11 results about "Variance function" patented technology

The invention discloses a self-calibration optimized image denoising method using convexity, which belongs to the technical field of image processing, is used for image denoising under the condition that the noise level is unknown, and comprises the following steps: generating a noisy image and a secondary noisy image pair, and constructing a conditional denoising network; based on a linear minimum mean square error estimation theory, deducing a denoising estimator and calculating a variance of an estimation error, and introducing a noise ratio parameter to carry out normalization analysis; determining an optimal synthetic noise variance through derivation or a ternary search algorithm by using convexity of an error variance function; performing empirical formula calibration on the optimal synthetic noise variance; and training and executing the denoising network by using the calibrated optimal variance, and outputting a final denoised image. According to the method, the robust training process is constructed under the condition that the noise level is unknown, noise statistical characteristics can be effectively learned, and the limitation that the noise level needs to be known in advance in a traditional method is overcome.

A method of semiconductor manufacturing process parameter optimization based on Bayesian average Kriging and evidence theory is proposed. By introducing Bayesian model averaging into the basic Kriging model, different variance functions are adaptively weighted, which effectively improves the robust estimation of spatial correlation structure and alleviates the bias caused by model mis-specification under small sample conditions. At the same time, the introduction of evidence theory not only realizes the fusion of multi-source model prediction results, but also explicitly describes the cognitive uncertainty at the model level, enhancing the interpretability of the prediction results and the value of process diagnosis. The method shows good applicability in high-cost and data-scarce manufacturing systems, and can provide reliable proxy modeling support for complex processes such as semiconductor etching.

This invention discloses a self-calibrated image denoising method utilizing convexity, belonging to the field of image processing technology. It is used for image denoising under conditions of unknown noise levels. The method includes generating a pair of noisy images and images with secondary noise, constructing a conditional denoising network; deriving the denoising estimator and calculating the variance of the estimation error based on the linear minimum mean square error estimation theory, and introducing a noise ratio parameter for normalization analysis; utilizing the convexity of the error variance function, determining the optimal synthetic noise variance through differentiation or a ternary search algorithm; calibrating the optimal synthetic noise variance using an empirical formula; training and executing the denoising network using the calibrated optimal variance, and outputting the final denoised image. This invention, by constructing a robust training process under conditions of unknown noise levels, can effectively learn the statistical characteristics of noise, overcoming the limitation of traditional methods that require prior knowledge of the noise level.

The invention provides a method for improving AutoFocus performance, which comprises the following steps of: S1, extracting an edge value by a sobel operator: extracting image edge information through a cv2. Sobel () instruction, and representing as sobelx = cv2. Sobel (img, cv2. CV32F, 1, 0) and sobely = cv2. Sobel (img, cv2. CV32F, 0, 1); s2, solving a variance: af = (np.var (sobelx) + np.var (sobely)) * N, np.var is a function for solving the variance, the formula is that the sum of the variances of sobelx and sobely is multiplied by a coefficient N, and the obtained af is the definition value of the picture; pixel values obtained after Sobel operator processing are distributed more dispersedly, so that the variance can be used as an evaluation criterion because the variance is the measure of the dispersion degree when a random variable or a group of data is measured. Compared with the prior art, the method has the advantages that the performance is improved while the effect is better than that of the prior art, the difference value between the peak value and the peak valley of the optimized method is larger in the aspect of the effect, and therefore the peak value can be found more easily, the calculation amount is small in the aspect of the performance, and the CPU utilization rate and the memory use can be reduced.

The application provides a GLV multi-beam direct writing path planning method and system for holographic anti-counterfeiting, and the method comprises the following steps: acquiring model analysis depth information and discretizing the model analysis depth information into an initial feature matrix; evaluating a thermal diffusion cumulative amount to establish a physical threshold model, performing truncation and energy compensation on a dense area exceeding the threshold to generate a thermal equilibrium distribution matrix; extracting residual energy features of a splicing overlap area, constructing a semi-variance function to generate a spatial variation convolution kernel for smoothing operation to generate a reference exposure control sequence; establishing and solving a compensation model related to power, speed and depth to generate a control sequence snapshot; and collecting an instantaneous mechanical speed matching snapshot in real time to drive direct writing. The application effectively improves the machining precision and restoration degree of the holographic anti-counterfeiting topography.

The application discloses an RCS full-probability size extrapolation method and system based on an electromagnetic scattering mechanism, and the method is as follows: 1, a backscattering electric field formula under different sizes is derived according to the electromagnetic scattering mechanism, and a target RCS calculation expression of an incident wave vector is obtained according to the definition of a radar cross section (RCS); the target RCS calculation expression is converted into a variant formula used for constructing a covariance function through a coefficient method and an integral mean value theorem; 2, the variant formula is inversely transformed and then is disassembled; 3, a polynomial covariance function is used to represent a polynomial part in the inversely transformed formula, and a spectral mixture covariance function is used to represent a cosine part in the inversely transformed formula; the two kinds of covariance functions are combined into an SPFPE covariance function; 4, the SPFPE covariance function is initialized by using a random number method; 5, hyperparameters of the SPFPE covariance function are optimized with a maximum log-likelihood as an optimization target, and an SPFPE-GPR method is obtained; 6, the target RCS is extrapolated by using the SPFPE-GPR method.

The invention relates to a laser peening inherent strain prediction method based on Gaussian process regression, which comprises the following steps: carrying out a laser peening orthogonal test, and establishing model training data; carrying out normalization processing on the training data by adopting a maximum and minimum normalization method; selecting a mean value function, a covariance function and a hyper-parameter optimization algorithm, and establishing a Gaussian process regression model; training the Gaussian process regression model by using the training data, and optimizing hyper-parameters of the Gaussian process regression model; and utilizing the trained Gaussian process regression model to predict inherent strain distribution under any process parameter combination so as to realize continuous correspondence between the process parameters and the inherent strain. According to the laser peening intrinsic strain prediction method based on Gaussian process regression, intrinsic strain distribution prediction under any technological parameter combination is achieved by establishing the Gaussian process regression model and optimizing hyper-parameters through training data, and the method has the advantages that the number of tests is reduced, and prediction efficiency and accuracy are improved.

The application provides a GLV multi-beam direct writing path planning method and system for holographic anti-counterfeiting, and the method comprises the following steps: acquiring model analysis depth information and discretizing the model analysis depth information into an initial feature matrix; evaluating a thermal diffusion cumulative amount to establish a physical threshold model, performing truncation and energy compensation on a dense area exceeding the threshold to generate a thermal equilibrium distribution matrix; extracting residual energy features of a splicing overlap area, constructing a semi-variance function to generate a spatial variation convolution kernel for smoothing operation to generate a reference exposure control sequence; establishing and solving a compensation model related to power, speed and depth to generate a control sequence snapshot; and collecting an instantaneous mechanical speed matching snapshot in real time to drive direct writing. The application effectively improves the machining precision and restoration degree of the holographic anti-counterfeiting topography.

This invention discloses a method and device for refining pseudorange observations for BDS non-geostationary orbit satellites. The specific method includes: analyzing the SCB characteristics of BDS non-geostationary orbit satellites using multipath combination analysis; eliminating the influence of time-invariant parameters in MP combination based on epoch difference, establishing an ED-SCB sequence for each satellite and frequency, and using the averaging method to achieve the best estimate of the ED-SCB for the equation; for satellite a at the i-th frequency, assuming the ED-SCB at the mid-elevation angle is θ, calculating the absolute value of the elevation angle E1; establishing an SCB correction model using a polynomial piecewise fitting algorithm, with the principle of minimizing the sum of the absolute values ​​of the residuals; deriving the DSC filter variance based on the CSC filtering principle, using the DSC variance function described by the error propagation law, and determining the smoothing window based on the principle of minimizing variance; to suppress Doppler cumulative integration error, introducing a balance factor to adjust the weights of DSC and the original pseudorange, establishing an RDSC variance equation based on the error propagation law, and establishing a dynamic weight adjustment model for RDSC based on the principle of minimizing variance.

The invention provides an annotation method and system for new energy multi-modal data, and relates to the technical field of data annotation, and the method comprises the following steps: obtaining multi-modal data of new energy equipment, preprocessing the multi-modal data, mapping the preprocessed multi-modal data to the same feature space, and obtaining mapped multi-modal data; based on the type of the multi-modal data, performing multi-modal spatial statistical analysis on the feature vector of the mapped multi-modal data by using a semi-variance function, and generating a cross-modal spatial constraint rule in combination with a cross-modal hash algorithm; and based on a cross-modal spatial constraint rule, labeling and classifying the multi-modal data of the new energy equipment, and generating a new energy multi-modal labeling result. According to the method, the accuracy and engineering applicability of new energy multi-modal data labeling are remarkably improved.

The invention relates to the technical field of ecological change evaluation, and discloses a land ecological quality interannual rating and change detection method to solve the technical problems of complex modeling, strong parameter dependence, tedious operation and the like in a conventional change extraction method, and a land ecological quality rating scheme is constructed by using the first two principal components of tasseled cap transformation, ALBEDO and LST. The ecological quality level of the land surface is quantified, and periodic monitoring of the ecological quality condition is achieved; the method comprises the following steps: introducing a multi-time sequence dynamic rate of ecological change, quantifying a land surface ecological quality change (improvement / degradation) level, combining a semi-variance function theory, constructing a function relationship between the ecological quality dynamic rate and a corresponding observation time interval according to different time intervals (from small to large) in sequence, and obtaining a dynamic rate mean value at the same time interval; and carrying out extreme value detection on the maximum / minimum value of the dynamic rate under different time intervals in the interannual observation period to obtain a multi-time-sequence dynamic rate product data set representing the change of the land ecosystem.