83 results about "Decoder architecture" patented technology

Disclosed in the present invention is a non-autoregressive Transformer-based modeling method for a 4-level pulse amplitude modulation high-speed transmitter. The method involves establishing a deep learning model having an encoder-decoder architecture to predict the behavior of a 4-level pulse amplitude modulation transmitter. An encoder processes unordered non-sequential inputs, including an input signal parameter and link parameters, to generate a context vector and then transmit same to a decoder. The decoder uses both the context vector generated by the encoder and a transmitter output signal sequence to generate a categorical probability distribution for each point in the sequence one by one. The model is trained using a random masking strategy, and inference is performed by means of non-autoregressive decoding and filtering, so that the model can perform parallel prediction on an output sequence, and perform a filtering process to predict an output signal. Compared to traditional simulation methods, the present invention achieves a significant acceleration effect, particularly when processing multi-link systems.

The invention relates to the technical field of infrared light machine system thermal deformation prediction, in particular to a multi-material structure thermally induced stress deformation prediction method based on a graph neural network. The method comprises the steps of data set establishment, graph structure establishment, graph neural network model establishment and training and model and parameter optimization. Finite element nodes correspond to graph nodes, finite element edges correspond to graph edges, an encoder-message passing-decoder architecture model is established, and node states are updated through a three-layer physical symmetry message passing mechanism. Physical constraint loss including minimum displacement smoothness constraint and stress continuity constraint is innovatively added into a loss function. Compared with traditional finite element calculation, the method has the advantages that the speed is increased by more than 100 times, high hardware adaptability is achieved, the black box limitation of a data-driven neural network model is broken through, thermally induced stress deformation analysis caused by different material coefficients can be processed, the adaptability to geometric changes is high, and good engineering application value is achieved.

The invention provides an industrial defect detection method based on a multi-scale Transform, and the method comprises the steps: employing a pre-training network to extract multi-scale deep features, introducing a multi-scale Transform encoder-decoder architecture, achieving the long-distance dependence modeling through a multi-head self-attention mechanism, and obtaining the global representation capability exceeding a convolution network. Compared with CNN-U-Net, lossless jump connection is designed, 512-dimensional features generated by an encoder can be completely transmitted to a decoder, and tiny defect information loss caused by a traditional dimension reduction mode is avoided; the calculation complexity of global attention is remarkably reduced through a three-level local attention decomposition mechanism and is only 0.47% of that of a standard Transform, and the industrial real-time detection requirement is met; and self-adaptive error normalization and multi-scale smooth fusion are combined, so that false alarms at complex textures are effectively inhibited, and patch boundary artifacts are eliminated. Finally, according to the method, a pixel-level abnormal score graph is generated based on reconstruction deviation, and high-precision positioning and detection of multi-scale defects are achieved.

The invention discloses a CBCT tooth segmentation method and system based on an anatomical perception cascade network, and the method comprises the steps: a first stage, carrying out the simplified dichotomy segmentation based on an original CBCT image and a coarse segmentation network taking 3D U-Net as a trunk, outputting maxillary and mandibular tooth probability graphs, and taking the maxillary and mandibular tooth probability graphs as prior information to guide the generation of an SDM; in the second stage, the original CBCT image and the calibrated maxillary tooth probability graph and the calibrated mandibular tooth probability graph are spliced together, a formed multi-channel input tensor is input into a fine segmentation network, the fine segmentation network takes Residual U-Net as a trunk, and an improved AGBR module and an improved SDMAA module are integrated in an encoder-decoder architecture of the fine segmentation network; and the decoder fuses all refined and re-calibrated feature maps, and upsamples and reconstructs 42 types of instance segmentation results with correct topology and clear boundaries. According to the method, the problem that in the prior art, when the inherent and local boundary fuzzy defect in CBCT is overcome, an effective pertinence mechanism is lacked is solved, and precise and robust 42-class instance segmentation can be achieved.

The invention discloses a hydropower station operation state intelligent prediction method and system based on multi-source data fusion. The method comprises the steps that heterogeneous data including but not limited to water level, flow, generating capacity, equipment operation parameters, environmental meteorological data, hydrological data and the like are collected in real time from multiple data sources of a sensor network, an SCADA system, a meteorological station and a hydrological station of a hydropower station; the method comprises the following steps of: preprocessing collected original data, extracting time sequence features, fusing multi-source data, taking a fused multi-dimensional feature vector as input, and learning and training through an encoder-decoder architecture of a model so as to capture a complex dynamic mode of an operation state of a hydropower station; and inputting the data acquired and preprocessed in real time into the trained AI prediction model, and generating prediction results of key operation parameters, equipment states, residual life and potential risks of the hydropower station. The method can more accurately capture the complex nonlinear relation in the operation of the hydropower station, and significantly improves the prediction precision of key parameters and risks.

The invention discloses a Conformer-based mixed self-attention and convolution improved speech recognition method and system. The recognition method comprises the steps of obtaining speech feature representation; the speech feature representation is input to a Conformer encoder for encoding processing, high-dimensional feature representation is obtained, and the Conformer encoder comprises a convolution module, an attention linear enhancement module and a feedforward network residual scaling module; according to the high-dimensional feature representation, bidirectional decoding is carried out through a bidirectional Transform decoder, and bidirectional decoding output is obtained; and carrying out merging processing on the bidirectional decoding output to obtain a speech recognition result. According to the method, the modeling capacity for long-time dependence and global context is enhanced through a bidirectional decoder architecture, the processing capacity of the model in a long sequence is improved through ALiBi relative position coding, and the model can be flexibly adjusted in a noise environment through a right decoder weighting adjustment mechanism.

The invention discloses a robot control system and method based on AI voice, and belongs to the technical field of robot intelligent control. The system comprises a local voice processing module, a dialogue management and context understanding module, an instruction mapping and generating module and a robot communication and control module. The local voice processing module is integrated with a 16 kHz sampling ASR engine, voice-to-text conversion is realized through a 6-layer LSTM network and beam search decoding, and the TTS engine adopts a WaveNet architecture to synthesize voice. A 7 billion parameter large language model is arranged in the dialogue management module, a 32-layer Transform decoder architecture is adopted, 10 rounds of dialogue history is maintained, and intention recognition and entity extraction are achieved through a multi-head self-attention mechanism. And the instruction mapping module performs trajectory planning by adopting quintic polynomial interpolation and quaternion spherical linear interpolation. According to the invention, full-link localization processing is realized, the speech recognition delay is 180 milliseconds, the position error is less than 0.5 mm, and the problems of high delay, high privacy risk and weak complex instruction understanding ability in the prior art are effectively solved.

The invention discloses an epileptic seizure video detection method based on a memory enhanced double-branch auto-encoder. The method comprises the following steps: firstly, preparing and preprocessing data; secondly, constructing an ME-DBAE model which comprises a query generation module, a multi-head memory module and a double-branch decoder module; training the model by using training set data, finally inputting a to-be-detected video into the trained model to obtain an epileptic seizure score, and judging whether the video belongs to an epileptic seizure sample or not through a threshold value. By designing the multi-head memory module, diversified normal motion prototypes are stored in parallel, and the limitation that a single memory structure is difficult to adapt to complex normal actions is broken through; and by constructing a double-branch decoder architecture, double tasks of input reconstruction and future frame prediction are cooperatively realized, and meanwhile, space structure and time trend characteristics of normal motion are captured, so that epileptic seizure actions and normal actions are effectively distinguished.

The invention discloses a Raw domain moire removing method based on Mama, and relates to the technical field of image signal processing. The invention relates to a Mamba-based Raw domain moire removal method, which comprises the following steps that: S1, a Mamba-based Raw domain moire removal model is constructed, and the model adopts a U-Net type multi-scale encoder-decoder architecture; s2, designing a selective scanning module based on time and space, and adaptively focusing on a space-time region with significant moire; s3, training the model by using a deep learning Pytorch framework, and repeatedly traversing the VDRawmoire data set until the model is converged; s4, inputting multiple frames of Raw domain moire images into the model, and outputting a single-frame RGB image reconstruction result after moire removal; according to the method disclosed by the invention, the Moire removal performance of the Raw domain is improved to a new height by utilizing proposed selective scanning based on time and space.

The invention discloses a radar signal sorting method based on an improved long short-term memory network, and the method comprises the steps: extracting a pulse arrival time sequence, namely, a TOA sequence, from a pulse description word stream of a radar receiver; discretization processing is conducted on the TOA sequence through a preset fixed time window width wunit, the TOA sequence is mapped into a binary vector, each time window corresponds to one position in the vector, if a pulse arrives in the window, the corresponding position is set to be 1, and otherwise, the corresponding position is set to be 0; an improved ILSTM neural network is constructed, the network adopts an encoder-decoder architecture, and an encoder is used for performing high-level feature extraction on an input binary sequence and outputting a context feature vector; the decoder performs sequence prediction or reconstruction by using the context feature vector so as to output a sorting result; training the improved LSTM neural network by using a binary TOA vector data set of a known radiation source label, and optimizing network parameters; and after processing the to-be-sorted interleaved pulse TOA sequence, inputting the to-be-sorted interleaved pulse TOA sequence into the trained ILSTM network, and outputting a radiation source sorting result corresponding to each pulse by the network. The method is suitable for radar signal sorting in high-density, complex-modulation and strong-interference environments.

The invention discloses a multi-mode body intelligent agent track prediction method, which comprises the following steps of: receiving and processing input data, and standardizing dynamic and static contexts; an encoder-feature fusion device-decoder architecture model is constructed, the encoder maps high-dimensional features to a low-dimensional space and keeps key information, the feature fusion device fuses the features, the decoder extracts a trajectory mode probability, Gaussian noise is injected to enhance variation capture, and a multi-modal three-dimensional trajectory sequence is generated through GRU; and dynamically selecting an optimal path in combination with real-time environment feedback to complete prediction. Through standardization processing, multi-feature fusion and noise injection, prediction accuracy and environmental adaptability are improved, and efficient trajectory prediction is realized.

The invention provides a time series data interpolation method based on graph attention coding and a cross attention diffusion model. The core of the method is that an asymmetric encoder-decoder architecture is constructed, covariant feature representation is enhanced through a graph neural network, priori knowledge is injected, and then enhanced condition information is dynamically fused into each step of denoising process of a diffusion model by using a cross attention mechanism. Therefore, high-precision and high-robustness missing data interpolation is realized.

A high-speed parallel GRAND-CRC decoding method for functional safety communication is presented. M A round-robin noise error mode allocation scheme for each memory cell, without affecting error correction performance, provides... N The system adds an all-zero error mode to the set E of noise error modes and allocates noise error modes to each memory unit in turn. A high-speed parallel decoder architecture design scheme for GRAND-CRC is proposed, including a distributed memory design strategy and the advantage of unrestricted parallelism. The parallelism can be easily set according to hardware resources, and the decoder throughput increases linearly with the parallelism. The system uses a corrector as the basis for logical judgment, and the corrector signal processing unit triggers the correct decoding output. Its features include different processing methods in three cases, especially when contention occurs, selecting the guessed sequence corresponding to the corrector with the smallest sequence number as the decoding output.

The invention discloses a cluster aided decision planning method based on a large model, and the method comprises the steps: a commander carries out the unified deployment and control according to the order placing demands of a user, and gives out a natural language instruction; acquiring a natural language instruction sent by a commander, and converting the natural language instruction into a hyper-parameter combination in a standardized format based on a retrieval enhancement generation technology and a large language model; analyzing the hyper-parameter combination, and if a predefined path segment is empty, selecting a vehicle-node pair by adopting a station encoder-vehicle encoder-decoder architecture and an autoregression alternating strategy to generate a dynamic path planning scheme; and if the predefined path segment is not empty, redefining a decoder process based on the predefined path segment, and realizing path completion and optimization through a meta-action sequence, a flag matrix and a mask matrix. According to the method, dynamic path planning of any vehicle and node scale is realized, and meanwhile, quick response to a dynamic environment can be realized through real-time planning.

The application provides an image enhancement model training method and an image enhancement method, and relates to the technical field of image processing. Through an encoder-decoder architecture and supplemented by adaptive fusion skip connection, effective extraction and fusion of multi-scale features are realized. The encoder is responsible for multi-scale feature coding of the input image, gradually extracts the abstract features of the image and reduces the spatial resolution. The decoder is responsible for gradually restoring the image resolution and using the multi-scale features provided by the encoder for fine reconstruction, and finally outputs the enhanced image. While highlighting the local details of the ink droplets, the global contrast of the image and the statistical characteristics consistent with the physical reality are effectively maintained, and the distinguishability of the ink droplets and the background is significantly improved, providing a high-precision and high-quality image basis for subsequent quantitative analysis of the ink droplets.

This invention discloses a high-speed communication link transmitter modeling method based on a Transformer decoder. This method parameterizes the input signal and establishes a deep learning model with an encoder-decoder architecture, including a non-sequential encoder and a Transformer sequential decoder, to predict the output signal of the transmitter given the input signal. The encoder processes the input signal parameters, generates a context vector, and then passes it to the decoder. The decoder uses the context vector and the transmitter output signal sequence to generate a class probability distribution for each point in the sequence. The model is trained using a random masking strategy and employs non-autoregressive decoding and filtering techniques for inference, achieving parallel output sequence prediction. The final output is obtained through a single signal filtering step. Compared with traditional simulation techniques, this method significantly improves simulation speed while maintaining a very small error.

The invention belongs to the technical field of image recognition, and particularly relates to a deep learning-based ocean floating object recognition method and system, an asymmetric double-branch encoder-decoder architecture is adopted to extract feature maps, and multi-scale RGB and IR feature maps are sent to a subsequent adaptive weighted fusion module for scale-by-scale fusion, so that the feature maps of the deep learning-based ocean floating object recognition method and system are obtained. And the fused multi-scale features are subjected to feature enhancement and transmission through an improved feature pyramid network, the enhanced features are sent to a YOLO detection head, and the category probability, bounding box coordinates and confidence of the target are output in parallel.

The invention relates to the technical field of production line scheduling control, and particularly discloses a flexible production line arrangement scheduling and communication method by using an improved simulated annealing algorithm, which comprises the following steps: carrying out mathematical modeling simulation on production scheduling of an assembly line, and converting a multi-target flexible job shop scheduling problem into a mixed integer programming model; solving the mixed integer programming model by using an improved simulated annealing algorithm; a Transform model of an encoder-decoder architecture is adopted, multi-modal feature fusion is achieved through position encoding, wavelet transformation and type embedding, and a production scheduling scheme is obtained and used for assembly simulation of the discrete mixed flow assembly line; designing a Chisel-based tunnel communication architecture, and combining TCP long connection with TLS bidirectional authentication; through deep integration of multi-target mathematical modeling, a feature-driven prediction model and a high-reliability communication system, breakthrough improvement of dynamic process scheduling efficiency and equipment task allocation capability is realized.

The application discloses a multi-person redirection walking and position prediction method based on an artificial potential field, and belongs to the technical field of virtual reality and robot navigation. In view of the problems of fixed repulsion force weight, high collision risk, frequent resetting and short-term position prediction accuracy deficiency leading to collision in the traditional redirection walking algorithm in a multi-user dynamic scene, a dynamic potential field weight optimization mechanism is used to adjust the weight distribution of repulsion force between the environment and the user in real time according to the user density, to enhance the repulsion force intensity when the user approaches an obstacle or a dense area, and to introduce a safety distance and a direction perception mechanism, to analyze the user motion direction and the obstacle angle, to preferentially guide the deviation from a high-risk area, and to reduce the number of invalid resets. A prediction model adopts an encoder-decoder architecture, and a multi-layer long short-term memory network is embedded with an attention mechanism to dynamically capture key timing features in the space-time trajectory, and to improve the short-term position prediction accuracy.

The invention provides a phase compensation optical communication system based on multi-feature decomposition and convolution enhancement, and provides an optical signal phase disturbance compensation network HiLoPruneNet, the network is based on a SimVP encoder-translator-decoder architecture, multi-feature decomposition is embedded in an encoder, inter-domain convolution enhancement is added to a translator, and joint reconstruction is supplemented to a decoder. All convolutional layers in the whole model adopt a structure rarefaction process to realize high-precision prediction. A global-local joint module is adopted to perform multi-feature decomposition, an inter-domain fusion SeparableConv2d module and high-frequency and low-frequency convolution collaboratively enhance spatio-temporal features, and finally through joint reconstruction balance and structure rarefaction, redundant weights in a convolution layer are gradually reduced, so that the parameter quantity and calculation complexity of a model are optimized, and a future phase screen is output. According to the method, global trend and local disturbance details are considered, so that an effective solution is provided for optical communication in a complex environment.

The present application relates to the technical field of data processing, and more particularly to a time series prediction method for cloud server cluster load scheduling and electronic equipment, comprising: inputting a historical time series into a constructed PAPT model, the model adopting an encoder-decoder architecture, in the encoder, the historical time series is decomposed into a trend item and a periodic item; adaptive polynomial basis trend fitting is used for the trend item to capture trend characteristics, and after being combined with the periodic item, an attention layer integrated with periodic attention position coding is input, the periodic attention position coding drives a rotation angle through multiple periods, and the rotation angle is applied to each position of Q / K in attention, so that the attention calculation is integrated with the position dependence of multiple periods; in the decoder, the output of the attention layer integrated with periodic attention position coding is used for cross-attention calculation together with the output of the encoder, and finally the prediction result is output. Compared with the prior art, the present application has the advantages of realizing accurate prediction of general time series and the like.

In various examples, multimodal image data may be used to generate a set of top-down tile images, which are applied to a deep neural network generator architecture model to produce lane marking-specific heatmap images corresponding to the set of top-down tile images. The multimodal sensor data may include LIDAR-captured intensity channel data, LIDAR-captured feature height channel data, and optical color image channel data. The set of top-down tile images may be processed by the generator model to automatically detect lane boundaries and navigation boundaries to generate pixel-level heatmap images that may classify lane markings by marking characteristics such as line type and / or color. The generator model may comprise an encoder-decoder architecture, with multiscale feature extraction and / or context extraction functional layers intervening between the encoder model and the decoder model.

The invention discloses a video anomaly detection method based on multi-task learning, a model comprises a space-time encoder, a prototype memory network, a double-decoder architecture and a reconstruction decoder, and the method comprises the following steps: carrying out hierarchical feature extraction on a training video frame sequence by using the space-time encoder; performing memory enhancement processing on the spatial-temporal characteristics through a prototype memory network; respectively carrying out future frame prediction and current frame reconstruction through a double-decoder architecture; calculating a multi-task loss function, and updating parameters of the space-time encoder, the prototype memory network and the double decoders through back propagation according to the multi-task loss function; according to the method, a more comprehensive self-supervised learning framework is constructed, and the generalization ability of the model is enhanced.

The invention relates to a high-speed parallel GRAND-CRC decoding method for functional safety communication, which provides an alternate noise error mode allocation scheme of M storage units, adds an all-zero error mode to a set E of N noise error modes and allocates a noise error mode to each storage unit in turn under the condition of not influencing the error correction performance, a GRAND-CRC high-speed parallel decoder architecture design scheme is provided, the scheme comprises a distributed memory design strategy and the advantage that the parallelism value is not limited, the parallelism can be very conveniently set according to hardware resources, and the throughput rate of the decoder is linearly increased along with the parallelism; corrects are used as a logic judgment basis, correct decoding output is triggered by a syndrome signal processing unit, and the method is characterized in that different processing methods are adopted under three conditions, and particularly, a guess sequence corresponding to the syndrome with the minimum serial number is selected as decoding output when competition occurs.

Techniques are disclosed for denoising videos. In some embodiments, video frames are denoised using a denoising model that includes an encoder-decoder architecture and attention modules. During training of the denoising model, the attention modules learn weightings to upweight certain dimensions of input features to help pixel registration, remove ghosting artifacts, and improve temporal consistency when the frames of a video are being denoised. The denoising model can also be used to train a student denoising model that has a same architecture as, but is smaller and faster than, the denoising model. After training, noisy video frames can be input into the denoising model and / or the student denoising model to generate corresponding denoised video frames.

The invention discloses an evolutionary window attention self-adaption system for image segmentation. The evolutionary window attention self-adaption system comprises a super network module, an evolutionary architecture search module, a sub-network training module and an image segmentation execution module, the super network module is in communication connection with the evolutionary architecture search module, and the evolutionary architecture search module is in communication connection with the sub-network training module; the sub-network training module is in communication connection with the image segmentation execution module; and the super network module adopts an encoder-decoder architecture and comprises an encoder, a decoder and a multi-fusion hopping connection unit. According to the method, accurate segmentation of images with different modalities and different anatomical structures is realized, the accuracy, robustness and calculation efficiency of a segmentation result are improved, and clinical application requirements are met.

The invention provides a D2D node communication power distribution method based on lightweight Graph U-Net, and the method comprises the steps: firstly obtaining a physical link diagram and link information of a D2D communication network, and constructing a weighted directed graph; inputting the GU-Net into the trained GU-Net, mapping the GU-Net into a high-dimensional feature map, and performing down-sampling through a screening mechanism based on a node interference degree threshold value to reduce the calculation cost; performing graph coding to obtain a coding aggregation feature, and performing up-sampling and graph decoding to obtain a decoding aggregation feature; and finally, reasoning a power prediction value of each node based on the high-dimensional feature map and the decoding aggregation feature. According to the GU-Net, a lightweight encoder-decoder architecture is adopted, gradient propagation can be stabilized, efficient optimization is achieved, the GU-Net has the multi-scale interference modeling capacity, and local and global features can be captured at the same time. According to the method, the efficiency is remarkably improved while the performance is kept, and the method is adaptive to different network topologies and has good universality and deployability.

The application discloses a multispectral imaging satellite data space-time fusion method, medium, equipment and product, relates to the technical field of image processing, and comprises the following steps: adding noise to satellite data serving as a label through a diffusion process, and constructing a data pair of two kinds of satellite data; an encoder-decoder architecture based on a U-Net structure is used to construct a noise predictor; the data with noise is input into a noisy feature encoder for multi-scale feature extraction, and time step information of the diffusion process is embedded in the noisy feature encoder; the data pair is input into a clean feature encoder for feature fusion and multi-scale feature extraction; at each scale, difference features between the noisy features and the clean features are calculated, and the difference features are input into a decoder for restoration to obtain predicted noise data; and the data with noise is denoised to obtain clean data. The predicted noise can stably guide the reverse denoising process, and the quality and reliability of the finally generated image are improved.

The invention provides a multi-station and multi-device universal photovoltaic power station power prediction method. The method comprises the following steps: S000, constructing a prediction model; s100, performing unified domain transformation on the power data of the multiple power stations; s110, static characteristic embedding and global modulation of the power station are carried out; s120, constructing a multi-power-station universal encoder-multi-head decoder architecture; s130, multi-power-station data missing self-adaptive processing is carried out; s140, carrying out zero power section joint learning; s150, carrying out two-stage training and online calibration; and S160, prediction result optimization and application adaptation are carried out. According to the method, the deployment and operation and maintenance cost can be remarkably reduced, and large-scale operation is facilitated.