An elastic graph dynamic link model for tropical cyclone pattern recognition

Autor: R.S.T. Lee, Jihua Liu
Rok vydání: 2003
Předmět:
Zdroj: ICONIP'99. ANZIIS'99 & ANNES'99 & ACNN'99. 6th International Conference on Neural Information Processing. Proceedings (Cat. No.99EX378).
DOI: 10.1109/iconip.1999.843982
Popis: In this paper, an elastic graph dynamic link model (EGDLM) is proposed to automate the satellite interpretation process (Dvorak technique) and provide an objective analysis for tropical cyclones. The method integrates dynamic link architecture (DLA) for neural dynamics and the active contour model (ACM) for contour extraction of tropical cyclone (TC) patterns. Using satellite pictures provided by National Oceanic and Atmospheric Administration (NOAA), 145 tropical cyclone cases from the period between 1990 to 1998 were extracted for the study. An overall correct rate for TC classification was found to be above 95%. For hurricanes with distinct "eye" formation, the model reported a deviation within 3 km from the "actual eye" location, which was obtained from the aircraft measurement of minimum surface pressure by reconnaissance.
Databáze: OpenAIRE