Augmented Small-Scale Database to Improve the Performance of Eigenface Recognition Technique

2. Face Recognition Techniques

Human face recognition has several applications in computer visions and image processing. However, the human face is a complex natural object that tends not to have easily (automatic) identified edges and features. Due to this, it is challenging to acquire a mathematical model of the face that can be used as prior knowledge when analyzing a particular image (O’Toole et al., 2007; Hiremath & Prabhakar 2006). This expands the effort of the face recognition issue and opens a way of raising new techniques and algorithms, aiming to achieve optimum performance. Many algorithms have been examined to attain worthy performance under various conditions (Raffaele, Dario, & Davide, 2002; Zhou et al., 2011; Turk, 2001). These algorithms may be classified in to three categorizations:

• 1.

  Traditional algorithms: This refers to the identification of faces through the extractions of landmarks, or features, from an image of the subject's face. Popular recognition algorithms include Eigenface, Fisherface, the Hidden Markov model, and the neuronal motivated dynamic link matching.

• 2.

  3-D algorithms: This refers to a newly emerging trend, claimed to achieve previously unseen accuracies.

• 3.

  Skin algorithms: This refers to texture analysis, which is another emerging trend using the visual details of the skin, as captured in standard digital or scanned images.