This work introduces the aspects of vegetation cover classification using records with different resolution. Experimental data was comprised of multispectral aerial images of third level processing. The original spatial resolution (0.2 m/pixel) was recalculated to 3 different resolution levels of 0.4; 0.8 and 1.6 m/pixel. The area of study is located in the eastern part of West Tatra Mts – Liptovské Kopy, where sites with training polygons were marked and geolocated in the field, respresenting training site classes. In total, 234 polygons in 14 classes were defined, including: (1) mountain pine, (2) arolla pine, (3) spruce, (4) gravel, (5) meadows 1, (6) soil degradation, (7) meadows 2, (8) meadows 3, (9) tarmac, (10) water, (11) fallen forests, (12) dead spruce, (13) larch, (14) disturbed mountain pine. The aerial images were processed using different methods of digital image processing (PCA analysis, vegetation indexes, image texture analysis), producing 30 new image files for each spatial resolution level. Discrimination analysis and calculation of transformed divergence was used to classify 7 selected image files. Spectral signatures were calculated for all selected files and classification was achieved by using maximum likelihood classification. The accuracy of classification was evaluated in order to find the most suitable resolution for an automated classification of vegetation cover. The highest accuracy was reached at the resolution of 0.8 m/pixel, determined by Kappa index 62.9 %.