Edge detection using determinant-variance trace algorithm based on Roots-Dimensional Mapping method
Pouria Khanzadi, Babak Majidi, Ehsan Akhtarkavan · 2017
Edge detection is an important issue in the field of computer vision and image processing. Detection of the edge segments in complex natural images requires the edge detection algorithm to be able to find the edges in high detail areas of the images. Furthermore, the ability of the edge detection algorithm to find high amount of details in the image will results in better object detection of in complex scenes. In this paper, a new algorithm for edge detection called Determinant-Variance Trace algorithm (DVTAR) is proposed. The proposed algorithm is capable of detecting higher amount of conceptual details in the images. This method is based on Maximum Determinant State (MDS) theory and Root-DImensional MApping (RDIMA) method. The comparison of the results with standard edge detectors shows that the proposed algorithm is capable of detecting fine details in the various categories of standard images.