In this paper, a syntactic analysis approach for target classification and a correlation analysis approach to detect target formations (patterns) are defined and evaluated. Examples of simulation results are also given. In the syntactic approach, the structural relationship of target primitives is analyzed through a grammar set. These primitives can be interior components of a large target or individual small targets in a target grouping. A contextfree grammar (CFG) is constructed to describe different targets and target formations. This approach was demonstrated on a large target of an airport and on a target cluster of cars on the highway. In target formation detection, spatial correlation between the target and all predefined reference formations is analyzed to determine the best match. This correlation analysis approach was evaluated on a "mesh" formation of residential houses, a "line" formation of cars on the highway, a "U" formation of residential houses, and an "L" formation of planes in a plane parking area.
Yun-Kung J. Lin,
"Tactical Targeting By Structural And Correlation Analysis", Proc. SPIE 0432, Applications of Digital Image Processing VI, (9 January 1984); doi: 10.1117/12.936684; https://doi.org/10.1117/12.936684