29results about How to "Show well" patented technology

The invention belongs to the technical field of computer vision and agricultural information, and particularly relates to a tea disease detection method based on an attention mechanism. The detection method comprises the following steps: using an improved YOLOv5s model as a basic model; an SE attention mechanism is introduced into a backbone network of the YOLOv5s model, and an original C3 module is replaced with a seC3 module; adding a coordinate attention mechanism at the tail end of the backbone network; collecting tea disease image data, and preprocessing the collected tea disease image data; carrying out disease area calibration on the preprocessed tea disease image data by using a labeling tool to generate a labeling data set; training the constructed YOLOv5-SE model by using the annotation data set to obtain a trained model weight; and applying the trained YOLOv5-SE model to an unlabeled tea disease image, and carrying out disease detection and identification. According to the method, key features in the tea disease image can be better captured and represented, so that the accuracy and reliability of disease detection are improved.

The invention relates to the field of agricultural machinery track classification, and discloses an agricultural machinery track data classification method and device and a storage medium, and the method comprises the steps: obtaining agricultural machinery track data; preprocessing the agricultural machine trajectory data, and constructing a classification model; extracting short-range features and long-range features from the preprocessed data, and fusing the short-range features and the long-range features to obtain comprehensive features; inputting the comprehensive features into a bidirectional long-short-term memory network, and carrying out deep-level time sequence mode learning; inputting the learned features into a classifier to obtain an agricultural machine track classification result; a classification result is compared with an actual result, and the classification model is optimized through a PDW loss function; and finishing agricultural machine trajectory data classification by using the optimized classification model. According to the method, the parameter difference weighting loss function is introduced, the category weight is dynamically adjusted, the model performance is optimized, and the accuracy and generalization ability of agricultural machine trajectory data classification are improved.

The present application discloses a method for generating a virtual image and related equipment thereof, the method comprising: after acquiring voice data to be processed, firstly extracting voice features of the voice data to be processed to obtain voice features to be used, so that the voice features to be used can represent the voice information carried by the voice data to be processed; then determining virtual image display data corresponding to the voice data to be processed according to the voice features to be used, facial parameters to be used, and a pre-constructed virtual image generation model, so that the virtual image display data can represent a virtual image that matches the voice data to be processed, so that the virtual image can display the facial state presented by the character when the voice data to be processed is emitted, thereby achieving the purpose of automatically generating a virtual image.

The invention provides a substation terminal network equipment defect elimination checking method and system based on deep learning, and the method comprises the following steps: constructing an equipment configuration heterogeneous graph network by a substation client through combining an equipment configuration file, an equipment operation log and equipment network flow data, all the transformer substation clients utilize a heterogeneous graph network to train a local model, the federated learning server updates a shared learning model by combining local training parameters of all the clients, and the clients needing to be checked continue to update the model according to updating parameters of the server; and identifying a configuration mode of local network equipment through the updated model and obtaining a configuration parameter mapping table, generating an equipment checking strategy in combination with the configuration mode and the configuration parameter mapping table, and finishing checking work of all the network equipment through the equipment checking strategy. According to the invention, the transformer substation can accurately complete network equipment checking.

The invention provides a network intrusion detection method based on a de-noising diffusion probability model and a double-attention residual network, and belongs to the field of network intrusion detection. The method comprises the following steps: step 1, selecting and extracting features of original network traffic through three different types of claddings, cleaning the extracted features, and representing the cleaned features in an RGB image form; step 2, based on the network traffic RGB image generated in the step 1, learning local / overall features and fine-grained features of the network traffic by using a denoising diffusion probability model, and generating a network traffic sample of an attack type with a small scale to balance distribution of different types of network traffic; and step 3, dividing the expanded network traffic RGB image into a training set, a verification set and a test set by using a uniform random division mode, putting the training set into the network intrusion detection model based on the double-attention residual network for training, and performing verification by using the verification set to obtain an optimal network parameter. And finally obtaining a pre-trained network intrusion detection model.

The invention relates to a sparse principal component analysis method, a quantum computing device and a medium, belongs to the technical field of data dimension reduction, and solves the problem that the sparse principal component analysis method is low in efficiency. According to the technical scheme, the method mainly comprises the steps that a target data matrix is obtained, and a covariance matrix is determined based on the target data matrix; setting a to-be-solved principal component vector, and determining an original objective function and a vector constraint condition; expressing each component of the to-be-solved principal component vector through a binary system, and defining a plurality of second binary variables to respectively express whether each component of the to-be-solved principal component vector is zero or not; determining constraint conditions of the first binary variable and the second binary variable according to the vector constraint conditions and the definition of the second binary variable; forming a secondary unconstrained binary optimization model; and solving the quadratic unconstrained binary optimization model to determine the value of each first binary variable.

The present invention discloses a pipeline leakage monitoring and early warning system, which belongs to the technical field of pipeline network leakage monitoring. The present invention calculates the number of sampling points by the collected audio sampling rate and the collection time. N , the signal is calculated by the feature recognition algorithm level Values, Fourier transform coefficients, center of gravity frequencies and pipeline leakage characteristic values ​​are generated. At the same time, time domain waveforms, spectrograms, and time-frequency graph data are generated. Through matching and comparison analysis of pipeline leakage characteristic values, associated data, and altitude location case libraries, characteristic parameter case libraries, leakage audio case libraries, and noise audio case libraries, it is accurately determined whether there is a pipeline leak. By continuously updating the altitude location case library, characteristic parameter case library, leakage audio case library, and noise audio case library, a pipeline leakage signal early warning model that can be used for different temperatures, different altitudes, different regions, equipment voltages, temperatures, and environmental conditions is established. The leakage signal early warning model is updated, which can significantly improve the early warning accuracy of pipeline leakage monitoring.

The embodiments of the present disclosure disclose an industrial knowledge graph question-answering method and a computer-readable medium based on a large language model. A specific implementation of the method includes: inputting at least one question information in the question information set into the initial question embedding network; inputting at least one question information into the initial question relationship chain feature extraction network; determining the semantic similarity between at least one question information feature vector and the corresponding initial question relationship chain feature vector; constructing at least one question information sample based on at least one question information and at least one initial question relationship chain feature vector; constructing at least one question information annotation sample based on at least one question information sample and the question-answering industrial knowledge graph; training the initial industrial knowledge graph question-answering model to obtain a trained industrial knowledge graph question-answering model. This implementation can better utilize existing knowledge and semantic information, thereby improving the accuracy and effect of answer reasoning.

The invention provides a data processing system, method and device based on micro-macro relation construction and a medium, and relates to the technical field of neural networks, the data processing system comprises a micro relation construction module, a macro relation construction module, a multi-scale image convolution module and a fusion module; the microscopic relation construction module is used for capturing the synchronous relation of different target objects under the whole brain scale; the macroscopic relationship construction module is used for capturing a pairwise relationship and a subtype relationship among different target objects; the multi-scale image convolution module is used for respectively carrying out convolution learning on the synchronous relation, the pairwise relation and the subtype relation to obtain preliminary representation information; and the fusion module is used for fusing different initial representation information into target representation information so as to classify cognitive competence. The method comprehensively considers and analyzes the microcosmic angle and the macroscopic angle, optimizes the representation of the brain complex signal, improves the test accuracy, and provides a new tool for the cognitive test.

This invention discloses a concept abstraction method and apparatus based on pooling networks. It automatically abstracts concepts by splitting and merging entity sets corresponding to triple relationships in the input graph to obtain a pooling graph, and then obtains an output graph based on the pooling graph. The representation of the pooling graph is initialized using the representation of the input graph learned by a graph neural network model. The representation of the pooling graph is updated using the graph neural network model and, together with the representation of the input graph, initializes the representation of the output graph. The representation of the output graph is then updated again using the graph neural network model. The representation of the output graph is used as input to a knowledge representation learning model to train the model, ensuring that the entity representations contain conceptual information. Comparative learning losses are calculated for the input graph-pooling graph, pooling graph-output graph, and input graph-output graph relationships. Finally, reliable candidate triples are determined based on the knowledge scores of the candidate triples, achieving more accurate automatic information completion.

The invention relates to the technical field of oil and gas field development, in particular to a three-dimensional fine geologic model vertical grid optimization method, which comprises the following steps of: analyzing the relative position of an interlayer in a small layer according to a logging interpretation data table, and judging the plane distribution range of the interlayer according to an adjacent well; the relative positions of the sand body and the interlayer interface are calculated, the relative position sorting data are screened, and stratigraphic grid division is completed; carrying out vertical grid division on the geological model according to the screened relative position sorting data, and establishing a three-dimensional structure model capable of depicting a minimum sand body and an interlayer; and predicting the sand body and the interlayer by adopting a sequential indication method to obtain a three-dimensional lithofacies model reflecting the sand body, the mudstone and the interlayer. According to the method, the problems of thin interbed reservoir and interlayer information loss and incompleteness and the like in a traditional geologic model can be effectively solved; a thin interbed reservoir and an interlayer can be better displayed in a reservoir model according to a structural model established by a fine geologic model vertical grid optimization method.

The application discloses a kind of aspect-level sentiment analysis method based on multilevel knowledge enhancement, belong to sentiment analysis field. Including the following steps: S1: text word vector is obtained using GloVe word embedding tool, the syntax dependency tree of sentence is constructed using Stanza, and the dependency graph is constructed accordingly;S2: the context representation of sentence is extracted by inputting word vector into BiLSTM, and the dependency graph is updated using sentiment dictionary and sensitive relationship set, to realize the sentiment and syntax enhancement of sentence;S3: the enhanced dependency graph is input into GCN to model node features, to obtain specific aspect representation;S4: aspect word is enhanced using concept atlas to obtain aspect word representation, which is fused with specific aspect representation to obtain aspect representation;S5: aspect representation and context representation are coordinated and optimized using interactive attention, to obtain the final representation of sentence, so as to determine aspect sentiment tendency.The application can effectively improve the accuracy of specific aspect sentiment classification, and help merchants accurately locate problems in products or services.

This utility model discloses a bedside table with a concealed compartment component, relating to the field of furniture technology. It includes a bedside table body with several drawer boxes inside. Each drawer box has a sliding rail on both sides, and sliders fixed to the left and right sides of the drawer box, sliding within the sliding rails. A concealed compartment is located at the rear of the drawer box, containing several upper and lower storage compartments. This utility model's sliding pull-out mechanism and multi-compartment layout achieve secure storage of personal items and privacy protection. The concealed compartments fit perfectly with the bedside table, ensuring concealment and effectively utilizing cabinet space, increasing storage capacity without increasing size. The sliding rail and slider structure facilitates access to each storage compartment while enhancing the overall uniformity and aesthetics. It breaks through the traditional bedside table design, meeting daily storage needs while providing private storage space, balancing functionality and aesthetics.

The application discloses a kind of sparse point cloud and GCN fusion's citrus defect detection method and system, the point cloud of citrus data set is input into GCN_PointNet++ model in the application, the GCN_PointNet++ model combines GCN with PointNet++ network and carries out feature transmission by PaConv module;GCN_PointNet++ model is distilled by BIFPN, and the total distillation loss function is obtained by combining BIFPN distillation label to guide student network training, and the best point cloud segmentation result is obtained, that is, citrus defect segmentation point cloud model;Based on citrus defect segmentation point cloud model, three-dimensional reconstruction is carried out on citrus, three-dimensional point cloud mesh file is established, and the area of defects is calculated according to the surface formed by the mesh.The application realizes the accurate segmentation of citrus defect point cloud and the accurate quantification of citrus defect area.

The invention discloses a drum granulation method for a water-soluble compound fertilizer. The drum granulation method comprises the following steps: (1) crushing raw materials of a basic fertilizer; (2) uniformly stirring the crushed raw materials and a water-soluble modified starch solution to obtain a uniform mixture; (3) putting the obtained uniform mixture into a rotary drum granulator, and granulating to obtain a fertilizer particle mixture; (4) feeding the fertilizer particle mixture into a drying cylinder for drying treatment to obtain a dried fertilizer particle mixture; (5) performing primary cooling and secondary cooling on the dried fertilizer particle mixture to obtain a cooled fertilizer particle mixture; and (6) carrying out primary screening and secondary screening on the cooled fertilizer particle mixture to prepare a qualified fertilizer particle finished product. According to the invention, the physical properties and nutrient utilization rate of the fertilizer are improved by accurately controlling the use amount of the drum granulation raw materials of the water-soluble compound fertilizer and optimizing process parameters; and through the synergistic effect of the raw materials, the product quality can be obviously improved.

The invention discloses a method for monitoring the engineering status of multi-target spacecrafts from a third-perspective on the surface of an extraterrestrial body, and belongs to the technical field of deep space exploration. First, a monitoring system applied to the method for monitoring the engineering status of multi-target spacecrafts from a third-perspective is established; then, monitoring system parameters are determined according to requirements and a camera field of view model is constructed, a spacecraft geometric model is constructed according to the structural dimensions of the spacecraft, and a space illumination environment of the surface of the celestial body is established according to the local time; imaging composition and timing are designed to obtain the positional relationship between each spacecraft and the camera and the relative posture of each spacecraft and the camera; finally, according to the space illumination environment of the surface of the celestial body, an imaging effect image of all spacecrafts in the camera field of view is obtained, and real-time monitoring of the engineering status of all spacecrafts is achieved according to the image; the invention can monitor and take group photos of the engineering status of multiple spacecrafts on the surface of an extraterrestrial body by using a completely independent third-perspective method for local and panoramic imaging.

According to the static stain analysis device and analysis method based on semantic harmless recognition, a traditional data flow analysis framework is abandoned, a corresponding pollution propagation rule is formulated, and a pollution path diagram related to a pollution variable relation is abstracted; and a user-defined Sanitizer set is recognized based on semantics, so that the pollution path diagram is further verified and pruned through the Sanitizer set. According to the analysis method, a corresponding pollution path diagram is constructed through a defined pollution propagation rule analysis program, and the device is only focused in a data stream of a stain source and a leakage point, so that the time cost for analyzing redundant nodes and edges is reduced. According to the method, the detection capability and analysis efficiency of static taint analysis on vulnerabilities in the Web application program can be improved, and more accurate input data is provided for subsequent security analysis.

The utility model discloses a propaganda and display device for workplaces, and relates to the technical field of propaganda and display devices. The display board comprises a display board body, two telescopic rods are arranged at the bottom of the display board body, a bottom plate is arranged at the bottom ends of the two telescopic rods, a lifting assembly is arranged between the bottom plate and the display board body, a channel is formed in one side of the display board body, and two extension boards are arranged in the channel in a sliding fit mode. A plurality of sliding grooves are formed in one sides of the display board and the two extension boards, a plurality of sliding blocks are arranged in the sliding grooves in a sliding fit mode, clamping plates are arranged on the sliding blocks in an elastic fit mode, a round rod is arranged on one side of a channel, a cylinder is arranged on the peripheral side of the round rod in a rotating fit mode, an adjusting assembly is arranged on the cylinder, and the adjusting assembly is matched with the two extension boards. By arranging the sliding blocks, the sliding blocks can move to drive the clamping plates to move, the clamping plates move to the needed positions to fix the display pictorial and the clamping plates, and therefore the needed clamping plates can be moved out according to the size of the display pictorial, and splicing display of a plurality of display pictorial is facilitated.

The invention relates to the field of molecular oil refining, in particular to a structure-oriented lumped data storage and retrieval method and device. The method comprises the following steps: according to chemical structures of oil product molecules, combining codes corresponding to basic structure units of the oil product molecules to obtain corresponding structure codes representing the oil product molecules; classifying the structure codes corresponding to the oil product molecules according to preset properties to obtain the structure codes of the oil product molecules under various properties; constructing a database for storing the oil product molecules according to the structure codes of the oil product molecules under various properties; according to a retrieval condition provided by a user, the structure code corresponding to the retrieval condition is inquired from the database, and the retrieval condition corresponds to a preset property. According to the scheme, data processing, storage and retrieval in the molecular oil refining process are optimized, and the oil refining efficiency and the product quality are improved.

The invention belongs to the technical field of advertising boards, and particularly discloses a convenient-to-install vehicle-mounted LED advertising board which comprises a vehicle body top, a wind resistance induction type self-adaptive lifting mechanism, a self-cleaning flow guide assembly and an installation fixing mechanism. The wind resistance induction type self-adaptive lifting mechanism comprises a main body bottom plate, a sliding guide assembly, a folding advertising board assembly and a synchronous symmetry assembly. The height of the advertising board body is automatically controlled through different wind resistances caused by different speeds, an advertising space can be well displayed when a city travels at a low speed, and the windward area of a vehicle can be reduced through automatic unfolding and descending of the advertising board body when the vehicle travels on an expressway or an expressway section, so that the safety of the vehicle is improved. Not only are the wind resistance and the energy consumption of the vehicle effectively reduced, but also the quiet riding experience can be remarkably improved.

A performance mechanism for biomimetic limb movement includes a torso, forelimbs, and hindlimbs. The forelimbs include a fore thigh, fore calf, and forefoot. One end of a first follower link is hinged to the fore thigh, and the other end is hinged to a first drive rocker. The first drive rocker is hinged to a first hinge seat fixed to the torso. The hindlimbs include a hind thigh, hind calf, and hindfoot. One end of a fourth follower link is hinged to the hind thigh, and the other end is hinged to a second drive rocker. The second drive rocker is hinged to a third hinge seat fixed to the torso. Drive assemblies connected to the first and second drive rockers are located at the front and rear of the torso, respectively. This performance mechanism features limbs that can be automatically and independently controlled. The limbs are compact, agile, and easy to adjust. They can perform biomimetic limb movements in a coordinated manner, achieving multiple biomimetic coordinated postures as needed, such as leisurely walking or vigorous running.

The present invention relates to the technical field of Beidou integrated machine, and discloses a Beidou AIS intelligent shipborne integrated machine, including a body shell, a power supply module, a collection module, an information transceiver module, a display module and an alarm module, wherein the power supply module is electrically connected to the collection module, the information transceiver module, the display module and the alarm module through wires, respectively, the display module is connected to the collection module and the information transceiver module through wires, and the alarm module is connected to the information transceiver module through wires. The present invention can not only measure the distance of the external signal transmitter through a wireless rangefinder, thereby facilitating the measurement of the distance between two ships, but also can drive the pressure plate to move through the elastic force of spring 1, and the movement of the pressure plate drives the connecting rod to move and squeeze the cylinder, so that the cylinder moves to assist in supporting the antenna, and can also separate from the rotating rod through the movement of the inclined rod, so that the rotating rod rotates, thereby moving the moving cylinder, which is convenient for the normal use of the camera.

This invention discloses a method for characterizing multi-stage carbonate reservoirs based on time-varying Gaussian window transient extraction transform, used for characterizing multi-stage carbonate reservoirs. The method includes: inputting the seismic signal to be analyzed, where is time; constructing a short-time Fourier transform containing the time-varying window width parameter to be optimized, where represents frequency; determining the optimal window width parameter at each time step based on the local Rényi entropy minimization criterion; substituting and calculating its time partial derivative; constructing a transient extraction operator based on the time partial derivative results, including a local maximum criterion, an energy derivative zero-point criterion, and a group delay migration criterion; and proposing a time-varying Gaussian window transient extraction transform under the action of this operator for characterizing multi-stage carbonate reservoirs.

The present application belongs to the field of computer vision and agricultural information technology, and specifically relates to a tea disease detection method based on an attention mechanism. The detection method comprises: using an improved YOLOv5s model as a basic model; introducing an SE attention mechanism into the YOLOv5s model backbone network, and replacing the original C3 module with an seC3 module; adding a coordinate attention mechanism at the end of the backbone network; collecting tea disease image data, and preprocessing the collected tea disease image data; using a labeling tool to label the disease area of the preprocessed tea disease image data to generate a labeled data set; training the constructed YOLOv5-SE model using the labeled data set to obtain the model weight after training; and applying the trained YOLOv5-SE model to unlabeled tea disease images for disease detection and identification. This method can better capture and represent the key features in the tea disease image, thereby improving the accuracy and reliability of disease detection.

The invention relates to the technical field of motor disassembling, and provides a motor disassembling auxiliary device for a scrapped new energy automobile and a disassembling method.The motor disassembling auxiliary device for the scrapped new energy automobile comprises a bottom plate, a supporting frame, a fixing ring, a drawing ring, a fixing mechanism, a drawing mechanism and an adjusting mechanism; the two fixing rings are both arranged in the supporting frame, the two drawing rings are both arranged on the supporting frame, the fixing mechanism is arranged in the supporting frame and used for fixing a motor shell, the drawing mechanism is arranged on the supporting frame and used for disassembling the motor shell, and the adjusting mechanism is arranged on the bottom plate. According to the technical scheme, the motor disassembling angle adjusting device solves the problems that in the prior art, when an operator pries up a motor shell manually, time and labor are wasted, and the shell is prone to being damaged.

The invention discloses a chemical molecule graph classification method and device based on structure enhancement and a medium, and belongs to the technical field of graph machine learning. Aiming at the problem of insufficient utilization of chemical molecule graph structure information, the method comprises the following steps of: 1, acquiring a chemical molecule data set, and constructing a corresponding chemical molecule graph for each chemical molecule in the chemical molecule data set by taking an atom as a node and a chemical bond as an edge; 2, constructing a chemical molecule graph classification model based on structure enhancement, wherein the model comprises three parts of structure enhancement, sub-graph representation and graph classification; and step 3, training and optimizing the graph classification model constructed in the step 2, and performing model performance evaluation and comparative analysis. The structure information of the chemical molecules can be more perfectly captured, and the classification accuracy of the chemical molecules is remarkably improved.

The invention belongs to the technical field of data processing, and relates to a method for generating a smart school sports meeting order manual, comprising the steps of: collecting a student list, class information, a teacher list, and basic information of venue facilities; setting competition items, registration conditions, competition rules, and scoring standards, and storing them in a database; configuring a mathematical model for event scheduling, the mathematical model comprising variables, constraints, and an objective function; initializing a population through a genetic algorithm, wherein each individual represents a possible schedule arrangement; calculating the fitness value of each individual and evaluating its pros and cons; generating a new generation of populations through selection, crossover, and mutation operations; repeating the selection, crossover, and mutation operations until a preset number of iterations is reached or a stop condition is met, and outputting an optimal schedule arrangement plan; configuring a message queue system, which is used to receive score update requests from referees and staff, and ensure timely transmission of information; and synchronizing the update information to a database through an API interface.

The invention belongs to the technical field of data processing, and relates to an intelligent school freight meeting order manual generation method, which comprises the steps of collecting a student list, class information, a teacher list and basic information of site facilities; competition items, registration conditions, competition rules and scoring standards are set and stored in a database; configuring a mathematical model of the match schedule, wherein the mathematical model comprises variables, constraint conditions and a target function; a population is initialized through a genetic algorithm, and each individual represents a possible match schedule; calculating the fitness value of each individual, and evaluating the advantages and disadvantages of each individual; a new generation of population is generated through selection, crossover and mutation operation; repeating selection, crossing and mutation operations until a preset number of iterations is reached or a stop condition is met, and outputting an optimal schedule arrangement scheme; the configuration message queue system is used for receiving score updating requests from referees and workers and ensuring timely transmission of information; and updating information is synchronized to the database through the API interface.

The present invention discloses a hierarchical text classification method based on a pre-trained generative model, the purpose of which is to solve the problems of label inconsistency, the need to manually set thresholds, and the inability to effectively learn the co-occurrence relationship between target labels in existing hierarchical text classification methods. The present invention transforms the hierarchical text classification problem into a hierarchical label sequence generation problem, and by constructing a label sequence that incorporates hierarchical information, the model can learn the hierarchical information and co-occurrence relationship between labels, thereby solving the label inconsistency problem. Moreover, the label generation process of this method adopts an autoregressive method, so there is no need to manually set a threshold in the process of predicting labels. Experiments show that the classification performance of the present invention is better than that of other hierarchical text classification methods.