Performance Comparison of Several Pre-Processing Methods in a Hand Gesture Recognition System based on Nearest Neighbor for Different Background Conditions

Regina Lionnie (Department of Electronic Engineering, Satya Wacana Christian University, Jalan Diponegoro 52-60, Salatiga, 50711, Indonesia)
Ivanna K. Timotius (Department of Electronic Engineering, Satya Wacana Christian University, Jalan Diponegoro 52-60, Salatiga, 50711, Indonesia)
Iwan Setyawan (Department of Electronic Engineering, Satya Wacana Christian University, Jalan Diponegoro 52-60, Salatiga, 50711, Indonesia)

Journal of ICT Research and Applications
Vol. 6 No. 3 (2012)

This paper presents a performance analysis and comparison of several pre-processing  methods  used  in  a  hand  gesture  recognition  system.  The  preprocessing methods are based on the combinations ofseveral image processing operations,  namely  edge  detection,  low  pass  filtering,  histogram  equalization, thresholding and desaturation. The hand gesture recognition system is designed to classify an input image into one of six possibleclasses. The input images are taken with various background conditions. Our experiments showed that the best result is achieved when the pre-processing method consists of only a desaturation operation, achieving a classification accuracy of up to 83.15%.