37results about How to "Efficient aggregation" patented technology

The invention provides a cross-domain transfer learning laser ultrasonic layering defect detection method suitable for a composite material and application thereof.The method comprises the steps that firstly, multi-channel input containing a frequency domain, a time-frequency domain and Koopman operator characteristics is constructed for the problem of data distribution deviation caused by the batch difference of the composite material; secondly, extracting and fusing high-dimensional spatial-temporal features through a multi-channel attention mechanism and a graph convolutional network; and finally, migrating the defect features of the source domain to the target domain in combination with dynamic distribution self-adaption and domain adversarial training strategies. The method can effectively overcome model performance reduction caused by material attribute change, realizes high-precision layering defect identification and imaging under the condition of no target domain label, and is suitable for composite material detection in a complex industrial scene.

This invention relates to the field of digital archives management technology, specifically to an artificial intelligence-based system for analyzing and warning of abnormal archive behavior. The invention uses a graph construction module to transform archive access logs into a relational topology graph; an attribute propagation module performs weighted transfer and accumulation of weights based on a graph traversal algorithm to calculate the cumulative relational index value of nodes; a connectivity retrieval module retrieves cross-regional links based on partition isolation rules; and an early warning output module generates early warning signals based on index values ​​and link status. This invention utilizes the flow mechanism of feature values ​​in a graph structure to aggregate minute risky behaviors within discrete long-term windows, accurately identifying complex abnormal patterns where a single operation may not violate regulations but the long-term cumulative risk exceeds the limit. This solves the problem that existing static statistical rules are unable to detect hidden logical violations, effectively reducing false alarms and false negatives.

This invention relates to the field of energy forecasting technology, and in particular to a method and system for short-term energy load forecasting based on an improved Transformer using SE-Block. The method includes reversible normalization preprocessing of acquired multivariate load sequence data; feature extraction and fusion of the preprocessed data using improved cross-scale interactive patching, including multi-scale feature extraction, cross-scale interactive alignment, residual correction, and dynamic fusion; and feature filtering of the fused features based on a channel attention mechanism, including feature response based on improved SE-Block and nonlinear interaction of context vectors. This invention addresses the non-stationarity of actual load caused by meteorological conditions and user behavior. By automatically eliminating noise interference among multiple variables, the model accurately depicts the fluctuation details of the load curve, demonstrating its robustness in multivariate load forecasting for integrated energy systems.

The utility model relates to a water supply sewage treatment filtering device which comprises a supporting assembly and a filtering barrel rotationally arranged on the supporting assembly, a barrel is coaxially arranged in the filter barrel, an annular filter cavity is formed between the barrel and the filter barrel, an annular sealing plate is arranged above the cavity, a plurality of layers of concentric circle filter materials are arranged in the cavity in the radial direction, and through holes communicated with the annular filter cavity are formed in a barrel body of the filter barrel and the barrel. 4-12 blades are arranged on the upper portion of the inner wall of the barrel in a circumferential array mode, an end cover is coaxially arranged above the barrel through a bearing, a water inlet is formed in one side of the end cover and connected with a water pipe, the water outlet direction of the water pipe is tangent to the inner wall of the side face of the end cover, and tangential water outlet is matched with the blades to drive the filtering barrel to rotate. The bottom of the filter barrel is conical or arc-shaped, and a drain outlet is formed in the center. The device solves the problems that a traditional filtering device is low in efficiency and needs extra power, hydraulic self-rotation driving and multi-stage annular filtering without external power are achieved, the filtering effect is remarkably improved, and the operation and maintenance cost is reduced.

The application provides a forest fire smoke detection method based on semantic segmentation, and belongs to the technical field of fire warning, and comprises the following steps: feeding a real forest fire data set into a feature extraction network, inputting the extracted features into an improved DeepLabV3+ network for training, and outputting detection and recognition results. According to the complex and changeable forest fire scene, the non-rigid structure, the opacity and the variable mode of smoke and other characteristics, the application deeply studies a feature fusion method, adopts an improved ASPP structure to extract multi-scale spatial features, simultaneously introduces a bottom feature extraction module and a double-path convolution aggregation module to effectively extract smoke boundary information, and uses a lightweight backbone network to replace the original feature extraction network of DeepLabV3+, so that the detection speed is accelerated by using fewer parameters. The application can solve the problem that the false alarm and missed detection rates of the smoke detection method are high, thereby providing technical support for forest ecological protection.

This invention discloses a real-time semantic segmentation algorithm for remote sensing images based on a dual-path structure, comprising the following steps: S1, acquiring Potsdam and GID-15 remote sensing image datasets and preprocessing them, then dividing them into training, validation, and test sets according to a set ratio; S2, constructing a real-time semantic segmentation network for remote sensing images based on a dual-path structure, including spatial path, context path, and dual-path feature aggregation; S3, inputting the preprocessed training and validation datasets into the real-time semantic segmentation network for training, calculating the cross-entropy loss function and performing backpropagation, updating the network parameters, obtaining a model that meets the requirements, and saving the model; S4, inputting the preprocessed test set into the model obtained in step S3, and outputting accurate segmentation of the remote sensing image. This invention improves the richness and effectiveness of semantic feature extraction while enhancing segmentation accuracy.

The invention discloses an encryption aggregation method and system for a household micro-grid, and is applied to the technical field of micro-grid group operation control, and the method comprises the steps: obtaining adjustable data of a plurality of household micro-grids; classifying the adjustable data of the plurality of household microgrids to obtain a plurality of to-be-adjusted data groups; and inputting the plurality of to-be-adjustable data groups into a micro-grid group system at least comprising a plurality of household micro-grids for processing to obtain to-be-aggregated ciphertext data, and performing threshold decryption processing on the homomorphic aggregated ciphertext by using a distributed key generation technology to obtain an aggregation result of the micro-grid group system. According to the method provided by the invention, the technical problems of privacy leakage and data security caused by plaintext uploading and user side sensitive operation data transmission in an existing household micro-grid aggregation mode can be solved, so that privacy protection of the original sensitive data of the household micro-grid is realized.

The application relates to an active load response system based on low communication demand, wherein a central control module and a plurality of frequency modulation modules are connected through distributed control communication; the frequency modulation module comprises a data receiving and transmitting unit, a data processing and storage unit and an active load response adjusting unit; the data receiving and transmitting unit is used for obtaining operation data of the central control module, other frequency modulation modules and power loads; the data processing and storage unit is used for determining consumed power of the active load response adjusting unit according to the operation data obtained by the data receiving and transmitting unit and the operation state of the active load response adjusting unit; and the active load response adjusting unit is used for adjusting consumed electric energy of the load through a communication mode or a direct adjustment mode of load power voltage and frequency. The application solves the technical problems of a large number of devices, small capacity and wide distribution in the prior art, improves the technical performance of device clusters, and provides frequency support and adjustment for a power system.

The present application relates to the technical field of address translation request scheduling, and particularly relates to an address translation request aggregation scheduling method and system. The present application maintains a pending request filter table, each table entry corresponding to a 4KB address range, recording address translation requests that have been sent but have not received responses. When a new request is received, the 4KB base address to which the virtual address belongs is calculated and the filter table is queried: if there is a corresponding table entry, the request is added to the associated waiting queue, and the new request is not sent; if there is no corresponding table entry, the new request is sent and a new table entry is created. Thus, during the waiting response period, subsequent requests in the same address range are aggregated, the number of repeated requests is reduced from N times to 1 time, the root complex load is significantly reduced, redundant transactions are reduced, and the hardware overhead is extremely low.

This invention proposes a method for monitoring network signal operation. By deploying a centralized management service in a cloud computing environment, it performs threshold judgments on aggregated signal data, dynamically acquires additional cloud resources to support monitoring, and simultaneously constructs a cloud-based signal monitoring dashboard to update remote signals in real time, identify abnormal patterns, and integrate them into the network monitoring system. For anomalies related to resource scheduling, it obtains optimized paths, enhances dashboard functionality, and ultimately adjusts management services based on communication feedback data. This method achieves efficient aggregation of signal data, accurate identification of anomalies, and intelligent resource scheduling, significantly improving the reliability and response speed of cross-regional communication monitoring and providing stable support for multi-regional network operation.

The application discloses an AI identification method, system and model training method for millimeter wave image overlapping targets based on a three-branch network, and solves the security target detection problem in millimeter wave images with low resolution, mutual interference or mutual overlap through collaborative work of three independent branch networks, namely, an analysis branch network, an aggregation branch network and a prediction boundary branch network. The analysis branch network saves the detailed information in a high-resolution feature map; the aggregation branch network realizes the feature position offset alignment of high and low resolution feature maps through a recursive fusion module; and the prediction boundary branch network extracts high-frequency features and enhances the perception and prediction of target edges. The multi-branch network structure effectively realizes the segmentation and identification of images through the fusion between branches. While improving the segmentation performance of the model, the model significantly reduces the calculation complexity and the detection error and missed detection probability in millimeter wave images, and finally realizes the accurate detection of overlapping objects.

The invention belongs to the technical field of multi-village government affair service optimization methods, and particularly relates to an internet-based multi-village government affair service optimization method and system, which adopts a mode of combining satellite images, unmanned aerial vehicle surveying and mapping and mobile terminal off-line acquisition APP, grabs land boundary and terrain space data, acquires population household registration and contract business data, and provides a multi-village government affair service optimization method and system. Historical paper archives are identified through OCR batch scanning, and information is extracted and input into an information management system; the method comprises the following steps: constructing an AI and RAG intelligent cleaning model, filling data missing values by applying a generative adversarial network, recognizing population repeated registration and land area conflict abnormal data through a local abnormal factor algorithm, training an entity recognition model by relying on a rural government affair vertical RAG library, and automatically labeling data attributes and association relationships.

The invention relates to the technical field of data integration, in particular to an internet information integration method and device and electronic equipment. The method comprises the following steps: preprocessing Internet multi-source information; performing data analysis and structuring on the Internet multi-source information to obtain a standardized data model; constructing a multi-source evaluation model based on the standardized data model to evaluate the credibility of the Internet multi-source information; performing information fusion and conflict resolution on the Internet multi-source information based on the comprehensive credibility score; and constructing an integrated database based on the Internet multi-source information after information fusion and conflict resolution. According to the invention, by constructing a multi-source acquisition engine, a heterogeneous analyzer, a credibility evaluation matrix and an intelligent fusion engine, automatic, structured and credible integration of enterprise-related internet information is realized.

The invention provides an alarm noise reduction method and system based on a large model of a power system, and the method comprises the steps: obtaining multi-source heterogeneous alarm data in the power system, and constructing a preprocessing data set; constructing a heterogeneous information network based on the preprocessed data set, generating a node sequence containing alarm semantics by using the heterogeneous information network, and aggregating a plurality of node sequences meeting a similarity requirement into an alarm cluster; generating an alarm cluster feature data structure based on the alarm cluster; obtaining a knowledge graph and historical alarm data of the power system, and constructing a large power threat detection model; and inputting the alarm cluster feature data structure into the large power threat detection model to generate an alarm research and judgment report of the power system. The problem that high-risk events in an existing power system are hidden in massive alarms and cannot be found in time is solved.

The invention belongs to the technical field of crop variety innovation, and particularly relates to a breeding method and application of wheat capable of resisting gibberellic disease and herbicide-tolerant starter propagation. According to the method disclosed by the invention, through combination of molecular marker-assisted selection and backcross breeding, excellent characters from different donors are rapidly introduced into a genetic background of an excellent koji-making variety, so that accurate and efficient polymerization of target characters is realized, and the screening efficiency and accuracy can be improved; meanwhile, the screening index of starter propagation quality runs through the breeding process, so that the bred material is not only excellent in resistance and agronomic character, but also can meet the specific requirements of the terminal industry. The variety bred by the method provided by the invention can reduce the use of pesticides for preventing and treating gibberellic disease and reduce the pollution risk of fungaltoxin; the herbicide-resistant characteristic facilitates field weed management, and the yield and quality stability are improved; the special quality guarantees the high additional value in the starter propagation industry, and creates the value of the whole industrial chain from the field to the workshop.

This application provides an image contrast enhancement method, system, electronic device, and storage medium. The method includes: separating luminance and chrominance components from the YUV data of an input image; training the luminance component for contrast enhancement using an enhancement network model to obtain an enhancement coefficient grid for the luminance component and compensation coefficients for the chrominance component; enhancing the luminance component using the enhancement coefficient grid and enhancing the chrominance component using the compensation coefficients; and obtaining an enhanced image based on the enhanced luminance and chrominance components. This application can significantly reduce computational complexity while maintaining contrast enhancement effects, and can also suppress noise and eliminate video flicker.

The invention belongs to the field of cassava molecular markers, and particularly relates to a cassava KASP marker and application of the cassava KASP marker in cassava variety distinguishing. The 27 KASP markers provided by the invention have good distinguishing ability in cassava genetic analysis, and can effectively reveal genetic differences among different materials. The DNA fingerprint spectrum of 184 parts of cassava materials is constructed based on 27 core KASP markers, and the purpose of rapid variety identification can be achieved. The KASP marker disclosed by the invention can be used for effectively identifying false selfing offspring. In conclusion, the KASP marker provided by the invention provides a new method and thought for genetic improvement and variety identification of cassava, provides a powerful tool for subsequent cassava breeding and genetic improvement research, and is beneficial to breeding of new varieties with excellent characters.

The present application relates to geothermal water treatment technical field, disclose a kind of geothermal water water gas separation device, including tank body, tank cover being located at the top of tank body, inlet being located at the top of tank body, drain outlet being located at the bottom of tank body and exhaust port being located at the top of tank cover, further including multistage water collecting assembly being located in the middle of the inner cavity of tank body and resonance separation assembly being located in the inner cavity of tank body;Multistage water collecting assembly includes three groups of arc water collecting pipes being spaced apart along the axial direction of tank body and spiral water injection pipe being communicated with inlet;Resonance separation assembly includes umbrella-shaped water catcher being fixedly installed in tank cover, vertical rod being fixedly connected in the center of umbrella-shaped water catcher and extending downward, and three groups of floating units being arranged on vertical rod;The present application is provided with floating unit, so that conical screen disc produces continuous vibration when fluid impact, and the vibration forms shear force and local disturbance in fluid, which can break the stable state of micro-bubbles in water body, promote its desorption and collision coalescence.

The invention discloses a plant traditional Chinese medicine qi detection device and detection method, and relates to the technical field of electromagnetic ray detection, the device comprises a detection cavity with a wave absorbing layer laid in a camera obscura, and a bearing platform for placing a potted plant; a reflection energy-gathering plate is hung on the wave absorbing layer around the bearing platform so as to reflect and gather weak electromagnetic waves, rays or particle flows naturally released by plants for multiple times; a power sensor is arranged at the top of the camera obscura and faces the bearing platform, and an original signal is output after an energy signal subjected to energy gathering is collected; a signal processing module is arranged outside the camera obscura and is used for carrying out amplification, filtering, broadband decomposition, characteristic energy level extraction and fusion analysis on the collected original signals and outputting processed data to a display module so as to display power data of plant gas field particles or traditional Chinese medicine gas indexes; therefore, ultra-weak energy signals released by the plant in a natural state are captured with high precision, broadband, multi-dimensional and multi-physical field coupling feature extraction and quantitative detection of plant gas field particles are realized, and the method has the advantages of high sensitivity and high signal-to-noise ratio.

The invention relates to a Mars hyperspectral image classification method based on CNN and GCN, and belongs to the technical field of remote sensing image classification. The method comprises the following steps: constructing a double-branch hybrid neural network comprising a CNN branch and a GCN branch; wherein the CNN branch is of a lightweight network structure and comprises a denoising and dimensionality reduction module, a depth separable convolution module and a multi-scale convolution module; the GCN branch is composed of an HGNNConv network; inputting a to-be-classified Mars hyperspectral image into the double-branch hybrid neural network, and respectively extracting pixel-level features and super-pixel-level features of the Mars hyperspectral image through a CNN branch and a GCN branch; and introducing a channel attention mechanism and a space attention mechanism, carrying out weight distribution and fusion on the pixel-level features and the super-pixel-level features to obtain comprehensive features, and completing classification of the Mars hyperspectral images based on the comprehensive features. The objective of the invention is to solve the technical problems that pixel-level and super-pixel-level features cannot be fused efficiently and distortion is easy to occur in a feature propagation process in the prior art.

A distributed battery pack monitoring method, device, and storage medium include: a battery monitoring sensor group uploading multi-frequency internal resistance, voltage, and temperature data to a main control unit via a daisy-chain communication unit; the main control unit sending real-time battery parameters to an LCD display for direct display; the main control unit processing and plotting the multi-frequency internal resistance using intelligent algorithms to obtain the battery's internal resistance spectrum; the main control unit sending the internal resistance spectrum to a battery capacity assessment host or battery monitoring software platform to estimate capacity and provide real-time temperature feedback; a storage unit storing the monitoring data; and a switch output unit outputting fault status to peripheral devices. This invention comprehensively monitors the real-time operating status of the battery by collecting multiple key parameters. The daisy-chain communication structure enables efficient aggregation and uploading of multi-point data. The intelligent algorithm analyzes and processes the internal resistance data to accurately plot the battery's internal resistance spectrum.

The invention discloses a two-stage compression denoising imaging method for a large-area array Geiger mode APD (avalanche photo diode). The method comprises the following steps: intercepting photon events in a pre-sampling time window at the front part of a data block to carry out global histogram statistics, fitting and estimating time domain characteristics of background noise based on Poisson distribution, screening and compressing full-frame data, and eliminating most noise; carrying out local joint histogram statistics and peak value extraction on the compressed data based on a pixel space neighborhood, and reconstructing a preliminary distance image and a preliminary intensity image; if the effective data in the neighborhood of the data meets the flat area condition, standard median filtering is executed; otherwise, starting a robust filtering algorithm based on absolute median difference MAD for processing. According to the method, the cooperation of front-end data efficient compression and rear-end accurate denoising and edge preserving is realized, and the method has the advantages of high compression ratio, high adaptability and excellent imaging quality, and is particularly suitable for real-time three-dimensional imaging of a large-area-array GM-APD laser radar system.

This invention relates to the field of wind turbine blade inspection technology, specifically to an automatic inspection method and system for offshore wind turbine blades using unmanned aerial vehicles (UAVs) based on multi-airport collaboration. Based on a state-aware dataset of offshore wind turbine blades, this invention generates health values ​​that accurately characterize the blade's health status through multi-indicator calculation and weighted fusion, enabling automatic anomaly detection. Abnormal blades are then spatially clustered to form spatially concentrated inspection groups. Furthermore, by quantitatively assessing the task scale and calculating the comprehensive suitability of each UAV, the optimal UAV resources are selected to construct a UAV formation. Under the premise of strictly avoiding various no-fly zones, a flight path conforming to the shortest path principle is planned for each UAV in the formation, thereby achieving full-link automation from anomaly detection to task execution. This significantly improves the intelligence level, operational safety, and overall economic benefits of offshore wind farm operation and maintenance.

The application discloses a large-scale two-dimensional label map aggregation node optimization method, relates to the technical field of computers, and comprises the following steps: when a map interactive operation or label node data changes, the zoom level and the viewport range of a current prompt map are acquired; based on the viewport range, target label nodes in the viewport range are screened from the label node data; according to the zoom level, the basic aggregation parameters for label node aggregation are determined; and combined with the node density of the target label nodes in the viewport range, the basic aggregation parameters are dynamically adjusted to obtain the aggregation parameters for the current viewport range. The large-scale two-dimensional label map aggregation node optimization method significantly reduces the memory occupation and the calculation burden caused by repeated creation of node objects, improves the overall operation stability and the resource utilization efficiency of the system, and thus better meets the comprehensive requirements of performance, real-time performance and display effect of a large-scale two-dimensional label map application scene.

The invention relates to the technical field of remote sensing image processing, and discloses a remote sensing image change detection method, which comprises the following steps: acquiring a dual-time-phase to-be-detected remote sensing image; inputting the dual-time-phase to-be-detected remote sensing image into a preset remote sensing image change detection model to obtain a detection result of the target area; the remote sensing image change detection model comprises a hybrid encoder and a multi-scale progressive fusion decoder, and the hybrid encoder is used for extracting a multi-scale feature map of a dual-time-phase remote sensing image to be detected; a change feature extraction module of the multi-scale progressive fusion decoder is used for extracting a change feature map of a corresponding scale from each scale feature map; and a multi-stage progressive attention module of the multi-scale progressive fusion decoder is used for carrying out feature enhancement on the change feature patterns of each scale, and carrying out channel splicing and fusion on each enhanced change feature pattern to obtain a detection result. The method can improve the accuracy of remote sensing image change detection. The invention further discloses a remote sensing image change detection device and electronic equipment.

The application discloses a preparation method of polyolefin ether, and belongs to the technical field of polymer preparation. An initiator is mixed with an olefin ether monomer, and after reaction, an inert solvent is used for precipitation to obtain the polyolefin ether; the initiator comprises trifluoromethyl sulfonate. The application aims at the defects of the initiator for cationic polymerization of ether monomers, such as high cost, difficulty in obtaining, difficulty in storage, unsuitability for industrialized scale generation of corresponding high molecular materials and the like, and proposes the trifluoromethyl sulfonate initiator which is low in cost, simple in structure, high in activity, very convenient to use, simple in cationic polymerization principle of ether monomers, easy to operate, capable of carrying out polymerization at room temperature, adjustable in product molecular weight and molecular weight distribution, excellent in performance, and applicable to the fields of thermoplastic plastics, rubber, elastomer, adhesive, paint and chemical additive and the like.

The application discloses a grassland fertility dynamic evaluation method and system based on carbon-nitrogen ratio and carbon-clay ratio, which comprises multi-source data acquisition, data preliminary processing, multi-source data fusion, fertility evaluation model construction and grassland fertility dynamic evaluation. The application obtains original data through multi-source data acquisition; adopts a data preliminary processing method of time and space benchmark unification, remote sensing image preprocessing, soil data quality control and data set segmentation; adopts a method of taking carbon-nitrogen ratio and carbon-clay ratio as core indexes, deducing them from discrete sampling points to continuous spatial surface and fusing multi-modal data, so that the generated surface data has spatial continuity and contains rich ecological correlation semantics, providing a reliable and physically meaningful data basis for subsequent evaluation; and adopts a method of constructing a modal interaction graph and using a graph neural network as an evaluation model, so that the evaluation is not only based on the properties of the unit itself, but also fully considers the synergistic influence of the surrounding environment and ecological background.

The invention relates to a cloud side end collaborative intelligent monitoring platform for a container yard and a scheduling method. The platform adopts a three-layer architecture in which a terminal sensing layer, an edge computing layer and a cloud intelligent layer are coordinated. The terminal layer is responsible for multi-source data acquisition; the edge layer is responsible for regional data fusion and real-time processing to realize low-delay control; the cloud is responsible for global modeling, macroscopic optimization and model training. The scheduling method comprises a task receiving and analyzing step, a cloud global pre-scheduling step, an edge local real-time scheduling and dynamic optimization step, a terminal instruction execution and state feedback step and a cloud learning and model iteration step. According to the platform disclosed by the invention, efficient aggregation, processing and intelligent application of data are realized by constructing a'cloud-edge-end 'three-layer cooperative computing architecture, so that the real-time performance and accuracy of global situation awareness of the storage yard are improved, equipment scheduling and path planning are optimized, and finally the purposes of improving the working efficiency of the storage yard, reducing the operation cost and enhancing the safety are achieved.

The invention provides a spectral absorption three-dimensional imaging system and method based on parallel light beams and area array detection, relates to the technical field of optical precision measurement and tomography, and aims to solve the problems that high time and high spatial resolution, low system complexity and low cost cannot be compatible at the same time and three-dimensional field distribution cannot be quickly reconstructed with high precision in the prior art. The system comprises a laser emission and modulation module, a beam shaping module, a to-be-measured area, an area array sensor and a data processing module. The light beam shaping module, the to-be-detected area and the area array sensor are arranged on the same plane axis, the light beam shaping module comprises a first collimating mirror and an imaging module which are sequentially connected along a light path, the first collimating mirror forms parallel light beams, and the imaging module performs multi-angle rear two-dimensional imaging; and the area array sensor is electrically connected with the data processing module, is processed and controlled by the upper computer, and reconstructs a two-dimensional image into a three-dimensional field. According to the invention, problems in the prior art are solved, and rapid and high-precision three-dimensional imaging is realized.

The invention provides a voiceprint anomaly recognition method and system based on dynamic threshold and multi-feature fusion, and relates to the technical field of audio signal processing, mode recognition and artificial intelligence. The method aims at solving the problem that in the prior art, a judgment threshold cannot adapt to environment changes and registration model credibility changes at the same time. The method comprises the following steps: acquiring a voiceprint audio to be detected and a registered voiceprint library; extracting and fusing at least two different types of acoustic features to generate fused features; inputting the fusion feature into a deep neural network model to generate a voiceprint embedding vector; calculating a similarity score; determining a dynamic threshold according to the at least one environment quality index of the voiceprint audio to be detected and the at least one data quality index of the registered voiceprint library; and comparing the similarity score with a dynamic threshold value, and outputting a judgment result. The invention further provides a system for executing the method. According to the scheme, the judgment standard can be adaptively adjusted, and the accuracy and robustness of voiceprint recognition are comprehensively improved.