68 results about "Frequency weighting" patented technology

The invention provides a text emotion classifying method in the stock field, belonging to the technical field of stock tendentiousness analysis. The text emotion classifying method is characterized in that feature selection is carried out on an enlarged stock emotion word through public news information comprising stock news and the like and by using an improved evaluation group; feature weighting selection is carried out on emotion word in the stock Chinese text by using absolute word frequency weighting; and finally, tendentiousness analysis is carried out on stock news by using a Bayes, K-NN or SVM (support vector machine) text emotion classifying algorithm. The method provided by the invention has the advantages of simplicity and feasibility, and is convenient for calculation.

Method for reconstructing subsurface Q models (110) from seismic data (10) by performing ray-based (60), centroid frequency shift (50) Q tomography. The seismic source waveform's amplitude spectrum is approximated by a frequency-weighted exponential function of frequency (40), having two parameters to adjust to fit the frequency shift data, thereby providing a better fit to various asymmetric source amplitude spectra. Box constraints may be used in the optimization routine, and a multi-index active-set method used in velocity tomography is a preferred technique for implementing the box constraints (100).

Described is a miniature sound level dosimeter that is less expensive and more user-friendly than conventional sound level dosimeters. The dosimeter includes a frequency-weighting circuit, an envelope follower circuit, an analog-to-digital converter and a microcontroller. The frequency-weighting circuit applies a predetermined frequency-weighting, such as A-weighting, to an electrical signal indicative of the detected sound level to generate a frequency-weighted signal. An analog envelope signal is generated in response to the frequency-weighted signal and converted to a digital output signal. The microcontroller determines the sound dose in response to the digital output signal and a time interval.

Disclosed is a method and system for discovering documents using a computer and providing a small set of the most relevant documents to the attention of a human observer. Using the method, the computer obtains a seed document from the user and generates a seed document vector using term frequency-inverse corpus frequency weighting. A keyword index for a plurality of source documents can be compared with the weighted terms of the seed document vector. The comparison is then filtered to reduce the number of documents, which define an initial subset of the source documents. Initial subset vectors are generated and compared to the seed document vector to obtain a similarity value for each comparison. Based on the similarity value, the method then recommends one or more of the source documents.

The invention discloses an absorbed blasting vibration energy comfort evaluation method of a multistorey building. The absorbed blasting vibration energy comfort evaluation method comprises the following steps: establishing an absorbed blasting vibration energy index, evaluating human body vibration characteristics by energy transfer and conversion, and analyzing blasting vibration energy frequency band representation by a power spectrum density value to obtain blasting vibration adsorption energy values in different frequency band ranges; analyzing the dynamic response characteristics of a building; obtaining the displacement amplification coefficient of a structural body; according to influence on the building by blasting vibration energy composition and different pieces of frequency band energy, carrying out frequency weighting on the vibration energy, and calling an energy weighting value as the absorbed blasting vibration energy of building frequency weighting; and multiplying each piece of frequency band energy by a human body vibration response weighting coefficient to obtain the absorbed blasting vibration energy of human body frequency weighting. By use of the absorbed blasting vibration energy comfort evaluation method, the magnitude and the frequency composition of total blasting vibration energy received by the human bodies in different floors of the building can be accurately calculated so as to achieve a purpose that blasting vibration energy comfort is accurately and objectively evaluated.

A method for performing time and frequency Signal-to-Noise Ratio (SNR) dependent weighting in speech recognition is described that includes for each period t estimating the SNR to get time and frequency SNR information ηt,f; calculating the time and frequency weighting to get γtf; performing the back and forth weighted time varying DCT transformation matrix computation MGtM−1 to get Tt; providing the transformation matrix computation Tt and the original MFCC feature ot that contains the information about the SNR to a recognizer including the Viterbi decoding; and performing weighted Viterbi recognition bj(ot).

A procedure for synthesizing a state-feedback gain matrix for a vehicle suspension system including active suspension components such as continuously variable semi-active dampers is disclosed. Sensors and / or estimation schemes provide feedback to the controller concerning the vehicle states. A set of frequency-weighted metrics are first quantified and used as part of a full car 7 degree of freedom vehicle model to construct a constrained multi-objective optimization problem. Using commercially available software, a mixed H2 / H∞ problem is iteratively solved to minimize a set of body control objectives subject to a set of physical control and wheel control related constraints to obtain data, preferably in the form of a plot of the trade-off curve between optimum wheel control and optimum body control. An initial design point is selected from the trade-off curve to calculate a state-feedback gain matrix that provides a reasonable balance between body and wheel control objectives. Additional points may be selected from the trade-off curve to iteratively provide an optimal solution.

The design of periodic screens for halftone patterns is based upon a series of constrained optimization problems. For each of a number of input levels, a pattern is found that minimizes a cost function. The cost function is a spatial-frequency weighted difference between an ideal, continuous-space pattern and a predicted reconstructed pattern. The frequency weighting is a parameterized trade-off between the preservation of periodic components and the introduction of non-harmonic components.

The invention relates to a vehicle noise sound quality roughness information processing method which comprises the following steps: (1), segmenting noise signals and windowing; (2), carrying out Fourier transformation to the noise signals after windowing, calculating out the exciting sound level of noise signals on a critical band; (3), designing mutually superimposed passages on the critical band, dividing the exciting sound level onto the passages so as to form sub noise signals, then carrying out inverse Fourier transform to the noise signals, so as to obtain time domain sub signals at the exciting sound level; (4), carrying out inverse Fourier transform after demodulation to the time domain sub signals at the exciting sound level, so as to obtain a signal time domain value after modulation, and then calculating out the modulation indexes of each passage; (5), calculating out the related coefficient and the carrier frequency weighting factor among the noise signals of all the passages; (6), calculating out the roughness of the noise signals on each passage and calculating out the total roughness. Compared with the prior art, the vehicle noise sound quality roughness information processing method provided by the invention takes auditory perception characteristics of human ears into consideration, and has the advantages of easiness in method understanding, high precision and the like.

The invention discloses a full-digital real-time noise measurement and analysis method and a system thereof. The noise measurement and analysis system provided with a noise measurement and analysis software PDA is connected with a noise collection system by a USB port. The method and system realize the digitization of frequency weighting, time weighting, octave and 1 / 3 octave processing, and can display time domain waveform and frequency domain waveform and provide system settings; and a user can select settings of sampling frequency and sampling time of the noise collection system, select L linearity, A and C types of frequency weighting and select time average, fast gear and slow gear of time weighting. In a graphic display mode, the user selects a frequency domain analysis method of signals and displays a corresponding frequency spectrogram according to needs, wherein the functions of FFT frequency spectrogram analysis, octave frequency spectrum analysis, 1 / 3 octave frequency spectrogram analysis and the like realize the measurement and analysis to real-time noise. The method and system utilize a PDA computer system to realize the full-digital real-time noise signal measurement and analysis with low cost and high quality.

The present invention relates to a method for speech recognition of a speech signal comprising the steps of providing at least one codebook comprising codebook entries, in particular, multivariate Gaussians of feature vectors, that are frequency weighted such that higher weights are assigned to entries corresponding to frequencies below a predetermined level than to entries corresponding to frequencies above the predetermined level and processing the speech signal for speech recognition comprising extracting at least one feature vector from the speech signal and matching the feature vector with the entries of the codebook.

Approaches are described for processing half-scan or full-scan cone beam image data using one or more half-ramp filtering operations. In one embodiment, the half-ramp filtering operations allow extraction and use of missing frequency data so as to generate a reconstructed image that is relatively complete in terms of frequency data and which has suitable temporal resolution. In addition, in certain embodiments, the reconstructed image may have uniform frequency weighting.

The invention provides a controlled source audio-frequency magnetotellurics one-dimensional inversion method using an improved genetic algorithm. By applying the genetic algorithm to CSAMT one-dimensional data inversion, a target function based on the combination of frequency weighting and phase weighting is proposed, a selection operator based on the combination of fitness value ranking selection and the simulated annealing method is proposed, segmented crossover probability upper and lower limits are proposed, and a crossover operator based on the combination of the self-adaptive crossover probability based on fitness values and a mutation operator based on the combination of fitness and gene weighting are proposed; by using the probability-based optimal individual preservation strategy, the individual with the smallest fitness value is replaced by the individual with the largest fitness value with a certain probability. Compared with the conventional CSAMT data inversion method, the method prevents inversion failure caused by the emergence of ill-conditioned matrixes, can avoid the problems of prematurity and the condition of being liable to be caught in a local extremum in the standard genetic algorithm, and enables chromosomes to move in the optimal solution direction rapidly to find the optimal solution.

A method and apparatus for obtaining segmented images of the stained regions may comprise quantifying the extent of the presence of staining of a biomarker in an original image of a sample, which may comprise selecting a domain swatch of data based upon a user specified domain knowledge; clustering the data within the original image by conducting a frequency weighted mean shift of the data within the original image to convergence, forming a hierarchical plurality of layers each having a different data resolution to form a hierarchical data pyramid; segmenting the plurality of mean shifted data images to determine in each mean shifted data image within the hierarchical data pyramid data not excluded as outside of the swatch; mapping the data not excluded as outside the swatch spatially back to the original image to create a final image; and, storing the final image on a storage medium for further analysis.

The invention discloses a digitized acoustic detection system, which comprises a microphone, a front end circuit, an analog-to-digital (A / D) converter, a signal processor, a touch screen and a keyboard. After the microphone converts an acoustic signal into an electric signal, the signal is input into the front end circuit. After the front end circuit amplifies the signal, the electric signal is input into the A / D converter for analog-to-digital conversion. The converted signal is input into the signal processor for signal processing. The touch screen and the keyboard are connected with the signal processor. A calibration module, a sound level measuring module, a spectrum analysis module and a fast Fourier transform (FFT) analysis module are arranged in the signal processor, wherein the calibration module is used for channel calibration and microphone calibration; the sound level measuring module supports A, C and Z frequency weighting; the spectrum analysis module is used for conducting real-time measurement by using 1 / 1 and 1 / 3 octave frequency bands within an allowable wide frequency range; and the FFT analysis module is used for measuring the transient sound level, the maximum sound level and the minimum sound level of the selected frequency range. The digitized acoustic detection system has the advantages that the basic detection of various sound levels and sound intensity calculation can be realized, the real-time spectrum and statistical analysis can be conducted to sound vibration signals, the sound vibration signals can be automatically monitored, the data redundancy of low-frequency parts is reduced and the spectrum leakage is reduced through windowing.

The invention relates to a method for detecting oil and gas by the aid of stacked seismic data. The method includes analyzing frequency spectra of objective intervals of seismic traces beside wells and determining parameters of sliding time windows; dividing high-frequency and low-frequency portions of the frequency spectra by the aid of weighted average frequencies, and solving frequency spectral energy of the low-frequency portions by the aid of a reverse frequency weighting process; acquiring energy coefficients by the aid of ratios of the energy of the low-frequency portions to energy of the high-frequency portions of the frequency spectra; creating seismic wave attenuation factors for indicating the oil and gas containing probabilities of strata by the aid of the energy coefficients and weighting frequencies; acquiring data volumes of the seismic wave attenuation factors of work areas and identifying beneficial oil and gas containing work areas in the work areas. The method has the advantages that low-frequency enhancement and high-frequency attenuation information in the frequency spectra of the seismic data can be sufficiently extracted and utilized, accordingly, the method is favorable for highlighting existence of oil and gas layers, the oil and gas layer identification ability can be enhanced, and the oil and gas detection sensitivity can be improved.

The invention relates to a sentence semantic distance measurement method. The method comprises the following steps: firstly, carrying out word segmentation and stop word removal preprocessing on a sentence data set; selecting a word meaning similarity scheme, and setting a threshold value to execute normalization of synonymous words and synonymous words; then, calculating the vector space distanceof the two statements by combining smooth inverse frequency weighting and common component removal; measuring the word order distance of the two statements according to the out-of-order degree; calculating the semantic dependency distance of the two statements by combining the semantic dependency quintuple features; and finally, carrying out hybrid weighting calculation on the vector space distance, the word order distance and the semantic dependency distance. Measurement is from three dimensions of sentence vector representation, sentence word sequence and sentence component dependency, andfinally a final semantic distance is obtained in a weighted summation manner. A word level calculation means is utilized, and a sentence level operation idea is absorbed, and through introduction andcreative combination of a vector space distance, a word order distance and a semantic dependency distance, the semantic distance of the sentences is more comprehensively and reasonably measured.

The invention discloses a soft glance microblog multi-emotion dictionary expanding method based on semanteme. The soft glance microblog multi-emotion dictionary expanding method based on semanteme comprises the following steps: establishing a candidate seed dictionary; filtering candidate seed emotion words by word frequency weighting and entropy weighting; acquiring candidate emotion words by word 2vec algorithm, and verifying by a statistical approach; and supplementing an emotion dictionary by a rule-based method. By the method, the multi-emotion dictionary is expanded effectively, and theproblem of imbalance of the number of the emotion words in the multi-emotion dictionary is solved.

The invention relates to the technical field of loudspeakers and provides a method and a system for audio fidelity of a loudspeaker. The method comprises the following steps of monitoring a working process of the loudspeaker to obtain working condition parameters of the current loudspeaker, wherein the working condition parameters include loudspeaker temperature parameters and loudspeaker resistance parameters; calling frequency compensation values corresponding to the working condition parameters of the current loudspeaker according to a corresponding relation between the pre-generated working condition parameters of the loudspeaker and the frequency compensation values; and adopting the called frequency compensation values to carry out frequency weighted compensation on a power amplifier connected to the loudspeaker. According to the method and the system, a problem that one audio file has different playing effects at different time periods is solved, the tone quality of the loudspeaker is promoted and the high-quality audio-visual experience is provided for a user.

The invention discloses a method and system for evaluating the vibration comfort of transient violent vibration of a train body. The method comprises the steps that a train body roll angle and the transverse displacement of a gravity center position in the train operation process are measured in a strong wind environment; frequency weighted filtering is carried out on the train body roll angle thetransverse displacement of a gravity center position; the train body roll angle obtained after weighted filtering and the transverse displacement of the gravity center position after weighted filtering are cut off according to a period, and the absolute value and peak-to-peak value of the average value of the train body roll angle the transverse displacement of the gravity center position are solved according to intervals; according to the absolute value and peak-to-peak value of the average value of the train body roll angle the transverse displacement of the gravity center position, evaluation indexes of the vibration comfort of a train are obtained through calculation. According to the method, the vibration comfort condition of the transient violent vibration of the train body in the train operation process can be more accurately reflected, and the actual feeling of a driver and passengers of the train can be reflected more accurately.

The invention discloses a wave beam forming method. The wave beam forming method comprises steps that a downlink pilot frequency symbol sequence and an MUB codebook are acquired by the signal emission equipment, the downlink pilot frequency symbol sequence is weighted according to the MUB codebook to acquire a downlink pilot frequency symbol set, the downlink pilot frequency symbol set is sent on all antennas in a pilot frequency weighting period, a pre-coding matrix indication set sent by the signal reception equipment is received, a reconstruction channel characteristic vector is calculated according to the pre-coding matrix indication set, wave beam forming of the downlink data is carried out according to the reconstruction channel characteristic vector, precision of directive wave beams can be improved, and wave beam leakage among users can be reduced. The invention further provides the signal emission equipment and the signal reception equipment capable of realizing the wave beam forming method.

Method for reconstructing subsurface Q models (110) from seismic data (10) by performing ray-based (60), centroid frequency shift (50) Q tomography. The seismic source waveform's amplitude spectrum is approximated by a frequency-weighted exponential function of frequency (40), having two parameters to adjust to fit the frequency shift data, thereby providing a better fit to various asymmetric source amplitude spectra. Box constraints may be used in the optimization routine, and a multi-index active-set method used in velocity tomography is a preferred technique for implementing the box constraints (100).

The invention discloses a single noise exposure protective device which comprises a digital microphone, a demodulation module, a frequency weighting module, a sound intensity calculating module, a noise dose accumulating module and an alarm output module. The digital microphone collects environmental noise and converts the noise collected into digital pulse signals. The demodulation module receives the digital pulse signals transmitted from the digital microphone, demodulates the digital pulse signals and converts the digital pulse signals into noise data. The frequency weighting module receives the noise data transmitted and handles the noise data in a frequency weighting mode. The sound intensity calculating module receives data signals transmitted from the frequency weighting module and obtains noise sound intensity after calculation and process. The noise dose accumulating module receives results of the sound intensity calculating module, conducts time accumulation and calculation for sound intensity in a working time range of the device, obtains total noise dose, and transmits the total noise dose to the alarm output module. The single noise exposure protective device is simple in debugging, stable in properties and suitable for being used by site workers.

The invention discloses a method for improving the vector distance classifying quality. The method comprises the following steps that the characteristic weighting process based on VSM is carried out, wherein the characteristic weighting process comprises the characteristic item word frequency weighting based on the word meaning and the characteristic item frequency weighting based on a document structure; word stems of English searching are extracted; a user query log is analyzed; a training corpus is corrected and expanded. The main defect of VSM formed by contradiction between the characteristic entry mutual independence requirement and the natural language diversity is overcome by constructing a segmentation dictionary, the classifying quality and the algorithm efficiency are improved while the classifying algorithm processing is simplified through the word stem processing technology, the query requirement in which a user takes great interest is obtained by analyzing the query log of the user and is used for guiding and amending the terminological dictionary, then the corpus can be dynamically changed along with the improvement of various special technologies through the amending and expanding of the training corpus, and a Robot program is guided to collect the latest special technical data.

The invention discloses a modular multilevel converter for high-voltage DC power transmission and a control method thereof. The modular multilevel converter includes three phase units. Each phase unit includes an upper bridge arm and a lower bridge arm which are symmetrical. Each bridge arm includes N sub modules and a reactor connected in series with the sub modules. The sub modules are double-clamped sub modules. According to the modular multilevel converter for high-voltage DC power transmission and the control method thereof, the MMC can be equivalent to two separate sub topological structures to achieve working states such as double switch-on, double switch-off, single switch-on single switch-off, and single switch-off single switch-on, and the sub modules can be sorted through a voltage-frequency weighting method to make the switching frequency of the IGBTs of the sub modules uniform.

The invention discloses a method for locating a mobile user based on cell frequency weighting in a period of time. The method comprises the steps of 1, carrying out data analysis through parameters such as service cell IDs and adjacent cell IDni in a wireless measurement report in a special period of time; 2, calculating cell locations according to parameters such as base station locations, cell antenna direction angles and tower height; and 3, training the special period of time by employing a machine learning method. According to the method, development of a light location platform is realized; the construction investment of the location platform is reduced; the problem that the location precision of a triangular location mode is poor is solved; and the method can be applied to acquisition of a user location attribute in a mobile environment, description of user tracks through assistance of time stamps and analysis of passenger flow volume.

Disclosed is a method and system for discovering documents using a computer and providing a small set of the most relevant documents to the attention of a human observer. Using the method, the computer obtains a seed document from the user and generates a seed document vector using term frequency-inverse corpus frequency weighting. A keyword index for a plurality of source documents can be compared with the weighted terms of the seed document vector. The comparison is then filtered to reduce the number of documents, which define an initial subset of the source documents. Initial subset vectors are generated and compared to the seed document vector to obtain a similarity value for each comparison. Based on the similarity value, the method then recommends one or more of the source documents.

The invention provides a frequency measuring method and a system of a distribution type power supply environment. The method comprises steps of sampling voltage of a power grid based on preset standard frequency; respectively obtaining voltage phase positions sampled at the 1st time and voltage phase positions sampled at the 1-1 time by use of complex value wavelet transform; and calculating voltage frequency according to the voltage phase positions sampled at the 1st time, the voltage phase positions sampled at the 1-1 time and a voltage frequency weighting calculation formula. In this way, the voltage phase positions of sampling voltage can be precisely obtained by use of the complex value wavelet transform; distributed power supply harmonic interference is reduced; precision of phase position information is greatly improved; and by performing voltage frequency weighting calculation, rapid and precise measurement of frequency of the power grid can be achieved.

The invention relates to a non-intrusive ultrasonic detection method of mud flow rate in a deepwater drilling riser. The method comprises the steps that 1) an emission sensor is close to the wall of a riser, and a receiving sensor is arranged on an ultrasonic wave propagation path in the direction vertical to the wall of the riser; 2) the actual work signal frequency and the echo signal intermediate frequency shift information of the emission sensor are acquired; 3) difference frequency signals are acquired, and the difference frequency signal frequency is calculated; 4) the difference frequency signal of the theoretical difference frequency signal frequency maximum is rejected; 5) time frequency conversion is carried out on the difference frequency signals to acquire the frequency domain information of amplitude-frequency or power-frequency; 6) difference frequency signals with amplitude value less than 2% of the maximum amplitude value or with power value less than 2% of the maximum power value are removed, and weighted integral is carried out on remaining frequency to acquire a frequency difference result; 7) a curve of the relationship between the mud flow rate and the frequency difference result is acquired; and 8) according to ultrasonic, the mud flow rate is calculated through the frequency difference result of the riser. The non-intrusive ultrasonic detection method provided by the invention can be widely used for the detection of the mud flow rate in the deepwater drilling riser.