Цифровая трансформация (Apr 2021)
Methods for Determining the Size of three-Dimensional Objects in Images
Abstract
Computer vision technology is used to obtain the necessary information from images. The two-dimensional image was transformed into its corresponding three-dimensional structure based on the reconstruction process and singular points. A review of software tools designed for scanning objects and measuring their dimensions is performed. A mathematical model and algorithm for constructing three-dimensional structures of objects in images are presented. Based on the design pattern, the architecture of the software application is developed. It consists of an image processing module, a module for calculating the size of objects, and a module for storing the received data. The fastest algorithm for finding singular points and their descriptors is used. In this paper, both homogeneous and Euclidean coordinates of characteristic points are calculated. The method of using the mobile application in automatic and manual modes is described. The results of scanning objects with the camera of a mobile device and determining their size in real time are presented.