SUMMARY
There are a lot of applications of pattern recognition which need input image with a certain size. The size effect the result of pattern recognition. Determining size of image adopts interpolation technique. Interpolated image’s quality depends on interpolation technique. Texture is the main feature which is used in image processing and computer vision to classify object. One of some methods that are used to characterize texture is statistical methods. Statistical methods characterize texture by the statistical distribution of the image density. This research compared 4 interpolation methods (Nearest Neighbor Interpolation, Bilinear Interpolation, Bicubic Interpolation and Nearest Neighbor Value Interpolation) and 6 features of 10 test images. Based on 6 features which are researched, skewness changes upto 800%, energy 90%, entropy 75%, smoothness 18%, standard deviation 10% and mean 0,9%.Index Terms—Interpolation, Statistical feature, NNI, Bilinear Interpolation, NNV, Bicubic Interpolation