Evaluation of rounding functions in nearest neighbor interpolation

Abstract

A novel evaluation study of the most appropriate round function for nearest neighbor image interpolation is presented. Evaluated rounding functions are selected among the five rounding rules defined by the Institute of Electrical and Electronics Engineers IEEE 754-2008 standard. Both full- and no-reference image quality assessment metrics are used to evaluate the influence of rounding functions on nearest neighbor interpolation image quality. The concept of the number of occurrences is used to determine the percentage of achieved occurrences, which is also equivalent to the sample proportion percentage used to calculate the margin of error. Inferential statistical analysis is used to deduce from a small number of images and draw a conclusion of the performance of each rounding function on a big number of images. Experimental results are provided to demonstrate with 95% confidence the minimum performance percentage of the most appropriate rounding function.