Preprocessing of Radiological Images; Comparison of the Application of Polynomic Algorithms and Artificial Neural Networks to the Elimination of Variations in Background Luminosity

摘要

One of the major difficulties arising in the analysis of a radiological image is that of non-uniform variations in luminosity in the background. This problem urgently requires a solution given that differing areas of the image have attributed to them the same values and this may potentially lead to grave errors in the analysis of an image. This article describes the application of two different methods for the solution of this problem: polynomial algorithms and artificial neural networks. The results obtained using each method are described and compared, the advantages and drawbacks of each method are commented on and reference is made to areas of potential interest from the point of view of future research.