368 results about "Weight prediction" patented technology

The invention discloses a crop growth prediction method based on multi-source data fusion analysis, and the method comprises the steps: carrying out the standardization processing and missing value filling processing of multi-source data, constructing a time window, calculating the correlation between data sources, and generating a current data credibility vector based on historical credibility evaluation data; extracting features by adopting multi-scale decomposition, detecting abnormal points and analyzing abnormal degrees, and constructing an abnormal conduction network to evaluate conduction influence; performing feature importance evaluation and feature fusion; initial weight distribution is calculated, a weight prediction model is constructed, and consistency optimization is carried out on a weight matrix; performing hierarchical prediction by adopting a multi-level prediction model, and performing collaborative integration based on a prediction result and an optimized weight matrix; and calculating a prediction result deviation and carrying out optimization adjustment, and outputting a final prediction result. According to the invention, accurate prediction of the crop growth state is realized, and the reliability and practicability of the prediction result are improved.

The invention relates to the technical field of data processing, in particular to a multi-source target trajectory fusion method and system based on deep learning. Comprising the following steps: constructing a space-time alignment training set containing radar echo data, satellite telemetry data and truth value trajectory data; designing a multi-channel deep learning network architecture which comprises a radar echo data feature extraction channel and a satellite telemetry data feature extraction channel, and establishing dual-channel feature association through an attention mechanism; optimizing network parameters through back propagation to obtain a weight prediction model; radar echo data and satellite telemetry data are received in real time, and space-time alignment of sampling data is achieved through a sliding window mechanism and coordinate system conversion; inputting the preprocessed double-source features into the trained weight prediction model, and generating a radar-telemetering fusion weight coefficient; and performing weighting processing and optimization on the double-source trajectory coordinates, and outputting a fused target motion trajectory. The problem that a traditional single-source track measurement system is insufficient in precision in a complex environment is solved.

The invention discloses a power transformer temperature prediction method and system based on multi-feature extraction and fusion, and the method comprises the steps: carrying out the time-space decoupling standardization and causal emergence clustering of all temperature variables in data, and obtaining a clustering result; extracting a common pattern from each class of the clustering result to obtain an information complemented multivariable time sequence, and performing down-sampling on the information complemented multivariable time sequence to obtain a plurality of subsequences; performing feature extraction on the multivariable time sequence and the sub-sequence after information completion through a dynamic time-frequency domain MLP mechanism to obtain overall features and local features, and fusing the overall features and the local features to obtain time sequence features; acquiring an external environment covariable, inputting the external environment covariable into the self-attention mechanism network to obtain a covariable self-feature, and performing feature fusion on the covariable self-feature and the time sequence feature through a covariable-driven causal convolution fusion mechanism to obtain a fusion feature; and inputting the fusion feature into a dynamic attention feature weighted prediction model for prediction to obtain a prediction result. And the precision and reliability of temperature prediction of the power transformer are obviously improved.

Embodiments of this disclosure can provide a system and method for training a perception model to perform an autonomous driving task. During operation, the system can obtain labeled training data comprising images captured by multiple cameras mounted at different locations on a vehicle, and the perception model can generate, in parallel, a prediction output associated with the task and a confidence score based on the labeled training data. The confidence score can indicate a level of uncertainty associated with the prediction output. The system can generate an uncertainty-weighted prediction based on ground truth indicated by the labeled training data, the prediction output, and the confidence score; compute a loss function based on the uncertainty-weighted prediction; and update the perception model based on the loss function.

The invention relates to a sea cucumber growth character recognition and measurement method based on machine vision and a measurement system thereof, and belongs to the field of image processing, and the method comprises the following steps: S1, image acquisition and preprocessing; s2, instance segmentation and morphological feature extraction; s3, measuring the length and width of the sea cucumber; s4, pixel-to-actual size conversion is carried out; s5, constructing a body weight prediction model; and S6, outputting a result. The method has the advantages that the sea cucumber image is obtained by using the high-definition image acquisition technology, the contour information of the sea cucumber is accurately extracted through the instance segmentation algorithm, and the problems of changeable and irregular sea cucumber shapes and the like are effectively solved by combining the optimized image processing technology, so that the measurement precision is remarkably improved. Meanwhile, a machine learning regression model is introduced, the weight of the sea cucumber is predicted based on various morphological characteristics, and the dimensionality and the utilization value of measured data are further enriched.

The various implementations described herein include methods and systems for encoding and decoding video. For example, a computing system receives a video bitstream that includes a current block and a syntax element. The current block is encoded using information from a set of reference blocks. The system determines, based on the syntax element, whether the current block is encoded in a compound weighted prediction (CWP) mode. The system decodes the current block using a set of weighted prediction factors when the current block is encoded in the CWP mode. When the current block is not encoded in the CWP mode, the system decodes the current block using an implicit masked blending mode based on a weighted average of reference values associated with the set of reference blocks.

The embodiment of the invention discloses a dynamic computing power scheduling method and device based on deep learning and a medium, belongs to the technical field of cloud computing, and solves the problem that cloud computing resources are seriously wasted due to the fact that a cloud computing scheduling method in the prior art is prone to unbalanced resource allocation. Comprising the following steps: preprocessing acquired cloud computing data to generate a structured feature vector; wherein the cloud computing data at least comprises cloud node state data, computing task attributes and environment data; through a multi-head attention mechanism, task resource space-time correlation features corresponding to the structured feature vectors are extracted; the task resource space-time correlation features are input into a weight dynamic prediction module and a resource matching degree scoring module at the same time so as to carry out parallel processing of weight prediction and matching degree scoring calculation; and solving a multi-objective optimization function corresponding to computing power scheduling based on a parallel processing result, and generating a computing power scheduling instruction based on a solving result.

The invention relates to the technical field of data processing, in particular to a legal case relation graph dynamic updating method based on a neural network. Comprising the steps that original legal case data are collected and preprocessed to obtain target legal data; entities in the target legal data are extracted to serve as nodes, node sub-graphs are obtained through a machine learning model, weight parameters corresponding to the nodes are selected from weight parameter sub-graphs, weights of node edges are obtained based on the weight parameters and a weight prediction model, and a first graph is formed; acquiring newly added legal data, extracting new nodes, acquiring newly added sub-graphs through a machine learning model, and fusing the newly added sub-graphs with the first graph to obtain a second graph; and generating a search vector according to key weight parameters of adjacent nodes and edges in the second map, and embedding the search vector into the second map. According to the method, the problems that the legal case relationship is complex and cannot be dynamically updated are solved, the legal case relationship graph can be dynamically updated, and the accuracy and timeliness of the graph are improved.

The invention relates to a multi-modal data feature layer fused sow weight estimation method, which relates to a detection method, and comprises the following steps: acquiring point clouds of sows at different angles through a sow point cloud acquisition device, and taking sow time sequence data such as variety, fetal number and reproduction stage and reconstructed sow real-time three-dimensional point clouds as input; a dual-branch network fused in a feature layer is established, meanwhile, deep features are automatically extracted from one-dimensional data and three-dimensional point clouds, time sequence data features and real-time data features are fused, and current body weight estimation of the sow and body weight curve estimation of the whole breeding stage are completed. Through the method, the accuracy of weight estimation is improved, the final output of the method is the weight real-time value of the current reproduction stage and the weight prediction value of the whole reproduction stage, and the real-time weight value is beneficial for obtaining the physical health condition of the sow at the current stage and adjusting the feeding strategy; therefore, the sows can reach the optimal health state and maintain the optimal production performance, and the minimum feed cost is maintained.

The invention relates to the technical field of voice signals, in particular to a two-step mixed sound source separation and de-reverberation method, which comprises the following steps of: firstly, carrying out separation network training by taking different types of signals as training targets in various separation networks for separating attention and the like; the invention provides an improved de-reverberation method based on a time convolution network-weight prediction error, multiple improvement strategies such as taking a scale invariance signal-to-noise interference ratio as a network loss function, adopting a transposition mechanism for an input signal and a mechanism for additionally adding a residual value to a network unit are used, and finally, the de-reverberation method based on the time convolution network-weight prediction error is obtained. And cascading the separation attention network with the best separation noise reduction effect with the time convolution-transpose-residual error-WPE network. According to the invention, the problem that the separation effect of the mixed audio signal with reverberation and noise signals is not good under the actual sound field condition is solved, compared with a single one-step or two-step existing separation, de-mixing and noise reduction network, the method is significantly improved, and the quality of the separated voice can be improved for later recognition and discrimination.

The invention relates to a multi-level sentiment classification method for network public opinions. According to the method, social media data are acquired and processed in real time through a distributed message queue, online clustering is performed by using a Streaming K-means algorithm, and an initial topic set is generated. A topic state space is constructed through a topic state analysis model, a proper classification level is selected by using a Q-learning algorithm, and an information entropy data set is constructed through dependency path analysis and information entropy calculation. And finally, generating a multi-level sentiment classification result through the level weight prediction model and the sentiment classification model, and using the multi-level sentiment classification result for network public opinion monitoring. The method can adapt to changes of network public opinions in real time, and the accuracy and timeliness of sentiment classification are improved.

The invention relates to a resource elastic scaling decision-making method, system and device and a medium. The method comprises the following steps: collecting real-time operation data of a security service node, and performing multi-dimensional security index analysis according to the real-time operation data to obtain a portrait data packet; predicting the security service weight value to obtain a prediction result, performing dynamic error compensation on the prediction result to generate a corrected weight prediction value, and generating a control instruction based on the corrected weight prediction value and the active session state; and when the instruction is a migration instruction, analyzing a session state snapshot of the instruction, calling a preset kernel state locking function to lock a memory session block of a source node, obtaining incremental state change data to generate a migration snapshot packet, and performing block verification injection operation on a target node. According to the method, by integrating multi-dimensional safety index analysis, prediction error compensation and stateful transition verification mechanisms, the accuracy and response efficiency of resource elastic scaling decision making are improved, and the continuity of stateful service transition and the consistency of safety strategies are enhanced.

The present disclosure provides methods, apparatuses and devices for coding and decoding. When determining to enable a weighted prediction for a current block, the method includes: for the current block, obtaining a weighted prediction angle of the current block; for each pixel position of the current block, determining a surrounding matching position pointed by the pixel position from surrounding positions outside the current block based on the weighted prediction angle of the current block; determining a target weight value of the pixel position based on a reference weight value associated with the surrounding matching position; determining an association weight value of the pixel position based on the target weight value of the pixel position; determining a first prediction value of the pixel position based on a first prediction mode of the current block; determining a second prediction value of the pixel position based on a second prediction mode of the current block; and based on the first prediction value, the target weight value, the second prediction value and the association weight value, determining a weighted prediction value of the pixel position; and determining weighted prediction values of the current block based on the weighted prediction values of all pixel positions of the current block.

The invention relates to the technical field of distributed learning and wireless communication, in particular to an efficient communication federated learning model training method and system based on residual adaptive compression. The client obtains and updates the global model; obtaining a local model residual error by utilizing the local model memory set and the updated model weight; carrying out adaptive sparse compression on the residual error and transmitting the residual error to a server; the server selects a reliable client to carry out model weight prediction and aggregation; and broadcasting the updated global model to the client, and repeating the steps until convergence. Through adaptive compression and reliable client selection, the communication overhead is remarkably reduced, and the model convergence speed and the final performance are improved. The method is particularly suitable for edge computing scenes with limited resources, effectively solves the key problems of communication efficiency, model performance, privacy protection and the like in federated learning, and provides new possibility for wide application of federated learning.

The invention discloses an engine group fault mode identification method and device based on a density clustering-support vector machine and multi-moment prediction weighting and electronic equipment, and the method comprises the steps: carrying out the dimension reduction processing of training engine performance parameters and testing engine performance parameters based on principal component analysis (PCA); training engine low-dimensional performance parameters and testing engine low-dimensional performance parameters are obtained; performing fault mode clustering processing on the low-dimensional performance parameters of the training engines based on a density clustering DBSCAN algorithm in predicting the single-moment fault mode of the engine to obtain fault mode classification labels corresponding to the training engines; establishing a single-moment fault mode classification model based on a support vector machine (SVM), and training the SVM-based single-moment fault mode classification model by using the low-dimensional performance parameters of the training engines and the fault mode classification labels corresponding to the training engines to obtain a trained SVM-based single-moment fault mode classification model; inputting low-dimensional performance parameters of the test engines into the trained SVM-based single-moment fault mode classification model to obtain a single-moment fault mode classification label of each test engine; and obtaining a final fault mode classification label of each test engine based on multi-moment fault mode weighted prediction and the single-moment fault mode classification label of each test engine.

An MRI image reconstruction system (100) enhances MRI image quality and accuracy by integrating an MRI scanner (102) that maps under-sampled k-space measurements to fully sampled MRI image (106) A deep learning model (108), controlled by a processor (110), drives the reconstruction. This model includes a dynamic weight prediction auxiliary network (112) and a base reconstruction convolutional neural network (118). The auxiliary network (112) processes acquisition context (114) to create context-specific weights (116), which the convolutional network (118) uses to process the under-sampled data (106). This process generates a context-specific fully-sampled MRI image. The final reconstructed MRI image (124) is obtained by utilizing the scanner settings or the acquisition context information together with the acquired under-sampled k-space (104)

The invention discloses an ultrahigh-definition panoramic image adaptive HDR fusion method based on data driving, and relates to the technical field of digital image processing, and the method comprises the steps: carrying out the coarse alignment and high dynamic range fusion of a low-resolution line graph sequence, and generating a low-resolution HDR panoramic image; carrying out sharpening processing on the low-resolution HDR panoramic image to obtain a low-resolution sharpened HDR image, and meanwhile, obtaining a multi-scale residual spectrum generated by sharpening; performing up-sampling on the multi-scale residual spectrum, and performing small-sample element learning fine tuning on the weight prediction network in combination with the low-resolution sharpened HDR graph to obtain a fine-tuned weight prediction network; performing weight prediction and layer-by-layer fusion on the registered high-resolution tile sequence on a multi-scale pyramid based on a fine tuning weight prediction network model; according to the method, the sharpening processing is implemented on the low-resolution HDR panorama, and the multi-scale residual spectrum is extracted, so that the detail sensitivity and adaptability in a complex scene are improved.

The invention discloses an AI and algorithm fused intelligent geological layering system and method for geotechnical engineering investigation, and the method comprises the steps: data importing and preprocessing, data verification and gross error elimination, automatic classification of related / non-related variables, cleaning and iterative updating of layer mean values and boundaries, and hierarchical division. According to the method, the data is automatically cleaned, the layer group is verified, the main layer / sub-layer / sub-layer / interbed is divided, the layering accuracy and efficiency are improved, the artificial intelligence technology is applied, and the layering efficiency is improved. Technicians are replaced to complete the heavy work flow of drawing grass sections, manually layering, inputting soil layer information into a computer, manually judging the rationality of in-situ test, geotechnical and rock test data and layering, modifying sketches, secondarily proofreading the rationality of results and completing division, and the exploration efficiency is improved.

The invention discloses a method and equipment for evaluating the meat yield of Bellamya piriforme, a medium and a product. The method comprises the following steps: acquiring a morphological data set of the Bellamya piriformis; the morphological data set of the Bellamya piriformis comprises character parameters of a plurality of sample Bellamya piriformis individuals and corresponding soft body weights; constructing an optimal KNN prediction model based on the morphological data set of the Bellamya piriformis; the optimal KNN prediction model is used for predicting the software weight according to the character parameters of the individual Bellamya piriformis; inputting the character parameters of the target Bellamya piriformis individual into the optimal KNN prediction model to obtain a soft weight prediction value of the target Bellamya piriformis individual; and dividing the soft body weight prediction value of the target Bellamya piriforme individual by the body weight to obtain the meat percentage prediction value of the target Bellamya piriforme individual. According to the method, nondestructive evaluation of the meat percentage of the Bellamya piriforme can be realized, and the meat percentage of the Bellamya piriforme can be predicted under the condition that the Bellamya piriforme survives.

The invention provides a multi-modal pig weight measurement method and system based on attention fusion, and the method comprises the steps: S1, data collection: employing a depth camera to synchronously obtain an RGB image and a depth image of a pig, and converting the depth image into point cloud data; s2, multi-modal feature extraction: extracting 2D features from the RGB image, and extracting 3D features from the point cloud data; s3, feature fusion: performing cross-modal feature fusion based on an attention mechanism on the extracted 2D features and 3D features to obtain attention fusion features; and S4, MLP regression prediction: inputting the attention fusion features into an MLP weight regression prediction model, and predicting the weight of the pig through the MLP weight regression prediction model. According to the method, the 2D features and the 3D features are fused, so that the adaptability of the system in a complex scene is enhanced, and the accuracy of weight prediction can be greatly improved.

The invention relates to the field of data processing, in particular to a concentrator error calibration method based on machine learning, which is used for calibrating errors of active electric energy and reactive electric energy by collecting and preprocessing related data of a concentrator and combining a random forest regression model. Firstly, a model is established by using historical data, an electric energy fluctuation time period is identified, and the influence of different characteristics on electric energy is determined by calculating the correlation degree of load power and electric energy. And adjusting the weight of each decision tree according to the trend correlation between each feature and the electric energy, and obtaining a weighted prediction value. And finally, calculating an electric energy error and carrying out dynamic adjustment based on a correction coefficient so as to realize real-time error calibration. According to the method, the electric energy error of the intelligent concentrator is automatically calibrated through the random forest algorithm in combination with feature correlation analysis and an error correction model.

Systems, methods, and apparatus for video processing are described that include weighted prediction for video blocks. An example method includes: performing conversion between a current slice of a current picture of a video and a bitstream of the video, wherein the bitstream conforms to a format rule, and wherein the format rule specifies the presence of a general constraint information syntax structure that includes one or more constraint flags indicating that constraints on explicit weighted prediction are enabled for slices of a set of pictures.

The invention discloses a logistics return transportation capacity intelligent matching system and method based on journey prediction, and the method comprises the steps: carrying out the transportation capacity departure time window prediction, journey time consumption prediction, space correction prediction, sudden factor correction prediction and bearable weight prediction of the multi-source heterogeneous data of each vehicle in the current collection period; obtaining a transport capacity departure time window, an optimal return path, return time consumption and an estimated loadable weight of the target vehicle; and generating a transport capacity label based on the transport capacity departure time window, the optimal return path, the return time consumption and the estimated bearable weight of the target vehicle, performing multi-dimensional adaptation on the transport capacity label and the order information to obtain a comprehensive score of the transport capacity label, and performing order matching and state updating operation of the target vehicle based on the comprehensive score of the transport capacity label. Accurate pre-judgment of the transport capacity state is achieved, and the invalid matching rate is greatly reduced.

The invention relates to the technical field of iron and steel industry, in particular to an internet of things weighing monitoring method and system based on the iron and steel industry, and the method comprises the following steps: continuously collecting the weight information of steel materials based on a weight sensor installed on a conveyor belt, recording the weight value and the corresponding timestamp of each material in real time, and removing the noise and abnormal values in the data; and obtaining a cleaned weight data set. According to the invention, the real-time change of the weight of the material is accurately monitored through the time sequence analysis and anomaly detection technology, so that the iron and steel industry can accurately control the use of the material, unnecessary waste is avoided, the real-time performance and accuracy of data processing are improved, production personnel are allowed to discover production deviation in real time through continuously updated weight prediction information, and the production efficiency is improved. According to the method, rapid intervention, rapid identification of abnormal weight points and reason analysis are carried out, material loss caused by misoperation or equipment faults can be greatly reduced, and the continuity of the production process is further ensured by adjusting production parameters in real time.

The invention belongs to the field of robot control, and provides a multi-view feature fusion operable part semantic segmentation method and system, and the method comprises the steps: obtaining the point cloud data of an operable part, and generating a corresponding multi-view image; carrying out target detection, extracting features for each view angle, and obtaining a two-dimensional bounding box and a semantic tag; based on the extracted two-dimensional bounding box, processing by using SAM to obtain a foreground mask; on the basis of the extracted two-dimensional bounding box, the relevance of the same semantic target under different visual angles is captured on the global scale by using the constructed global visual angle interaction module, and the feature consistency is enhanced; and processing the fused bounding box features obtained by the global view angle interaction module by using a weight prediction network, predicting the response weight of each bounding box at each super point, combining the obtained foreground mask and the predicted response weight, and back-projecting to a 3D point cloud space through view point information to obtain a final 3D semantic segmentation result. According to the invention, the segmentation accuracy is improved.

The invention discloses a method for realizing foreground multi-modal detection based on adaptive dynamic weight fusion, which comprises the steps of multi-modal data acquisition, multi-modal data preprocessing, multi-modal feature extraction, multi-modal feature fusion, foreground detection and prediction result post-processing. The multi-modal feature fusion step comprises the following steps: a) respectively inputting the extracted RGB features and IR features into a weight prediction sub-network, wherein the sub-network generates a spatial weight map for each modal; each pixel value in the weight map represents the local confidence degree of the mode for foreground discrimination at the corresponding spatial position; b) normalizing the spatial weight maps corresponding to the modalities to enable the sum of the weights of the modalities at the same spatial position to be 1; and c) multiplying the normalized spatial weight map by the corresponding modal features element by element, and adding the modal weighted feature maps element by element to obtain a final comprehensive feature map for foreground detection. According to the method, the foreground detection precision and robustness are remarkably improved.

The invention discloses a future pork pig group weight estimation method and system based on multi-model fusion, and the method comprises the steps: obtaining a historical structural data set of a pork pig breeding complete period, and employing a multi-modal fusion evaluation method of field priori knowledge weighted evaluation, data-driven feature importance calculation and an automatic code analysis tool. Extracting a key factor set influencing the weight of the pork pig, then based on the historical structured data set and the key factor set, respectively constructing an expert rule reasoning model, a data-driven prediction model and a code generation type dynamic model to predict the weight of the pork pig, inputting each model based on real-time data, and outputting a prediction value; weights are dynamically distributed according to historical errors, results are fused, abnormal values are removed in combination with discrete degree detection, and a final fusion prediction value is generated. The technical problem of improving the weight estimation accuracy of the pork pig group in the prior art is solved.

The invention belongs to the technical field of health management, and particularly relates to a body weight prediction system based on multi-modal food data. The body weight prediction system comprises a unified diet feature learning module based on CLIP, a time sequence prediction model and a diet-body weight loss calculation module. According to the invention, the influence of diet intake on weight prediction is explored, and the change trend of future weight is predicted by using short-term weight change and diet records. According to the method, unified features of historical food data are extracted through a diet feature learning module, and then the features and a historical weight change sequence are spliced to serve as input of a time sequence prediction model; the relation between diet intake and weight change is established through a diet-weight loss training diet feature learning module and a time sequence prediction model. The system is high in accuracy and adaptability for weight prediction; the method can be applied to the fields of health and medical treatment for personalized health management.

The invention discloses a bicubic interpolation image super-resolution reconstruction method based on dynamic weight prediction, which relates to the technical field of image super-resolution, and is characterized in that a lightweight convolutional neural network is trained through an improved bicubic interpolation algorithm based on dynamic weight parameter adjustment to obtain a dynamic super-resolution model to realize image super-resolution reconstruction; according to the dynamic super-resolution model, a bicubic interpolation algorithm is improved by establishing a dynamic weight adjustment mechanism based on local contrast, so that an optimized prediction global weight matrix is obtained, contribution of a high-contrast direction is enhanced in an original low-resolution marginal region, noise is suppressed in a flat region, and the dynamic super-resolution model is obtained. And keeping the balance smoothness and details of the texture region. The invention provides a bicubic interpolation image super-division reconstruction method based on dynamic weight prediction, which converts a high-dimensional pixel generation task into a low-dimensional weight optimization problem and remarkably compresses the complexity of a model.

The invention belongs to the technical field of cigarette primary processing, and discloses a tobacco cutting section tobacco leaf cumulative weight prediction method based on machine learning, which comprises the following steps: during batch production of a tobacco cutting section, inputting factors related to tobacco leaf cumulant, climate related factors and production time period variables into a trained online prediction tobacco leaf cumulant linear regression model; the online prediction tobacco leaf cumulant linear regression model outputs the real-time online prediction belt weigher cumulant; after batch production of the tobacco cutting section is finished, factors related to the tobacco cumulant, climate related factors and production time period variables are input into the trained off-line prediction tobacco cumulant linear regression model, and the off-line prediction tobacco cumulant linear regression model outputs the off-line prediction belt scale cumulant. According to the invention, high-precision prediction of the tobacco leaf cumulative weight in the shredding section is realized.