46results about How to "Suitable for real-time processing" patented technology

The invention provides a method for carrying out blind source separation on convolutionary aliasing voice signals. Firstly, a time domain convolutionary aliasing model is converted into a frequency domain multi-channel linear instantaneous convolutionary aliasing model, which can be realized by the following steps: firstly, converting convolutionary aliasing time domain signals into a frequency domain; then carrying out relatively independent ICA operations on each channel to obtain independent components. Next, the independent components are rearranged by an MSBR algorithm, which specificallycomprising the following steps: firstly, classifying signals of different frequency bands; then progressively obtaining transposed matrixes according to different object functions step by step, wherein the steps of rearrangement are mutually complementary. The MSBR algorithm utilizes the strong relevance of harmonic frequency to improve the iteration accuracy and solves the residual uncertainty of residual frequency bands according to the continuity of adjacent frequency bands and corresponding reference frequencies, and the computational complexity of the MSBR algorithm is approximately in direct proportion to the number of reference frequency bands. The invention improves the convergence efficiency and the accuracy, is more suitable for real-time processing, has good separation performance of convolutionary mixed voice signals and can also be applied to real phonetic environment.

A method for tracking an object in a sequence of video frames includes the following steps: creating a model with characteristic features for the object to be tracked; and performing a template matching algorithm in individual frames on the basis of the created model for determining a position of the object in the respective frame. An apparatus arrangement for performing the method includes at least one video camera (10, 12), at least one monitor (24, 26), one computer (20) and one input device (28) for an observer (22).

The invention discloses a mobile platform ground movement object detection method including the steps as follows: (1) respectively calculating SWIFT characteristic description of a (n-m)th frame image, an nth frame image and a (n+m)th frame image of an image sequence; (2) by taking the nth frame image as a reference, respectively registering the (n-m)th frame image and the (n+m)th frame image to obtain registration images; (3) respectively conducting space multiscale movement significance tests on the two registration images with the nth frame image; (4) calculating a difference absolute value image of the nth frame image and the (n-m)th frame registration image, and the difference absolute value image of the nth frame image and the (n+m)th frame registration image; (5) respectively calculating a difference image of the (n-m)th frame and the nth frame as well as a difference image of the nth frame and the (n+m)th frame; (6) mixing the two difference images to obtain a three-frame difference image and then conducting binaryzation; and (7) conducting morphological processing on a binary segmented image and then conducting region marking so as to obtain a final detection result image. The mobile platform ground movement object detection method solves the background compensation problem under the condition of a moving platform and has very good robustness on the light change and background interference in a scene as well as image deformation caused by platform movement.

The invention discloses a tumble detection method for a human body based on a three-axis acceleration sensor, which acquires human body three-dimensional acceleration signals through the three-axis acceleration sensor. The method adopts the three-axis acceleration sensor to acquire the acceleration signals of the human body in three directions, adopts two characteristic quantities of SVM and SMA to detect the tumble of the human body. The method comprises the following particular steps: 1) acquiring acceleration data of the human body; 2) performing data preprocessing; 3) calculating SVM value; 4) performing SVM wavelet analysis; 5) detecting SVM detection, if the SVMs of a plurality of consecutively acquired sample points are larger than or equal to the threshold value th1, confirming the tumble of the human body initially, and entering step 6), otherwise, continuing step 5); 6) calculating SMA value, if the SMA is larger than or equal to the threshold value th2, entering step 7), otherwise, determining as suspicious tumble, returning to step 5); 7) if the SMAs of a plurality of consecutively acquired sample points is larger than or equal to the threshold value th2, confirming the occurrence of tumble finally and alarming, otherwise, determining as suspicious tumble, and returning to step 5).

The invention relates to a method for detecting a crowded people flow in real time based on three-dimensional depth map data. The method comprises the following steps: acquiring three-dimensional depth map data by using a camera; acquiring preliminary head targets by using foreground extraction and head target area detection; tracking pedestrians and recording the movement loci; and determining the number of the pedestrians and moving direction according to the movement loci. The method performs image processing on a three-dimensional depth map, effectively solves a defect of false detection in two-dimensional image data processing, improves the accuracy of people flow statistics, reduces computation complexity, may detect people flow in real time, and is suitable for intelligent video monitoring field.

The invention discloses a sinus heart rate turbulence trend detection method based on piecewise linearization. The piecewise linearization is adopted for specifically analyzing whether the variation trend of sinus heart rate accelerates first and then decelerates after premature ventricular contractions or not. The method includes the steps: (1) electrocardiosignal preprocessing; (2) self-learning process for first 10 seconds; (3) HRT (heart rate turbulence) sample collection; (4) piecewise trend analysis; and (5) turbulence trend representation based on a cloud model. The variation trend of the sinus heart rate at the RR interval is detected by means of piecewise linearization, and the turbulence trend is further represented by natural language through the cloud model. By the aid of an MATLAB (matrix laboratory) simulation tool, signals in an MIT-BIH heart beat irregularity database are selected for verification, and the variation trend of the sinus heart rate after single-time ventricular premature beat can be detected correctly. In addition, using the exponential smoothing method to predicate QRS complex occurrence positions to facilitate detection of QRS complex, and using a template for judging sinus heart beat is simple to implement and suitable for real-time treatment.

The invention discloses a data processing method and device based on a rule engine, equipment and a storage medium. The data processing method comprises the following steps: receiving a data acquisition instruction comprising a processing target and a target data source, and acquiring a service data set from the target data source; preprocessing all the service data in the service data set according to a preset Sql processing rule to obtain a service data result set; calling a distributed computing engine to traverse the service data result set, and converting each group of service data in theservice data result set into a first object; determining a service rule set matching the processing target from a preset rule database; determining a service rule in the service rule set according tothe service data in the first object; and selecting a rule engine according to the service rule, calling the service rule through the selected rule engine to process the first object, and obtaining an output processing result. According to the data processing method based on a rule engine provided by the invention, the operation efficiency of non-professionals can be improved.

The invention relates to an airborne SAR image automatic target positioning method, which convert the coordinates of a target in the SAR image into an actual geographic coordinates of the target on the earth by utilizing the flight information (location, speed, etc.) of an aerial carrier and the SAR image information (resolution, center distance, central Doppler frequency, etc.). Based on a range-Doppler equation, the invention combines an earth model equation to carry out an target location, which not only can realize an absolute localization without a control point, but also considers the influence of the earth model on positioning accuracy; on the other hand, Taylor series expansion is used for carrying out the linearization of nonlinear equations; an iterative method is adopted to solve the equation, thus improving the positioning speed; meanwhile, the initial value point of an iterative computation is appropriately selected according to the operation mode of radar, thereby controlling a converging direction in the iteration and avoiding a target confusion. As a positioning method without the control point, the automatic target positioning method of the invention has the advantages of high accuracy and high speed, thereby being capable of carrying out a real-time processing and suitable to various airborne SAR systems.

The invention discloses an amplitude-phase error correction method for an MIMO radar transceiving array based on clutter, and mainly solves a problem that the detection performances of a radar are worsened because of the amplitude-phase inconsistency of array elements. The implementation process comprises the steps: constructing receiving, transmitting and transceiving combined guide vectors of aradar and a receiving guide vector and a transceiving combined guide vector of a radar receiving subarray; respectively calculating a projection matrix of the transceiving combined guide vectors of the radar and the radar receiving subarray; respectively constructing transmitting and receiving array error compensation coefficient cost functions through the projection matrix and the echo data, andsolving an optimal solution of each cost function; correcting the transmitting and receiving array guide vectors through transmitting and receiving array amplitude phase error matrixes, and achievingthe correction of the radar transceiving array. According to the invention, the number of information source targets does not need to be estimated, and the method just needs a single echo sample for effectively estimating the amplitude-phase error of the MIMO radar transceiving array. The method is simple in calculation burden, is low in complexity, and is suitable for the error correction and real-time processing of the MIMO radar transceiving array.

The invention relates to a method for generating a sound masking signal based on time reversal. In the method, a corresponding sound masking signal is acquired according to a target sound source signal and has a long-time amplitude spectrum similar to that of the target sound source signal and extremely low speech intelligibility. The method comprises the following steps of: picking up the target sound source signal by using a microphone or a microphone array and performing preprocessing to acquire a clean target sound source signal; performing framing based on a specific time window length according to the acquired target sound source signal; and performing time reversal on a time domain of each frame of signal according to y(t)=x(-t) to acquire the sound masking signal corresponding to a target sound source, wherein x(t) represents a frame of signal, and the y(t) represents an output signal of the frame. The time reversal sound masking signal has the long-time amplitude spectrum similar to that of the target sound source signal; the method has advantages in energy masking; and the time reversal sound masking signal is completely unintelligible or is partially intelligible and cannot become a new interference sound source.

The invention provides a real-time motion compensation method for an ultra-wide band synthetic aperture radar (UWB SAR), which is characterized in that motion parameters output by a single-point global positioning system (GPS) are used for carrying out real-time motion compensation of the UWB SAR, the flight time is segmented according to Ta, when the current moment is a starting point of an imaging synthetic aperture, an imaging coordinate system is built, and if not, the next steps are carried out: step 1, eliminating the single-point GPS position jumping errors; step 2, calculating the transverse motion errors according to the single-point GPS positioning information; step 3, calculating the heading speed errors; and step 4, calculating the sight line direction motion errors and compensating the motion errors in the airborne UWB SAR sight line direction. In the method of the invention, the information processing is only carried out on the output data of a navigation system, the motion parameter evaluation does not need to be carried out, the storage space and the calculation quantity are saved, and the large-area imaging compensation precision is high. In addition, the cost is low, and the realization is easy.

The invention provides an image processing method and an image processor, and relates to the field of image processing. The algorithm is simple, the image matting effect is lightly affected by change of lighting, and the method is suitable for real-time processing of an image. The method comprises the following steps: acquiring a brightness component and a chrominance component of each pixel point in an input image; counting chrominance component values of pixel points of the input image and determining the color corresponding to the pixel points with the highest occurrence rate as background color of the input image; dividing a chrominance coordinate system into a foreground region, a background region and an unknown region according to the background color; determining whether each pixel point in the input image belongs to the foreground region, the background region or the unknown region; determining whether the pixel point belonging to the unknown region belongs to the foreground region or the background region.

The invention discloses a method for estimating the state of charge (SOC) of a secondary battery. Modeling on dynamic characteristics of the different kinds of batteries is achieved by using the strong learning capacity of a non-linear auto-regressive exogenous neural network, and the state of charge of the battery is estimated on line based on the built model by using an extended Kalman filteringalgorithm. The process of modeling on the dynamic characteristics of the battery based on NARX is offline, and the SOC estimation is performed on line in real time. The instantaneity of SOC estimation is not affected by the training process of the model. According to the method provided by the invention, the accurate simulation of the dynamic characteristics of the battery can be achieved withouttoo much training set sample data, the accurate estimation of the battery SOC can be achieved; the method has the characteristics that the online estimation is small in calculated quantity, and the cheap digital processor can be used for operating, is applicable to the battery management system based on the low-cost microcontroller to predict the SOC of the battery, and has the advantages of realtime, high efficiency and low cost.

The invention discloses a direct sequence spread spectrum signal interception detection method. The method comprises the steps of 1, obtaining a data sequence x (n); 2, obtaining a self-correlation sequence r (m); 3, smoothing the r (m); 4, setting a search step j; 5, estimating an accumulation amount C (j); 6, searching the maximum value C (jmax) of the accumulation amount C (j); and 7, carrying out detection judgment and pseudo code cycle estimation. According to the detection method provided by the invention, the direct sequence spread spectrum signal interception detection can be realized only by use of the one-time self-correlation estimation result; through accumulation of correlation peaks of the self-correlation sequence and step variable search, the processing gain is improved; the influence of information codes is reduced; the detection performance is improved; the process is simple; and the prior information is unnecessary.