1096 results about "Noise detection" patented technology

The invention discloses a weak supervision semantic segmentation method based on large model auxiliary supervision, and belongs to the technical field of image segmentation in computer vision, and the method comprises the following steps: (1) initializing a visual Transform and a large model CLIP, and respectively extracting visual features; and (2) respectively generating class activation mapping based on the features of the visual Transform and the CLIP, and obtaining a classification result of the visual Transform. And (3) semantic affinity information is extracted based on multi-head self-attention of CLIP, an auxiliary mask is generated, segmentation prediction is acquired from a visual Transform, and auxiliary loss is calculated with the mask. And (4) performing noise detection on the auxiliary mask by using a Gaussian mixture model, and measuring distribution difference to filter noise pixels. And (5) fusing CLIP and visual Transform class activation mapping, refining the class activation mapping by using a post-processing technology to obtain a pseudo tag, and training a semantic segmentation model by using the pseudo tag. According to the method, the CLIP is introduced as auxiliary supervision information, a more reliable pseudo tag is generated, and the semantic segmentation performance is effectively improved.

The invention discloses an equipment data acquisition and query optimization method based on a large model enabling ERP-MES system, and relates to the technical field of integrated enterprise resource planning and monitoring management systems, and the method comprises the following steps: carrying out the dynamic adaptive data acquisition and preprocessing of equipment: employing an edge node to deploy a lightweight noise detection model, identifying an electromagnetic interference mode in sensor data collected by the equipment in real time, and dynamically compensating and correcting by combining environment temperature and humidity data; according to the method, an edge node lightweight noise detection model is combined with environmental data compensation and correction, and a BERT large model is utilized to analyze a protocol document design middleware, so that accurate acquisition and efficient processing functions of dynamic self-adaptive data of equipment are realized; preprocessed data are mapped and fused through a multi-modal large model to construct a unified data warehouse, and a mapping strategy is optimized through reinforcement learning, so that the equipment data quality is improved, and the data fusion and utilization efficiency is enhanced while heterogeneous data fusion is achieved.

A system using global and local-aware denoising framework for commonsense knowledge graph noise detection is provided. The system includes an input receiver, a triple encoder, a global noise detector, a local noise detector, and a comprehensive evaluation scorer. The triple encoder receives the CSKG information from the input receiver and extract semantic information from free-text formatted nodes in the CSKG information, so as to obtain structured triples. The global noise detector identifies high-frequency patterns through rule mining and uses a rule encoder to generalize learned rules and guide noise detection. The local noise detector utilizes the structured triples output by the triple encoder to aggregate neighboring triple information around both the head and tail nodes via graph networks. The comprehensive evaluation scorer provides a comprehensive score based on global noise detection and local noise detection, so as to indicate a noise level for the CSKG information.

The invention discloses a transformer noise detection method based on season trend decomposition and time sequence prediction, and belongs to the technical field of transformer noise anomaly detection. Comprising the following steps: S1, acquiring noise data of a transformer, and preprocessing the noise data; s2, adopting an STL decomposition model to obtain a seasonal component, a trend component and a residual component; s3, performing Transform time sequence prediction on the seasonal components and the trend components to obtain prediction results of the seasonal components and the trend components; s4, performing anomaly detection on prediction results of the seasonal components and the trend components, and outputting a detection result; and S5, carrying out weighted composition on the detection results of all the components, and outputting a diagnosis result. According to the method, the time sequence data of the noise of the transformer substation is decomposed, predicted and subjected to abnormal point detection, and reliable technical support is provided for state monitoring and fault prevention of transformer substation equipment.

A noise detecting playback system is provided to facilitate adaptive output of audio content by an audio player in response to detecting noise interferences such as external noise outside of the playback. The detection playback system assesses microphone signals and sensor data of the playback system for signs of attention requiring features to trigger playback of the audio content. Microphone signals may provide volume level of a particular external noise, which is contrasted with output volume of the audio content, other noises in the environment, and / or a lack of detected noise in the environment to determine that the particular external noise is interfering or potentially interfering with the audio content. Image sensor data of the surroundings of the listener acquired by a camera coupled to a personal listening device may also be assessed for attention features associated with an external noise.

The invention provides a high-resolution positioning method for noise source imaging. The high-resolution positioning method comprises the following steps: designing a sensor array layout and selecting sensors; a multi-channel acquisition synchronous acquisition sensor and records signals to obtain acquisition data; performing signal processing on the acquired data by using adaptive filtering, machine learning calibration, time-frequency analysis and intelligent beam forming technologies to obtain standard data; based on the standard data, performing high-resolution direction estimation by using an optimized multi-signal classification algorithm, and removing a sidelobe effect by using an optimized CLEAN algorithm to complete processing of the standard data; and based on the processed standard data, reconstructing a spatial distribution image of the noise source by using an inversion technology, and dynamically adjusting the resolution of the reconstructed spatial distribution image by using an adaptive algorithm to realize high-resolution positioning of noise source imaging. According to the invention, high-resolution positioning can be realized, the reliability and precision of noise detection are improved, and noise control and management are more effective.

The invention provides an elevator door abnormal sound detection and identification method based on a variational auto-encoder (VAE) and a Transform network. The method comprises the steps of data acquisition and preprocessing, model construction and training, abnormal sound detection and identification, method performance evaluation and the like. The method specifically comprises the following steps: acquiring sound signals in the operation process of an elevator door by installing an audio sensor, continuously recording to form time sequence data, and cleaning and normalizing the time sequence data; then, feature reconstruction is carried out through a VAE-Transform network, a reconstruction error is calculated to obtain an abnormal score, abnormal feature vectors are continuously marked, finally, abnormal segments are input into a classifier, and the abnormal sound types are judged according to the confidence score of each abnormal sound type. According to the method, unsupervised network training is introduced to complete abnormal sound detection, and then the detected abnormal sound is further classified and identified, so that the calculation amount of data can be significantly reduced, the efficiency is greatly improved, the specific fault type is rapidly identified, and accurate maintenance suggestions are provided, thereby improving the fault processing efficiency and accuracy, and improving the fault processing efficiency and accuracy. The problem that abnormal sound fault data are scarce is effectively solved, and the method is suitable for abnormal sound fault monitoring of the elevator door in a complex environment.

The invention discloses a complex underwater side-scan sonar exploration detection method and device based on a multi-dimensional attention collaborative lightweight anti-noise detection framework, and the device comprises an underwater side-scan sonar imaging device which is used for obtaining a sonar image of a detected target; the computer is connected with the underwater side-scan sonar imaging equipment and comprises a backbone feature extraction network module which is used for processing an input sonar image through a multi-scale edge refining module and outputting a three-scale feature map; the check feature fusion network module is used for realizing cross-channel and cross-space information fusion through the focusing space adaptive local attention module, performing adaptive modulation and deep information aggregation on multi-scale features through a channel frequency aggregation and attention mechanism, and outputting three enhanced feature maps; and the YOLOhead detection head module is used for generating a target detection frame according to the enhanced feature map, and outputting a detection result after non-maximum suppression processing.

The embodiment of the invention discloses a road section noise control method and system of an intelligent guardrail, relates to the technical field of noise control, and is used for solving the problem that the current road section noise control mode is difficult to provide enough noise reduction in real time. The method comprises the following steps: determining noise high-incidence time by acquiring noise detection frequency spectrums of all road sections in a current area; acquiring environmental data by utilizing guardrail detection equipment, and establishing a digital twin model of noise of each road section in combination with a noise detection frequency spectrum; and determining a noise source and characteristics thereof based on the model so as to determine a to-be-adjusted physical structure of the guardrail, and determining adjustment data and a noise reduction value thereof through simulation adjustment. And in combination with the noise reduction value, the noise source and the characteristics, determining the operation parameters of the sound wave offset device to perform active noise reduction, adjusting the guardrail physical structure according to the adjustment data to realize physical noise reduction, and realizing noise control of each road section in the current area.

The invention relates to the technical field of image processing, in particular to a denoising enhancement processing method and system for sensor image data, and the method comprises the steps: carrying out the preprocessing of original image data, and carrying out the noise type detection and classification of the preprocessed image data through a deep learning model; constructing a high-fidelity denoising and enhancing strategy which comprises noise domain decomposition, dynamic region weighted denoising and multi-scale enhancement and texture compensation; constructing a self-adaptive multi-domain collaborative denoising and enhancement strategy, and combining noise domain self-adaptive projection based on a variational auto-encoder, multi-scale detail enhancement under a pyramid structure and cross-domain texture migration and fusion; and through a mapping relation between a noise detection result and an enhancement strategy, selecting an optimal method to carry out de-noising and enhancement on the image. According to the method, high-fidelity denoising, detail recovery and quality improvement of the image in a complex scene can be realized, and a reliable solution is provided for sensor image data processing.

The invention provides a marine radar image denoising method and device based on a multistage filtering algorithm, and the method comprises the steps: carrying out the preprocessing of a to-be-processed marine radar image, and obtaining a first image; performing edge gradient detection on the first image based on the gradient direction corresponding to each pixel point in the first image and the target identified in the first image, and performing noise detection on the first image based on the gray value corresponding to each pixel point in the first image after the edge gradient detection is passed; after the noise detection is passed, processing the first image by adopting bilateral filtering to obtain a second image; and image enhancement is carried out after image sharpening is carried out on the second image, and a denoised marine radar image is obtained. According to the method, multiple filtering technologies such as Gaussian filtering, median filtering and bilateral filtering are combined, multiple kinds of noise in the marine radar image are effectively removed, especially image edge and target information can be kept in a complex environment, and the target detection and navigation precision is remarkably improved.

The embodiment of the invention provides a vehicle abnormal sound detection method and device, a vehicle and a storage medium, and the method comprises the steps: obtaining an audio signal of the vehicle, and converting the audio signal into initial image data; judging whether a preset graph exists in the initial image data in a preset abnormal sound interval or not; when it is judged that the initial image data has the preset graph in the preset abnormal sound interval, a preset abnormal sound similarity detection model is adopted to perform graph similarity detection on the preset graph in the initial image data, and similarity information of the audio signals is determined, the abnormal sound similarity detection model comprises a first detection sub-model for carrying out similarity detection by adopting graphic comparison and / or a second detection sub-model for carrying out similarity detection by adopting RGB (Red, Green, Blue) comparison; and carrying out abnormal sound detection on the vehicle based on the similarity information. According to the embodiment of the invention, the abnormal sound similarity detection model is adopted to carry out abnormal sound recognition on the image converted from the audio signal, and compared with manual detection, the method is more objective and accurate.

The invention provides a magnetotelluric data denoising method and device based on a depth fusion model, is applied to a magnetotelluric data denoising system, and relates to the technical field of magnetotelluric data processing. A deep fusion model with a'noise feature recognition-signal selective reconstruction 'framework is established, input signals are pre-classified through a front noise detection module, the problem of false attenuation of effective signals in traditional end-to-end denoising is effectively avoided, a denoising auto-encoder containing a residual structure is utilized, and the denoising accuracy is improved. The deep feature extraction capability of the residual structure and the signal reconstruction advantage of the denoising auto-encoder are combined, the fidelity and integrity of the reconstructed signal are remarkably improved, the denoising process does not depend on manual intervention, and the efficiency and accuracy of data processing are greatly improved.

The present invention relates to the field of fault detection technology, and specifically discloses a stepper motor fault detection method and system based on noise, the method comprising obtaining the working state of the stepper motor, obtaining the audio information of the stepper motor under different working states based on the noise detection point, creating a running audio model as an audio feature; counting the working state and the audio feature, creating a mapping relationship from the audio feature to the working state; in the actual application stage, obtaining the working period, obtaining the actual audio information based on the noise detection point, determining the audio feature according to the actual audio information and the working period, and obtaining the working parameters based on the mapping relationship as the prediction result. The present invention installs an audio detector at the noise detection point, and the audio detector obtains the working audio of the stepper motor, and then compares it with the preset conditions to determine in real time whether there is a fault; in this process, the preset conditions can be adjusted by the user, thereby improving the accuracy.

A system may include a resistive-inductive-capacitive sensor, a driver configured to drive the resistive-inductive-capacitive sensor at a driving frequency, a measurement circuit communicatively coupled to the resistive-inductive-capacitive sensor and configured to measure phase information and amplitude associated with the resistive-inductive-capacitive sensor, and a noise detection circuit communicatively coupled to the measurement circuit and configured to determine a presence of external interference in the system based on at least one of the phase information and the amplitude information.

The invention discloses a heart sound signal noise detection method based on a multi-scale convolutional network. The method comprises the following steps: collecting heart sound signal information; cutting into segments with fixed length as network input; then, the cut heart sound fragments are preprocessed; evaluating the quality of the heart sound fragment based on the spectral density ratio, and correcting the label of the low-quality fragment; generating a spectrogram; effective features in spectrograms of different scales are extracted through a convolutional neural network of a hierarchical multi-scale structure, and accurate recognition and classification of heart noise are achieved; training the multi-scale convolutional neural network; and according to an output result of the multi-scale convolutional neural network, outputting a diagnosis result of the cardiac murmur. According to the method, the hierarchical multi-scale convolutional neural network is introduced, so that multi-scale features are effectively fused, and the accuracy and robustness of heart sound signal noise detection and clinical result classification are improved. And the quality of the heart sound fragments is evaluated based on the spectral density ratio, and the labels of the low-quality fragments are corrected, so that the classification performance is further improved.

The invention discloses a motor abnormal sound detection method and system based on contact acquisition and small sample learning, and the method comprises the steps: directly coupling a motor housing through a contact vibration sensor, collecting an original vibration signal, and generating an anti-interference vibration signal; inputting the anti-interference vibration signal into a nonlinear resonance enhancement module to generate an enhanced sound signal; performing wavelet packet decomposition on the enhanced sound signal, extracting a multi-scale frequency band energy entropy, and generating a motor abnormal sound feature matrix by combining singular value decomposition dimension reduction; on the basis of a dynamic weight distribution element learning algorithm, a small number of normal samples and abnormal samples are utilized to construct an abnormal sound classification model; and inputting the motor abnormal sound characteristic matrix into an abnormal sound classification model, detecting transient abnormality through a sliding window time sequence matching algorithm, and outputting an abnormal sound judgment result. According to the embodiment of the invention, high-precision and low-false-alarm motor abnormal sound detection under complex working conditions can be realized.

The invention relates to a helical gear detection method based on waviness information and a gear detection system. The helical gear detection method based on waviness information comprises the following steps: extracting full-tooth waviness information of a helical gear; processing the full-tooth waviness information to obtain a spectrogram; calculating an angle alpha of waviness on any single tooth of the helical gear; and according to the angle alpha of the waviness on the single tooth and the spectrogram, judging whether the gear has noise requirements or not. According to the method, through full-tooth order analysis and single-tooth topology three-dimensional angle analysis of the involute helical gear, important factors influencing gear meshing noise are calculated and extracted, tooth surface waviness information can be effectively detected, ghost frequency can be effectively identified and found, and a ripple line angle can be effectively analyzed; the method greatly improves the fitting quality of data, thereby improving the detection precision of the gear, enabling the gear to meet the noise detection requirements, and providing a brand new detection method for high-speed and low-noise gear production.

The invention relates to an intelligent AI information processing method and system and an intelligent terminal, and relates to the technical field of information processing, and the method comprises the steps: obtaining sound detection information in a preset detection range and triggering information of a user; inputting the sound detection information into a preset noise model to match noise detection information; updating the sound detection information through a preset sound processing method according to the noise detection information; inputting the updated sound detection information into a preset audio processing system, and determining an output text according to the updated sound detection information when the audio processing system outputs the text; obtaining a selected domain image based on the trigger information; determining a selected text field according to the selected field image; and performing matching combination according to the selected text field and the output text to obtain final uploading information, and uploading the final uploading information to a preset storage database for storage. The information processing method and device have the effect of improving the information processing accuracy.

The present disclosure relates to a video noise detection method and apparatus, and a device and a medium. The video noise detection method includes: extracting a first video frame and a second video frame from a target video, wherein the first video frame and the second video frame are adjacent video frames; performing differential processing on the first video frame and the second video frame, so as to obtain an inter-frame differential image between the first video frame and the second video frame; performing flat area detection on the first video frame and the second video frame, to obtain an intersection of flat areas in the first video frame and the second video frame; and calculating a time-domain noise value corresponding to the first video frame by using pixel information of the intersection of the flat areas in the inter-frame differential image.

A first clamping rod and a second clamping rod are arranged on the two sides of a collecting assembly respectively, one end of a supporting assembly is provided with a supporting frame used for supporting a steering gear, and the other end of the supporting assembly is movably connected with a rack so that the supporting assembly can drive the steering gear to abut against the collecting assembly. The acquisition assembly is provided with a vibration sensor, and the force application assembly is provided with a bidirectional rotating motor used for driving an input shaft of the steering gear. The supporting assembly drives the steering gear to move towards the collecting assembly, after the steering gear abuts against the vibration sensor, the first driving piece drives the first clamping rod to abut against one end of a steering gear rotating shaft, the second driving piece drives the second clamping rod to abut against the other end of the steering gear rotating shaft, and the force applying assembly simulates a steering wheel and applies force to an input shaft of the steering gear. And the vibration sensor acquires vibration information of the steering gear, so that whether the steering gear is qualified or not can be tested, and the problem that abnormal sound is brought to the whole vehicle to affect the progress of a project is avoided.

The invention provides a signal recovery method integrating noise detection and prior estimation, which comprises the following steps of: acquiring pre-estimation of an original signal by utilizing a zero interference equalization module based on an observation signal and a channel matrix; carrying out iterative optimization on the estimated value of the signal by utilizing a self-adaptive signal processing module; based on pre-estimation of the original signal and information extracted from noise observation of the original signal generated by the adaptive signal processing module, pre-estimation of signal distribution is obtained through a semi-parameter Bayesian module, and estimation of the original signal distribution is updated in each iteration process of the adaptive signal processing module; and the estimated value of the noise variance is updated through the iterative noise estimation module, and is updated in each iteration process of the adaptive signal processing module. According to the method, the difficulty of inaccurate signal recovery caused by noise fluctuation brought by environmental factors is overcome, and meanwhile, the signal can be accurately recovered when the prior information of the original signal is lacked.

The embodiment of the invention relates to the field of time series anomaly detection, in particular to a time series data anomaly detection method and device of a dynamic condition diffusion model based on gradient noise. A specific embodiment of the method comprises the following steps: determining a GNDC-DM model; performing trend noise detection processing on the acquired initial industrial equipment data based on a trend noise diffusion model included in the GNDC-DM model to generate trend noise reduction data; performing seasonal noise detection processing on the obtained initial industrial equipment data based on a seasonal noise diffusion model included in the GNDC-DM model to generate seasonal noise reduction data; and performing hybrid anomaly detection processing on the acquired initial industrial equipment data, trend noise reduction data and seasonal noise reduction data based on a hybrid anomaly diffusion model included in the GNDC-DM model to generate equipment noise reduction data. According to the embodiment, the detection based on the reconstruction score can be more effective.

The invention relates to the field of video processing, and aims to provide a video noise AI (Artificial Intelligence) generation processing method which can be used for improving the denoising performance, reserving details, reducing the processing cost and improving the processing efficiency. According to the technical scheme, the method comprises the steps that S1, data preprocessing is conducted, a video file is decoded into image frames, the format of the image frames is converted, a data set is obtained after standardization processing and data enhancement are conducted, and the data set is divided into a training set, a verification set and a test set; s2, establishing a model: establishing a noise detection model, a generative adversarial network and a detail enhancement model; s3, model fusion: combining the noise detection model, the generative adversarial network, the detail enhancement model and the self-attention module into a joint model; training, verifying and testing the joint model; and S4, video noisy point generation processing is carried out, the collected video is subjected to data preprocessing, the joint model is input to obtain a noiseless image frame sequence, and a noiseless video is output after merging.

The invention provides a high-voltage level shift circuit with an adaptive dead zone. The level shift circuit comprises a level shift module, a dv / dt noise detection module, a bank signal control module and an output module. According to the invention, the problem that dv / dt noise affects output through a signal chain is prevented by using a bank turn-off technology; the Blank time of dv / dt noise with different intensities is adaptively adjusted through the dv / dt noise detection module, so that the limitation on the minimum conduction time of an input signal in the application is solved; a low transmission delay, high dv / dt noise resistance and low minimum input signal conduction time are obtained, and the application requirements of a DCDC system and a driving circuit system are met.

A squeezed light generator may include a triply-coupled optical resonator having a triplet resonance with a split-resonant frequency spacing Ω, where the triplet resonance receives light at frequencies associated with at least one split-resonant frequency in the triplet resonance and generates squeezed light with at least another split-resonant frequency in the triplet resonance.

The invention discloses a multispectral and hyperspectral laser radar weak multi-echo signal extraction method. According to the method, a B-spline model is adopted to establish models for transmitted pulses and received echoes. Compared with a traditional Gaussian model and a B-spline model, the method has local control characteristics, is more suitable for modeling and analysis of complex waveforms, improves the adaptability to diversified signals, remarkably reduces the probability that noise is misjudged as echo signals by conducting preliminary screening and positioning on received echoes of all channels, and improves the accuracy of noise detection. Therefore, error accumulation is effectively reduced. The process not only improves the accuracy of signal identification, but also greatly reduces the complexity of subsequent data processing, reduces the calculation burden of the system, and lays a solid foundation for refined waveform analysis.

The invention discloses a method and a device for evaluating the learning concentration degree of a user, electronic equipment and a storage medium, relates to the technical field of image analysis and visual computing, realizes comprehensive and detailed concentration degree evaluation of the user, can adapt to diversified learning scenes while improving the accuracy and intelligence of learning concentration degree evaluation, and improves the learning experience of the user. And the risk of misjudgment is reduced. The method comprises the following steps: acquiring a video stream acquired when a user learns, and extracting a plurality of to-be-detected image frames from the video stream; based on the plurality of to-be-detected image frames, detecting eyes and behavior actions of the user during learning to obtain an action detection result; based on the plurality of to-be-detected image frames, detecting articles placed in a learning environment where the user is located during learning, and based on the video stream, carrying out environment noise detection to obtain an environment detection result; and evaluating the concentration degree of the user during learning by using the behavior action detection result and the learning environment detection result to obtain a concentration degree evaluation result.

The invention discloses a noise measuring device and a noise filtering method thereof, and belongs to the technical field of acoustic signal processing. The noise measuring device comprises an annular microphone array, directional microphones and a processor; the annular microphone array and the directional microphone are electrically connected with the processor; the annular microphone array comprises a plurality of omnidirectional microphones arranged in an annular array, one part of the omnidirectional microphones correspond to the non-interested area, and the other part of the omnidirectional microphones correspond to the interested area; the directional microphone is arranged in the center of the annular microphone array, the directional microphone has a target sound receiving direction, and the target sound receiving direction corresponds to the region of interest; by adopting a plurality of omnidirectional microphones which are annularly arranged and a directional microphone, the device can comprehensively utilize the information of a plurality of microphones, and a beam forming algorithm is combined, so that a noise source in a factory boundary is accurately positioned. The method has flexibility and adaptability, and has a wide application prospect in the aspects of noise detection, voice recognition and audio monitoring.

The invention provides a method and device for carrying out noise detection on a train, which can be applied to the field of railway and urban rail transit, and the method comprises the following steps: determining the initial noise reverberation duration and the sound source position of noise; iteratively executing the following operations to obtain the dynamic reverberation duration: determining the nth reverberation duration according to the sum of the (n-1) th noise reverberation duration and the preset duration; processing the nth noise reverberation duration and the reference reverberation duration according to tunnel shape parameters for the pipe tunnel to obtain an nth room coefficient; processing the nth room coefficient by using a target noise detection model corresponding to the sound source position to obtain an nth sound pressure level increment value of the noise; under the condition that a difference value between the Nth sound pressure level increment value and a preset standard sound pressure level increment value meets a preset convergence condition, determining the Nth noise reverberation duration as a dynamic reverberation duration; and processing the reference reverberation duration and the dynamic reverberation duration according to a reverberation noise prediction algorithm to obtain a noise value of the noise.