We illustrate and extend the techniques of computational mechanics in
explicating the structures that emerge in the space-time behavior of e
lementary one-dimensional cellular automaton rule 54. The dominant reg
ular domain of the cellular automation is identified and a domain filt
er is constructed to locate and classify defects in the domain. The pr
imary particles are identified and a range of interparticle interactio
ns is studied. The deterministic equation of motion of the filtered sp
ace-time behavior is derived. Filters of increasing sophistication are
constructed for the efficient gathering of particle statistics and fo
r the identification of higher-level defects, particle interactions, a
nd secondary domains. We define the emergence time at which the space-
time behavior condenses into configurations consisting only of domains
, particles, and particle interactions. Taken together, these techniqu
es serve as the basis for the investigation of pattern evolution and s
elf-organization in this representative system.