|
One of the main parameters describing the quality and interpretational usefulness of satellite and aerial remote sensing images is their spatial resolution. Imagery characterized by a high ground resolution enable the interpreter to conduct a more detailed analysis and more thorough interpretation than possible with images of lower resolution. When working with digital sensors we can distinguish two parameters which define the spatial resolution of images: GRD (Ground Resolved Distance) and GSD (Ground Sampling Distance). GSD is the sampling frequency and only describes the size of the pixel on the ground. The GRD parameter describes the smallest length which can be recognized on the image. The ground resolved distance is determined based on especially constructed calibration targets. These targets can take several forms, shapes and sizes. They are characterized by a difference in contrast between its individual segments. Calibration targets have been used for many years to determine the ground resolved distance of analogue sensors. The newest research on digital sensors based on the same traditional calibration targets have shown that some of these targets are now inadequate. In order to correctly determine the resolution of digital sensors, changes must be made to the structure of the targets. The GRD value calculated based on existing targets is laden with large errors caused by artifacts occurring on the acquired images. These are the result of the way energy is registered by a digital sensor. A research team from the Remote Sensing and Photogrammetry Department of the military University of Technology has conducted a number of experiments which have allowed to determine the optimal parameters of a calibration target for establishing the ground resolved distance of digital sensors. Additionally, the experiments allowed description of a new methodology for analyzing the acquired imagery. In these experiments a number of parameters specific to digital imagery had been analyzed.
|