Wavelet analysis, which is regarded as a mathematical microscope, has been widely used in many disciplines. Being of unique finitude and symmetry, B-spline function is used as wavelet basic function. Based on an analysis of white noise and its wavelet transform characteristics, this paper showed that B-spline wavelet could effectively eliminate the white noises in observed data. The effectiveness of B-spline wavelet noise elimination was compared with that of least square curve fitting on observed susceptibility data in terms of denoising degree, computation efficiency, flexibility of operation, and agreement with well and geologic data. B-spline wavelet noise elimination was superior to least square curve fitting in depicting geology and reservoir.
论文详情