invention provides a video time sequence modeling and action recognition method and system based on a double-flow structure, and the method comprises the steps: constructing a double-path space-time model, enabling a low-speed path module to be responsible for extracting stable semantic and structural features, enabling a rapid branch feature modeling module to focus on capturing high-frequency time change information, and enabling a high-speed path module to be responsible for extracting stable semantic and structural features; video information is perceived in parallel on different time granularities, deep fusion of spatio-temporal features is realized, the time sequence feature extraction capability of a high-frame-rate video path is effectively improved, the calculation complexity and parameter redundancy brought by traditional three-dimensional convolution are effectively reduced, the modeling precision and real-time performance in a video analysis task are effectively improved, and the real-time performance of the video analysis task is effectively improved. According to the method, a more accurate long-range action recognition capability, a more efficient emergency detection scheme and more flexible deployment adaptability are provided for video time sequence modeling and action recognition, an innovative solution is provided for efficient extraction of video rapid dynamic features, and the method has remarkable theoretical value and wide application prospects.