%0 Generic %A Dahle, Felix %D 2024 %T Training data semantic segmentation historical Images %U %R 10.4121/052fa1dc-cf10-4bef-8b64-aeda99fbe3a4.v1 %K Antarctica %K Machine Learning %K Semantic Segmentation %K Historical Images %X <p>This dataset contains historical images and their labelled counterparts used to train a network for semantic segmentation of historical images. The images are historical aerial imagery from Antarctica, made between 1940 and 1990.</p><p>The images are black and white and can be nadir (V) or oblique (L or R).</p><p>The labelled images are labelled manually and are the segmented equivalents of the historical images. They contain 6 classes:</p><p>1: ice, 2: snow, 3: rocks, 4: water, 5: clouds, 6:sky, 7: unknown.</p><p>A pixel value of 0 means unlabelled.</p><p><br></p> %I 4TU.ResearchData