Image Processing for Neutron Radiography

Abstract

Neutron radiography is a relatively technological development in the annals of non- destructive evaluation. It plays a role of increasing importance as the technology is further developed. In a neutron radiographic formation system, the object to be imaged is illustrated by the object attenuates the neutron beam and a shadow-image, a projection through the object, is observed at the image plane. Due to the finite size of the neutron source, the resulting degradation is modeled by a convolution operation with the system point spread function. Several spreading functions were tested. The estimation of the deconvolution function from the spreading function by the nonlinear solvers,and developement of two- dimensional deconvolution methods are included in this wok. The approaches are catagorized into indirect method in two-dimensional problem. Results of the restoration model based on the half-plane and quarter-plane deconvolution methods to recover the details of the degraded images are in good agreement with the original simulated true data. In addition, the effect of noise during the restoration process is investigated also. This method is expected to be generally useful far neutron radiography.