Moving edge detection with sub pixel accuracy in 1-D images

Abstract

Image processing includes many various procedures and one of the mostly used is edge detection. Sometimes is desirable to improve precision of edge detector to sub-pixel range. In our paper we deal with precise localization of edge which is moving during the exposure time. For such the edges we tested three edge detectors with sub-pixel precision in 1-D images: algorithm that uses function erf to approximate real image samples and two edge detectors based on gray level and spatial moments of the image function. This paper presents results of simulations with noisy images, we have chosen the standard deviation of error of edge position as the precision criterion.