| 摘要 |
Pictures (4, 5) are taken at times (t-dt) and (t) of a particle environment such as clouds and skeletons (9,10,11,12) which characterize their shapes are produced from an algorithm. A prediction (6) of the cloud evolution for a time (t+dt) is produced by generating vectors (13,14,15,16) which connect corresponding skeleton points at the two previous times and extending them by an equal amount to produce predicted skeletons (21,22) |
