<title>Examination of ERS and IRS-1C images for interactive image interpretation</title>

Abstract

In the field of remote sensing, efficient image analysis necessitates defined image requirements for the tasks which have to be solved. Image interpretability scales, such as the National Imagery Interpretability Rating Scale (NIIRS), define different levels of image quality or interpretability by the types of tasks an analyst can perform with imagery of a given rating level. These scales result from subjective observer interrogations and exist for the sensor types visible, infrared, SAR (synthetic aperture radar) and multispectral. Another approach is persecuted at the IITB. Within objective observer experiments image interpreter have to carry out defined tasks on images with defined sensor parameters. Experiment results are performance parameters which give information about the solvability of the tasks on the examined image parameters. Two commercial satellite sensors (optical and SAR) with a resolution of 25 meters were examined. Goal of the investigation was to find out which tasks can be solved with the optical and/or the SAR image. The inspected tasks were deduced from tasks used in the NIIRS. Also it was examined if the accuracy of targeting decisions can be enhanced by providing the image interpreters both pictures.