The regression method for the initial lead identification is analyzed,and Grubbs rule method is proposed.

A new method are proposed to decide the adaptive decomposition order and the threshold of wavelet space of each order based on Grubbs rule, and a proved threshold estimation model of wavelet coefficients is given upon the hard-thresholding and soft-thresholding models.

In this paper,the biginning time of precursory anomaly is determined for the residual sequence of fixed point deformation data with the Laiyite and Grubbs criterions.

describe the Grubbs criterion and Dixon criterion of rejecting rough error through the flow chart and brief program,then limit the use of these ways in the end.

To deal with the K value excursion and the gross error in measurement,self-checking function and Grubbs criterion were adopted.

To get an insight into the planting structure of the agropastoral region,the planting area and yields of main crops such as soybeans,corn and wheat are analyzed in combination with the related meteorological data by means of Bayesian criteria.

Based on the fuzzy set theory, a membership function was introduced according to Bayesian criteria.

According to Bayes rule,all pixels of the abdominal image were partitioned into the corresponding Gaussian components of the mixture model,and the correct rate and the misjudgement rate of partitioning all pixels of each organ were obtained.

Adaptive wavelet denoising threshold for based on bayesian rule;

Four data fusion methods have been investigated in this paper,including "AND" rule,"OR" rule,the largest posterior probability?rule and Bayesian rule.