Feature Line Detection
Arthur Ardeshir Goshtasby · 2017
This chapter provides details of well-known line detection methods in the literature. The characteristics of the methods are examined and their performances are compared quantitatively and qualitatively. To measure the performances quantitatively, synthetic images containing known lines are used, and to evaluate the performances qualitatively, various natural images of man-made scenes are used. The chapter discusses steps involved in line detection by the Hough transform using the polar equation of lines. It describes a method that uses the parametric equation of lines to find lines among a set of points. In the chapter, the performance measures of some line detection algorithms were determined on a set of synthetically generated images. These performance measures are compared to determine the strengths and weaknesses of the algorithms. Also, the chapter compares the quality of lines detected by the algorithms in natural images.