Gabor Wavelet Based Edge detection on Hexagonal sampled Grid
P Vidya · 2011
Image processing in hexagonal grid is very much advantageous than in the conventional rectangular grid. The primary motivation behind using a hexagonal pixel grid model is that retina of the eye closely resembles a hexagonal grid space. Thus by modeling and processing images on such a grid space, one can mimic natural behavior to realize Computer Vision. Edge detection is one of the important preprocessing steps in many of the image processing applications. Wavelet based edge detection is found to be a better technique for various applications. In this paper directional Gabor Wavelet based edge detection is performed on hexagonal domain. For performance evaluation peak signal to noise ratio and mean square error is considered. Edge detection on hexagonal domain gives better results and better visual appeal of images compared to edge detection on rectangular domain. The paper is organized as follows. Section II deals with edge detection using Gabor wavelets on hexagonal sampling grid. Section III deals with performance evaluation and along with comparative study are performed with rectangular sampled grid.Section IV results and discussion and section V deals with conclusion. II.Hexagonal resampling and edge detection A. Hexagonal resampling This paper considers hexagonal grid based on half-pixel shift method [4]. For each odd line, find the midpoint between two adjacent pixels by simple linear interpolation (i.e., mid = ((left + right) / 2). Discard the left and right, keeping only the mid values. This gives us a hexagonal mapping from a regular square or rectangular grid [5].