1-D syntactic pattern recognition for the detection of bright spots

Kou-Yuan Huang, King Sun Fa

研究成果: Paper同行評審

摘要

In a seismogram, the wavelets of a bright spot have some specific structural pattern; so syntactic pattern recognition approach is proposed for the detection of bright spots. Testing traces are selected from the input seismogram, and tree classification techniques are used in the extraction of bright spot wavelet patterns. The system for one-dimensional (1-D) syntactic pattern recognition includes the likelihood-ratio test, optimal amplitude-dependent encoding, probabili- (Figure Presented) ty of detecting the signal involving in the global and local detection, and threshold setting. The relation between error probability and the minimum Levenshtein-distance classification is proposed. The system is used to detect the candidate bright spot, trace by trace, in a simulated seismogram as well as real seismograms of Mississippi Canyon and High Island.

原文English
頁面534-536
頁數3
DOIs
出版狀態Published - 1 1月 1984
事件1984 Society of Exploration Geophysicists Annual Meeting, SEG 1984 - Atlanta, United States
持續時間: 2 12月 19846 12月 1984

Conference

Conference1984 Society of Exploration Geophysicists Annual Meeting, SEG 1984
國家/地區United States
城市Atlanta
期間2/12/846/12/84

指紋

深入研究「1-D syntactic pattern recognition for the detection of bright spots」主題。共同形成了獨特的指紋。

引用此