Abstract
The genetic algorithm was already applied with success to the particle pairing process of the 3-D stereoscopic particle tracking velocimetry by the present author. In the present study, the algorithm has been further improved for more accuracy of particle pairing and reduced computation time. The point of the improved algorithm is a new-concept genetic encoding scheme, according to which the gene strings of each individual do not start from globally random sets of particle ID's but from pre-selected candidate pairs of ID's. The crossover and mutation schemes are likewise modified from conventional methods in order to meet with this new strategy. Thanks to the pre-selection of candidate pairing particles, the algorithm converges more quickly and the number of erroneous pairs is reduced.
