I.J. Information Technology and Computer Science, 2015, 07, 19-27 Published Online June 2015 in MECS (http://www.mecs-press.org/) DOI: 10.5815/ijitcs.2015.07.03 Copyright © 2015 MECS I.J. Information Technology and Computer Science, 2015, 07, 19-27 A Gender Recognition Approach with an Embedded Preprocessing Md. Mostafijur Rahman Institute of Information Technology, University of Dhaka, Dhaka, Dhaka-1000, Bangladesh Email: bit0312@iit.du.ac.bd Shanto Rahman, Emon Kumar Dey, Mohammad Shoyaib Institute of Information Technology, University of Dhaka, Dhaka, Dhaka-1000, Bangladesh Email: {bit0321@iit.du.ac.bd, emonkd@iit.du.ac.bd, shoyaib@du.ac.bd} Abstract—Gender recognition from facial images has become an empirical aspect in present world. It is one of the main problems of computer vision and researches have been conducting on it. Though several techniques have been proposed, most of the techniques focused on facial images in controlled situation. But the problem arises when the classification is performed in uncontrolled conditions like high rate of noise, lack of illumination, etc. To overcome these problems, we propose a new gender recognition framework which first preprocess and enhances the input images using Adaptive Gama Correction with Weighting Distribution. We used Labeled Faces in the Wild (LFW) database for our experimental purpose which contains real life images of uncontrolled condition. For measuring the performance of our proposed method, we have used confusion matrix, precision, recall, F-measure, True Positive Rate (TPR), and False Positive Rate (FPR). In every case, our proposed framework performs superior over other existing state-of-the-art techniques. Index Terms— Contrast Enhancement, Gender Recognition, Feature Extraction, Classification, Preprocessing I. INTRODUCTION Nowadays gender recognition from facial image has become an active research area in the field of computer vision for different applications such as biometric authentication, surveillance systems, market analysis, security systems, etc. Gender can be recognized in different ways such as from human body shape, gait component [29], facial image [3-15, 28], etc. There have been a number of gender recognition approaches but none of these able to provide satisfactory result in some uncontrolled situations. Moreover, some facial images are so confusing that most of the time human beings fail to recognize gender from facial images. Fig.1 represents some such types of image which are taken from LFW database [17]. So, there is a wide scope of improving the performance of gender recognition approaches. This paper presents a novel approach for gender recognition from facial images. (a) (b) Fig. 1. Real life confusing image (a) male but looks like female, (b) female but looks like male Gender recognition from facial image can be broadly categorized into two groups such as appearance-based approach and feature-based approach. Appearance-based approach uses the global feature of the whole face image as an attribute for classification whereas feature-based approach uses a set of discriminative facial features such as nose, eye-brow, cheek, etc which are extracted from facial image as classification attributes. Most of the time the male faces have prominent nose, eye-brow over the female faces and on the contrary the female faces have clear cheeks. Moreover, regions of nose and chin show a greatest difference between male and female faces [15]. Lapedriza et al. [8] classified facial features into two parts such as internal and external. Eyes, nose and mouth were considered as internal parts in where hair, chin and ears were taken as the external part. Noteworthy, the more essential information can be gained from external facial portion. For their experimental purpose, they used FRGC database. Baluja et al. [10] proposed a boosting sex identification performance method in where gender recognition is performed based on boosting pixel comparison. They used FERET database and SVM on images of 20×20 pixels. Makinen et al. [11] proposed a gender recognition method by considering FERET database. All of the above studies have some universal problems because they used FERET database in where images are captured under controlled situation such as frontal faces, consistent lighting, etc. However, it is almost impractical to capture images in real time gender recognition applications in controlled situation due to