The Multi-Modal Human Identification Based on Smartcard in Video Surveillance System

摘要

Recently, the use of security surveillance system including CCTV is increasing due to the increase of terrors and crimes. However as the data recorded through the video surveillance system is exposed, the invasion of privacy is raised. The research on the technology of protecting privacy from the surveillance system in addition to these organizations is actively carried out. At this time, to distinguish the human of privacy protection, the identification technique is necessary. For identification, the proposed system uses soft-biometrics such as color and height information based on smart card. It can obtain reliable feature information using smart card. In this paper, representative colors are extracted by applying octree-based color quantization technique to the clothing region and height is extracted from the geometrical information of the images. The identification is accomplished by comparing the similarities between two data based on Euclidean distance. Through identification by using multiple information rather than single information, the proposed system demonstrates the possibility of human identification under the surveillance camera environment.