English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Among the numerous biometric systems presented in the literature, face recognition systems have received a great deal of attention in recent years. The main driving force in the development of these systems can be found in the enormous potential face recognition technology has in various application domains ranging from access control, human-machine interaction and entertainment to homeland security and surveillance (Struc et al., 2008a). While contemporary face recognition techniques have made quite a leap in terms of performance over the last two decades, they still struggle with their performance when deployed in unconstrained and uncontrolled environments (Gross et al., 2004; Phillips et al., 2007). In such environments the external conditions present during the image acquisition stage heavily influence the appearance of a face in the acquired image and consequently affect the performance of the recognition system. It is said that face recognition techniques suffer from the so-called PIE problem, which refers to the problem of handling Pose, Illumination and Expression variations that are typically encountered in real-life operating conditions. In fact, it was emphasized by numerous researchers that the appearance of the same face can vary significantly from image to image due to changes of the PIE factors and that the variability in the images induced by the these factors can easily surpass the variability induced by the subjects’ identity (Gross et al., 2004; Short et al., 2005). To cope with image variability induced by the PIE factors, face recognition systems have to utilize feature extraction techniques capable of extracting stable and discriminative features from facial images regardless of the conditions governing the acquisition procedure. We will confine ourselves in this chapter to tackling the problem of illumination changes, as it represents the PIE factor which, in our opinion, is the hardest to control when deploying a face recognition system, e.g., in access control applications. Many feature extraction techniques, among them particularly the appearance based methods, have difficulties extracting stable features from images captured under varying illumination conditions and, hence, perform poorly when deployed in unconstrained environments. Researchers have, therefore, proposed a number of alternatives that should compensate for the illumination changes and thus ensure stable face recognition performance. 12

From Gabor Magnitude to Gabor Phase Features: Tackling the Problem of Face Recognition under Severe Illumination Changes