The combined method of improving image segmentation

Бесплатный доступ

The article deals with the problem of automatic segmentation of digital images via piecewise constant approximations with a limited number of connected segments, constituted of averaged pixels. The solution is achieved by combining of global improvement of approximation across entire image field and local improvement of approximation fragments. The effectiveness of the combined method is demonstrated on the actual example of scene snapshots segmentation. Shots were taken from different angles and then merged into a single image.

Image, approximation, segmentation, standard deviation, minimization

Короткий адрес: https://sciup.org/148183096

IDR: 148183096   |   DOI: 10.18097/1994-0866-2015-0-9-118-124

Статья научная