Gait with a combination of swing arm feature extraction for gender identification using kinect skeleton

Biometric is crucial in improving security. Many cases of crime to identify the perpetrators can be identified by biometrics, which are difficult to identify with other technologies. Research on biometric focus on the physical characteristics of a human being or human behavior. The way humans walk is one of human behavior that has uniqueness, the uniqueness of these patterns can be used to gender identification. In this study proposes a variety of features to identify gender based gait. These features are taken from the skeleton provided by kinect camera. There are four features are measured, there are three features of gait were recorded from the side that is the width of the foot when walking, the width of the swing arm and high ankle when lifted from the floor. The last feature is the width between 2 feet (from left ankle to right ankle) were taken from the front of the camera. Result, these features to 80% can be used to identify gender of a person.