의료영상처리를 위한 캐니 연산자 기반의 향상된 윤곽선 검출법

초록

This paper proposes an improved edge detection method based on Canny operator and its application for medical image processing. First, Non-Local(NL) means algorithm replaces Gaussian filtering for denoising. And then, 3D Sobel operator is applied to detect edge component for 3D data set of MR Images. In particular, NL-means approach outperforms other conventional filters for removing noise. Also, 3D Sobel operator is more suitable than 2D operator for MR Images which are reconstructive data set of 3D object. The experimental results demonstrate that the proposed method surpasses the conventional edge detector in MR Images.