151 results about "Transform coding" patented technology

The invention discloses a super-resolution image reconstruction method, system and device based on dynamic frequency domain adaptive coding and contrast constraint optimization, and a medium. The method comprises the following steps: creating an original training data set; image data features and application scenes are considered, and feature extraction is performed on image data by adopting a modular design method; in combination with discrete cosine transform and inverse transform thereof, adaptive transform coding is carried out on different frequency components in the image features through a dynamic weight adjustment mechanism; the U-Net and Transform architectures are combined, isotropic design is optimized, a diffusion model and a feature fusion and reconstruction technology are applied, and diffusion process reconstruction is achieved; by introducing a contrast loss function, contrast learning is carried out between images with different resolutions, and the feature learning ability is enhanced. The image reconstruction method is more intelligent and efficient, and can effectively deal with frequency feature recovery and detail reconstruction problems in a complex image scene.

The invention discloses a traffic large model construction and decision-making method and device based on multi-modal two-way map reasoning, and the method comprises the steps: constructing a multi-modal data set of a text, an image and a track, generating fusion features through spatial-temporal clustering and cross-modal Transform coding, carrying out the two-way map reasoning in combination with a traffic knowledge map, and carrying out the decision-making of the traffic large model. The method comprises the following steps: generating an embedded representation through a forward graph neural network, reversely mapping a decision scheme generated by a language model to a graph to verify consistency, outputting knowledge to enhance embedding, fusing multi-modal features and knowledge embedding by adopting an LoRA multi-task joint fine tuning technology, adapting to traffic field tasks, deploying a real-time inference engine, and carrying out real-time inference on the traffic field. And processing the dynamic data flow through an aging perception attention mechanism, and outputting traffic event identification, path planning and scene question and answer results in parallel. Compared with the prior art, the method has the advantages that the problems of insufficient multi-source heterogeneous data fusion, low knowledge utilization efficiency and poor real-time decision consistency can be solved, and the semantic understanding and decision accuracy of the traffic large model is effectively improved.

The invention discloses a mechanical arm path planning method based on Transform and a diffusion model, and belongs to the technical field of robots, and the method comprises the following steps: collecting environment information to generate a reference path, and constructing a training data set; a conditional diffusion Transform prediction network is constructed, features are extracted, and multi-modal path prediction is realized; gaussian noise is applied to the reference path, and denoising training is carried out on the conditional diffusion Transform prediction network; in combination with a diffusion model and a cost guidance mechanism, optimizing a noise path sampled from Gaussian distribution until a smooth collision-free path is generated; the candidate paths are evaluated, and a mechanical arm joint control instruction is generated; and the mechanical arm executes the planning track and carries out real-time sensing and online re-planning. According to the method, environment perception, Transform coding and diffusion generation are organically combined, a plurality of feasible tracks which are smooth and capable of avoiding obstacles are rapidly generated in a complex obstacle scene, and the method has good generalization ability and can adapt to different scenes and dimension changes.

The invention relates to the field of medical image segmentation, and discloses a precision surgery-oriented robot medical image segmentation and feature extraction method, which comprises the steps of constructing a dynamic segmentation network model of a bidirectional attention architecture, and inputting a preprocessing module for feature extraction and generating a multi-scale feature pyramid; the Transform coding branch is used for time sequence feature modeling and outputting a time sequence enhancement feature; the convolutional coding branch is used for enhancing anatomical features and surgical instrument features and outputting spatial enhancement features; the multi-stage feature fusion unit is used for performing multi-stage iterative fusion and outputting final fusion features; the decoding output module is used for decoding and generating pixel-level segmentation masks of the anatomical structure and the surgical instrument; training the dynamic segmentation network model; and performing medical image segmentation and feature extraction based on a medical image video sequence input in real time by using the trained dynamic segmentation network model. Accurate segmentation of anatomical tissues and dynamic instruments in an operation scene is realized.

The invention discloses a landslide risk assessment method based on an extreme rainfall and geology coupling model, and relates to the technical field of geological disasters. Comprising the following steps: S1, constructing a three-dimensional probability density field of a fracture network and a non-Gaussian random field model of a permeability coefficient tensor; s2, setting a physical kernel layer according to the non-Gaussian random field model, setting a data driving layer through space-time Transform coding, and constructing a graph attention network model; s3, generating an adversarial network through physical information, constructing extreme rainfall coupling data, and updating the non-Gaussian permeability coefficient random field model according to the graph attention network model; and S4, acquiring an entropy generation rate according to the mechanical field data, the seepage field data and the temperature field data, and determining a risk level. Physical interpretability grading early warning of landslide risks is realized, and meanwhile, risk space distribution can be visually displayed through a sliding surface probability cloud picture, so that accurate decision support is provided for disaster prevention and control.

A system and method for adaptive data processing combining compression and encryption. The system analyzes input data characteristics, compares probability distributions, and creates a transformation matrix to convert data into a dyadic distribution. It generates a main data stream of transformed data and a secondary stream of transformation information. The system dynamically selects and applies processing techniques, including transformation, encoding, compression, and encryption algorithms, based on analyzed characteristics and real-time performance metrics. It compresses the main data stream using Huffman coding and implements security measures to protect the output. A feedback loop monitors technique effectiveness, updates a knowledge base, and influences future selections. The system can operate in lossless, lossy, or modified lossless modes, adapting to different application requirements. This approach offers an efficient solution for scenarios where both data reduction and security are critical concerns.

Devices, systems, and methods for digital video coding, which includes matrix-based intra prediction methods for video coding, are described. In a representative aspect, a method for video processing includes performing a conversion between a current video block of a video and a bitstream representation of the current video block according to a rule, where the rule specifies a relationship between applicability of a matrix based intra prediction (MIP) mode or a transform mode during the conversion, where the MIP mode includes determining a prediction block of the current video block by performing, on previously coded samples of the video, a boundary downsampling operation, followed by a matrix vector multiplication operation, and selectively followed by an upsampling operation, and where the transform mode specifies use of a transform operation for the determining the prediction block for the current video block.

The invention discloses a dynamic graph convolution-based target point cloud completion method for a battery replacement robot, and the method comprises the steps: 1, reconstructing a complete target fastener model from a multi-view image through SFM and MVS technologies, and obtaining complete point cloud data; 2, constructing an incomplete-complete point cloud pair as training data by using a geometric constraint-based adaptive cutting strategy; 3, multi-resolution point cloud processing is adopted, and feature extraction and fusion are carried out according to three-level resolution; 4, on the basis of DGCNN dynamic graph convolution, in combination with multi-stage Edge Conv dynamic edge convolution and a Transform coding module, local geometric feature capture and global relation perception are realized; 5, point cloud generation adopts a pyramid step-by-step refining method, geometric details are added to each layer based on a previous layer result, and the number of points is gradually expanded; and 6, the training process is guided through a multi-target joint loss function, and the point cloud quality is improved while the precision is ensured. The geometric structure integrity of the point cloud of the target fastener is improved, and the operation error of follow-up operation of the battery replacement robot is reduced.

The invention discloses a hydropower station water level detection method and system. The method comprises the steps that an original water flow characteristic parameter set is obtained through multiple sets of Doppler effect radar flow measuring devices; performing segmentation extraction to obtain a water flow feature sub-data set; inputting a Transform network coding layer to generate a water flow feature coding vector; outputting a preliminary water level predicted value sequence through a decoding layer; a radar flow measurement model is called for correction to obtain a correction sequence; time sequence integration is carried out to obtain a water level change curve. The system comprises a multi-source Doppler radar signal acquisition unit, a spatial-temporal feature segmentation extraction unit, a Transform coding processing unit, a water level preliminary prediction unit, a radar model correction unit and a time sequence integration output unit. According to the method and system, multi-source parameters are deeply fused, the environmental adaptability is improved, the continuity and reliability of a detection result are enhanced, the water level of the hydropower station can be accurately monitored, and support is provided for scheduling decision and safety prevention and control.

The invention discloses a visual defect detection method based on multilevel Transform. The method comprises the following steps: firstly, extracting multi-scale features of an input image through a pre-trained convolutional neural network, and performing hierarchical fusion; then, a multi-level Transform coding and decoding structure is used for carrying out deep reconstruction on the fusion features, and long-distance dependency relationships of different granularities are captured through hierarchical patch segmentation and a self-attention mechanism; meanwhile, introducing a standard deviation to estimate branch learning reconstruction uncertainty; and finally, a pixel-level anomaly score graph is generated by calculating a normalized residual error and adopting a multi-scale anomaly score aggregation strategy, so that accurate anomaly positioning and discrimination are realized. According to the method, multi-scale feature representation and global semantic modeling capability are fused, the accuracy, robustness and generalization capability of anomaly detection are effectively improved, only normal sample training is needed, and the method is suitable for complex scenes such as industrial visual detection and has a good application prospect.

A video coder receives data for a block of pixels to be encoded or decoded as a current block of a current picture of a video. The video coder receives a set of transform coefficients of the current block. The video coder identifies multiple transform hypotheses. Each hypothesis includes two or more predicted transform parameters. The video coder computes a cost for each hypothesis by performing inverse transform on the transform coefficients of the current block according to the predicted transform parameters of the hypothesis. The video coder signals or receives a codeword that identifies a first transform mode of a first transform parameter. The codeword is assigned to the first transform mode based on the calculated costs of the multiple transform hypotheses. The video coder encodes or decodes the current block by reconstructing the current block according to the identified first transform mode.

Devices, systems and methods for digital video coding, which includes matrix-based intra prediction methods for video coding, are described. In a representative aspect, a method for video processing includes performing a conversion between a current video block of a video and a bitstream representation of the current video block according to a rule, where the rule specifies a relationship between applicability of a matrix based intra prediction (MIP) mode or a transform mode during the conversion, where the MIP mode includes determining a prediction block of the current video block by performing, on previously coded samples of the video, a boundary downsampling operation, followed by a matrix vector multiplication operation, and selectively followed by an upsampling operation, and where the transform mode specifies use of a transform operation for the determining the prediction block for the current video block.

The invention discloses a novel digital hybrid coding method based on a large model. The method comprises the following steps: S1, preprocessing to-be-coded data; s2, performing feature coding on the preprocessed data by adopting a diffusion model driven feature learning method; s3, performing neural network coding on the feature coding data, and dynamically adjusting the coding bit number based on a variable bit rate quantization method; s4, performing adaptive entropy coding on the neural network coding data, and performing probability density transformation based on a normalized flow entropy coding method; s5, discrete transform coding is carried out on entropy coding data, and hierarchical quantization is carried out on different frequency components by adopting a hierarchical sub-block quantization method; and S6, based on the code length distribution and entropy coding parameters of the A3C optimization transformation coding data, extracting a lightweight coding model from the diffusion model by using a knowledge distillation method, and generating final optimization coding data. Through the hybrid coding and dynamic optimization strategy, the data compression efficiency and the coding quality are remarkably improved.

The invention discloses a single-pixel multi-encryption imaging method and device and a storage medium. The method comprises the steps of generating a single-pixel orthogonal basis transform coding illumination pattern, and exchanging the sequence of a coding illumination pattern sequence by using a time sequence key and a remapping algorithm; the coded illumination pattern is processed by using a space key, and at the moment, an image sequence is encrypted in two aspects of a space domain and a frequency domain; projecting a preset coding illumination pattern to the surface of an object to be encrypted and imaged; measuring and recording the light intensity of the reflected or transmitted light by using a detector; decrypting the signal by using the combination of the space key and the time sequence key to obtain a plaintext signal; and reconstructing the image by using a corresponding single-pixel imaging method. According to the method, the single-pixel imaging technology is utilized, the image encryption signal can be obtained under the condition of low sampling rate, calculation only needs to be carried out when the encryption coding illumination pattern is generated, and the defect that a traditional image encryption method based on a camera is large in calculation amount is overcome.

The invention relates to the technical field of fault prediction, and relates to a flash furnace fault prediction method based on graph fusion and multi-stage learning, which comprises the following steps: establishing a static graph and a dynamic graph, describing the influence intensity and dynamic change between sensor data according to the static graph and the dynamic graph, and obtaining spatial-temporal characteristics; inputting the spatial-temporal characteristics into a Transform coding model, and obtaining prediction data; the Transform coding model is trained by using a training set, and model parameters are adjusted by minimizing a prediction error to obtain the Transform coding model; the prediction data is coded, the coded data is stretched, translated and rotated, each data point is mapped to a high-dimensional space, the distribution position of abnormal points is obtained, the abnormal score of the data is calculated, the abnormal score is judged, and the fault result of the nickel flash furnace system is obtained. According to the method, the static relation and the dynamic relation can be mined and fused from various sensor data, so that fault prediction is realized.

The invention discloses a rolling bearing life prediction method based on a secondary attention mechanism Transform coding layer and an improved time convolution network, and the method comprises the steps: collecting the full-life vibration signals of a bearing, and constructing a multi-domain and multi-class normalized degradation feature set; then, global information of the service life process of the rolling bearing is obtained through a secondary attention mechanism Transform coding layer; and obtaining key local information of the whole life stage of the rolling bearing by constructing an improved time convolution network with one-dimensional convolution layers connected in series with different scales, and carrying out smoothing processing to obtain a final life prediction result of the rolling bearing. According to the method, the overall change trend of the health state of the bearing can be effectively captured in a self-iteration form of the attention mechanism, and local time features of different scales are extracted by the improved time convolutional network, so that relatively comprehensive rolling bearing health state time feature expression is obtained, and the precision of a rolling bearing life prediction result is effectively improved.

The invention relates to the technical field of multi-modal data prediction, and discloses a fashion preference prediction method and device, and the method comprises the steps: obtaining multi-modal data comprising an image, a text and a user behavior time sequence, carrying out the feature extraction, and obtaining an image feature, a text feature and a time sequence feature; constructing spatial features based on the city embedded table and the corresponding regional culture label codes; aligning the dimensions of the spatial features and the time sequence features, and fusing the aligned spatial features and time sequence features by using a gating attention mechanism to obtain space-time fusion features; splicing the space-time fusion feature with the image feature and the text feature to obtain a multi-modal fusion feature; decoupling the multi-modal fusion feature to obtain a material decoupling feature, a style decoupling feature and a scene decoupling feature; and after feature splicing is carried out on the multiple decoupling features, the fashion preference prediction probability is obtained through Transform coding and MLP classification.

The invention discloses a perspective distortion image-oriented monocular dressing human body reconstruction method, and relates to the technical field of three-dimensional vision. The invention provides a three-dimensional human body reconstruction framework fusing distortion correction, three-dimensional geometric representation and pseudo multi-view constraint, and image distortion is relieved by uniformly partitioning an image and using virtual view transformation to decouple the relevance between an image position and perspective distortion; a multi-scale attention enhancement network is designed, and a hourglass model and a channel attention mechanism are utilized to extract block features, so that the model can process perspective distortion images; the human body parameter model is coded into a frequency domain coefficient through Fourier transform, and the frequency domain coefficient serves as the basis of geometric representation; in order to further optimize high-frequency details, the invention provides a pseudo multi-view module, and the robustness of the network to perspective distortion is enhanced by simulating multi-view data enhancement. Experimental results show that compared with the prior art, the method disclosed by the invention has the advantages that the quantification and the qualitative property on the Thuman2.0 data set and the CumHumans data set are optimal.

The invention discloses a dynamic space-time CNN-Transform emotion brain-computer interface decoding method, and relates to the technical field of brain-computer interfaces, a dynamic time feature extraction module is designed according to the sensitivity of multi-scale convolution to an electroencephalogram sequence along a time dimension, and electroencephalogram sequence time feature information is mined by using convolution kernels of different sizes; constructing a local-global spatial feature extraction module by referring to close correlation between asymmetry of left and right brain regions of the brain and an emotional state, and sequentially extracting spatial feature information of the left brain, the right brain and the whole brain of the electroencephalogram sequence; a spatial-temporal feature fusion module is designed, and spatial-temporal feature relations among the left brain, the right brain and the whole brain are mined; the long-time dependency relationship in the electroencephalogram sequence is captured through an attention mechanism by referring to the advantage of Transform on long-time sequence processing, so that the emotion electroencephalogram decoding precision is effectively improved; and finally, performing emotion recognition on the feature sequence subjected to Transform coding by using a multi-layer perceptron to realize end-to-end emotion electroencephalogram decoding.

The invention discloses a gold mine target region optimization method and system based on spatial constraint modified embedded geochemical anomalies and a memory. The method comprises the following steps: acquiring geochemical data of a target area, and preprocessing to obtain a spatially serialized standardized data matrix and an adjacent matrix; according to the method, global Transform coding modeling is adopted to obtain a hidden state sequence of global element interaction characteristics of each sampling point of a target area, then unitary potential energy and binary potential energy are constructed in sequence, a geochemical abnormal area identification model of each sampling point of the target area is established in combination with CRF, and the geochemical abnormal area marked by the model is subjected to abnormal scoring and marking. And through comprehensive evaluation, gold mine target region optimization is completed. Based on the system containing the computer program memory provided by the method, the problems of large calculation amount and low accuracy in the prior art can be effectively solved, and through tests, the area under the ROC curve of the gold mine target region optimization process by adopting the system can reach 0.92, and the obvious advantages of the system are shown.

The invention discloses a high-precision remote sensing image cultivated land boundary extraction method, which comprises the following steps of: acquiring high-resolution remote sensing image data, preprocessing the data, and obtaining cultivated land parcel vector data through manual annotation; constructing a multi-task depth semantic segmentation model based on a coding and decoding structure; constructing a mixed loss function; acquiring a high-resolution remote sensing image of a cultivated land area to be detected and performing data preprocessing; inputting a high-resolution remote sensing image of a cultivated land area to be detected into the trained high-resolution remote sensing image cultivated land extraction model based on multi-task learning; and converting the cultivated land semantic segmentation result into cultivated land parcel vector data containing latitude and longitude coordinates. According to the method, through staged Transform coding based on overlapping patches and efficient sequence reduction, efficient multi-scale coding is realized, the problem of insufficient remote dependence capture is solved, and the computing power bottleneck of high-resolution self-attention is relieved.

The invention discloses a nuclear magnetic image Alzheimer's disease diagnosis method and system based on a three-dimensional Transform. The method comprises the following steps: firstly, collecting data to construct a diagnostic data set, and preprocessing the data; an Alzheimer's disease diagnosis network model based on Transform is constructed and trained, and the Alzheimer's disease diagnosis network model based on Transform comprises a voxel attention module, a plaque attention module, a global-local feature extraction module, an auxiliary classification module, a Transform coding module and an overall classification module. After an Alzheimer's disease structure image to be diagnosed is preprocessed, the trained Alzheimer's disease deep diagnosis model is used for diagnosis classification. The method is beneficial to rapid and accurate Alzheimer's disease identification, improves the accuracy and efficiency of related disease identification, is beneficial to early prevention and multi-dimensional intervention, creates convenience for Alzheimer's disease patients, and reduces the burden of a public medical system.

The invention discloses a multi-modal data fusion and intelligent analysis method and system for advanced early warning of wind and light storage equipment, and the method comprises the steps: deploying data collection equipment, synchronizing the data collection equipment to a virtual equipment model through a digital twin interface, and generating a physical enhancement feature; recombining the physical enhancement features into a three-dimensional feature tensor; orthogonal constraint tensor decomposition is carried out on the three-dimensional tensor, and cross-scale fault features are extracted; constructing a fault propagation graph network FPGN based on the core tensor, and outputting an early warning decision through graph convolution and Transform coding; according to the early warning deviation, a gradient descent method is adopted to adjust the physical constraint weight, and the model is retrained; the three-dimensional feature tensor recombination realizes unified characterization of multi-modal data, orthogonal constraint tensor decomposition forces different dimension features to be independent, a cross-scale coupling mode of a fault is extracted, and the effects of dimension reduction and efficiency improvement are achieved; according to the FPGA network, graph convolution and Transform time sequence coding are fused, and the propagation intensity of modeling faults in an equipment topology network is realized, so that early warning from local anomalies to system-level risks is realized.

There is provided a computer-implemented method for verifying the robustness of a neural network classifier with respect to one or more parameterised transformations applied to an input, the classifier comprising one or more convolutional layers, the method comprising: encoding each layer of the classifier as one or more algebraic classifier constraints; encoding each transformation as one or more algebraic transformation constraints; encoding a change in an output classifier label from the classifier as an algebraic output constraint; determining whether a solution exists which satisfies the classifier constraints, transformation constraints and output constraints, and determining the classifier as robust to the local transformations if no such solution exists. A perception system and a computer readable medium are also provided.

The invention discloses a knowledge graph multi-modal document analysis and image table semantization knowledge recall method, and belongs to the technical field of knowledge engineering and information retrieval. The invention provides an innovative scheme for fusing a visual language model, semantic abstract generation and knowledge graph modeling. The method comprises the following steps: constructing a vertical domain knowledge graph by adopting a BERT-BiLSTM-CRF model; according to the method, multi-modal document analysis is realized through models such as DocLayout-YOLO, TableMaster, UniMERNet and the like; the method comprises the following steps of: segmenting an image into 16 * 16 block sequences by adopting a vit-gpt2-image-adaptation model, and realizing image semantization through 768-dimensional vector space mapping and Transform coding; constructing a document summary tree based on DBSCAN clustering and LLM recursive summary; and designing a hybrid retrieval space fusing semantic vectors and structured vectors, and reordering by adopting a double-attention mechanism. According to the method, the knowledge base document retrieval recall rate is increased to 99%, the question and answer accuracy rate reaches 90% or above, the index construction time is shortened by 60%, and the problem that semantic understanding and recall of non-text elements in complex documents are difficult is effectively solved.

The invention discloses a text classification method and device and a processor. A normalized text vector is obtained by preprocessing and coding an input text. Then utilizing a multi-layer structure in a pre-trained bad text recognition model, firstly carrying out first-level category classification on the input vector through a first Transform coding layer and a classification layer, and meanwhile, carrying out related semantic clustering through a first clustering layer to generate a first clustering text vector; then, a second Transform coding layer and a classification layer complete finer-grained secondary category classification based on the first clustering text vector, and continuous aggregation is performed through a second clustering layer to generate a second clustering text vector; and finally, the third Transform coding layer and the classification layer are combined with the second clustering vector to output a global classification result. And finally, all classification results are screened by setting a confidence threshold, high-confidence categories are reserved, and accurate classification results with clear levels are obtained.

A method comprising: processing a video frame; determining a reference block for a current block of the video frame; predicting the current block with an intra block copy method; deriving intra prediction information for the current block based on the reference block; selecting a transform coding for the current block based on the intra prediction information; and applying the selected transform coding to the current block.

Methods, systems and devices for implementing an adaptive color transform (ACT) during image / video encoding and decoding, including: determining, for a conversion between a video comprising a block and a bitstream of the video, that a size of the block is greater than a maximum allowed size for an ACT mode, and performing, based on the determining, the conversion, wherein, in response to the size of the block being greater than the maximum allowed size for the ACT mode, the block is partitioned into multiple sub-blocks, and wherein each of the multiple sub-blocks share a same prediction mode, and the ACT mode is enabled at a sub-block level.

The invention discloses a visual Transform-based feature focusing image semantic segmentation method, which adopts a feature pyramid structure to generate four feature maps with different sizes, and uses an overlapping block embedding module in a feature map embedding process to ensure that local information is continuous to a certain extent while the feature maps are partitioned. A position code generation module is used for dynamically generating corresponding position codes according to local neighborhood information, translation invariance needed in computer vision is kept, and then a fusion module combining a focusing attention mechanism and convolution operation is adopted in the Transform coding process. A self-attention mechanism is utilized to grasp a dependency relationship between an overall structure and a long distance, and convolution attention is introduced to enable the model to maintain sensitivity to local information, so that the model can integrate global and local detail information at the same time. And finally, inputting the feature maps of different sizes generated in each stage of the model into a decoder for segmentation. According to the method, the segmentation performance of a target boundary and a complex scene is improved.

The invention discloses a science popularization content intelligent retrieval interaction system and method combined with a knowledge graph. The system comprises a knowledge graph construction unit, a multi-layer multi-head attention mechanism collaborative Transform coding unit, a knowledge graph embedding and Transform feature fusion unit, a retrieval result sorting unit based on fusion features, an interactive intention capture unit and a result optimization output unit. The method comprises the following steps: constructing a knowledge graph for multi-source science popularization data, extracting retrieval keyword features by applying Transform coding, fusing the knowledge graph and the retrieval keyword features, and calculating the similarity with science popularization content to finish initial sorting; and meanwhile, capturing user interaction behavior analysis intention change in real time, and readjusting and outputting an initial result in combination with knowledge graph semantic association. The method corresponds to operation steps of all units of the system, and intelligent retrieval interaction of popular science content is achieved. According to the system and the method, the problems of insufficient semantic understanding, poor interaction experience and the like of traditional retrieval are effectively solved, and the accuracy and the interaction intelligence of popular science content retrieval are improved.