120 results about "Curve fitting algorithm" patented technology

The invention provides a method for tracking a small target with high precision under a complex background and a low signal-to-noise ratio (SNR). The method comprises pre-treating an image under the complex background and low SNR, portioning and extracting the target based on a target adaptive threshold of the binomial distribution judgment rule, adopting a curve fitting algorithm based on Kalmanfilter thinking improvement to carry out motion prediction on the target, using data fusion of infrared and visible light transducers to improve detection probability of the target and reduce false-alarm probability, and when the shape of the target alters, using a shape of which edge characteristics are normalized to identify and seek characteristic invariables so as to achieve precise tracking of the target.

An automatic-tracking device of vehicle movement orbit consists of vehicle terminal, service end and client end. It is featured as selecting position coordinate automatically by vehicle terminal according to preset movement path to set up highway enclosing wall, confirming specific position relative to highway enclosing wall by vehicle terminal according to highway position coordinate and actual position coordinate for judging whether vehicle is in the scope of highway enclosing wall or not.

A system and method for estimating the distance between a borehole and a subsurface boundary of interest in a geophysical region. In one embodiment, available existing sensor data for the geophysical region is used to create a resistivity model of the region, with the model reflecting changes in resistivity across the boundary. A hypothetical borehole is defined in the model, with the hypothetical borehole having a number of segments along its length that are spaced-apart from the boundary by a number of different, preselected distances. Based on the existing data, resistivity curves corresponding to the hypothetical boundary are plotted. The ratio between two selected resistivity curves in each of the respective spaced-apart segments is computed, and these ratio values are plotted as a function of distance from the boundary. A curve-fitting algorithm is applied to the ratio plot to derive an equation which correlates the selected sensor readings with distance from the boundary. This equation may then be applied to actual sensor data from a sensor package present during an actual drilling operation to provide the drilling operator with quantified estimates of distance from an actual borehole and the boundary.

The invention provides a method for facula center positioning of a CCD image measurement system. According to the method, a facula image generated by a CCD sensor is simulated by means of software, denoising is carried out on a laser facula generated by simulation by means of median filtering, binarization is then carried out on the denoised image, edge extraction is carried out on the image treated with binarization by means of the Gauss-Laplace operator, then primary positioning is carried out by means of the gray level gravity method, and finally accurate positioning is carried out on the primarily positioned image by means of the Gaussian curvefitting algorithm. According to the method, due to the introduction of the image pre-processing step, noise in the image can be effectively filtered, and interference resistance is improved; due to the fact that only useful data points in the image are calculated and analyzed, the calculated amount is small, and then realization of the software is easy and resources of the system are saved greatly.

The invention relates to a method for apperceiving and reconstructing a non-vision structural form of a near space vehicle model. The method comprises the following steps: on the basis of a distributed embedded dispersed bragg fiber grating sensing array, sensing and detecting strain information of each measuring point of a test model, and utilizing a certain curvature conversion algorithm to convert the strain information into curvature information; on the basis of a curvature interpolation, calculating the related coordinate value of each point of wings and frame of the test model according to a three-dimensional space curve fitting algorithm of orthogonal curvature information; utilizing a coordinate fuse algorithm to obtain an absolute coordinate value of each point of the model; and lastly, reconstructing the deformation state and the low-frequency vibrating form of the test model according to a computer graphic processing technology and realizing the three-dimensional visual real-time display for the change in the model form. An experimental system for the method comprises a simulated near space vehicle model, a fiber grating network analyzer with a network service function, a high-performance computer and a display, wherein the fiber grating sensing array is embedded into the surface of the simulated near space vehicle model.

Improved calibration method for vector network analyzers stores sparse calibration data, interpolates systematic error data using the sparse calibration data from each measurement made by the vector network analyzer, and from the systematic error data at each measurement frequency and the uncorrected Measurement Data Creates calibrated measurement data. Sparse calibration data may be generated by measuring every Nth frequency step over the calibration frequency range greater than the specified measurement frequency range, or by measuring every frequency step over the calibration frequency range and compressing the resulting measurement data. Interpolation can be achieved by using a curve fitting algorithm such as a parametric polynomial curve fitting algorithm.

The invention relates to a battery analysis method and system based on electrochemical parameters. According to the method, collected battery macroscopic parameters are mutually coupled with electrochemical-based microcosmic parameters obtained by further calculation through a curve fitting algorithm, so that the feature parameters input into a trained machine learning model contain more effectiveinformation; two-stage data cleaning is carried out through edge calculation and cloud calculation, so that the prediction precision of the machine learning model is fundamentally improved; through analysis of battery big data, fault early warning and out-of-control alarm can be realized according to outlier detection of real-time data; and a relatively accurate prediction model can be obtained through training of massive historical data, and accurate estimation of a battery state is realized.

The invention discloses an air conditioner device refrigerating capacity test method based on a multi-dimensional curve fitting algorithm, and belongs to the technical field of air conditioner refrigerating capacity test methods. According to the principle on which the test method is based, a reference prototype physical model of an air conditioner device is established and used as a research object, and a refrigerating capacity mathematical model of the air conditioner device under any work condition is obtained by establishing a multi-dimensional curve fitting mathematical model between the refrigerating capacity of a reference prototype of the air conditioner device and all influencing parameters and conducting training simulation through multiple sets of sample data including the normal condition. By means of the mathematical model, the refrigerating capacity of the reference prototype of the air conditioner device under any work condition can be obtained and used as the theoretical value of the refrigerating capacity of a tested air conditioner device of the same type. Under the same test work condition, the actual refrigerating capacity of the tested air conditioner device is obtained. A refrigerating capacity test device or instrument manufactured through the test method can be used for refrigerating capacity tests of the air conditioner device and is widely used for air conditioner device refrigerating capacity tests of traffic vehicles, ships, civil use and industrial application.

The invention provides a background signal extraction method in radar scattering cross section measurement. The method includes measuring echo signals used for obtaining radar transmission signals reflected by a dihedral angle reflector, isolating the echo signals into in-phase channel signals and orthogonal channel signals, obtaining first complex signals through reconstruction of the in-phase channel signals, obtaining second complex signals through the orthogonal channel signals, respectively obtaining real part signals and imaginary part signals of the first complex signals and the second complex signals by means of a curve fitting algorithm, obtaining the real part signals of background signals according to the real part signals and the imaginary part signals of the first complex signals, and obtaining the imaginary part signals of background signals according to the real part signals and the imaginary part signals of the second complex signals. Due to the fact that the background signals are extracted through the curve fitting algorithm in high precision, the background signal extraction method in the radar scattering cross section measurement solves the problem that robustness is poor in practical engineering application.

The invention provides a breast molybdenum target image breast area segmentation and calcification point detection method. According to the method, artificial interfering substances in a breast molybdenum target image is rapidly removed by the aid of an image gradient weight calculation method based on neighborhood pixels, a breast and chest muscles are initially segmented by the aid of an image segmentation method based on pixel aggregation, and breast boundaries are accurately detected and fitted by the aid of a straight line detection algorithm based on Hough transformation and a curve fitting algorithm based on a polynomial. By the method, marking precision of boundaries among the breast, a background and the chest muscles in the breast molybdenum target image is remarkably improved, calcification points in the breast are detected by the aid of texture filtering, so that breast area calibration and calcification point detection accuracy is improved, the breast area can be automatically segmented and calibrated, and the calcification points can be automatically detected, marked and measured.

This document relates to melting curve analysis of nucleic acids. The method according to first aspect of the invention comprises analyzing a nucleic acid melting curve measured from a sample, the melting curve comprising a sum signal of at least two nucleic acid melt signals and a background signal as a function of temperature. The method further comprises optimizing at least one constant in a temperature-dependent exponential correction function so as to minimize the variation of the nucleic acid melting curve at a temperature region where the target nucleic acids in the sample remain essentially double stranded, and generating a corrected nucleic acid melting curve representative of the nucleic acid melt signal by applying said exponential correction function over the region of the measured melting curve where the strands of the nucleic acids dissociate. According to further aspects, the invention relates to a curve fitting algorithm for precise estimation of melting peak areas and a mathematical transformation for linearization of calibration curve data to enhance the linear measuring range of a competitive PCR assay. The invention provides a powerful tool for analyzing PCR-amplified sample containing two or more different nucleic acids having similar but distinguishable melting temperature.

The invention discloses a robot stable motion control method based on curve fitting modeling and multi-constraint foot point estimation. The method comprises the following steps: (1) a local environment of a swing leg foot end is modeled based on a curve fitting algorithm; (2) a reachable area of the swing leg foot end is fitted by combining with working space of the swing leg foot end according to the fitted local environment; (3) according to the reachable area of the swing leg foot end fitted in the step (2), a multi-constraint foot point estimation algorithm is combined to estimate and analyze a plannable foot point in the reachable area of the swing leg foot end; and (4) according to the steps (1) to (3), a multi-gesture conversion algorithm is combined to analyze a robot joint output position, and the polynomial interpolation operation is performed for a joint corner to finish the stable motion control of a robot under non-structural environment. The stable motion control method is high in terrain adaptability and high in efficiency.

The invention discloses a wavelet transform and curve fitting based pulse wave preprocessing method. The pulse wave pretreatment method includes the following steps of 1), subjecting to acquired pulse condition signals to noise suppression with the wavelet global threshold method; 2), rectifying baseline drift and cycle extraction of the pulse condition signals; 3), smoothing the pulse condition signals on the basis of fitting of Gaussian curves. Compared with the prior art, the pulse wave pretreatment method has the advantages that cycles of the pulse condition signals can be effectively extracted, fitting of pulse condition data is completed only in 120 seconds by operating 3.1GHz on a machine according to the curve-fitting algorithm, average fitting error of 97.62% of pulse conditions is smaller than 0.35, and for the fitting result that the error is larger than 0.35, morphological characteristics of the pulse condition signals are also kept, and no diagnostic information is lost.

The invention discloses a remote monitoring device for mountain cracks. The device consists of a sensor node and a monitoring host. The device uses a GSM network to perform information transmission, adopts a cyclic working mode from working to stopping to monitor the mountain cracks, and meanwhile adopts digital filter and curve fitting algorithm to improve the precision of data acquisition; therefore, the device can implement high-precision and long-distance mountain crack monitoring, and greatly prolongs the working time.

The invention discloses an optical fiber radiation thermometer based on a colorimetric method, and belongs to the field of high temperature measurement. The thermometer comprises an optical system and a circuit system, wherein lenses of a lens group are connected with a high-power Y-type power transmission optical fiber to form the optical system; the circuit system adopts signal dual-channel processing and comprises photoelectric detectors, a preprocessing module, a preamplifier, an AD sampling module, a signal processing module, a display module, a keyboard input module and an information interaction module. Signals are divided into two paths via the power transmission optical fiber and transmitted to the corresponding photoelectric detectors, optical signals are converted into current signals, the current signals are converted into voltage signals via the preprocessing module and the pre-amplification module, the signal processing module adopts a curve fitting algorithm for analysis, sampling signals are obtained, a corresponding temperature value is calculated through calibration, and difference setting is further improved through the interaction module. The optical fiber radiation thermometer structure based on the colorimetric method can be applied to measurement of a crystal furnace thermal field.

The invention discloses an iris positioning method and equipment. The method comprises steps that an iris identification system obtains an iris image which contains an iris and a pupil to be positioned; that the position of a light spot in the pupil is determined, wherein the light spot is formed by irradiating the pupil with an external light source; that according to the position of the light spot and the gray information of the iris image, a first boundary circle and a second boundary circle used for representing the iris to be positioned in the iris image are determined, wherein the first boundary circle comprises the center of circle and the radius of the first boundary circle, and the second boundary circle comprises the center of circle and the radius of the second boundary circle; and that according to the first boundary circle and the second boundary circle, the iris in the iris image can be positioned. Therefore, the iris identification system can rapidly position the position of the iris by a simple image processing technology and a curve fitting algorithm according to the obtained gray information of the iris image, so that the working efficiency of the iris identification system is improved.

The invention discloses a spectrum extraction method and device for a hyperspectral collection system, and the hyperspectral collection system comprises a slit, a smartphone, a light-splitting module, and a collimating module. The method comprises the steps: providing various types of monochromatic light through a monochromator, enabling the monochromatic light to reach the collimating module through the slit and becomes parallel light, enabling the parallel light to irradiate the light-splitting module, to pass through a convergence lens, and to pass through a camera lens of the smartphone and finally to be converged at the array detector, and obtaining a plurality of RGB images; obtaining the brightest point in each of the RGB images, enabling the brightest points to serve as the positions of spectrum lines, and determining the displacement values of all spectrum lines through employing an interpolation method; enabling a collected RGB image of a real object to be converted into a gray-scale image, and then plotting a spectrum curve through combining with the displacement values of all spectrum lines; carrying out the compensatory plotting of the spectrum curve according to the actual spectral transmittance through employing a curve fitting algorithm, and obtaining the actual spectrum curve. The method can complete the high-precision extraction of a spectrum, is simple in algorithm, is high in accuracy, and is of great significance to the development of spectral imaging.

The invention relates to a three-dimensional spinal lateral protrusion reconstructing method based on orthophoric and lateral phases, belonging to the field of computer-aided measurement. The three-dimensional spinal lateral protrusion reconstructing method based on the orthophoric and lateral phases comprises the following specific steps of: firstly, setting beginning and ending seed points at gravity centers of a first thoracic vertebra and a last lumbar vertebra in an X-ray film in the orthophoric and lateral phases respectively; secondly, acquiring gravity centers of other vertebras by reasoning transversely and longitudinally; thirdly, fitting all the gravity center points into a curve by a curve fitting algorithm; fourthly, computing the included angle between the normal of each gravity center point on the curve and the horizontal line and the offset of each gravity center point relative to the first gravity center point, wherein the inclined angle between two gravity center points of orthophoric and lateral phases of a same vertebra and the offsets of the two gravity center points form a spatial three-dimensional parameter; and finally, according to the spatial three-dimensional parameter of each vertebra, translating and rotating the corresponding vertebra in a standard spinal model according to the three-dimensional parameter so as to form a spinal spatial three-dimensional model for pathological changes of a patient suffering from the spinal lateral protrusion. According to the three-dimensional spinal lateral protrusion reconstructing method, the time for performing X-ray scanning on the human body is shortened.

The present invention relates to a drawing method of isopiestic line air pressure slot line and ridge line. The method comprises the steps of: determining whether an isopiestic line is a closed curveor not, if yes, extracting slot points and ridge points by employing a chord contact method, or else, extracting slot points and ridge points according to the position where the isopiestic line is located; and tracking the slot points and ridge points, and connecting the slot points and the ridge points. In the invention, extraction of the slot points and the ridge points of an air pressure fieldcan be tested based on the chord contact method, and the least square method curve fitting algorithm is employed to provide a unique ligature method of the slot points and the ridge points to obtain agood automatic identification result of the slot points and the ridge points. The connection result of the slot points and the ridge points meets the requirements of the weather system slot and ridgeline analysis to reflect the trend of the slot points and the ridge points, and the test result shows that the algorithm can complete the automatic analysis and identification of the slot lines and ridge lines of the weather system.

The invention discloses a method and a device for determining a moving trace of an obstacle, and belongs to the field of intelligent traffic. The method comprises the following steps: determining a transverse coordinate of the obstacle in each frame of image of continuous n frames of images acquired by a vision sensor, wherein n is an integer larger than 1; performing curve fitting on the transverse coordinates of the obstacle in the continuous n frames of images according to a preset curve fitting algorithm, thus obtaining the moving trace of the obstacle. According to the method and the device, the images acquired by the vision sensor arranged on a vehicle are subjected to analytic processing, so that the moving trace of the obstacle is determined; the problem that of high complexity during data processing in the relevant technology is solved, and the effects of reducing the detection cost of the obstacle and improving the data processing efficiency of the sensor are achieved. The method is used for determining the moving trace of the obstacle.

The invention discloses an LED plant growth spectrum fitting algorithm. The algorithm is characterized by comprising the following steps: 1, establishing spectrum radiation model of a single LED: performing approximation on a distribution model of radiation intensities along with wavelengths in a unit solid angle on an optical axial direction of the single LED by use of a Gauss distribution function, f(x)=ae<-(x-b)<2> / C<2>>; and 2, carrying out a curve fitting algorithm: according to a superposition principle of spectrums, obtaining a basic mathematic model of spectrum synthesis of multiple LEDs: L(lambda)=sigma KiSi(lambda); and in a fitting process, setting a first wavelength lambda0 and a coefficient K0, and successively solving lambda (i) and Ki by use of a least square algorithm. According to the invention, any one fitting spectrum with quite high similarity with a target spectrum can be obtained through spectrum fitting.

The invention provides a human body key point detection and self-learning method and device. The method comprises the steps that image information of multiple continuous frames is extracted through cameras or videos, multiple key points of a human body are detected for each image, wherein the key points have coordinate information and confidence coefficient information, and for the key points withthe confidence coefficient smaller than 0.6, an adjacent edge curve fitting algorithm is used for completing coordinate prediction and confidence coefficient calculation of the key points with the confidence coefficient smaller than 0.6. According to the method, the coordinates of the shielded key points of which the confidence coefficients are smaller than 0.6 are predicted, so that the accuracyof human body key point detection is improved, and human body key point information can be returned more accurately.

A system and method estimates the distance between a borehole and a subsurface boundary of interest in a geophysical region. In one embodiment, available existing sensor data for the geophysical region is used to create a resistivity model of the region, with the model reflecting changes in resistivity across the boundary. A hypothetical borehole has a number of segments along its length that are spaced-apart from the boundary by different, preselected distances. The ratio between two selected resistivity curves in each of the respective spaced-apart segments is computed, and these ratio values are plotted as a function of distance from the boundary. A curve-fitting algorithm is applied to derive an equation, which may be applied to actual sensor data from a sensor package.

The invention discloses a farmland seedling ridge identification method and system based on semanteme and instance segmentation, and the method comprises the steps: collecting a current farmland image, coding, carrying out instance segmentation and semantic segmentation, clustering, carrying out fitting of feature points, drawing to obtain a seedling line and a furrow line, and the like, whereinthe system comprises a function module for achieving all above steps. According to the invention, seedling ridge detection is carried out by using an end-to-end visual image segmentation training method. The required network scale is reduced to a certain extent, the operation speed is effectively improved. Meanwhile, the embedded vector easy to cluster is obtained through the deep network, then the seedling line and the furrow line are efficiently recognized through the simple clustering algorithm and the curve fitting algorithm, and big data samples can be used for continuously supplementingand improving the accuracy and robustness of the detection method.

The invention discloses a method and device for generating a dental panoramic image on the basis of a CBCT image. The method comprises steps of: determining a seed point in a current axial surface image of a CBCT apparatus, processing the seed point by using a curve fitting algorithm to generate a dental arch line; using a curve surface where the dental arch line is located as a central curve surface, computing body tissue density CT values corresponding to the pixels in the central curve surface according to a set sampling thickness, and generating initial dental panoramic images according to the body tissue density CT values; and determining the gradient factor of an anisotropic diffusion filter formula and processing the initial dental panoramic images with different scales by using the anisotropic diffusion filter formula to generate an final dental panoramic image. The method and device achieve a purpose of obtaining the two-dimensional panoramic image of the entire teeth by using the image processing algorithm on the basis of CBCT three-dimensional data.

The invention discloses a method for recovering abnormal topographic data of lunar craters. The method comprises the following steps of: obtaining elevation value data; screening abnormal areas by adoption of a section elevation abnormal value detection algorithm; carrying out fitting to obtain an elevation distribution curve on the current section line by adoption of a section line curve fitting algorithm, and predicting elevation values of the abnormal areas on the section line according to the fit curve; and obtaining new recovery elevation value to complete the recovery of the original topographic data. According to the method disclosed by the invention, the data loopholes and data abnormal values in lunar topographic DEM data can be recovered, and the recovered topographic data is smooth in surface, so that the processing demands of Chang'e lunar basic spatial data are fully satisfied and the effect is apparent.

The invention discloses a portable optical fiber radiation thermodetector and a measuring method of the portable optical fiber radiation thermodetector, and belongs to the field of pyrometry. The thermodetector comprises a high-power optical fiber coupler, an energy transmitting optical fiber, a photoelectric detector, a pre-amplification module, an AD sampling module, a signal processing module, a display module, a keyboard input module, a digital-to-analog conversion module and a 4-20mA current loop. Thermal radiation energy at a broadband spectrum is coupled by the high-power optical fiber coupler and transmitted to the photoelectric detector by the energy transmitting optical fiber to convert optical signals into electrical signals, the electrical signals are sampled by the AD sampling module, obtained sampled signals are analyzed by the signal processing module through the curve-fitting algorithm, and a corresponding temperature value can be worked out finally through calibration. The portable optical fiber radiation thermodetector is very suitable for measuring crystal oven thermal fields due to the structure, easy and convenient to use and very practical.

The invention relates to a method for improving the ultrasonic thickness measurement accuracy by using curve fitting. The method comprises the following steps: according to the characteristics that an ultrasonic signal is similar to a straight line in a short time period above and below a zero point, and when the signal amplitude is higher, the signal is more similar to the straight line, collecting two adjacent sampling points above and below the zero point at the two sides of the positive maximum peak or the negative maximum peak of an echo signal, fitting a straight line through the sampling points, working out corresponding coordinates of a feature point on the fitted curve, sending the coordinates to a thickness calculating module and calculating the thickness of an object to be measured. The method has the advantages that a curve fitting algorithm is used for compensating sampling precision, so that the ultrasonic thickness measurement accuracy is effectively improved under the condition that the sampling clock frequency is constant so as to meet the requirements on the thickness measurement accuracy in certain occasions, the complexity of hardware implementation is reduced, and the reliability of the implementation is improved.

The invention provides a method for measuring the spatial orientation of a pipeline based on a CMOS image sensor. The method includes the following steps that: a laser light source and a CMOS image sensor are arranged at mechanical centering fixed ends at two ends of a universal joint, an attenuation sheet and a bubble fluorescent ring are arranged between the laser light source and the CMOS image sensor, the initial position of the laser light source is set, so that the initial position of the laser light source can be located at the center of the pipeline, and laser emitted by the laser light source is overlapped with the central axis of the pipeline; the laser light source moves, and the CMOS image sensor acquires one frame of image when the laser light source moves a certain distance each time from the initial position; laser spots and bubble fluorescent rings on the acquired images are processed, so that the pixel positions of the laser spots on a photosensitive surface and the directions of bubbles on the fluorescent rings are obtained, the center position of each laser spot and the position of each fluorescent ring bubbles are recorded; the center of each pixel light spot is converted into the actual physical position coordinates of a light spot center; and the orientation curve of the pipeline is fitted through using a least squares method-based curve fitting algorithm according to the actual physical position coordinates of the light spot center and the directions of the bubbles of the fluorescent rings.

The invention relates to a real-time monitoring device for the convergence and deformation of surrounding rock in the shield area of a shield machine. The device is composed of a push rod motor, an H-bridge circuit, an ARM chip and a power supply module which are fixed on a mounting plate in a protection box, a laser radar fixed to a push rod of the motor, a hardware part such as a portable WIFI module embedded in the side wall of the protection box, and a software part such as a control program and a communication protocol written for the device. The device filters the measurement data by thesignal processing technology and the curve fitting algorithm to enable the cross-section shape of the surrounding rock to reappear. The box body is embedded in the inner wall of the shield through lugs on both sides, and the laser radar pushes the tempered glass at the top of the box body to lean out of the shield to carry out measuring. The real-time monitoring device for the convergence and deformation of surrounding rock in the shield area of the shield machine can acquire the precise change curve of the surrounding rock in real time, and monitor and early warn the breakdown of the shieldmachine caused by the extrusion deformation of the surrounding rock, thereby reducing the huge economic loss caused by the breakdown. The real-time monitoring device is convenient to assemble and disassemble and small in overall size, adopts the wireless transmission mode, and is suitable for various complex geological environments.