80 results about "Parallel encoding" patented technology

The invention discloses a dynamic compression coding method, which is applied to a sparse matrix, and comprises the following steps: S1, obtaining the sparse matrix, and generating a bitmap with the same dimension as the sparse matrix, a value in the bitmap being used for indicating whether an element at a corresponding position in the sparse matrix is zero; s2, a mark sequence is generated according to the bitmap, and each mark corresponds to multiple continuous elements in the sparse matrix and is used for representing whether each element in the multiple elements is zero or not; s3, extracting non-zero elements in the sparse matrix according to the mark sequence; and S4, combining the flag sequence with the non-zero elements to form a data stream after compression coding. The invention also discloses a dynamic compression coding system, a storage medium and a device. According to the dynamic compression coding method, the index storage cost can be reduced, the data access efficiency can be improved, the storage and bandwidth occupation can be reduced by supporting parallel coding and decoding, and finally the calculation throughput can be improved.

The invention relates to the technical field of load prediction, and provides a power load prediction method and system based on time sequence decomposition and an attention mechanism, and the method comprises the steps: carrying out the adaptive time sequence decomposition of an obtained original load sequence, calculating the sample entropy of each decomposed component, and carrying out the clustering; constructing a group of encoder and decoder networks for each piece of clustered data, performing parallel encoding to extract features, performing serial decoding reconstruction on the features from low frequency to high frequency, and outputting prediction data from low frequency to high frequency step by step; the weight is initialized based on the sample entropy, the trained weight is obtained through optimization in the encoder and decoder network training process, and the predicted value of the power load is obtained through weighted fusion. According to the method, adaptive time sequence decomposition, a weight mechanism guided by sample entropy and an attention-enhanced encoder-decoder structure are introduced, so that multi-component collaborative modeling and cross-scale dynamic prediction are realized, and the prediction accuracy and stability in complex load data and small sample scenes are effectively improved.

The invention discloses a real-time intrusion detection system and method based on deep learning, and aims to solve the problem that when the existing deep learning intrusion detection technology processes high-dimensional network flow data, the information fidelity and the real-time processing efficiency are difficult to consider at the same time. According to the system and the method, a heterogeneous feature decoupling and fusion network architecture is adopted, high-dimensional feature vectors are losslessly decomposed into feature subsets of static state, time sequence, graph topology association and the like, deep learning is carried out by a parallel coding subsystem, cross-modal fusion attention self-adaptive aggregation is carried out, and finally classification judgment is realized. According to the scheme, on the basis of keeping the integrity of high-dimensional data, the detection capability and the real-time processing efficiency of unknown attacks are remarkably improved, and the self-adaptability and the robustness are improved.

This invention discloses a knowledge- and data-driven time series forecasting method and system, belonging to the field of industrial time series forecasting technology. To address the technical problems of existing technologies when processing strongly non-stationary data, such as key signal attention dilution, systematic underestimation of peak amplitude, incomplete capture of multi-scale dynamic patterns, and difficulty in embedding physical domain knowledge, this invention constructs standardized multi-dimensional time series data, performs heterogeneous feature grouping and parallel encoding to obtain multiple sets of feature representations, then extracts and integrates features from multiple sources to obtain intermediate state features, and finally performs multi-branch prediction and adaptive weighted fusion based on the intermediate state features to obtain the final time series forecast result. This invention can achieve active focusing on abrupt events in industrial sequences and accurate characterization of extreme value amplitudes, significantly improving prediction accuracy, physical consistency, and robustness to extreme conditions.

This invention provides a lossless compression and decoding method for depth images. Lossless compression includes: dividing sequential data into multiple independent image blocks according to predetermined partitioning rules; selecting a corresponding thread model based on the characteristics of the depth image; calling a thread based on the thread model and using a reversible encoding algorithm to independently encode each image block; wherein, at the start of encoding for each image block, the encoder state is reset so that the encoding of each image block does not depend on the data of the preceding image block; generating metadata containing position index and compressed data length information and assembling it with the encoded image block for output. Lossless decoding includes: reading the metadata and decoding at least one image block based on the metadata. Through the above methods, parallel encoding and decoding are achieved; precise positioning via physical offsets and random access and on-demand region decoding are supported, improving the processing response of high frame rate depth streams under low computing power environments while ensuring lossless restoration.

The present application relates to the field of substation wiring, especially to a wiring system and method for skid-mounted electric control integrated substation, the method comprising: constructing time data sequences of cabinet temperature, cabinet temperature and humidity outside the substation, and comprehensive hidden features of the last time step, and passing them through a multivariate attention mechanism and a parallel LSTM encoding model to generate encoding features; passing the time data sequences through a water vapor mass conservation equation to generate physical prediction features; and passing the encoding features and the physical prediction features through a fusion mapping layer, a time attention mechanism and a control model to generate dehumidifier working power and heater working power. The present application realizes intelligent control of the dehumidifier and the heater in the anti-condensation wiring of the skid-mounted electric control integrated substation, and ensures appropriate heat dissipation and dehumidification effects.

Various implementations involve methods for encoding a media image. A video encoder encodes media images in a video stream by dividing the media images into blocks of pixels and encoding numerical coefficients associated with the blocks to generate a bitstream. As part of the encoding process, the video encoder estimates the bitrate required to encode a set of blocks of the media image. The video encoder produces more accurate bitrates for encoding the blocks by performing the bitrate estimation based on various context-based parameters of the coefficients, rather than just the coefficients themselves. This context-based bitrate estimation method can improve the visual quality of the encoded video stream.Furthermore, the video encoder encodes blocks using various simplifying assumptions to eliminate inter-block dependencies without unduly reducing visual quality. As a result, each block can be encoded independently and in parallel with the encoding of other blocks.

The invention discloses a method and a system for detecting rice bacterial leaf blight based on unmanned aerial vehicle spatial spectrum fusion. The method comprises the following steps: firstly, constructing a rice bacterial leaf blight detection data set; a rice bacterial leaf blight severity grading detection double-branch model is constructed, and the model adopts a spectrum-space double-branch parallel coding and fusion framework and comprises a spectrum branch module, a space branch module and a cross attention fusion mechanism module. And inputting the complete image of the rice field into the trained rice bacterial leaf blight severity grading detection double-branch model, and obtaining severity grading detection results of the bacterial leaf blight in different rice growth periods of the target field. According to the invention, deep complementation and dynamic information balance of spectrum and spatial features of the multispectral unmanned aerial vehicle are realized, so that the accuracy of disease detection and the generalization ability of the model are effectively improved.

This invention discloses an adaptive end-to-end network learning-driven polarization descattering depth estimation method. It proposes a physics-driven end-to-end network to simultaneously acquire clear reconstructed images and high-precision depth images in scattering environments. The method first acquires linearly polarized images of the scene in four directions, extracting light intensity and linear polarization feature maps through decoupling calculations. Then, a dual-task network including a polarization-guided fusion module is constructed. Utilizing the physical sensitivity of DOLP to scattering particle concentration and optical path length, it transforms these into a spatial attention mask, physically constraining and enhancing the shallow texture of the intensity map in the feature space. Finally, dehazing and depth estimation branches are encoded and decoded in parallel. This invention overcomes the intensity-depth ambiguity of traditional RGB images, effectively suppressing the interference of non-uniform fog on geometric perception and visual texture, and achieving high-precision scene depth estimation and high-fidelity dehazing imaging under adverse weather conditions.

The application discloses a synchronization method of a multi-core video encoder. Each encoding core of the video encoder corresponds to a piece of encoding instance reconstructed frame state information. A piece of encoding instance reconstructed frame state information in an invalid state is configured according to information of a video frame to be encoded. As the encoding is performed, the encoding core updates an image block row number of a reconstructed image in the corresponding encoding instance reconstructed frame state information in real time. In the process of inter-frame parallel encoding, each encoding core does not need to be synchronized with other encoding cores, or the encoding core corresponding to a video frame M(N) only needs to be synchronized with the encoding core corresponding to a video frame M(N-1). When the video frame M(N) is encoded, the encoding core only needs to be synchronized with the encoding state of the video frame M(N-1) at most, so that the synchronization relationship and the synchronization operation among the encoding cores are greatly simplified.

The application discloses an FSE encoding method, device, equipment and storage medium, comprising: obtaining a to-be-encoded symbol sequence corresponding to to-be-encoded data, and determining a to-be-encoded symbol with no dependency relationship between a symbol state and other symbol states from the to-be-encoded symbol sequence to obtain a target to-be-encoded symbol; splitting the to-be-encoded symbol sequence based on a position of the target to-be-encoded symbol in the to-be-encoded symbol sequence to obtain a plurality of to-be-encoded subsequences corresponding to the to-be-encoded symbol sequence; and separately encoding the plurality of to-be-encoded subsequences to obtain the to-be-encoded data after encoding. The application splits the to-be-encoded symbol sequence by taking a symbol with unique state as a splitting point, and then obtains a plurality of subsequences, which can realize parallel encoding of the subsequences compared with serial encoding of the entire to-be-encoded symbol sequence after splitting, thereby improving the encoding rate.

The invention provides a positive sample anomaly detection method and system based on dense features, and relates to the technical field of anomaly detection, and the method comprises the steps: collecting first multi-dimensional data corresponding to a positive sample and second multi-dimensional data of to-be-detected industrial equipment; extracting a first dense feature from the first multi-dimensional data through a pre-constructed visual large model, and extracting a second dense feature from the second multi-dimensional data; respectively carrying out parallel coding processing on the two types of features to generate first feature data and second feature data; performing adaptive adjustment on the first feature data and the second feature data through a pre-constructed feature migration model to generate third feature data and fourth feature data; and on the basis of the third feature data, standard feature data of the target industrial equipment in a normal operation state is determined, the standard feature data is compared with the fourth feature data to generate an anomaly detection result of the to-be-detected industrial equipment, and the generalization ability and discrimination precision of positive sample anomaly detection in a complex industrial environment are improved.

The invention relates to a multi-modal MRI brain tumor segmentation method based on convolution attention, and the method comprises the steps: carrying out the parallel coding of multi-modal medical image data, and extracting the specific low-layer space features of each modal; performing feature fusion and high-level coding on the low-level spatial features to obtain shared features; carrying out convolution attention processing on the shared features to obtain enhanced features; and carrying out decoding processing on the enhanced features to obtain a segmentation result. According to the method, more accurate tumor boundary segmentation and tissue differentiation can be realized in medical image analysis such as brain MRI (Magnetic Resonance Imaging), meanwhile, the parameter efficiency, the calculation speed and the generalization ability under limited data of the model are remarkably improved, and the clinical practicability is enhanced.

This invention discloses a hyperspectral image classification method and related apparatus based on a dual-branch Mamba-Former network. First, the hyperspectral image data to be classified is input. Then, spectral compression and local spatial feature extraction are performed using the SS-ResNet module, and classification tokens are added to generate a token sequence. The token sequence is then input into the parallel encoding branch, MAPE Block, for feature encoding. Further, the classification tokens from the parallel encoding branch are extracted and deeply fused using Fusion-MLP to obtain the final discriminative feature representation. Finally, the fused features are input into a classifier to predict the land cover categories in the hyperspectral image. This invention effectively integrates the local inductive bias of CNNs, the global alignment of the global context branch MHSA, and the linear, efficient long-range modeling capabilities of Mamba, improving the accuracy of deep learning applied to hyperspectral image classification.

The invention discloses a code rate control method for parallel coding based on virtual subframes. The method comprises the following steps: S1, dividing a to-be-coded video frame into one or more sub-frames which are independent from one another and are not overlapped with one another; and S2, calculating a target coding bit number of the to-be-coded sub-frame. And S3, dividing the to-be-coded subframe into a plurality of virtual subframes according to the number of the coding cores participating in parallel coding in the subframe and a division rule of the virtual subframes. And S4, scheduling the coding cores participating in parallel coding in the sub-frames based on the virtual sub-frames, and respectively scheduling each virtual sub-frame in the sub-frame to be coded to different coding cores to carry out parallel coding in the sub-frames. According to the invention, the result consistency of multiple times of coding of the sub-frame is ensured; and meanwhile, strict synchronization is not needed.

The invention discloses an audio magnetotelluric two-dimensional deep learning inversion method integrated with physical prior, and belongs to the technical field of electromagnetic data inversion processing. Constructing two-dimensional geological resistivity simulation models in batches through a parametric modeling method; further forming a training set sample containing simulation model-response data-physical prior benchmark; constructing a deep learning network comprising a heterogeneous double-channel parallel encoder; a space self-adaptive dynamic gating aggregation module is introduced into the network; constructing a loss function of cooperation of data and a physical mechanism; performing parameter iteration and optimization on the deep learning network by taking minimization of the loss function as a target; according to the method, through a physical prior guidance and dynamic feature adaptive fusion mechanism, the defects of divergence and fuzziness of anomalous body boundaries are effectively overcome, the recognition precision and noise resistance of complex underground structures and deep concealed ore bodies are greatly improved, and an efficient and accurate technical scheme is provided for deep mineral resource exploration.

The invention discloses a parallel stepped convolutional code space coupling coding and decoding (PSCBCC) method and device, belongs to the technical field of channel coding, and is suitable for a large-bandwidth communication scene. According to the method, a frame of data is divided into multiple groups of sub-data, parallel convolutional coding of three paths of input of current sub-data, interleaved previous sub-data and an interleaved previous check sequence is realized by utilizing a parallel coding unit, a current check sequence is generated, and the sub-data and the check sequence are respectively reserved to different subsequent moments to form asymmetric coupling; and during decoding, a parallel decoding unit is matched with a coding structure, and a plurality of groups of log-likelihood ratios are input to realize iterative updating. The throughput rate is improved through parallel processing, the bit error rate performance is enhanced through asymmetric coupling, multi-code-rate adaptation is supported, and the performance is superior to that of existing BCC and 4G / 5G standard codes.

The invention discloses a training method of a cue word quality evaluation model and a quality evaluation method.The training method comprises the steps that a cue word training text is obtained, text sequence word segmentation is conducted on the cue word training text, and a semantic accuracy sequence, an expression integrity sequence and a logic continuity sequence of the cue word training text are obtained; according to the semantic accuracy sequence, the expression integrity sequence and the logic continuity sequence, performing multi-dimensional parallel coding fusion processing to obtain a target cue word representation vector; performing double-branch prediction fusion on the target cue word representation vector to obtain fused prediction data; and according to the fusion prediction data, performing parameter updating on the initialized cue word quality evaluation model to obtain a trained cue word quality evaluation model. According to the training method, a cue word quality evaluation model can be provided, and the model can be beneficial to improving the interpretability and accuracy of cue word quality evaluation. The invention relates to the technical field of natural language processing.

The application provides a well logging productivity prediction system and method based on an improved attention mechanism, relates to the field of oil and gas exploration and development, and comprises a semi-structured data conversion module, a depth alignment and resampling module, a training data construction module, a hybrid model construction and training module and an unknown well yield prediction module. The system automatically analyzes WIS logging files, extracts multiple logging curves, realizes depth unification, missing value filling and interactive feature construction; based on a parallel coding structure of a one-dimensional convolution network and a multi-head attention mechanism, the local features and long-range dependencies of the logging curves are jointly modeled by combining a multi-scale residual network, so that high-precision prediction of the oil test yield is realized. The system can generate a depth-yield curve according to the prediction result, divide the yield into five grades, and realize automatic evaluation of the productivity of a new well.

Methods and apparatuses are described to encoded data into a bitstream and to decode data from a bitstream. The method is able to perform parallel encoding and decoding efficiently and avoids padding of substreams thus reducing the amount of bits within the bitstream. Portions of input data channels are multiplexed and encoded into substreams. During the multiplexing shuffling methods are applied in order to obtain substreams of more uniform lengths. The amount of bits within the substream may be further reduced by including only the relevant significant bits within the trailing bits of the encoding process.

This invention provides an efficient parallel encoding and decoding method for BCH codes, belonging to the technical field of BCH encoding and decoding circuit implementation in the areas of memory and error correction coding. Unlike previous lookup table methods, for an (n, k, t) BCH code with an n-bit codeword length, k data bits, and t-bit error correction capability, this invention only needs to store the values ​​of k n-k bit parity check matrix column vectors. These k n-k bit parity check matrix column vectors are then XORed with the S-synonym value for t rounds. The error pattern corresponding to the received codeword is obtained from the bitwise XOR value and corrected. This invention is a hardware-level encoding and decoding implementation, which can be completed within one clock cycle, reducing the multi-cycle decoding delay caused by iterative algorithms, achieving parallelization of BCH encoding and decoding, simplifying the encoding and decoding process, and reducing resource consumption.

The invention discloses a method and system for accelerating construction of a constellation model and a storage medium. The method comprises the following steps: decomposing an algorithm for constructing a constellation model into an algorithm serial part and an algorithm parallel part according to a preset decomposition mode; coding modification suitable for the CPU process is carried out on the serial part of the algorithm, and parallel coding modification suitable for the AI accelerator process is carried out on the parallel part of the algorithm. And merging and reconstructing the transformed algorithm serial part and algorithm parallel part, and executing the merged and reconstructed algorithm for constructing the constellation model by using an on-board computer carrying an AI accelerator. According to the method, the AI accelerator is used for parallel computing, the constellation model construction algorithm is optimized in parallel, and the number of serial cycles of a CPU is reduced, so that the problem of low operation efficiency of the model construction algorithm is solved, and the acceleration effect is achieved.

This invention discloses a multi-scale spatiotemporal semantic object state representation method based on brain-inspired biomimetic mapping, belonging to the field of brain-inspired computing technology. Addressing the problems of cross-level semantic abstraction fragmentation, low efficiency of topological association, and semantic redundancy accumulation in city-level video swarm intelligence perception, this invention constructs a hierarchical mapping architecture from edge to middle layer to cloud, sequentially performing progressive abstraction representations from pixel level to individual level, group level, and city level. A multi-dimensional weighted sensor network topology model is constructed, generating a full-cycle continuous spatiotemporal state vector through cross-sensor association matching and trajectory stitching, and completing global optimization and deduplication of cross-camera samples. Simultaneous parallel encoding of metadata and feature streams is performed for swarm intelligence semantic compactness, globally deduplicating and differentially encoding duplicate features and redundant samples. This invention achieves efficient multi-scale spatiotemporal semantic expression through edge-cloud collaboration, significantly reducing transmission and storage overhead and improving the accuracy of cross-regional long-term trajectory stitching.

The invention discloses a fault diagnosis method based on multi-scale joint feature extraction, and relates to the field of fault diagnosis, and the method comprises the steps: dividing preprocessed multi-source time series data into a source domain data set and a target domain data set; performing feature extraction on multi-source time series data in the source domain data set and the target domain data set, and performing fusion encoding on extraction results through a parallel encoder to generate fusion features; inputting the fusion features into a constructed fault diagnosis model, and training and optimizing the fault diagnosis model in combination with a multi-core maximum mean difference algorithm and a dynamic balance strategy to obtain an optimal fault diagnosis model; and according to the full connection layer and the classification layer of the optimal fault diagnosis model, performing cooperative processing on the fusion features of the target domain data to obtain a fault diagnosis result. According to the invention, by introducing a multi-scale recursive trajectory fusion mechanism, the detection sensitivity and recognition accuracy of the fault diagnosis model to early weak faults and composite faults are improved.

This invention relates to the field of computer vision and image processing technology, specifically to an industrial image anomaly detection method and system based on multi-source feature fusion and synthetic anomaly enhancement. This invention aims to solve the problems of insufficient feature utilization and pixel-level segmentation edge blurring in existing unsupervised industrial image anomaly detection methods. By combining a non-parametric memory prior with a parametric fine-tuning filter network, and introducing a spatial attention mechanism to guide multi-source feature fusion, this method can effectively detect minute defects on industrial surfaces with only normal sample training. The technical solution includes: constructing a frozen multi-scale feature extractor; establishing a non-parametric feature memory; generating a coarse anomaly prior map; designing a multi-source feature parallel encoding network; constructing an attention-guided feature fusion and decoding module; generating synthetic anomaly samples based on a Berlin noise field; performing gradient-truncation hybrid forward propagation; and performing forward inference to obtain anomaly localization results.

The present disclosure relates to a video encoding method and device, electronic equipment, storage medium, computer program product and a method of generating a bitstream. The video encoding method comprises: obtaining real encoding information of an encoded frame in a to-be-encoded video; obtaining quantization scales and complexity information of parallel frames in the to-be-encoded video which are encoded in parallel with a current frame, wherein the encoding order of the current frame is after the parallel frames; predicting encoding information of each parallel frame based on the quantization scales and complexity information of each parallel frame; predicting a quantization scale of the current frame based on the real encoding information of the encoded frame and the predicted encoding information of each parallel frame; and encoding the current frame based on the predicted quantization scale of the current frame.

A method of video encoding includes, prior to encoding a first tile of s plurality of tiles of a current picture, initializing a shared row buffer that is shared among multiple processor threads associated with the first tile. The method also includes encoding a first unit of a plurality of units in a first row of the first tile by a first processor thread and using a corresponding first HMVP buffer. The method also includes, when all of the plurality of blocks in the first unit have been encoded, copying contents of the first HMVP buffer into the shared row buffer, copying contents of the shared row buffer into a second HMVP buffer, starting encoding of a unit in a second row of the plurality of rows by the second processor thread using the second HMVP buffer, and resetting the first HMVP buffer.

The embodiment of the invention provides an EFEC encoder, an EFEC decoder, a method, equipment and a system, and relates to the technical field of optical communication, in the EFEC encoder, a first bit width conversion module converts the bit width of input data into the bit width of an RS domain; the first control module performs blocking and value filling operation on the first RS domain data; the RS coding module performs parallel RS coding on each symbol included in the second RS domain data through a preset integer number of RS coding channels; the second control module deletes the filling value in the second RS domain data; the second bit width conversion module converts the bit width of the first RS domain data and the RS check code into the bit width of the BCH domain; the BCH coding module carries out parallel BCH coding on each symbol included in the BCH domain data through a preset integer number of BCH coding channels; and transmitting the BCH domain data and the BCH check code. According to the scheme, the encoding and decoding speed of the EFEC can be improved under the condition of saving hardware resources.

The application provides a frame parallel multimodal state level communication method, device and equipment for long video understanding and a storage medium, and relates to the technical field of artificial intelligence and computer vision. The method comprises the following steps: dividing an input video into a plurality of frame subsequences; based on a plurality of provider large language models, encoding the plurality of frame subsequences in parallel when sharing the same text context or task instruction to generate multimodal intermediate state representations; extracting visual key-value states therefrom, and performing distinguishing, screening and normalization processing to form a multimodal state key-value pair set; injecting the visual key-value states into a text condition key-value cache of a carrier large language model; based on the key-value cache after completing state injection, performing inference calculation by the carrier large language model to generate video understanding related output results. The application can guarantee communication efficiency while retaining and fusing key visual and timing information in the video more fully, and improve the overall inference performance in the long video understanding task.

The invention discloses a multi-scale Transform-based computing network resource load feature representation method and device and a storage medium, and the method comprises the steps: obtaining multi-source computing network resource load related historical time sequence data, and dividing the data according to different time scales of hours, days and weeks to form a multi-scale data sub-sequence set; for the data sub-sequence of each time scale, respectively carrying out feature embedding and position coding processing to obtain a feature vector sequence reflecting the load change under the scale; a plurality of groups of parallel Transform encoders are designed, and depth feature extraction is carried out on the feature vector sequence of each scale; a scale attention module is introduced and is used for establishing a relation among representations of different scales and fusing global dependency information under each scale; the output features of the Transformer encoders of all scales are combined into a uniform load feature representation vector according to a certain mode, and the uniform load feature representation vector serves as comprehensive representation of the current load state of calculation network resources; and the load characteristic representation is input into the prediction module for realizing prediction of the load at the future moment. According to the method, the characteristics of the computing network resource load data under different time scales can be effectively fused, and the accuracy, the stability and the generalization capability of load modeling are improved.