Abstract
This paper describes a method that can improve the binarization of stripy patterns like wood rings or fingerprints while strongly suppressing noise. The method first divides the original image into small blocks, and then applies 2 kinds of frequency filters, a band pass filter and a least squares filter, to each block by using FFT. The application of the least squares filter to each block enhances the local dominant frequency components which can be considered as the local stripe's frequency components. The structure of the method is very simple because it doesn't need to discriminate the local direction of the stripe. The method can be applied to patterns whose stripe period varies like wood rings because it operates on the stripe's frequency in isolated areas. Experimental results indicate the effectiveness of proposed method.
