46 results about "Value noise" patented technology

The invention discloses an OFDM channel estimation method for de-noising which is based on wavelet unbiased risk threshold value, which relates to the technical field of communication, the purpose of the invention is that the problem that an LS method is easy to be influenced by noise under the condition without channel information is effectively solved. The realizing process of the method comprises first, obtaining channel frequency response which contains noise according to the LS channel estimation method which is on the basis of an OFDM system which is based on pilot frequency, second, doing inverse discrete Fourier transform for the channel frequency response which is obtained and to get channel time-domain impulse response which contains noise, third, de-noising the channel time-domain impulse response the method for de-noising which is based on wavelet unbiased risk threshold value, and getting channel time-domain impulse response after de-noising, fourth, doing inverse discrete Fourier transform for the channel time-domain impulse response after de-noising, and getting frequency domain response of channel, and compensating the received signals by the channel time-domain impulse response after de-noising. The method can be used in OFDM channel estimation techniques under various multi-path channels.

The invention relates to an angle zone scanning angle value jump point suppression method based on pole number table checking, and a device. The method comprises the following steps: a single-pair pole hall sensor and a multi-pair pole hall sensor measure an angle value signal, an A/D converter performs analog-to-digital conversion for a hall signal, and a single-pair pole angle value theta 1 anda multi-pair pole angle value theta 2 are obtained through a calculation module; an angle valve extreme value zone tabulation module performs tabulation for the pole number Pcheck of the multi-pair poles and the number of zones Zcheck according to the single-pair pole angle value theta 1 and the multi-pair pole angle value theta, the number of the current multi-pair pole angle value zones is judged through a zone judging module; by a zone scanning comparison module, comparison and judgement are executed for the table checking zone Zcheck and the zone value Zjudge wherein the multi-pair pole angle value theta 2 is, and a pole number adjustment value Pmod is output according to a judgement result; and multi-pair pole angle value fine segmentation processing is executed according to the Pmodand the Pcheck, and a high-resolution angle value theta duo-org is output. With the method and the device provided by the invention, a multi-pair pole angle value jump point problem caused by the reasons, such as angle value noise and temperature variation, can be suppressed effectively, reliability and resolution of a multi-pair pole coder are improved.

The invention provides a satellite navigation and positioning observation value weighing method and device. A BDS satellite observation value and observation quantity residual after pre-adjustment areacquired, and a weight ratio of observation values of different track types of the BDS is determined; according to an original observation value, pseudo-range observation value noise of each satellite in the current epoch is solved in real time; and by using the observation value weight ratio and pseudo-range observation value variance, an observation value variance matrix is classified and solved in real time. In consideration of ionospheric disturbance, multipath effects and observation noise, an observation value with better quality is selected to participate in solution, the pseudo-rangeobservation value noise of each satellite in the current epoch is solved, convergence is accelerated, a real-time and reliable positioning result is obtained, and the solution precision is improved. In consideration of effective information in real-time data, the contribution of a different type of observation quantity in positioning and solution is clarified, the positioning stability is improved, the method is easy to extend, and better position reference is provided for GNSS real-time navigation positioning and atmospheric inversion application.

The invention discloses a data driving threshold value noise-reduction method for rotary machine vibration signals. According to the method, firstly, wavelet transformation is carried out on collected vibration acceleration signals, and the signals are decomposed into different frequency bands. Secondly, noise estimation is carried out on the signals at each frequency band, and data driving threshold values adapting to the signals are obtained. Thirdly, the signals are segmented through a sliding window technology. Finally, threshold value noise reduction is carried out on each segment of signals by utilizing the data driving threshold values, the signals are reconstructed, and time-domain signals after noise reduction are obtained. The data driving threshold values are from noise estimation of the signals, and the threshold values can be set in a self-adaptive mode according to the intensity of the noise. Compared with traditional threshold values, the data driving threshold values adapt to the signals, the threshold values are set more accurately, and weak fault signals are reserved while noise is reduced. The data driving threshold value noise-reduction method for the rotary machine vibration signals combines the advantages of wavelet transformation, the noise estimation algorithm, the sliding window technology and the 3sigma principle, can effectively extract failure characteristics, and achieves failure diagnosis of mechanical equipment.

The invention discloses a combination precise single-point positioning method and system based on inter-constellation difference. The combination precise single-point positioning method comprise the steps that original observation values and auxiliary parameters, observed on an observation station, of the different constellations are obtained; the original observation values are preprocessed; thepreprocessed original observation values are subjected to non-ionizing combination; inter-constellation difference is conducted on non-ionizing layer combination observation of a single constellation;an error observation equation is determined based on an observation model for inter-constellation difference; the preliminary position of the observation station and observation value noise are obtained; the elevating angles of all satellites are determined based on satellite orbit parameters and the preliminary position of the observation station; a random model of each set of inter-system difference observation values is determined on the basis of the elevating angles of all the satellites and the observation value noise; a weight matrix is determined; and position parameters, inter-constellation difference ambiguity parameters and troposphere parameters are estimated through a least square method on the basis of the error observation equation and the weight matrix, and thus single-point positioning is achieved. According to the combination precise single-point positioning method, high-precision observation station positioning information and inter-system time difference parameter information can be obtained.

The embodiment of the invention discloses a fingerprint identification method and device. The method comprises the following steps: continuously carrying out a fingerprint image collection on a target finger at least twice, and obtaining at least two copies of initial fingerprint image data of the target finger; and calculating the mean value of pixel values of corresponding pixel points in each copy of initial fingerprint image data, mapping the mean value into the corresponding pixel points to form target fingerprint image data, and carrying out fingerprint identification according to the target fingerprint image data. According to the fingerprint identification method and device, multiple copies of initial fingerprint image data are obtained; the mean values of the pixel values of the corresponding pixel points in multiple copies of initial fingerprint image data are calculated; and the mean values are mapped into the corresponding pixel points, so as to form the target fingerprint image data. According to the method, the fingerprint image identification rate is improved by a mean value noise reduction method; the problem of low fingerprint image identification rate caused by noise points generated when a fingerprint sensor collects the fingerprint image data of the target finger is solved; and the use experience of a user is improved.

The invention relates to a noise removing method for contact net insulator leaked currents. According to the method, an extreme value prolongation method is used for reducing end point effect, signals are resolved through an improved EMD method so as to obtain components of frequency from high to low, a power ratio method is used to remove false components, noise in high-frequency ingredients in the remaining components is removed in a small wave threshold mode, and high frequency signals after noise removing and the remaining components are reconstructed, so that signals after noise removing can be obtained. The EMD method has adaptivity compared with small wave conversion. The problem of base function selection does not need considering. Compared with the EMD method, the improved EMD threshold method reduces effect of end point effect and false components on noise removing effect. Further, threshold value noise removing is conducted on high-frequency components, useful signals can be maintained to a largest degree, and signal distortion is reduced.

The invention discloses a daily behavior recognition method based on myoelectric wavelet coherence and a support vector machine. Myoelectric signals of human body relevant muscles are collected through a myoelectric signal collection instrument; sample data of two paths of myoelectric signals is obtained; pretreatment is performed by using a modified wavelet threshold value noise reducing method;the wavelet coherence coefficient of the two paths of myoelectric signals is calculated. The obtained wavelet coherence coefficient is used as a feature vector to be input into the support vector machine for classification recognition; different daily behaviors are successfully recognized; higher recognition rate is realized. A method of combining the myoelectric features of the wavelet coherencewith the support vector machine is used; higher recognition rate and reliability is realized on the human body daily behavior recognition. Experiment results show that the method provided by the invention reaches the average sensitivity of 96.17 on going upstairs, going downstairs, standing, walking, running and falling; the average specificity degree reaches 92.29; the values are higher than those of a conventional method.

A close loop method for measuring head SNR, for a storage device comprising a storage media and a head, comprising steps of: (a) loading the head on the media with a dynamic fly height; (b) measuring an initial environmental temperature value Ti and measuring the head signal signal_load; (c) unloading the head; (d) adjusting a power which controls the dynamic fly height until a real-time environmental temperature value T2 is equal to the initial environmental temperature T1; (e) measuring the head noise value noise_unload, (f) calculating the head SNR with the follow equation:

The method of the present invention can obtain a fair condition between the signal and noise measurement, thereby a reliable and accurate head SNR can be obtain. The present invention also provides a close loop method for measuring media SNR.

The invention discloses a signal-dependent noise parameter estimation method based on improved density peak clustering. A sample is extracted from an image containing noise through a sliding window, and a mean value, entropy and gradient are calculated to serve as feature data. And inputting into a clustering algorithm for clustering, and distinguishing weak texture samples from strong texture samples. In the clustering process, the concept of relative density is introduced, comparison ranges are divided through the distance between data points and surrounding data, the relative density of the data points in each comparison range is calculated, and finally the data points with high relative density are selected as clustering centers. The problem that when a traditional DPC algorithm is used for clustering a data set with non-uniform density, a sparse cluster center is often neglected, so that the clustering precision is influenced is solved. And according to a clustering result, pixel level estimation and noise estimation are carried out on a sample whose cluster label is weak texture, and finally a pixel value-noise variance estimation pair is fitted through a least square method, so that a noise parameter estimation value of an original image is obtained, and the preparation work of denoising is realized.

The invention provides a nonlocal median value blind noise reduction method based on visual perception, comprising the following steps: constructing an impulse noise blind detector based on the visual outlier measure of the pixels in a digital image, wherein the visual outlier measure is obtained by quantizing the visual similarities in impulse noise of different models and fusing different visual feature quantization results; extracting the nonlocal information of the image, and constructing a nonlocal median value calculation model; calculating regularization parameters according to the visual outlier measure and the nonlocal information, and establishing a nonlocal median value regularization item; and building a nonlocal median value noise reduction functional model, and adaptively repairing the noise pixels in the image. According to the blind noise reduction method of the invention, impulse noise of different models and densities in the digital image is processed in a unified manner according to the visual features of impulse noise, image self-similarity and outlier data mining, and the problem that noise pixels are hard to repair effectively due to unknown impulse model, high-density noise and image multi-modal complexity in an actual noise reduction process is solved.

The invention belongs to the field of imaging automatic target recognition, and particularly discloses a fixed value noise determination method and a fixed value noise removal method for an area arraylaser detector, which comprises the following steps: acquiring multiple frames of images of a target through an area array laser; counting histograms of the multiple frames of images; and determininga distance value corresponding to the fixed-value noise through adjacent distance value difference calculation and threshold judgment based on a statistical histogram, and not counting data at the distance value in subsequent accumulated detection according to the fixed-value noise distance value, thereby obtaining a laser distance image and an intensity image of which the fixed-value noise is removed. According to the invention, through histogram statistics, a fixed value noise distance value in a laser image is highlighted; counts corresponding to adjacent distance values in a statistical histogram are subtracted and compared, the position of fixed value noise is located, the method can be effectively used for removing the fixed value noise and recovering part of target distance information at the fixed value noise, surface information is more complete, and therefore the imaging effect of the Gm-APD laser detector can be effectively improved.

The invention discloses a device and a method for detecting quality of a beef sauce seasoning liquid. The method comprises the following steps: detecting each specimen of qualified beef sauce seasoning liquids by a sensor array to obtain m signal-to-noise ratio spectrum characteristic value noise widths, and defining the average value of the signal-to-noise ratio spectrum characteristic value noise widths as a threshold value Thr; obtaining the signal-to-noise ratio spectrum characteristic value noise width sigma of a beef sauce seasoning liquid sample A; calculating a dynamic classification parameter delta of the beef sauce seasoning liquid sample A by using a nonlinear self-calibration dynamic classification model; and judging whether the beef sauce seasoning liquid sample A is qualified or unqualified by a computer. The device and the method have the characteristics of capabilities of detecting the quality of the beef sauce seasoning liquid accurately and providing reliable data support for a manufacturer to manufacture beef sauce with stable quality.

The invention discloses a device and a method for detecting varieties of beef sauce seasoning liquid. The method comprises the following steps: firstly, dividing n types of beef sauce seasoning liquids into m beef sauce seasoning liquid specimens with equal mass; secondly, detecting each specimen of each sample by a sensor array sequentially to obtain n threshold values of Thr1, Thr2, ..., Thrn which are related to the varieties of the beef sauce seasoning liquids; thirdly, detecting a beef sauce seasoning liquid W to be detected so as to obtain the signal-to-noise ratio spectrum characteristic value noise width sigma of the beef sauce seasoning liquid W; fourthly, calculating a dynamic classification parameter delta of the beef sauce seasoning liquid W by using a nonlinear self-calibration dynamic classification model; and finally, determining that the variety of the beef sauce seasoning liquid W is variety j by a computer. The device and the method have the characteristics of capability of quickly and accurately differentiating the varieties of the beef sauce seasoning liquids.

The invention relates to a two-third weighing noise reduction method and belongs to the technical field of image processing. The method comprises the steps that images are collected through a camera at first; the digitized images are binarized, and binarized images are obtained; appropriate windows are selected to carry out noise reduction processing on the binarized images, and noise reduction effect pictures are obtained; noise reduction effects are compared according to the combination of the noise reduction effect pictures and a grey level histogram, and the most appropriate window is found; the effect picture corresponding to the most appropriate window is binarized. The two-third weighing noise reduction method is simple, the data volume is reduced, the computing speed is high, parameters such as the neighborhood set value, the mid-value noise reduction window size and the number of images used for multi-picture average do not need to be tested and screened; in the binarized images, the method is large in noise reduction weighing coefficient, good in reliability and particularly applicable to the field of image processing for images with the requirement for only black and white colors.

The invention discloses a flow velocity validity identification and correction method for frequency domain space-time image velocity measurement, and the method comprises the following steps: 1, detecting a target signal peak value and a corresponding direction in an amplitude spectrum, searching a valley value and a direction from the peak value to a nearest boundary of the amplitude spectrum, taking an amplitude symmetric point of a valley value point relative to a signal peak value point, and calculating a peak value of the peak value; the main direction of the frequency spectrum is relocated after the peak symmetry is judged; 2, taking the valley value as background mean noise, calculating confidence, and calculating a starting point distance of a velocity measurement line and a flow velocity value on the velocity measurement line; 3, acquiring a confidence threshold according to an adaptive confidence mechanism to set a validity flag bit of a corresponding flow rate, and performing interpolation correction on an error vector according to the validity flag bit; and 4, calculating the effective data rate according to the effective flag bit. The effectiveness of the flow velocity measurement value can be effectively identified, random noise and gross errors caused by environmental disturbance under the field monitoring condition are eliminated, and the effective data rate is improved.