A new approach for open-close eye states detection: Complex wavelet transform and complex-valued ANN

Abstract

A novel method for open-close eye states detection, based on complex wavelet transform (CWT) and complex-valued artificial neural network (CVANN) is proposed in this study. Firstly, color information of images is used. Red images for eye are chosen as intensity image of color image. After getting the red image of seperately right and left eye, the color information is used to feature extraction with CWT. Features of eyes are extracted using CWT with 4th level and image size is reduced. After then, four statistical features (maximum value, minimum value, mean value and standard deviation) are obtained from extracted features. These new statistical features are presented to CVANN as inputs. Image set including ten person images with open and close eye states is used in this study, CVANN detected eye states with % 6.7 numerical test error. Classification results shown that, one of ten images is misclassified for two states.