The Automatic Analysis of Aerial Photographic Data

Abstract

As the volume of aerial photographic data increases in various applications, there is increasing demand for automated data analysis. Faster and larger computer alone is not the solution. In this paper effective and computationally efficient techniques for computer processing of the aerial photographic data are presented. They include: (1) picture data compression, (2) feature exttraction using the histograms of the original and sharpened pictures, (3) sequential target and classification, (4) threshold selection, and (5) Walsh power spectrum analysis. All of these techniques may be incorporated in a fully automated data analysis system to meet certain real-time on-line system requirement.