Analysis of Kriging Applied to Resolution Enhancement of Digital SatelliteImagery

摘要

To aid the viewer in interpreting the information contained in an image, spatialresolution of the image can be increased. One way to increase the resolution is to improve the optics and sensors of the imaging system. Another is to process the data on the ground by computer. Enhancement techniques can be employed to alter the image and extract necessary data. Current techniques to manipulate and enlarge an image are nearest-neighbor, bilinear interpolation, and cubic convolution which can give greater clarity and readability. To satisfy the growing demand for quality (high-resolution) remotely sensed images, new techniques to enhance images need to be developed. Kriging is the geostatistical spatial estimation technique developed by Matheron that optimally predicts spatial data using observations at nearby locations. This research investigates the application of kriging as a technique of digital image resolution enhancement by adapting its use for the personnel computer and validating its capability and reliability. Twenty-seven percent of the images tested produced significantly lower means for kriging than cubic convolution with 45 percent of the kriged images producing lower variances.... Kriging, Semi-variogram, Variogram, Resolution, Cubic convolution, Bilinear Interpolation, Nearest neighbor.