
Accession Number : ADA294919
Title : Hybrid Pyramid / Neural Network Vision System.
Descriptive Note : Research and development status rept. for 1 Mar31 May 95,
Corporate Author : DAVID SARNOFF RESEARCH CENTER PRINCETON NJ
Personal Author(s) : Pearson, John ; Sajda, Paul ; Spence, Clay
PDF Url : ADA294919
Report Date : 31 MAY 1995
Pagination or Media Count : 10
Abstract : An artificial problem was constructed and described in the last report. During this quarter, another, more difficult, artificial problem was constructed. As before, the objects to be found and some potential false positives each have component patterns. Each positive has two different component patterns chosen from three types. In addition, the rectangles which form the objects and the subpatterns all have artificial shadows added (Figure 1). The angle of the shadow is randomlychosen. The potential false positive objects either have two component patterns of the same type, or only one pattern of some type, and one or both of the component patterns may not have a shadow. Thus the pattern tree must detect all three types of components with shadows to be able to detect a positive. As in the previous simpler problem, the positives and potential falsepositives are 18by11 pixel rectangles, but now their brightness is randomlychosen between 136 to 247. The subpatterns have a brightness that is independently chosen at random from the same range, but both patterns (if there are two in an object) have the same brightness. The component patterns are threebythree x, +, and square patterns. (KAR) P. 2
Descriptors : *NEURAL NETS, *PATTERNS, *STATISTICAL ANALYSIS, ANGLES, PROBLEM SOLVING, TREES, HYBRID SYSTEMS, VISION, SHADOWS, PYRAMIDS(GEOMETRY).
Subject Categories : Statistics and Probability
Cybernetics
Distribution Statement : APPROVED FOR PUBLIC RELEASE