54results about How to "Reduce memory consumption" patented technology

The invention discloses a fan unmanned aerial vehicle autonomous inspection method and system, and the method comprises the steps: building a fan three-dimensional model based on the collection data of a laser radar for a fan, and enabling the fan three-dimensional model to be constructed through a recurrent neural network; planning an inspection route of the unmanned aerial vehicle; setting an intelligent inspection APP, and determining an autonomous flight path of the unmanned aerial vehicle; automatically identifying the to-be-detected image through the fan defect detection module, determining the fan defect type and positioning the fan defect position; and counting the defect positions of the fan and outputting an analysis inspection report to finish inspection. According to the routing inspection method, the laser radar, the edge calculation module and the like are integrated, the hardware requirement of fan routing inspection is met, and unmanned routing inspection work of the wind driven generator is greatly facilitated.

The invention relates to a GPU-based high-IO perceptibility long-sequence text injection method, medium and equipment, and the method comprises the steps: obtaining and coding a long-sequence text input, processing a coded matrix vector through a multi-head attention mechanism, and realizing text injection, the matrix vector comprises a query Q, a key K and a value V. The method is characterized in that the value V is equal to the query Q, the key K and the value V. In the multi-head attention mechanism, the similarity is calculated by using an L2 distance, the matrix vector is segmented into blocks to realize thread block parallel processing during forward propagation, and the similarity-based result matrix and the gradients of the query Q, the key K and the value V are recalculated by using a recalculation strategy during back propagation. Compared with the prior art, the method has the advantages of reducing the overhead of attention calculation memory access, improving the operation efficiency and the like.

The invention relates to an operation log recording method, device and equipment based on an SPI mechanism. The method comprises the steps that a loading operation log package is started, and log variable parameters are read; starting a log component to form log information, and determining a log dropping mode according to the number of logs to be processed; scanning, loading and defining an SPI function, and dropping the to-be-processed log into a disk according to the log dropping mode; a monitor is initialized, and whether the system runs normally or not is judged through the monitor. At least code readability and maintainability can be improved, memory consumption is reduced, performance is improved, and compatibility is improved.

The invention discloses a novel power system output regulation and control strategy generation method and device and a control system, and belongs to the technical field of power system regulation and control. The output regulation and control strategy generation method comprises the following steps: representing net load historical data corresponding to net load climbing demand uncertainty characteristics by adopting a Gaussian mixture model; the uncertainty of the net load climbing requirement is reduced; inputting the net load day-ahead prediction data corresponding to the net load historical data and the net load uncertainty scene set into a two-stage risk avoidance model based on the lowest operation cost and the worst condition risk probability assessment theory; the two-stage risk avoidance model divides the whole strategy generation process into day-ahead optimization scheduling plan generation and intra-day real-time scheduling plan generation; the day-ahead optimal scheduling plan is optimized by using the obtained intra-day real-time scheduling plan to obtain a safe and economic output scheduling strategy; and furthermore, the novel power system is controlled in different scenes to realize efficient, stable and economical operation.

The invention discloses a traveling salesman problem solving method based on a quantum self-attention neural network, and belongs to the field of quantum computing. A classic encoder-decoder structure is adopted, a quantum self-attention layer is designed in the encoder and comprises a quantum multi-head attention layer and a quantum full-connection feed-forward layer, and a variable component sub-circuit is used in a sub-layer to replace a full-connection layer of a classic neural network. Firstly, an encoder receives an input data set, node features are embedded into a quantum self-attention layer, and a complex relation between nodes is extracted through a quantum multi-head attention mechanism. The decoder then stepwise outputs the sequence of nodes to form an optimal path according to the embedding generated by the encoder. In the model training stage, a REINFORCE deep reinforcement learning algorithm is utilized, the total path length serves as an optimization target, and model parameters are updated through strategy gradients. Compared with the existing algorithm, the method has the advantage that the performance is well improved.

The invention discloses a data association relationship determination method, system and device, a medium and a program product applied to an artificial intelligence technology, and the method comprises the steps: taking to-be-associated data of a data association task and all source data in associated data information as node features, and inputting the node features into an association relationship recognition model; performing graph sampling on the incidence relation identification model, aggregating feature information of each sampling node and adjacent neighbor nodes based on a target polynomial, and determining an incidence relation identification result between the sampling nodes corresponding to the current graph sampling according to the current aggregation feature; and determining target data with an association relationship in the association data information for the to-be-associated data according to each association relationship identification result. The problem that task precision and task execution efficiency cannot be considered at the same time in the prior art can be solved, and on the basis of improving the task execution efficiency determined by the data association relationship, it is ensured that a high-precision data association result is output.

The invention discloses a method and equipment for reconstructing a three-dimensional scene, and relates to the field of computer graphics. The method comprises the following steps: simulating the process of light from the optical center of a camera to a three-dimensional scene through a light advancing module, and determining space coordinates of a plurality of sampling points in the three-dimensional scene; decomposing the color feature space and the density feature space, and determining a matrix and a vector pair corresponding to the space coordinates of each sampling point; interpolating each matrix and vector pair, and determining a color feature vector and density of each sampling point; determining a spherical harmonic feature vector of each sampling point according to the color feature vector of each sampling point and a spherical harmonic code of surface modeling; determining an anti-aliasing feature of each sampling point according to the color feature vector and the cone code of each sampling point; splicing the spherical harmonic feature vectors and the anti-aliasing features, and inputting a predicted color of each sampling point output by a target multi-layer perceptron; and carrying out volume rendering on the predicted color and density to determine the reconstructed three-dimensional scene, and reconstructing the three-dimensional scene with high efficiency and high quality.

The application discloses a low-depth sequencing population genotype filling calculation memory optimization method, which comprises the following steps: carrying out block division on a single nucleotide polymorphism site (SNP), and setting a checkpoint value according to the block division; and carrying out calculation on a forward auxiliary variable and a backward auxiliary variable according to the set checkpoint value, and calculating a haplotype observation conditional probability according to the block division. The application has the advantages of simple principle, effective reduction of calculation memory consumption, improvement of genotype filling efficiency and the like.

The invention discloses a lightweight fire hazard multi-target detection method in an electric power environment. The method comprises the following steps: S1, constructing a lightweight fire hazard multi-target detection data set in the electric power environment; s2, a preset RT-DETR model is improved, and an improved RT-DETR model is obtained; s3, training the improved RT-DETR model by using the data set, and optimizing model parameters through an improved loss function to obtain a lightweight fire hazard multi-target detection model in the electric power environment; and S4, performing multi-target detection by using the lightweight fire hazard multi-target detection model in the electric power environment. According to the invention, high-precision disaster and hidden danger multi-target detection is realized.

This invention discloses an adaptive filtering method based on the multi-kernel Nystrom method, belonging to the field of kernel adaptive filtering. The multi-kernel Nystrom method combines multiple kernel functions, which can include not only Gaussian kernels but also other different kernel functions, mapping the original data to multiple different independent feature spaces. Compared with the traditional single-kernel Nystrom method, the multi-kernel Nystrom method has better filtering accuracy and becomes less sensitive to the choice of kernel parameters. Compared with other multi-kernel adaptive filtering methods, it can significantly reduce computational complexity, time, and memory consumption. Subsequently, this invention introduces the proposed multi-kernel Nystrom method into kernel adaptive filters for the first time. Experiments show that this algorithm can achieve better filtering accuracy than current state-of-the-art kernel adaptive filtering algorithms with relatively low computational complexity.

The RWKV-based novel translation model inference method and system includes the following steps: 1) Collecting novel translations and extracting parallel corpora, and using dynamic MicroBatch concatenation technology for sequence compression; 2) Introducing a lightweight group query attention mechanism on the basis of the RWKV architecture, directly obtaining KV information from the Embedding layer to build the model; 3) Employing a sublinear complexity hybrid parallel training mode, combined with a global scalar scaling FP16 mixed precision strategy for training; 4) Applying a hierarchical distributed heterogeneous architecture, offloading the optimizer to low-performance devices and performing gradient compression transmission; 5) Outputting the translation using a joint decoder and dynamic batch inference technology. This invention, through the above method and system, effectively reduces the computational complexity and memory usage in the long text translation process, improves model training efficiency and inference throughput, and significantly improves the translation efficiency and contextual coherence of ultra-long texts.

The application discloses a power maintenance action prediction method based on target consistency screening and a bidirectional state space, first constructs an observation side video feature sequence, then carries out action semantic token coding, probability weighted mapping and time sequence modeling, obtains action semantic context representation, constructs a joint input vector based on the action semantic context representation, carries out time sequence modeling and distribution construction through a gate recurrent unit, obtains a feature side target distribution, and extracts an observation action representation; subsequently, a future action prediction model based on a bidirectional selective state space is used for prediction, and multiple candidate future action sequences are output; finally, an action side posterior distribution is constructed, a target consistency score is calculated, and the candidate future action sequence with the minimum score is selected as an optimal action prediction result. The application explicitly models statistical correlation and potential target distribution of maintenance actions, introduces a bidirectional selective state space diffusion generation architecture, and realizes accurate and logical prediction of power maintenance actions.

The embodiment of the invention provides a large result set suppression method and system, and belongs to the technical field of data query. The method comprises the following steps: acquiring a data query request of a user in response to a data query trigger signal of the user; based on the data query request, performing structured query language statement object recognition in a corresponding filtering event adapter; applying a predetermined maximum row value to the identified structured query language statement object, and executing structured query language statement object transmission applying the predetermined maximum row value; and returning the result data of the corresponding number based on the predetermined maximum row value. According to the scheme, a large result set generated due to unreasonable query conditions is avoided, the memory consumption and the garbage collection frequency are reduced, and the application stability is improved.

The present invention provides an object rendering method and device, a storage medium, and an electronic device, wherein the method comprises: allocating a rendering texture space in real time in a video memory, wherein the rendering texture space is used to store a plurality of rendering textures, each rendering texture corresponding to a display resolution; obtaining camera parameters of a virtual camera in a current virtual scene, wherein the virtual scene includes a scene object; selecting a target rendering texture in the rendering texture space according to the camera parameters; and rendering the scene object to a display screen using the target rendering texture. The present invention solves the technical problem of high resource overhead in the related art when performing 3D image rendering, reduces memory consumption in the 3D rendering process, improves the utilization efficiency of hardware resources, and reduces resource overhead.

The application discloses a point pair registration method and system based on feature guidance and double constraints, and belongs to the technical field of computer vision and robot grabbing. The method aims to solve the problems of recognition failure and insufficient estimation accuracy of traditional methods when parts are blocked and stacked, and point cloud features are not obvious. The scheme adopts a two-stage optimization strategy. Firstly, in the coarse registration stage, an adaptive sampling PPF algorithm guided by geometric features is proposed. By preferentially sampling points in the edge, curvature and other feature obvious areas, the robustness and efficiency of matching are improved. Secondly, in the fine registration stage, an ICP algorithm with multi-constraint optimization is proposed. The pose result obtained by coarse registration is used as the initial value. Through rotation and angle constraint, a multi-objective optimization function is constructed to guide the fine registration process to converge in the correct direction and avoid falling into local optimum, thereby ensuring the accuracy and physical feasibility of the final pose estimation. The application significantly improves the accuracy of industrial part pose recognition.

This invention discloses a brain-inspired optimization method for sentence structure-based self-attention mechanisms, comprising the following steps: acquiring a large-scale unlabeled corpus, preprocessing the corpus text, and identifying sentence boundaries; constructing structured attention constraints based on sentence boundaries: constructing intra-sentence attention constraints for ordinary characters to allow direct attention computation between characters within a sentence, constructing intra-sentence compression constraints for sentence boundary positions to form sentence-level compressed representations, and constructing cross-sentence constraints to allow both ordinary characters and boundary positions in the current sentence to access the boundary positions of the previous sentence; applying the structured attention constraints to the self-attention modules of each layer of a pre-trained language model for training or fine-tuning based on the corpus; and performing performance testing and brain-likeness analysis on the optimized model. This invention reduces the computational scale and memory consumption of the attention matrix while retaining long-range information integration capabilities, thus lowering computational complexity.

This invention discloses a large-scale power grid simulation method, program product, and storage medium based on algebraic multigrid optimization. The method includes: parsing the netlist of a large-scale power grid to establish a power grid topology matrix; extracting global feature vectors based on the netlist and the power grid topology matrix; compressing the power grid topology matrix to obtain a compressed feature matrix; constructing and training a regression model, which is used to predict the solution time based on the compressed feature matrix, global feature vectors, and coarsening parameters; using the minimum output of the regression model as the objective function, and employing simulated annealing to find the optimal coarsening parameters; establishing a power grid simulation system Ax = b; and solving the power grid simulation system using the algebraic multigrid method with the optimal coarsening parameters to obtain the power grid unknown vectors, thus realizing large-scale power grid simulation. This invention improves solution efficiency and reduces memory consumption while ensuring accuracy.

The present invention discloses a method and device for adding flame special effects to a picture, so as to solve the problem of how to reduce the consumption of hardware resources while accurately expressing the visual effects of the surface of an object. The scheme of the present application includes: obtaining a three-dimensional model to which flame special effects are to be added; setting at least one flame source on at least part of the surface of the three-dimensional model; projecting the three-dimensional model including at least one flame source to a target special effects plane; generating a flame special effects layer based on a preset flame simulation algorithm and at least one flame source in the target special effects plane; overlaying and rendering the flame special effects layer with the three-dimensional model to obtain a picture after adding flame special effects. The scheme of the embodiment of the present invention projects the three-dimensional model onto a two-dimensional image for flame special effects processing, which can effectively reduce the resources consumed by image processing, and projects the three-dimensional image onto a two-dimensional plane for processing to achieve data dimensionality reduction, which is conducive to the addition of real-time special effects.

The application discloses a large model inference method and device based on quantized perception fine-tuning and a medium. The method comprises the following steps: performing structural modification on an original parameter matrix of a large model, determining a horizontal vectorization vector, a vertical vectorization vector and a low-bit fixed matrix corresponding to each channel original parameter matrix of the large model; performing parameter quantization fine-tuning pre-training on the large model layer by layer based on the horizontal vectorization vector, the vertical vectorization vector and the low-bit fixed matrix, and obtaining horizontal vectorization vector values and vertical vectorization vector values of each channel of the large model; determining a deployment parameter matrix of the large model for deployment and inference according to the horizontal vectorization vector values, the vertical vectorization vector values and the low-bit fixed matrix of each channel of the large model; and performing inference analysis on input data by using the large model deployed with the deployment parameter matrix to obtain an inference result of the input data.

The invention relates to a page display method and device, electronic equipment and a storage medium. The method comprises the following steps: determining a notification message corresponding to an operation instruction on an application layer by using a document object model service in response to the detected operation instruction on a page element in a quick application page; wherein the notification message carries a to-be-rendered document object node identifier and event information corresponding to the to-be-rendered document object node, and the to-be-rendered document object node at least comprises a node corresponding to a page element operated by the operation instruction; sending a notification message to a local service layer; rendering the node associated with the document object node identifier through the local service layer according to the event information in the notification message; and displaying the fast application page rendered by the node. Through the method, the memory occupation of the application framework layer is reduced, and the performance of the fast application is improved.

The application discloses an acquisition method and device for frequent binomial sets, terminal equipment and a storage medium. The number of occurrences of an element of a data set is continuously detected and counted. If the count value of the number of occurrences of the element reaches a preset threshold value, the element is counted into a partial frequent one-item set. The preset threshold value is the minimum support degree set by a user according to actual needs. Item pairs formed by the data set according to the partial frequent one-item set are screened. Hash operation is performed on the screened item pairs. The frequent binomial set is acquired according to the result of the hash operation and the preset threshold value. Thus, the acquisition method for frequent binomial sets provided by the application improves the PCY-Multistage algorithm. Hash operation is only performed on item pairs that meet the partial frequent one-item set, thereby reducing the device memory consumption when calculating the frequent binomial set and improving the user experience.

The application discloses an intelligent garbage classification method and system based on an ARM architecture, and belongs to the technical field of image processing.The application embeds an AlexNet neural network into the ARM architecture, and is used for garbage classification and identification, so that the requirements of high accuracy, short identification time and model updatable are met.Meanwhile, online knowledge distillation technology is used for category incremental learning, so that the performance of the embedded AlexNet network is further improved, and the forgetting problem caused by the incremental learning is avoided.In addition, a plurality of characteristic data are collected through an external multi-sensor, and a self-adaptive weighted fusion algorithm is used, wherein the weight uses a Q-learning algorithm to avoid misjudgment caused by hidden image information and unobvious image characteristics, and the accuracy and generalization ability of garbage classification are improved.

The present invention discloses a commodity inventory method based on visual recognition, comprising the following steps: an inventory personnel moves an inventory device along the commodity display position, and places the commodities to be counted on the equipment scale body one by one; after the weight value is stable, it is determined as the weight value of the commodity, and then the image recognition unit collects the image of the weighed commodity, and transmits the image of the commodity to the identifier for recognition, and obtains the category of the weighed commodity; after the weight or quantity of the commodity is confirmed, the inventory data of the commodity is formed; after the inventory personnel complete the inventory of all commodities, if necessary, the review personnel check the difference between the book inventory and the actual inventory, and if there is an error, the inventory is adjusted, and then the result is formed to complete the inventory. The solution provided by the present invention optimizes the process of inventory operations, reduces the number of people involved in inventory, improves the efficiency and accuracy of inventory, and saves the operating costs of business units.

This application relates to a map construction method. Addressing the technical problem that existing environmental occupancy map construction methods, while aiming to improve query and update efficiency and significantly reduce memory usage, struggle to support real-time, large-scale autonomous navigation tasks in dynamic environments, this application provides a boundary representation-based map construction method and related apparatus. The map employs a global-local map framework. The local map includes the occupancy status of surrounding voxels centered on the robot, while the global map includes the voxel occupancy status in the boundary graph. The voxel occupancy status in the boundary graph is determined by using adjacent voxel pairs to represent the boundary surface layer separating free voxels from adjacent voxels, thus obtaining boundary voxels and constructing the boundary graph. A data structure combining two-dimensional hashing and one-dimensional arrays is used to manage the boundary voxels. Under this data structure, the occupancy status of any voxel is determined through the boundary graph. This application effectively improves query and update efficiency and significantly reduces memory usage.

The present disclosure provides a user abnormal behavior detection method and system that fuses data enhancement and dynamic graphs, relates to the technical field of deep learning anomaly detection, and includes: obtaining original behavior data of a user; inputting preprocessed original behavior data into a conditional variational autoencoder model, using the conditional variational autoencoder to extract latent space variables in a time series of the original behavior data, using a decoder to convert the latent space variables back to reconstructed values of the original time series, generating new time series data, and fusing the new time series data with the original behavior data to obtain enhanced data; inputting the enhanced data into a dynamic graph model after converting the enhanced data into graph structure data, using a space-time separation attention mechanism in the dynamic graph model to perform attention calculation on decoupled time and space dimensions, obtaining a feature graph representation, determining abnormal behavior by calculating the cosine similarity of the feature graph representation through a KNN algorithm, and obtaining an abnormal behavior determination result.

This invention relates to a method and system for identifying and repairing holes in geometric models for fluid simulation, belonging to the field of fluid simulation technology. It solves the problems of unfriendly operation and low repair efficiency in existing technologies. The method includes: acquiring the geometric model of the fluid simulation and mapping the geometric model to obtain model data; receiving a hole identification command, identifying holes based on the model data, acquiring a set of hole data, and displaying the number of holes, a positioning button, and a repair button in the hole repair attribute panel; receiving a command for any positioning button, displaying the hole in the geometric model in the view area based on the hole corresponding to the positioning button; receiving a command for any repair button, repairing the corresponding hole based on the repair button, and updating the model data. This achieves rapid identification and repair of holes in the geometric model.

The application provides a global data reconstruction method and device based on position mapping and a storage medium, and relates to the technical field of data processing. The method comprises the following steps: for each source data in at least two source data, a mapping relationship between global data and the source data is respectively constructed, and the mapping relationship is stored as index data independent of the source data, wherein the index data corresponding to different source data is different; in response to a reconstruction instruction, the positions of target data segments corresponding to each data point in the global data in the source data are determined based on the index data; and the target data segments are extracted from the source data for fusion to generate the global data. The application realizes the decoupling of the mapping logic and the source data, avoids the residence of full data in the memory, significantly reduces the memory consumption, and improves the calculation efficiency and data reusability.

The invention belongs to the field of computer vision, and relates to an automatic segmentation method and system of a three-dimensional digital dental model based on super points. The method comprises the following steps: carrying out oral cavity three-dimensional scanning through a three-dimensional oral cavity scanner to obtain a three-dimensional dental model; inputting the three-dimensional tooth jaw model into a three-dimensional tooth jaw model automatic segmentation network based on super points to obtain an FDI label of each point of the three-dimensional tooth jaw model; and according to the FDI label of the three-dimensional dental model and the label color mapping table, performing color assignment visualization processing on each point of the three-dimensional dental model, and outputting to obtain a final segmentation model. According to the method, the local geometric information and the global semantic relationship of the tooth can be efficiently captured, the segmentation precision can be improved, the memory consumption can be reduced, and a tooth model segmentation task with higher precision can be supported.

The application discloses a data processing method, comprising: obtaining a first embedding vector for representing a known data unit and a known data unit position, and a second embedding vector for representing a data unit position to be predicted, processing the first embedding vector through a target encoder to obtain an output vector, and processing the output vector and the second embedding vector through a target prediction network to obtain the data unit to be predicted. The application does not need to separately set additional M position information as the input of the target encoder, and the number of hidden variables of the intermediate output of the target encoder is consistent with the number of the input embedding vectors, thereby reducing the calculation amount and memory consumption of the target encoder.

The present invention discloses an intelligent washing control method, device and washing machine based on clothing image recognition, the method comprising the steps of: obtaining original photos of clothing in different postures in a washing cabin; preprocessing the obtained original photos to obtain a local sample set; sending the local sample set to a pre-trained neural network for a pre-estimation classification of the local sample set; performing secondary classification on the estimated result using a pre-built decision tree model to obtain a clothing material identification classification result; and determining a corresponding washing program according to the material identification classification result. The present invention starts from the local and takes the global into consideration, that is, it ensures the information provided by a higher resolution, and uses local images for model learning, which accelerates the image processing efficiency and reduces the hardware requirements. It does not require a large video memory to cope with the large matrix caused by high resolution. Through secondary classification, more accurate clothing material identification can be obtained, and a built-in targeted washing program can be applied according to the material.