Biometric modals for visual surveillience applications

• Main Author: Wai, Mon Kyaw
• Other Authors: Teoh Eam Khwang
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Biometrics
• Online Access: http://hdl.handle.net/10356/20721
• Institution: Nanyang Technological University

Video surveillance applications are increasingly developed using the biometric modals of human being such as face features, face postures and others soft biometric modals i.e. skin, hair and etc. Many developments have to be done on the human recognition such as estimation of age and gender, consistent human tracking and learning of movement with behavior. The main focus of this project is to develop the system to recognize similar people in video images using facial biometric features. Firstly, segmentation of the skin is to reject as much “non-face” of the image as possible. Conventionally, the face images are not always in straight eyes aligned position owing to the dependency of situation the images are taken and varying face postures. To generalize the face postures of all the test images, face alignment process using the coordinates of two eyes position has to be done as a pre processing.The comprehensive objective of this project is to focus on classification of family members from non family members using Gabor facial features of each family and select the classifier with best performance using the AdaBoost feature selection algorithm which is statistically robust, computationally efficient and global image processing algorithm.