Lying Facial Recognition Using Content Based Image Retrieval Approach In The Compressed Domain

Kategori : Komputer
muhammad said hasibuan

ABSTRAK

In this paper we explore content based image retrieval to identify lying facial in the compressed domain. The work was conducted by constructing an application or tool which automatically detect and recognize someone dishonesty accurately based on face recognition. In this research we used more than 30,000 face images which consist of normal face expression and three lying facial categories. We wrote two face detection and face recognition algorithms to decide whether someone lies or honest. This work was carried out to bulit an automatic tool of lying facial detection and recognition fastly and accuratelly. Face detection done to separate between face area and non face area in order focus on face recognition. In this work we used face image with normal pose without obstacles and facing toward with 90 degree. In the experiments we apply more 150 queries which 50 images from each category of total of three categories (lying facial A, B and C categories). Our algoritms demonstrate good effectiness of image retrieval in term of Precision and recall.

Kata Kunci : recognition, matching, retrieval, lying facial,

Download File    16 - 02 - 2012