Title :
Parametric optimization for EPGVF snake using ANOVA and Taguchi method
Author :
Deng, Runhong ; Fox, Martin D.
Author_Institution :
Connecticut Univ., Storrs
Abstract :
In this study, Taguchi experimental design method is used to determine the optimal parameters for EPGVF snake segmentation algorithm. Optimum factor levels of processing parameters as well as quantitative contributions of each parameter are obtained by performing analysis of mean (ANOM) and analysis of variance (ANOVA), respectively. A verification test is also performed to prove the effectiveness of Taguchi technique after the optimum levels of parameters are determined. The results show better performance is obtained by using optimal parameters than the empirical parameters.
Keywords :
Taguchi methods; biomedical ultrasonics; edge detection; image segmentation; medical image processing; optimisation; ANOVA; EPGVF snake; Taguchi Method; carotid intimal-medial; edge preserving gradient vector flow; image processing; object boundary detection; parametric optimization; snake segmentation algorithm; ultrasound images; variance analysis; Analysis of variance; Biomedical engineering; Computed tomography; Design for experiments; Optimization methods; Performance analysis; Performance evaluation; Power engineering and energy; Testing; Ultrasonic imaging;
Conference_Titel :
Bioengineering Conference, 2007. NEBC '07. IEEE 33rd Annual Northeast
Conference_Location :
Long Island, NY
Print_ISBN :
978-1-4244-1033-0
Electronic_ISBN :
978-1-4244-1033-0
DOI :
10.1109/NEBC.2007.4413302
