This paper presents an automatic feature recognition method based on center-surround difference detecting and fuzzy logic that can be applied in ground-penetrating radar (GPR) image processing. Adopted center-surround difference method, the salient local image regions are extracted from the GPR images as features of detected objects. And fuzzy logic strategy is used to match the detected featur...

Gaussian process regression (GPR) is a powerful non-linear technique for Bayesian inference and prediction. One drawback is its O(N) computational complexity for both prediction and hyperparameter estimation for N input points which has led to much work in sparse GPR methods. In case that the covariance function is expressible as a tensor product kernel (TPK) and the inputs form a multidimensio...

the subsoil and groundwater were suspected to be polluted by leakage from underground tanks around the oil storage facility in Dhanbad, Jharkhand, India. to delineate the seepage zone, integrated geophysical methods comprising Ground Penetrating Radar (GPR) and 2D electrical Resistivity Imaging (eRI) methods were employed over the suspected zone and in the close vicinity. the studies were condu...

An important problem of marble-quarry management is assessing the quality and the homogeneity of quarry blocks before excavation. In this study, we decided to image the limestone, which we studied in a marble quarry, in terms of layer thickness, discontinuities and cavities using the ground-penetrating-radar (GPR) method. The method was successfully applied to detect and map the fractures with ...

Prediction uncertainty has rarely been integrated into traditional soft sensors in industrial processes. In this work, a novel auto-switch probabilistic soft sensor modeling method is proposed for online quality prediction of a whole industrial multi-grade process with several steady-state grades and transitional modes. Several single Gaussian process regression (GPR) models are first construct...

Abstract Monitoring of water quality through accurate predictions provides adequate information about management. In the present study, three different modelling approaches: Gaussian process regression (GPR), backpropagation neural network (BPNN) and principal component (PCR) models were used to predict total dissolved solids (TDS) as indicator for The performance each model was evaluated based...

Learning based super-resolution (SR) methods, which predict the high-resolution pixel values but not directly provide an estimation of uncertainty, are typically non-probabilistic and have limited generalization ability. Gaussian processes can provide a framework for deriving regression techniques with explicit uncertainty models, but Gaussian Process Regression (GPR) has a significant drawback...