Issues of uncertainty in super-resolution mapping and their implications for the design of an inter-comparison study

摘要: Super-resolution mapping is a relatively new field in remote sensing whereby classification undertaken at finer spatial resolution than that of the input remotely sensed multiple-waveband imagery. A variety different methods for super-resolution have been proposed, including pixel-swapping, simulated annealing, Hopfield neural networks, feed-forward back-propagation networks and geostatistical methods. The accuracy all these approaches has tested, but tests tended to focus on technique (i.e. with little benchmarking against other techniques) used measures accuracy. There is, therefore, need greater inter-comparison between various available, study would be welcome step towards this goal. This paper describes some issues should considered design such study.