280 results about "Temporal information" patented technology

The embodiment of the invention discloses a safety monitoring method and system for building construction, and the method comprises the steps: collecting the original video data of a construction site, carrying out the denoising, illumination correction and frame rate adjustment, and outputting a standardized video data stream; extracting attitude features of the constructors and representing the attitude features as a feature matrix to form an attitude feature matrix set containing spatio-temporal information; the sensitivity and correlation of the feature matrix are analyzed, calibration noise is added after dimension reduction, and feature data conforming to differential privacy are generated; and extracting data advanced representation by utilizing a pre-training model, and completing behavior classification, dangerous area judgment and safety violation detection. And the violation risk is evaluated in combination with the risk level of the construction area, graded early warning is generated, and meanwhile violation information is recorded to form a traceable management mechanism. According to the embodiment of the invention, accurate and efficient safety violation behavior detection is realized, and intelligent technical support is provided for safety management of building construction.

An apparatus is configured to classify indicator lights of surrounding vehicles as either active or inactive. The apparatus may use a Siamese network to determine respective feature vectors from respective images captured by a vehicle at different times. The apparatus may also embed speed and time information in the respective feature vectors based on odometry information associated with the vehicle at a time each respective image was captured, and fuse, using a temporal attention mechanism, features and the speed and time information from the respective feature vectors to produce a fused feature vector. The apparatus may further process the fused feature vector using capsule modules to produce an indicator feature vector, calculate a similarity metric from the indicator feature vector, and process the similarity metric with a classifier to output an indicator classification.

The invention provides a long video understanding method capable of relieving time sequence illusion in a video language large model. The long video understanding method is based on a static bias adaptive frame selection mechanism and a cross-modal feature fusion strategy. According to the static bias mechanism, inter-frame similarity is evaluated through a discriminator, redundant frames are identified, key frames are selected or a complete sequence is reserved, so that calculation overhead is reduced, and spatio-temporal information integrity is kept; a video frame and a text are mapped to a shared semantic space, the single-frame semantic understanding ability is enhanced, then an embedded sequence serves as a soft prompt to be input into a large language model, and a final answer is generated in an autoregression mode. According to the method, the efficiency and accuracy of long video understanding and video question and answer tasks can be remarkably improved; the problem of low training and reasoning efficiency caused by time sequence dependence redundancy and excessive computing resource consumption is effectively relieved; and through a dynamic multi-modal task processing framework and a space-time memory bank compression mechanism, the modeling capability and generalization performance of the model on a long video sequence are further improved.

The invention relates to the technical field of intelligent monitoring of protective forests, and discloses a protective forest monitoring and evaluating system based on intelligent visual identification. According to the system, synchronous perception information of vision, terrain and weather is captured through a multi-source information synchronous acquisition module, and a vegetation growth space field containing canopy form topology and canopy density change process is generated through multi-dimensional feature extraction operation. A degeneration characteristic evolution graph constructed according to the method can present a degeneration plaque contour, a spreading track and an intensity distribution thermodynamic diagram. The map is subjected to deep feature analysis through a trained multi-level degradation identification network to obtain a degradation level and a trend vector, and finally a comprehensive evaluation report with spatio-temporal information is formed. According to the invention, high-precision and automatic monitoring and evaluation of the dynamic degradation process of the protection forest are realized.

The invention discloses a ground wire full life cycle management method, relates to the technical field of power equipment asset management, and aims to solve the problems that manual recording is easy to tamper, cloud diagnosis delay is high and an alarm mode is single and easy to neglect in traditional ground wire management. According to the method, operation fingerprints and spatio-temporal information during grounding wire hooking are recorded through the block chain technology, and it is ensured that data cannot be tampered; a built-in lightweight AI model of the edge computing unit is used for analyzing sensor data in real time to carry out localized fault diagnosis; and triggering a multi-mode sound-light alarm controlled by the PWM signal according to the diagnosis result grade. The system realizes full-life-cycle credible traceability, sub-second fault response and high-recognition-rate alarm of the state of the grounding wire, and is suitable for intelligent operation and maintenance of the grounding wire in the field of a transformer substation and the like.

The invention relates to the technical field of motion detection alarm, in particular to a multi-agent fusion perception area intrusion abnormity alarm method and system, and the method comprises the following steps: collecting a target coordinate and a timestamp, integrating and sorting tracks, analyzing the direction and time change, judging the stability, recognizing an induction behavior, and screening and verifying the abnormity. According to the method, through integration of the spatial position and time information of the multi-source sensing intelligent agent, the continuous track of the target is reconstructed, the integrity identification of the motion behavior is enhanced, and the track direction change and time interval linkage analysis is carried out, so that the sensing depth of the behavior trend is improved; multi-dimensional recognition of abnormal behaviors is achieved through track stability and regional stay state linkage judgment, continuous conditions are introduced into abnormal verification to filter accidental interference, alarm output is matched with a regional strategy based on a verification result, the accuracy and robustness of intrusion detection are effectively improved, and the practical requirement in a high-dynamic scene is met.

The invention belongs to the technical field of computer vision and robot collaborative perception, and relates to a 3D human body posture estimation method and system based on space time sequence information fusion. The method comprises the steps that a multi-view image sequence is collected, and 2D joint coordinates of all view angles are extracted from the multi-view image sequence through a posture detector; position embedding, global embedding and edge embedding are carried out based on the 2D joint coordinates, and spatial features, including position information, global association information and skeleton edge information, of each joint point are obtained; fusing the features of each visual angle in the spatial features through a cross-visual-angle attention mechanism to obtain global feature representation; performing spatio-temporal feature enhancement and time sequence mixing on the global feature representation to obtain features integrating time sequence information, space information and channel information; and performing 3D posture regression on the features of the integrated time sequence information, the space information and the channel information to obtain a 3D human body posture. According to the invention, accurate and robust multi-view 3D human body posture estimation can be realized.

The invention discloses a video key frame extraction method fused with self-supervised deep learning, and the method comprises the following steps: carrying out the standardized sampling of an input video according to a fixed interval, extracting a SuperPoint local key point and a Video MAE global semantic feature, and generating a dense descriptor and a semantic vector; constructing multi-dimensional change indexes such as local matching degree and global similarity; constructing a soft distribution matrix and a matching point set based on the fusion features; dynamically judging the key frame through a self-adaptive multi-threshold rule; and outputting the key frame set. The method fuses local and global spatio-temporal information, has robust feature extraction and key frame discrimination capabilities under a weak supervision condition, and can effectively improve the efficiency and precision of video compression, abstract and event detection.

An electronic device may include a communication circuit, a memory, and at least one processor. The memory may store instructions that cause the electronic device to execute an application, determine a property of the application, receive a plurality of data packets for the application from an external device through the communication circuit, determine whether at least one of data throughput information on the external device, state information about a channel transmitting the plurality of data packets, or signal round trip time information satisfies a specified condition, apply artificial intelligence learning-based algorithms to merge and process the plurality of data packets to the plurality of data packets, based on the specified condition being satisfied, and refrain from applying the artificial intelligence learning-based algorithms to the plurality of data packets, based on the specified condition not being satisfied.

A system and method may be provided for real-time RF fingerprinting, that includes obtaining residual data from each transmission via residual data preprocessing, processing at least a portion of the residual data using a photonic hardware-compatible continuous-time recurrent neural network (CTRNN) model to correlate temporal information and generate informative features, classifying the residual data based on the informative features using a convolutional neural network (CNN) model, and outputting a prediction of which of the plurality of devices at least one of the plurality of adjacent data units was transmitted from. The CNN model may be configured to convolute the generated informative features using at least one convolution-1D layer, for each of which it may select a maximum value for every two consecutive sequential points output from the convolution-1D layer using a max-pooling layer, and flatten and fully connect the output from a last max-pooling layer using a fully-connected layer.

The invention discloses a heating monitoring method for a stainless steel seamless tube in a heat treatment furnace, which belongs to the technical field of metal heat treatment and specifically comprises the following steps: acquiring an electromagnetic induction signal of a steel tube in rotation advancing in real time and binding spatio-temporal information; kinematics inversion is carried out on the spatio-temporal information based on the movement speed of the steel pipe, and an electromagnetic parameter evolution track of each physical point on the surface is reconstructed; continuously deviated abnormal points are identified based on the evolution trajectory and clustered to extract electromagnetic characteristics of the heating abnormal area, temperature deviation is obtained through a preset electromagnetic-temperature mapping model, and a compensation instruction is generated; and predicting space-time coordinates of the abnormal area reaching the next heating area, and driving the corresponding independent heating unit to execute dynamic energy compensation in the matched time window. Real-time, global and accurate sensing and closed-loop regulation and control of the heating uniformity of the moving workpiece are achieved, the uniformity and stability of heat treatment quality are remarkably improved, and energy consumption is reduced.

The invention provides an electric vehicle charging demand prediction method and system and electronic equipment, and relates to the technical field of deep learning prediction. Comprising the following steps: acquiring historical observation data and corresponding time characteristic data of an electric vehicle charging station; an MIFM prediction model is constructed, an initial input information coding module receives initial data and carries out parallel embedding processing, and unified input representation containing node specific adaptive embedding is constructed; the spatial-temporal information dynamic fusion module obtains the unified input representation, captures explicit geographic space dependence and implicit function correlation in parallel through a two-channel dynamic graph learning mechanism, dynamically integrates multi-layer spatial-temporal characteristics by using an adaptive Kalman filtering fusion strategy, and generates an enhanced fusion signal; and the space-time dependency modeling and prediction module learns the spatial relationship representation based on the enhanced fusion signal by using a stacked graph attention network, captures time dynamics through a time sequence decoding module, and generates a charging demand prediction result of a future time step.

The invention relates to the technical field of underground pipeline detection, in particular to an underground nonmetal pipeline detection positioning method and system. The method comprises the steps that multi-source heterogeneous data such as road surface vibration, echo imaging, a thermal radiation sequence and environment field intensity in a target area are acquired and preprocessed, and an original observation set containing spatio-temporal information is constructed; time-frequency joint representation of vibration and echo is realized based on a self-supervised multi-mode feature coding network, thermal radiation texture anomaly is extracted by a space-time convolution converter, and domain-invariant high-dimensional features are formed; conformal attention cross-modal alignment is adopted, the features are mapped to a unified hidden space, key boundaries are screened through gating mutual information, and a comprehensive feature map containing dielectric constant gradient and elastic wave guiding characteristics is generated through fusion; and carrying out topological reasoning by combining a graph neural network with a as-built graph and facility position priori, generating a candidate pipeline network, checking pruning by ray tracing arrival time consistency and a thermal diffusion time delay rule, and outputting pipeline space distribution and three-dimensional coordinates.

The invention relates to the technical field of artificial intelligence, and provides an audio marking method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring an audio to be processed; determining that there is a multi-person dialogue in the audio to be processed; determining a starting time point of a conversation timeline associated with the to-be-processed audio according to the starting time information of the multi-person conversation; starting from the starting time point, extracting a line segment carrying first starting and ending time information from the to-be-processed audio along the dialogue timeline; and understanding and labeling the audio to be processed by taking the line segments as units. According to the audio annotation method and device, the electronic equipment and the storage medium provided by the invention, audio understanding annotation can be efficiently and accurately performed on long-duration audio involving complex contexts and multi-role interaction scenes.

The invention discloses an electric power intra-day price prediction method based on dynamic holiday weight and multi-source fusion, and the method comprises the steps: obtaining multi-source historical data, and carrying out the time synchronization processing; holiday and festival time information is acquired, and a dynamic weight is generated based on the influence of holidays and festivals and upstream and downstream dates on the power load and the electricity price; constructing a multi-dimensional predictive factor matrix based on the multi-source historical data after time synchronization processing; and according to the multi-dimensional predictive factor matrix, on the basis of a collaborative optimization multi-model combination comprising a feedforward neural network model and a bagged regression tree model, intra-day joint prediction is executed, the intra-day joint prediction refers to a process of predicting the power load and the electricity price hourly, and an hourly prediction result of the power load and the electricity price is output. A dynamic holiday weight mechanism is introduced, a multi-source fused high-dimensional predictive factor matrix is constructed, and a multi-model combined predictive strategy of collaborative optimization of a feedforward neural network and a bagged regression tree is adopted, so that the precision and stability of intra-day electricity price and load prediction of the electricity market are improved.

A computing platform may train, using code specific information, temporal information, and compute specific information, a CO2 minimization model to output, for a given processing request, a resource allocation recommendation. The computing platform may input a processing request into the CO2 minimization model to output a resource allocation recommendation indicating processing resources for use in executing the processing request with a minimal amount of CO2 emissions. The computing platform may cause the task execution platform to execute the processing request using the processing resources of the resource allocation recommendation. The computing platform may: dynamically monitor the task execution platform to collect, in real time, metadata associated with execution of the processing request, generate, using the CO2 minimization model, updated resource allocation recommendations based on the real time metadata, and cause the task execution platform to shift the processing for the processing request to the updated resources.

An elevator health assessment method based on a time sequence knowledge graph and deep learning belongs to the field of elevator system health management, and comprises the following steps: firstly, constructing and dynamically updating an elevator fault time sequence knowledge graph, collecting structured operation data and unstructured maintenance texts of an elevator, preprocessing, respectively extracting entities and relationships, and calculating the elevator fault time sequence knowledge graph; constructing a tetrad time sequence knowledge graph fused with time information in the graph database; secondly, based on a time sequence knowledge graph, precise evaluation of the elevator health state is achieved, a time sequence DistMult method is adopted for carrying out embedded representation on the graph, and a vector sequence fused with time-space information is generated; time sequence dependence features are extracted through a bidirectional gating circulation unit, and key degradation information and feature dimensions are dynamically focused in combination with a time step attention and channel attention mechanism; and finally, designing a residual multi-layer perceptron classifier to perform health level classification on the fusion features, introducing an increment fine tuning mechanism, and outputting probability evaluation results of each health level of the elevator. The accuracy is improved.

The invention discloses a road surface and inside crack disease space-time alignment joint detection method for pavement health management, and aims to solve the problems of surface and inside disease separation, space-time data misalignment and high microcrack omission ratio in traditional detection. The method comprises the following steps: driving double digital speedometers to synchronously rotate through a wheel linkage mechanical structure, realizing time-space synchronous acquisition of road surface images and radar signals, and recording accurate time-space information in combination with a GNSS and a clock synchronization module; a YOLO deep learning model is adopted to identify a road surface crack, and the small target detection precision is improved; through data preprocessing, coordinate conversion and space matching, space-time alignment of data in the table is achieved, and the crack direction is judged. According to the method, the detection frame rate is larger than or equal to 20 FPS, the small target precision is improved by more than 30%, the space-time alignment precision is smaller than or equal to 0.1 m, new disease categories are supported, and efficient and accurate technical support is provided for road health management.

The invention relates to the technical field of unmanned aerial vehicle resource management, and discloses a heterogeneous task-oriented multifunctional unmanned aerial vehicle resource adaptive allocation method. The method comprises the following steps: collecting real-time multi-source demand data of a heterogeneous task, performing time synchronization and noise filtering, and generating a clean data stream; performing multi-scale feature extraction on the data stream, and fusing spatio-temporal information to form unified feature representation; inputting the features into a probabilistic reasoning unit, performing uncertainty quantization and abnormal mode recognition, and outputting a resource demand feature vector with confidence; constructing a dynamic resource load topology based on the vector, and predicting a load migration trend; according to the trend, a learning mechanism is adopted to generate a self-adaptive resource allocation strategy, and parameters are adjusted through real-time feedback; and finally, real-time task requirements are matched with the strategies, and allocation is executed. According to the method, the accuracy and foresight of resource allocation and the overall robustness of the system are improved by quantifying demand uncertainty and predictive load scheduling.

The invention discloses a geospatial data geometric topological relation evaluation method and system based on a large language model, and belongs to the technical field of spatio-temporal information processing, and the method comprises the following steps: S1, data preprocessing and model construction: processing a training data set containing geospatial data and a corresponding geometric topological relation thereof, and constructing a model; the geometric topological relation of the geographic space data is coded into text or vector representation which can be understood by a large language model; the coded training data is utilized to finely adjust a pre-trained large language model, so that the pre-trained large language model can evaluate a topological relation according to input geographic space data description; s2, geographic space data input and feature extraction; and S3, large language model reasoning and topological relation evaluation. According to the method, by combining the language understanding ability of the large model and the spatial topology calculation technology, intelligent and high-precision evaluation of the geographic spatial data geometrical relationship is achieved, and a brand-new and self-adaptive evaluation technology is provided for the field of geographic spatial data processing.

The invention discloses a smart power grid fault detection method and device based on a dynamic graph neural network. Firstly, multi-source electrical measurement data are collected in real time, and preprocessing and spatial-temporal feature extraction are performed on the data; then, constructing a dynamic graph structure taking the power grid equipment as nodes and the electrical connection relation as edges, and establishing a dynamic graph neural network model; and fusing the spatial topological features and the time sequence features of the power grid data in a feature fusion layer of the dynamic graph neural network model to obtain feature representation containing spatio-temporal information. And outputting a node-level fault probability prediction result based on the fusion features, and carrying out online incremental learning updating on dynamic graph neural network model parameters according to real-time data streams. According to the method, the natural advantages of the dynamic graph neural network are utilized, and the inherent spatial dependency relationship of the power distribution network and the dynamic time rule of the measurement data are effectively modeled, so that the fault detection accuracy is remarkably improved, and the false alarm rate and the missing report rate are reduced.

The invention relates to the field of automatic driving, and particularly discloses a lane line detection method based on time sequence curvature and multi-scale context. The method comprises the following specific implementation steps of: giving a data set image and preprocessing the data set image; lane features of an image are extracted through a backbone network, a multi-scale feature map is generated, and the multi-scale feature map is continuously processed by a double-branch detection framework. According to the framework, two parallel branches of time sequence modeling and attention enhancement are fused, firstly, a feature map is sent into a multi-scale attention branch, multi-scale channels and spatial features are fused, and semantic information is enriched; and meanwhile, the time sequence optimization branch captures prior knowledge by using time sequence information, extracts context information and models curvature change. And finally, carrying out weighted fusion on the output features of the two branches and the original trunk features, processing and refining lane prediction through a loss function, and outputting a final result. According to the method, the problems that visual clues of lane lines are lacked in a complex scene and a long-distance dependency relationship is difficult to model in a curve scene are effectively solved, and the robustness and accuracy of the model are enhanced.

Embodiments of the present disclosure provide a data alignment method, a differential protector and a differential protection system. The data alignment method comprises: at a first time node, obtaining first sampled current data from a first sampling device; at the first time node, receiving a second message from a second differential protector, the second message comprising second sampled current data and a sampling time mark thereof, first time information about a receiving time difference of the second differential protector from receiving a first message to the second time node, and second time information about a second sending processing delay of the second differential protector from the second time node to sending the second message, the second time node being a time point at which the second sampling device obtains the second sampled current data; when time synchronization is maintained, calculating and storing a time calculation deviation between a third time node and a first calculation value of the second time node; when time synchronization is lost, determining the third time node according to the stored time calculation deviation.

This invention provides a method and apparatus for correcting sea surface temperature (SST) predictions based on spatiotemporal axial attention. The method includes inputting target SST data into a target SST prediction correction model, comprising a convolutional input layer, an encoder, and a decoder. The convolutional input layer performs feature extraction, temporal encoding, and positional encoding on the target SST data to obtain a target feature vector. The encoder performs attention calculations on the target feature vector in three dimensions to obtain a first target output vector. The decoder outputs the target SST prediction result based on the first target output vector and historical prediction values ​​output by the decoder. Temporal and positional encoding are performed during the model input stage to enhance the representation of temporal and positional information contained in the original data. By performing attention calculations on the target feature vector in the spatiotemporal, longitude, and latitude dimensions respectively by the encoder, effective fusion of features in different dimensions is achieved, further improving feature representation capabilities and increasing the accuracy of SST prediction.

This invention relates to the field of nursing technology for disabled individuals, and particularly to a method and device for identifying the risk of aspiration in disabled individuals. The method includes: establishing a multimodal, non-invasive data acquisition system to simultaneously collect temporal information on laryngeal vibration, respiratory status, and facial movements; preprocessing to extract effective signals; constructing an individual-specific physiological baseline model using historical normal physiological samples and binding it to the individual's identity; segmenting and processing real-time signals to construct a three-dimensional feature collaboration matrix, calculating mutual information entropy, and dynamically iterating the associated baseline; temporally aligning swallowing and respiratory abnormality anchor points, marking key intervals, and screening suspected risk events; and outputting the final aspiration risk identification result after environmental interference filtering, feature weight adaptation, and verification. This invention enables accurate, real-time identification and early warning of aspiration risk without requiring any wearable devices and without affecting the daily care and rest of disabled individuals.

This invention belongs to the field of video processing and is a generative video encoding and decoding method, comprising the following steps: S1: In a video frame, an inter-frame is generated based on the temporal structure and content complexity of the video frame. The inter-frame is set as a recognition frame, and key information of the recognition frame is extracted. The key information includes the temporal information of the recognition frame and the correlation information between the preceding and following frames; S2: According to the information of the recognition frame, the preceding and following frame images of the recognition frame are extracted, and after feature processing, an information code carrying the feature information of the preceding and following frame images is generated by a hash algorithm. The information code is fused with the feature-processed preceding and following frame images to obtain a recognition image; S3: The recognition image is inserted into the original video frame sequence corresponding to the temporal position of the original recognition frame, and the video frame sequence after inserting the recognition image is re-encoded to generate a new video.

This invention relates to the field of brain function image classification and processing technology, specifically to an adaptive high-order fusion learning method for brain function image classification, comprising the following steps: S1, acquiring the brain function image to be classified and performing image data preprocessing; S2, constructing a high-order functional brain network matrix sequence; S3, adaptively learning the weights of different-order functional brain networks, weighted summing them to further output features; S4, capturing the contextual dependencies of the high-order functional brain network matrix sequence; S5, further extracting features and performing weighted fusion of features; S6, predicting and outputting the classification result; S7, parameter tuning. This invention, through end-to-end adaptive high-order functional brain network fusion learning, utilizes adaptive learning of different-order functional brain network weights to enhance the interpretability of the inference process between input features and prediction results, and captures temporal information before and after the generation of the high-order functional brain network matrix through a self-attention mechanism, accurately and efficiently classifying and identifying images of normal brain function and brain disorders.

The invention belongs to the technical field of informatization prevention and control of diseases and insect pests, discloses a multi-element crop disease and insect pest monitoring and early warning method based on big data, and aims to solve the problems that traditional monitoring and early warning are one-sided and low in efficiency. The method comprises the following steps: firstly, constructing a sky-ground integrated network, and collecting satellite remote sensing, unmanned aerial vehicle images, internet-of-things perception, weather forecast and agricultural knowledge graph data; constructing a spatio-temporal feature vector through preprocessing and feature engineering; then, through a GCN-LSTM hybrid model, fusing spatio-temporal information to predict a future disease and pest occurrence probability; based on a prediction result, starting four-level early warning and generating a precise prevention and treatment prescription; and finally, model closed-loop optimization is realized through prevention and treatment effect evaluation. According to the invention, three-dimensional monitoring and accurate prediction are realized, the early warning time is advanced, the pesticide usage amount is reduced, the crop yield is guaranteed, and agriculture is promoted to be converted into data drive.