An Improved Canny Edge Detection Algorithm for Detecting Brain Tumors in MRI Images

Abstract: Medical image processing represents an important research topic. Computer aided diagnostic application both, reduce time and improve quality of diagnostic process. In this paper an edge detection algorithm, specially adjusted for processing brain MRI images is presented. LoG filer was introduced as the first step of improved Canny algorithm. Also, gradient magnitude and kernel gradient were adjusted specially for edge detection in brain MRI images. This algorithm is based on the improvement of the traditional Canny algorithm. The proposed method was compared with other standard edge detection methods, and it was noticed that it produces more detail edge detection. The simulation results have shown that the proposed algorithm is more noise-resilient and better in edge and detail detection than standard Canny algorithm.