134 results about "Polynomial regression" patented technology

A thermal control system for a liquid crystal cell is presented that may utilize a time division scheme to multiplex temperature sensing and heating functions across active thermal elements such that the cell may generally be kept at a constant temperature. A calibration process characterizes the profile of the cell and generates a polynomial regression formula that provides the voltage drive output for a temperature and cell state input. The control system stores the state of the liquid crystal cell, the regression formula, and reads the temperature of the liquid crystal cell to compute end assert the temperature compensated voltage drive.

A color correction apparatus and method are provided. The color correction apparatus includes: a chromatic region determination module determining a first chromatic region to which an input pixel constituting an input image belongs and a second chromatic region neighboring the first chromatic region on a basis of specified chromatic region division information; a polynomial regression operation module performing polynomial regression on the first chromatic region and the second chromatic region; and a blending module providing corrected pixel information for the input pixel by giving weight values to results of the polynomial regression on the first chromatic region and the second chromatic region. The color correction method includes: determining the first and second chromatic regions; performing polynomial regression on the first and second chromatic regions; and providing corrected pixel information for the input pixel by giving weight values to results of the polynomial regression on the first and second chromatic regions.

The invention discloses a method for calibrating colors of multiple cameras in an open environment, comprising the specific steps as follows: firstly, the white balance of a single camera is calibrated by adopting an X-Rite 24 color standard color card so as to solve the problem that physical parameters of the cameras are inconsistent with one another; secondly, data of the X-Rite 24 color standard color card of different cameras under different monitoring scenes is obtained, the position of the color card is obtained in a man-machine interaction way, the position and the color average value of each color lump in the color card are calculated to obtain 24 color values in two scenes with multiple cameras, color mapping relation among the cameras is calculated in a polynominal regression method, and each camera is subjected to color calibration; and finally, areas with similar colors in the two scenes are found according to an H-S color matching method, and color calibration parameters are adjusted in a self-adaption way by detecting the change of pixel colors in the matched areas with the similar colors through on-line detection, thereby ensuring that the multiple cameras in a monitoring system are consistent in color in the open environment.

A method for recognizing objects in images is disclosed. The method of the present invention comprises the following steps. First, acquire a digital image. Then, select one or more objects from the image according to a certain characteristic. Next, generate an x-axis histogram and/or a y-axis histogram from the segmented image. Then, find the zeroes and maxima for the x-axis histogram and/or the y-axis histogram and use the polynomial regression analysis to determine the shape, shape and location of each of the objects in the segmented image according to the zeroes and maxima. If the two curves linking two zeroes and one maximum in the x-axis histogram and the y-axis histogram are two sloped line, the corresponding object may be determined to be a triangle. If each of the four curves linking two zeroes and two maxima is a line, the corresponding object may be determined to be a rectangle.

The testing method of high-temp. component creep life adopts the simulation pipe-bursting test machine or durable strength testing machine to make creep rupture acceleration simulation test of new materials which have different ageing extents and are not aged under the condition of a series of temp. and stresses. It adopts polynomical regressive analysis method to process data, and obtain the relationship formula of non-aged material between ageing factor, temp. and life time. By utilizing test, data processing and formula operation the material creep service life can be obtained.

The invention discloses a color digital camera white balance correction method based on a color adaptation model. According to the method, a color correction matrix is calculated through adoption of a root-polynomial regression method, so device related response values RGB under a plurality of common light sources (hereinafter called the correction light sources) are converted into CIE1931 tristimulus values XYZ unrelated to devices under the same light source. RGB response values of an unknown light source (hereinafter called a test light source) in a practical scene are converted into an XYZ color space through utilization of a pre-calibrated color correction matrix, the corresponding colors of the RGB response values under a reference light source are calculated through utilization of a CAT02 color adaptation model, and the corresponding colors are the colors sensed by a visual system after an observer carries out color adaptation to the test light source.

The invention discloses a geometric correction method of an airborne imaging hyperspectrum of an unmanned aerial vehicle. At present, an image obtained according to POS (Positioning and Orientation System) data exhibits an overlarge deviation with practicality, and the point, line and surface distortion of scenery is large and is difficult in geometric fine correction through polynomial correction. The method comprises the following steps: collecting the position gesture information of a low-precision POS sensor of a current unmanned aerial vehicle to correct a collimation axis error; according to the angular point and outline information of ground object characteristics, preprocessing the POS data to obtain corresponding elements of exterior orientation; carrying out collinearity equation correction; and through a ground correction point, carrying out polynomial correction. By use of the geometric correction method, a relationship between natural ground object characteristics and the own error of the sensor is considered, correction accuracy is improved, the aerial photography low-precision POS data of the unmanned aerial vehicle is optimized, the aerial photography image of the unmanned aerial vehicle can be accurately corrected only by carrying the low-precision POS sensor and a hyperspectral imager, and technical support is provided for the wide application of the current low-cost imaging hyperspectrum of an unmanned aerial vehicle.

The invention relates to a method for partitioning glutenite sedimentary period based on a Fischer diagram. The method comprises the following steps: selecting logging curve combination, discriminating a lithological change surface, discriminating a high-frequency base-level cycle, determining the preferred period of the high-frequency base-level cycle, determining the accuracy of partitioning the high-frequency base-level cycle, making Fischer diagrams, and partitioning a long-term base-level cycle. The method can be used for constructing the corresponding relationship between lithofacies and the logging curves by analyzing the lithofacies, can be used for selecting the logging curve combinations which can well reflect the lithologic characters to carry out clustering analysis, main composition analysis and polynomial regression analysis so as to determine the lithological change, and can be used for partitioning a high-frequency cycle base level by using the logging curve combination, lithological change level and discrete wavelet high-frequency component, and adopting time-frequency analysis technique to acquire the high-frequency preferred cycle period of the logging curve to inspect the partitioned high-frequency cycle base level. On this basis, the Fischer diagrams are made to obtain curves reflecting the change of the base level so as to partition the sedimentary period.

The invention discloses a partition power grid bus load prediction system based on weather information. Real-time weather information and prediction weather information are used in the system, load prediction of all buses of converting stations of 500 kV and 220 kV is achieved, and recognition of power grid partition and partition load prediction are achieved. A prediction algorithm used by the system comprises a classical algorithm and an intelligent prediction algorithm, the classical algorithm comprises unary linear regression, quadratic polynomial regression, self-adaptation index prediction, index prediction, increasing rate prediction, nonhomogeneous index prediction, a B. Compertz model and a logistic model, the intelligent prediction algorithm comprises an optimized BP neural network algorithm and an optimized particle swarm algorithm, and the system selects a prediction algorithm in a preferential mode during a prediction process. The system is a day-ahead bus load prediction system, bus load and partition load of each time interval from morrow to multiple days in future are predicted, and prediction content is active load of 96 points of a predicted day.

The method for forming the optimal characteristic curve of a solar cell comprises the steps of: providing a first acceptable error; and determining a current-voltage polynomial regression equation, whose square root of the residual sum of square is less than the first acceptable error for a set of solar cell measured data. The order of the current-voltage polynomial regression equation is gradually increased until the square root of the residual sum of square of the current-voltage polynomial regression equation is less than the first acceptable error.

The invention discloses a shield tunneling machine tunneling parameter prediction method based on polynomial regression. According to the invention, the big data processing method based on machine learning is applied to shield engineering data processing for the first time; characteristic learning is carried out on shield engineering big data through a machine learning method after cleaning and statistical processing. By means of the data high-level feature extraction capability of a machine learning algorithm, operation parameters and operation efficiency are efficiently modeled, so that instructive suggestions are made for parameter adjustment of a shield in engineering, and the characteristics of artificial intelligence are achieved.

The invention discloses a spectral reflectivity reconstruction method combining principal component analysis and a regularized polynomial. The method comprises the following steps: 1) collecting spectral data of training samples by employing a multi-spectral imaging system; 2) carrying out dimensionality reduction on the spectral data of the training samples by means of a principal component analysis method to reduce the spectral data quantity of the training sample; and 3) constructing a training sample set through the spectral data of the training samples processed in the step 2), then carrying out polynomial regression channel responding expansion on the training sample set, then constructing a target function by taking the minimum error between the reconstructed spectral reflectivity and the actual value as a target and adding constraining items into the target function by employing a Tikhonov regularization method, and finally, solving the target function to obtain the reconstructed spectral reflectivity so as to reconstruct the spectral reflectivity combining the combining principal component analysis and the regularized polynomial. The method can be used for reconstructing the accurate spectral reflectivity.

The invention discloses a method for detecting an event region based on a splay tree. The method comprises the steps of: randomly arranging sensor nodes, describing a sensing network topology by using a delta network based on a Voronoi image and Delaunay, building the splay tree through three kinds of information Beacon, Probe and Join, fusing data based on polynomial regression and the built splay tree, and finishing the reliable detection of the event region at the same time. The test proves that the method of the invention is feasible and can increase the detecting accuracy in the field ofdetecting the event region.

The invention provides a color correction method and evaluation method for a panorama camera and belongs to the technical field of image processing. The method comprises the steps of S1, collecting raw images comprising standard color cards in preset environments; S2, preprocessing the raw images; S3, extracting R, G and B mean parameters of the preprocessed images; S4, obtaining correction coefficients of reference images and observation images based on a polynomial regression correction model; S5, establishing color correction models of the reference images and the observation images and obtaining color correction coefficients; and S6, correcting camera images online through utilization of the trained color correction models and correction coefficients. According to the method, the constancy of the image can be effectively kept, colors of a practical scene can be effectively restored, and the image quality of the panorama camera is effectively improved.

The invention discloses a machinery device fault combination prediction system and a method, which are stable and reliable, can perform early warning and has high prediction accuracy and realizes the scientific planning and healthy management on the device. The method comprises steps of conditioning and amplifying collected signal through installing a sensor at a suitable test point in the machinery device to collect parameter information of frequency, amplitude and temperature of a spindle and a bearing bush, converting the collected signal to suitable digital signals through A/D conversion and then transmitting the signal to an upper computer, establishing a man-machine interaction interface through an Labviw software, performing collection, processing and storage on data, and integrating with polynomial regression prediction, small wave gray prediction, small wave neural network prediction and the prediction method of the particle cluster optimization support vector machine to perform combination prediction. The invention can not only analyze and predict the machine fault in order to provide feasible solution, but also provide theoretical meaning and application significance to improvement of the monitoring analysis function and the fault diagnosis accuracy of the rotation machinery device and to the enrichment and development of the fault prediction theory.

The invention discloses a battery SOH evaluation method, which comprises the steps of performing charging and discharging test on a battery, and recording test data; performing data processing on thetest data to obtain a V-dQ/dV data point set, and performing filtering and noise reduction; identifying a peak value of the data point set, carrying out segmenting according to the peak value point, and performing polynomial regression on each segment of the V-dQ/dV to obtain an IC curve for qualitatively analyzing a battery attenuation mode; and integrating and then converting the IC curve to obtain a Q-V curve numerical model, performing data verification, and quantitatively analyzing the battery attenuation behavior by calculating optimal voltage curve reconstruction model parameters. According to the method, filtering and nonlinear regression operations are carried out on test data and a calculation result; influence of measurement noise on an analysis result is reduced, the method ishigh in adaptability and is used for various types of lithium ion battery systems, the battery attenuation mode is analyzed in a set and quantitative mode, a reasonable independent variable step length can be set for quantitative analysis, and the calculated amount is greatly reduced. The invention further provides a battery SOH evaluation system.

The invention discloses an indirect prediction method for the remaining life of a power battery. With equal charge voltage rise time and equal discharge voltage drop time as characteristic health factors, a time sequence for characteristic health factors and cycle times is built; global degradation and local fluctuation are decoupled by empirical mode decomposition, particle filtering and polynomial regression are combined to predict the remaining life of the battery, wherein the particle filtering is used for tracking a local fluctuation phenomenon, and the polynomial regression is used for fitting a global degradation trend. Experiment results show that the errors between the prediction result in the method and lithium battery cycle life experiment data from an NASA Ames superior prediction center are within 4%, and good accuracy and good practicability are achieved.

The embodiment of the invention relates to the technical field of automatic driving, and discloses a method and system for adjusting a steering angle of a vehicle steering wheel and a vehicle. The method comprises the following steps: when the vehicle runs on a curve, acquiring the real-time speed of the vehicle and the real-time track curvature of the vehicle, substituting the real-time speed, the real-time track curvature and the real-time factor coefficient of the vehicle into a preset polynomial model, solving a target steering wheel steering angle of the vehicle, and adjusting the steering wheel steering angle of the vehicle according to the target steering wheel steering angle, wherein the real-time factor coefficient is obtained by performing polynomial regression processing on thepolynomial model by using steering wheel steering angle sample data, vehicle speed sample data and track curvature sample data of the vehicle. By implementing the embodiment of the invention, the steering wheel steering angle can be self-adaptively adjusted, and the complicated manual debugging steps are reduced.

The invention discloses a big data prediction method and a system based on a macroscopic factor. The method comprises the following steps of setting a range of collecting basic data; according to a set range, collecting the basic data related to influenza; according to the collected basic data, using a time sequence prediction method and a polynomial regression method to establish a plurality of prediction models; according to the plurality of established prediction models, acquiring corresponding influenza prediction results respectively; and integrating the influenza prediction results of the plurality of prediction models so as to acquire a final prediction result. Therefore, based on information that is relatively easy to acquire, the models are established and the influenza can be predicted.

The present invention provides a prediction method of a power battery SOC, and relates to the field of the storage battery test. The prediction method comprises the following steps: acquiring a battery static state open-circuit voltage and a SOC value with respect to different open-circuit voltages of the battery; establishing a relation curve of the static state open-circuit voltage and the SOC; determining the initial SOC of the power battery through adoption of the polynomial regression analysis method; and acquiring a battery remaining capacity in the process of driving of a vehicle through the time measurement method. Through adoption of polynomial regression analysis, an SOC is corrected according to an open-circuit voltage to make the estimation of the SOC at a high accuracy in the whole time phasing and have real time after a power battery BMS is started.

The invention discloses a fingerprint detail information hiding and recovering method based on set polynomial conversion and harmonics. According to the basic thinking, a triangular quantization expression with displacement and rotation invariance is built through the Delaunay triangulation algorithm on the basis of fingerprint detail characteristics, so that a characteristic polynomial is built on a quantization expression set, assessment elements of non-characteristic space are introduced by means of the harmonics thinking of the set, and information hiding and verification recovering of the fingerprint detail characteristics are achieved. The method overcomes the defect that a finger template of an intelligent card stores information with plaintext and is attacked easily, improves system safety, effectively protects privacy of the biological characteristics of fingers of a user, and can be widely applied to a plurality of application scenarios like passports, identification cards and entrance guard cards.