Gender identification from facial images using global features
Salma Aouled El Haj Mohamed, Nahla Nour, Serestina Viriri · 2018
Human face presents a unique feature of human being, and identifying the gender of a person by these features seems to be an easy task for humans, but it is a challenging task in computer vision. In this paper we used efficient global gender classification methods. The face part of the image is segmented using Viola and Jones face detection technique which excludes unwanted area from the image and thus reduces image size, Histogram equalization is performed to normalize the illumination effect, the features are extracted using four methods: Zigzag and Block Discrete Cosine Transform (DCT and Block-DCT), Discrete Wavelet Transform (DWT) and hybrid algorithm, which combine (DWT) and (DCT). K-Nearest neighbour (KNN), fuzzy of KNN and support vector machine (SVM) are used for classification. The face images used in this study are taken from FERET and ESSEX datasets. The result showing that hybrid DWT-DCT algorithm performs much better when we applied the F-KNN and SVM classifiers in both datasets. But when we applied KNN classifier the hybrid DWT-DCT achieved higher accuracy with ESSEX dataset but the DCT achieved higher with FERET dataset.