Title :
Fingers segmentation and its approximation
Author :
Jetensky, Pavel ; Marek, Jaroslav ; Rak, Josef
Author_Institution :
Fac. of Electr. Eng. & Inf., Univ. of Pardubice, Pardubice, Czech Republic
Abstract :
Aim of the paper is to find an algorithm for detecting finger´s knuckle in the binary image with high precision and accuracy. At our disposal are measured coordinates of points on the finger, which are obtained from Microsoft Kinect. We form a suitable model for finger´s characterization. It is based on the transformation of coordinates and regression models. A regression model gives a non-smooth function with change point. The main goal is to find the change point of the approximation function corresponding to the knuckle.
Keywords :
approximation theory; image segmentation; regression analysis; Microsoft Kinect; approximation function; binary image; change point; coordinate transformation; finger characterization; finger knuckle detection algorithm; finger segmentation; nonsmooth function; regression model; Algorithm design and analysis; Approximation algorithms; Coordinate measuring machines; Estimation; Function approximation; Image segmentation;
Conference_Titel :
Radioelektronika (RADIOELEKTRONIKA), 2015 25th International Conference
Conference_Location :
Pardubice
Print_ISBN :
978-1-4799-8117-5
DOI :
10.1109/RADIOELEK.2015.7129026
