PROPOSAL OF AN IMAGE INTERPRETABILITY INDEX – ANALYSIS OF VEGETATION MAPPING
ISBN 978-85-88783-11-9
Authors
1Teixeira, A.; 2Barros, R.; 3Madureira, C.; 4Rangel, R.
1IBGE Email: alexandre.teixeira@ibge.gov.br
2UFRJ Email: rbarros@ufrj.br
3UFRJ Email: cmad@ufrj.br
4UFRJ Email: romulorangel17@gmail.com
Abstract
The purpose of this study is to present and discuss the concept of interpretability of orbital images used in the production of a reference cartographic base. In this context, interpretability means the suitability of an image for purposes of interpretation, in such a way that its content allows the detection and identification of objects on Earth’s surface. This study proposes to assess vegetation data derived from classification of images obtained by satellites Landsat-8 OLI, ALOS/AVNIR-2 and RapidEye. The choice of particular methods of quality control allowed the construction of an interpretability indicator. In geometric and interpretation terms, it is important to detail the knowledge on issues that are crucial to define which sensors are suitable to reach the expected geometric and thematic accuracy for a given cartographic scale. In general, when the images meet the requirements for positional accuracy for a given scale, they rarely offer satisfactory conditions for feature extraction on the same scale. This suggests that the positional and thematic accuracies do not necessarily indicate the same analysis scale. Added to the assessment of an image positional accuracy, the possibility of generating an interpretability indicator can help in choosing the most appropriate inputs for cartographic production. Finally, this paper deeply analyzes the knowledge on key issues in the definition of the most appropriate sensors to reach the expected accuracy for a given cartographic scale.
Keywords
Interpretability, ; positional accuracy; Cartographic