Accession Number : AD0381131

Title :   THE INVERSION OF THE NORMAL EQUATIONS OF ANALYTICAL AEROTRIANGULATION BY THE METHOD OF RECURSIVE PARTITIONING.

Descriptive Note : Interim rept. Apr 66-Jan 67,

Corporate Author : BROWN (D) ASSOCIATES INC EAU GALLIE FL

Personal Author(s) : Gyer, Maurice S.

Report Date : APR 1967

Pagination or Media Count : 58

Abstract : The feasibility of the simulataneous reduction of large blocks of aerial photographs by analytical techniques reported by various sources had led to the requirement for a deeper analysis of the properties of the solution. Such an analysis requires as a minimum, the standard errors of the elements of exterior orientation of the photographs and the ground coordinates of the pass points as determined by the analytical aero-triangulation. An impediment, until now in obtaining these data was the large computational effort involved in computing the full inverse of the normal equations and the degradation of the results due to cumulative round-off error during the computation. This study reports a highly successful effort to overcome these difficulties by using a method which systematically takes into account the highly patterned structure of the reduced normal equations. The reduced normal equations, involving only the elements of exterior orientation, are primarily banded about the diagonal. The Method of Recursive Partitioning consists of an algorithm which systematically limits the computation to those entries of the matrix which are required for determining the standard errors of the unknown parameters. The savings in time and increase in computational precision over conventional approaches to direct inversion and solution make the operational implementation of the method highly desirable for solving the system of equations as well as the inversion of the associated coefficient matrix. (Author)

Descriptors :   (*PHOTOGRAMMETRY, AERIAL PHOTOGRAPHY), TRIANGULATION, AERIAL PHOTOGRAPHS, ERRORS, COMPUTER PROGRAMMING, DIGITAL COMPUTERS, REDUCTION, PHOTOGRAPHY, MATRICES(MATHEMATICS), PHOTOGRAPHIC ANALYSIS, ALGORITHMS.

Subject Categories : Cartography and Aerial Photography

Distribution Statement : APPROVED FOR PUBLIC RELEASE