A spectral histogram descriptor computes a set of marginal distributions based on the filter bank’s responses, and further encodes them into the images. The encoding process for local image structure takes place during the filtering stage, whereas the encoding process of global image feature is conducted during the histogram stage. One drawback of spectral histogram descriptors is their performances will be greatly deteriorated when the filter bank’s responses are not stochastically independent. To tackle this problem, a computational technique named Enhanced Independent Spectral Histogram Feature (EISHF) is proposed. EISHF is composed of four working modules: (1) unsupervised independent filter bank responses computation, (2) binary hashing, (3) XOR bitwise operation and feature encoding, and lastly, (4) block-wise histogramming. To ensure the performance of ordinary spectral histogram descriptors, an XOR operation has been delicately adopted to increase the independency of the filter responses. Tested on three public face databases, the experimental results have substantiated the performance of EISHF in handling different kinds of facial expressions, illuminations, time spans as well as facial makeup effects.
Bibliographical noteFunding Information:
This research work was supported by Fundamental Research Grant Scheme (FRGS) under the Ministry of Education and Multimedia University, Malaysia (Project ID: MMUE/140020) and Multimedia University Mini Fund (Project ID: MMUI/170037).
The FERET database was assembled to support the United State (U.S.) government in exaggerating face recognition technology. This FERET program was sponsored by the Department of Defense’s Counterdrug Technology Development Program through the Defense Advanced Research products Agency. Its primary mission is to develop automatic face recognition system to assist security, intelligence and law enforcement authorities in their duties. The data collection was performed within three years, from 1993 to 1996. Images of an individual were taken on different days. For some individuals, their photography took for a period of two years, i.e. first and last sessions were over two years lapse. This is to study the facial appearance changes of the individuals. Samples of the FERET images are illustrated in Fig. .
© 2017, Springer Science+Business Media, LLC.
All Science Journal Classification (ASJC) codes
- Media Technology
- Hardware and Architecture
- Computer Networks and Communications