Imaging spectroscopy is a relatively new technique in remote sensing which has become available in electronic form thanks to new matrix solid state detectors and improvements in optical technology. Imaging spectrometers have been recognized to be powerful instruments for the remote sensing of solar system bodies and have been included, as part of the payload, in many planetary missions. Electronic imaging spectrometers were firstly used for terrestrial remote sensing (Vane et al. 1984). The first spaceborne instruments designed for planetary observation were the Galileo Near - Infrared Mapping Spectrometer (NIMS), launched aboard the U.S. Galileo mission on 1990 with the aim to study the geochemistry of Jovian satellites and the atmosphere of Jupiter (Aptaker 1987) and the Phobos Infrared Spectrometer (ISM), flown on the Soviet Phobos space probe on 1988 (Bibring et al. 1989). To date, an other instrument has been developed for planetary exploration: the VIMS, Visible Near Infrared Mapping Spectrometer, designed for the CRAF (Comet Randevouz Asteroid Flyby) mission (Juergens et al. 1990) but now intended to observe the Saturnian system of rings and satellites from aboard the U.S. Cassini spacecraft. When solar radiation interacts with a planetary surface, the reflected component contains absorption bands which are diagnostic of the mineralogic composition of the surface (Adams 1974). An imaging spectrometer acquires many images in several (>100) narrow spectral bands, thus providing the spectrum of each pixel in the scene. Combining spectroscopy and imaging together is then possible to generate compositional maps of large areas. If an atmosphere is present between the surface and the instrument, the air turbulence degrades every monochromatic image, lowering the signal to noise ratio and worsening the spatial resolution. Moreover, in the case of push-broom imaging spectrometers which form a bidimensional image by scanning the slit across the planetary surface and acquiring the spectra of the regions imaged by the slit, the scanning mechanisms introduce an error caused by the noise in the driver electronics and tolerances in the mechanics. All these effects contribute to the degradation of imaging performance. These problems are not present in Fourier transform imaging spectrometers which do not have scanning elements even if pointing errors can still be present. Moreover, push-broom imaging spectrometers have an intrinsic limitation in the maximum achievable spatial resolution. This parameter, in fact, competes with the maximum achievable signal to noise ratio. Owing to the fixed scanning velocity, high spatial resolution images require relatively low integration times yielding, consequently, low signal to noise data. In the first section of this paper, the basic concepts of the methodology used to increase the content of spatial information of imaging spectroscopy data is discussed. The details are illustrated in the second part. The rest of the paper reports some results of the method applied on imaging spectroscopy data of the Moon. The last section discusses the possible extension of the technique to planetary remote sensing from space vehicles.
Copyright The European Southern Observatory (ESO)