Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

55results about How to "Address underutilization" patented technology

Personalized recommendation system based on semantic analysis

The invention discloses a personalized recommendation system based on semantic analysis, and relates to the technical field of personalized recommendation systems.The personalized recommendation system comprises a data acquisition module, a dynamic situational semantic modeling module and a recommendation engine module, the data acquisition module transmits acquired user behavior data and situational data to the dynamic situational semantic modeling module, and the dynamic situational semantic modeling module transmits the acquired user behavior data and situational data to the recommendation engine module; the dynamic situational semantic modeling module extracts sequence features of user behavior data through a behavior context analyzer, generates context embedding through a situational semantic embedding network, completes cross-modal semantic alignment of multi-modal data through a multi-modal fusion layer, and generates multi-level semantic embedding features. According to the method, the dynamic situational semantic modeling module is designed, the semantic alignment function of the situation and the behavior is achieved, the problem that a traditional recommendation system is insufficient in situation data utilization is solved, the adaptability of the recommendation result to the dynamic situation is enhanced, and the method is suitable for dynamically changing user demand scenes.
Owner:HEFEI ZHONGKE CLOUD NUMBER TECH CO LTD

Bearing fault diagnosis method and device

The invention discloses a bearing fault diagnosis method and device, and relates to the technical field of bearing fault diagnosis, and the method comprises a model training step, a model integration step and a fault diagnosis step, and the model training step provides a trained diagnosis model for the model integration step. The anomaly detection model output in the model integration step is used as the model algorithm input in the fault diagnosis step, and the model integration step is used for integrating a plurality of models to improve the diagnosis accuracy and the system robustness; the model integration step comprises the steps of multi-model integration, dynamic model selection and anomaly detection. According to the bearing fault diagnosis method, the model integration step is designed, the advantage integration function of multiple models is achieved, the limitation problem when a single model processes complex and changeable fault modes is solved, fault diagnosis is difficult to comprehensively and accurately conduct, the accuracy and robustness of a bearing fault diagnosis system are improved, and the fault diagnosis efficiency is improved. Fault detection under complex working conditions can be better handled, and the probability of false alarm and missing alarm is reduced.
Owner:YANCHENG SHURONGZHISHENG TECH CO LTD +1

Non-structured scene oriented foot type robot visual navigation system

The invention discloses an unstructured scene-oriented foot type robot visual navigation system, relates to the technical field of robot navigation, and establishes a set of multi-layer local sensing map capable of describing ground features in order to solve the problem of inaccurate navigation caused by insufficient utilization of ground feature information in the prior art. Description of ground features is added on the basis of a traditional local map. And secondly, on the basis of the local perception map, the ground feature parameters are fused into the trajectory optimization process, so that the problem that the foot robot cannot fully utilize the ground feature information is solved. And the accuracy of visual navigation of the robot is improved.
Owner:HARBIN INSTITUTE OF TECHNOLOGY SUZHOU RESEARCH INSTITUTE +1

Asynchronous federated learning system and method for turnout switch machine fault diagnosis model

The invention provides an asynchronous federal learning system and method for a turnout switch machine fault diagnosis model. The system comprises the following steps: acquiring mechanical vibration signals at three positions of a speed reducer shell, a contact seat and an actuating rod of a switch machine by using a mechanical vibration sensor; information fusion of the vibration signals of the three point locations is carried out; a lightweight dual-scale neural network model; an asynchronous federated learning process resistant to time delay and packet loss; and an asynchronous federated learning system structure. An asynchronous federal learning process resistant to time delay and packet loss is adopted, communication delay and the remaining receiving time of single uploading of an edge computing unit are fully utilized, and the extra computing time for fusion after all units are uploaded is shortened; the asynchronous federated learning system structure clarifies the hierarchical relationship and specification of federated learning of point switch fault diagnosis in a railway system. Distributed training of the switch machine fault diagnosis model is realized, and reliable guarantee is provided for railway transportation safety.
Owner:BEIJING JIAOTONG UNIV

Electrolytic tank for producing hydrogen by electrolyzing water and method for producing hydrogen by electrolyzing water

The invention provides an electrolytic cell for producing hydrogen by electrolyzing water and a method for producing hydrogen by electrolyzing water. The electrolytic cell for producing hydrogen by electrolyzing water is provided with an anode, a cathode, a diaphragm, an anode runner, a storage device and an electrolyte conveying device, the electrolyte inlet of the electrolyte conveying device is communicated with the electrolyte outlet of the storage device, the electrolyte outlet of the electrolyte conveying device is communicated with the electrolyte inlet of the anode runner, and the electrolyte outlet of the anode runner is communicated with the electrolyte inlet of the storage device. According to the invention, the electrolyte dispersed with the anode catalyst circularly flows in the anode flow channel and the storage device, so that the polarization performance and the stability of the electrolytic cell for hydrogen production by water electrolysis during hydrogen production by water electrolysis are remarkably improved.
Owner:TAN KAH KEE INNOVATION LAB +1

Micro-expression recognition method based on multi-modal fusion

The invention discloses a micro-expression recognition method based on multi-modal fusion, which relates to intelligent micro-expression recognition and comprises the steps of acquiring input data of a model from a micro-expression database, performing feature extraction on each piece of input data by using ViT and a CNN (convolutional neural network) to obtain a feature vector Va and a feature vector Vf; and performing similarity measurement on the feature vector Va and the feature vector Vf to obtain an output feature. The micro-expression data is classified by fusing peak frames, optical flow diagrams and key points, and the problem of insufficient utilization of the micro-expression data in the prior art is solved by utilizing data features of different modes; for data of different modes, a network architecture is adopted, Vision Transformer (ViT) is used for extracting global features of peak frames, a convolutional neural network (CNN) is used for extracting local features of an optical flow graph, fused features are optimized through comparison loss and classification loss, the similarity of the same expression sample features is improved, and recognition accuracy and robustness are enhanced.
Owner:BAOJI UNIV OF ARTS & SCI

A liquid phase method for growing silicon carbide single crystal and a growth method thereof

The present invention relates to a growth device and a growth method of a silicon carbide single crystal by a liquid phase method, and belongs to the technical field of silicon carbide single crystal production. The growth device includes a growth crucible, a seed crystal rod fixed with a seed crystal, a laser rangefinder for determining the position of the growth raw material, a movable graphite block, a braking device, and a blocking graphite; a reserved hole is provided in the crucible cover of the growth crucible corresponding to the laser rangefinder directly below the light exit hole; at least four rows of holes are provided on the side wall of the growth crucible, and each row of holes is evenly spaced from top to bottom, and a hole is provided at the center of the bottom, and the movable graphite block is filled in the hole, and one end of the blocking graphite is embedded in the hole, and the other end is sleeved on the outer wall of the growth crucible; an opening is provided in the center of the blocking graphite for the braking device to pass through and connected to the movable graphite block, and the braking device is used to control the movable graphite block to move forward and backward relative to the growth raw material. The device of the present invention can realize the determination and adjustment of the high temperature line position before crystal growth, can weaken the self-circulating convection in the growth solution, and improve the utilization rate of carbon.
Owner:BEIJING LATTICE SEMICONDUCTOR CO LTD

Vehicle body grid parameter dynamic optimization method and system based on model simulation

The invention discloses a vehicle body grid parameter dynamic optimization method and system based on model simulation, and belongs to the crossing field of computer-aided engineering and artificial intelligence technology. According to an existing vehicle body grid optimization scheme, deep mining and analysis of grid data cannot be achieved, and consequently the accuracy and reliability of a grid optimization result are not high. According to the vehicle body grid parameter dynamic optimization method based on model simulation, a vehicle body historical database module, a vehicle body grid optimization module and a domain knowledge base module are utilized to match vehicle body grid data to be optimized with vehicle body historical data to obtain the most similar vehicle body historical data; the information is used as experience information and verification basis of vehicle body grid parameter optimization; and then experience information and a verification basis are reused, the to-be-optimized grid data of the vehicle body are analyzed and processed, and optimization suggestion information is generated, so that historical data can be deeply mined and analyzed, and the accuracy and reliability of a grid optimization result can be effectively improved.
Owner:ZHEJIANG YUANSUAN TECH CO LTD

Growth device and growth method for liquid phase method silicon carbide single crystal

The invention relates to a growth device and a growth method for a liquid-phase silicon carbide single crystal, and belongs to the technical field of silicon carbide single crystal production. The growth device comprises a growth crucible, a seed crystal rod fixed with a seed crystal, a laser range finder used for measuring the position of a growth raw material, a movable graphite block, a brake device and blocking graphite; a preformed hole is formed right below a crucible cover of the growth crucible corresponding to a light outlet hole of the laser range finder; at least four rows of holes are formed in the side wall of the growth crucible, each row of holes are uniformly distributed at intervals from top to bottom, a hole is formed in the center of the bottom and filled with the movable graphite block, one end of the plugging graphite is embedded into the hole, and the other end of the plugging graphite sleeves the outer wall of the growth crucible; the center of the plugging graphite is provided with an opening for the brake device to pass through and connected with the movable graphite block, and the brake device is used for controlling the movable graphite block to move back and forth relative to the growth raw material. According to the device, the position of the high-temperature line before crystal growth can be determined and adjusted, self-circulation convection in a growth solution can be weakened, and the utilization rate of carbon is increased.
Owner:BEIJING LATTICE SEMICONDUCTOR CO LTD

Method for fairly scheduling multiple processes of computer

The invention discloses a computer multi-process fair scheduling method, and relates to the technical field of multi-process resource allocation in a parallel computer architecture, the processes are dynamically divided into an I / O intensive type and a CPU intensive type through process classification and feature recognition, and the types of the processes are judged by monitoring the CPU use time and the I / O waiting time of each process. A scheduling strategy is set for each process, different priority adjustment strategies are formulated for different types of processes in dynamic priority adjustment, the length of time slices is dynamically adjusted according to the types of the processes, short time slices are allocated for I / O intensive processes, frequent I / O operation is allowed, returning is conducted in time, and waiting time is shortened. Compared with the prior art, the method has the advantages that long time slices are distributed for CPU intensive processes, frequent context switching expenses are reduced, the CPU utilization rate is increased, the execution requirements of different types of processes are better balanced by dynamically adjusting the lengths of the time slices, and the resource utilization efficiency of a system is improved.
Owner:SHANGHAI HAOXIN HAOYI INTELLIGENT TECH CO LTD

Peak-valley difference arbitrage energy storage control method considering peak and deep valley electricity prices

The invention discloses a peak-valley difference arbitrage energy storage control method considering peak and deep valley electricity prices, which comprises load distribution prediction and charge and discharge starting time distribution based on the load distribution prediction, the load distribution prediction comprises charge prediction, discharge prediction and a prediction mode, the charge prediction comprises preconditions, opportunity and content, and the discharge prediction comprises preconditions, opportunity and content. The precondition is that the local electricity price has a deep valley; the charging prediction opportunity is prediction before the valley period starts; the charging prediction content comprises the amount of electricity capable of being charged in deep valleys and the amount of electricity capable of being charged in valley sections; the discharge prediction comprises opportunity and content, and the discharge prediction opportunity is predicted before the peak segment starts; the discharge prediction content comprises the discharge capacity of the peak and the peak section; the prediction method comprises the following steps: S1, finding an intra-day load change rule; s2, calculating a reference load base line of the day and comparing the reference load base line with the current load base line in each time period; s3, solving in a proper time period to obtain a predicated value of the chargeable and dischargeable electric quantity of each time period; the load prediction distributes charging and discharging starting time, and the charging logic and the discharging logic are included.
Owner:ZENERGY TECH CO LTD

Turbofan engine RUL prediction method and system based on historical data fusion

The invention discloses a turbofan engine RUL prediction method and system based on multi-scale historical information fusion, and the method comprises the steps: collecting various sensor data during the operation of a turbofan engine, and carrying out the preprocessing, and obtaining a plurality of pieces of time window data with RUL labels; processing data by using three sliding windows of different scales, superposing all historical time window data before the current moment on a third dimension, performing dimension reduction on the superposed data by using a principal component analysis method, and finally fusing the obtained data of three scales to obtain training and testing samples; an ISEformer network framework is adopted, and an RUL prediction model is obtained through training; and outputting a prediction result by the prediction model according to the operation data of the to-be-detected turbofan engine. Through the innovative method combining historical data fusion and the deep learning technology, the RUL prediction performance of the turbofan engine under the complex working condition is remarkably improved.
Owner:WUXI UNIV

Electrochemical reactor with multiple electrode groups connected in parallel and through cathode liquid

The invention relates to the technical field of water treatment, in particular to a catholyte-through multi-electrode-group parallel electrochemical reactor which comprises a reaction tank and a plurality of groups of electrode assemblies arranged in the reaction tank, the electrode assembly includes an anode; the cathode is annularly arranged outside the anode; the separation net is arranged between the anode and the cathode and is configured to isolate the catholyte from the anolyte; the base and the cover are provided with grooves for mounting the anode, the separation net and the cathode and are respectively mounted at the bottoms and the tops of the anode, the separation net and the cathode; the positive terminal is inserted into the cover and is connected with the anode; the negative terminal is inserted into the cover and is connected with the cathode; the acid extraction opening is formed in the center of the cover and penetrates through the cover to extend into the anode area of the anode; and the external pump is connected with the acid pumping port through a pipeline. The problem of insufficient utilization of hydroxyl in the existing water softening technology is solved, the number of the electrode groups can be flexibly adjusted according to actual requirements, the industrial-scale water treatment requirements are met, and meanwhile, the operation cost is effectively reduced.
Owner:DALIAN UNIV OF TECH

Image analysis method based on double-branch multi-modal sequence learning

The invention provides an image analysis method based on double-branch multi-modal sequence learning, and belongs to the technical field of medical image analysis, and the method comprises the steps: constructing a multi-modal embedding module, filling missing multi-modal disease data by using learnable parameters, fusing the filled multi-modal features, and obtaining the coding features of a subject; a dual-branch structure comprising a static information analysis module and a dynamic information analysis module is utilized to analyze the coding features, and dual-branch multi-modal image analysis features are obtained; fusing the dual-branch multi-modal image analysis features to obtain image features in the longitudinal queue data; and stacking the image features in all the longitudinal queue data, and analyzing the image features in all the stacked longitudinal queue data by using the multi-modal data except the filled multi-modal disease data to obtain an image analysis result. According to the method, the problem of insufficient utilization of multi-modal information and time sequence evolution information in a multi-modal longitudinal queue image analysis method is solved.
Owner:BEIHANG UNIV

A Method and System for Monitoring Electricity Trading Risks Based on Multi-Source Signaling Mechanism and Sparse Hybrid Expert Model

PendingCN122089090ASolve the lack of generalization abilityImprove stabilityEnsemble learningBiological modelsData setElectricity market
This invention proposes a method and system for monitoring electricity trading risks based on a multi-source signal mechanism and a sparse hybrid expert model. The method includes: collecting historical day-ahead electricity market price data and raw data on influencing factors; performing data preprocessing to construct a multi-source fusion signal from price spread signals and bidding space signals; using a random forest algorithm for feature selection; constructing a set of similar day data based on the bidding space signal and incorporating it into the training set; using a sparse hybrid expert model as the model framework, with the price spread signal and bidding space signal as targets, training different prediction models; and searching for an optimal declaration curve based on the Northern Eagle optimization search algorithm by predicting the price spread signal and bidding space signal, using the short-term power prediction curve as a benchmark. This invention effectively solves the problem of insufficient model generalization ability, realizes risk monitoring in electricity spot trading, and improves the accuracy of electricity spot trading risk assessment while enhancing the stability of the power system.
Owner:STATE POWER RIXIN TECH CO LTD

Oil gas recovery intelligent control system and method based on multi-source data fusion

The invention discloses an oil gas recovery intelligent control system and method based on multi-source data fusion, and relates to the technical field of intelligent control. The system comprises a data acquisition and classification module, an influence intensity judgment module, a priority dynamic adjustment module, a multi-source fusion control module and an equipment execution module. The data acquisition and classification module divides direct and indirect data, the influence strength judgment module dynamically captures data association strength without a preset rule, the priority dynamic adjustment module matches data priorities according to strength sorting, the multi-source fusion control module fuses high-priority data to generate a control instruction, and the equipment execution module adjusts operation parameters. Through dynamic collection of classified data, real-time judgment of influence intensity, dynamic adjustment of priority, multi-source fusion control and adaptation to multi-medium complex working conditions, the control accuracy is improved, the high recovery rate is stably maintained, emission reaches the standard, and the operation and maintenance risk is reduced.
Owner:QINGDAO FEIPUSI ENVIRONMENTAL PROTECTION TECH CO LTD +1

A method and system for nonlinear carrier phase removal of fringe projection images

ActiveCN116363035BAccurate detectionAccurate nonlinear carrier phaseSingular value decompositionProjection image
The application discloses a kind of nonlinear carrier phase removal methods and systems of fringe projection image.The method obtains the fringe projection image of first and second view angle with nonlinear carrier;Including the following steps: (1) the fringe projection image of first and second view angle is fused using image splicing algorithm;(2) using singular value decomposition, the first principal component obtained is regarded as background component, and other components are regarded as foreground component;(3) detect foreground region and background region;(4) polynomial curved surface fitting is carried out to the background region of first and / or second view angle, and the first and / or second nonlinear phase is obtained;(5) subtract respective carrier phase and the respective nonlinear phase obtained in step (4).The application obtains global fusion fringe projection image without obstruction by double camera fringe imaging system, and detects background region based on singular value decomposition, so that nonlinear carrier phase is accurately removed.
Owner:GUANGXI UNIV

Inclined vibration device for lithium battery cell stock solution barrel

ActiveCN222181805URealize efficient use of functionsAddress underutilizationBottle emptyingControl engineeringElectric machinery
The utility model provides an inclined vibration device for a lithium battery cell stock solution barrel, which comprises a placing platform used for placing the lithium battery cell stock solution barrel, a vibration mechanism located at one corner of the placing platform, a lifting mechanism and the vibration mechanism located at the same side of the placing platform, a first rotating frame installed on the ground, and a second rotating frame installed on the ground. The second rotating frame is installed on the placing platform, the first rotating frame is provided with a vertical turbine screw rod, the turbine screw rod is connected with a horizontal transmission plate and can drive the transmission plate to move up and down, the second rotating frame is fixedly connected with the transmission plate, the turbine screw rod is connected with a lifting motor, and the vibration mechanism comprises a vibration motor. According to the technical scheme, the function of efficiently utilizing the stock solution hung on the inner wall of the stock solution barrel is achieved. And moreover, the operation is simple, convenient and safe, the vibration process of the stock solution barrel can be realized only by turning on the switch, the problem of insufficient material utilization rate in the prior art is solved, the original solution method is optimized, and the time consumption is reduced.
Owner:江苏国轩新能源科技有限公司

A method for finding K most similar neighbors of proteins based on graph neural network

The present invention proposes a method for searching K most similar neighbors of a protein based on a graph neural network. The purpose of this method is to find the K neighbors that are most similar to the query point feature in the feature space of the sample, so as to mine the potential similarity relationship of the protein. The proposed method is a graph neural network model with node feature embedding, which can make full use of the network topological characteristics of the interaction relationship between proteins, the amino acid sequence, physicochemical properties and geometric properties of the protein, and generate the feature vector of the protein node. The K most similar neighbors of the protein are solved according to the node feature vector. The graph neural network consists of a graph convolutional neural network, a graph attention network, an objective function and a nearest neighbor solving module. The K most similar neighbors are the K nodes with the shortest Euclidean distance to the target node or the most similar to the target node. Experimental results show that the method proposed by the present invention can obtain a higher accuracy rate in mining the K most similar neighbors of proteins.
Owner:BEIJING INST OF TECH

Large-scale generation method and system for interpretable question and answer pairs facing vertical field

The invention provides a large-scale generation method and system for interpretable question and answer pairs for the vertical field, and relates to the technical field of data processing, and the method comprises the steps: 1, constructing a cross-modal heterogeneous meteorological knowledge graph; heterogeneous graph neural network modeling is carried out on the cross-modal heterogeneous meteorological knowledge graph, an embedded vector of each node is learned, a knowledge graph with node embedded vectors is obtained, semantic community division is carried out on an entity based on the embedded vectors, and an optimized knowledge graph is generated. According to the method, through knowledge graph optimization, conical semantic space construction, structured evidence chain retrieval and pre-training large language model generation, large-scale generation of interpretable meteorological question and answer pairs in the vertical field is realized, and interpretability and credibility of question and answer results are improved.
Owner:XIAMEN SHIBAO NETWORK TECH CO LTD

Aluminum-silicon flue gas waste heat cooperative power generation system and energy saving method

The invention relates to the technical field of flue gas waste heat recovery, and provides an aluminum-silicon flue gas waste heat cooperative power generation system and an energy saving method.The system comprises a gas-water heat exchanger, a hot water pump, a waste heat boiler, a steam turbine, a power generator, a condenser, a circulating water pump, an aluminum electrolysis cell group, an ore-smelting electric furnace group, a dust removal system and a flue gas purification system; the aluminum electrolysis cell group is connected with the gas-water heat exchanger; hot water subjected to heat exchange of the gas-water heat exchanger is pumped into a deaerator through a hot water pump to be deoxidized and then enters a waste heat boiler, high-temperature flue gas of the ore-smelting electric furnace group is introduced into the waste heat boiler to heat the hot water, and then the hot water is discharged after being treated by a dust removal system; high-temperature steam generated by the waste heat boiler enters the steam turbine, and the steam turbine is matched with the power generator to generate power; after being condensed by the condenser, steam discharged by the steam turbine enters the gas-water heat exchanger through the circulating water pump for water replenishing and heat exchange. The system saves energy and reduces consumption, waste heat waste is reduced, greenhouse gas emission is reduced, and the requirement for green and low-carbon production is met.
Owner:SHANGHAI UNIV OF ENG SCI +1

Building energy calibration and prediction method based on multi-task Gaussian process proxy model

The invention relates to a building energy calibration and prediction method based on a multi-task Gaussian process proxy model. The building energy calibration and prediction method is characterized by comprising the following steps: S1, data acquisition; s2, constructing an initial model; s3, sensitivity analysis and parameter screening; s4, sampling parameters; s5, performing batch simulation; s6, carrying out data sub-sampling; s7, constructing an agent model; s8, Bayesian parameter calibration is carried out; s9, predicting a physical model; s10, judging whether a preset standard is met or not; s11, correcting model deviation; and S12, final output is carried out. The method has the advantages that efficient and accurate calibration of building parameters and effective correction of model deviation are realized by constructing a weather decoupling agent model and a sequential calibration framework, and processing of high-dimensional data and construction of digital twinning can be supported.
Owner:SHUNDE POLYTECHNIC

Production of bacterial protein biofeed by two-step fermentation of corn husk

This invention belongs to the field of bio-fermentation feed technology, specifically involving a two-step fermentation method for producing microbial protein bio-feed from corn husks. The invention provides a method for preparing microbial protein bio-feed, specifically: inoculating a first fermentation substrate with Aspergillus niger inoculum for a first fermentation to obtain a starter culture; and inoculating a second fermentation substrate with a compound microbial agent for a second fermentation to obtain the microbial protein bio-feed. The preparation method described in this invention can maximize the degradation of cellulose and hemicellulose in corn husk raw materials while considering cost, and increase its protein content through secondary fermentation, providing the feed industry with a method for preparing high-quality protein feed.
Owner:INNER MONGOLIA UNIV OF TECH

A device for repairing existing garbage landfill

A device for repairing an existing landfill comprises a waste gas treatment device, an evaporator, a vapor gas treatment device, a gas-liquid separation regulating chamber, a fixed biofilm reactor, a mobile biofilm reactor, a reverse osmosis treatment device and a power distribution room; the waste gas treatment device, the evaporator and the vapor gas treatment device are arranged side by side at the front end; the waste gas treatment device is connected to the evaporator and the gas-liquid separation regulating chamber respectively through pipelines; the evaporator is connected to the vapor gas treatment device through pipelines; the gas-liquid separation regulating chamber is arranged at the rear of the waste gas treatment device, the evaporator and the vapor gas treatment device; the fixed biofilm reactor is arranged at the rear of the gas-liquid separation regulating chamber through pipeline connection; the mobile biofilm reactor is arranged at the rear of the fixed biofilm reactor through pipeline connection; the reverse osmosis treatment device is arranged at the rear of the mobile biofilm reactor; the power distribution room is arranged at the rear of the reverse osmosis treatment device.
Owner:CHINA RAILWAY ENVIRONMENTAL TECH ENG CO LTD

A wildfire anomaly detection method based on multi-frame dynamic processing of positive samples

The present invention relates to the field of image safety supervision, and provides a wildfire anomaly detection method based on multi-frame dynamic processing of positive samples. The purpose is to improve the accuracy and stability of the wildfire detection algorithm through innovative dynamic processing and positive sample strategies, especially in the application of power transmission scenarios, so as to effectively reduce false alarms and improve detection efficiency. The main scheme includes obtaining a fine data set and training a detection model by annotating wildfire data. The video frame is differentially processed to identify the changed area, and the specific size area is extracted and mapped to N equally divided masks to achieve dynamic target detection. The sky segmentation model and OpenCV process the image, save the outline of the sky area, and filter out false alarms of clouds. The scoring mechanism determines the alarm, and an alarm is triggered when the continuous score exceeds the specified threshold. After the alarm, the fire alarm score is reset, the cooling period is started, and the alarm is suspended.
Owner:SICHUAN SHUJU INTELLIGENT MFG TECH CO LTD

A method and system for reusing convolutional discarded information for fine-grained image classification with small samples

This invention discloses a method and system for reusing discarded convolutional information in small-sample fine-grained image classification, belonging to the field of computer vision. The method includes the following steps: acquiring images of a support set and a query set, and extracting the backbone feature descriptors of the images using a convolutional neural network; after at least one convolutional block in the convolutional neural network, retrieving the discarded feature information of the convolutional block through a feature retrieval path to generate a retrieved feature descriptor; fusing the backbone feature descriptor with the retrieved feature descriptor to obtain a high-dimensional feature descriptor in a unified format; calculating the similarity between the images of the query set and the support set based on the high-dimensional feature descriptor, and performing small-sample fine-grained image classification based on the similarity, thereby realizing the reuse of discarded convolutional information. This invention can effectively retrieve discarded feature information in convolutional neural networks, improving the performance of small-sample fine-grained image classification.
Owner:SHAANXI UNIV OF SCI & TECH

Preparation method and application of hydrophobic confined microbeads based on Ni-based bimetallic biochar

The present invention belongs to the technical field of material preparation and environmental pollution control, and discloses a method for preparing hydrophobic confined microbeads based on Ni-based bimetallic biochar and its Fenton catalytic application. The steps are: (1) culturing sweet osmanthus; (2) preparing Ni-based bimetallic biochar Ni / Fe-MSBC; (3) preparing Ni / Fe-MSBC@hydrophobic confined microbead reactor based on phase inversion technology. The Ni-based bimetallic biochar prepared by utilizing the property of sweet osmanthus that it is easy to enrich metals, and the porosity of the biochar and the dispersion of the metal are further improved by the molten salt method, so that the active sites are more evenly dispersed inside and on the surface of the biochar. In addition, the hydrophilic polymer and the hydrophobic polymer are used as carriers of the photo-Fenton catalyst, and the porous microbead reactor HCBRs with a hydrophobic confined microenvironment is prepared by regulating the blending of the hydrophilic polymer and the hydrophobic polymer. The HCBRs have excellent flexibility and stability, can effectively improve the catalytic activity and catalytic stability, and efficiently degrade antibiotic pollutants.
Owner:SUZHOU UNIV OF SCI & TECH

Human behavior recognition method based on graph structure skeleton data of replay method and related device

PendingCN122712172AAchieve orderly learning abilitySolve the problem of being unable to adapt to dynamic continuous learning scenarios
This application discloses a method and related apparatus for human behavior recognition based on graph-structured skeletal data using the replay method, relating to the field of computer vision technology. The method includes inputting target skeletal data into a trained action recognition classifier to obtain human behavior recognition results. The process of determining the trained action recognition classifier involves: dividing the graph-structured skeletal dataset into multiple disjoint subsets based on a partitioning criterion; treating each subset as a task; and utilizing a combination of buffer compression, composite pruning, and mask-based autoencoder reconstruction tasks, processing each task sequentially in chronological order to ultimately obtain the trained action recognition classifier. This application significantly improves the model's ability to retain memories of old tasks and its learning and adaptation capabilities for new tasks under fixed buffer resource constraints, effectively enhancing the accuracy and robustness of the final action recognition.
Owner:SHANDONG NORMAL UNIV

Resource transfer method and device, computer program product and electronic equipment

The embodiment of the invention relates to the technical field of computers, in particular to a resource transfer method and device, a computer program product and electronic equipment. The method comprises the following steps: acquiring resource transfer signing information of a signing user, and marking the signing user according to the resource transfer signing information of the signing user to obtain marking information; generating resource transfer limiting conditions according to the resource transfer signing information, and determining a target resource transfer limiting condition corresponding to the signed user in the resource transfer limiting conditions according to the marking information of the signed user; in response to a resource transfer request, determining a hit rate that request information in the resource transfer request hits the target resource transfer limiting condition; and determining whether to intercept the resource transfer of the signed user according to the hit rate and the mark information of the signed user. According to the invention, the error rate of resource transfer is reduced.
Owner:HANGZHOU NETEASE CLOUD MUSIC TECH CO LTD

Feeding device for observing vital signs of experimental rats

PendingCN119453082AReduced range of movementquick grabAnimal housingAnimal scienceMechanical engineering
The invention relates to the technical field of experimental animal monitoring, in particular to a feeding device for experimental rat vital sign observation, which comprises a placement frame, a plurality of breeding assemblies placed in the placement frame, a separation assembly fixedly mounted at the lower end of a breeding box, and a limiting assembly movably arranged in the breeding box, the overturning assembly is fixedly installed on the surfaces of one sides of the multiple push plates. According to the invention, a grating cover plate above the breeding box is turned over, so that the grating cover plate drives a plurality of partition plates below the grating cover plate to move upwards, the breeding box is divided into a plurality of spaces, a plurality of white rats are separated, and meanwhile, the partition plates move upwards to drive a push plate to turn over to a vertical state and pull the push plate to move, so that the activity space of the white rats is limited; and when the push plate moves, the stirring roller rotates, strikes and stirs the padding at the bottom of the breeding box and scatters and turns over the padding, the situation that the padding on the lower layer is not fully utilized due to the fact that the padding on the upper layer is bonded into blocks is avoided, and the utilization rate of the padding is increased.
Owner:LIAONING UNIV OF TRADITIONAL CHINESE MEDICINE