Shenyang Ligong University, Shenyang, China

---

Abstract—Aimed at the problem how to detect the circular region and determine the parameters of circular in the digital image, two ways of the current common round recognition methods are introduced—one is based on the Hough transform, and the other is on the geometric characteristics. A fundamental theory of determining the radius and the center of the circle by scanning and counting the number of pixels vertically and horizontally through circular contour pretreated is exhaustively studied. A new circular region detection algorithm based on the geometric characteristics is proposed. Many experimental results show that the new circular region detection algorithm has a good effect both on processing efficiency and accuracy.

Index Terms—Geometric characteristics, circle determining, Hough transform, Freeman chain code.

[PDF]

Cite: Yueqiu Jiang, Xiangwen Fu, Hongwei Gao, "A New Circular Region Detection Algorithm based on the Geometric Characteristics," Journal of Software vol. 8, no. 11, pp. 2899-2907, 2013.