The Image Analysis Using Fast Radial Symmetry Transform in Control Systems Base on the Computer Vision
Stella A. Lyasheva, Mikhail P. Shleymovich, R. M. Shakirzyanov · 2019 International Multi-Conference on Industrial Engineering and Modern Technologies (FarEastCon) · 2019
Currently computer vision-based control systems are widely used. In such systems, procedures for detecting specified objects in images, calculating their characteristics and generating information for decision making are implemented. Mostly, in practice, the detected objects have the property of radial symmetry. For example, round markers are used as such objects, providing precise positioning in robotic systems. For their detection various methods of processing and analyzing images can be used. In this paper, we consider an approach based on the fast radial symmetry transform. As its application example, an algorithm for detecting centers of radially symmetric objects is considered. It is shown that this algorithm allows detecting the centers of radially symmetric objects in the conditions of noise that interfere with scene analysis in real conditions of the control systems operation. Besides, it shows how the algorithm is used in combination with the circles-detecting Hough method. The Hough method has a number of parameters, the setting of which is a time-consuming process as they are very sensitive to the size and location of the circles. The application of the algorithm for the circles' centers detecting makes it possible to refine the found circle by the degree of proximity of the center, obtained using the Hough method, to one of the centers obtained by fast radial symmetry transform.