Разработка алгоритма обнаружения объектов по изображению с телевизионной камеры подводного аппарата

Вадим Тришин, машиностроения и транспорта Санкт-Петербургский политехнический университет Петра Великого. Институт металлургии, Сергей Чупров, Владимир Уланов
2019
The object of the research is the algorithms for object detection in lowcontrast images and especially on images obtained from a camera under water. The aim of the work is to develop an algorithm for detecting objects that does not require training and is applicable on low-power machines. The paper discusses options for existing methods for detecting objects, different options for the conditions of the underwater environment, formulated recommendations on the use of methods in different scenes.
doi:10.18720/spbpu/3/2019/vr/vr19-775 fatcat:cp56t7pfv5atvbfsqdfru77kqe