240results about How to "Improve denoising effect" patented technology

The invention provides a distribution line fault waveform denoising analysis method based on an improved wavelet threshold function and computer equipment, and relates to the technical field of power system fault detection and signal processing. The method comprises the following steps: S1, obtaining a fault waveform of a distribution line through actual measurement or analogue simulation, and converting the fault waveform into a fault signal; s2, through denoising effect comparison, selecting an applicable wavelet basis function and a decomposition layer number, and carrying out wavelet transformation to decompose a fault signal; s3, carrying out wavelet decomposition coefficient threshold processing on the fault signal by adopting a self-adaptive adjustment mechanism of a threshold and an improved threshold function; and S4, performing discrete wavelet inverse transformation on the wavelet decomposition coefficient subjected to threshold processing to reconstruct a signal so as to obtain a denoised fault signal. According to the method, through the improved threshold function and the multi-scale analysis technology, the fault waveform denoising effect can be remarkably improved, the accuracy and the real-time performance of fault detection are ensured, and therefore the reliability and the safety of a distribution line are improved.

The invention discloses an image denoising method based on double-branch feature fusion, and the method specifically comprises the steps: 1, carrying out the collection of an image, and carrying out the preprocessing of the collected image; 2, a denoising network with double-branch feature fusion is built, the network comprises a generation network and a discrimination network, and authenticity discrimination is carried out on a denoised image generated by the generation network and a real image through the discrimination network; step 3, adversarial training is carried out on the generative network and the discriminant network to obtain a trained dual-branch feature fusion denoising network; and step 4, performing denoising processing on the image by using the trained generative network. The method is remarkably superior to a traditional denoising method in denoising quality, model generalization ability and operation time, different types of noise can be effectively removed, and more image details and structure information are reserved.

The invention relates to the technical field of spectrum analysis, in particular to an intelligent protein quality detection method and system based on spectrum analysis, and the method comprises the following steps: collecting infrared spectrum data of a preset number of protein samples; determining a local fluctuation coefficient of the transmittance of each wavelength; determining a fluctuation coefficient of the infrared spectrum data of each protein sample; screening the infrared spectrum data of all the protein samples based on the fluctuation coefficient to obtain fluctuation data of each suspected impurity; equally dividing each piece of suspected impurity fluctuation data into each fragment; determining an abnormal index of the suspected impurity fluctuation data in each segment; correcting a spatial domain weight and a value domain weight of each filtering window in a bilateral filtering algorithm based on the abnormal index, and denoising infrared spectrum data of the protein sample; and determining a difference coefficient of the infrared spectrum data of each protein sample, and detecting the protein quality. Therefore, the accuracy of protein quality detection is improved.

The invention discloses a generative adversarial network sonar image denoising method and device fused with comparative learning, and the method comprises the steps: constructing a generative network, inputting a noisy image, and carrying out the image denoising processing through a fusion attention mechanism module and a comparative learning module; constructing a discrimination network, inputting a noise-free image without noise and a de-noised image generated by the generation network, and carrying out image discrimination; s3, training an adversarial network composed of the generative network and the discriminant network, and storing final parameters after training is finished; and S4, inputting the noise image into the trained generative network to obtain a denoised image. The fusion attention mechanism module and the contrast learning module can improve the ability of the generative network to understand and process the input image, enhance the ability of the network to capture and retain the image details, and accurately restore the main body features and detail information of the noise image.

The invention relates to a power transmission line multi-defect detection method and system based on improved YOLOv8s, and the method comprises the steps: collecting insulator defects and bird nest images on a power transmission line, and obtaining an initial data set; adopting an SAM large model to preprocess the initial data set to obtain a target data set; dividing the target data set into a training set, a verification set and a test set; taking the YOLOv8s network model as a reference model to construct an improved YOLOv8s network model as a power transmission line multi-defect detection network model; inputting the training set into a model for model training; inputting a verification set into the trained power transmission line multi-defect detection network model for model evaluation; and inputting the test set into the model to obtain a power transmission line defect detection result, and evaluating the model. According to the improved YOLOv8s network model, the problems of missing detection and false detection of defects can be avoided, the detection precision of the model is improved, and the light weight of the model is considered at the same time.

The invention relates to the technical field of geophysics, in particular to a deep reflection seismic data random noise denoising method based on multi-scale feature fusion, and the method is executed by a processor, and comprises the steps: generating a training set based on a deep reflection random medium forward modeling model and seismic random noise of different frequency bands; building a training network, and performing training on the training set to obtain trained model parameters; and carrying out transfer learning on the trained model parameters, and carrying out actual deep reflection seismic data random noise reduction. According to the method and the device, the problems of low denoising quality of random noise of deep reflection seismic data, too complex implementation process and low efficiency in the prior art are solved.

The invention discloses a magnetotelluric signal denoising method and system based on self-supervised comparative learning, and the method comprises the steps: constructing a noisy signal sample, obtaining a noise contour sample and / or a pure signal sample of the noisy signal sample, and training a pre-training denoising model based on a comparative learning framework through the noisy signal sample; a small part of data with labels is taken and sent to the trained pre-training model, values of all layers in the learning framework are fixed and compared, only a regression layer is finely adjusted, and a trained signal-to-noise mapping model is obtained; and finally, inputting the to-be-denoised signal into the signal-to-noise mapping model to obtain a denoised magnetotelluric signal. According to the method, a self-supervised contrast learning framework is introduced into a magnetotelluric signal noise reduction technology, data augmentation is realized through a time domain augmentation mode and a frequency domain augmentation mode, strong interference in the magnetotelluric signals is effectively removed, slow change information of a low-frequency band of the signals is reserved, and a good noise reduction effect is achieved.

The invention relates to the technical field of image processing, and discloses an image denoising method and device, computer equipment and a storage medium, and the method comprises the steps: carrying out the preliminary denoising of a noise-containing image based on a diffusion coefficient function model, and obtaining a preliminary denoised image; and performing second denoising processing on the preliminarily denoised image by using a denoising convolutional neural network, and outputting a denoised image. And according to the feature information of the image, adjusting the threshold parameter and the number of iterations of the diffusion coefficient function so as to adapt to the noise characteristics of different images, and finally outputting an optimized target de-noised image. According to the image denoising method, noise in the image can be effectively removed, and detail and edge information of the image can be accurately reserved. The optimal balance between noise removal and detail protection is realized by adaptively adjusting the threshold parameter and the number of iterations, and the denoising effect is remarkably improved. And the finally generated image is higher in quality, richer in details and clearer in edge.

The invention relates to the technical field of part profile detection, in particular to an automobile part profile detection method, device and system. The method comprises the following steps: firstly, according to corresponding curvature values and reflection score values of all data points in a block region, obtaining the reflection interference degree of the block region; according to the curvature values of the block area and other block areas, obtaining defect expression measurement of the block area; adjusting the Gaussian filtering standard deviation by using the reflection interference degree and the defect expression measurement of the block regions, and obtaining the adjusted Gaussian filtering standard deviation of each block region; performing three-dimensional Gaussian filtering on the point cloud initial data to obtain point cloud filtering data of the automobile part; and according to the point cloud filtering data, carrying out profile quality detection on the automobile part. According to the method, the standard deviation of Gaussian filtering is adjusted according to the noise degree and the defect expression degree of different areas in the point cloud initial data, so that the denoising effect is improved, and the accuracy of automobile part profile quality detection is improved.

The invention relates to the technical field of digital signal generation, in particular to a broadband sine wave digital signal generation method and system, and the method comprises the steps: obtaining an initial broadband sine wave digital signal; adopting a signal decomposition algorithm to obtain each modal component and residual component of the initial broadband sine wave digital signal; determining an interference influence factor of each modal component; determining a period difference degree and a superposition influence coefficient of each modal component; determining a relative difference factor and a relative influence coefficient of each modal component; determining a step parameter of each modal component; and in combination with an adaptive filtering algorithm, reconstructing the filtered modal component and residual component to obtain a de-noised broadband sine wave digital signal. According to the invention, the step length parameter of the adaptive filtering algorithm can be adjusted, the denoising effect of the adaptive filtering algorithm is improved, and then the high-quality broadband sine wave digital signal is generated.

The invention discloses a QUIC network abnormal behavior identification method and system and a storage medium. The method comprises the following steps: firstly, acquiring multiple segments of network traffic signals based on a sliding window principle; secondly, decomposing and reconstructing the multiple segments of network traffic signals by adopting a wavelet transform method to obtain network traffic reconstruction signals; when the Hurst index is smaller than a first abnormal threshold value, predicting and estimating the network traffic reconstruction signal based on Kalman filtering to obtain a prior prediction value and a posterior optimal estimation value corresponding to each segment of network traffic signal; finally, judging whether the error between the priori predicted value and the posterior optimal estimated value is greater than or equal to a second abnormal threshold value or not; and if the first abnormal threshold value is greater than or equal to the second abnormal threshold value, indicating that the data jitter abnormal behavior exists in each section of network flow signal, and giving an alarm or visually displaying. According to the method, the wavelet algorithm is combined with the Kalman filtering, so that the accuracy, universality and efficiency of network abnormal behavior recognition are improved, and a better denoising effect is achieved.

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 relates to the technical field of fault diagnosis, and discloses a fault diagnosis method and system applied to an automobile air conditioner compressor. The method comprises the following steps: firstly, introducing auxiliary noise to decompose a vibration signal, calculating a decomposition result by using a sample entropy method to obtain a sample entropy, and identifying and removing noise; secondly, variational mode decomposition is carried out on the vibration signals after noise elimination, an improved crayfish optimization algorithm is used for optimizing a decomposition scale and a penalty factor in the variational mode decomposition process, and a final IMF component signal set is obtained; mapping the final IMF component signal set to polar coordinates through a symmetric polar coordinate transformation method to form a fault diagnosis image set; and finally, training to obtain a convolutional neural network fault diagnosis model, identifying a fault diagnosis image set, and outputting a fault diagnosis result of the automobile air conditioner compressor. The vibration signal of the automobile air conditioner compressor is processed and analyzed, the purpose of fault diagnosis is achieved, and the method is objective and accurate.

The invention discloses a lightning low-frequency three-dimensional positioning method for a complex environment, and the method comprises the following steps: constructing a signal processing frame which comprises a graph learning module, a signal path optimizer, a signal decision agent module and a parameter updating module to carry out the preprocessing of a lightning signal; performing pulse matching on the preprocessed lightning signal to obtain initial positioning; and fitting the preliminary positioning through an improved particle swarm optimization (PSO) algorithm to obtain a final positioning result. The method focuses on three key technologies of signal processing, pulse matching and initial result optimization in lightning low-frequency positioning, and through deep research and innovation of the three key links, the precision and reliability of lightning low-frequency positioning in a complex environment are improved.

The invention discloses a physical intelligent magnetic resonance diffusion weighted image denoising and quantifying method, and relates to diffusion magnetic resonance weighted imaging. Comprising the following steps: 1) acquiring diffusion weighted magnetic resonance data acquired by using a multi-diffusion-direction planar echo imaging sequence; 2) designing a physical intelligent magnetic resonance diffusion weighted image denoising and quantification deep learning network model and a loss function; 3) using the training data set obtained in the step 1) to train a physical intelligent magnetic resonance diffusion weighted image denoising and quantification deep learning network model; and 4) inputting diffusion weighted data to be denoised into the trained network to generate a final denoising and quantification result. Through combined application of physical regularization loss and low-b-value data, the problem of low signal-to-noise ratio in magnetic resonance diffusion weighted imaging is effectively solved, image denoising is realized, and an accurate quantitative result is further provided. And the image quality and the ADC quantitative accuracy can be maintained and even improved while the acquisition times are reduced.

The embodiment of the invention discloses a data denoising method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining an annotation data set, wherein the annotation data set comprises a plurality of pieces of sample data and original labels; extracting sample features corresponding to the sample data, clustering the sample features to obtain at least one first cluster, and determining target features based on the sample features failed in clustering; inputting the sample data into a first large language model for content generation to obtain a first reference tag of a first cluster, splicing the sample data and the first reference tag, then performing feature extraction, and updating the first cluster based on a feature extraction result; clustering the target feature and the updated first cluster to obtain a target cluster; determining a second reference label of the target cluster, and denoising the labeled data set based on the difference between the original label and the second reference label; according to the embodiment of the invention, the clustering accuracy can be improved, so that the denoising effect is effectively improved.

The invention relates to the technical field of signal processing, and discloses a harbor equipment vibration signal processing method, which comprises the following steps: constructing a comprehensive function, and calculating a harbor equipment vibration signal based on the characteristics of a frequency scale, a second wavelet basis function and a threshold value; the concentration ratio of energy distribution, the calculation complexity of second wavelet basis functions under the frequency scale, the orthogonal value between the second wavelet basis functions under the frequency scale, the sparse value of second wavelet coefficients under the frequency scale and the signal-to-noise ratio of time domain signals reconstructed after threshold processing are comprehensively considered. The selection of the optimized frequency scale, the optimized second wavelet basis function and the optimized threshold is realized. The innovative method gets rid of the limitation of a single or optimized algorithm, fixed parameter setting or predefined rules in the prior art, and significantly improves the precision and robustness of low-frequency vibration signal processing.

The invention provides a transverse wave leakage noise suppression method and device based on multi-component pattern recognition, and belongs to the field of seismic data processing, and the method comprises the steps: obtaining a P component, an X component, a Y component and a Z component of original seismic data; respectively rotating the X and Y components to obtain an R component and a T component; performing pattern recognition on the P component and the Z component to obtain a noisep noise model; performing pattern recognition on the P component and the R component to obtain a noisser noise model; performing pattern recognition on the P component and the T component to obtain a noisset noise model; and sequentially subtracting the three models from the Z component to obtain the Z component after transverse wave leakage noise suppression. Through the method provided by the invention, other noises in the Z component can be better extracted, and a better denoising effect is achieved.

The invention provides an adjustable and controllable image denoising method and system based on a cascade neural network, and particularly relates to the technical field of image processing, and the scheme comprises the steps: collecting a plurality of to-be-processed images; and based on preset noise reduction intensity, performing noise reduction processing on each to-be-processed image by using the trained cascade noise reduction network to obtain a noise reduction image corresponding to each to-be-processed image. According to the scheme, the noise reduction intensity of the to-be-processed image can be dynamically adjusted so as to adapt to the images with different noise levels and characteristics, and the personalized requirement of the cascaded noise reduction network for noise reduction processing of the image can be improved.

The invention discloses a seismic data blind denoising method based on adaptive distribution estimation constraint. The method comprises the following steps: S1, constructing an initial noise distribution prediction network model and an initial generation diffusion denoising network model; s2, collecting seismic data with noise, and obtaining an ideal noise distribution prediction network model; s3, inputting a test data set in the noisy seismic data into the ideal noise distribution prediction network model for processing to obtain noise distribution estimation data; s4, processing the noise distribution estimation data through an adaptive noise distribution matching algorithm to obtain a matching time step; s5, inputting the test data set, the noise distribution estimation data and the matching time step into the initial generation diffusion denoising network model to obtain an ideal generation diffusion denoising network model; and S6, inputting the test data set into the ideal generative diffusion denoising network model to obtain the seismic data with the high signal-to-noise ratio.

The invention provides a radar signal denoising method based on an adaptive threshold and an improved threshold function, and relates to the field of signal denoising. The method comprises the following steps: S1, carrying out multi-scale wavelet transformation on an original radar signal to obtain a wavelet coefficient under multiple scales, and solving a logarithmic energy entropy of a wavelet coefficient matrix; s2, segmenting the original radar signal s (n) into a normal radar signal and a noisy radar signal; s3, processing a normal radar signal; s4, processing the noise-containing radar signal; s5, reconstructing the wavelet coefficients processed in the steps S1 to S4; and S6, obtaining a de-noised radar signal d (n). When wavelet denoising is carried out, an original radar signal is divided into a normal radar signal and a noisy radar signal, and denoising processing is carried out on the noisy radar signal, so that resource waste caused by blind denoising processing on all signals is avoided, and denoising is more targeted; based on the adaptive wavelet threshold and the improved semi-soft threshold function, the signal-to-noise ratio of the radar signal is improved, and the denoising effect is improved.

The invention discloses an ultraviolet spectrum denoising method based on an improved SWO optimization VMD-KPCA algorithm, and the method comprises the steps: firstly enabling a spectrum signal to serve as an input, importing the spectrum signal into an improved SWO algorithm (ISWO), and outputting an optimal parameter combination of VMD according to a fitness function through the algorithm; setting variational mode decomposition optimal parameters to perform signal decomposition to obtain intrinsic mode functions (IMF), and performing feature extraction on each IMF by using KPCA (Kernel Principal Component Analysis); and distinguishing the IMF by using a variance contribution rate (VCR) to screen out an effective signal, and finally reconstructing an effective signal component to finally obtain a denoised ultraviolet-visible absorption spectrum signal. The method has the advantages that through the improved SWO algorithm, optimal KPCA and VMD parameters can be selected in a self-adaptive mode, and the denoising effect is improved; the advantages of KPCA and VMD are combined, signals and noise can be effectively separated, the quality of spectral data is improved, the optimized algorithm has the good convergence speed, the calculation complexity is reduced, and the method is suitable for processing large-scale spectral data.

The invention relates to an image denoising method, a model training method and device, electronic equipment and a computer program product, relates to the technical field of image processing, and can be applied to a scene for denoising an image containing noise. The method comprises the steps of obtaining a to-be-processed image, and performing decomposition transformation processing on the to-be-processed image to obtain a plurality of frequency sub-graphs; performing feature extraction processing based on the plurality of frequency sub-images to obtain intermediate image features; performing up-sampling processing on the intermediate image features to obtain recovered image features, the up-sampling processing being an inverse process of decomposition transformation processing; and generating a de-noised image according to the restored image features and the to-be-processed image. According to the method, video memory occupation in the image processing process is effectively reduced, the calculation efficiency of the model is improved, and meanwhile, detail and texture information of the image is reserved; moreover, the image denoising effect is optimized, and the information loss in the denoising process is reduced.

The invention discloses an intelligent control method and system for a mobile lighting device, and relates to the technical field of mobile lighting. An intelligent control system of a mobile lighting device comprises a lighting target preprocessing module, a lighting target identification module, a lighting scheme output module and a lighting device intelligent control module. According to the invention, through combining a plurality of image processing layers of the trained residual learning network and the deep convolutional network, image enhancement operation is carried out in the illumination target preprocessing model, which is helpful for identifying a dynamic target object under various environmental conditions; by using the illumination target identification model, motion prediction can be carried out on an identification target, accurate positioning information is output, and necessary target positioning information is provided for intelligent illumination control; the optimal intelligent control scheme of the lighting device is calculated and output by adopting an improved group optimization algorithm, so that intelligent control of the lighting device is realized.

The invention discloses a highway bridge safety state assessment method and system, and relates to the technical field of highway bridge state assessment, and the method comprises the steps: collecting the time sequence vibration data of a plurality of key nodes of a highway bridge and the overall surface image data; inputting each time sequence vibration data into the improved generative adversarial network, generating denoised time sequence vibration data, and performing initial state evaluation according to the denoised data to obtain an initial evaluation value; wherein a generator in the network adopts an encoder-decoder structure, and a channel attention mechanism is additionally arranged between feature encoding and feature decoding; identifying all cracks in the surface image data, performing region division on the cracks, calculating according to the crack of each region to obtain a secondary evaluation value, and using the secondary evaluation value to assist in adjusting the initial evaluation value; and for each adjusted evaluation value, integration is carried out based on the support degree between the data, and an overall evaluation result is obtained. The accuracy and efficiency of highway bridge safety state evaluation can be effectively improved.

The invention discloses an unsupervised method for removing ring artifacts with enhanced interlayer complementarity. According to the method, a polynomial coefficient compensation equation is defined to optimize the non-uniform response of the detector based on the principle that ring artifacts are generated based on the non-uniform response of the detector. In consideration of randomness of detector response and independence in the vertical direction, a 2.5 D data input format is designed, remarkable interlayer information between adjacent slices is utilized, interlayer information constraint is realized by applying a TV minimization algorithm, and preliminary ring removal and denoising are carried out. In the polar coordinate system, the ring artifact is represented as a vertical stripe, the gradient of the stripe in the horizontal direction is large, a horizontal differential operator is designed to extract the polar coordinate image gradient, the loss is calculated, and the compensation coefficient is updated. According to the method, hyper-parameters do not need to be adjusted, images do not need to be trained, the ring and noise removal effect is good, and the automatic process is completely achieved.

The invention discloses a critical dimension scanning electron microscope image processing method and electronic equipment, and the method comprises the steps: employing a pre-trained image processing model to carry out the processing of an input image, and obtaining an optimized image; the pre-trained image processing model comprises a de-noising branch and an edge enhancement branch, the two branches are both formed based on a U-Net network architecture, the de-noising branch comprises a first encoder module and a first decoder module which are connected in sequence, and the first encoder module and the first decoder module both adopt an SC block module to extract local and global features of an image; the edge enhancement branch comprises an edge information extraction module, and a second encoder module and a second decoder module which are connected in sequence; and the feature fusion module is used for fusing all levels of output of the first encoder module and all levels of output of the second encoder module to serve as all levels of input of the first encoder module. Compared with a traditional image denoising algorithm, the algorithm solves the problem of edge blurring caused by excessive smoothness of the image while denoising.

The invention relates to the technical field of data processing, in particular to a blowing flow adjusting system in the metal powder machining process. The system comprises the steps of collecting flow data, obtaining a mutation factor of each data in a main body flow data sequence, and obtaining a noise factor of each data in the main body flow data sequence according to the mutation factor of each data in the main body flow data sequence; based on the noise factor, adaptively obtaining a judgment threshold value of a Mahalanobis distance of each datum in the main body flow data sequence; according to the judgment threshold value of the Mahalanobis distance of each data in the main body flow data sequence, noise data is obtained for smoothing, the denoising effect is improved, and the influence of abnormal flow data on noise data recognition is avoided.

The present application relates to the field of one-way valve check detection technology, and specifically to a one-way valve check performance detection method and system, which specifically includes: obtaining the first and second check fluctuation sets of each local check sequence by classifying the data in the neighborhood of the air pressure difference at each moment, and constructing the local check fluctuation value based on the data discreteness in the first and second check fluctuation sets; classifying the elements in the local check sequence based on the data anomaly and data discreteness of the local check fluctuation value, and obtaining the replacement value of the central element of the local check sequence based on the classification result; obtaining the pressure difference correction sequence based on the sequence composed of the replacement values; and performing the one-way valve check performance detection based on the difference between the pressure difference correction sequences under all kinds of air pressure. The influence of vibration noise on the pressure difference data is suppressed, the accuracy of the air pressure difference data at both ends of the one-way valve is improved, and the accuracy of the one-way valve check performance detection is thereby improved.

This application discloses an image processing method and apparatus in the computer vision field of the artificial intelligence field. The image processing method includes: obtaining a channel image corresponding to a to-be-processed raw (raw) image, where the channel image includes at least two first channel images; performing image restoration processing on the first channel image to obtain the restored first channel image; and performing image restoration processing on the to-be-processed raw image based on the restored first channel image to obtain a raw image obtained through joint demosaicing and denoising processing. The technical solution of this application can improve image quality of the raw image obtained through joint demosaicing and denoising processing.