185 results about "Smoothing" patented technology

The invention discloses an airborne laser radar sounding data processing method and system, and relates to the technical field of marine surveying and mapping and earth observation, and the method comprises the following steps: collecting original full-waveform data of an airborne laser radar sounding system, the original full-waveform data comprising infrared laser channel data and blue-green laser multichannel data; joint denoising processing is carried out on the original full-waveform data to obtain denoised waveform data, and the joint denoising processing comprises background noise modeling and removing based on machine learning and random noise filtering based on frequency domain low-pass filtering; a collaborative joint denoising strategy is formed by fusing background noise modeling based on machine learning and frequency domain low-pass filtering based on signal-to-noise ratio optimization. According to the method, complex background noise can be accurately estimated and removed, random noise can be adaptively filtered out, weak underwater echo signals are reserved to the maximum extent, and the problem of water depth information loss caused by excessive smoothness is effectively avoided.

The invention discloses an industrial defect detection method based on self-supervised pre-training and feature space generation. Firstly, features are extracted and mapped to a unified potential space; secondly, introducing a SimMIM framework, which is one of mainstream technologies in the current industrial vision pre-training field, to carry out self-supervised pre-training to improve the representation capability, and using an OfficientForme network to improve the reasoning speed and reduce the memory demand; secondly, performing abnormal synthesis in a feature space by adopting a generative adversarial network and enhanced Perlin noise, outputting a feature increment and a soft mask by a generator, and performing linkage updating with a pixel-level mask and an image-level label; thirdly, a segmentation-classification double-head framework is adopted, a segmentation head outputs a pixel-level anomaly graph, and a classification head outputs an image-level anomaly score; and finally, training optimization is carried out through strategies such as abnormal graph up-sampling smoothing, a grouping learning rate, multi-stage scheduling and the like. The method realizes accurate and controllable synthesis of the abnormal region under the scene without, with or with mixed supervision, and significantly improves the robustness and real-time performance.

The invention discloses a GRACE and Swarm time-varying gravity field fusion filtering method based on a state space model, and the method comprises the steps: taking a spherical harmonic coefficient as a state quantity, constructing a random walk process equation, introducing three types of observations, employing a quantization parameter for observation noise and process noise covariance, constructing according to an order / power law, and carrying out the self-adaptive updating along with the monthly; a Nelder-Mead method is adopted to search for spectral index parameters, and an EM algorithm and a statistical method are utilized to update other parameters in a closed / quasi-closed mode; obtaining the state and posterior covariance of a full time sequence by using Kalman filtering and RTS smoothing; in the GRACE and GRACE-FO window period, continuous reconstruction is carried out by means of a process model and Swarm; and outputting quality evaluation information including monthly gravity field coefficients, posterior covariance, innovative variance ratio, residual whitening test, space power spectrum, uncertainty band and the like. According to the method, while physical rationality and calculation feasibility are ensured, a continuous and stable monthly time-varying gravitational field sequence with quantifiable uncertainty is realized.

The invention relates to the technical field of vehicle suspension system control, and discloses a multi-source data fused shock absorber self-adaptive regulation and control method and system.The method comprises the steps that multi-source signal data are synchronously collected through a sensor, and a preliminary signal set is generated in combination with environment change data; analyzing vibration signal peak amplitude and displacement data integral variation trend, distinguishing high-frequency small-acceleration and low-frequency large-amplitude pavement excitation characteristics, and determining signal contribution degree distribution; dynamically adjusting the signal weight and introducing temperature compensation to correct stiffness drift to generate a weighted vector; carrying out weighting processing, smoothing, noise suppression and nonlinear amplitude estimation to generate fusion data, and optimizing the weight to obtain a refined weighted vector; after secondary fusion, a cross-frequency-band transition smoothing technology and a least square method are adopted to determine damping characteristic parameters; according to the vehicle suspension system self-adaptive regulation and control method and system, the self-adaptive regulation and control accuracy and the working condition adaptability of the vehicle suspension system are improved.

The proposed technique introduces embodiments of a computer-implemented method for interpreting image delineations as vector objects and topological extraction from segmentation by visual computational methods applied to sections (slices) of seismic volumes, in order to aid geological interpretation and sampling of parameters originating from the fracture network and its topology. Embodiments of a developed method integrates a software / application that allows the loading and generation of statistical data related to the fracture network while maintaining georeferencing and scale of the two-dimensional input data. In addition, a fracture segmentation method is shown that uses pyramid image smoothing (decomposition into hierarchical levels of resolution) in order to reduce the amount of details and aid the identification of main faults or fractures. The segmentation after this smoothing is based on adaptive thresholding segmentation.

The invention relates to a method and device for controlling the nanometer morphology of a double-sided thinned silicon wafer, and the method comprises the steps: obtaining the warping degree data of each position on the double-sided thinned silicon wafer, and obtaining a warping degree data curve; performing multi-stage differential smoothing processing on the warping degree data curve to obtain morphology parameters of the silicon wafer after double-sided thinning; and comparing the morphology parameters of the silicon wafer after double-sided thinning with a preset morphology parameter threshold value, and adjusting parameters of double-sided thinning equipment according to a comparison result. Batch production of defective products can be avoided.

The invention relates to the field of data analysis, in particular to a multi-source monitoring data-oriented dynamic trend assessment method, which comprises the following steps of: acquiring and preprocessing multi-source monitoring data to obtain a historical data sliding window and a basic smoothing coefficient for trend assessment; performing weighted correlation analysis on the disturbance variable and the measured variable change sequence to obtain a working condition response decoupling factor; obtaining a trend stability factor by evaluating the geometric morphology of the smooth trajectory of the measured variable; performing working condition response and trend stability combined correction on the basic smoothing coefficient to obtain a dynamic smoothing coefficient; a smooth trend value and a heat exchanger sub-health early warning signal are obtained by performing exponential weighted moving average and change rate judgment on a measured variable, and the problem that a fixed parameter EWMA cannot distinguish a working condition adjustment response and a tiny fault trend under a multivariable complex working condition is solved.

The invention relates to an intelligent optimization method for sludge deep dehydration process parameters. Key parameters are collected in real time through an Internet of Things architecture, a multi-dimensional dynamic working condition data set is constructed by using a sliding time window and standardization processing, data quality is improved through abnormal value filtering and noise smoothing, a CNN-GRU hybrid model is input, and precise prediction of a working condition evolution trend is realized. On the basis of a prediction result, a trend guide feedback channel, a preference evolution feedback channel and a stability feedback channel are constructed, a multi-objective optimization weight is adjusted in a self-adaptive mode through normalization and weighted fusion, and a high-correlation optimal process parameter combination and closed-loop feedback control are obtained by combining NSGA-II algorithm multi-objective optimization and trend alignment degree screening. According to the scheme, the data perception and multi-target adaptive optimization capability of the sludge deep dehydration system is effectively enhanced, and the operation efficiency, the energy consumption utilization and the system stability are improved.

The invention discloses a track precision measurement method based on design file matching. The method comprises the steps of design file resolving, time-space synchronization of multiple sensors and segmented matching of design files, inertial navigation resolving initialization, inertial navigation resolving, indoor and outdoor scene discrimination, data fusion resolving and reverse smoothing. The problem that an existing scheme cannot be suitable for various detection scenes is solved; the method has the advantages that automatic identification and full coverage of the track detection scene can be achieved, the reliability is improved through a software and hardware combination mode for different scenes, design files and historical observation data are effectively integrated, the method is suitable for various detection scenes on the basis that the measurement precision is met, and establishment of a track detection database is facilitated. The invention further discloses a track precision measurement system based on design file matching and a track precision measurement device based on design file matching.

Embodiments disclose a method and a system for robust transformation of input with a neural network. The method comprises processing the input data with a variational neural network (VNN) trained with ML to produce static parameters including noise level for the input data, injecting a set of random noises sampled on a probabilistic distribution according to the statistic parameters defined by the VNN to produce a set of perturbed input samples. The method comprises processing each of the set of perturbed input samples with a transformation neural network to produce a set of transformations and outputting a combination of the set of transformations as the robust transformation of the input data. Some embodiments consider training the variational neural network and transformation neural network by using adversarial examples from an attack model via alternating, explicit, and implicit gradient frameworks.

The invention discloses a multi-scale time sequence modeling video abstract generation method fusing semantic enhancement and boundary perception, and belongs to the technical field of computer vision. The method comprises the following steps: carrying out feature coding on an input video frame sequence, and extracting a corresponding time sequence feature representation; inputting the extracted time sequence features into a video abstract generation model, and outputting importance scores by the model; performing time sequence smoothing and peak value detection on the importance score, and determining a key frame candidate set; and generating a final video abstract fragment based on the key frame candidate set in combination with a non-maximum suppression and fragment combination strategy. Experimental results obtained on reference data sets SumMe and TVSum prove the advancement of the method. According to the video abstract generation method provided by the invention, the problems of fuzzy event boundary and insufficient multi-scale time sequence understanding in the existing video abstract generation method can be effectively solved.

The invention belongs to the technical field of automated guided vehicles, and particularly relates to a multi-source positioning fusion and navigation correction method and system for an AGV, and the method comprises the steps: carrying out the pose estimation through the bidirectional coupling of extended Kalman filtering and particle filtering, and the estimation uncertainty is fed back to the extended Kalman filter to dynamically adjust the noise parameter. And further introducing sliding window factor graph optimization to carry out back-end smoothing. In the navigation stage, a dynamic decision maker based on deep reinforcement learning is adopted, and optimal control parameters are generated according to real-time positioning, map semantics and historical performance. The system can also enable the filtering model and the decision strategy to adapt to the current environment through periodic online fine tuning. Through innovatively and deeply fusing a classical state estimation method and a leading-edge machine learning method, the positioning precision, the environment understanding capability and the navigation intelligence of the AGV in a complex dynamic scene are remarkably improved.

A navigation path can be determined for an object using one or more neural networks. In various embodiments, image data is obtained that is representative of an environment in which the object is to be navigated. Relevant features are identified from the image, and a curve fit to those features. Loss values for the potential paths are scaled based at least in part upon the distance of those features in the real world. This can include, in at least some embodiments, performing the scaling as a function of the curvature of the curve fit to the features. Temporal smoothing can be performed with respect to prior path predictions in order to prevent sudden changes in the predicted path. The paths are analyzed to select a path with a highest confidence value that also at least satisfies a minimum confidence criterion. The path can be converted into three-dimensional navigation information.

The invention discloses an industrial safety cross-domain collaborative analysis system based on time series data and visual data, and relates to the technical field of cross-domain data collaborative processing. Comprising a collaborative data acquisition and processing module used for acquiring multi-source monitoring data in real time, extracting a hot area characteristic parameter set and carrying out data preprocessing; the infrared thermal image time sequence response extraction module is used for quantifying a thermal response change rate and carrying out signal smoothing; the cross-domain time difference compensation alignment module is used for carrying out trend term elimination and offset scanning on the multi-source monitoring data, carrying out correlation quantification on the multi-source monitoring data and a thermal response change rate, obtaining the most accurate offset and realizing multi-modal signal time alignment; the cross-domain collaborative mutation analysis module is used for quantifying the mutation synchronization intensity of the multi-modal signal and identifying potential anomalies; and the leakage judgment grading feedback module is used for carrying out grading judgment and implementing response measures. The problem that the leakage early-stage characteristics are difficult to identify synchronously due to the time difference of the multi-mode response in the hazardous chemical substance pipeline and the storage tank area is solved.

The invention relates to the technical field of industrial internet-of-things monitoring, and discloses an internet-of-things-based remote monitoring method and system for asphalt mixture stirring equipment. The method comprises the following steps: acquiring key component operation parameters and external environment variables, and performing time domain statistical filtering to obtain an initial operation sequence; performing adaptive window length smoothing and frequency domain noise suppression to obtain a clean operation sequence; performing environment equivalent correction on the clean operation sequence by using the external environment variable to obtain a calibration feature vector; performing multi-dimensional feature correlation analysis according to the calibration feature vector, and determining a working condition switching label; determining a dynamic reference range according to the working condition switching label; calculating a preliminary abnormal probability according to the calibration feature vector and the dynamic reference range; and performing historical trend weighting and abnormal integration to obtain a final health assessment result. According to the method, through environment correction and dynamic reference following, environment interference and working condition misjudgment are eliminated, and accurate evaluation of the equipment health state is achieved.

The invention provides a high-fidelity state observation visual positioning system and method for a high-vibration platform, belongs to the field of machine vision and image processing, and is suitable for external closed-loop control of high-vibration platforms such as an unmanned aerial vehicle with arms. In order to solve the problems of measurement failure, large data noise and multi-sensor calibration errors caused by motion blur in the prior art, a single RGBD camera (a view field covers an external target and a mechanical arm visual label) is adopted, and a process of locking a region of interest by deep learning-measuring a pose by a high-precision algorithm-multi-label weighted fusion-time domain low-pass smoothing is combined. And outputting a high-fidelity pose signal. According to the method, multi-sensor calibration errors are eradicated, the anti-fuzzy capacity is high, signals are smooth to meet the control requirement, the positioning precision of the mechanical arm can be improved to the 3 mm level, and the dynamic task success rate is remarkably increased.

The invention discloses a narrow space automatic driving path planning method based on artificial intelligence, and the method comprises the following steps: S1, collecting laser radar point cloud and image data, executing semantic segmentation and clustering, and building a topological graph; s2, mapping topological graph nodes into pulse coupling neural network neurons, and constructing a double-domain coupling mechanism comprising a structure coupling item and a state coupling item; s3, generating a path direction vector, constructing a path intention gating factor, and dynamically adjusting a coupling input value; s4, periodically updating the neuron membrane potential, and adaptively adjusting the discharge threshold; s5, recording a first discharge node number and time, and generating a path activation sequence according to a discharge sequence; s6, performing connectivity check and curvature smoothing on the path activation sequence to generate a continuous track; and S7, converting the trajectory into a control instruction, and driving the vehicle to execute path tracking. According to the invention, high-precision, high-connectivity and high-control-stability output of path planning in a narrow space is realized.

A method for processing at least one decoded image region. At the exit of a decoding loop that has decoded at least one current image region, the decoded current image region is processed with the aid of a boundary-smoothing module that uses metadata relating at least to a boundary between the current image region and a neighboring image region.

The invention relates to the technical field of image processing, and discloses a road-air cooperative pavement disease digital management and control system based on a three-dimensional model, and the system comprises a data collection module which carries out the graying and smoothing of an unmanned plane image, so as to achieve the multiplicative detrending, and obtains a detrending brightness graph; a window generation module calculates a structure tensor based on the de-trending brightness graph, determines a wheel track main direction, and generates a window set according to a preset window size and a step length in a vertical direction; a priori graph generation module calculates normalized double coherence in each window and performs aggregation to obtain window feature values, and a priori graph is formed through interpolation; an intermediate feature map generation module generates a spatial gain map by using the prior map, and performs point-by-point gating on the intermediate feature map; the training sample construction module determines a sampling probability according to the window characteristic value and selects a sample; and the network training module carries out weighted segmentation on the loss training network through the pixel weight, and outputs a pavement disease detection result in a reasoning stage.

The application discloses a backdoor attack method based on color frequency injection and adaptive local enhancement, and relates to the technical field of machine learning and artificial intelligence security. The method comprises the following steps: introducing low-frequency color offset and weak high-frequency signal into an image in a CIELAB color space to perform global color-frequency injection; using a pre-trained proxy model to locate a high-sensitive perception domain of the model through mixed evaluation of gradients and class activation maps, and generating a binary mask; in an HSV color space, respectively applying nonlinear stretching factors to saturation and brightness of the sensitive domain based on the mask to perform adaptive local enhancement; using Gaussian smoothing, adaptive noise and histogram matching to eliminate edges and statistical abnormalities caused by local enhancement, completing compensation color enhancement to generate a poisoned image; and modeling a trigger core parameter as a constrained optimization problem, and using a particle swarm optimization algorithm to jointly dynamically update the trigger core parameter to obtain an optimal strategy. The application anchors the trigger feature depth in the core semantic area of the model and lurks in the normal data manifold, guarantees a high attack success rate, realizes extreme visual and feature concealment, and has strong anti-defense robustness.

The invention provides a multi-radar information fusion measurement method for multiplex rocket suspension recovery, which comprises the following steps: acquiring a distance measurement value of the surface of a rocket through a plurality of millimeter-wave radars arranged on a rocket recovery device, and screening out effective measurement points based on a preset effectiveness threshold value; based on the screened effective measurement points, solving an initial estimation value of a circle center position by using circular geometric constraints of the cross section of the rocket; and constructing a state vector containing the center position and the motion state of the rocket, and performing multi-frame data fusion on the initial estimation value by using a Kalman filtering algorithm to obtain the optimal estimation of the center position after smoothing. According to the method, effective data of multiple radars are fused, geometric constraints of the cross section of the rocket are introduced to carry out space domain circle fitting, and dynamic smoothing and tracking are carried out in a time domain in combination with Kalman filtering, so that the purpose of improving the measurement accuracy of the landing pose of the rocket is achieved.

This invention relates to a data processing system for CTD (Conductivity, Temperature, Depth) instrument observations, comprising: building a fully domestically developed technology foundation, employing a combination of Spring Boot, MyBatis-Plus, DM database, and Vue3 to achieve domestic adaptation of the front-end, back-end, and database; incrementally accessing and storing raw CTD data files based on the technology foundation, generating dual traceability identifiers of FileInfoID and DataID; performing multi-step sequential processing on the stored data, including zero-error correction, outlier filtering, data smoothing, salt density and acoustic calculation, and standard file generation, with each step's algorithm optimized for the characteristics of domestically produced CTD data; storing raw data, intermediate data, and output data through a multi-level data table design, utilizing traceability identifiers to achieve full-process correlation and traceability, and providing visual display and standardized output. This invention achieves fully domestic processing from data access to output, effectively improving the accuracy and traceability of data processing and ensuring marine data security.

The present application relates to the technical field of industrial internet of things monitoring, and discloses a remote monitoring method and system for asphalt mixture mixing equipment based on the internet of things.The method comprises the following steps: collecting key component operating parameters and external environmental variables, performing time domain statistical filtering to obtain an initial operating sequence; performing adaptive window length smoothing and frequency domain noise suppression to obtain a clean operating sequence; using the external environmental variables to perform environmental equivalent correction on the clean operating sequence to obtain a calibration characteristic vector; performing multi-dimensional feature correlation analysis according to the calibration characteristic vector to determine a working condition switching label; determining a dynamic reference range according to the working condition switching label; calculating a preliminary abnormality probability according to the calibration characteristic vector and the dynamic reference range; performing historical trend weighting and abnormality integration to obtain a final health assessment result.This method eliminates environmental interference and working condition misjudgment through environmental correction and dynamic reference following, and realizes accurate assessment of the health state of the equipment.

The application relates to a wave height parameter correction method and system for low-angle laser radar observation data, and relates to the technical field of ocean observation data processing. The method comprises the following steps: firstly, performing smoothing and data cleaning processing on original wave height data to construct a fitting model of polynomial and Gaussian global trend analysis. Then, through dynamic fusion of double models, the depth of capturing short-term fluctuations of the polynomial and capturing long-term trends of the Gaussian is combined, and the model generalization ability is improved. Finally, intelligent reference interval positioning is introduced to adapt to the wave height distribution changes in different periods and positions, and a correction factor is introduced to avoid over-correction, so that accurate correction data is obtained, the problem of wave height attenuation with distance is solved, and the efficiency and reliability of the correction processing are significantly improved.

The application discloses a kind of unstructured scene under the method for dividing and generating reference line of passing boundary, it is related to the technical field of car intelligence planning, and its technical solution main points are: obtaining discrete path data, pre-processing, obtaining pre-processed data, and calculating the heading and curvature information of reference line;Receive road network information and input A* algorithm, obtain global rough path, convert global rough path from Cartesian coordinate system to Frenet coordinate system, then establish driving space node graph, input driving space node graph into dynamic programming algorithm, obtain left and right rough boundary, realize smoothing involving equality constraint initial solution in combination with quadratic programming model;Establish the kinematic model of vehicle motion of arc length parameter and discretization AL-iLQR solver combined with multi-constraint optimization, generate smooth reference line.The application significantly improves the path modeling efficiency and noise resistance, optimizes the real-time and safety of passing boundary smoothing, and efficiently generates kinematically feasible trajectory.

The invention discloses a two-stage adaptive numerical distribution reconstruction method based on local differential privacy. And the terminal equipment applies a local differential privacy perturbation mechanism to the held numerical private data to generate perturbation data and sends the perturbation data to the aggregation server. After an aggregation server collects data, first-stage estimation is executed firstly, and initial smooth distribution with low variance characteristics is generated based on an EMS algorithm to serve as a guide map; then, a non-uniform self-adaptive bucket dividing strategy is constructed according to cumulative distribution characteristics of initial distribution, and equal probability interval division is achieved; and finally, second-stage estimation is executed, re-statistics is carried out on noise data based on a self-adaptive bucket dividing strategy, accurate reconstruction is carried out by applying an expectation maximization algorithm with space kernel smoothing, and weighted updating is carried out on threshold-controlled space smoothing by utilizing a kernel function based on a bucket center physical distance. According to the method, the problem of over-fitting of discretization deviation of uniform bucket division and an expectation maximization algorithm in a local differential privacy high-noise environment is effectively solved through a two-stage strategy, and the accuracy and robustness of value distribution estimation are remarkably improved.

The present application relates to the technical field of data processing, more particularly, the present application relates to a kind of hydraulic system oil circuit fault early warning method and system, comprising: collecting parameter data in hydraulic system oil circuit, the parameter data includes pressure in oil circuit;Pressure prediction value at any time is predicted using improved exponential smoothing method, and according to the size of pressure prediction value at current time, fault early warning is carried out, wherein the initial smoothing coefficient of the pressure actual value collected at each time is modified in the improved exponential smoothing method.The present application not only considers the inherent characteristics (such as mutation degree and fluctuation degree) of pressure data when calculating reliability, but also introduces the environmental purity index of hydraulic oil, more comprehensively evaluates the reliability of data, can more accurately reflect the change trend of pressure data by dynamically adjusting the smoothing coefficient of pressure actual value at each time, reduces the influence of prediction deviation and data noise, so as to improve the accuracy of fault early warning.

The application relates to the technical field of region division, and discloses a region division method, system, terminal and storage medium based on edge segmentation and clustering analysis, the method comprising the following steps: obtaining a target building of a target region, performing statistical processing on the grid flow of the target building, performing edge segmentation, and obtaining an initial division result; constructing a boundary buffer zone, performing gradient calculation to obtain gradient data, and performing position correction on boundary nodes of an initial spatial boundary to obtain a division optimization result; performing region clustering processing on the division optimization result according to clustering parameters to obtain a clustering result, and performing screening processing on the clustering result to obtain a target clustering result; and performing smoothing optimization on the boundary of all clustering regions of all levels in the target clustering result to obtain a target division result. Through integrated processing of initial grading, edge correction, density clustering and geometric smoothing, the stability of data statistics and the accuracy of the division result are improved.

The present application relates to the technical field of image processing, and provides an image smoothing processing method and a target contour extraction method, the image smoothing processing method first acquires a to-be-smoothed image; then substitutes the to-be-smoothed image into a smoothing model, and solves the smoothing model to obtain a first smoothing result corresponding to the to-be-smoothed image. Thanks to the respective characteristics of the norm and the Laplacian operator, and the pre-performed texture removal filtering operation, the smoothing model can effectively remove various interferences such as noise and texture, the structure of the smoothing result is more prominent, the smoothing effect is improved, the subsequent application of the smoothing result is guaranteed, and the accuracy and robustness of the subsequent contour extraction and positioning process of the target object are enhanced.

The present application relates to the field of image processing, in particular to a real-time monitoring method and system for injection molding production line based on machine vision, which divides an image into multiple analysis regions, calculates the region noise degree representing the brightness instability of the region, calculates the pixel noise confidence in combination with the deviation degree of the pixel gradient and its neighborhood, and then fuses the multi-dimensional information to obtain the real noise probability of each pixel. Finally, the smoothing strength of bilateral filtering is adaptively adjusted based on the probability, and strong smoothing is applied to the pixels with high noise probability. The present application can accurately identify and process high light noise, effectively denoise while retaining the real defect details to the maximum, and significantly improve the monitoring image quality and defect detection accuracy.