93 results about "Multi feature" patented technology

This invention discloses a clutter discrimination and suppression method based on multiple echo features, comprising: acquiring radar front-end clutter and target echo detection video data and dividing it into several connected regions; extracting multi-dimensional features and labeling prior information for each connected region; using the Relief feature selection algorithm to calculate and filter the weights of the multi-dimensional features, removing redundant features to obtain the optimal feature vector; using this feature vector to train a support vector machine classifier, and obtaining the optimal parameter model through cross-validation; acquiring measured clutter data, extracting corresponding features and inputting them into the model; and determining whether to remove clutter and retain targets based on the output. This invention effectively filters out dynamic clutter spots, avoids false deletion of weak targets, significantly reduces the false alarm rate of the system, and alleviates the computational burden of subsequent track processing.

This invention discloses a warehouse management control system and method based on reinforcement learning, relating to the field of intelligent warehouse management technology. The method includes first acquiring the warehouse spatial location structure, product location information, and product set; then performing multi-feature processing on this information to obtain a joint feature vector. Next, using a product and location selection function, the corresponding selection probability is calculated based on this vector. Then, through simulated allocation and calculation of action rewards, the rewards are substituted into a loss function to update parameters, and this process is repeated until the probability change rate reaches a target, obtaining the final state probability. Finally, the actual physical allocation of products is performed accordingly. This scheme leverages multi-feature processing and function calculation to quantitatively analyze selection, solving the problem of high-dimensional state space representation; it addresses the action space explosion by iteratively optimizing parameters, reducing computational complexity; and it uses simulated allocation action rewards to drive parameter updates, overcoming the limitations of sparse rewards, improving training efficiency, and thus enhancing warehouse management control efficiency.

The application relates to the field of point cloud registration, and provides a point cloud registration method and system based on a multi-feature point point set. The method comprises the following steps: preprocessing point cloud by statistical noise reduction based on Gaussian distribution and voxel downsampling; extracting multi-feature points by adopting a directional bounding box algorithm, an improved internal morphological descriptor algorithm and a geometric centroid point algorithm; constructing a multi-feature point point set; and obtaining an initial transformation matrix by adopting a principal component analysis-based coarse registration algorithm; performing fine registration by adopting a G-ICP algorithm based on L2LOSS robust loss function optimization; and outputting a registration result. The application improves the accuracy and efficiency of point cloud matching and reduces the registration time.

This invention relates to the field of new energy power generation control technology, and discloses a state space construction and maximum power point tracking system based on multi-feature information. The system includes: a state vector construction module for constructing the state vector of the current control cycle; a state space mapping module for determining the current unique operating state unit; a candidate interval generation module for generating candidate voltage intervals for the maximum power point; an interval verification and retention module for determining the retained voltage interval for the next control cycle; a target adjustment control module for determining the target voltage and the duty cycle of the next control cycle; a hysteresis boundary determination module for outputting a state hold command or a relocation trigger command; and a lock-and-relocate module for determining the maximum power point lock voltage or triggering relocation. This invention achieves stable tracking and relocation control of the maximum power point under complex operating conditions.

This invention discloses a method for extracting ski slopes from remote sensing images based on a multi-feature collaborative enhancement network, belonging to the field of satellite remote sensing image processing and application technology. It aims to solve the problems of discontinuity, detail loss, and insufficient generalization ability in existing methods when extracting long, curved ski slopes with complex backgrounds in ski resorts. Specifically, the network first designs a cloud and snow perception module and a spatial context perception module to effectively suppress spectral interference and fuse multi-scale environmental features. Second, it innovatively introduces a bi-branch graph convolution module to model long-range dependencies in both the spatial and channel dimensions, enhancing the topological continuity of the ski slope target. Furthermore, it embeds a direction perception module in the encoder to capture the multi-directional linear patterns of the ski slope, and employs bidirectional dynamic serpentine convolution in the decoder to adaptively recover complex curve structures. Finally, it progressively restores the feature map resolution and refines the target contour reconstruction to obtain a binary segmentation map.

This invention discloses a method and system for judging the anti-interference threshold of power system measurement point data. The method includes: collecting raw data from power system measurement points and preprocessing it; inputting the preprocessed measurement point data into a Kalman filter model; setting a normal data threshold range and a fluctuation threshold according to the type and operational requirements of the power measurement points, and calculating the fluctuation characteristic parameters of the data; establishing anti-interference judgment rules, comparing the fluctuation characteristic parameters with preset thresholds, and distinguishing between normal data fluctuations, external interference, and real anomalies through multi-feature collaborative judgment; if interference or data anomaly is determined, triggering corresponding early warning and verification mechanisms; if the verification is correct, outputting early warning information; if the verification is abnormal, adjusting the filter parameters or threshold range. This invention can improve the accuracy and stability of power system measurement point data monitoring.

This invention discloses an intelligent method for judging tea category and grade based on multi-feature selective interaction, comprising: 1. collecting and preprocessing tea image data to construct a tea dataset; 2. constructing a multi-feature selective interaction classification network, including: a deep image semantic feature extraction module, a color-texture low-level feature extraction module, a feature alignment module, a selective interaction module, and a random forest ensemble learning classification module, and processing the tea data to obtain prediction results; 3. constructing a joint loss function based on the prediction results and the true labels to train the classification network, obtaining a trained classification model to achieve tea category and grade classification. This invention can effectively suppress redundant information in the fusion process through the selective interaction mechanism, improving the accuracy and robustness of tea category recognition and grade classification.

This application provides an intelligent image comparison method and related equipment for the hydrophobicity of synthetic insulators using multiple features, belonging to the field of power equipment testing technology. The application includes: acquiring multiple reference standard images and multiple images of insulators to be tested; performing image preprocessing on the reference standard images and the images of insulators to be tested respectively to obtain multiple preprocessed reference images and multiple preprocessed images to be tested; constructing multiple image pairs; for each image pair, extracting multiple image features and calculating a set of single-feature similarities corresponding to the multiple image features; performing weighted fusion on the set of single-feature similarities based on a dynamic weighted fusion algorithm to obtain a comprehensive similarity score for the image pair; determining the hydrophobicity level of the insulator image to be tested relative to the reference standard image in the image pair based on the comprehensive similarity score of each image pair, and outputting the comparison result. This application can improve the accuracy of judgment through multi-feature fusion and comprehensive comparison of multiple reference samples.

The application discloses a kind of based on multi-feature decomposition's electricity sales volume prediction method, device, equipment and medium, the method includes: obtaining the electricity data and exogenous influence data of target area, and the electricity data and exogenous influence data are preprocessed, obtain the standardized target electricity sequence;Based on target electricity sequence and exogenous influence data, exogenous feature is constructed, and target electricity sequence is decomposed in multiple ways, and multiple scale sequence component is obtained;According to the feature of each multi-scale sequence component, the component prediction is carried out to the corresponding prediction model, and each component prediction result is obtained, and each component prediction result is synthesized according to decomposition rule, and initial electricity sales volume prediction result is obtained;Based on rolling prediction mechanism, initial electricity sales volume prediction result is evaluated and selected, and final electricity sales volume prediction result is obtained.The technical scheme can improve the prediction accuracy of electricity sales volume, enhance the business adaptability and stability of prediction result.

This invention discloses a method for classifying defects in patternless wafers based on multi-feature joint decision-making, belonging to the field of wafer defect detection technology. The method includes the following steps: irradiating the surface of a patternless wafer with a probe beam having a preset polarization state; acquiring spatial scattered light intensity information of the scattered light from the wafer surface under different combinations of polarization states based on a probe array distributed at multiple spatial angles; extracting local optical features of the candidate defect regions and constructing a feature tensor based on these local optical features; analyzing the candidate defect regions according to the polarization degradation evolution law of the feature tensor in the spectral dimension and outputting the defect classification result. This approach avoids the low classification accuracy caused by highly similar scattering features of small defects, significantly improving the accuracy of the detection results.

The application discloses a kind of based on multi-feature fusion's gas turbine rub-impact fault diagnosis method, step one, the running data of gas turbine is collected in real time by multiple sensors;Step two, feature extraction is carried out to vibration data, and the key time-frequency features related to rub-impact are calculated;Step three, for speed data, the rub-impact energy index is calculated according to slight collision energy model;Step four, feature extraction is carried out to temperature data, and whether temperature anomaly appears is judged according to temperature feature condition;Step five, the weight of each feature is adjusted, and the weighted algorithm is used to fuse multiple features;Step six, the fault decision threshold is determined using ROC curve, and whether the fault occurs is judged;The application can improve the precision of gas turbine rub-impact fault diagnosis, monitor the running state of potential machine in real time, give early warning to abnormal condition, and adjust the operating condition of gas turbine according to the diagnosis result, so as to reduce the probability of rub-impact fault.

This invention relates to the field of target range definition technology, and in particular to a method for target range definition based on gesture recognition. The technical solution includes: target definition and command issuance can be completed through gestures, improving the naturalness and convenience of human-computer interaction; through multi-feature point recognition, perspective correction, anomaly point screening, and edge smoothing, the accuracy and stability of the defined range are improved, enhancing the accuracy and robustness of target definition; the system has automatic and manual adjustment modes, and can autonomously optimize or prompt user intervention in different environments, improving the overall reliability of the system and enhancing its adaptability and fault tolerance; the system can not only define the range, but also analyze multiple targets within the range and automatically select the optimal target to follow based on a preset strategy; through a layered architecture design, gesture recognition, data processing, and motion control are organically combined to ensure that the robot can accurately understand the user's intentions and execute stable and continuous following actions.

The application relates to the technical field of gene sequencing, in particular to a CNV detection method based on an isolated forest algorithm, a medium and equipment. The method comprises the following steps: obtaining genomic sequencing data of a CNV event introduced in a reference genome from a Bam file of single-cell DNA sequencing; adopting a variable window strategy to divide the reference genome into windows, and adjusting the windows through a consistent read quantity to obtain a segmentation site information file; calculating RD signal values in each window according to the segmentation site information file and the genomic sequencing data, and extracting PEM signals to obtain PEM signal values; performing multi-feature calculation and analysis on the RD signal values and the PEM signal values based on the isolated forest algorithm; and identifying CNV events in each window according to an analysis result. The application can effectively identify CNV events in single-cell DNA sequencing data, not only overcomes the limitations of traditional methods in single-cell DNA sequencing data, but also improves the accuracy and reliability of CNV event detection.

The present application relates to the technical field of machine vision, in particular to a kind of robust visual detection system of ton bag goods based on multi-cluster feature fusion and posture correction drive, the system includes non-local mean enhancement module, multi-feature clustering fusion module, geometric feature extraction module, dominant posture clustering module, posture correction transformation module and goods segmentation determination module, non-local mean filtering is carried out to the original image of ton bag goods to obtain enhanced image, the present application is equipped with non-local mean enhancement module and multi-feature clustering fusion module, can realize the efficient enhancement of ton bag goods image and multi-dimensional feature fusion, the space alignment weighted superposition of color cluster and texture cluster is generated Fusion segmentation chart, let the feature representation of ton bag goods homogeneous region be more comprehensive, accurate, effectively improve the integrity and effectiveness of geometric feature extraction, so that the feature analysis ability of system to ton bag goods image is significantly improved.

This invention relates to the field of wireless communication system technology, specifically to a discriminative enhancement fusion modulation recognition method for confused modulation categories, comprising the following steps: preprocessing the signal to construct multi-channel signal features for the input model; constructing a received signal model including a discriminative enhancement residual convolutional feature extraction module, a confused sensing temporal discriminative coding module, and a discriminative enhancement classification module, with the multi-channel signal features serving as input for the final classification. On the one hand, the signal is preprocessed to construct multi-channel signal features as the input basis for subsequent models; on the other hand, through multi-feature collaborative modeling, a discriminative enhancement mechanism is introduced to allocate the modeling capability of the model to confused modulation categories. By making targeted modifications to key functional modules and performing collaborative fusion at the overall network level, the model can progressively enhance its discriminative capability for confused modulation categories at multiple stages, including feature extraction, temporal modeling, and classification decision.

The application discloses the field of reciprocating mechanical state anomaly detection, and particularly relates to a drilling pump anomaly detection method based on multi-feature distance fusion and lateral comparison, which comprises the following steps: S1, collecting a reciprocating drilling pump fluid end vibration acceleration signal; S2, waveform abnormal vibration signal preprocessing; S3, calculating multi-dimensional time domain features and lateral comparison values; S4, calculating drilling pump fluid end abnormal state representation indexes; S5, calculating drilling pump fluid end abnormal state threshold values; and S6, deployment and application. The drilling pump anomaly detection model based on multi-feature distance fusion and lateral comparison calculates the degradation degree of different cylinder multi-dimensional time domain features of the vibration signal, quantifies the distance of the equipment offset health state center, enhances the sensitivity of the drilling pump fluid end degradation representation, and achieves the monitoring purpose of identifying the drilling pump fluid end abnormal state.

The application discloses a kind of based on the fault identification method of double-fed wind farm sending line of multi-feature combination, comprising the following steps: S1.builds double-end transmission line simulation model, sets different fault conditions and generates simulation fault data, according to simulation fault data, the label mark of fault type is carried out, obtains the voltage and current simulation fault data of double-fed wind turbine;S2.according to voltage simulation fault data and current simulation fault data, extract fault voltage feature vector and fault current feature vector, and form multidimensional fault feature vector;S3.construct the neural network of fault identification, multidimensional fault feature vector is input into neural network, neural network is iteratively trained, until the error of the output of neural network and label tends to be stable;S4.acquire double-fed wind turbine real-time fault data, extract fault voltage and fault current feature vector, form multidimensional fault feature vector input neural network, neural network outputs fault type probability, the greatest fault type probability is determined as fault category.

This utility model discloses an intelligent switchgear operation control device based on multi-feature information monitoring, including a cabinet and a detection device. The detection device includes a control device and temperature, humidity, smoke, and pressure sensors. The cabinet has a camera, fire extinguishing device, cooling fan, and rain protection device. The temperature sensor works with the control device to adjust the fan speed; the humidity sensor controls the rain protection device to shield the fan from rain; the smoke sensor triggers the fire extinguishing device; and the pressure sensor starts the camera to record video and upload it to the monitoring terminal when the cabinet door is opened. This device realizes intelligent monitoring and control of multi-feature information of the switchgear, improves operational safety and reliability, adapts to rural exterior wall installation environments, and reduces failure and maintenance costs.

The application discloses a footprint optical image retrieval method based on a multi-feature expansion residual convolutional network, and steps are as follows: 1, data collection and data preprocessing of the footprint image; 2, establishing a footprint optical image retrieval model based on a multi-feature expansion residual convolutional network enhanced by an attention mechanism, including: an initial feature extraction module, an expansion residual convolutional network module, a multi-scale feature fusion module and an attention enhancement module; 3, training the footprint optical image retrieval model to obtain a trained footprint image retrieval model, which is used for footprint retrieval and identity matching of a footprint image to be retrieved. The application can deeply mine various hidden features of the shallow layer and the deep layer of the footprint image, so that the retrieval accuracy and the retrieval efficiency of the footprint image are improved.

This invention discloses a mirror feature enhancement training method and apparatus based on a prototype network, relating to the field of intelligent signal feature analysis and recognition technology. The method includes: splitting the end-to-end deep learning model to be enhanced into a generalized encoding layer, a generalized decoding layer, and the original loss function; constructing a mirror prototype network training framework with left and right branches, where the left branch has the same structure as the original model, and the right branch consists of an isomorphic encoder and a distance calculation layer; optimizing the framework through a combination loss function; and completing feature enhancement training using an alternating fixed branch weight strategy. This invention does not require significant modifications to the original model, allows the model to acquire more feature data, and the framework is pluggable to adapt to automated training frameworks, is compatible with multiple models, has strong adaptability, and can be extended to other small-sample recognition scenarios for data processing.

An automated email processing method based on multi-feature hierarchical classification includes the following steps: Step 1, retrieving emails received from a third-party email address in real time or at regular intervals and storing them in a database; Step 2, periodically loading emails containing predetermined information from the database into a cache; Step 3, matching the corresponding information in the cached emails with the identification code in our system and distributing them to our system; Step 4, replying and / or forwarding the emails through the third-party email address based on the information distributed by our system.

The application discloses a kind of multi-feature supported spillway remote sensing identification method and system, and the object level target is obtained by image segmentation algorithm;Spectrum, shape, geometric stereo feature of object level target is extracted, and preliminary extraction of spillway object is realized by multi-feature joint decision;The spatial direction and distance relationship between objects are comprehensively considered, and the fine spot around the preliminary identification result is identified and integrated again, contour modification is carried out simultaneously, and finally the relatively neat spillway boundary vector is output.This method specifically designs the geometric stereo feature parameters of spillway to highlight the unique features of spillway in three-dimensional space, effectively overcoming the difficulty of distinguishing spillway from general buildings.At the same time, the method also provides an automatic remote sensing spillway identification process to improve the accuracy of spillway identification, which has important practical application value in the fields of spillway remote sensing monitoring and water conservancy construction.

This invention discloses an unmanned vehicle path planning and decision-making system and method based on multi-feature map fusion, relating to the field of autonomous driving road planning technology. It addresses the technical problem that existing single-feature map-based planning and decision-making methods cannot meet the requirements of special scenarios such as construction sites and mining areas where structured roads are defined but the environment is constantly changing and new areas requiring brief exploration may arise. The key technical points of this invention include: acquiring road point data and constructing structured roads based on a positioning and perception module; obtaining a raster map through 3D point cloud map conversion; and invoking the planning program based on the nearest points of the vehicle's current position and the structured roads, as well as the nearest points of the destination position and the structured roads. This invention, through modular encapsulation, can directly perform path planning on structured roads that conforms to vehicle dynamics characteristics, while simultaneously enabling rapid path planning for target points outside the road using A-satellite navigation. This improves the generalization of autonomous navigation under structured maps, solves the planning problem for target points outside structured roads, and achieves highly efficient, generalized navigation decision-making based on structured roads.

This application provides an intelligent noise removal method and apparatus based on multi-feature deep fusion, relating to the field of petroleum geophysical exploration data processing technology. The method includes: designing an observation system according to a pre-defined geological task; dividing seismic gather data into N regions and selecting low signal-to-noise ratio (SNR) shots; identifying data above the first arrival time window as the first type of noise data; selecting labeled shots and standardizing and quality-evaluating the labeled shot data; adding noise data from a noise database to the labeled shots according to a set weight ratio, forming noisy shot gather data through data fusion technology, adding gain, and standardizing; constructing a deep neural network model; generating training samples based on the labeled shot data through data augmentation and sample partitioning techniques; and removing noise through feature analysis and learning of the input data using the model to obtain high-quality data. The modular process and flexible parameter configuration of this application facilitate integration with existing processing systems, demonstrating strong engineering practicality and promotional value.