Using the wiener predictive filter approach to improve measurements of soil density obtained with an X- ray computed tomography scanner

This paper presents a method for improving the measurements made with direct transmission X-ray, which generate the projections that allow the use of the reconstructive tomography method. It builds on the use of the predictive Wiener filtering, which is a nonlinear filtering. Such method allows a better accuracy of the measurement obtained by the reconstructed images of Computed Tomography (CT) scanner, i.e., improving the figure of merit Signal/Noise Ratio (SNR). Results show a consequent improvement in measures of the soil densities, which is of interest to a better management and monitoring of the physical parameters of agricultural soils.