45results about How to "The amount of calculation is not large" patented technology

The invention discloses an indoor robot SLAM method. The method comprises the following steps: obtaining image data through a camera, wherein the image data comprises RGB images and depth images; using a corner detection algorithm, an LK characteristic point tracing algorithm, and a RANSAC algorithm to process the RGB images and depth images to adjust the position and angle of the camera so as to obtain the RGB image data information under a robot operation system; converting the depth images in a world coordinate system to a ground coordinate, traversing the depth images that are projected on the ground coordinate, setting the gray value of the grid where a barrier stays as a first characteristic value, carrying out traversing to obtain a 2D barrier grid map; through a single line laser radar scanning mode, searching the 2D barrier grid map, when the grid with a grey value equal to the first characteristic value is found, feeding back the distance between the grid and the camera to obtain the distance between a robot and the barrier, obtaining the depth images under a robot operation system, and obtaining an environment 2D map.

The invention discloses a speech enhancement method using a speech existence probability, after the processing of the invention, the speech quality is higher, and the noise reduction amount is larger.The method is realized by the following technical scheme that: on the basis of a MIS measure speech enhancement method, the speech existence probability is used, an inputted noisy speech is sampled,framed and windowed, and then a noisy speech spectrum is obtained through fast Fourier transform (FFT); the noise estimation of the obtained speech spectrum is carried out, based on a non-stationary noise minimum search algorithm of statistical information, the smoothing between front and back frame noise estimation values is carried out by using inter-frame correlation, and a noise power spectrumis estimated; at the same time, speech prior signal-to-noise ratio estimated values obtained in several front and back frames are smoothened; then the speech existence probability and a MIS measure gain factor are combined, the obtained noisy speech spectrum is multiplied by a new gain factor, and a spectrum of the enhanced speech is obtained; and the inverse fast Fourier transform (IFFT) is performed to obtain an enhanced time domain speech signal.

Disclosed are an image flicker detection method and device, and an image capturing device. The image flicker detection method includes the following steps: carrying out exposure imaging on an imaging target under irradiation of a light source so as to form image frames; based on mean luminances of pixels of different rows in the image frames, judging whether relative motion of any two image frames exists in an exposure imaging process; if no relative motion exists, based on mean luminances of pixels in different rows in the two image frames, detecting whether flicker signals exist; and if the flicker signals exist, according to the detection result, determining the frequency of the flicker signals. The technical scheme solves a problem that the prior art is incapable of simply, rapidly and accurately detecting image flicker under a motion scene.

The invention discloses a method for providing augmented reality. The method comprises the following steps: acquiring an on-site scene image through a terminal by using a camera, and displaying the image on a terminal screen; continuously detecting a bar code in the on-site scene image when a target mark is detected in the on-site scene image; analyzing the detected bar code, and acquiring data information in the bar code, wherein the data information comprises marks of corresponding virtual contents; acquiring relevant virtual contents according to the data information of the bar code; overlapping the virtual contents onto a scene displayed on the terminal screen. The invention discloses a system and a terminal for providing augmented reality. Due to adoption of the method, the system and the terminal, more virtual contents can be provided under the condition of limited numbers of augmented reality marks.

The invention discloses an indoor robot positioning method by combining a visual odometer and a physical odometer. The method is characterized in that by collecting the images, ORB characteristics are extracted for image coupling, camera position and pose estimation, and closed loop detection to accurately position the robot. The visual odometer is added in a known environment for closed loop detection on the robot, the accumulated error of the physical odometer based on particle filter is eliminated, the global error of the odometer is changed to staggered accumulation, and an closed is established based on the above process. The visual odometer is combined for effectively solving the problem of error accumulation of the physical odometer, and realizes automatic positioning and accurate re-positioning of the robot in the known environment, the increased computation burden is not large, efficiency and real-time performance are guaranteed, indoor navigation requirement can be preciously satisfied, and the indoor robot positioning method is effective for solving the problem of inaccurate positioning of the robot under large environment in the prior art.

The invention discloses a synchronization method for orthogonal frequency division multiplexing system, including: the launcher firstly constructs the originated / synchronous sequence that is sent out with OFDM data symbols together. Based on the correlation of CP effective length in OFDM symbols form the receiving end, the correlation result conducts a time domain processing to the correlation sequence energy to get the best synchronization points set after one synchronization. Based on the originated sequence and CP, the departure estimation and departure compensation of fractional part DF several times are processed. During a range of synchronization points set output at once, the long pre-symbol received after the compensation is made time-domain correlation with the symbol to find the second time synchronization point. Then after time-domain of the short pre-symbol and the received pre-sequence is made frequency departure compensation, time-domain-related operations is conduct to get the DF departure estimation of integer times, and departure compensation is conduct.

The invention discloses a photovoltaic transmission line fault phase selection method based on an IMF energy entropy. The method comprises steps that a three-phase fault component voltage and a zero sequence fault component voltage of a photovoltaic power station transmission line are acquired; EMD of the three-phase fault component voltage and the zero sequence fault component voltage is respectively carried out, IMF components under each frequency band are adaptively acquired; a correlation coefficient sequence between each IMF component and an original signal is solved, and effective IMF components are screened through a set threshold; energy entropy fault characteristic parameters are calculated according to the effective IMF components, in combination with short circuit fault characteristics and different characteristics of the characteristic parameters, phase selection criteria are constructed, and fault phase selection is realized. The method is advantaged in that the EMD methodis combined with the energy entropy concept, the method is suitable for photovoltaic power station transmission line photovoltaic side fault phase selection, accurate motion can be carried out at thephotovoltaic side, characteristics which are not influenced by the fault position, transition resistance and a fault initial angle are realized, and accuracy of photovoltaic transmission line fault phase selection is improved.

The invention discloses an inverse time lag over excitation protection method of a generator and a transformer. The method comprises the following steps: carrying out segment handling by making a given inverse time lag over excitation action characteristic curve, wherein the given inverse time lag over excitation action characteristic curve is obtained by a plurality of groups of input given values (generally 8 groups to 10 groups), calculating excitation multiple according to voltage and frequency, and adopting a segment Lagrange polynomial interpolation method to work out the corresponding action time of over excitation protection on the excitation multiple. Therefore, the inverse time lag over excitation protection method realizes high-precision inverse time lag over excitation protection.

The invention relates to a determination method for attitudes of a spinning stabilized meteorological satellite based on a coarse-fine attitude relation model. The method is characterized by comprising the following steps that 1: the coarse attitude of the satellite is calculated; 2: a random error in the coarse attitude of the satellite is eliminated; and 3: the coarse attitude is modified by utilizing the coarse-fine attitude relation model. The method has the benefits that 1, the precision is high, and a calculated fine attitude result is basically consistent with the practical fine attitude of the satellite; 2, the applicability is high, and the fine attitude of the satellite can be calculated by the method to achieve accurate positioning of an observation cloud picture when the fine attitude cannot be obtained within 24h after attitude orbit control of the satellite or in an area observation mode; and 3, a processing procedure is distinct, the operand is not great, and implementation is facilitated.

The invention provides a non-peer image encryption method based on saliency detection, and the method comprises the steps: obtaining an original image, and generating a saliency image according to thesignificance mapping combination of preset features of the original image; dividing the original image into an important region and a non-important region according to the saliency map; encrypting the important region by adopting a DNA encryption algorithm to obtain a first encrypted image; performing compressed sensing encryption on the non-important area to obtain a second encrypted image; andrecombining the first encrypted image and the second encrypted image to obtain a final encrypted image. According to the method provided by the invention, the security of the image can be ensured, thecalculation amount of image encryption operation can be reduced, and the data volume of image transmission and storage is reduced.

The invention provides a secondary radar answer code extracting and confidence interval marking algorithm, synthetically using position of answer pulse, sigma and Delta information and adjacent answer information, adopting an idea of establishing a pulse information list for each answer pulse, convenient to implement software programming. It has the characters of high processing efficiency, strong self-adapting ability, convenience for real-time implementation, etc. It can be used in processing answer signals for secondary radar, enemy-self identifier, and civil aviation management system.

The invention discloses a positioning correction method based on distance measurement. The method comprises the following steps of: 1, acquiring a distribution condition of a distance measurement error through tests; 2, estimating a probable range of an actual distance value corresponding to a distance measurement value with errors according to the error distribution during distance measurement each time; 3, positioning according to the probable range of an actual distance between a target node and a plurality of other nodes, and calculating a probable range of an actual position of the target node; 4, taking the center of the probable range as a positioning result if the probable range is not greater than an expected value, otherwise repeating the step 3 according to a distance measurement result of the target node and the other nodes until three groups of probable ranges of the actual position of the target are obtained; and 5, estimating the probable range of the actual position according to the probable ranges and taking the center of the probable range as a corrected position of the target node. The method has the characteristics of small computation increase, no need of extra distance measurement and high positioning precision after correction.

The invention discloses a three-dimensional reconstruction method of a scanning electron microscope image. The method comprises the following steps that a threshold value of the number of images is set; a template area is selected for the first frame of the image captured by the SEM; the predefined step size is used to move the sample platform to obtain the real-time image, and the template regionof the first frame is used for template matching; the matched image is added into the matched image sequence; for the matching image sequence, the region of interest is selected and the modified Laplace transform is used to compute the focus measurement of each pixel; Gaussian interpolation is used to calculate the maximum focus measurement value of each pixel, and the height of the correspondingpoint is calculated according to the maximum focus measurement value to obtain the three-dimensional point cloud; according to the three-dimensional point cloud, the interpolation fitting is carriedout on the three-dimensional graphics to achieve the three-dimensional reconstruction.

The invention discloses a transcoding method used for similar video code streams, which comprises the following steps: firstly, analyzing sound code stream data, namely extracting video data from the file format, analyzing according to the sound code stream format, obtaining the parameter and the transformation control parameter which an existing decoder needs, secondly, transforming the sound code stream data, namely transforming the sound code stream data according to transformation control information, thirdly, resetting a sound code stream, namely carrying out the corresponding format arrangement to the data which is transformed in the second step or data which is directly transmitted according to the requirements of the format of a target code stream, and then obtaining the target code stream needed by the existing decoder. The method greatly expands the use range of an existing decoder in maximum, fully utilizes the existing decoder, and improves the availability ratio of a device with a decoder, in particular to an inlaid device with the hardware decoder.

The present invention discloses a rapid correction method of scanning electron microscope image drift. The method comprises the steps of: performing feature detection of a reference image of a scanning electron microscope; performing feature detection of real-time images of the scanning electron microscope; performing feature first matching of a feature detection result of the reference image anda feature detection result of the real-time image, and rejecting a mistake matching pair; calculating and obtaining a homography matrix between the reference image and the real-time image; performinghomography matrix frame mapping of feature points of the real-time image, rejecting the mistake matching pair again, and obtaining an accurate matching pair; and recalculating and obtaining a homography matrix between the reference image and the real-time image, and performing perspective transformation of the real-time image. The rapid correction method is simple in algorithm and fast in speed, can achieve real-time feedback and is good in image drift correction effect.

The invention provides a wireless communication receiver, an interference elimination method and device thereof and a signal demodulation method. The wireless communication receiver at least comprises two receiving antennas. The interference elimination method comprises the following steps: carrying out channel estimation on an interference cell based on received signals; calculating correlation of interference cell emission signals on each receiving antenna through the channel estimation result of the interference cell; and carrying out interference elimination operation on signals received by the receiving antennas according to the correlation of interference cell emission signals on each receiving antenna. According to the technical scheme, the interference elimination is performed on the received signals through the wireless communication receiver under the condition of not increasing too much operation; the wireless communication receiver is allowed to have a certain interference inhibition capability; and thus the reception performance is improved under the condition of same frequency interference.

The invention provides a multi-cell channel estimation method for a time division-synchronous code division multiple access (TD-SCDMA) system. The method specifically comprises the following steps of: performing rough channel estimation on each cell by adopting single-cell Steiner channel estimation; determining the total number of access users according to a rough channel estimation result; judging whether the total number of the access users meets multi-cell joint channel estimation conditions, performing multi-cell joint channel estimation on the access users of each cell of the total number of the access users meets the multi-cell joint channel estimation, otherwise performing multi-cell serial interference cancellation channel estimation on the access users of each cell; and performing de-noising processing on a multi-cell joint channel estimation result or a multi-cell serial interference cancellation channel estimation result. By the method, operational complexity can be greatly reduced and the needs of the access users in any number can be satisfied under the condition of ensuring high channel estimation accuracy, thereby better solving the problem of severe inter-channel interference of the TD-SCDMA system on cell edges and providing quality assurance for multi-cell joint detection.

An image flicker detection method and device, and an image capture device. The image flicker detection method includes: performing exposure imaging on an imaging target illuminated by a light source to form an image frame; judging whether any two image frames have relative motion during the exposure imaging process based on the average brightness value of pixels in each row in the image frame; If there is no relative motion, it is detected whether there is a flickering signal based on the pixel brightness average value of each row in the two image frames; if there is a flickering signal, then the frequency of the flickering signal is determined according to the detection result. The technical solution solves the problem that the prior art cannot simply, quickly and accurately detect image flickering in a moving scene.

The invention discloses a synchronous phase detection method for a static frequency converter with direct voltage compensation. The method comprise: (1) measuring the AC-side three-phase line voltage of the converter in real time, and eliminating the waveform distortion of the AC-side three-phase line voltage in a commutation area by directly pre-estimating and compensating the waveform of actual frequency conversion line voltage so as to obtain a sine estimated amount; (2) using a software phase-locked loop to perform real-time phase locking on the estimated amount to achieve robust synchronous phase detection of the static frequency converter. The method uses only a single PT to realize the accurate tracking of the synchronous phase of the static frequency converter under the variable frequency conditions, so as to ensure that the a converter bridge does not generate commutation failure, to ensure the safe equipment operation, and to improve the power factor on the output side of the static frequency converter. The algorithm does not depend on a complex hardware filter circuit, and is small in software calculation amount, and is low in cost.

The invention relates to a harmonic source identification method based on data association analysis, which comprises the following steps: acquiring harmonic voltage of a point of common coupling (PCC)and harmonic current data of each feeder within a period of time, generating a waveform, and carrying out preliminary positioning on a harmonic source; carrying out segmented clustering analysis on the harmonic voltage and the harmonic current, and outputting clustering analysis results; comparing the harmonic voltage with the clustering result of each section of each group of harmonic current, judging the similarity degree of each feeder harmonic current and the harmonic voltage on the whole, and selecting the harmonic current with the highest similarity degree as a judged harmonic source; and comparing whether the judged harmonic source is consistent with the preliminarily positioned harmonic source or not, and if so, indicating that the harmonic source is successfully positioned. The method does not depend on power grid parameters and models, is not affected by other conditions, is simpler in calculation process, is less in calculation amount, is higher in speed, and can achieve the quick recognition of the harmonic source.

The invention discloses a phase modifier frequency conversion starting process short circuit fault protection method and device. The method comprises the following steps of acquiring a phase modifiermachine end side voltage, a phase modifier neutral point side current, a computer end side voltage amplitude and frequency, and a ratio of the machine end side voltage to a frequency per unit value, and calculating a neutral point side current amplitude and a neutral point side current amplitude variation; in an initial constant magnetic flux starting stage, when the ratio of the phase modifier machine end side voltage to the frequency per unit value is reduced to exceed a setting value and a neutral point side current amplitude is increased to exceed the setting value, determining to be a short-circuit fault; or in a constant-voltage starting stage, when the machine end side voltage is reduced to exceed the setting value and the neutral point side current amplitude is increased to exceedthe setting value, determining to be the short-circuit fault; and after the short-circuit fault is judged, carrying out short-time delay protection on alarming or tripping. In the technical scheme, short-circuit fault protection criteria in the frequency conversion starting process are proposed in a segmented mode, and sensitivity of short-circuit fault protection in the phase modifier frequency conversion starting process can be improved.

The invention discloses a phase modifier excitation loss protection method and device based on a break variable criterion. The method is characterized by comprising the following steps of collecting stator voltage and current of the phase modifier and rotor voltage, calculating stator voltage, reactive power, rotor voltage value and break variable amount of the phase modifier, according to the stator voltage, the reactive power value, the rotor voltage and the amplitude and the direction of the variation of the phase modifier, comprehensively judging an excitation loss fault of the phase modifier, and performing alarm or tripping after short-time delay protection, thereby improving the protection sensitivity of the excitation loss fault of the phase modifier and the reliability under otherabnormal operation or faults.

The invention discloses a method for extracting systematic frequency fluctuation signals. The method includes steps of calculating a positive sequence voltage in a real-time sampling mode, carrying out lowpass filtering on the positive sequence voltage; adopting a zero-passage method to calculate frequency of the filtered positive sequence voltage, and obtaining frequency signals with a large number of noises; and processing the obtained frequency signals through a digital filter, and obtaining smooth frequency fluctuation signals. The method can reduce a complex transfer function design step with a current / power as damping signals, greatly improves sensitivity with frequency as damping signals, and improves a damped oscillation function of a dynamic reactive power compensation device in a complex power grid.

A mass amount adjusting mediation decoupling control method suitable for a pitch peak heating station, and belongs to the technical field of controlling of heating stations. The mass amount adjusting mediation decoupling control method suitable for the pitch peak heating station mainly solves the problem that an existing heat supply process control method is not suitable for controlling a pitch peak heating station which is of nonlinear multivariable coupling system control. The mass amount adjusting mediation decoupling control method suitable for the pitch peak heating station comprises the following steps: modeling a pitch peak heating station heat supply coupling system; utilizing the method combing typical signal response with least squares to build heat supply mass amount adjusting main adjusting channel models G11, G12 and G13; carrying out discretization of an obtained main channel modular form (2), and utilizing a proportion integral derivative (DIP) neural network to learn and obtain the non-linear relationship G13, G21 and G22 combining with the discretized main channel model; and utilizing the PID neural network to carry out decoupling control of the mass adjustment and quality adjustment of the pitch peak heating station. The mass amount adjusting mediation decoupling control method suitable for the pitch peak heating station utilizes the advantages of solving the problems of nonlinearity, strong coupling and multiple variables of the PID neural network to carry out multivariable decoupling control of a quality adjusting circuit and a mass adjusting circuit, and the mass amount adjusting mediation decoupling control method suitable for the pitch peak heating station is applied to pitch peak heating stations.

The invention discloses a transcoding method used for similar video code streams, which comprises the following steps: firstly, analyzing sound code stream data, namely extracting video data from the file format, analyzing according to the sound code stream format, obtaining the parameter and the transformation control parameter which an existing decoder needs, secondly, transforming the sound code stream data, namely transforming the sound code stream data according to transformation control information, thirdly, resetting a sound code stream, namely carrying out the corresponding format arrangement to the data which is transformed in the second step or data which is directly transmitted according to the requirements of the format of a target code stream, and then obtaining the target code stream needed by the existing decoder. The method greatly expands the use range of an existing decoder in maximum, fully utilizes the existing decoder, and improves the availability ratio of a device with a decoder, in particular to an inlaid device with the hardware decoder.

The invention discloses a generator low-frequency starting process protection method and device. The method comprises the steps of collecting the generator end side voltage of a generator; calculatingthe machine end side voltage frequency of the generator; when the calculated machine end side voltage frequency value is higher than a frequency braking low-frequency input constant value, keeping the original protection constant value of each protection unchanged; and when the calculated machine end side voltage frequency value is lower than the frequency braking low-frequency input constant value, increasing the original protection constant value by a corresponding multiple, wherein the corresponding multiple calculation mode is that a frequency deviation value is multiplied by a ratio braking coefficient constant value. By improving the protection constant value of the low-frequency starting process protection, the reliability of the initial extremely-low frequency stage protection inthe variable-frequency starting process of the generator can be improved, and when the frequency of the generator is higher than the set value, the protection can take a sensitive constant value to ensure the sensitivity of the protection.

The invention provides a multi-cell channel estimation method for a time division-synchronous code division multiple access (TD-SCDMA) system. The method specifically comprises the following steps of: performing rough channel estimation on each cell by adopting single-cell Steiner channel estimation; determining the total number of access users according to a rough channel estimation result; judging whether the total number of the access users meets multi-cell joint channel estimation conditions, performing multi-cell joint channel estimation on the access users of each cell of the total number of the access users meets the multi-cell joint channel estimation, otherwise performing multi-cell serial interference cancellation channel estimation on the access users of each cell; and performing de-noising processing on a multi-cell joint channel estimation result or a multi-cell serial interference cancellation channel estimation result. By the method, operational complexity can be greatly reduced and the needs of the access users in any number can be satisfied under the condition of ensuring high channel estimation accuracy, thereby better solving the problem of severe inter-channel interference of the TD-SCDMA system on cell edges and providing quality assurance for multi-cell joint detection.

The invention discloses a positioning correction method based on distance measurement. The method comprises the following steps of: 1, acquiring a distribution condition of a distance measurement error through tests; 2, estimating a probable range of an actual distance value corresponding to a distance measurement value with errors according to the error distribution during distance measurement each time; 3, positioning according to the probable range of an actual distance between a target node and a plurality of other nodes, and calculating a probable range of an actual position of the target node; 4, taking the center of the probable range as a positioning result if the probable range is not greater than an expected value, otherwise repeating the step 3 according to a distance measurement result of the target node and the other nodes until three groups of probable ranges of the actual position of the target are obtained; and 5, estimating the probable range of the actual position according to the probable ranges and taking the center of the probable range as a corrected position of the target node. The method has the characteristics of small computation increase, no need of extra distance measurement and high positioning precision after correction.