74 results about "Spatial encoding" patented technology

The application discloses a real scene video deblurring system and method based on a single-step video diffusion model, comprising an encoding module, a denoising module and a decoding module, wherein the encoding module is used for respectively performing latent space encoding on each frame in a blurred video sequence to be recovered, generating a frame-by-frame latent space representation corresponding to the input video frame by frame; the denoising module is used for performing single-step denoising on the frame-by-frame latent space representation to obtain a latent space representation corresponding to a clear video; the decoding module is used for decoding the latent space representation corresponding to the clear video into image frames frame by frame and outputting according to the original time sequence of the input video to obtain a deblurred video result. Through frame-by-frame latent space encoding, frame-by-frame blur differences can be preserved, and through single-step diffusion distillation, reasoning delay can be reduced, and deblurring quality and reasoning efficiency are considered.

An end-to-end polarization hyperspectral image classification method and system, belonging to the field of deep learning and optical imaging technology, solves the technical problems of slow reconstruction speed, limited accuracy, low information utilization, and weak feature extraction capabilities in existing technologies. The method involves capturing images of a target using a snapshot-style spatially coded hyperspectral polarization imaging system. The captured images are then encoded by the system to obtain two-dimensional aliased data. A trained reconstruction network is used to reconstruct the two-dimensional aliased data into a polarization hyperspectral data cube. The classification network is then trained based on the polarization hyperspectral data cube to obtain a trained classification network. The trained reconstruction network and the trained classification network are jointly fine-tuned to obtain a polarization hyperspectral image classification model, which is then used to classify polarization hyperspectral images. This invention achieves high-quality reconstruction and accurate classification of polarization hyperspectral images.

The application discloses a source-load joint probability prediction method and system of a physically constrained graph attention network. The method collects multi-dimensional feature data of source-load nodes in a prediction area to construct an initial node feature matrix. A similarity matrix is generated through differentiable graph structure learning. A dynamic adjacency matrix is generated through normalization and introduction of a sparse mask. Spatial feature aggregation is performed through a multi-head graph attention network to obtain node spatial encoding. The node spatial-temporal hidden state is output through an encoder. The node spatial-temporal hidden state is input into a probability prediction head to output Gaussian distribution parameters of the source-load node power. A joint loss function is constructed. The joint loss function is used for soft constraint training to output a probability prediction result. Posterior projection hard constraint correction is performed in an inference stage to obtain a corrected prediction result. The application solves the problems of lack of physical consistency and inability to quantify uncertainty in the prior art.

The application discloses an intelligent commercial site selection method and device based on big data analysis and a medium, relates to the technical field of big data analysis, and comprises the following steps: collecting a commercial site selection coupling data set, performing feature alignment, and forming a standard space-time data stream; adopting an Apache Flink stream processing engine to perform complex event mode recognition on the standard space-time data stream, and forming a reference point address digital portrait; inputting the reference point address digital portrait and the commercial site selection coupling data set into a deep learning site selection model; performing time series modeling and space encoding through a feature coding layer; performing space clustering analysis and similarity calculation through a similarity measurement layer; and outputting a site selection similarity cloud map. Through the Apache Flink stream processing engine and the deep learning site selection model, the application enhances the fusion precision between multi-source data and significantly improves the dynamic prediction capability of commercial site selection.

The application discloses a remote sensing image binary change detection network based on frequency domain feature interaction. The network is aimed at the pseudo change problem caused by light, season, sensor noise and complex background in high-resolution remote sensing images, and a binary change detection network composed of a hybrid encoder, a frequency domain feature interaction module, a difference feature enhancement module and a lightweight decoder is constructed. The hybrid encoder extracts multi-scale spatial detail features through a spatial encoder and extracts global semantic context through a semantic encoder, so as to give consideration to local boundary texture and advanced semantic information. The frequency domain feature interaction module converts the double-time-phase features into the frequency domain, generates a content-aware filter according to the semantic context, and applies asymmetric frequency domain filtering to the two time phases respectively, so as to suppress the pseudo change caused by the difference in imaging conditions. The difference feature enhancement module further extracts robust change difference features through multi-scale spatial interaction, frequency-aware double-gating and lightweight refinement, and finally outputs a binary change mask by the decoder. The application can improve the change region recognition accuracy, boundary positioning ability and robustness in complex remote sensing scenes.

The application discloses a kind of feedforward event camera three-dimensional reconstruction method and system based on space-time feature aggregation, comprising: obtaining at least two asynchronous event streams, convert each event stream into space-time voxel tensor, space-time voxel tensor includes multiple time boxes, each time box corresponds to the event accumulation in a short time fixed time slice;Space-time voxel tensor is input into time attention encoder, the feature of each spatial position is aggregated in time sequence on different time boxes by self-attention mechanism, obtain space-time feature map rich in time context information;Space-time feature map is input into the space encoder-decoder based on feedforward architecture, and the globally aligned three-dimensional point graph is generated by regression.The method and system can effectively extract the motion information and geometric clues in event data, can accurately predict three-dimensional point graph, and are significantly better than existing event camera methods in depth estimation, camera pose estimation and three-dimensional reconstruction tasks.

The application discloses a blood pressure detection system control method, equipment and storage medium, and belongs to the technical field of blood pressure detection. The method comprises the following steps: performing multi-modal decomposition on a to-be-detected physiological signal to form a space-time input matrix, inputting the space-time input matrix into a space-time fusion model, calculating the attention weights between channels in the space-time input matrix through a space encoder, performing weighted aggregation on channel features based on the attention weights to obtain a feature sequence, identifying long-range time sequence dependence in the feature sequence through a time sequence encoder to obtain a space-time fusion feature vector, and performing regression mapping on the space-time fusion feature vector through a multi-task output layer to output blood pressure detection data. The application adaptively enhances the signal components related to blood pressure through a space-time encoder, and suppresses noise interference. Meanwhile, based on the multi-task output layer, the robustness to posture changes and breathing interference is improved, and the accuracy of blood pressure detection is improved.

This invention relates to a remote sensing interpretation method and system that integrates multi-source spatiotemporal spectral features and a visual model. First, optical imagery, harmonic data, and synthetic aperture radar (SAR) data of the target area are acquired and preprocessed. Next, spatiotemporal features of the optical imagery and harmonic data are extracted separately and fused using a cross-attention mechanism to generate fused features. Then, the fused features and SAR data are subjected to image serialization, temporal encoding, and spatial encoding processing, and spatiotemporal features are extracted using a self-attention mechanism. Finally, the spatiotemporal features are decoded and the multi-source feature convolution results are fused to generate pixel-level land cover classification results. This invention introduces harmonic data to capture the temporal variation trend of ground features, utilizes the visual Transformer self-attention mechanism to capture spatiotemporal interdependencies, and combines multi-source data to capture multi-dimensional features of ground features, avoiding the limitations of a single data source. Furthermore, the fusion of multi-source temporal features allows the model to adapt to different geographical environments, improving generalization performance.

The application discloses a delivery data consistency checking method based on space coding and related equipment, comprising the following steps: extracting attribute information related to a to-be-checked space unit from project contract data, design model data and field measurement data respectively, and uniformly associating each extracted attribute information to the same target level space identifier corresponding to the to-be-checked space unit; performing consistency checking on the to-be-checked space unit based on the attribute information associated with the target level space identifier; in response to deviation in the consistency checking, generating a checking result containing the target level space identifier and a description of the deviation, and triggering a preset disposal process.

This invention discloses an infrared anomaly region detection and segmentation method based on a cascaded architecture, belonging to the fields of infrared image processing and computer vision technology. Addressing the problems of high false detection rate and low segmentation accuracy in existing infrared anomaly region detection methods, this method employs a three-stage cascaded detection-filtering-segmentation architecture: After preprocessing the infrared image, it is input into a target detection model to generate candidate bounding boxes; multi-bounding box fusion technology is used to calculate spatial cue confidence; a secondary confidence assessment and false positive filtering are performed based on the channel attention mechanism and inverse residual structure of a lightweight feature extraction network, generating a filtering score; high-confidence detection boxes are converted into effective spatial codes and used as spatial cue words input into the segmentation model to complete pixel-level fine segmentation. This invention improves positioning accuracy through a multi-bounding box fusion confidence mechanism, reduces the false detection rate through secondary filtering, and achieves high-precision pixel-level segmentation without manual annotation through effective spatial code conversion.

Systems and methods of clustering audio objects. Example systems and devices are described that include a cluster selection module and a speaker drop-out monitoring module. The systems and devices can also include a speaker optimization process that is selectively enabled when the speaker drop-out monitoring module determines that one or more speakers in an object-based audio system are experiencing a drop-out issue. Example methods for clustering audio objects are described that can include the steps of receiving an input audio block that includes a plurality of audio objects, and calculating object-to-speaker gains for the plurality of audio objects. The methods can include identifying a speaker as experiencing a drop-out issue based on the object-to-speaker gains, and clustering the audio objects based on whether the speaker is experiencing a drop-out issue.

This invention provides a method for extracting image watermarks based on a single-step inversion diffusion model, comprising: acquiring a synthetic image, wherein the synthetic image is generated by a VAE decoder based on the latent variables output by multi-step denoising of initial noise according to the diffusion model, and the symbol of the initial noise is a symbol mask obtained by encrypting the watermark information used for tracing through an encryption algorithm; mapping the synthetic image to the latent space where the diffusion model denoising operation is located to obtain latent space coding features; performing a symbol classification task based on the latent space coding features without using continuous initial noise regression to predict the symbol mask prediction value of the initial noise; acquiring the decoding algorithm corresponding to the encryption algorithm; and extracting watermark information from the symbol mask prediction value according to the decoding algorithm.

The application discloses a depression evaluation method based on audio-visual multi-modal data fusion, comprising the following steps: step 1, obtaining the facial video of a subject and the audio of the subject; step 2, obtaining the low-level visual features and low-level audio features of the subject; step 3, inputting the low-level visual features and low-level audio features obtained in step 2 into a parallel multi-scale bridge fusion depression evaluation PMBFN network for processing and obtaining a depression rating; by constructing a spatial coding module of a visual and audio branch, a parallel multi-scale dynamic convolution module and a space-time attention pooling module, multi-scale deep features are quickly and efficiently extracted from audio-visual multi-modal data, dynamic performance of depression behavior is comprehensively captured, and under the adjustment of a multi-modal bridge fusion module, data between modes is fully interacted, the utilization rate of multi-modal data is improved, and therefore the accuracy and efficiency of automatic depression evaluation are improved.

The application discloses an information processing method and device for brain dysfunction auxiliary analysis, equipment and medium, relates to the technical field of data analysis, and comprises the following steps: acquiring initial functional magnetic resonance data, and preprocessing the initial functional magnetic resonance data to obtain target functional magnetic resonance data; generating feature data of each brain region based on the target functional magnetic resonance data, and determining the target importance of each brain region and the target correlation between each brain region based on the feature data; determining the target spatial coding of each brain region based on the comparison result of each target correlation and a preset threshold, and constructing a target graph structure coding based on the target importance, the target correlation and the target spatial coding of each brain region; generating enhanced features based on each target graph structure coding and each feature data, determining a classification result based on the enhanced features, and generating and outputting a brain dysfunction analysis result by using the classification result. The application can improve the accuracy of functional magnetic resonance data classification and meet the medical image analysis demand.

This application discloses a method, device, medium, and product for short-to-medium-term ship motion prediction based on a spatiotemporal fusion Informer model, relating to the field of ship motion prediction. The method includes: acquiring short-to-medium-term ship motion parameters, environmental parameters, and wave spectrum parameters to generate time-series data; performing spatial and positional encoding on the time-series data, and obtaining spatiotemporal fusion features based on the spatial and positional encoding results; constructing a spatiotemporal fusion Informer model by introducing a multi-head ProbSparse attention layer and a Distilling layer; and obtaining ship motion prediction results based on the spatiotemporal fusion features using the spatiotemporal fusion Informer model. This application can improve the efficiency and accuracy of short-to-medium-term ship motion prediction, providing key technical support for intelligent maritime aviation operation systems.

According to embodiments of this disclosure, a method and apparatus for recognizing vehicle turn signals are provided. The method includes processing a target image associated with a vehicle using a spatial encoder in a recognition system to determine a first spatial feature associated with a first turn signal and a second spatial feature associated with a second turn signal; generating a first temporal feature corresponding to the first turn signal and a second temporal feature associated with the second turn signal using a time encoder in the recognition system based on the first spatial feature, the second spatial feature, and time information corresponding to the target image; and determining a first temporal signal for the first turn signal and a second temporal signal for the second turn signal based on the first and second temporal features to determine the vehicle's light status information. Based on this approach, embodiments of this disclosure can more accurately recognize the vehicle's light status information.

The application relates to the technical field of unmanned aerial vehicle data processing, and provides an unmanned aerial vehicle anomaly detection method and system.The unmanned aerial vehicle anomaly detection method comprises the following steps: acquiring a sensor data sequence, constructing a static graph through an MIC correlation coefficient to obtain a node subset; adopting TCN and causal transformation for coding to obtain time coding; constructing a dynamic graph, fusing the static graph and the dynamic graph to obtain space coding; synchronizing the end semantics of the time coding and the space coding to obtain an aligned representation, and obtaining prediction data; extracting real data to obtain an anomaly score; acquiring a training data sequence and a test data sequence to obtain a global threshold; acquiring an adaptive multiplier, extracting a cruise data sequence to obtain an adaptive threshold, and obtaining a decision threshold; and comparing the anomaly score and the decision threshold to obtain an anomaly detection result.Through the above method, the stability and reliability of unmanned aerial vehicle anomaly detection and alarm can be improved.

This application belongs to the field of visualization, specifically disclosing a method and system for visualizing a maritime spatiotemporal knowledge graph. The method includes: dividing the projection plane of Earth's latitude and longitude space into multiple spatial grids and encoding them; dividing it into equal-length time segments in the time dimension and encoding them; merging the spatial and temporal codes to obtain a spatiotemporal grid code; using the spatiotemporal grid code as a hash key to construct a hash index table associating the spatiotemporal grid code with the unique identifier of the entity; rendering the entity as a three-dimensional sphere located within the corresponding three-dimensional cube according to the hash index table, and drawing three-dimensional curves between the three-dimensional spheres of the associated entities according to the triple data; dynamically switching the current spatiotemporal grid and updating the visualization content in response to user interaction commands. Through this application, the spatiotemporal relationships and triple relationships between entities in the graph can be clearly expressed, improving the comprehensiveness of knowledge graph visualization.

This invention provides a multi-scale localization cell modeling method for biomimetic SLAM, belonging to the field of biomimetic localization and mapping technology. The scale parameters of the grid-head-oriented cell network at each scale are determined by the modulus coprime of the remainder system. The location cell firing activity is obtained by jointly decoding the multi-scale GHD cell network through population vector decoding and remainder decoding. During the decoding process, a multi-scale grid cell activity stabilization mechanism is designed to stabilize the firing activity of the multi-scale GHD cell network, obtaining stable GHD cell firing activity. Without significantly increasing the computational complexity of the localization cell model, this invention can significantly improve the spatial encoding capability of the localization cell module, thereby improving the localization and mapping performance of the biomimetic SLAM system in complex environments.

This invention discloses a single-cell spatial transcriptome sequencing method based on spatial coding, belonging to the field of spatial transcriptome sequencing technology. Two microfluidic chips with mutually perpendicular channels are used to add two sets of spatial coding sequences to tissue slices. Unique spatial coding combinations are formed at the intersection of the two channels, marking the spatial location information of cells. After coding, cell nuclei are extracted from the tissue slices, and high-throughput single-cell nuclear sequencing is performed to add cellular coding to single cells. Through sequencing analysis, transcripts are mapped to spatial locations according to spatial coding, and to single cells according to cellular coding, reconstructing a single-cell spatial transcriptome atlas. This method achieves single-cell spatial coding in a simple and low-cost manner, belonging to spatial transcriptome sequencing technology with single-cell resolution.

The application discloses a neurosurgery operation scene intelligent understanding method, which is realized through a trained neurosurgery operation scene intelligent understanding network; the network comprises a time coding layer and a space coding layer, and the output end of the space coding layer is connected with at least one fuzzy attention module; the time feature output by the time coding layer and the fuzzy space feature output by the fuzzy attention module sequentially pass through the cross-modal fusion operation of a decoding layer and the dimension reduction processing of a feedforward neural network, and output an operation stage recognition prediction result; the fuzzy attention module evaluates and quantifies the reliable degree of the space visual feature at each position in the feature sequence output by the space coding layer, dynamically adjusts the attention weight and gives the space visual feature to form a fuzzy space feature; the method effectively improves the recall rate and Jaccard coefficient while maintaining high accuracy and high precision, and proves that the method has the effect of significantly optimizing the fuzzy recognition and prediction accuracy.

This invention relates to the field of pedestrian flow prediction technology and discloses a method for predicting urban pedestrian flow based on a spatiotemporal dataset of urban points of interest (POIs). The method includes: acquiring the spatiotemporal dataset of urban POIs; dynamically fusing multimodal features using a hierarchical perceptual attention fusion mechanism to obtain POI feature vectors; performing spherical spatial dependency modeling using a region-adaptive spherical convolutional coding method to obtain spatial coding vectors; performing business hour constraint coding using a periodic mask generator to obtain temporal feature vectors; extracting spatiotemporal coupling features based on the POI feature vectors, spatial coding vectors, and temporal feature vectors; performing region-aware temporal series modeling on the spatiotemporal coupling features; and generating predicted pedestrian flow values ​​and congestion level probability distributions through multi-task shared prediction. This invention improves the performance of urban pedestrian flow prediction by employing a method based on a spatiotemporal dataset of urban POIs.

This invention relates to the field of visual inspection technology and discloses an online labeling quality inspection system and method, including a spatial coding light projection module, an image acquisition module, a light intensity pre-regulation unit, and a signal processing module. The light intensity pre-regulation unit determines the conjugate modulation site of the specular reflection region based on the imaging and projection optical center and the transmission trajectory of the tested container. By superimposing an asymmetric spatial energy flow attenuation field at this site, the incident brightness is pre-controlled, so that the charge accumulation of the reflected image is maintained within the full-well charge capacity limit. The signal processing module analyzes the phase information and reconstructs the three-dimensional morphological features of the labeling region to establish evaluation indicators. This invention suppresses high-light interference before interaction, eliminates cross-frame adjustment lag, realizes feature extraction and three-dimensional reconstruction of labeling defects against a strongly reflective curved surface background, and improves the detection accuracy under high-speed flow conditions.

This invention relates to the field of optical inspection technology for sheet metal surfaces, specifically to an optical-based method and system for detecting the flatness of sheet metal surfaces. The method includes: projecting an optical pattern with a specific spatial coding structure onto the surface of the sheet metal to be tested, and acquiring the original deformed image by obtaining a surface-modulated deformed pattern. The image is preprocessed to eliminate environmental interference and extract the effective area. Based on the spatial coding structure, the effective area image is decoded to reconstruct a two-dimensional height displacement field of the sheet metal surface relative to a reference plane. Based on this height displacement field, the local curvature distribution of the surface at multiple scales is calculated and the global undulation trend is analyzed. By comparing with a preset flatness threshold, various defect areas are accurately identified. The defect areas are classified and parameterized to generate a detailed inspection report containing information on defect type, location, size, and depth. This invention improves the accuracy, anti-interference capability, and interpretability of defect analysis in flatness detection.

This invention discloses a method and apparatus for predicting interactive dynamic states based on mesh representation and atomic skill graphs, belonging to the field of robotics technology. It solves the problem of the disconnect between abstract action commands and underlying physical states in machine-object interactions. The method includes: extracting key interactive state nodes during the manipulation of a target object by an end effector; acquiring the environmental visual image, distributed time-of-flight (ToF), and ontological state information of the end effector at each state node; fusing visual and ToF information to reconstruct the topological mesh of the target object, and jointly constructing an atomic skill graph encompassing physical structure and contact features with the ontological state, as the model input; importing the model input into a graph neural network for latent space encoding and multi-round message aggregation and propagation, and outputting the dynamic state prediction result for the next moment through graph decoding. This invention achieves a unified mapping between abstract skills and physical states, improving the accuracy of interactive state prediction in the face of high-dimensional deformation and complex multi-point contact.

This invention provides a cross-domain flood prediction system based on spatial coding and coupled bidirectional LSTM, belonging to the field of flood prediction technology. The invention constructs a three-layer collaborative architecture of sensing nodes, edge gateways, and a cloud engine. Sensing nodes collect and preprocess multimodal hydrological data; the edge gateway performs spatial interpolation on discrete sensor data to generate a watershed raster, extracts multi-scale spatial features through spatial pyramid pooling, and reduces the dimensionality to a spatially encoded vector; the cloud engine initializes this vector as the cell state of a forward long short-term memory network, and injects the forward hidden state into the gating units of the reverse long short-term memory network in real time through a coupling matrix, achieving synergy between forward and reverse information, and finally outputting the predicted future runoff value. This invention can preserve the spatial heterogeneity of the watershed, fully utilize future meteorological forecast information, and significantly improve the accuracy of cross-regional forecasts and the system's disaster resilience.

This invention discloses a method and system for structured coding of safe space based on LiDAR point clouds, relating to the field of environmental perception technology. The method includes: acquiring local LiDAR obstacle point cloud data of the robot at the current moment, and preprocessing the local LiDAR obstacle point cloud data to obtain an obstacle sphere; based on the obtained obstacle sphere, using ellipsoid-based convex decomposition to generate polyhedral constraints, obtaining convex domain intersections, and constructing a local safe space based on the convex domain intersection representation; and performing structured coding on the convex domain intersections through a shared mapping network to generate a fixed-dimensional safe space code. This invention uses LiDAR point clouds as input, first performing convex decomposition on the local point cloud to construct a convex polyhedron composed of an unordered, variable-length set of constraint surfaces, and then performing symmetric aggregation of the constraint set through a shared mapping network to generate a fixed-dimensional safe space structured code for subsequent path planning, obstacle avoidance decision-making, and other modules.

The application discloses a vehicle trajectory prediction method based on multi-vehicle space-time interaction relationship. First, according to the highway driving data set, the speed relationship graph, the acceleration relationship graph and the driving intention relationship graph of the target vehicle and the surrounding vehicles are generated, and then a new environment feature graph aggregating neighbor features is generated through SuperGATConv fusion; secondly, the new environment feature graph is input into an environment intention encoder constructed by a dilated convolution network to generate a spatial coding feature; the vehicle data information of the target vehicle to be predicted is input into an LSTM as an input feature to obtain a time coding feature. Then, the spatial coding feature and the time coding feature are spliced and input into a space-time decoder to obtain a future state feature. Finally, the future state feature is input into a prediction module constructed by an LSTM and an MPL to generate a predicted future trajectory. The application obtains rich interaction information between vehicles, ensures that local spatial information is not lost, and improves the prediction performance of the vehicle trajectory.

A high-throughput parallel testing method based on imaging flow cytometry and spatially coded reagents combines the high-throughput characteristic of an imaging flow cytometer and the high coding capacity characteristic of spatially coded reagents, thereby achieving high-throughput parallel testing of nucleic acid sequencing, protein testing and single cell analysis, and being capable of being used for liquid biopsy or other biomedical use requirements for performing high-throughput synchronous testing on different biomarkers.