54 results about "Wavelet transform analysis" patented technology

In the present invention, a multi-person pose recognition system has been developed. This system includes a body pose detection module, a CC2420DBK board and a multi-person pose monitoring software module. The body pose detection module includes a triaxial accelerometer, a Zigbee chip and an 8-bit microcontroller. Several body pose detection modules and the CC2420DBK board form a Zigbee wireless sensor network (WSN). The CC2420DBK board functions as the receiver of the Zigbee WSN and communicates with a robot onboard computer or a host computer through a RS-232 port. The multi-person pose monitoring software monitors and records activities of multiple users simultaneously. The present invention provides a pose recognition algorithm by combining time-domain analysis and wavelet transform analysis. This algorithm has been implemented in the microcontroller of a body pose estimation module. Through the algorithm, the system can recognize seven body poses: falling, standing, sitting, lying, walking, going upstairs and going downstairs.

The invention discloses a video-analysis-based smoke detecting method, which is realized by taking a digital camera as a sensor and under support of a digital signal processing chip. The method is characterized by comprising the following steps: collecting digital video by utilizing the digital camera; screening out a foreground part containing a movement part; screening out an area with likeness by using a support-vector-machine detector; analyzing high-frequency signal change by applying wavelet transform, and screening out a digital image of which the background is in gradually fuzzy change; and screening out digital images with smoke texture characteristics by using an Adaboost cascade classifier pairs. By combining moving object extraction by background modeling, grey area screening by a support-vector-machine (SVM) classifier, high-frequency signal change analysis by wavelet transform, the smoke texture characteristic cascade classifier and other methods, the method greatly improves the accuracy and efficiency of smoke detection, reduces the false alarm rate and has high robustness and higher practicality.

An implantable heart-monitoring device designed to distinguish two differing heart rhythms. The device comprises electrodes, a template, and first, second and third electronic mechanisms. The first electronic mechanism converts electrical representations of heartbeats from the electrodes into digital data and performs calculations including at least a partial discrete wavelet transform upon the digital data to generate a subset of discrete wavelet transform components including components for distinguishing the two differing heart rhythms and also demonstrated to be relatively low in variability from one patient to another. The template contains a corresponding subset of discrete wavelet transform components captured from at least one individual whose heart was beating in accordance with one of the two differing heart rhythms. The second electronic mechanism correlates the subset of transform components against the template components and provides a correlation value. The third electronic mechanism mathematically examines a time series of the correlation values and gives an indication of whether the first or second of the heart rhythms is present.

A non-invasive diagnostic tool and method for detecting an obstruction in a coronary artery, the diagnostic tool including a signal processor adapted to receive signals corresponding to a heart beat from a plurality of acoustic sensors attached to the chest of a patient. The signal processor is programmed to identify a diastolic portion of the signals for a plurality of heartbeats and conduct a wavelet transform analysis on the diastolic signals for determining the existence of, as well as the severity and the location of, an obstruction in a coronary artery of the patient.

A transient noise removal system removes or dampens undesired transients from speech. When the transient noise removal system receives a speech frame, the system performs a wavelet transform analysis. The speech frame may be represented by one or more wavelet coefficients across one or more wavelet levels. For a given wavelet level, the transient noise-removal system may determine a wavelet threshold. The transient noise removal system may compare the threshold corresponding to a wavelet level to the wavelet coefficients within that level. The transient noise removal system may attenuate each wavelet coefficient based on a comparison to a threshold.

The invention provides a detecting system for an incompletely-blocked position of a pipeline. The detecting system comprises a pressure wave pulse transmitting device, a pressure sensor and a sensor data collecting system connected with the pressure sensor. The pressure wave pulse transmitting device generates pressure wave pulse signals so that the pressure wave pulse signals can be transmitted in the pipeline through media in the pipeline. The pressure sensor senses the pressure wave pulses signals, transmitted by the media in the pipeline, in the pipeline and outputs the pressure wave pulsesignals to the sensor data collecting system. The invention further provides a detecting method for the incompletely-blocked position of the pipeline. Wavelet transform is adopted for analyzing datacollected by the sensor data collecting system, and the incompletely-blocked position of the pipeline is calculated. By means of the detecting system and method for the incompletely-blocked position of the pipeline, the blocked position of a blockage can be effectively detected in the incompletely-blocked initial stage of the pipeline, measures are taken in time, and safety running of the pipelineis ensured.

The invention discloses a power cable field partial discharge signal noise reduction method. The method comprises the steps that S1 a field power cable partial discharge signal is collected, and digital processing is carried out to form the time domain signal f (t) of the field partial discharge signal; S2 the time domain signal f (t) is converted to a corresponding frequency domain function F (omega) through fast Fourier transform, and mathematical processing is carried out on the frequency domain function F (omega) to acquire a fast Fourier power spectrum described in the description; S3 anAGFCM algorithm is used to classify noisy signal power spectrum data into two categories; S4 narrow band interference peak is positioned; the positioned narrow band interference peak is compressed; inverse fast Fourier transform is carried out on P (omega) after narrow band interference peak filtering to acquire a narrow band noise filtered signal g (t); and S5 wavelet transform analysis is carried out on g (t); mother wavelet is selected; a threshold is determined; white noise is filtered out to acquire a final denoised signal. According to the invention, periodic narrow-band noise can be suppressed; the waveform distortion rate is reduced; and the signal-to-noise ratio of the denoising result is improved.

A method for extracting data from an overall signal, generated by a plurality of impulses, by use of a computer using wavelet transform analysis of timing and strength of each of said plurality of impulses is disclosed. The method includes using a Discrete Wavelet Transform analysis or a Continuous Wavelet Transform analysis.

The invention provides a strong interference signal detection method based on power spectrum and multiple dimensioned wavelet transformation analysis, comprising: a receiving terminal sampling received signals, calculating a sampled received signal power spectrum, and performing smooth filtering on the received signal power spectrum through a sliding filter; performing gauss curve fitting on the power spectrum after smooth filtering, and performing multiple dimensioned wavelet transformation on the fitted power spectrum to obtain each scale wavelet transformation coefficient; calculating a sum wavelet coefficient and a threshold thereof to obtain a new sum wavelet coefficient; employing an extreme point searching algorithm to search for a new sum wavelet coefficient maximum point; and verifying the accuracy of the searched maximum point. The method overcomes the problem of difficult threshold setting of a traditional threshold detection method, has the advantages of high detection precision, strong robustness, and convenient realization, and can be used for detection and identification of a plurality of interference signals in a communication system.

The invention discloses a method for detecting a dynamic activity rule of the overburden of a western shallow coal seam during mining, and the method comprises the steps: selecting a typical working face side of a western mine, and mastering the specific mining parameters of a working face; processing the earth surface radar satellite SAR image data of the whole mining process of the working face through employing a D-InSAR image high-precision registering algorithm under a low coherence condition, and obtaining the earth surface deformation condition of the working space in the mining process; detecting the dynamic developmental situation of mining-induced fractures in the underground working face overburden with radon; analyzing the dynamic coupling relation among the specific mining parameters of the working face, the mining earth surface deformation and the mining fracture development, wherein the dynamic coupling relation represents the dynamic activity rule of the overburden of the western shallow coal seam during mining. The method introduces a satellite remote sensing method and a radioactivity measurement method to the field of underground coal mining, and can reveal the dynamic activity characteristics of the overburden of the western shallow coal seam accurately, quickly and effectively during mining.

The invention discloses a blind watermark embedding and extracting method of a stereoscopic video image; the stereoscopic video image is divided into image blocks which are not overlapped mutually, discrete cosine transformation and two-level discrete wavelet transformation are carried out to all image blocks, the relation of numerical values of the discrete cosine transformation direct-current coefficients and two-level discrete wavelet transformation low-frequency coefficients of a left viewpoint view and a right viewpoint view of the stereoscopic video image is analyzed, so as to define the internal relationship of the image blocks and interblock relationship of two image blocks in the stereoscopic video image; a watermark embedding mode is determined though the relationships and a waythat whether the watermark information is consistent, wherein the data of the primary stereoscopic video image is not changed when relational embeddedness is carried out, the discrete cosine transformation direct-current coefficient of the image block with small correction magnetite of the discrete cosine transformation direct-current coefficient is corrected when quantization embeddedness is carried out, in this way, the quality of the stereoscopic video image is ensured, when watermark is extracted, no primary stereoscopic video image is needed, so as to realize blind detection.

The invention discloses a wavelet analysis-based grinding machining working condition detection system and a wavelet analysis-based grinding machining working condition detection method. The system comprises a sensor information acquisition module, a corresponding table module for working conditions and acoustic emission frequency band power, and an intelligent working condition judgment and output module. An acoustic emission sensor is arranged on a grinding machine and transmits acoustic emission signals to the corresponding table module for the working conditions and the acoustic emission frequency band power and the intelligent working condition judgment and output module respectively; the corresponding table module for the working conditions and the acoustic emission frequency band power acquires the standard power intensity of each frequency band by using a wavelet transformation analysis method, and constructs a corresponding table between standard working conditions and standard acoustic emission frequency band power intensity Pst; and the intelligent working condition judgment and output module performs wavelet transformation analysis on the acoustic emission signals to acquire current power intensity, matches the current power intensity with the Pst, calculates a fit error value and determines and outputs a current grinding working condition.

The invention relates to a non-woven fabric flaw detection and classification method and system based on machine vision. The method comprises the following steps: (1) image acquisition; (2) image preprocessing; (3) extracting detail signals of the image: analyzing and extracting the detail signals of the defect image of the non-woven fabric by adopting wavelet transform; (4) calculating a texture feature image and reconstructing a feature image: calculating texture information of detail signals of the image to obtain the feature image with the capability of distinguishing a normal image and a defect image of the non-woven fabric, and reconstructing the feature image to obtain the reconstructed feature image; (5) calculating an abnormal score: calculating the abnormal score of the non-woven fabric feature image so as to judge whether flaws appear or not; (6) differential calculation of a flaw region: performing differential calculation on the original to-be-detected image and the reconstructed feature image to obtain the flaw region, and performing post-processing on the flaw to obtain a complete flaw region; and (7) flaw region classification: constructing a feature vector and training a classifier, and classifying the flaw region.

Embodiments of the invention provide a border gateway protocol (BGP) stability exception detection method and device and an electronic device. The method comprises the steps of acquiring a pre-storedbase autonomous domain set; counting update packets corresponding to each first source autonomous domain included in the base autonomous domain set in different time periods; analyzing exceptions of each first source autonomous domain in different time periods through wavelet transform according to the quantity of update packets corresponding to the first source autonomous domain; and inputting exceptions corresponding to each first source autonomous domain into a pre-trained exception prediction model. Through adoption of the BGP stability exception detection method and device and the electronic device, calculation efficiency in the stability exception detection is improved, and fewer calculation resources are used.

The invention relates to a novel photon probability mapping method based on terrace forming, which belongs to the technical field of mapping method. The method comprises the following steps: matchingaccording to channel conditions and generating an optimal Gaussian beam by a laser; performing probability matching by a probability matcher, and changing the shape of the Gaussian beam by a spatial phase array to obtain a novel beam formed in terrace; performing the 64QAM projection mapping on the data by the novel beam; concentrating and converging the 64 constellation point optical signals; performing pre-emphasis; amplifying the optical signals after passing through a channel, detecting by a photodetector, performing low-pass filter, orthogonalizing and normalizing, and performing data convergence and equalization; matching probabilistic solutions, removing step noise by wavelet transform analysis, performing constellation demapping; estimating channel conditions according to the received data, feeding back channel state information to the probability matcher, and performing real-time dynamic regulation on the probability matcher. The novel photon probability mapping method based on the terrace forming can effectively reduce the signal transmission power, reduces the bit error rate, improves the frequency band utilization, and lengthens the transmission distance.

The invention provides a phase multi-path correction method based on a non-difference and non-combination PPP model. According to GNSS original observation values, a non-difference non-combination PPPobservation equation is constructed, phase residual errors are calculated through a sequential least square algorithm, and coordinates serve as constant estimation in order to prevent multipath errors from being absorbed by coordinate parameters. And a mathematical relation between phase residuals of different frequencies is established, spatial distribution conditions of multiple paths of different frequency bands are analyzed by utilizing continuous wavelet transform, and an equal-arc-length hemispherical grid model considering spatial distribution characteristics of the multiple paths is constructed according to the spatial distribution conditions. A user searches a multi-path grid model according to a satellite altitude angle and an azimuth angle to obtain a multi-path correction number of a certain frequency, and calculates multi-path correction numbers of other frequencies according to a mathematical relationship of residual errors among the frequencies to correct multiple pathsin a phase observation value. According to the method, the multipath error is processed based on the observation value domain, only the phase multipath model of one frequency needs to be constructed,and the complexity of PPP data processing is greatly reduced.

The invention provides an intelligent substation relay protection device fault diagnosis system and method, and relates to the field of electrical equipment and electrical engineering. The method comprises the steps that a data obtaining module obtains the state quantity data of a relay protection device; a database obtains state quantity data; a data processing module is used for filtering and deleting redundant data and invalid data in the acquired data; a state evaluation module analyzes and evaluates the state quantity data and finally obtains the health state grade of the device; a faultprediction module judges the state trend and the residual life of the device and performs fault prediction; and a starting element hidden fault analysis module is used for carrying out wavelet transform analysis on the obtained fault recording data to obtain a real fault node so as to judge the execution performance of the starting element, judge whether a hidden fault exists or not and judge whether a starting threshold needs to be modified or not. The problems of improper maintenance, excessive maintenance and the like of a relay protection device can be effectively avoided, the safety condition of the device is improved, and the service life of the device is prolonged.

The invention discloses a welding process quality evaluation method for shielded metal arc welding. Discrete welding voltage signals and discrete welding current signals are collected in a welding process, and low-frequency voltage signals and low-frequency current signals are obtained through wavelet transformation analysis. The welding process is divided into m equal time slots. The average value of the low-frequency voltage signals in each time slot is worked out and then the average value of m low-frequency current signal average values is worked out. A variation coefficient of the m low-frequency voltage signal average values is worked out and compared with a series of variation coefficients in a welding process judgment database, and a basic score of the welding process is worked out. Then, long short circuit time is adopted for correcting the basic score, and a final quantitative evaluation score of the welding process is obtained and serves as a basic criterion for reflecting operational capacity. The operating skills of welders can be judged objectively and standardly.

The invention discloses a homology cluster identification method based on wavelet transformation and an apparatus thereof. The method comprises the following steps of acquiring a power angle curve of each generator in an electric power system; carrying out wavelet transformation analysis on the power angle curve of each generator respectively; carrying out feature extraction on the power angle curve of each generator after the wavelet transformation analysis respectively to acquire a homology identification key characteristic of each generator; according to the homology identification key characteristic of each generator, carrying out cluster analysis based on a density so as to acquire a plurality of homology clusters. By using the method, the homology clusters during a transient process in the electric power system can be rapidly and accurately identified and real-time discrimination of the homology clusters is realized so that an important reference is provided for scheduling personnel to adopt corresponding emergency control measures and a stable control level of the electric power system is increased.

The invention relates to a line power stealing detection method based on wavelet analysis of power data. The method can judge in real time whether a power stealing behavior exists on a power line through wavelet transform analysis based on power data. The specific steps comprise: calculating hourly output power of a distribution station area and hourly power consumption of each user according to the power data, wherein the power data comprises hourly power consumption of total electricity meters in the distribution station area, and hourly power consumption of each user; calculating and normalizing a hourly line loss rate curve according to the calculated hourly power; performing local wavelet transform with a sliding window on the hourly line loss rate curve, and extracting a singular value vector after the wavelet transform; and calculating an eigenvalue according to the singular value vector, and comparing with a power stealing detection threshold to detect in real time whether there is a power stealing behavior and the power stealing time.

The invention relates to a mechanical fault feature extraction method based on wavelet transform and topological data analysis, and the method comprises the steps: carrying out the feature analysis of an original signal through the wavelet transform and topological data analysis, carrying out the parallel fusion of two groups of features, and inputting the two groups of features into a subsequent classification network; on the one hand, wavelet transform analysis is adopted to consider the time domain and frequency domain characteristics of mechanical fault data, original vibration signals are analyzed from multiple scales, and better characteristics are extracted from non-stationary original time domain signals; on the other hand, topological data analysis is applied to the field of fault diagnosis, the deep topological relation between fault data sets and the shape in hidden data can be effectively mined out through the topological data analysis method under the influence of noise, more robust and more important data features are extracted, and the subsequent fault type classification accuracy can be improved.

The present invention provides a stratigraphic sequence division method and a device. The method comprises the steps of according to the seismic reflection characteristics of a seismic profile, identifying a two-level sequence interface and a three-level sequence interface; respectively positioning the two-level sequence interface and the three-level sequence interface at the corresponding depths of a well logging in the seismic profile; subjecting the logging curve of the well logging at the corresponding depth point of the two-level sequence interface to primary reconstruction to generate a first-level reconstruction curve; subjecting the first-level reconstruction curve at the corresponding depth point of the three-level sequence interface to secondary reconstruction to generate a secondary reconstruction curve; and subjecting the secondary reconstruction curve to wavelet analysis according to a preset scale so as to generate a stratigraphic sequence division result. According to the technical scheme of the invention, based on the long-term and accurate identification of stratigraphic sequences, the medium-term and long-term low-frequency sequence division results and the short-term high-frequency sequence division results of a sequence curve are ensured to be accurate and reliable. The logging curve is reconstructed twice to obtain a secondary reconstruction curve based on seismic data, and the wavelet transform analysis at different scales is conducted. Therefore, the true combination of well-log and seismic data during the stratigraphic sequence dividing process is realized.

The invention discloses a method for detecting a cable early fault of a low-current grounding system. The method comprises the steps: analyzing the characteristics of the zero-sequence voltage waveform signal sampled due to the early fault of the cable through wavelet transform, extracting a zero-sequence voltage fundamental frequency band signal, judging the early fault of the cable from the dimensions of the root-mean-square value and the duration of the zero-sequence voltage fundamental frequency band, and taking the duration of the zero-sequence voltage exceeding a threshold value as the duration of the early fault, thereby determining the early fault of the cable.

The invention relates to an electric energy quality monitoring and electric appliance fault analysis system based on a wavelet neural network and a working method thereof, and the method comprises thesteps that (1) an intelligent circuit breaker obtains a current signal in a power grid, and transmits the current signal to a main control system; (2) the main control system receives the current signal, and obtains harmonic information of the current signal through calculation of the trained wavelet neural network; (3) electric energy quality monitoring: comparing the harmonic information with an electric energy quality model in a preset database one by one, calculating similarity, and judging a problem type; (4) electric appliance fault analysis: comparing the harmonic information with an electric appliance fault model one by one, calculating similarity, and judging a fault type; and (5) a judgment result is transmitted to a far-end server and is stored by the server, and a user can check information in real time through a web page end or a mobile phone APP end. Wavelet transform analysis is realized by adopting the three-layer neural network, the wavelet transform accuracy can be improved, and the harmonic analysis precision is further improved.

The invention discloses a gamma energy spectrum set analysis method for deducting a background based on a WTSVD algorithm. The gamma energy spectrum set analysis method comprises the following steps:S1, performing grading according to a wavelet transform adaptive degree; S2, analyzing the time domain and the frequency domain of the wavelet, judging whether artificial nuclide interference exists or not, if yes, entering the step S3, and if not, directly entering the step S4; S3, eliminating data interfered by artificial nuclide, and performing supplementary measurement; S4, constructing a background training matrix Bn * p, and decomposing the wavelet through Mallat; S5, measuring a gamma energy spectrum, and constructing a gamma energy spectrum set S; S6, performing SVD decomposition, namely unitary matrix decomposition, on the background training matrix Bn * p, and extracting a principal component matrix U; and S7, performing noise reduction on the energy spectrum data set S through bit operation based on the principal component matrix U to obtain a noise reduction energy spectrum set D. The influence of artificial nuclide on the background energy spectrum is accurately reduced through wavelet transform analysis.

The invention provides a dynamic compensation method for harmonic current of a power grid based on wavelet transformation. The method comprises the steps of: performing wavelet transformation analysison a sampled power grid current, decomposing power grid current data by adopting a Matllat algorithm, and separating a wavelet coefficient of a fundamental current in the power grid current; selecting three wavelet coefficients with the phase difference of 120 degrees from the wavelet coefficients of the fundamental current, and operating the three wavelet coefficients by utilizing the sine wavephase characteristics to obtain a fundamental current compensation coefficient; reconstructing the wavelet coefficient of the fundamental current, calculating the fundamental current in the power gridcurrent, and compensating the fundamental current by adopting the fundamental current compensation coefficient to obtain a high-precision real-time fundamental current; and subtracting the fundamental wave current from the power grid current to obtain a high-precision harmonic current in the power grid. The dynamic compensation method solves the defect of a poor dynamic performance when the real-time detection of the harmonic current is performed by adopting the wavelet transform, can effectively improve the real-time detection precision of the harmonic current of the wavelet transform powergrid and can improve the dynamic response performance of the harmonic current.