Title :
A Peak-Trough Detection Algorithm Based on Momentum
Author :
Harmer, K. ; Howells, G. ; Sheng, W. ; Fairhurst, M. ; Deravi, F.
Abstract :
This paper presents a simple, yet novel, approach to peak-trough detection using a rudimentary model of Newtonian mechanics. Based on the line-searching technique also employed in artificial neural network technology to determine global minima, the momentum is used to find both peaks and troughs of a signal. This algorithm provides a fast alternative to the traditional techniques, which uses contextual information in order to determine prominent peaks and troughs without requiring smoothing or thresholding.
Keywords :
Artificial neural networks; Change detection algorithms; Detection algorithms; Friction; Gravity; Neural networks; Noise robustness; Shape; Signal processing algorithms; Smoothing methods; Peak Trough Detection Momentum;
Conference_Titel :
Image and Signal Processing, 2008. CISP '08. Congress on
Conference_Location :
Sanya, China
Print_ISBN :
978-0-7695-3119-9
DOI :
10.1109/CISP.2008.704
