A new structure of decision tree based on oriented edges gradient map for circles detection and the analysis of nano-particles

In this paper, we propose a new approach to detect circles and nano-particles based on an oriented-edges gradient map and a decision tree. The decision tree is calculated from geometric constraints based on particular right triangles inscribed in a circle. Use of the proposed accumulator and dynamic storage matrix radii shows the robustness of our algorithm in terms of results and execution time. This robustness can also be enhanced in the event of prior knowledge. Indeed, we can enable or disable intermediate nodes or a part of nodes of the proposed decision tree to strengthen both the detection results and the execution time of the algorithm. Our approach makes it possible to detect circles and analyse the distribution of the nano-particles which is evaluated using four databases which include TEM, synthetic, real and complex images.