DETECTION OF CONTOURS IN HYPERSPECTRAL IMAGES
V. V. Shipko1, E. A. Samoylin1, V. E. Pozhar2, A. S. Machikhin2
1Military Educational Scientific Center of the Air Force "Air Force Academy named after Prof. N.E. Zhukovsky and Yu.A. Gagarin", Voronezh, Russia 2Scientific and Technological Center of Unique Instrument Engineering, Russian Academy of Sciences, Moscow, Russia
Keywords: hyperspectral images, detection of contours, gradient, correlation coefficient, interspectral difference
Abstract
The main approaches to identifying the contours of objects in hyperspectral images are considered. Algorithms for extracting the contours of spectrally selective objects based on the spatial spectral correlation and interspectral difference of gradients are presented. The efficiency of the proposed algorithms for processing real hyperspectral images under the condition of additive Gaussian noise is demonstrated.
|