1521 results about "De noising" patented technology

A method and apparatus for de-noising weak bio-signals having a relatively low signal to noise ratio utilizes an iterative process of wavelet de-noising a data set comprised of a new set of frames of wavelet coefficients partially generated through a cyclic shift algorithm. The method preferably operates on a data set having 2N frames, and the iteration is performed N−1 times. The resultant wavelet coefficients are then linearly averaged and an inverse discrete wavelet transform is performed to arrive at the de-noised original signal. The method is preferably carried out in a digital processor.

The invention discloses a convolution network arithmetic unit, a reconfigurable convolution neural network processor and an image de-noising method of the reconfigurable convolution neural network processor. The reconfigurable convolution neural network processor comprises a bus interface, a preprocessing unit, a reconfigurable hardware controller, an SRAM, an SRAM control module, an input caching module, an output caching module, a memory, a data memory controller and the convolution network arithmetic unit. The processor is featured by few resources and rapid speed and can be applicable to common convolution neural network architecture. According to the unit, the processor and the method provided by the invention, convolution neural networks can be realized; the processing speed is rapid; transplanting is liable to be carried out; the resource consumption is little; an image or a video polluted by raindrops and dusts can be recovered; and raindrop and dust removing operations can be taken as preprocessing operations for providing help in follow-up image identification or classification.

A unmanned plane automation detection target and tracking method is disclosed. An unmanned plane, a sonar distance detector installed on the unmanned plane, an illumination compensation module, an image de-noising module, a face detection module, a fuselage face identification module, a far-end face identification module, a target tracking module, a flight control module and a console module are included. The illumination compensation module carries out illumination compensation on an image. The image de-noising module carries out de-noising processing on the image. The face detection module carries out face detection on the received image. The fuselage face identification module identifies the detected face image and adds an identification result of the far-end face identification module. The far-end face identification module identifies the face image which can not be processed by a fuselage. The target tracking module tracks a target. The flight control module controls a movement track of the unmanned plane. The console module is manually monitored and emits various kinds of commands.

The invention provides a method for carrying out shale gas reservoir evaluation and finding a dessert area for shale gas exploration and development comprehensively through comprehensive application of rock physics, log data and omnibearing or wide-azimuth three-dimensional seismic data. The method is characterized by analyzing the characteristics of the dessert area in a shale reservoir by applying comprehensive geophysical exploration technologies of laboratory rock geophysical measurement, log data comprehensive analysis, rock physical modeling, vertical seismic section data, omnibearing or wide-azimuth three-dimensional seismic data high precision surface layer comprehensive modeling static correction processing, prestack de-noising processing, amplitude relative fidelity processing, resolution enhancement processing which is carried out by utilizing well constraint and well seismic data to drive ground seismic data, fine excision, iteration velocity analysis, interval velocity field analysis, prestack depth migration processing, high resolution processing, azimuth classification processing, high-precision anisotropy processing, prestack and post-stack data inversion processing, and neural network analysis and the like, and thus gas bearing characteristic prospect of the shale gas reservoir is evaluated accurately and the dessert area for shale gas exploration and development is delineated.

The invention relates to a removing method for a glasses frame in a face image. The method comprises the steps of: detecting and positioning eyes in the input face image with a pair of glasses, geometrically correcting the face image on the basis of the positioned result to form a first image; extracting an eye estimation area from the first image to obtain a second image, and conducting mathematical morphology processing on the second image so as to determine the position of a glasses frame; based on the first image, establishing a face skin color model; in the second image, interpolating each pixel point of the glasses frame the position of which is determined, correcting the interpolation result in a part of pixel points of the glasses frame according to the skin color model, and conducting de-noising processing to form a ninth image; and replacing the eye estimation area in the first image by the ninth image, synthesizing and outputting the face image the glasses frame of which is removed. According to the technical scheme, the glasses frame in the face image with a pair of glasses can be conveniently and effectively removed, and the face recognition rate can be further improved.

The invention discloses a laser three-dimensional imaging device based on a single-photon detector, belonging to the technical field of photoelectric instruments. A target to be detected is irradiated by the laser pulse emitted by a pulsed laser via a scanning system; the returning photons are received by a receiving / emitting co-axial optical system, i.e., the returning photons are received by a double-gating single-photon detecting module via a spectral filter and a spatial filter and an arriving pulse is outputted, so that the photon flight time of the target measuring point can be measured by combining the laser emission detection and the multi-photon arriving pulse time; and a data processing unit is used for carrying out the coordinate conversion based on the position and attitude data, scanning mirror targeting data, and photon flight time of the three-dimensional imaging device, de-noising and three-dimensional image construction and correction, so as to output the reliable target three-dimensional range image. The invention solves the problems that the existing laser three-dimensional imaging device is incapable of penetrating vegetation and camouflage and being miniaturized when conducting long-distance operations.

The invention provides a binocular camera-based panoramic image splicing method. According to the method, a binocular camera is arranged at a certain point of view in the space, the binocular camera completes photographing for once and obtains two fisheye images; a traditional algorithm is improved according to the defect of insufficient distortion correction capacity of a latitude-longitude correction method in a horizontal direction; corrected images are projected into the same coordinate system through using a spherical surface orthographic projection method, so that the fast correction of the fisheye images can be realized; feature points in an overlapping area of the two projected images are extracted based on an SIFT feature point detection method; the search strategy of a K-D tree is adopted to search Euclidean nearest neighbor distances of the feature points, so that feature point matching can be performed; an RANSAC (random sample consensus) algorithm is used to perform de-noising on the feature points and eliminate mismatching points, so that image splicing can be completed; and a linear fusion method is adopted to fuse spliced images, and therefore, color and scene change bluntness in an image transition area can be avoided.

This invention relates to a method for de-noising digital radiographic images based upon a wavelet-domain Hidden Markov Tree (HMT) model. The method uses the Anscombe's transformation to adjust the original image to a Gaussian noise model. The image is then decomposed in different sub-bands of frequency and orientation responses using a dual-tree complex wavelet transform, and the HMT is used to model the marginal distribution of the wavelet coefficients. Two different methods were used to denoise the wavelet coefficients. Finally, the modified wavelet coefficients are transformed back into the original domain to get the de-noised image.

The invention relates to an expressway audio vehicle detection device and a method thereof. By the aid of the detection device, a microphone array signal acquisition module acquires an audio signal on a lane, the audio signal is subjected to band-splitting filtering and framing through a signal processing module after being subjected to de-noising processing through a noise suppression module, cross-correlation processing is conducted among sub-band signals, an audio space spectrogram is obtained, a vehicle detection module tracks a track of the maximum value on the audio space spectrogram and judges whether a vehicle passes, and the vehicle type and the vehicle speed are obtained through a vehicle type recognition module and a vehicle speed recognition module if the vehicle passes. The detection method is based on the device, a minimum statistical noise estimation method of the adaptive window length is adopted, the signal which is subjected to noise suppression processing is subjected to band-splitting filtering and framing processing, then cross-correlation processing is conducted among same sub-band signals, cross-correlation results are summed after being subjected to amplitude compression and are unfolded along a timer shaft, and an audio signal space-time spectrum is obtained. The method and the device have the advantages of being low in cost, low in energy consumption,easy to construct, interference resisting, capable of working in all weather and the like.

The present invention discloses a laser pen indicating and light spot trajectory identification method which belongs to the interactive reorganization technical field. Step 1, according to laser pen indicating information, a user obtains video information of a display screen from a video input device; step 2, an embedded device extracts light spots from the acquired video information, and determines the positions of the light spots after a de-noising process; step 3, the position coordinates of the light spots are transmitted to a computer which controls a target screen; step 4, according to the position coordinates of the light spots, the computer which controls the target screen recognizes the light spot trajectory; with the specific light spot information, the trajectory information indicated by the user is obtained; when the recognition is succeeded, step 5 is carried out; step 5, after the trajectory recognition is succeeded, the trajectory of the recognition result of the original trajectory is optimized, so that a new trajectory is obtained; the optimized result is displayed with visually optimized effect on the target screen to replace the indication effect of the step 3. The method can be applied in various interactive recognition display circumstances so as to carry out highly effective light spot analysis and trajectory recognition.

The invention discloses a vehicle ranging method based on monocular vision, and belongs to the field of target detection and ranging. The method comprises the steps of installing a monocular camera on a vehicle, measuring the height and the pitching angle of the camera, and determining focal distance parameters of the camera; acquiring a video image in an expressway environment; performing preliminary de-noising and filtration on the video image by adopting Gaussian filtration; performing interest region segmentation preprocessing before detection of a target vehicle on the video image; performing vehicle detection in a segmented video image region, wherein an Haar feature for increasing a wheel feature and a tail feature is adopted in the target vehicle detection, so as to effectively improve the accuracy of target vehicle identification; and measuring the distance of the target vehicle, wherein a ranging method based on pin-hole imaging is adopted within a short-distance range, while a ranging method of data fitting is adopted within a long-distance range, so that the error rate is reduced, and a real-time ranging effect can be achieved. The method has the advantages of high detection speed, high accuracy, strong real-time property and low cost.

Apparatuses and methods for extracting, de-noising, and analyzing electrocardiogram signals. Any of the apparatuses described herein may be implemented as a (or as part of a) computerized system. For example, described herein are apparatuses and methods of using them or performing the methods, for extracting and / or de-noising ECG signals from a starting signal. Also described herein are apparatuses and methods for analyzing an ECG signal, for example, to generate one or more indicators or markers of cardiac fitness, including in particular indicators of atrial fibrillation. Described herein are apparatuses and method for determining if a patient is experiencing a cardiac event, such as an arrhythmia.

The invention relates to a radar image de-noising method based on a GAN (Generative Adversarial Network). The method includes: the establishment of a radar simulation data set; the addition of the noise in a radar time frequency image; the establishment of the GAN, wherein the GAN is comprised of a generative network and discrimination network, a convolution nerve network structure is adopted by the two networks, a convolution nerve network is established as the generative network based on a residual error network, the front two convolution layers are designed as a convolution kernel with a step size of 2 so that the down-sampling is realized, the operand is reduced and the abstractness feature is extracted, the unit that keep the scale of the feature image in the residual error network isfollowed, and the pooling layer in the network is replaced by the convolution layer with invariant scale of feature image; a fully connected layer is deleted which is located after the convolution layer and used for the classification of the image, the up-sampling for the image is realized by utilizing two deconvolution layers, the output feature of the shallow convolution layers and the output feature of the deep convolution layers are added and further processed, the reuse of the feature is realized. The radar image de-noising method also includes training of the GAN.

The invention discloses a wind-power gear box fault diagnosis method based on wavelet medium-soft threshold and electronic-magnetic diaphragm (EMD), which includes: A. measuring a wind-power gear box by utilizing an acceleration sensor to obtain vibration accelerator signals; B. performing wavelet analysis on the signals, performing de-noising preprocessing through a wavelet medium-soft threshold method to eliminate interfering noise and obtain de-noising signals; C. performing EMD analysis on the de-noising signals to obtain each intrinsic mode function (IMF) component; D. judging true and false of the IMF components according to the related coefficients between the IMF components and original signals, rejecting false IMF components, and analyzing true IMF components to obtain components containing fault characteristics; and E. obtaining envelope spectrums containing fault characteristic components to draw fault characteristics. The wind-power gear box fault diagnosis method combines wavelet medium-soft threshold de-noising and the EMD and is effective in fault characteristic information drawing.

The invention discloses a method for identifying a disturbance event in a distributed type optical fiber pipeline security early-warning system. When the disturbance event exists, wavelet de-noising is conducted on two routes of sampling signals. The characteristic values of one sampling signal where wavelet de-noising is conducted are extracted, wherein the characteristic values include the vibration fragment length, the time domain energy, the k-order original point distance, the k-order center distance, the skewness, the kurtosis and low frequency wavelet coefficient energy Ej, obtained through wavelet decomposition, of all layers, and j ranges from 1 to 7. The thirteen extracted characteristic values are sent to a decision tree classification device, and the type of the disturbance event is obtained through the decision tree classification device. Man-machine interaction incremental learning is achieved by changing the type of the disturbance event stored in a database under the condition that a new type of the disturbance event appears or the type, obtained through the decision tree classification device, of the disturbance event is wrong, and online training is conducted on the decision tree classification device according to the modified type of the disturbance event. By means of the method, the type of the disturbance event can be accurately obtained.

The invention provides a rolling bearing fault diagnosis algorithm based on time-frequency domain multidimensional fault feature fusion. Aiming at the respective features of vibration signals of a rolling bearing in a normal state, a roller fault state, an inner ring fault state and an outer ring fault state in a time-frequency domain, through extraction of time domain and frequency domain features, redundancy removal and re-fusion, fault features are described in an optimal way to obtain an intelligent judgment result. First, wavelet de-noising is performed on extracted original rolling bearing vibration data; then, time domain feature vectors are extracted to form a time domain feature matrix, and coefficient energy moments after wavelet packet decomposition and reconstruction are extracted to form a frequency domain feature matrix; and the time and frequency domain matrixes are further fused to obtain a time-frequency domain multidimensional fault feature matrix. Redundancy of the multidimensional feature matrix is eliminated to obtain a new multidimensional feature matrix. Then, information of multidimensional features is fused with a weighted feature index distance, and a state judgment result of the rolling bearing is obtained through the feature index distance obtained through fusion.

The invention discloses an abnormal vibration analysis-based GIS (gas insulated switchgear) mechanical fault diagnosis method and system. The GIS (gas insulated switchgear) mechanical fault diagnosis method includes the following steps that: vibration signals on a GIS are acquired; de-noising processing is performed on the acquired signals through using a threshold de-noising method; various kinds of feature information contained in the signals are extracted by using a narrowband noise aided multivariate empirical mode decomposition (EMD) method; the power feature of the signals is extracted by using a power density function, and the maximal amplitudes of the power spectra of all IMF of each channel are calculated so as to form power feature matrixes under measured conditions; and a power feature matrix under a normal condition and / or power feature matrixes under various kinds of fault conditions which are obtained through tests are adopted as judgment criteria of faults. Compared with a traditional method, the method and system of the invention have no influence on the normal operation of a whole power system, and can monitor the running state of the GIS safely and reliably, and are suitable for being applied to GIS substations of various voltage levels.

The invention discloses a water supply pipe leakage detection locating signal processing method. The method includes: firstly determining the burst interference exists to the acquired information, if the burst interference exists, performing burst interference removing process, if the burst interference does not exist or the burst interference has been removed, performing noise suppression process to the signal; as a whole de-noising pretreatment process is completed, determining whether the leakage exists according to the processed signal, and adding a function of identifying the difference from the true leakage to the fixed interference sound source during the determination, if the leakage does not exist, directly giving out the detecting result, if it passes the identification and the true leakage exists, further performing leak point positioning, and giving out the final test results and the leakage point position. The invention has beneficial technical effect of improving signal quality by removing burst interference in the acquired signal and noise restraining and the like de-noising processes when performing leakage detection to the water supply pipe, thereby enhancing the pipeline leak detection accuracy and leak-point positioning accuracy under complex noise environments; having practical value that it firstly determines whether true leakage acoustic singles are included in the acquired signal and then locates the leakage points.

The invention provides the small-scale microphone array speech enhancement system and method. The system in the invention consists of two-part, signal acquisition and signal processing. Signal acquisition part comprises the microphone array and the AD converter component. Noisy voice signal is collected from the microphone as analog signals, through the AD converter, turning into digital signals, getting signal enhancement in the DSP chip, and then through USB port, transmitting to the computer for preservation. Signal processing part comprises the microphone consistency calibration module, the adaptive noise cancellation module, the SNR (Signal Noise Ratio) estimation module, and acoustic source location adaptive module, etc. The system in the invention can remove the introduced environmental noise in collection process, increase the collected acoustic source SNR, and obtain better acoustic source signal. In most cases the sound source signal is the speech signal, speech intelligibility improves a little after de-noising process.

The invention discloses a real time panorama video splicing method based on ORB characteristics. The real time panorama video splicing method based on the ORB characteristics comprises the following steps: acquisition of multipath synchronized video data is started; pretreatment is carried out on images in various paths at a same moment, and color images are changed into gray scale images of 256 levels, and a de-noising processing is carried out on the images through employing a Gaussian filter; the ORB feature extraction algorithm is employed to carry out feature point extraction on the images in the various paths at the same moment, and ORB characteristic vectors of the feature points are calculated; through the adoption of the nearest neighborhood matching method and the RANSAC (random sample consensus) matching method to determine a homography matrix array between corresponding frames of the synchronized videos; frame scene splicing is carried out according to the homography matrix array; and finally spliced videos are output. The real time panorama video splicing method based on ORB characteristics and the apparatus are advantageous in that: the feature extraction speed and the coupling effect are improved in the image splicing process.

An autonomous obstacle-avoiding planning method of a tour detector based on a binocular stereo vision is characterized by: generating DEM data of a local topography; constructing a raster map of a field range of a binocular stereo vision system, carrying out DEM data statistics to each grid in the map and carrying out de-noising processing to the DEM; carrying out topography traversability analysis so as to generate a suitability map; using the suitability maps generated during back and forth two planning to carry out data fusion and obtaining the fused suitability map which is used in the obstacle-avoiding planning; placing several alternative path arcs on the suitability map; evaluating the each alternative path arc from two aspects of a distance from the alternative path arc to a target point and a safe obstacle-avoiding ability; selecting the alternative path arcs whose evaluation values are higher a setting threshold from all the alternative path arcs and selecting the path arc who has the highest evaluation value as a result of the obstacle-avoiding planning. The method of the invention is simple and complete. Security is high. Planning efficiency is high. Engineering realization is easy to be achieved. The method can be used to perform a tour detection task of a complex terrain.

The invention discloses a wavelet difference algorithm-based cable fault localization method. The method comprises the following steps of: 1, detecting a signal in real time and synchronously uploading the signal; 2, performing signal acquisition; 3, performing cable fault localization by steps of performing signal pre-treatment which comprises normalization treatment, de-noising treatment and fault signal modulus transformation, judging a wavelet packet entropy fault region, determining the fault point by selecting a one-dimensional signal, performing wavelet transformation, solving a wavelet coefficient, detecting a modulus maximum point, sorting singular points after the maximum point, performing first-order difference operation and difference comparison, and performing localization operation on the cable fault position by using a single-end fault localization method free from influence of the wave velocity; and 4, synchronously outputting the treatment result. The method has the advantages of reasonable design, simple operation, convenient implementation and high localization precision, and effectively solves the practical problems that the fault region cannot be identified and the localization precision is relatively low and easily influenced by various factors in the conventional cable fault detection method.

The invention discloses an image de-noising method based on a convolution pair neural network, comprising the following steps: building a convolution pair neural network model, wherein the convolution pair neural network model includes multiple convolution pairs and corresponding activation layers; selecting a training set, and setting the training parameters of the convolution pair neural network model; according to the convolution pair neural network model and the training parameters thereof, training the convolution pair neural network model with the goal of loss function minimizing to form an image de-noising neural network model; and inputting a to-be-processed image to the image de-noising neural network model, and outputting a de-noised image. Through the image de-noising method based on a convolution pair neural network disclosed by the invention, the learning ability of the neural network is enhanced greatly, accurate mapping from noisy images to clean images is established, and real-time de-noising is realized.

The invention discloses a method for acquiring and predicting road traffic flow. According to the method, an annular sensing coil, a vehicle detection module, a traffic flow acquisition module and traffic flow data pre-processing and prediction software are adopted. The road traffic flow data pre-processing and prediction software runs in an upper computer (a personal computer (PC)) and is used for reading traffic flow data in the traffic flow acquisition module (a security digital (SD) card) through a network interface. In order to improve the reliability of prediction, a method for pre-processing and predicting the road traffic flow data comprises the following steps of: firstly, de-noising the traffic flow data by using a wavelet analysis method and a least square method; and secondly, establishing a traffic flow prediction model by using an improved back propagation (BP) neural network, so that prediction of the traffic flow is realized, and reference is provided for a control timing scheme for road traffic optimization and road traffic planning. The method has the advantages that: road traffic parameters such as vehicle flow, average vehicle speed, occupancy and traffic density in a rated period can be obtained, so that the prediction of the road traffic flow is realized, and the accuracy of data acquisition and the accuracy of prediction of the road traffic flow are improved.

Acquisition of data by managing crosstalk interference with sector designs and unique sweeps is conducted and the resultant data are processed in 3D common receiver domain to attenuate crosstalk noise while preserving the signals for high source and receiver density acquisition designs. High-amplitude spectral amplitudes are attenuated and inter-ensemble statics or structural time delays are applied to achieve optimum filter performance. If the spectral amplitudes have been attenuated to a level consistent with non-simultaneous acquisition, conventional surface consistent processing can be performed to correct for statics and amplitude variations. A 3-point filter in different frequency bands may then be applied to remove any remaining residual crosstalk noise.

The invention discloses a method for automatically monitoring water quality indexes on line and a device for implementing the method. A method of combining an ultraviolet and visible spectroscopy and various sensors is adopted, dozens of water quality indexes including chemical oxygen demand and ammonia nitrogen can be measured at one time, measurement indexes can be configured in a building block mode according to requirements, and chemical agents are not required. According to the device, the constructed digital optical fiber spectrometer is taken as a core, ultraviolet and visible absorption spectrum data of a water sample is processed in a mode of sequentially combining wavelet de-noising, principal component analysis and a support vector machine, and water quality indexes such as chemical oxygen demand and biochemical oxygen demand of water are acquired. Various physical and electrochemical sensors acquire water quality indexes such as ammonia nitrogen, dissolved oxygen and conductivity. All hardware and software for implementing the method is put in a cabinet to form the device, and the device analyzes the introduced water sample under the control of an embedded industrial control computer system, and automatically monitors the water quality indexes in real time.

The invention discloses a method for detecting changes of SAR (synthetic aperture radar) images on the basis of multi-scale product and principal component analysis ( PCA ), mainly solving the problems that the adaptability is poor, the application range is narrow and the change detection results are subject to image misregistration. The method comprises the following specific implementation procedures: firstly, conducting the logarithmic ratio operation on two inputted time phase SAR images to obtain a difference image; carrying out the wavelet transform on the difference image; carrying out the multi-scale product de-noising on the high-frequency information of each decomposition layer; then, combining the de-noised images of each layer and carrying out the PCA transform, wherein, a first PCA image is used as a new difference image; and finally classifying the new difference image by using the minimum error ratio threshold value of the generalized Gaussian model to obtain the final result image of changes. The experiment shows that the invention can enhance the change information, have strong antinoise performance and reduce the influence of image misregistration, thus having high applicability and can be applied to the disaster detection of SAR images.