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

19results about How to "Solve the robustness problem" patented technology

Road obstacle recognition method based on machine vision

The invention discloses a road obstacle recognition method based on machine vision, and the method comprises the steps: carrying out the precise positioning of the spatial position of an obstacle through a multi-sensor information fusion algorithm according to the azimuth and distance information of the obstacle estimated through a sound source positioning algorithm and the position information outputted by an obstacle detection model, and improving the positioning precision; the method comprises the following steps: acquiring guide mark information, such as road signs and lane lines, extracting guide mark features through a semantic segmentation algorithm, carrying out dynamic fusion on the guide mark features and an obstacle recognition result, and improving the accuracy and robustness of obstacle recognition by using prior information; in order to solve the problem of insufficient parallax analysis precision in long-distance, high-speed motion and other scenes, a monocular depth estimation algorithm based on deep learning is introduced, pixel-level depth information is predicted from a single image through a convolutional neural network, and the precision of obstacle distance estimation is improved.
Owner:CHONGQING NORMAL UNIVERSITY

Multipath channel parameter reconstruction and beam coverage prediction method

PendingCN121966764AHigh physical fidelityOvercoming the problem of high coherenceSpatial transmit diversityTransmission monitoringAlgorithmChannel parameter
The invention discloses a multipath channel parameter reconstruction and beam coverage prediction method, and relates to the technical field of wireless communication. Aiming at the problems of low data utilization efficiency, limited environment modeling precision, strong coupling of measured data and antenna beam configuration and the like in the prior art, the method comprises the steps of configuring antenna beams and orientation parameters, collecting reference signal receiving strength, performing deterministic channel modeling simulation, establishing a mapping model, performing sparse recovery, constructing a channel propagation model, evaluating a channel map and the like. In combination with a coherence perception weighting network (CARE-Net) algorithm and ray tracing (RT) physical prior guidance, accurate reconstruction from a low-dimensional reference signal receiving strength (RSRP) measurement value to a high-dimensional multipath parameter is realized. The method has the advantages that the reconstructed multi-path parameters are decoupled from the antenna configuration, the method has the cross-beam / cross-configuration generalization prediction capability, the channel coverage performance under different antenna configurations can be accurately predicted, a reliable basis is provided for wireless network optimization, and the method is low in cost, high in precision and strong in generalization.
Owner:XIAMEN UNIV

A permanent magnet synchronous motor control method and system based on second-order active disturbance rejection control

ActiveCN113726240BRealize drive controlSolve the robustness problemElectronic commutation motor controlVector control systemsControl signalMathematical model
The present invention relates to a permanent magnet synchronous motor control method and system based on second-order auto-disturbance rejection control, which belongs to the field of permanent magnet synchronous motor control technology and solves the problems of poor robustness and weak anti-disturbance capability of traditional auto-disturbance rejection controllers. Based on the established d-q axis mathematical model of the permanent magnet synchronous motor, a state equation is obtained; a second-order auto-disturbance rejection controller is established; based on a model reference adaptive observer, an estimated speed ω of the permanent magnet synchronous motor is obtained. r ; Set the speed of the permanent magnet synchronous motor ω r and speed given value ω r * At the same time, input the second-order ADRC to obtain the q-axis voltage prediction value u q ; and, the result obtained by subtracting the d-axis current value of the permanent magnet synchronous motor from the reference value of the d-axis current is input into the PI controller to obtain the d-axis voltage prediction value u d ; The PWM control signal is obtained and input into the inverter connected to the permanent magnet synchronous motor to realize the drive control of the permanent magnet synchronous motor. The robustness and anti-interference ability of the controller are improved.
Owner:BEIJING MECHANICAL EQUIP INST

Sleep detection method and device, electronic equipment and storage medium

The invention provides a sleep detection method and device, electronic equipment and a storage medium, and belongs to the technical field of artificial intelligence, and the method comprises the steps: inputting an image frame in an obtained video stream into a target detection model, so as to output a human body frame and a face frame of a target object in each frame of image; extracting a human body image from the human body frame, inputting the human body image into a human body skeleton point detection model to obtain a skeleton point information sequence, and inputting the skeleton point information sequence into a human body resting state detection model to obtain human body state parameters; extracting a face image from the face frame, and respectively inputting the face image into a face key point detection model and a health monitoring model to obtain corresponding eye activity parameters and physiological activity parameters; and determining whether the target object is in a sleep state or not according to at least one of the human body state parameters, the eye activity parameters and the physiological activity parameters. According to the method, the problems of robustness and accuracy in user sleep detection are effectively solved.
Owner:IFLYTEK CO LTD

Router fault automatic repair method, system, device and medium

InactiveCN119652818ASolve the robustness problemImplement automatic repairTransmissionPathPingPublic network
The invention provides a router fault automatic repair method, system and device and a medium, and the scheme comprises the steps: responding to the occurrence of a fault, and obtaining log data corresponding to the current fault in real time; obtaining a problem classification result and a problem mark of the current fault by using a pre-trained fault classification model based on deep learning based on the obtained log data; based on the problem classification result and the problem mark, accessing a pre-constructed problem solving large model according to a preset public network service path to obtain a solution of the current fault; and executing the currently obtained solution to realize automatic repair of the router fault. According to the scheme, by designing the problem solving large model, the router can automatically repair most of self faults, and by periodically updating the problem solving large model, the robustness of the problem solving large model can be effectively improved.
Owner:SHANDONG YOU INTERNET OF THINGS CO LTD

Image processing method, apparatus and system

The application discloses an image processing method, device and system. The method comprises the following steps: receiving a first image; processing the first image by using a processing model to obtain a second image, wherein the resolution of the second image and the resolution of the first image satisfy a first preset condition, the processing model is used for inputting the first image into a first generation network to obtain a third image, inputting the third image into a second generation network to obtain the second image, and the resolution of the third image and the resolution of the first image satisfy a second preset condition; and displaying the second image. The application solves the technical problem that the robustness and stability of the image processing method in the related art in a real scene are poor.
Owner:ALIBABA GROUP HOLDING LTD

Flight control-oriented OTA upgrade method, system, device and program product

PendingCN122507388ASolve transmission reliabilitySolve the robustness problemControl orientedInterference resistance
The application discloses a flight control oriented OTA upgrading method, system, device, computer readable storage medium and program product, belongs to the field of computer science and technology, and aims to solve the problems of insufficient safety of an existing upgrading process, unreliable data transmission and weak batch management capability for a heterogeneous architecture. The method establishes a first connection, sends a start request frame containing a specific identification code and an inverse code, sets an upgrading flag and resets the flight control, and disconnects the connection. After resetting, a second connection is established, firmware is transmitted and written, an end request frame with the same frame structure is sent, the flight control is finally reset and starts new firmware, and the connection is disconnected. The frame structure containing the identification code and the inverse code, the instruction code and the check code enhances the communication reliability and the anti-interference capability. The stage resetting and the flag mechanism ensure that the upgrading process is controllable and recoverable, avoids firmware damage or startup failure, and improves the safety and stability of the flight control system remote upgrading.
Owner:HARBIN UNITED AIRCRAFT TECH CO LTD

Circulating system equipped with submersible type cylindrical linear induction electromagnetic pump

The invention provides a circulating system equipped with a submersible type cylindrical linear induction electromagnetic pump, which is characterized in that the electromagnetic pump is arranged in a container filled with liquid metal along the longitudinal direction, an inert gas tank is arranged for sealing, and a radiating fin structure is arranged on the outer side of an external stator of the electromagnetic pump. According to the submersible electromagnetic pump circulation system, the occupied area and weight of the liquid metal circulation loop are remarkably reduced, the service life of the liquid metal circulation loop under different working condition requirements is greatly prolonged, and the requirements of deep sea, deep space and other extreme environments for robustness and light weight of equipment can be effectively met.
Owner:JIANGSU UNIV

Noise correlation cross-modal matching loss adjustment method and device based on meta-learning

The invention discloses a noise correlation cross-modal matching loss adjustment method and device based on meta-learning. The method comprises the following steps: constructing a pairing sample set containing a multi-modal data pair; quantifying semantic correlation of each multi-modal data pair in the pairing sample set by using a cross-modal model, and selecting an initial loss function; constructing a loss adjustment network, and balancing the noise sensitive loss function and the noise insensitive loss function to obtain an adaptive loss function; optimizing the loss adjustment network by using a meta-learning algorithm; updating model parameters of the cross-modal model on the target noise data set by using the optimized loss adjustment network; and inputting the query target and each object in the matching database into the updated cross-modal model to obtain the similarity of the query object relative to each object in the matching database. The problem that the robustness and the matching accuracy of the cross-modal model are difficult to consider at the same time in the prior art is solved, and powerful support is provided for landing of the cross-modal model in an actual scene.
Owner:XI AN JIAOTONG UNIV

Method for adapting file processing of structured and unstructured data

The application relates to the technical field of file processing, and discloses a file processing method suitable for structured and unstructured data, which comprises the following steps: S1, identifying the type of an input file and classifying the input file, including structured data, semi-structured data and unstructured data; S2, pre-processing and feature extraction are performed on different types of data; S3, the correlation between different data modalities is quantified based on multi-modal mutual information optimization; S4, the interaction consistency between data modalities is verified, and the fusion weight is dynamically adjusted. Through classification and feature extraction of structured data, semi-structured data and unstructured data, efficient fusion of multi-modal data is realized by using mutual information optimization and a dynamic weighting strategy. Compared with a traditional method, the application can significantly improve the precision and robustness of data processing, and provides an efficient solution for multi-modal data fusion in a complex scene.
Owner:BEIJING HANXINSHENG TECH CO LTD

Newborn birth weight prediction method and system based on XGBoost model

The invention discloses a newborn birth weight prediction method based on an XGBoost model, which learns a large amount of newborn birth weight information, antenatal ultrasonic examination biometric data within a week from NT to birth in a newborn period, and information such as height and weight of parents through the model, and realizes accurate prediction of the newborn birth weight before antenatal. After a doctor obtains a newborn antenatal ultrasonic measurement index and parent information, the method can accurately predict the birth weight of the newborn. According to the method, the technical problem that the newborn birth weight prediction is inaccurate due to the fact that an existing regression model established based on ultrasonic measurement data is low in prediction accuracy, high in misdiagnosis rate, poor in robustness, not considering parent factors, insufficient in individuation and the like can be solved.
Owner:SHENZHEN LONGHUA DISTRICT MATERNAL & CHILD HEALTH HOSPITAL (SHENZHEN LONGHUA DISTRICT MATERNAL & CHILD HEALTH & FAMILY PLANNING SERVICE CENT SHENZHEN LONGHUA DISTRICT HEALTH EDUCATION INST)

Line feature matching method based on Fourier representation and posterior probability estimation

The invention discloses a line feature matching method based on Fourier representation and posterior probability estimation, and the method comprises the steps: detecting and extracting line features on an image sequence pair, and calculating a binary descriptor; obtaining an initial line feature matching pair; a compact Fourier series is introduced to represent a mapping relation satisfied between adjacent images; prior maximum likelihood estimation is established; iteratively solving a maximum post-test estimation parameter satisfied by the line feature matching pair; establishing a uniform distribution probability model; screening image line feature matching pairs meeting a probability threshold value; and searching residual feature matching pairs, completing iteration and realizing accurate and steady matching of image line features. According to the method, the problem of accurate and robust matching of the image line features is solved only through the basic line feature descriptors, the mapping relation between the images is represented through Fourier, the likelihood estimation model is established in combination with the geometrical relation met by line feature matching, and accurate and efficient matching of the line features is achieved based on EM algorithm iterative convergence.
Owner:NANJING UNIV OF SCI & TECH

A remote sensing image scene classification method based on uncertainty visual language alignment

PendingCN122347796ASolve the robustness problemGeneralizableFuzzy mappingComputer vision
This invention relates to a remote sensing image scene classification method based on uncertain visual-language alignment, comprising the following steps: acquiring remote sensing images as training and testing sets; using a CLIP dual encoder to map the original image and text to a shared point representation space; the uncertain encoder predicting each embedding vector along two paths; internally, the uncertain encoder performing two-level interactions on the input embedding vectors; globally modeling the image and text as two Gaussian distributions based on the mean and variance vectors, maximizing the image-text similarity that should be correctly matched in the entire batch; acquiring new remote sensing images and inputting them into the trained model, which then classifies the scenes in the remote sensing images. This application captures fuzzy mappings between modalities in the probability space, providing a more generalized representation for downstream remote sensing image scene classification tasks.
Owner:CENT SOUTH UNIV

Opportunistic array radar directional diagram reconstruction method of dynamic multi-node platform

The invention provides an opportunistic array radar directional diagram reconstruction method for a dynamic multi-node platform. The opportunistic array radar directional diagram reconstruction method comprises the steps of initializing a task target and constraint condition parameters; a fuzzy number is selected according to requirements, and a fuzzy opportunity constraint programming model for dynamic multi-node opportunity array radar directional diagram reconstruction is established; and solving the model by adopting a multi-area-array approximation matrix pencil method in combination with fuzzy simulation. According to the method, corresponding planning models can be established under different task requirements by utilizing the constructed general model and the solving method, so that the problem of system robustness caused by dynamic environment influence involved in practical application of a directional diagram reconstruction problem is effectively solved; node resources, position information and the like of the opportunity array radar can be dynamically adjusted along with environment tasks, practical application is better met, change of the opportunity array radar along with encountering is achieved, namely, the characteristic of opportunity adjustment of system parameters is achieved according to different task requirements, and reference is provided for modeling and solving of other problems related to the opportunity array radar.
Owner:NANJING UNIV OF INFORMATION SCI & TECH

A log anomaly detection method based on LDA topic features

The present invention is a log anomaly detection method based on LDA topic features. The method includes two stages: model training and anomaly detection. In the model training stage, a log parser is used to parse a system log into a log template set and a log triple set, and the log template set is used to train an LDA model to obtain a log template topic classification model; the log triple is converted into a process log template topic using an LDA-CM model, and then a training sample is constructed using a sliding window mechanism, and finally the training sample is input into an LSTM model to train and generate a log anomaly detection model. In the anomaly detection stage, the process log to be detected is converted into a corresponding template topic sequence, and then input into an LSTM-ADM model to realize anomaly detection for the process log. The present invention can more accurately learn the semantic relationship features between logs and more effectively detect abnormal behaviors of processes or systems through unstructured log records.
Owner:NANJING UNIV OF POSTS & TELECOMM

Voice endpoint detection method, storage medium and program product

PendingCN121983094ASolve the robustness problemimprove accuracySpeech analysisNonlinear filterFeature extraction
The embodiment of the invention provides a voice endpoint detection method, a storage medium and a program product. The method comprises the steps that a voice signal to be detected is acquired, framing and windowing processing is carried out on the voice signal, a framing and windowing signal is obtained, and the voice signal comprises noise; first nonlinear filtering processing and second nonlinear filtering processing are carried out on the framing windowing signal, a first enhanced signal and a second enhanced signal are obtained, and a delay parameter corresponding to the first nonlinear filtering processing is smaller than a delay parameter corresponding to the second nonlinear filtering processing; performing cepstrum domain feature extraction processing and fusion processing on the first enhanced signal and the second enhanced signal to obtain fusion features; and according to the fusion feature, the voice signal is detected to obtain a detection result, and the accuracy and stability of voice endpoint detection are improved.
Owner:RDA MICROELECTRONICS SHANGHAICO LTD

Carbon emission monitoring method and device applied to cement production line, computer equipment, readable storage medium and program product

The invention relates to a carbon emission monitoring method, device and equipment applied to a cement production line, a storage medium and a program product, and relates to the technical field of carbon emission monitoring. According to the invention, the measurement accuracy of the carbon emission can be improved. The method comprises the following steps: acquiring non-electrical characteristic data, equipment state data and equipment power consumption data of key preset type equipment in a cement production line, and integrating and sorting the non-electrical characteristic data, the equipment state data and the equipment power consumption data to obtain a target data set; according to the target data set, constructing a joint feature vector corresponding to each piece of non-electrical feature data; inputting the joint feature vector into a load monitoring model to obtain power load information of the equipment; equipment operation information of the key preset type equipment and material characteristic change information in the production process of the cement production line are obtained, the power load information, the equipment operation information and the material characteristic change information are input into the carbon emission monitoring model, and the sum of carbon emission generated by the cement production line is obtained.
Owner:南方电网能源发展研究院有限责任公司 +1

Method and apparatus for 3d scene reconstruction based on neural network and multi-view consistency

This invention discloses a method and apparatus for 3D scene reconstruction based on neural networks and multi-view consistency. The method includes: extracting image frames from video data and downsampling them; using a pre-trained monocular absolute depth estimation model to predict the downsampled images; using a scale offset recovery module to recover the inconsistent scale and offset between downsampled absolute depth maps; projecting the obtained scale offset consistent depth maps between downsampled image frames using optimizable camera parameters; selecting local and global keyframes; supervising optical flow information, geometric consistency, and regularization terms; optimizing depth and camera parameters; inputting the relevant parameters into an image-depth fusion module for processing; and outputting the 3D scene reconstruction result. This invention reduces the reliance on sensors in existing methods, improves the density and robustness of 3D scene reconstruction methods, and solves the problem of difficulty in achieving both robustness and density in existing 3D scene reconstruction methods.
Owner:ZHEJIANG UNIV

Garbage classification method based on intelligent vision, related device thereof and storage medium

The invention discloses a garbage classification method based on intelligent vision, a related device thereof and a storage medium, and the method comprises the steps: obtaining garbage image data, and carrying out the preprocessing of the garbage image data, and obtaining the preprocessed data; and classifying the preprocessed image data based on a pre-trained garbage classification model to obtain a classification result. The method comprises the following steps: firstly, acquiring garbage image data in real time through a camera mounted at a garbage throwing port, and carrying out gray processing, noise filtering, edge detection, image enhancement and normalization processing on an image; secondly, a complex network structure comprising a multi-scale feature extraction module, a feature fusion module, a recurrent neural network module, an attention module and a multi-task learning module is adopted, multi-level and multi-angle feature extraction and classification are carried out on the preprocessed image data, and efficient and accurate classification of garbage is achieved.
Owner:HUIZHOU HAIHE INTELLIGENT CONTROL TECHNOLOGY CO LTD