130 results about "Median filter" patented technology

The invention discloses a construction method of a multi-underwater robot cooperative control system, which relates to the field of underwater robots, and establishes a multi-parameter coupled data acquisition and preprocessing mechanism by using a Doppler current profiler ADCP, an inertial measurement unit IMU and controller software. A combined algorithm of sliding median filtering and Kalman filtering is adopted, and data disturbance caused by environment noise and attitude drift is effectively removed; and through abnormal point triple standard deviation determination and double adjacent point interpolation completion, stable correction of sampling data is realized. A standardized data set obtained after normalization processing unifies parameter scales of different physical magnitudes, so that the local average flow velocity, the flow velocity fluctuation root mean square and the average acceleration can be directly compared in a feature space. Therefore, according to the method, the high-consistency expression of the water flow disturbance characteristics in time and space is realized, so that the synchronous measurement precision of multiple underwater robots in a non-uniform flow field is remarkably improved.

The invention discloses a coal mine production anomaly detection system and method based on time series decomposition and prediction, and provides an end-to-end decomposition-prediction anomaly detection framework, the framework firstly constructs a deep STL iterative decomposition network, and uses a differentiable median filtering and iterative orthogonalization mechanism to obtain a deep STL iterative decomposition network; an original power load sequence is decoupled into three orthogonal components, namely a trend component, a season component and a residual error in a robust manner, and modal aliasing is effectively inhibited; then, a parallel multi-branch prediction architecture is adopted, aiming at the physical characteristics of each component, an implicit neural predictor based on a polynomial basis / Fourier basis and a hierarchical frequency domain Transform-CNN are respectively used for special modeling and prediction, and anomaly judgment is realized based on a prediction error; according to the method, objective and tampering-resistant power data can be fully utilized, the anomaly detection performance of the non-stationary sequence is remarkably improved, and a reliable non-intrusive intelligent monitoring solution is provided for coal mine production safety.

The invention discloses a microstructure mapping modeling and mechanical simulation method suitable for a two-phase structure alloy, and the method comprises the steps: carrying out the self-adaptive median filtering denoising and Retinex enhancement of a metallographic diagram, so as to improve the gray scale comparison of a matrix and a second phase; training a U-Net + + segmentation model by using the enhanced image to complete pixel-level phase region extraction; a two-phase geometric model is established according to the segmentation result, a finite element feature data set is obtained through multi-physics field coupling calculation, lossless fusion and topological consistency verification are conducted on the finite element feature data set and a prior model, and a fusion model is obtained; simulation parameters are set based on actual working conditions, mechanical simulation is executed, and mechanical property parameters and response curves are output. The method can accurately reflect the mechanical behaviors of the two-phase structure alloy under different working conditions, provides a scientific basis for material design and performance optimization, has high universality, and can be widely applied to performance analysis and evaluation of various types of alloys.

The invention relates to the technical field of data processing alignment, in particular to a long-distance photoelectric sensor focusing calibration method and system, and the method comprises the following steps: starting a laser transmitter with an end to monitor a photoelectric array, obtaining the light intensity and three-dimensional coordinates of multiple sampling points, and carrying out the filtering; and extracting edge and radial light intensity, calculating a gradient vector through derivative difference and generating an offset direction vector, extracting symmetrical sampling points based on the vector to construct a balance ratio sequence, generating an offset angle through a mean square error model, and mapping the angle and the vector to generate a calibration instruction. According to the method, multi-sampling-point light intensity is associated with three-dimensional coordinates, median filtering is executed to remove noise, gradient is extracted through derivative difference to achieve spatial offset direction mapping, an offset angle value is finely solved in combination with a balance ratio sequence and a mean square error model, and parameter stability is ensured; the offset vector and the angle are correlated to execute coordinate mapping operation to compensate the angle deviation, real-time automatic correction of the pose of the transmitting end is achieved, and the long-distance alignment efficiency and the signal intensity of the sensor are improved.

The invention discloses a printer motor control method and system based on a current detection front-end conditioning circuit, and relates to the technical field of current detection, and the current detection front-end conditioning circuit comprises a precise sampling resistor and a multi-stage RC filter network; the method comprises the following steps: collecting a motor phase current signal through a current detection front-end conditioning circuit, and converting the motor phase current signal into a digital current signal after differential amplification and filtering; obtaining a direct-axis current component and a quadrature-axis current component through coordinate transformation; setting a reference quadrature-axis current, and setting a direct-axis current reference value as 0; calculating a direct-axis voltage and a quadrature-axis voltage, and generating a switching signal for driving an inverter; in each control period, performing median filtering on the current sampling signal to adjust the integral gain of the PI controller; and identifying the abnormal state of the motor based on the time domain characteristics of the digital current signal so as to adjust the operation strategy of the 3D printer. High-precision current sensing and abnormal state recognition are both considered, and the printing quality of the 3D printer is improved.

The application relates to the technical field of quantifiable test indexes of new energy vehicles, and particularly discloses a method for constructing a typical road spectrum working condition based on a genetic algorithm and new energy platform data, data acquisition, invalid data section elimination rules including static data section elimination and abnormal continuous duration elimination; data abnormal value identification rules: speed abnormal values are defined and a correction method is set; data smoothing and denoising rules: a median filter is used to preliminarily denoise a speed sequence; characteristic parameter calculation: including kinematic characteristic parameters and new energy light truck characteristic parameters; construction of an initial section library; multi-objective genetic algorithm optimization; road spectrum synthesis and post-processing: optimal sections filtered by the algorithm are spliced, and smoothing processing is carried out to ensure the continuity of a speed curve; multi-objective genetic algorithm selection: selecting an NSGA-III algorithm as a multi-genetic objective algorithm condition; the application realizes rapid construction and iteration of the road spectrum, and is high in efficiency.

This application relates to a pipeline inspection method, apparatus, computer equipment, storage medium, and computer program product. The method includes: acquiring a multi-channel raw signal sequence collected within a target pipeline; performing median filtering on the raw signal sequence based on a preset sliding window to obtain an intermediate observation sequence; performing dynamic prediction filtering on the intermediate observation sequence to obtain a reference filtered signal; wherein the dynamic prediction filtering is used to predict candidate sequences based on the intermediate observation sequence; and fusing the intermediate observation sequence and the candidate sequences; identifying abnormal sequences in the raw signal sequence based on the reference filtered signal; correcting the abnormal sequences to obtain an initial corrected sequence; and performing secondary filtering on the initial corrected sequence to obtain a target filtered signal. This method can effectively improve the signal-to-noise ratio and anomaly identification accuracy of pipeline inspection data.

The application relates to a Beidou RTK measurement point cloud self-adaptive filtering processing method, belonging to the field of high-precision three-dimensional positioning technology based on the Beidou system. The method firstly determines whether it is suitable to directly take the median filtering result as the filtering output according to the data amplitude span state of any measurement point in the Beidou RTK measurement point cloud and the effective satellite quantity when the point cloud is acquired, and quantifies the reliability of the median filtering result through the data fluctuation and direction change of the local neighborhood of the measurement point when it is not suitable, so that the fusion filtering result determination of the median filtering result and the original measurement value is completed in the case that the direct median filtering result is not suitable for filtering output. The application can avoid the reverse filtering optimization effect of the median filtering in the Beidou RTK measurement point cloud, and improve the RTK measurement point cloud data filtering accuracy of the median filtering.

This invention belongs to the field of power monitoring technology and discloses a brightness balancing and regional smoothing fusion algorithm for power monitoring images. The method includes: image denoising, using mean filtering and median filtering to smooth the image; image registration, performing optimal matching on two or more images obtained from different sensors, at different times, or from different angles; and image region fusion, smoothly fusing images with different angles and brightness contrasts. This method is applicable to panoramic video monitoring systems. Designed specifically for the complex lighting and high electromagnetic interference characteristics of power operation sites, this invention effectively eliminates on-site noise through adaptive preprocessing, and achieves a balance between global brightness uniformity and local detail preservation through layered brightness balancing, solving the problems of incomplete brightness balancing and detail loss in power scenarios encountered by general algorithms.

This application provides a TDLAS gas sensor and its anti-interference filtering and noise reduction method, relating to the field of gas concentration detection technology. The method includes: acquiring the original analog-to-digital conversion data of the gas to be measured; using an adaptive median filtering algorithm to filter, reduce noise, and correct the baseline of the original data to obtain a time-domain signal; and calculating the gas concentration based on Beer-Lambert's law. The adaptive median filtering algorithm uses multi-resolution analysis to decompose the signal into low-frequency and high-frequency components, identifies noise regions and absorption peak regions by calculating local gradient changes, and performs multi-scale reconstruction after adaptive threshold classification. This invention addresses the problem of TDLAS gas sensors being susceptible to interference from environmental temperature and pressure fluctuations. By protecting the absorption peak region while filtering out the noise region, it significantly improves the signal-to-noise ratio and detection stability, and has the advantages of strong anti-interference capability, fast response speed, and significant noise reduction effect.

The present application relates to the technical field of infrared imaging, and relates to an infrared image adaptive denoising method and system based on FPGA, which is suitable for finding a blind element algorithm module to capture a certain number of frames of data at the FPGA end, performing multi-directional gradient detection on each pixel point, setting a condition to determine the position of the blind element and marking it; a convolution kernel adaptive replacement blind element module is used to compare the convolution kernel slider and the gradient at the PL end of the FPGA, multiple reuse of the adaptive replacement blind element module is used to ensure complete coverage of the large blind element group, and the adaptive median filter module is used to intelligently select and replace the pixels that meet the requirements by automatically sorting the center pixel and the surrounding pixels; surface blur filtering is performed through a surface-blur algorithm module, so that the smooth areas in the image are effectively smoothed, while the details of the edge areas are retained, blind elements in the video stream can be effectively detected and compensated, and traditional noise can be inhibited through multi-stage filtering while the image details are retained.

The application belongs to but is not limited to the technical field of electrocardiogram anomaly detection, and discloses an ST segment anomaly detection method based on reverse transformation and fusion attention, adopts a median filtering method to perform denoising processing on input data samples, performs data augmentation on abnormal ST data, and utilizes a convolutional neural network model fused with reverse transformation and channel prior attention mechanism to classify electrocardiogram signals. The application utilizes reverse transformation and time sequence channel convolution to ensure that features can be effectively extracted and fused, and combines a channel prior attention mechanism to enable the model to accurately capture the correlation of each dimension space in the electrocardiogram signal while retaining relevant time sequence information, thereby improving the accuracy of ST segment anomaly detection.

The present application relates to the technical field of data preprocessing, and provides a power system frequency measurement data preprocessing method and system, which comprises: obtaining measurement data and dividing into several intervals; for each interval, detecting pulse noise points, determining the size of a first window according to the ratio of the number of pulse noise points to the total number of all data points in the interval, and performing median filtering on the detected pulse noise points in the interval by using the first window according to the size of the first window; selecting a second window, and suppressing Gaussian noise by weighted mean filtering according to the similarity between any data point in the second window and the center point of the window. The three factors of noise reduction effect, feature protection and calculation speed are considered.

The invention discloses a photovoltaic cell photoluminescence image noise identification and suppression method and system, and relates to the technical field of photovoltaic cell defect detection.The method comprises the steps that photoluminescence images of a photovoltaic cell under dark field and bright field conditions are collected through an image sensor; pre-processing the image, wherein the pre-processing comprises blind pixel correction based on a manual calibration method and black level correction based on absolute difference operation; based on the preprocessed image, fixed pattern noise is identified through a multi-frame comparison technology, and a noise position mapping matrix is generated; and based on the matrix, carrying out noise suppression on the image by adopting a selective multi-stage median filtering strategy, and outputting a corrected image. According to the method, the problems of image quality reduction and low defect identification precision caused by fixed-mode noise interference in the prior art are effectively solved, efficient processing and detail reservation are taken into account, and the accuracy and reliability of photovoltaic cell defect detection are remarkably improved.

This invention discloses a traffic control method and system for an inspection robot and an electric shuttle bus. The method includes: S1. collecting laser ranging data of the wheel axle cross-section of the electric shuttle bus; S2. performing large-step sliding median filtering; S3. filtering valid data segments containing axle data; S4. extracting valid data segments and performing small-step sliding median filtering on them; S5. calculating the mode of the amplitude in the valid data segments and replacing abnormal amplitudes exceeding a preset threshold with the mode; S6. identifying abrupt change points in the wheel axle profile in the valid data segments; S7. filtering abrupt change points; S8. calculating the position and forward speed of the electric shuttle bus; S9. transmitting the position and forward speed of the electric shuttle bus to the robot scheduling system; S10. making traffic control decisions based on the real-time position and speed prediction results of the electric shuttle bus and the real-time task path of the robot. This invention has advantages such as improved operational efficiency and safety.

This invention relates to a membrane tension compensation method based on the detection of uniformity in the thickness of water-soluble membranes produced by dishwashing pods. Addressing the problem that membrane thickness distribution is sensitive to nonlinear tension disturbances under high-speed operation conditions during water-soluble membrane production, and that traditional thickness control methods struggle to achieve precise closed-loop compensation, this invention proposes a method that continuously collects cross-sectional thickness information of the membrane material, constructs a time-space thickness sequence, extracts thickness gradient topological features, combines a multidimensional tension sensitivity feature vector with a dynamic thickness gradient topological graph, and uses a layer-by-layer sparsely connected lightweight graph convolutional network to infer the tension compensation amount. After median filtering and denoising, the result drives the actuator to adjust in real time, achieving high-precision closed-loop compensation for the uniformity of the water-soluble membrane thickness. This method improves the sensitivity and stability of thickness control under nonlinear disturbance environments, while also possessing both real-time performance and computational efficiency.

The application discloses a noise suppression method and system for hyperspectral medical image acquisition, and relates to the technical field of hyperspectral imaging signal processing. An initial hyperspectral medical image is obtained by pretreating a sample to be detected; a pre-trained CNN network is used to screen the initial hyperspectral medical image; a denoising algorithm combining median filtering and wavelet transform is used to perform median filtering on the screened initial hyperspectral medical image, then the filtered image is decomposed through wavelet transform to obtain a wavelet coefficient matrix, a new wavelet coefficient matrix is generated according to the principle of median filtering, image reconstruction is performed through the obtained new coefficient matrix, and finally, a final hyperspectral medical image after noise suppression is obtained according to wavelet threshold denoising. The application can enhance the edge and detail features of the hyperspectral medical image and improve the imaging quality.

The application discloses a kind of visual feedback drive's water tank cleaning robot intelligent control method and system, belong to robot control technical field.It includes: obtaining water tank inner wall original image, generates feature enhancement image by median filter and contrast limited adaptive histogram equalization;Input semantic segmentation model identifies stain area, calculates stain abundance index;According to stain abundance index, determine pollution grade, dynamically adjust cleaning motor speed and electric pressure push rod displacement, to change brush head positive pressure;In the process of advancing, collect cleaned area image and calculate residual stain abundance index, if greater than threshold value, then record non-compliance area three-dimensional coordinates, complete global scanning and plan compensation path to compensate cleaning.The application effectively improves cleaning efficiency and compliance rate, reduces energy consumption and mechanical loss.

The invention relates to the field of image processing, in particular to an airborne infrared image ship target segmentation method based on an improved beetle antennae optimization algorithm, and the method comprises the steps: obtaining an airborne infrared grayscale image, and carrying out the median filtering of the image; enhancing image target edge features; obtaining an edge detection image; removing a connected region with a small number of pixels in the edge detection image, and performing morphological closed operation on edge information; selecting the longest edge as a coastline, generating a segmentation mask image by using the coastline, and performing negation on the segmentation mask image to generate a sea-land segmentation mask image; segmenting the airborne infrared grayscale image by using an improved beetle antennae optimization algorithm to obtain a preliminary segmented image; and multiplying the sea-land segmentation mask image and the preliminary segmentation image pixel by pixel to obtain a ship identification result image. According to the method, the robustness of the algorithm is enhanced, and target segmentation requirements under different sea conditions are met.

Embodiments of the present application disclose a satellite telemetry data feature selection method and device based on a multi-objective optimization algorithm and a medium; the method can include: splicing the original telemetry data of the satellite, and screening out constant values in numerical data and time alignment, and based on the Rayleigh criterion, the 53H algorithm and the median filtering algorithm, removing outliers and noise in the original telemetry data to obtain preprocessed telemetry data; improving the traditional fast non-dominated sorting genetic algorithm NSGA-II according to the non-uniform arithmetic crossover operator and the mutation operator based on the fitness level; determining the objective function for feature selection according to the feature quantity, the root mean square error and the correlation function; based on the objective function, selecting feature data for data completion, satellite state prediction and fault diagnosis from the preprocessed telemetry data by using the improved NSGA-II algorithm.

The invention relates to a multi-beam data screening optimization processing method based on image sonar. Aiming at the problems of weak target signal annihilation and insufficient image sharpness caused by a traditional arithmetic average method, the method adopts a maximum value screening strategy to carry out multi-beam data fusion. The method comprises the following specific steps: carrying out preprocessing and TVG compensation on sonar original echo data; obtaining a plurality of beam data through parallel beam forming; for each distance unit, traversing and comparing the signal intensity of all beams, and selecting the maximum value as the final pixel value of the unit; and then median filtering noise reduction and configurable point picking processing are carried out. According to the method, the detection probability of a weak target and a point target can be remarkably improved, a sonar image with high contrast and high sharpness is generated, and meanwhile, different detection tasks are flexibly adapted by integrating multiple fusion and point selection modes. The method is efficient in calculation, is easy to realize on embedded platforms such as a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array) and the like, and has good engineering application value.

This disclosure provides a method for locating and measuring the height of a display panel ink-coated barrier, relating to the display field. The method includes: acquiring an original barrier region image including an inner barrier and an outer barrier; performing anti-interference preprocessing on the original barrier region image using an anti-interference preprocessing module to obtain a preprocessed barrier region image, which includes: the anti-interference preprocessing module using db wavelet decomposition to zero high-frequency fringe coefficients and then inverse transforming to suppress periodic interference fringes; the anti-interference preprocessing module using variable window adaptive median filtering to remove residual noise and display substrate scratch images; inputting the preprocessed barrier region image into an improved Unet network to obtain pixel-level coarse boundaries of the barrier; and locating the pixel-level coarse boundaries of the barrier based on a camera calibration module to locate the sub-pixel-level boundaries of the inner and outer barriers, thereby improving the anti-light source interference capability during the positioning process of the display panel ink-coated barrier.

The application discloses a kind of based on random sampling consistency and change point detection atomic clock integrity monitoring method, comprising: for the frequency difference data of atomic clock in current sliding window, using random sampling consistency algorithm to construct temporary trend model, based on the inner point residual statistics of this model corresponding to set dynamic threshold to judge all data points in current sliding window, determine abnormal point;And the data value of abnormal point is replaced with the predicted value fitted by temporary trend model;Until all sliding windows are traversed, complete data repair and output all abnormal point positions;The frequency difference data of atomic clock is median filtered, and the noise median absolute deviation estimate value is calculated;Based on the value and the total number of data points in current sliding window, a penalty item benchmark is constructed to generate a penalty value sequence;Then the PELT algorithm is used to select the selected change point number to determine the change point position, so that a kind of high-precision atomic clock integrity monitoring method is realized.

The application discloses an alloy surface detection system based on image recognition, comprising: an alloy surface image acquisition unit, which is used for reading alloy surface image information through an image acquisition machine; an alloy surface image recognition and extraction unit, which is used for pre-processing alloy surface image information, obtaining an alloy surface grayscale image and performing median filter noise reduction processing; obtaining contour information through an edge detection algorithm; an analysis unit, which is used for obtaining structural features of the contour information, analyzing alloy surface quality by detecting the structural features of the contour information, and more conveniently grasping problems of abnormal image information of the image surface by detecting the structural features of the contour, so as to reduce a loss rate in the production process of the alloy and facilitate systematic grasping of alloy surface detection.

The application provides a kind of electronic fence design method and device for construction safety control under near electric environment, comprising: through the ranging communication of mutually executed UWB base station that has been deployed and the generation of calibration parameters combined with preset calibration point, coordinate calibration is completed;Base station and personnel positioning tag bidirectional ranging obtain round-trip time delay raw data;Round-trip time delay raw data is protocol analyzed, median filtered and discarded based on statistical threshold exception data, to obtain pure ranging data set;Pure ranging data set is input into positioning engine, and real-time coordinates are obtained by trilateration and least square solution;Space relationship calculation is carried out with preset electronic fence model, and the final real-time position state and risk level are generated through buffer and hysteresis mechanism and risk judgment logic based on uncertainty;Trigger corresponding alarm and push management platform.In this way, the coordinate precision of positioning engine solution is improved, and millisecond-level accurate identification and hierarchical active warning of construction personnel approaching dangerous area are realized.

The invention discloses an image measurement system, which comprises an input device for inputting an image to be measured, and a processing device integrated with a preprocessing unit, a morphology fitting unit and an angle calculation unit, the preprocessing unit is used for performing two-dimensional convolution filtering and sliding window median filtering on the to-be-detected image through the parallel filtering architecture to obtain a first image; the morphology fitting unit is used for obtaining morphology fitting parameters of the target structure in the first image through least square fitting; the angle calculation unit is used for comparing the morphology fitting parameter with a preset reference model parameter to obtain an angle measurement result of the target structure and a specific direction; the system further comprises an output device for outputting and displaying angle measurement results. According to the scheme, under the condition that influence factors such as noise, gray fluctuation and local interference exist in the image quality, parameters reflecting the real morphology of the target structure can still be stably obtained, more reliable angle measurement is achieved accordingly, and the accuracy of an angle measurement result is improved.

The invention belongs to the technical field of vehicle signal processing, and particularly relates to a wheel acceleration signal processing method based on median filtering and low-pass filtering, and the method comprises the steps: collecting a wheel speed signal of a vehicle, carrying out the cascading of median filtering and low-pass filtering, and giving full play to the cooperative advantages of two algorithms. The combination fundamentally overcomes the inherent defects of a single filtering strategy, so that the signal-to-noise ratio of the finally output wheel acceleration signal is greatly improved, control misjudgment, system instability or function failure caused by signal noise are fundamentally avoided, and the safety and stability of a vehicle are greatly improved; the current working condition is intelligently judged according to the real-time running state of the vehicle, and the median filtering window length N and the low-pass filtering cut-off frequency are dynamically adjusted according to the current working condition. According to the mechanism, the filtering parameters are not fixed any more, but can be optimally matched with the driving state of the vehicle, so that the robustness and the comprehensive performance of the control system are remarkably improved.

The application relates to the technical field of signal analysis, and particularly discloses a chip single-wire IIC signal analysis method based on function testing, which comprises the following steps: step S1: performing dynamic window median filtering on an input signal, effectively eliminating signal jitter and burrs through dynamic expansion of a sliding window and median replacement, and outputting a smooth digital signal; step S2: performing virtual clock generation, obtaining a time interval sequence based on jump edge detection, calculating interval variance by using a sliding window, screening effective intervals according to a variance threshold, and finally equally dividing to generate a virtual clock; and step S3: sampling at the center of a signal steady state interval by using the clock, and analyzing sampling data according to an IIC protocol state machine to complete data decoding; the whole process is realized by a pure software algorithm, the hardware limitation is broken through, and the analysis precision and equipment compatibility are significantly improved.

The application discloses a wide field of view video image change detection method and device, and the method comprises the following steps: using an improved adaptive fast guided filter for video image change detection; after filtering the multi-temporal video image, a logarithmic ratio operator considering neighborhood information is proposed, the mean ratio difference graph generation mode is improved, and a difference graph is generated; the improved MR image and the improved LR image are subjected to image fusion through discrete wavelet transform, and a fused difference graph is obtained; a soft threshold function is used to perform initial classification on the fused difference graph, and an initial change area and an unchanged area are obtained; a cumulative distribution function is used to compress the pixel value of the classification result from [0, 255] to [0, 1]; a super-fast robust constraint fuzzy C-Means clustering algorithm is proposed, and an improved adaptive median filter is used for denoising processing. The device comprises a processor and a memory.