In this paper, we present a practical and potential useful approach to spatial spectral feature extraction of hyperspectral imagery. Many new hyperspectral imaging sensors have collected hyperspectral data cubes in optical -infrared wavebands. The data cubes have two spatial dimensions and one spectral dimension providing hundreds of images of the same scene in different wavebands. Radiance clutter may interfere with the detection of specific target signals in the data cubes. But, target signals and background sources generally have different spatial features in different spectral bands. For example, target signals may have a higher contrast than the local background in one band but not in a different band. We exploit these band differences to detect the target signals and extract spatial and spectral features. In our analysis we use real image cube data from sensors such as the Fourier Transform Hyperspectral Imager (FTHSI). We combine traditional spatial processing, such as frame differencing and adaptive filters, but apply them to different image bands instead of different images of the same scene obtained at different times. We compute the probabilities of detection and false alarms for targets of a given strength against the measured optical clutter. We compare target detection algorithms using only one band with those using multiple bands.A proliferation of new hyperspectral image sensors has lead to new approaches to remote sensing and automatic target recognition 2, 3, 4, 5 These sensors have compiled archives of hyperspectral data at visible and infrared wavelengths. These data cubes possess unique properties that allow new methods of detecting desired target signatures in the presence of background clutter. Generally, the hyperspectral image cubes contains two spatial dimensions and many wavelength bands with a two-dimensional image for each band. Thus, if a spectral signature of target is known and it is different than the background clutter, detection algorithms that combine target spectrum with the target shape can enhance the probability of detection while decreasing the probability of false alarm.